Assignment 8

tushardeep05

Central_line_associated_bloods.pdf

Home >Nursing homework help >Assignment 8

CENTRAL LINE ASSOCIATED BLOODSTREAM INFECTIONS IN SOUTH

CAROLINA: A CORRELATIONAL STUDY OF BED CAPACITY, SURVEILLANCE

TECHNIQUES, AND THERAPEUTIC PROCESSES

Donald M. Peace

ROSSLYN BYOUS, DPA, Faculty Mentor and Chair

RAY BORGES, DHSc, Committee Member

JULIA MOORE, PhD, Committee Member

Charles Tiffin, PhD, Dean, School of Public Service Leadership

A Dissertation Presented in Partial Fulfillment

Of the Requirements for the Degree

Doctor of Philosophy

Capella University

April 2011

UMI Number: 3449355

INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a complete manuscript

and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion.

UMI 3449355

unauthorized copying under Title 17, United States Code.

ProQuest LLC 789 East Eisenhower Parkway

P.O. Box 1346 Ann Arbor, MI 48106-1346

Abstract

Central line associated bloodstream infections (CLABSI) is a concern for healthcare

providers, hospital administrators, third party payers and the patient. This research

determined what surveillance practices and therapeutic regimes were effective to

ameliorate the incidence of hospital-acquired infections associated with central line

catheters. Effective surveillance and therapeutic best practice strategies offer the greatest

means of minimizing infections for patients in acute care hospitals that require central

lines. Infection control programs are important for CLABSI prevention by developing

clinical strategies that modify risk factors. Prevention strategies typically cost very little

but make a significant impact on CLABSI. A quantitative research design with a

correlational approach to conduct the study is demonstrated in this dissertation. The

study compared samples of all reported CLABSI cases in the study population with bed

capacities of 50 or less, 51 to 200, 201 to 500, and 501 or more beds from January 1,

2008 through December 31, 2008. Multivariate Analysis of Variance (MANOVA) was

employed since more than one dependent variable existed in the study.

In conclusion, active surveillance programs in conjunction with therapeutic regimes as

observed with bundle packaging were proven effective adjuncts to care for those

requiring central lines. The integration of multiple therapeutic measures used in central

line bundling was associated with lower infection rates when compliance of use was high.

This research determined that hospitals must target bundled implementation and

compliance as opposed to integrating policies on infection control. Results identified that

of the 63 acute care hospitals, the mean CLABSI rate was 6.62 per 1000 central line days

with a Standardized Infection Rate (SIR) of 1.0004 for hospitals with 50 beds or less.

Infection rates of 3.93 and a SIR of 2.981 was seen in hospitals of 51 to 200 beds,

infection rates of 2.55 and a SIR of .9981 in hospitals of 201 to 500 beds and infection

rates of 4.21 and a SIR of .9996 in hospitals with greater than 500 beds. Larger facilities

had more resources human, material, and policy and reflected the best infection rate and

lowest SIRs that revealed lower CLABSI incidences.

iii

Dedication

This dissertation is dedicated to my wife, Marie. She has encouraged my lifelong

goal of achieving a doctoral degree. Her steadfast support inspired me to complete the

research and writing necessary to achieve my educational goals and to complete this

study of hospital-acquired infection.

Acknowledgments

I would like to thank my mentor Dr. Rosslyn Byous for her guidance,

encouragement, and support throughout this dissertation process. Additionally, I would

like to thank Dr. Becky Campbell, Mrs. Dixie Robert, R.N., and Dr. Dana Giurgiutiu at

the South Carolina Department of Health and Environmental Control for their guidance,

time, and resources necessary for me to complete this dissertation. I truly appreciate the

academic guidance provided by Dr. Ray Borges and Dr. Julia Moore who helped guide

research and subject matter for this dissertation. Finally, thank you to my children

Whitney, Lauren, Kristin, and Austin who provided me with inspiration and

encouragement through this doctoral process. Lastly, I would like to honor my new

grand-daughter, Makayla. She lights up my life and provides purpose and hope for a new

generation.

Table of Contents

Acknowledgments iv

List of Tables x

List of Figures xii

CHAPTER 1. INTRODUCTION 1

Introduction to the Problem 1

Background, Context and Theoretical Framework For the Problem 4

South Carolina History, Community, and Health Perspective 6

Statement of the Problem 9

Purpose and Significance of the Study 11

Research Design 12

Research Questions and Hypotheses 12

Rationale for the Study 14

Relevance of the Study 15

Significance of the Study 15

Nature of the Study 17

Hypotheses 18

Assumptions and Limitations 18

Definition of Terms 22

Quantitative Definition of Terms 24

Expected Outcomes 25

CHAPTER 2. LITERATURE REVIEW 27 Introduction to the Literature Review 27

Theoretical Orientation of the Research 30 Review of the Research 31 Morbidity, Mortality, and Cost of Prevention 35 Who is at Greatest Risk for Bloodstream Infection 38 Risk Level with Catheter Types 40 Pathogenesis of Catheters 42 The Most Common Hospital Acquired Bloodstream Infections 44 Antibiotic Resistant Microorganisms 45 Research and Metrics that Minimize the Incidence of CLABSI 47 CLABSI and HAI Rates Estimation in Healthcare Organizations 55 Pharmacologic and Therapeutic Measures Prevent and Treat CLABSI and HAI 57 Organizational Measures that Effectively Contain CLABSI and HAI 59 Financial Analysis: Cost and Benefit 60 Direct Medical Cost Associated with CLABSI and HAI 60 Impact of Hospital Acquired Infection on Cost of Care 62 Review of Research Literature and Methodological Literature 63 Review of Research Regarding Acute Care Hospitals 65 Review of Research Regarding Risk Factors in CLABSI 67 Review of Methodological Factors 67 Synthesis of Research Findings 69 Critique of Previous Research 70 Chapter 2 Summary 72

vii

CHAPTER 3. METHODOLOGY 74 Purpose of the Study 74

Problem Statement 77

Research Design 77

Target Populations, Sampling Methods, and Related Procedures 79

Target Populations 79

Procedures 80

Sampling Methods 80

Sample Size 81

Sampling Procedures 81

Recruitment 82

Setting 82

Protection of Participants 82

Operationalization of Variables 83

Instrumentation 85

Research Questions and Hypotheses 85

Data Collection 89

Field Testing and Pilot Testing 89

Data Analysis 90

Limitation of the Research Design 92

Internal Validity 93

External Validity 93

Expected Findings 94

viii

Researcher’s Position Statement 94

Ethical Issues in the Study 95

Conclusion 96

CHAPTER 4. DATA COLLECTION AND ANALYSIS 97

Introduction 97

Description of the Sampled Data 99

Research Methodology and Data Analysis 100

Hypothesis 101

Presentation of Data and Results of the Analysis 104

Hypothesis Data Analysis 139

Summary 141

CHAPTER 5. RESULTS, CONCLUSIONS, AND RECOMMENDATIONS 143

Introduction 143

Summary of Results 144

Demographic Analysis 144

Hospital Population Data 145

Discussion of the Conclusions in Relation to the Literature 156

Recommendations for Further Study 158

Limitations, Delimitations, and Recommendations 159

Recommendations Based on Delimitations 159

Conclusions 160

REFERENCES 162

APPENDIX A. HIDA HOSPITAL INFECTION CONTROL PROCESSES REPORT 178 APPENDIX B. NHSN BLOODSTREAM INFECTION FORM 191

List of Tables

Table 1. Hospital Bed Capacity of 0 to 50 Receiving Central Lines 106

Table 2. Hospital Bed Capacity of 51 to 200 Receiving Central Lines 108

Table 3. Hospital Bed Capacity of 201 to 500 Receiving Central Lines 109

Table 4. Hospital Bed Capacity of 500 or More Receiving Central Lines 110

Table 5. Case Summary Bed Category, Infection Rate per 1000 Clinic Days 111

Table 6. Multivariate Test to Define Hospital Bed Capacity and Infection Rate per 1000 Central Line Days 112

Table 7. Statistical Analysis of Hypothesis 1 Dependent Variables 114

Table 8. Data Correlation of the Infection Rate to Hospital Bed Capacity 116

Table 9. Multivariate Test to Define Correlations between Monitoring Methods, Bed Capacity, and CLABSI 117

Table 10. Comparative Data to Evaluate Effectiveness of Increased Numbers

Of Infection Control Practitioners and Hours of Surveillance Activities 118

Table 11. Data Correlation Between Infection Rate, Bed Capacity, and ICP 119

Table 12. Multivariate Test to Determine ICP and Bed Capacity 120

Table 13. Comparative Data to Evaluate Effects of Surveillance and Therapeutic Activities to CLABSI Event 120

Table 14. Data Correlation to Evaluate Effectiveness of Surveillance and Therapeutic Activities to CLABSI Events 122

Table 15. Multivariate Test to Determine Effectiveness of Surveillance and

Therapeutic Activities to CLABSI Events 124

Table 16. Comparative Data to Evaluate Effects of Surveillance and Therapeutic Activities to CLABSI Events 125

Table 17. Methodological Processes to Evaluate CLABSI, Clinical Interventions and Hospital Bed Capacity 127

Table 18. Methodological Processes to Evaluate Intercepts, Infection Rate, and Bed Capacity and their Effects on Decreased Rates of Central Line Infection 128

Table 19. Statistical Analysis Between CLABSI and Hand Hygiene Measures 129

Table 20. Correlation Between Rates, Surveillance, and Review 131

Table 21. Correlation Between Infection Rate, and Hand Hygiene Evaluation 133

Table 22. Statistical Analysis Between CLABSI, Rates, and Clinical Intervention 135

Table 23. Correlation Between CLABSI Rates and Clinical Interventions 138

xii

List of Figures

Figure 1. Statistical Decision Tree™ 113

CHAPTER 1. INTRODUCTION

Introduction to the Problem

Healthcare organizations desire information to ameliorate the negative aspects of

hospital-acquired infections (HAI) in central line associated bloodstream infections

(CLABSI). Hospital-acquired infections in the United States account for approximately

1.7 million infections annually with an associated death count of 98, 987 (Klevins et al.,

2007). Approximately 30,665 of these deaths are bloodstream associated. This research

defines what surveillance measures can be utilized to decrease the incidence and

prevalence of CLABSI in acute care hospitals in South Carolina. The research provides

healthcare executives and clinicians in acute care facilities information to aid their

attempts to decrease CLABSI and to improve quality of care. CLABSI rates have

decreased significantly over the past ten years in some, but not all critical care

environments (Burton, Edwards, Horan, & Fridkin, 2008). Further study was needed to

determine what efforts make the greatest impact on the decrease of CLABSI. In critical

care units, CLABSI rates dropped from 7.8 per 1,000 to 3.7 per 1,000 central line days in

the span of time from 1997 to 2007 (Burton et al., 2008). Surgical critical care units saw a

decrease from 5.4 per 1,000 to 3.1 per 1,000 central line days in the same time span.

Medical/surgical units within a teaching affiliation dropped from 6.0 per 1,000 to 2.6 per

1,000 central line days and those without a teaching focus decreased from 4.1 per 1,000

to 1.9 per 1,000 central line days. A CLABSI baseline prevention practice assessment

tool was used to obtain information for these reports to help states establish HAI

prevention collaborative partnerships. This tool has 30 questions that were administered

to participating hospitals for the purposes of establishing baseline information that helped

measure change in therapeutic and epidemiologic processes (Centers for Disease Control

(CDC), 2009a).

HAIs in hospitals are the primary cause of morbidity and mortality in the United

States (Klevins et al., 2007). This degree of morbidity and mortality comes with

increased expense to the U.S. healthcare system at a time that the cost of healthcare is

continually rising and exceeding the general rate of inflation (Griffith & White, 2007).

Federal and State governments are experiencing an erosion of available funds and

employers found that increased insurance premiums reduce their profits. Hospitals

receive reimbursement of only 1% to 5% of the actual cost associated with treatment of

hospital-acquired infection (Chen, Chou, & Chou, 2005). The direct cost of medical care

in the U.S. ranges from $28.4 to $33.8 billion in 2007 dollars (Scott, 2009). The payer

system drives healthcare organizations’ desire to curtail the cost of care. Healthcare costs

reflect the consumption of economic resources in the delivery of healthcare (Shi & Singh,

2008). By 2015, healthcare expenditures are presumed to account for 20% of the Gross

Domestic Product (GDP). Researchers suggest that increased prevalence of CLABSI in

hospitals contributes to increased healthcare cost (Tarricone, Torbica, Franzetti, &

Rosenthal, 2010). Nosocomial or hospital-acquired infections are associated with

increased length of stay (LOS), but for CLABSI it was unknown if these infections are

caused by clinical deterioration of the patient due to nosocomial contamination, or if a

hospital-acquired infection resulted from the prolonged hospitalization (Tarricone et al.,

2010).

HAI is a common life-threatening complication observed in the critically ill.

Complications exist in hospitals for those requiring central line catheterization and other

invasive therapies used as an adjunct to care. These complications increase the length of

hospitalization for those with HAI. The average length of stay (ALOS) in hospitals is a

mathematical computation used for health planning purposes; ALOS represents how long

a patient is confined to the hospital (Shi & Singh, 2008).

Historically, infections develop while receiving invasive medical treatments in

hospitals, nursing homes, outpatient surgery centers, and dialysis clinics. HAI is a major

public health problem in the United States. These infections are healthcare-associated or

acquired in the hospital environment. Infection may occur from routine care, surgery, or

from complications associated with medical devices such as ventilators, urinary catheters,

and indwelling invasive catheters or lines. Nosocomial infections may often be a side

effect of the overuse of antibiotics (Jarvis et al., 2009). The U.S. Centers for Disease

Control and Prevention (CDC, 2006) estimates that 1.7 million nosocomial infections

occur yearly in U.S. healthcare facilities. Hospital-acquired infections result in

approximately 99,000 deaths and cost $20 billion in additional healthcare costs each year

(Richards, Brennan, & Straub, 2009). Reports indicate that 250,000 of these infections

occur yearly in the United States with an estimated death toll of 30,000 to 62,000 (South

Carolina Hospital Association, 2010).

In this research, background information regarding CLABSI infection was

described when risk factors were present. Research indicates that there are independent

risk factors for CLABSI. Factors associated with increased risks include prolonged

hospitalization before catheter placement, prolonged duration of catheter placement,

increased microbial colonization at the catheter insertion site, increased microbial

colonization of the catheter hub, internal jugular cannulation, neutropenia, birth at an

early gestational age, the requirement of total parenteral nutrition (TPN), and receipt of

poor or substandard catheter care (Marschall et al., 2008). The susceptibility of critically

or seriously ill patients to become colonized by these organisms inhibits their ability to

ward off infection, which often leads to infection that is resistant to antimicrobial therapy

(Burton et al., 2009; Johanson & Dever, 2003).

Background, Context, and Theoretical Framework for the Problem

The provision of effective surveillance activities and identification of best

practices to minimize the effect of CLABSI and HAI are problematic to healthcare

organizations and to physicians providing their care. Those who have other health risk or

co-morbid conditions are considered vulnerable and are at greater risk. Since hospital

stays are shorter, HAI rates per 1000 patients have actually increased (Duffy, 2002).

Conversely, research finds that long stays in intensive care units are associated with an

increase in hospital-acquired infection and increased cost of care (McGee et al., 2006).

The review of national databases and benchmarking techniques provide evidence that

clinical management influences hospital-acquired infections thereby affecting provision

of care and case management concerns (McGee et al., 2006). An additional impact is the

type of hospital that provided care for these patients. Studies conducted by Khan et al.

(2006) acknowledge that an increase in hospital volume was observed in a higher volume

hospital and is associated with improved survival rates. With adjusted in-house mortality

ratios, patients in high bed capacity hospitals have a 25.5% mortality rate as opposed to

patients in smaller bed capacity hospitals with rates nearing 34%.

Fundamental to the development of critical care techniques and healthcare

management processes was an understanding of the importance and significance of early

intervention therapies, advanced equipment, excellent clinical skills provided by

practitioners that provide care for large numbers on a routine basis, and the management

acumen provided by executives in hospitals. The conceptual basis for the study was

developed from clinical data from hospitals that indicate that healthcare facilities with

higher bed capacity provide care differently than healthcare organizations that do not

have the opportunity to provide advanced techniques on a routine basis as often seen in

smaller bed capacity hospitals (Liu, Zingmond, & McGory, 2006). HAIs are associated

with prolonged hospitalizations when invasive procedures are used in patient care

(Bootsma et al., 2002). The clinical concept of care at larger bed capacity hospitals

versus smaller bed capacity hospitals provided a probabilistic framework for empirically

evaluating the effectiveness of medical strategies in these two hospital groupings. This

provided an analytical study of the effects of errors in surveillance and prevention

strategies; since large capacity hospitals typically use more assets to positively affect

outcomes. Clinical and management research from hospitals that have CLABSI was

advanced by analyzing data from various hospitals with differing bed capacities through

assessment of reported health data to the State Department of Health and Environmental

Control’s Division of Acute Disease Epidemiology. By researching this data and

analyzing clinical concepts and surveillance techniques used by South Carolina’s

hospitals, a determination of best practices was pinpointed so that all facilities regardless

of size and capability can provide excellence in care as described in literature and

reported in this research. The research determined if existing clinical data satisfied

deficits in knowledge of CLABSI rates and surveillance, prevention, and treatment

methodologies, and if this information extends the opportunities to further study types of

facilities other than acute care hospitals.

South Carolina History, Community, and Health Perspective

South Carolina, the Palmetto State is ranked 40th in land mass in the United States

and encompasses 31,113 square miles. The state has over 187 miles of coastline that

borders the Atlantic Ocean on the east and the Blueridge Mountains of North Carolina

and Georgia and the 238 mile long Savannah River on the western side of the state (SC

Department of Parks and Tourism, 2010).

Travel and tourism contributes approximately $18.4 billion yearly to the economy

of South Carolina (US Travel Association, 2010). This industry provides ten percent of

all jobs in the state and accounts for $1.2 billion in state and local revenue. South

Carolina’s citizens contributed $9.9 billion to the state in 2008 in travel and tourism.

South Carolina is a strong agrarian state with strong revenues being generated in

agriculture, livestock, manufacturing, mining, fishing, and services (Bureau of Labor

Statistics, 2010; Netstate, 2009). South Carolina enjoys a Gross State Domestic Product

(GSP) of $156,384 (millions in 2008 adjusted dollars) (Kaiser Family Foundation, 2010).

South Carolina has approximately 4.5 million inhabits (U.S. Census Bureau,

2010). The majority of the state’s population (67%) lives in metropolitan areas of the

state while 33% live in rural or suburban areas (Kaiser Family Foundation, 2010). The

2009 estimates identify 6.8% of the population is under age five, 23.7% are under age 18,

and 13.7% are persons 65 years of age and older. Whites account for 68.9% of the

population followed by blacks at 28.2%, then Hispanics at 4.5%. Seventy-six percent of

the state’s inhabitants are high school graduates and 20.4% have Bachelor’s degrees or

higher. Home ownership accounts for 72.2% with a median home value of $94,900. The

Census Bureau indicated that the median household income in 2008 was $44,695 and

persons below poverty level were 15.7% (U.S. Census Bureau, 2010). The

unemployment rate reported at 11% for South Carolina down from 12.1% in August

2009. Of those, 36.3% of persons employed in South Carolina consider themselves as

white-collar workers and 63.7% are blue-collar workers. Those considered permanent

workers account for 95.2% of the population and temporary workers at 4.8%. The total

number of workers in the state represented by unions accounted for 5.4% (Kaiser Family

Foundation, 2010). Although the Kaiser Family Foundation (2010) reported a higher

percentage of high school graduates America’s Health Rankings (2009) report that the

state has a lower high school graduation rate with 61.0% of incoming ninth graders who

graduate within four years as problematic. The state recognizes that a high violent crime

rate at 730 offenses per 100,000 is concerning.

South Carolina’s statistics show that 16.1% of the state’s population is without

insurance with 14.1% being uninsured children. Medicaid enrollment accounts for 20%

of those in the state and Medicare enrollment at 17%. Health status indicators indicated

that South Carolina has an infant mortality rate of 9.0 % and children who are considered

obese or overweight are reported at 33.7%. Approximately 14% of the states adults are

classified as having disabilities (Kaiser Family Foundation, 2010).

The state’s health related strengths include high immunization coverage with 78.8

percent of children ages 19 to 35 months having received all of their immunizations. At

the beginning of this study, public health was moderately funded at $82 per person

annually. Significant improvements were observed in the past decade as noted by the

decreased rate of preventable hospitalizations falling from 76.2 to 70.6 discharges per

1,000 Medicare enrollees. The prevalence of smoking declined from 25.5% to 20.0% of

the population in the past five years. Although in the same five year span of time, the

incidence and prevalence of obesity increased from 24.5% to 30.6% of the population.

Other significant improvements observed in health status indicators found that rates of

deaths from cardiovascular disease decreased from 398.6 to 301.3 deaths per 100,000 in

the state’s population over the past ten years (Americas Health Rankings, 2009).

In South Carolina, health disparities are found with the health risk of obesity.

Prevalence of obesity is 30.6% of the overall population in South Carolina. Obesity is

more prevalent among blacks at 40.1% than non-Hispanic whites at 26.1%. The

incidence and prevalence of diabetes also vary by race and ethnicity in South Carolinians;

13.1% of blacks have diabetes compared to 8.4% of non-Hispanic whites. Consequently,

mortality rates vary in South Carolina. Statistics indicate that 1,071 deaths per 100,000

occur in the black population with diabetes as compared to whites who experience 848

deaths per 100,000 population (America’s Health Ranking, 2009).

In their Access Health SC document, the South Carolina Hospital Association

indicated that one of every six South Carolinians is without healthcare insurance. The

lack of healthcare coverage often forces the sick to postpone medical treatment or

promotes primary care to be delivered in the emergency room. In 2007, South Carolina

provided $1.3 billion in unpaid healthcare services (SCHA, 2009).

Other health concerns exist such as the state’s increased infant mortality rate at

8.9 deaths per 1,000 live births as well as a higher than the acceptable premature death

rate with 9,549 years of potential life lost before age 75 per 100,000 population as

observed in the state’s statistical records (America’s Health Rankings, 2009). There are

identified significant shortfalls in health profession indicators (Trust for America’s

Health, 2010). In the 2009 statistical summary, South Carolina ranked 28 in primary care

providers, 33 in mental health professionals, 26 in dental care and 22nd regarding nursing

shortages.

Statement of the Problem

Bloodstream infections are clinically linked with the use of central lines. These

infections occur commonly and are acknowledged as a potentially preventable source of

morbidity in the hospital setting. The purpose of this study was to determine methods

used to decrease bloodstream infections that were associated with the use of central lines.

Infection control practitioners desire to implement central line bundle processes

that may result in a significant decrease in line-associated bloodstream infections.

Percutaneous catheterization is a frequently used procedure to access the central venous

circulation. The use of central line catheterization provides adjunctive support when a

need exists to monitor cardiovascular pressures, obtain blood samples, or to create an

access for medications (CDC, 2009b). A central line is an intravascular catheter that lies

at or near the heart, or within one of the great vessels. Great vessels are identified as the

aorta, pulmonary artery, superior vena cava, inferior vena cava, brachio-cephalic veins,

internal jugular veins, subclavian veins, external iliac veins, common iliac veins,

common femoral veins in adult and neonates, and umbilical artery, or vein in neonates

only (CDC, 2010).

Morbidity is an issue; however, financial consequences are acknowledged in

healthcare institutions when the patient's length of stay is prolonged after developing a

hospital-acquired infection while receiving treatment with a central line. Healthcare

executives and medical practitioners alike recognize the importance of intervention to

ameliorate the impact of financial losses when the length of stay is prolonged following

the development of CLABSI (CDC, 2010). The research evaluated surveillance and

practice patterns in higher bed capacity hospitals in comparison to fewer bed capacity

hospitals in order to determine effective clinical methods that create a positive healthcare

outcome for those who develop HAI or CLABSI.

CLABSI is defined as a hospital-acquired bloodstream infection. CLABSI often

appears to be more serious since the patient's defense mechanisms against infection may

be impaired during hospitalization. Many disease-causing organisms live within a

hospital environment and are often more dangerous than those in the community. Rates

of septic shock and increased mortality from hospital-acquired infections compared to

those considered community-acquired are not increased by impaired innate immunity;

instead, proinflammatory factors are recognized as contributors to poor patient outcomes.

Specifically, it was determined that pro-inflammatory mediators C3a, IL-6, and

procalcitonin are contributing factors to increased rates of acquired infection, causing a

poor immune response (Groeneveld & Hack, 2008). CLABSI is a primary bloodstream

infection (BSI) that is confirmed by diagnostic culture and is not secondary to an

infection from another body site (CDC, 2010).

Purpose and Significance of the Study

The purpose of this study was to determine methods utilized in hospitals that

positively and negatively affect care and clinical outcomes. The study evaluated hospital

data provided in aggregate form comprised of CLABSI information from those receiving

care in high bed capacity acute care hospitals versus fewer bed capacity acute care

hospitals in South Carolina. The research was proposed so that healthcare executives and

physicians might have a better understanding of variances in duration of hospitalization

in patients with HAI or CLABSI and the resources required to provide active surveillance

practices and therapeutic regimes for care. The purpose of the study was to explore and

examine surveillance and clinical considerations of care in four hospital groups. These

groups are hospitals with bed capacities of 50 or less, 51 to 200, 201 to 500, and 501 or

more. Research indicated that healthcare organizations with larger bed capacities have

greater resources and integrative policies allocated to infection control programs. In the

course of this study, research and analysis of various clinical and epidemiological

processes were used to identify and to minimize the effect of CLABSI. From this data

analysis, healthcare executives will better understand financial, staffing and policy

considerations that meet patient care requirements. This study analyzed data from

records submitted to the South Carolina Department of Health and Environmental

Control’s Division of Acute Disease Epidemiology (SCDHEC DADE) by acute care

hospitals in South Carolina with CLABSI.

Research Design

In the study, quantitative research was conducted with the goal of determining the

correlation of records from acute care hospitals with varying bed capacities (independent

variables) and the incidence of hospital-acquired infection (dependent variables) in the

population observed who develop central line associated bloodstream infections

(Creswell, 2009). The quantitative research design was descriptive and used a

correlational approach to conduct the study. To accurately estimate the relationship

between variables, a descriptive (correlational) study usually needs a larger sample of

subjects. In this research, records from all hospitals of varying size were evaluated from

January 2008 through December 2008.

Research Questions and Hypotheses

In the correlational design research, a determination of the following facts

occurred. Is there a relationship between rates of CLABSI reported in hospitals and the

four hospital groups with capacities of 50 or less, 51 to 200, 201 to 500, and 501 beds or

more? Is there a significant difference between responsive measures used by hospitals to

minimize the effect of CLABSI when two or more infection control practitioners are

employed in a full-time capacity? Is there a correlation between monitoring and

compliance methods and incidence rates of CLABSI infection? Is there a higher rate of

CLABSI observed when direct observation is conducted during catheter insertion? Is the

CLABSI rate lower when staff members are authorized to stop invasive procedures when

they observe protocols not being followed? Does the use of a preestablished procedure

checklist decrease the rate of CLABSI in acute care hospitals in South Carolina? Does it

make a difference if acute care hospitals with larger bed capacities have well defined

syndromic surveillance programs that are automated with their laboratories to provide

care? Does the number of person-hours used to identify and observe technique, evaluate

insertion site care, and to perform surveillance of hand washing technique decrease the

number of infections? Does a relationship exist between an increased rate of CLABSI in

acute care hospitals and integration of prevention processes in facility protocols? Is there

a positive correlation of a daily review of line necessity and prompt removal of

unnecessary lines on the incidence and prevalence of CLABSI? Does monitoring of

optimal catheter site selection minimize the incidence of CLABSI infection? Does the

use of chlorhexidine skin antiseptics during site preparation and care lower incidence of

CLABSI infection? Do acute care hospitals that use maximal barrier precautions upon

insertion observe a lower incidence of CLABSI infection? Is there a CLABSI rate

increase or decrease when hygiene monitoring is conducted more frequently? Are higher

CLABSI rates observed when the total number of hospital wide monthly observations is

increased?

The research determined if there was a significant difference in incidence and

prevalence rates of CLABSI as observed in submitted data files from various bed

capacity hospitals in South Carolina. The correlation of key surveillance and treatment

factors were used to determine efficacy of care.

Long stays in the critical care unit are associated with high costs and burdens on

participants and their families, insurers, and the society at large. Although factors that

affect the length of stay and outcomes of care in the intensive care unit were studied

extensively in the literature, the conclusions reached have not been reviewed to determine

whether they reveal an organizational pattern that might be of practical use in reducing

the length of stay in the critical care unit. The research consensus is that length of stay in

the intensive care unit is increased by several medical, social, psychological, and

institutional factors. HAI and CLABSI may be proven one of those causative factors

(McGee et al., 2006).

Rationale for the Study

This research was needed because healthcare executives and physicians who

practice medicine primarily in rural communities need reliable information regarding

how surveillance and therapeutic processes can affect HAI and CLABSI rates.

Institutions of various bed capacities provide sound medical care based upon available

resources. This information assist clinicians in better clinical management strategies,

improved selection of antibiotics, improved aseptic technique, and even better selection

criteria for procedures being performed on the patient. The healthcare executive may

find opportunities for improved diagnostic coding techniques, improved utilization of

personnel, or improved negotiations with insurance companies. The research identified a

causal relationship between care techniques for those with CLABSI and a hospital’s

preventative and surveillance measures; moreover, successful strategies evaluated in

hospitals that have high bed capacities as compared to those healthcare facilities that have

fewer beds.

The incidence and prevalence of CLABSI has decreased by an estimated 8-10%

per year (CDC, 2010) with an average bloodstream infection rate of 2.71 per 1000 patient

days (Klevins et al., 2007). The mortality percentages are significantly higher with

increased length of stay, and increased cost of care results. Research indicated that

healthcare associated infections are one of ten leading causes of death in the United

States. These life-threatening infections account for healthcare cost that is approximately

$28 to $33 billion dollars yearly (Herzer & Seshamani, 2010). Further research was

needed to determine the mechanism of the relationship between hospital capacity and

outcome among those with a critical illness (Khan et al., 2006).

Relevance of the Study

The relevance of the study is significant to the PhD program in Public Leadership

and to the Health Services Administration specialization at Capella University. New

knowledge aids the future healthcare executive who reads this study in management

methodologies and will assist clinicians practicing critical care and utilize central line

catheterization in their clinical practice. Creating the right atmosphere decreases hospital

length of stay, decreases the chance of developing hospital-acquired infection, improves

clinical care, and effectively decreases cost within the hospital environment.

Significance of the Study

Research on the impact of CLABSI and the clinical efficacy of surveillance

techniques and preventative measures in large bed capacity hospitals versus lower

volume acute care institutions is minimal in South Carolina. This study bridges the gap

in knowledge between these hospital environments and provides substantive data that

offers clinical and managerial information that directs physicians and executives to better

clinical and operational management paradigms. Knowledge provided in this study

contributes to critical care medicine and healthcare management by providing sound

methodologies based upon statistics from data of healthcare organizations in South

Carolina. Clinical research developed the scientific foundation for medical practice, and

the effect of research on medical practice is observed now the research has been

completed. Research findings influence both medical practice and administrative

protocols, and the practice of research, shapes the practice of medicine and business.

High volume healthcare organizations that care for large volumes of patients are

associated with improved survival rates, especially in those with co-morbidities since

they provide care to high volumes of patients consistently and are well practiced in their

care (Khan et al., 2006). The research bridges any debate found in the literature by

providing clinical and financial data measuring the variances between the specific

populations of patients suggested by the record files in the research. This research

prompts additional research from other geographic regions of the United States, as well

as study of those with co-morbid conditions or emerging practice trends in critical care.

The research contributed to the body of knowledge in infection control techniques

and bridged the gap in medical and management literature by providing salient concepts

in clinical, academic, and managed care arenas. This research might encourage other

academicians, clinicians, and healthcare executives to continue additional research in the

areas of nosocomial infection, management of care, cost, and identification of best

environments of care for these complex concerns.

Nature of the Study

This research was conducted utilizing a quantitative methodology in a

correlational study design. Quantitative research permitted a better understanding of the

prevalence of characteristics within the population and an opportunity to evaluate

accurately and reliably components of this population by using statistical analysis.

Quantitative research is used when the researcher desires to profile a group of people or

database with shared characteristics (demographics). By using statistical techniques such

as correlation, regression, cluster analysis, or factor analysis, quantitative research is

implemented as method of evaluating objective theories by examining the relationship

among variables in the study (Creswell, 2009).

The goal of correlational research was to determine the relationship of existing

differences among hospital groups. Correlational research involved collecting data on

two or more variables in these groups. This type of research also employed the collection

of data on independent variables for two or more groups (Leedy & Ormrod, 2005).

Correlational research evaluated relationships between the variables topically without

probing deeper into the data for causation. Data correlations exist when a variable

increases or decreases and a second variable either increases or decreases in a predictable

manner (Leedy & Ormrod, 2005). Correlational research is defined as non-experimental

research when the manipulation of variables does not exist. Variables may be predictive

in a correlational study, but correlation does not prove causation (Bordens & Abbott,

2008).

This research method used ex post facto; or after the fact data, as conducted in

causal-comparative research (Leedy and Ormrod, 2005). The data were collected

through mandatory submission from acute care hospitals to the South Carolina

Department of Health and Environmental Control (SCDHEC) for purposes other than the

research analysis.

Hypotheses

The hypotheses of the research on CLABSI infection rate at acute care hospitals

with varying bed capacities was focused on surveillance interventions used by South

Carolina’s hospitals and was measured against results of CLABSI rates indicated by each

hospital. In the correlational design research, a determination of the following facts

assured that the research hypotheses are supported.

Hypothesis 1. There is a significant variance in the rates of CLABSI reported in

hospitals with capacities of 50 or less, 51 to 200, 201 to 500, and 501 beds and more.

Hypothesis 2. There a correlation between monitoring and compliance methods

and incidence rates of CLABSI infection.

Hypothesis 3. A relationship exists between an increased rate of CLABSI in acute

care hospitals and integration of prevention processes and strategies when included in

their facilities’ protocols.

Assumptions and Limitations

Assumptions

An assumption underlying the study was a basic comprehension by healthcare

professionals that insert or provide care of central lines and knowledge of infection

control practices and an understanding of the role that pathogens play in the seriously or

critically ill patients as well as how those with central lines may become infected. It was

expected that all records in the study be from patients who had a central line inserted at

time of their CLABSI. It was assumed that all records were reported from one of the 63

acute care hospitals in South Carolina from January 2008 through December 2008.

Another assumption was that not all participants that developed CLABSI were infected at

the time of admission. It was assumed that all hospitals that submit data complied with all

rules and regulations and provided accurate information to SCDHEC when submitting

the Hospital Infection Disclosure Act (HIDA) reports.

Methodological assumptions acknowledge that historical data was collected over a

span of time that was defined by the researcher. This span of time is from January 2008

through December 2008. The year 2008 was the initial year of data collection by South

Carolina DHEC. There was no manipulation of variables as observed in experimental

designs. Assumptions include assignment to the study based upon preexisting

characteristics. It was assumed that there is no intervention, manipulation, or random

assignment.

Theoretical assumptions indicate that raw data was provided to the researcher by

the SCDHEC and has not been altered. Assumptions are conditions that are taken for

granted as being true when data collection occurs (Leedy & Ormrod, 2010). Assumptions

exist when there is a potential correlation in the data collected and bed capacity of the

reporting hospital.

Topic specific assumptions occurred when specific information provided for

research was desired to improve the quality of care, reduce nosocomial infection, identify

best practices for patient care in all acute care hospitals, and improve collection of

adequate surveillance information to assure that best medical care was provided.

Assumptions related to the sample are that all raw data was accurate and reliable

and that the data has been collected ex post facto and in a time span required by the

researcher. It was assumed that data is accurate and no extraneous factors would make

the information null and void.

Assumptions include the use of quantitative protocols for the development of a

reasonable and practical research topic with a precise set of ideas or concepts.

Quantitative assumptions included a focused literature review, and a refined and clearer

understanding of the research question or problem to be researched and then a well

developed research design for the question. An assumption existed that research was

implemented to create a correlational relationship to predict patterns in the research

population. It was assumed the research sample size in a quantitative approach would be

reasonably large, and that the research sample would be a subset of a larger population of

all patients with the same population characteristics. It was assumed that data analysis

and interpretation was completed after data collection. It was assumed that quantitative

data analysis and interpretation are primarily deductive. It was finally assumed that

research findings will be presented in an appropriate format, and that the data collected

supported the Hypothesis for the research.

Assumptions regarding ethical considerations included preestablished norms for

conduct that identified and promoted acceptable behavior. Another ethical consideration

included the knowledge of appropriate methods, procedures, and perspectives that guide

how information was analyzed. Ethical assumptions included promoting the goals of

research under the concepts of knowledge and truth, as well as avoiding error in the

research process. Since research involved cooperation in a coordinated effort among

many people and disciplines, ethical standards promoted the values that are essential to

collaborative work. These components included concepts such as trust, accountability,

mutual respect, and fairness. It was also assumed that research performed will not be

stolen or disclosed prematurely and full credit will be awarded appropriately. It was

assumed that ethics guided the researcher’s accountability to the public through policies

on research conduct and misconduct, on conflicts of interest, and on protection of human

subjects.

Limitations

There was a reasonable possibility the study would not provide solutions. The

research did not include data from rehabilitation facilities, nursing homes, or alternate

care centers or other tertiary care sites that may provide central line catheterization as a

modality of care. The research will not include data from organizations outside the State

of South Carolina. The study will not include data collected before January 1, 2008 nor

after December 31, 2008. Limitations also include no data from pediatric or neonatal

populations who required central line catheterization. The research will not include data

that has not been collected through normal data collection processes mandated by

SCDHEC Division of Acute Disease Epidemiology.

Definition of Terms

Acute care facility. An acute care facility is defined as a healthcare center in a

system that has multicasualty capabilities that can appropriately match needed resources

for an injured patient with special needs (Institute of Medicine, 1999).

CDC. The Centers for Disease Control is an agency of the U.S. Department of

Health and Human Services. Headquartered in Atlanta, Georgia, this organization has a

mission focused on preventing and controlling disease and promoting environmental

health and health education (CDC, 2010).

CLABSI. Central line associated bloodstream infections result when a patient gets

a bloodstream infection after having a central line placed and this infection is not related

to any other type of infection (SCDHEC, 2009).

Comorbid conditions. Comorbid is defined as disease that worsens or affects a

primary disease (Valderas et al., 2009).

HAI. Hospital-acquired infections are infections that occur either systemically or

locally and must not be present upon admission to the hospital (SCDHEC, 2009).

ICP. Infection Control Practitioners are professionally trained in the scope of

infection control prevention and monitoring (SCDHEC, 2009).

Immunocompromised. Immunocompromised means that the immune system is

incapable of normal, full reaction to pathogens or tissue damage because of a disease

(Serra et al., 2008).

Morbidity. Morbidity is defined as the state of being ill or diseased (Murphy et al.,

2007).

Mortality. Mortality is defined as being dead (Murphy et al., 2007).

MRSA. Methicillin Resistant Staphylococcus aureus infections are a family or

strain of bacteria that develops resistance to antibiotics used to treat normal

staphyloccocus infections (Coia et al., 2006).

NHSN. National Healthcare Safety Network is a data reporting system used by

hospitals in South Carolina. Hospitals must submit data at least twice yearly. NHSN is a

secure system that is internet based and used for the purposes of reporting and monitoring

data (SCDHEC, 2009).

Nosocomial. Nosocomial is defined as an infection acquired in a healthcare

facility, and it is also known as hospital-acquired infection or HAI (Stone et al., 2007.

Patient. A patient is defined as a person who has an illness or disease and is

receiving treatment or medical care (Sultz & Young, 2006).

Record. A medical record is a document either in paper form or electronically

that represents the clinical care provided to a patient over the course of an illness (Sultz &

Young, 2006). For the purposes of this research, a record did consist of aggregate data in

electronic form that has been de-identified and provided in a composite, spreadsheet

format.

Surveillance. This is the process of identifying and documenting infections within

a hospital environment. The preferred method is active surveillance. Active surveillance

includes prospective and patient based observation of infectious processes performed by

trained professionals (SCDHEC, 2009).

Quantitative Definition of Terms

The distinct relationship between variables in the research question was defined

by clinical and managerial values. The use of quantitative research provided an

opportunity for the researcher to gather information in a manner that allowed for easy

justification by statistical analysis. Quantitative research attempted to remove all

subjectivity from the study by carefully planning and minimizing midstream deviations

(Patten, 2009). The use of correlational research allowed for an in-depth study of the past

to evaluate the linkages and trends in the data, but this research process did not determine

causation (Leedy & Ormrod, 2005). A correlational designed study identifies an

alternative approach that a researcher may use to investigate how a specific independent

variable may affect the dependent variable. The dependent variables in this research were

directly related to the independent variables of varying bed capacities hospitals and

surveillance and therapeutic procedures as expressed in the research question (Leedy &

Ormrod, 2005). In the research, 63 hospitals were be the sources used for record review

to collect the necessary data to complete the study. Since there was more than one

dependent variable, research did use a parametric process or conventional statistical

procedure and utilize Multivariate Analysis of Variance (MANOVA). This process or

analysis was used when more than one dependent variable existed and when the

dependent variables cannot be combined into one variable. This type of analysis was

used when the independent variables have a significant effect on the dependent variables.

MANOVA attempts to identify the interactions between the independent variables and

the relationships among dependent variables (Cronk, 2006). MANOVA was used when

there are several dependent variables and when one or more independent variables

existed (Field, 2009).

Expected Outcomes

The research expected to reveal a correlation in the records between hospitals of

varying bed capacities and their rates of CLABSI. Information obtained from literature

review suggested that a significantly higher incidence of CLABSI was observed in

hospitals of higher volume than those with lower bed capacities. It was expected that

variances existed in hospitals that have two or more Infection Control Practitioners

(ICPs) available in a full-time capacity for surveillance purposes. It was expected that a

correlation existed between monitoring, compliance methods, and incidence rates of

CLABSI infection in the hospitals in South Carolina. It was also expected that a higher

rate of CLABSI was observed when direct observation of insertion and site management

was conducted during catheter insertion by an infection control or other healthcare

practitioner. When staff members were authorized to stop invasive procedures as they

observed protocols not being followed, non-compliance incidence rates of CLABSI are

lower moreover; the use of preestablished procedure checklist decreased the rate of

CLABSI in acute care hospitals in South Carolina.

Acute care hospitals with larger bed capacities that have well-defined syndromic

surveillance programs and automated laboratories were expected to have fewer CLABSI

infections per month as compared to those facilities that do not. Hospitals that have more

full time infection control practitioners on staff who conduct more hours of infection

control activities per month may also record fewer CLABSI infections. Expectations of

this study also indicated that the number of person-hours used to identify, observe

technique, evaluate insertion site care, and perform surveillance of hand washing

technique decreased the number of infectious processes.

Finally, it was expected that the daily review of central line necessity and the

prompt removal of unnecessary central lines minimized the rate of CLABSI in South

Carolina’s hospitals. Monitoring of optimal catheter site selection was thought to

minimize the incidence of CLABSI infection. Acute care hospitals that use chlorhexidine

skin antiseptics during site preparation and care are expected to have a lower incidence of

CLABSI infection. Acute care hospitals that use vigorous hand hygiene monitoring

program and maximal barrier precautions during insertion are expected to observe a

lower incidence of CLABSI infection.

CHAPTER 2. LITERATURE REVIEW

Introduction to the Literature Review

This study seeks to determine if a relationship existed between hospital-acquired

bloodstream infections and the size of hospitals when central lines were used in acute

care hospitals in South Carolina from January 2008 through December 2008. Therapeutic

interventions and preventative measures used by these healthcare organizations must be

examined to determine the effect surveillance and preventative measures had on the

incidence and prevalence of central line associated bloodstream infection (CLABSI).

Approximately half of all critical care patients require a central line (Weinstein, 2009).

This accounts for 15 million central line days per year. As a central line permeates the

integrity of the skin, the patient becomes vulnerable to both bacterial and fungal

infections. These infections often result in CLABSI in these critical care patients.

Approximately 250,000 to 500,000 CLABSI infections result from central lines annually

in the United States and are an important cause of both morbidity and mortality for

patients receiving care (Weinstein, 2009). CLABSI infections cause a significant toll on

human life. In the U.S., CLABSI infection accounts for an estimated 1.7 million

infections annually and 99,000 deaths in 2002 (Elixhauser & Steiner, 2007). Mortality

rates range from 4% to 20% annually in that population of patients. This estimate

accounts for approximately 500 to 4000 patients who die from CLABSI infections in

hospitals in the United States (Weinstein, 2009). Variances within these two studies

encourage researchers to understand the true morbidity and mortality of disease within

their patient population. Prevention of CLABSI infections is a national priority (Yokoe

et al., 2008). A cadre of interested groups including healthcare and professional

organizations, governmental, regulatory agencies and accreditation bodies, patient

advocates, and payers leads initiatives for preventative measures. Organizations such as

the Society for Healthcare Epidemiology (SHEA) and the Infectious Diseases Society of

America Standards and Practices Committee (IDEA) have taken the mantle of leadership

to focus acute care hospitals into prioritizing efforts to minimize the impact of CLABSI

and other hospital-acquired infections (HAI). Since each CLABSI episode costs $3700 to

$29,000, the minimization of CLABSI infections in hospitals is significant (Weinstein,

2009). The morbidity, mortality, and cost of care regarding CLABSI created a zero

tolerance culture in many American hospitals and encouraged clinicians and

administrators alike to find a positive impact on this illness by developing an active

surveillance and therapeutic regime for care. Similar concerns in a collaborative cohort

study determined that hospitals that developed evidence based interventions have found a

sustained reduction up to 66% in their rates of CLABSI in their 18-month study

(Pronovost et al., 2006). Implantation of foreign objects is an indispensable part of all

fields of medicine. These implants are associated with a greater risk of bacterial and

fungal infection. Specific clinical guidelines are effective in reducing the incidence of

CLABSI (Von Eiff et al., 2005). These strategies, both technically and non-technically

based, were described over the course of this research.

According to the South Carolina Department of Health and Environmental

Control (SCDHEC) (2009), hospitals in South Carolina have not compared their data

with other hospitals in the state. Each hospital’s infection rate for CLABSI was compared

with the historical experience in the standard population of all National Healthcare Safety

Network (NHSN) users in the United States. The South Carolina Hospital Association

(SCHA, 2010) indicated that CLABSI is a significant problem to residents of the state.

The development of a CLABSI reduction and prevention project led by the South

Carolina Healthcare Alliance for Infection Prevention (SCHAIP) created an integrative

partnership with the state’s hospitals entitled Every Patient Counts. Their effort STOP

BSI (Stop Bloodstream Infections) focused hospitals on quality improvement strategies

and educational programs to help ameliorate the impact of CLABSI.

The literature review presented a conceptual framework for the study that

included a rationale and a basis for developing a correlational study that addressed the

research questions presented in the study. The framework of the research developed from

the following key sources: (a) pathogenesis of central line associated bloodstream

infections (CLABSI), (b) patient clinical management strategies, (c) prophylaxis, and

measures that ameliorate the impact of infection, and (d) prevention strategies (Memoz et

al., 2007; von Eiff, Jansen, Kohnen, & Becker, 2005). Central line associated

bloodstream infections (CLABSI) were determined by clinical manifestations and

laboratory confirmation of the pathogenesis and the effectiveness that appropriate

antibiotic usage had on the elimination of pathogens.

While not a primary focus of this study, cost of patient care is the financial impact

that results from an increased length of hospital stay and severity of illness, and it may be

an opportunity for future research. The research developed measures that may influence

surveillance and prevention strategies in hospitals in South Carolina. The research

framework was part of the literature review. This review followed an evaluation of

potential research designs and a conclusion to the research. The literature review was

conducted through the search engines at Capella University library, Centers for Disease

Control and Prevention (CDC), Google Scholar, as well as through other online search

engines. Healthcare journals and medical textbooks provided large volumes of

information synthesized for the support of this research.

Theoretical Orientation of the Research

Ajzen and Fishbein first formulated the theory of reasoned action (TRA) and

theory of planned behavior (TBP) in 1980. This theory was formulated and based upon

attitude research from the expectancy value models. The authors formulated the TRA

when attempting to estimate the existing variances between behaviors and attitude. The

theory of planned behavior is a tool that predicts deliberate behavior since that behavior

is planned.

The theory of reasoned action suggests that behavior is determined by intention to

perform a certain behavior, and therefore it affects attitude toward this behavior. The

best indicator of behavior is intention. Intention is acknowledged as a cognitive process

that represents a person's readiness to perform a given behavior (Ajzen & Fishbein,

1980). Intention is defined by these characteristics: a person’s attitude toward the

behavior, and their ability to develop behavioral control over their actions (Kretzer &

Larson, 1998). In this research, the planned behavior is the ability to manipulate actions

through an attitudinal process of hand hygiene, as well as observational characteristics of

practitioners during invasive processes that could minimize the impact of infections and

associated bloodstream infections due to poor aseptic techniques or policies.

Review of the Research

Central lines or central catheters are plastic tubes inserted into a large vein. Veins

used for this procedure are located in the neck, chest, arm, or groin. Central lines are

typically utilized to obtain blood samples, provide fluids or medications or to perform

hemodynamic monitoring. Catheters may be maintained in these sites for several weeks.

Bloodstream infections may develop when bacteria or pathogens travel from the catheter

and enter the blood. Patients who develop a CLABSI may become severely ill with fevers

and chills, or the skin around the catheter insertion site may become reddened and

painful. These central line associated infections are serious, but are often managed

effectively with antibiotics. In order to decrease the risk of continued infection, the

catheter may be removed as a precautionary measure and replaced in another site (CDC,

2009c).

Research indicated that healthcare organizations and practitioners providing care

to the patient supported successful HAI control programs. Healthcare providers assure

that progress is made toward national HAI prevention targets as indicated by the NHSN,

the Society for Healthcare Epidemiology of America (SHEA), Association for

Professionals of Infection Control and Epidemiology Inc. (APIC), and other

organizations that were established to control or monitor infections. Practitioners must

embrace the use and quality of metrics and supporting systems that ameliorate the impact

of CLABSI and HAI. Healthcare organizations must develop priorities and broadly

implement current evidence based prevention strategies within their organization.

Organizations such as NHSN encouraged the development of HAI plans to enhance

prevention strategies. These plans are focused around (a) HAI program development and

implementation of an organized infrastructure, (b) epidemiological surveillance programs

that assist with detection, reporting, and response; (c) strategies that promote prevention

efforts, and (d) planned evaluation and communication skill development (Maine Center

for Disease Control and Prevention, 2010).

CLABSI may be local or systemically based infections, and can be suspected or

laboratory confirmed to meet their operating definition (Turcotte, Dube, & Beauchamp,

2006). Based upon the CDC recommendations, three categories of infections exist: (a)

exit site infection, (b) catheter colonization, and (c) catheter-related bloodstream

infection. CLABSI is the clinical development of bacteria within the bloodstream that

occurs in the absence of an obvious source of infection other than the invasive catheter,

this bacteremia is proven when the same disease causing bacterium is isolated from the

catheter and from blood cultures. From a morbidity perspective, these are the most

serious effects of CLABSI (Turcotte et al., 2006).

Intravascular catheters are linked with bloodstream infection in critical care units

(Marik, 2010). One or more of the following catheter types cause infections: a) non-

tunneled and tunneled central venous catheters (CVC); b) peripherally inserted central

venous catheters (PICC lines); c) arterial lines; d) hemodialysis catheter, tunneled and

non-tunneled; and e) subcutaneous ports (Linenberger, 2006). The pathogenesis of

CLABSI is based on the ability of dangerous microorganisms to adhere to and then

colonize on the catheter. These foreign bodies promote the formation of biofilm on the

surface of the catheter that in turn promotes the growth of bacterial and fungal infections

(von Eiff et al., 2005). Biofilm is identified as structured communities of microbial

organisms whose function is dependent on symbiotic interactions (Hansen, Rainey,

Haagensen, & Molin, 2007). Bacteria use basic survival strategies. Bacteria’s ability to

colonize on surfaces is greater than its ability to grow in a solution. Bacterial cell walls

excrete adhesins that allow them to colonize on many different types of surfaces.

Bacterial cells secrete a polymer comprised of sugar or exopolysaccharides that allow

attachment to the catheter to occur. The cells multiply and form colonies that rapidly

spread over the surface, forming bacterial populations embedded in a gel-like

polysaccharide matrix (Stickler, 2008).

Healthcare providers introduce a variety of therapeutic modalities into the patient

care paradigm to minimize the effect of healthcare associated infections and associated

cost of care and to decrease the length of stay for patients developing HAI while

receiving a central line associated with their medical care (Cohen et al., 2008). The

Healthcare Infection Control Practices Advisory Committee (HICPAC) offers

recommendations or publishes comments associated with aseptic insertion techniques of

these vascular catheters as well as guidance for appropriate maintenance of the devices

after placement. HICPAC is a federal advisory committee that provides guidance and

advice to the Centers for Disease Control and Prevention (CDC) and the Secretary of the

Department of Health and Human Services (HHS). Recommendations focus on the

practice of healthcare infection control and clinical surveillance strategies that prevent

and control HAI in the nation’s healthcare facilities. The committee issues

recommendations for preventing and controlling HAI through guidelines and informal

communications (Centers for Disease Control and Prevention, 2009a; Cohen et al., 2008).

As a means of effective monitoring and management of CLABSI related

infections, the HICPAC was created to provide oversight and guidance to the CDC.

HICPAC is comprised of liaison representatives from multiple professional organizations

and other federal agencies. These organizations and agencies include the Association for

Professionals of Infection Control and Epidemiology Inc. (APIC), the Society for

Healthcare Epidemiology of America (SHEA), the Association of Peri-Operative

Registered Nurses (APRN), the Center for Medicaid and Medicare Services (CMS), the

Food and Drug Administration (FDA); and other agencies deemed necessary. Typically,

these experts are from the fields of infectious diseases, healthcare associated infections,

nursing, surgery, epidemiology, public health, health outcomes, and related areas of

expertise (CDC, 2009c).

The role of HICPAC is to make recommendations to healthcare organizations

regarding prevention strategies by developing and issuing guidelines as resolutions

through formal and informal communication chains. HICPAC serves as a formal conduit

of information within Centers for Disease Control (CDC) and through integrative

relationships with other CDC advisory committees such as the National Center for

Infectious Diseases Board of Scientific Counselors, the Advisory Counsel on Elimination

of Tuberculosis and the Advisory Committee on Immunization Practices (CDC, 2009a).

Data collected in the study improves the quality of care and minimizes the impact

of HAI and CLABSI in South Carolina. Improved quality measures support care through

integrated research that positively affects how care is provided in healthcare facilities in

the state. Intervention strategies as recommended by agencies such as SHEA, HICPAC,

CDC, and others were evaluated and measured against standards of South Carolina’s

hospitals in the study.

Morbidity, Mortality, and Cost of Prevention

Of the thousands of hospital-acquired infections that occur each year in the U.S,

80,000 are catheter associated bloodstream infections (Chittick & Sherertz, 2010). This

is significantly lower than the estimate by Weinstein (2009) that reports of 250,000 to

500,000 annually. Regardless, CLABSI results in considerable morbidity, mortality, and

high human and financial cost. The increased length of hospitalization associated with

hospital-acquired infections is estimated to be “1 to 4 days for urinary tract infections, 7

to 8.2 days for surgical site infections, 7 to 21 days for bloodstream infections, and 6.8 to

30 days for pneumonia” (Chittick & Sherertz, 2010, p. 552). The mortality rate is

significant with hospital-acquired infections, mortalities linked with bloodstream

infections and pneumonia are “23.8% to 50% and 14.8% to 71%, or 16.3% to 35% and

6.8% to 30%, respectively” (Chittick & Sherertz, 2010, p. 552). Research analysis of

eight studies with 2,540 critical care facilities revealed that mortality increased in critical

care patients with catheter related bloodstream infection compared to those without

catheters (Siempos, Kopterides, Tsangaris, Dimopoulou & Armaganidis, 2009).

Therefore, basic infection control programs are considered cost-effective for both the

patient and the healthcare provider.

Mehta et al. (2007) attempted to determine the CLABSI rate, microbiological

profile of existing pathogens, resistance of these microorganisms, length of stay, and

increased mortality rates in twelve critical care units in seven Indian cities. The study

occurred from July 2004 to March 2007 and was comprised of 10,835 patients who were

hospitalized for 52,518 days collectively. These patients developed 476 CLABSI

infections that indicated an overall rate of infection of 4.4% and 9.06 CLABSIs per 1,000

critical care days. Mehta et al. (2007) reported their CLABSI rate as 7.92 per 1,000

catheter days. The most prevalent strains of pathogens observed in this study were

Methicillin Resistant Staphylococcus aureus (MRSA), Enterobacteriaceae, and

Pseudomonas aeruginosa. The average length of stay (LOS) for CLABSI patients was

9.4 days. In a separate study by Lin and Hayden (2010), MRSA and vancomycin-resistent

enterococcus (VRE) were found to be the primary causative microorganism colonizing

the critical care units in their study. Researchers recognized that both pathogens share

similar epidemiologic characteristics. These characteristics encouraged clinicians to

utilize similar surveillance and infection control strategies to minimize the impact of the

outbreak. The study was confirmed by a prospective cohort study where

recommendation included a consideration for a targeted active surveillance program

combined with culturing of infection sites. This was conducted as a means of developing

a highly sensitive and simple prediction tool to identify subpopulations of patients that

were considered high risk (Furuno et al., 2006). In MRSA patients, the chemotherapeutic

options for treatment are limited. A study conducted by Osih et al. (2007) indicated that

even though the impact of rapid and appropriate empirical treatment with antimicrobials

has not been clearly linked with positive patient outcomes, a seven percent decrease in

hospital length of stay could be linked when therapies are instituted. Their research

suggested that appropriate use of antimicrobials might not be critical to patient outcomes,

as other studies have indicated.

Surveillance of HAI, especially in high-risk areas such as critical care units,

encourages infection control practitioners in healthcare organizations to develop and

implement efforts that improve infection control standards and healthcare quality.

Implementation of an active surveillance program has proven to be an effective adjunct in

the prevention of HAI. Considered sentinel work in the field of surveillance, the Study of

the Efficacy of Nosocomial Infections Control, a study by Haley, Quade, Freeman, and

Bennett (1980) improved the efforts of researchers worldwide to develop ideal standards

in the field of epidemiology and surveillance. Increasing evidence existed that HAIs are a

major cause of patient morbidity and mortality in developed countries. Device associated

infections, especially CLABSI pose the greatest threat to critical care patients (Moreno et

al., 2006).

Acinetobacter and Pseudomonas aeruginosa species show an elevated risk of

infection in critical care patients, a need existed for laboratories to provide molecular

epidemiological support to infection control personnel. The use of contact isolation

precautions minimized the effect of the infectious process since both isolated pathogens

appear resistant to al β-lactam and quinolones antibiotics (Paterson, 2007). β-lactamases

emerged as a major source of gram-negative pathogens. They possess multidrug

resistance to a variety of antimicrobials including fluoroquinolones, aminoglycosides,

trimethoprimsulfaethoxazole, and β-lactam and β-lactamase inhibitors. This report

indicated that broad-spectrum β-lactamase resistance among some gram-negative bacteria

was associated with an increased mortality rate as well as an increased length of

hospitalization and hospital cost (Scwaber et al., 2006).

A significant burden exist for healthcare facilities in the United States caused by

MRSA with the increase in hospital-acquired strains of Staphylococcus aureus that

developed into resistant pathogens (Calfee et al., 2008). Even with the report of a

decrease of cases, MRSA related HAIs are associated with morbidity and mortality in the

critical care population (Farley, 2008). Patients who exhibited MRSA have almost twice

the mortality rate, increased hospitalization, and higher median hospital cost than those

having methicillin susceptible Staphylococcus aureus. The data suggested that this

increase in morbidity and mortality is not due to increased virulence of these pathogens,

but due to delays in the initiation of effective antimicrobial therapies or to higher severity

of illness secondary to infection caused by resistant strains (Calfee et al., 2008).

Sound statistical information exists on the number of hospital-acquired infections

that were observed each year in U.S. healthcare organizations and associated cost of care.

Themes presented in research have a direct relationship to the goals of the study and

assists the researcher of the study by laying a firm foundation for the necessity of

additional research on hospital-acquired infection (Jarvis, 1996; Heyland, Cook, &

Griffith, 1999). Research from Jarvis (1996) and Heyland et al. (1999) supported data in

the study by providing information that shows the care and financial impact of hospital-

acquired infection in the healthcare environment.

Who is at Greatest Risk for Bloodstream Infection?

Anyone who is a patient in a hospital is at risk for developing a HAI (CDC,

2007). Guidelines are available for healthcare organizations that provide comprehensive

recommendations for detection and prevention strategies to minimize the risk and impact

of healthcare-associated infections. These documents provide clinicians and healthcare

organizations with practical recommendations when implementing central line–

associated bloodstream infection (CLABSI) prevention efforts.

Patients at the greatest risk for CLABSI are those receiving care in acute care

facilities such as intensive care unit (ICU), surgical wards, and other similar critical care

environments. Primary reasons for increased risks to these patients are frequent insertion

of catheters, and the use of catheters that are exclusively inserted in critical care patients

(arterial catheters, Swan-Ganz, CVP catheters). These catheters are typically and

frequently inserted in emergency circumstances. They are usually accessed frequently

each day, and often left in place for extended periods (Marschall et al., 2008). Although

there are no absolute values on average duration of central line catheters, research

indicates that a study of 1,753 patients indicates that the average duration of

catheterization before infections ensued is (a) femoral, 6 days, range 2-9 days; (b)

internal jugular, 8 days, range 4-13 days; (c) subclavian vein 14 days, range 10-28 days;

(d) radial artery, 6 days, range of 4-8 days; (e) percutaneous (PICC) line, 17 days, range

of 8-32 days (Shannon et al., 2006).

Research indicated an increase in cases of CLABSI in the non-critical care unit

population (Marschall et al., 2008). Recent data indicated that greater numbers of patients

required central lines outside the critical care unit where the risks are greater. An

increased length of hospitalization is another associated factor that increases the

likelihood of hospital-acquired CLABSI.

Research indicated there are independent risk factors for CLABSI. Factors

associated with increased risks include prolonged hospitalization before catheter

placement, prolonged duration of catheter placement, increased microbial colonization at

the catheter insertion site, increased microbial colonization of the catheter hub, internal

jugular cannulation, neutropenia, and birth at an early gestational age. Other indicators

include the requirement of total parenteral nutrition (TPN), and receipt of poor or

substandard catheter care (Marschall et. al., 2008).

Risk Level with Catheter Types

Multiple types of CVC catheters exist. A peripherally inserted central catheter

(PICC) is a CVC inserted into a vein in the groin or arm rather than a vein in the internal

jugular in the neck or subclavian vein in the chest (Scales, 2010). Tunneled catheters are

surgically implanted into veins in the neck or chest and buried under the skin. The distal

end of the catheter is passed through the skin into the vein. On the external portion of the

catheter, a port exists through which medicines can be given or blood samples can be

obtained. Tunneling the catheter under the skin helps keep the catheter securely in place,

and provides a more stable line. An implanted port is similar to a tunneled catheter, but is

secured completely under the skin. Medications may be injected through the skin into the

catheter via the planted port. Implanted ports usually have a small reservoir that can be

refilled in the same way. This reservoir provides for slow and sustained release of

medication into the bloodstream (Ash, 2008; Pittiruti, Hamilton, Biffi, Macfie, &

Pertkiewicz, 2009).

Research suggested the use of tunneled/cuffed CVCs and peripherally inserted

CVCs typically pose lower risk of developing CLABSI related infections than do non-

tunneled catheters (Marik, 2010; Linenberger, 2006; Scales, 2010). Likewise, tunneled

hemodialysis catheters are associated with fewer CLABSI events than non-tunneled lines

for dialysis. Over 70% of patients that have chronic hemodialysis in the United States

have a tunneled catheter. Tunneled catheters allow for higher blood flow rates at

moderate pressure drops without occlusion and with fewer traumas to the vascular system

and typically fewer infections (Ash, 2008). Arterial lines are less likely to result in

infection than uncuffed CVC catheters. Additionally, reports that the incidence of CVC

related CLABSI is 1-5 per 1,000 catheter days. The incidence of infection in arterial lines

is lower and varies from 1-2 per 1,000 catheter days (Marik, 2010). The long-term use of

CVCs causing bacteremia was estimated at 2.8-14 episodes per 1000 catheter days

(Fernandez-Hidalgo et al., 2006). Rosenthal et al. (2006) also supported this data as they

report HAI from invasive vascular devices as the largest threat to patient safety. Focused

surveillance and evaluation of vascular device associated infection rates per 1000 device

days provided a tool for benchmarking with other hospitals of similar size and

composition. In a CDC study, the United States pooled mean rate for CVC related

catheters was 4.0 per 1000 CVC days (CDC, 2010). A similar study found that CLABSI

rates were reported to be as low as 1.2 per 1000 catheter days and as high as 14.7 per

1000 catheter days in critical care units that focus their care on burn patients. When

using PICC lines, the CLABSI incidence rate documented in cohort studies was 1.1 per

1000 catheter days to 2.5 per 1,000 catheter days (Turcotte et al., 2006). Subclavian and

internal jugular catheters have not been studied for their CLABSI rates.

The earlier work of Maki, Kluger, and Crnich (2006) supports Marik’s

assumptions in their analysis of 200 published studies of adult patients that acquired

bloodstream infections with a variety of intravascular devices. Their findings concluded

that peripheral catheters accounted for only 0.1% and midline catheters account for only

0.4%. Higher rates of infection were apparent in patients that received short term, non-

cuffed and non-medicated CVCs (4.4%). Arterial lines implemented for the purposes of

hemodynamic monitoring indicated a rate of 0.8%; PICC lines, 2.4%. Surgically

implanted long-term catheters that were cuffed and tunneled accounted for the highest

percentage of CLABSI at 22.5%. Researchers observed central venous ports at a rate of

3.6% (Maki et al., 2006).

Practitioners who utilize cuff and tunneled double lumen CVCs had the lowest

incidence and prevalence of catheter related bloodstream infection. The data was based

on prospective studies where all intravascular devices were evaluated for the presence of

infection with laboratory confirmation of the pathogen. All intravascular devices pose a

risk of bloodstream infection and can be used as a benchmarking tool (Maki et al., 2006).

Historically, national efforts have been focused on reduction of intravascular related

bloodstream infection when short-term CVCs are used in a critical care environment.

Infection control efforts should be equally applied to all CVC types and uses.

Pathogenesis of Catheters

Collecting blood cultures from patients thought to have CLABSI is a means of

determining the presence of dangerous pathogens in the bloodstreams of patients.

Positive blood cultures can be indicative of colonization in the bloodstream and can

provide a definitive diagnosis thereby arming the practitioner with information that

improves their ability to eradicate specific organisms through therapy. In recent years, the

increased utilization of central venous catheters and other indwelling devices has

complicated medical care since they typically increase the risk of bacteremia (Hall &

Lyman, 2006).

CLABSI resulting from either non-tunneled PICC or CVC line is often caused by

colonization of pathogens on the outer catheter. This is thought to occur when the

process of transcutaneous insertion of the catheter results in contamination of the flora of

the skin. Tunneled catheters considered contaminated on the catheter hub are associated

with intralumenal infections and are implicated as the causative route of contamination

(Marik, 2010). Factors thought to increase the risk of CLABSI include duration of

catheter insertion, number of lumens in the catheter, number of stopcocks in the

connected tubing, use of these catheters for transfusion of blood products, and the use of

these vascular lines for parenteral nutrition infusion (Marik, 2010; Pittiruti et al., 2009;

Turcotte et al., 2006).) The use of CVCs for the administration of chemotherapeutic

agents significantly increased the incidence of bacteremia (Fernandez-Hidalgo, et al.,

2006).

Hospital-acquired Gram-negative bacteremia is a leading cause of both morbidity

and mortality in critically ill patients. In a 2005 study, 44 patients with Gram-negative

bloodstream infections had bloodstream infection (BSI). Bloodstream infection

accounted for rates of 6.9 per 1,000 admissions and 11.3 per 10,000 hospital days. Seven

primary pathogens were isolated in this population. CVC patients accounted for 22.2% of

the population. An average of 50.5 days was the average length of stay identified in this

study compared to 6.13 days for all other patients (Sligl, Taylor, & Brindley, 2005).

Increased rates of MRSA and VRE in critical care units indicated that a common

vector source existed in that healthcare environment. Healthcare workers’ hands are

identified as the culprit in the spread of infection. It was determined that physicians and

nurses otherwise referred to as clinicians exacerbate patient-to-patient transmission of

these pathogens (Lin & Hayden, 2010). MRSA is endemic in many hospitals in the

United States. MRSA colonization is seen as a risk factor for clinical infection that is

associated with poor clinical outcomes and high cost for care. A prevalence density

study of hospital-acquired MRSA was conducted in a 12-month study of three hospitals

within a healthcare system. These hospitals were comprised of 850-beds and had 40,000

annual admissions to this system. Upon introduction of universal admission surveillance

and integration of active surveillance programs for MRSA, a large reduction in MRSA

was noted among patients with existing central lines in the hospital and 30 days after

discharge. Upon integration of a polymerase chain reaction (PCR) test following

decolonization, contact isolation patients were confirmed laboratory positive. Laboratory

surveillance found 17.8% of actual MRSA patient days; in the critical care units, 33.3%

were identified with PCR (Robicsek et al., 2008).

Polymerase chain reaction or PCR provides a means for researchers and

laboratorians to produce millions of copies of a Deoxy-ribonucleic acid (DNA) sequence

in a short period of time of usually two hours or less. This laboratory process eliminated

the need to use bacteria for amplifying DNA samples (Smittgen & Livak, 2008).

The Most Common Hospital Acquired Bloodstream Infections

The frequency of HAI varies depending on the location within the body. The

CDC reports that of 1.7 million infections reported by healthcare organizations in the

United States, the most common HAI are urinary tract infections (UTI). UTIs account for

32% of all HAIs in the U.S. patient population. Surgical site infections account for 22%

of HAIs followed by pneumonias for 15% and bloodstream infections for 14% (CDC,

2007).

The Morbidity and Mortality Weekly Review (MMWR) stated the 2002 estimate

for all infections combined does not compare to earlier estimates. HAI rates appear to be

decreasing. A reduction in CLABSI and surgical site infections was observed among

hospitals participating in National Nosocomial Infection Surveillance (NNIS) from about

1992 to 2004. In a 32 hospital consortium in Pennsylvania, CLABSI rates dropped by

68% during a four year period (CDC, 2005).

In the Mehta research, their CLABSI study found that the most common

pathogens are Methicillin Resistant Staphylococcus aureus (MRSA), Enterobacteriaceae,

and Pseudomonas aeruginosa (Mehta et al., 2007). Lin and Hayden (2010) identify the

most common pathogens observed in CLABSI as MRSA and Vancomycin-resistant

enterococcus (VRE). Researchers recognize that both pathogens share similar

epidemiologic characteristics. These characteristics encourage clinicians to utilize similar

surveillance and infection control strategies to minimize the impact of the outbreak.

Antibiotic Resistant Microorganisms

An increase of HAI caused by antibiotic resistant pathogens creates much concern

in the medical environment. New strains of microorganisms resistant to antibiotics

provide an opportunity to evaluate the emerging use of biofilm in device-associated

infections and the observance of increased antibiotic resistance (Pierce, 2005). Two

microorganisms consistently identified as problematic are Pseudomonas aeruginosa and

Candida albicans. These pathogens have shown their ability to form biofilm on most

implantable devices. These microorganisms also have an increased resistance or tolerance

to antibiotics that are associated with biofilm, appear to be pathogenically opportunistic,

and are persistent in hospital environments (Pierce, 2005).

In a 2005 survey of 891 infection control practitioners who were asked about

persistent bacteremia resulting from MRSA, Vancomycin continued to be the first line

choice to eradicate MRSA related bacteremia. When treating patients having an increased

frequency of illness and especially those reaching the upper threshold of clinical safety

with maximum dosages of Vancomycin, most infection control practitioners stated they

would most likely switch to a newer antimicrobial agent for patient support and treatment

(Hageman et al., 2006; Yang et al., 2010).

The frequency of patterns of selected antimicrobial resistance among pathogens

causing CLABSI is reported by hospitals in the National Healthcare Safety Network

(NHSN) (Hidron et al., 2008). In this study, CLABSI, urinary tract infections, ventilator-

associated pneumonia, and surgical site infections data are reported to the NHSN in the

time span of January 2006 to October 2007. Results of susceptibility of antimicrobial

agents of up to three pathogen isolates per HAI by a hospital were evaluated to define

antimicrobial-resistance in the pathogenic isolates. Incidence rates were calculated for

antimicrobial-resistance (Hidron et al., 2008). In this study, 463 hospitals reported one or

more HAIs. These hospitals reported 28,502 HAIs among 25,384 patients. The most

commonly identified pathogens were coagulase-negative staphylococci (15%),

Staphylococcus aureus (15%), Enterococcus species (12%), Candida species (11%),

Escherichia coli (10%), Pseudomonas aeruginosa (8%), Klebsiella pneumoniae (6%),

Enterobacter species (5%), Acinetobacter baumannii (3%), and Klebsiella oxytoca (2%)

(Hidron et al., 2008).

As many as 16% of all HAIs were associated with multidrug-resistant pathogens

including methicillin-resistant S. aureus (8% of HAIs), Vancomycin-resistant

Enterococcus faecium (4%), carbapenem-resistant Pseudomonas aeruginosa (2%),

extended-spectrum cephalosporin-resistant Klebsiella. pneumoniae (1%), extended-

spectrum cephalosporin-resistant Escherichia coli (0.5%), and carbapenem resistant A.

baumannii, Klebsiella pneumoniae, Klebsiella oxytoca, and Escherichia coli (0.5%)

(Hidron et al., 2008).

Research and Metrics that Minimize the Incidences of CLABSI

Research indicated that interventions that include education awareness are

observed as a primary means of reduction in CLABSI. Evidence based medical practices

and interventions are usually successful especially when implemented into a medically

diverse practitioner group such as a physician’s practice, surgical suite, or critical care

unit in a hospital setting. The Agency for Health Research and Quality (AHRQ, 2007)

found that active educational interventions seem most effective at reducing CLABSI rate.

These educational interventions include the use of web-based and video tutorials that

focus the clinician on prevention strategies and how they might more effectively adhere

to those interventions. Their study was based upon two controlled pre and post studies.

These studies used education and checklist to guide healthcare providers in their

adherence to insertion and care practices of the catheter (AHRQ, 2007). Results of this

study were echoed in another conducted by Ranji, Shetty, Posley, Sundaram, Galvin, and

Winston (2007). These researchers determined that active educational interventions in the

healthcare environment surrounding central line insertion and care appear to be effective

in reducing CLABSI rates. Of the sixty-four studies that met all inclusion criteria, 19

were associated with CLABSI prevention. Three studies target prevention of multiple

HAIs. These studies reveal a baseline of HAI rates above the median rates reported by

the Centers for Disease Control and Prevention’s National Nosocomial Infection

Surveillance System (NNIS). These studies were based on two controlled pre and post

investigations; one interrupted time series, and four simple pre and post educational

interventions focused on educating clinicians. These interventions appear to be most

effective at reducing CLABSI rates in this population.

Monitoring of critical care environments has significant impacts on clinical

outcomes. Healthcare facilities employ four primary metrics to monitor CLABSI and

HAI related matters especially when multi drug resistant organisms (MDRO) are

suspected. MDRO are microorganisms that are resistant to one or more therapeutic

classes of antibiotics. Some antibiotic classes appear to be associated with higher rates of

colonization with resistant strains compared to other antibiotic classes (Cohen et al.,

2008). An increased risk for colonization occurs if there is a lack of antibiotic sensitivity

or if resistance of the microorganism to the antibiotic used and high tissue antibiotic

penetration and broad-spectrum activity against desirable bacteria are present. Increased

rates of MRSA infections are seen with antibiotic classifications of cephalosporin,

glycopeptides, and quinolones (Tacconelli et al., 2007).

The use of an epidemiological line listing tracked specific patients who have HAI.

Researchers used an antibiogram to monitor susceptibility patterns in laboratory cultures

isolated from patients (Siegel et al., 2007). These researchers objectively measured the

incidence of hospital-acquired MDRO bacteremia by laboratory confirmation as well as

clinical cultures of blood and catheters to measure the incidence of infection or

colonization.

There is evidence that clinician education, hand hygiene, the utilization of sterile

technique during insertion and dressing change along with appropriate skin cleansing

measures has positive effect on decreasing HAIs. It is also recommended that

appropriate site selection prior to insertion of the catheter, the use of antimicrobial

impregnated catheters, dressing change, antibiotic lock solutions, frequency of catheter

changes, and needleless connection devices to minimize contamination of the system

(Chittick & Sherertz, 2010). The study by Lin and Hayden (2010) supported these

theories of infection control and encouraged adequate staffing levels and the assurance of

proper environmental cleaning as metrics of monitoring to be followed. They

recommended the use of bundled intervention strategies to decrease site infections as

groups of interventions also known as evidence-based medicine are proven clinically to

improve care. These interventions are known to improve care when used independently

and have substantially improved outcomes when used together. Bundled strategies are

hand hygiene, sterile barriers, chlorhexidine use for skin antisepsis, catheter site selection

with subclavian vein access the preferred route for non-tunneled application, and daily

determination of need for continued catheter access and prompt removal when no longer

required (Institute for Healthcare Improvement, 2009). Other interventions include active

surveillance, decolonization, and daily chlorhexidine bathing of all critical care unit

patients (Lin & Hayden, 2010). Active surveillance was defined by the CDC as an

assessment of the population of interest and identification of the greatest risk factors for

the outcomes of interest. Outcomes or processes are identified for surveillance, a

determination of an observation period was implemented, a surveillance methodology

was identified, monitoring of outcomes or processes were implemented, and denominator

data was collected if rates were to be calculated. Analysis of data was then conducted and

a report was issued on the surveillance information (CDC, 2006).

Bleasdale et al. (2007) conducted a study to determine if patients who bathed with

chlorhexidine gluconate versus soap and water on a daily basis had lower incidences of

bloodstream infections. Results indicated that patients in the chlorhexidine gluconate

intervention component of the study were significantly less likely to develop bloodstream

infections (4.1 versus 10.4 infections per 1,000 patient days). Clinical findings indicate

that daily bathing with chlorhexidine-impregnated cloths was both simple and effective in

reducing the rate of primary bloodstream infections in medical intensive care units at

Cook County Hospital, Chicago, Illinois.

The utilization of catheters impregnated with chlorhexadine and silver

sulfadiazine (CH-SS) decreased catheter colonization of pathogens in cohort studies

(Jaeger et al., 2005; Turcotte et al., 2006; & Douglas, 2009). They found this process

very helpful in those patients already known to be immune-compromised because of

hematomalogical malignancies, those receiving chemotherapeutic agents or persons who

were compromised by severe and sustained neutropenia. In a randomized, prospective

clinical trial with 106 patients, research compared colonization of catheters and

bloodstream infections using CH-SS coated CVCs and standard uncoated triple lumen

catheters. Fifty-one patients were in the CH-SS group and fifty-five in the control group

receiving the standard catheter. Patients in this study were being actively treated for

multiple myeloma (n=7), non-Hodgkin’s Lymphoma (n=10), and acute leukemia (n=89)

and considered immunocompromised. All patients were categorized according to age,

sex, disease and co-morbidities, catheter insertion site, and duration of diagnosis of

neutropenia (Jaeger et al., 2005).

In this study, CVCs were in place for an average of 14.3 ± 8.2 days in the study

group and 16.6 ± 9.7 days in the control group. Colonization with pathogens was

observed less frequently in the study group or CLABSI group. These researchers found

that patients exhibiting severe neutropenia who had CH-SS coated CVCs had

significantly lower rates of catheter colonization and CLABSI (one vs. eight patients;

p=0.02) (Jaeger et al., 2005).

Recommendations included the use of chlorhexidine-based solutions as a skin-

cleansing agent prior to insertion of CVCs rather than povidone-iodine or alcohol based

preparations as a means to minimize skin contamination on insertion. This recommend-

ation resulted from a random study of 538 CVC inserted catheters with 89.4 % or 481

cultures results evaluated. The use of chlorhexidine solutions was associated with a 50%

reduction in catheter colonization. Their results indicate a 95% confidence interval, an

adjusted relative risk of 2.01. Povidone iodine solution usage identified an adjusted

relative risk of 1.87, 95% confidence interval, and a reduction of 22.2% (Mimoz et al.,

2007).

For those patients who developed CLABSI, recommended treatment included

removal and replacement of the intravascular catheter, as well as the initiation of

systemic antimicrobial therapy as a means of therapeutic care for those with HAI. In

their study of 115 episodes of CLABSI, 98 patients (82%) were identified as having

effective treatment (Fernandez-Hidalgo et al., 2006).

When removal of the catheter is not an option, the use of antimicrobial lock

therapy (ALT) may be determined to be an effective treatment. In a 24-month

retrospective study that provided analysis for ALT recipients, 30 days post treatment

blood cultures were conducted. Twenty-six cases were included in this study for data

analysis. Seventy-seven percent of these cases received either chemotherapeutic agents or

parenteral nutrition by CVC. Most of these cases received vancomycin, daptomycin, or

gentamycin in a heparin lock. The researcher obtained blood cultures confirming

sterilization in 69.2% of cases. Sterilization and CVC retention were noted in 42.3% of

these cases (Bookstaver, Gerrald, & Moran, 2010). In a comparative study of

practitioners who treat using only antibiotics eliminate about one-third of CLABSI. This

inadequate measure often forces the provider to remove the catheter and replace it with

another. Their finding stated that most long-term catheter infections occur within the

lumen of the device and antibiotics cannot effectively treat this site. Eliminating

infection at the biofilm site is an appropriate measure, but requires a high concentration

of an antibiotic to eradicate the bacteria. The utilization of catheter lock solutions often

prevents the removal and replacement of the catheter. These researchers found that using

agents such as minocycline-edetate calcium disodium (MEDTA), taurolidine (2%)-

polyvinylpyrolidine (5%) (T/PVP), and ethanol provided a broad-spectrum solution to

kill bacteria at the site more effectively than traditionally used agents such as

ciprofloxacin, vancomycin, and rifampin (Sherertz et al., 2006). Serial interventions

conducted by healthcare practitioners reduce hospital-acquired MRSA infections. In a

retrospective study of four infection control studies in an 800-bed facility with eight

critical care units, clinical interventions were systematically introduced one at a time over

a nine-year period. An evaluation of clinician compliance was conducted to determine: a)

use of maximum sterile barrier precautions during the insertion of central venous

catheters, b) use of alcohol-based hand cleaners for hand disinfection, c) the inclusion of

a hand hygiene campaign, and d) the implementation of a routine nares (nostril)

surveillance cultures for MRSA in all critical care patients on admission to that unit and

weekly thereafter during their critical care stay. Those patients with positive cultures

were placed on contact isolation precautions. Researchers conducted segmented

regression analyses to evaluate variations in incidence and prevalence rates of MRSA

bacteremia on a monthly basis and compared these results to their predicted values.

Researchers monitored MRSA related bacteremia as a control in this study (Huang et al.,

2006).

Researchers indicated that with surveillance cultures routinely performed and

patients placed on contact isolation precautions, there were substantial reductions in

MRSA bacteremia in both critical care and non-critical care environments. Research

indicated the incidence of MRSA bacteremia decreased by 75% over a 16-month span of

time and by 40% in non-critical care environments. This organization showed a 67%

decrease in the incidence of MRSA bacteremia. Rates of bacteremia secondary to

methicillin susceptible Staphylococcus aureus remained stable during this period. Other

associated interventions did not yield statistically significant changes in MRSA

bacteremia.

Additional research conducted by Burton et al. (2009) supported the study by

Huang et al. (2006) as concerns about rates of methicillin-resistant Staphylococcus

aureus (MRSA) HAI infections prompted healthcare organizations to require mandatory

screening or reporting in efforts to reduce MRSA infections. The objective of this study

was to evaluate trend data in the incidence and prevalence of MRSA related CLABSI in

U.S. critical care units.

Data reported by healthcare organizations to the CDC from 1997-2007 found the

data were analyzed to determine the CLABSI incidence rates for seven adult, pediatric,

and neonatal critical care units. In this research, the definitions of percentage of MRSA

were identified as the proportion of Staphylococcus aureus CLABSIs that were

methicillin resistant. The use of regression modeling estimated the percent changes in

CLABSI metrics over the study evaluation period. This study determined the incidence

rate of CLABSIs per 1000 central line days (Burton et al., 2009).

That same study identified 33,587 CLABSIs in 1,684 critical care units. This

study represents 16,225,498 patient-days of surveillance. From this data, 2,498 reported

CLABSIs or 7.4% were MRSA related and 1,590 or 4.7% were identified as methicillin-

susceptible Staphylococcus aureus (Burton et al., 2009). In this research, it was

determined that of the critical care unit types,

surgical, nonteaching-affiliated medical-surgical, cardiothoracic, and coronary

units experienced increases in MRSA central line–associated bloodstream

infection (BSI) incidence in the 1997-2001 period; however, medical, teaching-

affiliated medical-surgical, and pediatric units experienced no significant

changes. (Burton et al., 2009, p. 727)

The time span of 2001 through 2007 saw MRSA CLABSI incidence declining

significantly in all critical care environments except in pediatric units, where typically

incidence rates remained unchanged.

Decreases in MRSA CLABSI rates “ranged from −51.5% in nonteaching-

affiliated medical-surgical ICUs to −69.2% in surgical ICUs. A confidence rate of 95%

and P_.001 existed for both groups. In all critical care environments, methicillin-

susceptible Staphylococcus aureus CLABSI incidence rates decreased from the years

1997 through 2007. Incidence rates during this period ranged from “−60.1% in surgical

ICUs to −77.7% in medical ICUs. A confidence rate of 95% and P_.001 existed for both

group in the years 1997 through 2007 (Burton et al., 2009).

CLABSI and HAI Rates Estimation in Healthcare Organizations

The Centers for Disease Control states that they estimate rates of HAI by three

specific data sources. The first source the National Nosocomial Infections Surveillance

System (NNIS), which is a voluntary network of hospitals in the United States that report

surveillance data. The second source is the National Hospital Discharge Survey (NHDS).

This source provides an annual survey of characteristics of inpatients that are discharged

from hospitals. Finally, the American Hospital Association (AHA) provides an annual

survey of hospitals and their HAI characteristics (CDC, 2007).

The development of the National Healthcare Safety Network (NHSN) in 2005

established a mechanism for healthcare organizations to voluntarily report healthcare

acquired infection clinical data into a national database. The reason for doing so was to

provide a source to determine the significance and magnitude of HAI; the analysis of

trends in HAI data; comparison of risk-adjusted data useful in quality improvement

activities; and provision of guidance for healthcare organizations to develop surveillance

programs for their facilities. Trend analysis encourages hospitals to recognize patient

safety problems and to integrate rapid interventions and corrective measures (Edwards et

al., 2008). The NHSN report provided summary data that was grouped into three

specific categories (a) device associated data, (b) procedure associated data, and (c)

medication associated data that helps determine antimicrobial susceptibility. Data

reported by the NHSN fall into four components that include “patient safety, healthcare

personnel safety, biovigilance, and research and development” (Edwards et al., 2008, p.

783). By December 2007, there were 923 hospitals enrolled into the NHSN registry, and

646 had submitted monthly reports indicating their desire to follow the prescribed patient

safety modules. Since the inception, a larger number of smaller hospitals are now

reporting data to the NHSN. Mandatory reporting by the state’s department of health was

first observed in New York, South Carolina, and Vermont. This mandatory reporting was

the most likely cause for the disproportionate volume of smaller hospitals reporting since

now many smaller hospitals are compliant with reporting efforts. Now that NHSN was

open to all hospitals and more states are electing to require mandatory reporting,

enrollment numbers have increased and the type of hospitals, including long-term care

and ambulatory surgery centers, is more varied. Not only do many acute care hospitals of

varying size and patient capacity provide data for analysis but now these additional

facility types are providing more diverse group for data analysis.

The Code of Laws of South Carolina, 1976, Chapter 7, Article 20, Title 44

mandates that South Carolina hospitals comply with the Hospital Infections Disclosure

Act (HIDA). Section 44-7-2430(A) (2) states, “Hospitals also shall report completeness

of certain selected infection control processes, as recommended by the advisory

committee and defined by the department, according to accepted standard definitions.”

Goals, objectives, and priorities of a public reporting system must be specific and

data must be monitored and measurable. This level of information assures that data is

credible and ensures accountability. Choosing appropriate measures and patient

populations was essential in obtaining useful data for the public and healthcare

organizations, which use this data to improve the quality of care. The researcher must use

standardized case-finding procedures and data validity checks (McKibben et al., 2005).

Appropriate support for infrastructure, resources, and infection control professionals;

adjustment for underlying infection risk; and production of useful and accessible reports

for stakeholders, with feedback to healthcare providers must be available. Planning and

management of this reporting process was recommended to be evaluated routinely by a

multidisciplinary group composed of public health professionals, healthcare providers,

infection control practitioners, as well as consumers of healthcare services (McKibben et

al., 2005).

Pharmacologic and Therapeutic Measures Prevent and Treat CLABSI and HAI

Healthcare providers make an impact on HAI by adhering to recommended

infection control practices. These practices include adherence to standard, contact,

droplet, and airborne precautions (CDC, 2007). Changes to earlier recommendations for

Standard Precautions include Respiratory Hygiene/Cough Etiquette and safe injection

practices (Siegel et al., 2007). Their recommendation includes the use of a mask when

performing specific high-risk and long duration procedures. After the SARS event of

2003, researchers described how clinicians failed to implement simple source control

measures with patients, visitors, and healthcare personnel. This failure may have

contributed to SARS corona virus transmission. The promotion of recommended

practices is evidence based. It was observed in some clinical procedures that respiratory

droplets might contribute to infectious processes; this necessitates the use of masks in

protracted clinical procedures. Evidence existed that environmental controls (gloves,

gown, and mask) diminish risk associated with the spread of fungal infections in severely

immunocompromised persons (Siegel et al. (2007).

Measures recommended by the CDC provide options for care that may prevent

HAIs and CLABSI in the healthcare settings. These measures include properly choosing

a vein where the safe insertion of the catheter can be performed and where infection risk

is minimal. Another measure includes healthcare providers properly washing their hands

with soap and water or using alcohol-based hand rub before gloving. Healthcare

practitioners should wear barriers to infection such as a mask, cap, sterile gown, and

sterile gloves prior to insertion so that the catheter remains sterile. The use of a sterile

sheet protects the sterile environment and minimizes the risk of contamination of the

sterile field. Proper cleaning of the patient’s skin at the insertion site with an antiseptic

cleanser is required prior to catheter insertion (CDC, 2009b).

After insertion of the catheter, hand hygiene measures are indicated and vital

when using the central line and providing care for the insertion site, bandage changes, or

equipment changes. The healthcare provider must wear gloves and clean the catheter

port with an antiseptic solution before using the catheter to draw blood or give

medications. It is imperative that physicians providing care evaluate on a daily basis the

necessity of continuous catheter placement. Catheter removal should be performed as

soon as it is no longer required (CDC, 2009b; Turcotte et al., 2006).

Infection control recommendations of care indicate that catheter insertion should

take place within the sterile confines of an operating room or similarly clean

environment. Other than PICC placement, bedside insertion should not be performed

except in an emergency. They recommend aseptic technique be maintained by providing

vigorous skin cleansing with a chlorhexidine gluconate 2% in alcohol or aqueous solution

prior to catheter insertion. The use of antibiotic/antimicrobial impregnated catheters with

chlorhexidine or silver sulfadiazine should be considered for appropriate risk groups to

minimize risk of infection (Bishop et al., 2007). Recommendations indicate that

practitioners should not use antibiotics as a prophylaxis for infection and flushing the

system with heparin versus normal saline remains controversial. Recommendations

include routine replacement of central line catheters as a means to reduce infection rates

(Bishop et al., 2007).

Organizational Measures that Effectively Contain CLABSI and HAI

Evidence exist that characteristics such as an aggressive organizational culture of

safety, as well as adequate nursing staff levels and composition positively influence other

healthcare personnel and their adherence to recommended infection control practices

(Siegel et al., 2007). These are seen as important factors in breaking the chain of

pathogen transmission. These concepts led healthcare executives to focus efforts

administratively to support the development and support of aggressive infection control

programs.

Financial Analysis: Cost and Benefit

Bloodstream infections linked to the insertion of intravascular devices are

acknowledged as the most life-threatening type of infection related to invasive medical

devices. Patients that develop CLABSI require increased levels of treatment that create

significant economic loss. Additionally, effects of CLABSI cause extended illness, loss

of personal income, and other substantial intangible costs related to a diminished quality

of life.

The Centers for Disease Control estimate the cost of one CLABSI is $29,156 in

2007 dollars. This accounts for a total cost of $2.68 billion in excess costs annually for

healthcare organizations (Umscheid et al., 2009; CDC, 2005). The MMWR from the

CDC reports costs associated with CLABSI are approximately $25,000 per episode as

indicated in the Pittsburgh Regional Healthcare Initiative.

Direct Medical Cost Associated with CLABSI and HAI

Healthcare costs have increased at an alarming rate. HAIs in healthcare

organizations create significant economic encumbrances on the nation’s healthcare

system. The most comprehensive national estimate of the annual direct medical costs due

to HAIs was published in 1992; the Study on the Efficacy of Nosocomial Infection

Control (SENIC) was conducted in the mid-1970s. Information contained in this

document was based upon the fact that approximately 4.5 HAIs exist for every 100

hospital admissions. Based upon this information, the annual direct cost on the

healthcare system was estimated to be $4.5 billion in 1992 dollars. When economists

adjusted for the rate of inflation, this was approximately $6.65 billion in 2007 dollars.

Research indicated that underlying epidemiology of HAIs in hospitals changed

significantly since the original SENIC study, as did the costs of treating HAI. The cost-

benefit ratio to manage HAI must be considered in the care of patients who develop

hospital-acquired bloodstream infections. Evaluating the economic impact of HAI is

important to the financial status of healthcare organizations and may determine whether

or not specific treatment plans are implemented (Scott, 2009). A causal-comparative

study conducted indicated infections acquired in hospital are likely to affect the duration

of hospitalization. Statistical methods used to estimate the additional length of stay spent

in hospitals because of hospital-acquired infections allow reproduction of the patient mix

in the hospitalized population. This modeling helps identify the timing of events as a

failure when accounting for important covariates; failure to model adequately the timing

of events may lead to biased results (Schulgen, 2000).

When economists apply two separate Consumer Price Index (CPI) adjustments,

the annual direct medical costs of HAI in U.S. healthcare organizations ranges from

$28.4 to $33.8 billion after economic adjustment and an increase from $35.7 billion to

$45 billion after adjustment to 2007 dollars using the CPI for inpatient hospital services.

After adjustment for the range of effectiveness of possible infection control interventions,

the benefits of prevention range from a low of $5.7 to $6.8 billion to a high of $25.0 to

$31.5 billion (Scott, 2009).

This research was founded on the basis that HAIs directly affect the cost of care

in acute care hospitals. Research indicated that hospital length of stay is linked with

varying types of infection (Schulgen et al., 2000). Their causal comparative study

provided a basis for the need to do additional research in the area of CLABSI for this

study. In a study of Argentinean hospitals, researchers found that hospital-acquired

infections accounted for a significantly high volume of morbidity and resulted in an

increase in cost of care due to the additional length of stay (Rosenthal et al., 2005).

Themes presented by these studies have a direct relationship to the goals of the study and

assist in research of the study by laying a firm foundation to determine the cost associated

with hospital-acquired infections. The data collected by the Argentinean study supports

the research by providing information that shows prevention methodologies and financial

impact of hospital-acquired infections in the hospital environment.

Impact of Hospital Acquired Infection on Cost of Care

Economic evaluation has become increasingly important in healthcare and

infection control in order to ameliorate the significant impact on cost of care in an already

financially stressed system. In the United States, hospital-acquired infections are a major

health problem and have a significant financial impact on the cost of patient care. The

increased incidence and prevalence of hospital-acquired illnesses have a great impact on

increased length of stay and hospital costs (Chen et al., 2005). Identifying patients who

are at high risk of developing hospital-acquired illness can help providers and

administrators effectively manage the costs of care (Baker, Lambert, Poulos, & Feldman,

2000). In conclusion, the research will identify patterns in records for those who are at

high risk for the development of CLABSI and therefore will assist healthcare executives

and physicians in determining those who may have expected increased cost of care.

Analysis of data obtained from South Carolina Department of Health and Environmental

Control linked and compared CLABSI data in acute care hospitals and determined if

patients were equally at risk for this type of nosocomial infection based upon the

therapeutic intervention received. Research findings provided information to support the

economic need to improve quality of care in medical, surgical, and mixed medical and

surgical critical care units in the U.S. Their findings provided the financial data that

supported the need for advanced studies in hospital-acquired infection by addressing

appropriate methodologies to care for patients in these acute care facilities (Chen et al,

2005). Financial considerations associated with CLABSI in critical care patients other

non-specific cost associated with increased length of stay, there are also non-inflation

adjusted costs associated with CLABSI that are reported to vary from $3,700 to $29,000

per clinical event (Marschall et al., 2008).

Review of Research Literature and Methodological Literature

Healthcare organizations want data that will assist them in finding solutions to

decrease the high cost of patient care and to ameliorate the financial impact of hospital-

acquired infections. Hospital cost containment, cost reduction, and alternative care

delivery systems continue to be on the minds of healthcare providers, payers, employers,

and policy makers. The increased cost of care was linked to an increased length of stay

when a patient develops CLABSI and other hospital-acquired infections. Utilizing tools

that provide a risk index is an effective way of identifying who may develop hospital-

acquired infections (Taheri, Butz, & Greenfield, 2000).

The Department of Health and Human Services awarded $17 million in 2009 to

diminish cases of HAI. Increased awareness and healthcare improvement initiatives have

little positive effect on curbing HAIs as based upon findings of the Agency for

Healthcare Research and Quality (AHRQ). In fact, AHRQ reported increases in

postoperative urinary catheter and postoperative sepsis rates and no change in the rate of

central line associated bloodstream infections (McKinney, 2010). In South Carolina,

improvements were made to the state’s tracking system, and healthcare providers were

encouraged to acknowledge patterns and implement innovative strategies to minimize the

number of CLABSI in the state’s healthcare facilities. In South Carolina, healthcare

practitioners embraced safety efforts and that team based approaches to CLABSI have

decreased rates by approximately 45% since the program’s inception in 2007 (McKinney,

2010).

Approximately 2 million hospital-acquired infections occur each year in the

United States with considerable morbidity, mortality, and cost (Jarvis, 1996). Schulgen

et al. (2000) developed a causal-comparative study that indicated that infections acquired

in the hospital are likely to affect the duration of hospitalization. This increase in length

of stay resulted from patient complications that may result from patients’ having a high

acuity, an increased use of therapeutic invasive devices, immunocompromised clinical

status, or antimicrobial resistance (DeAngelis, Murthy, Beyersmann, & Harbarth, 2010).

Their recommendations were to implement statistical methods to estimate the additional

length of stay spent in the hospital because of hospital-acquired infections. Their finding

from the causal-comparative study allowed modeling of the patient mix in the

hospitalized population. Modeling may define time variation when event failures occur

and promote prevention strategies for further treatment plans.

Failure to model adequately the timing of events may lead to biased results.

Three approaches were used in the past to estimate the extra hospital stay (a) a

comparison of duration of stay of infected and uninfected patients, (b) matching of

infected and uninfected patients with respect to potentially important determinants of the

length of hospital stay, (c) and matching for time-to-infection in addition to the other

factors (Schulgen et al., 2000). Even though these approaches provide an opportunity

for the evaluation of the patient population mix, inadequate timing of events may

overestimate the effect of hospital-acquired infections. On the other hand, statistical

methods analyze time-to-event data and derived alternative methods to estimate the extra

stay that appropriately accounts for patient mix and timing.

Review of Research Regarding Acute Care Hospitals

Economic evaluation has become increasingly important in healthcare and

infection control (Chen et al., 2005). That study evaluated the impact of hospital-acquired

infections on cost of illness in critical care units. In their retrospective cohort study, an

evaluation of medical, surgical, and mixed medical and surgical critical care units in a

tertiary-care referral medical center were evaluated. Estimates of the cost for patients

who acquired a hospital-acquired infection were computed using a stratified analysis and

regression approach. During the study period, 778 patients were admitted to the critical

care environment. Total costs for patients with and without hospital-acquired infections

were $10,354 and $3,985, respectively for hospitalization in this study. Costs stratified by

infection site and primary diagnosis were also significantly different except for surgical-

site infection. In the study, covariates were adjusted for in the multiple linear regressions.

Hospital-acquired infection increased the total costs by $3,306 per patient per admission.

Their study revealed that each additional day spent in the critical care unit increased the

patient’s daily cost by $353.

Prevention of hospital-acquired infections should reduce direct costs and decrease

the length of stay. In the United States, treatment cost in critical care negatively affects

the gross national product by 1%, approximately $62 billion in 1992 inflation adjusted

dollars (Chen et al., 2005). Researchers usually use the hospital’s perspective for

economic analysis of infection control. Preventing hospital-acquired infections usually

does not reduce indirect costs; however, direct costs can be reduced if the patient does not

acquire a hospital-acquired infection. In the study by Chen et al., (2005) researchers used

matched control techniques and determined that the costs of hospital-acquired infections

are higher than the costs for patients in the control group. This cost had a range from

$422 to $34,508 per episode. The wide range may be due to differences in methods,

populations, and severity of illness.

Research supports the findings that patients who had hospital-acquired infections

in critical care units incurred excess costs (Chen et al., 2005). In this research a final

determination indicated that hospital-acquired infections were significantly associated

with increased costs on crude analysis, stratified analysis of infection site and diagnosis,

and multiple regression analysis. After adjustment for all variables, hospital-acquired

infection significantly increased the total costs of care by $3,306 per patient per visit.

Costs of these infections range from $558 to $593 for urinary tract infection. Surgical site

infections cost on average $2,734, and pneumonia cost on average $4,947. The most

significant cost is for bloodstream infection with a range from $3,061 to $40,000,

Hospitals lose from $583 to $4,886 for each hospital-acquired infection in a

prospective payment system (Heyland et al., 1999). As healthcare executives focus their

efforts on cost containment, increased support for infection control programs should be

provided so that preventable hospital-acquired infections and their associated

expenditures can be minimized.

The total costs for patients with and without hospital-acquired infections were

$10,354 and $3,985, respectively (Chen et al, 2005; Rosenthal et al., 2005). Jarvis (1996)

and Heyland, et al., (1999) in their respective studies echoed many of the same findings

even though the study was performed almost ten years prior. The studies presented in this

proposal support the need to collect more data to evaluate therapeutic processes that

would serve as a mechanism to decrease the financial impact on hospitals in South

Carolina. The financial support for healthcare organizations decreases each year with

severe financial cutbacks and fewer financial resources to meet the ever-increasing

material, operational, and human resource needs observed in hospitals of all sizes.

Therefore, studies need to be conducted that evaluate clinical attributes of care that may

provide financial saving to hospitals. These studies are an important undertaking.

Review of Research Regarding Risk Factors in CLABSI

Review of methodological factors

Healthcare research includes studies that often combine both quantitative and

qualitative methodologies. In this research, the quantitative process was employed.

Research regarding hospital-acquired infection in patients requiring central line catheters

and associated therapeutic treatment modalities was predominantly a quantitatively

researched topic. Quantitative research, which is generally used in the biological

sciences, employed statistical methods to evaluate collection of data (Patten, 2009). In

the study, the data being collected include both CLABSI and HAI data.

The application of research methods and designs to specific health and medical

research problems encountered in human service disciplines is generally of quantitative

design. Study design in this research was correlational. In the quantitative research, the

approach to the study questions was defined by the research questions, hypotheses, and

variables to measure. As the research data was systematically obtained, the researcher

analyzed the data with statistical tools. The data were then analyzed and interpreted, and

the hypotheses that were set forth either are accepted or rejected (Creswell, 2009).

Quantitative research was linear and deductive; this allowed for clearer and more concise

management of data in medical and healthcare environments. Healthcare disciplines rely

on quantitative research to enhance knowledge about the technical world. The

quantitative process used statistical tests to analyze collected data. There were rigid rules

for the use of tests and interpretation of test results. This data management process was

developed over many decades of successful use of the scientific method in health and

medical care. Medical and health researchers collect data that is measurable. These data

were analyzed and rational conclusions were drawn from the interpretation of the

resulting numbers. Researchers do not have a theory at the beginning of the process. The

theory emerged from the data in the statistical analysis; it is grounded in the data

collected and thus called grounded theory (Patten, 2009). In this study, correlational

research was conducted. Correlational research provided an opportunity to determine the

relationships that existed even though full causation may not be fully determined.

Correlational studies do not identify causes or the effects of those causes. Correlational

research tries to find the possible relationships or associations that exist within the data.

Correlational research was described as a means of determining the degree of a

relationship (Patten, 2009). Correlational and causal-comparative research are two types

of non-experimental designs predominantly used when a researcher in interested in

causation or patterns in data.

Synthesis of Research Findings

Research was conducted on central line associated bloodstream infections. There

are studies that inform the reader of risk factors associated with CLABSI and other

hospital-acquired infections so that effects may be lessened to this malady through the

research process. It was determined that healthcare organizations want data that will

assist them in finding solutions to decrease the high cost of patient care and to ameliorate

the financial impact of hospital-acquired infection in those requiring central lines as part

of their care meanwhile improving the quality of provided care. Throughout research,

data were clear that hospital-acquired infection is a common complication in those

requiring central lines as an adjunct to their care and is associated with an increased

volume of morbidity and mortality (Arozullah, Khuri, Henderson, & Daley, 2001).

Utilizing tools that provide a risk index is an effective way of predicting those patients

who may develop hospital-acquired infection. Utilizing projection criteria to identify

who is the most likely candidate is a practical approach to alert caregivers to a serious

and potentially deadly illness. This information of potential risks allows healthcare

providers to become more cognizant of key issues and concerns and allow them to

implement best practices to ameliorate the significance of those risks.

Critique of Previous Research

In the review of research regarding CLABSI and HAI in hospitals, multiple

studies were used to support this component of the research. The study performed by

Taheri et al., (2000) was conducted to evaluate hospital infection control strategies

through the universal metric for CLABSI and HAI. The purpose of their study was to

assess how hospitals integrate epidemiological evaluation and infection control into their

clinical processes.

The study by Schulgen et al. (2000) evaluated the impact of hospital-acquired

infection on duration of hospital stay; this information relied on estimates obtained in

prospective cohort studies. The authors perceived that the statistical methods used to

estimate the extra length of stay were not adequate. A comparison of duration of stay in

infected and non-infected patients was not adequate to estimate the extra hospitalization

time due to hospital-acquired infections. Duration of stay prior to infection was

compensated in part by the bias of the methods used. New model-based approaches were

developed to estimate the excess length of stay. Statistical models based on multivariate

counting processes provide an appropriate framework to analyze the occurrence and

impact of hospital-acquired infections. The methods were developed by using data from a

cohort study on 756 patients admitted to intensive care units at the University Hospital in

Freiburg.

The study conducted by Chen et al. (2005) was an economic evaluation that

studied the impact of hospital-acquired infection on cost of illness and length of stay in

critical care units. The design of the study was a retrospective cohort study conducted in

medical, surgical, and mixed medical and surgical critical care units. Estimates of the cost

and length of stay for patients who developed a hospital-acquired infection were

computed using a stratified analysis and regression approach for the 778 patients that

were admitted to the critical care units. Strengths of this study design included the

adjustment for co-morbid conditions, severity of illness, and diagnoses of studied

patients. This credible study was approved by the Department of Medical Research and

Education and used operating definitions from the Centers for Disease Control. Statistical

analyses were performed by the use of Student’s t-test, the chi square, and the Mann-

Whitney U test. Multiple linear regression analysis was performed on total cost and

length of stay as dependent variables. Researchers used Eigen values and condition

indices for co linearity diagnostics and the Durbin-Watson test for serial correlation of

the residuals.

The study conducted by Jarvis et al. (2009) evaluated the healthcare associated

bloodstream associated infections (HA-BSI). Their study compared HA-BSI in critical

care units at five hospitals located in the United States and Australia using Centers for

Disease Control definitions. HA-BSI rates and prevention practices were compared

throughout the test period when using specialized needle connectors. This is a lesser

study that specifically looks at only one component of the research. Their research was

based upon a causal correlation study that would be used for my study. The rigor of the

study design and the sample size was difficult to assess leading to unsure determination

of the weaknesses or strengths of the reviewed study.

These cited works supported research in the field of hospital-acquired infection

and the effects this illness has on healthcare organizations. All studies were conducted in

a credible manner and based upon a need to understand the effects of bloodstream

infections and the effects these infections have on persons requiring central lines. The

methodological strengths of all studies were clinical in origin and based upon case

analysis and document review in a quantitative design. Many attributes of these studies

are building blocks and provide strength for my study. These studies have adequate

sampling, were conducted with a rigorous design, and with data qualified tools and

research processes. Information obtained from these conducted studies indicates a need

for additional research in the field of hospital-acquired infection and cost of medical care,

therapeutic and surveillance processes.

Chapter 2 Summary

When a patient develops CLABSI, there is an appreciable increase in the

prolongation of hospital stay and increased morbidity. The literature further indicates that

central line associated bloodstream infections are a primary cause of death in critically ill

patients in the United States. Research suggested that infection control programs are

important in the prevention of these illnesses by developing clinical strategies that modify

the risk factors for hospital-acquired infection. Many of these prevention strategies cost

very little but make a significant impact. Examples of these strategies are the use of hand

hygiene, chlorhexidine solution usage, adequate clinical staffing, and/or bundled

interventions that decrease site-specific critical care infections (Pittet, Allegranzi, &

Boyce, 2009).

Research surrounding hospital-acquired infection was based upon the historical

need to improve quality of medical care, and to reduce the cost of medical interventions

by utilizing better surveillance methodologies and employing bundled processes for

improved clinical care outcomes. Limited current data was available; additional studies

conducted with academic research methodologies was needed to assure the integrity of

data that was presented. It appears that a strong correlation in written research exist that

links hospital-acquired infection, catheter pathogenesis, and clinical opportunities for

improvement. The research provided an opportunity to evaluate these core concepts for

the benefit of acute care hospitals in South Carolina, the South Carolina Department of

Health and Environmental Control, and for the South Carolina Hospital Association.

There were many challenges to the study of hospital-acquired infection; data

collection must be conducted in clinical environments with strict guidelines and ethical

considerations in place. Additional research provided conclusions that can be integrated

into clinical practice patterns and best practice concepts. Future research will focus on

existing controversies with plans to improve diagnostic tools and approaches, treatment

plans, and prevention strategies.

CHAPTER 3. METHODOLOGY

Purpose of the Study

The purpose of the study was to determine surveillance processes utilized in

hospitals that affected the incidence and prevalence of central line associated bloodstream

infections (CLABSI). The research focused on variances that existed in hospitals within

South Carolina when a patient develops hospital-acquired infections (HAI) while using

central lines. The study evaluated secondary data containing records reported by hospitals

to the South Carolina Department of Health and Environmental Control’s Division of

Acute Disease Epidemiology (SCDHEC DADE). These records were comprised of de-

identified data from patients that developed CLABSI. The records were from data

submitted by infection control practitioners in 63 acute care hospitals in South Carolina.

The research will assist healthcare executives and physicians to better understand the

variances in care that occurs during hospitalization for patients with CLABSI and patients

who do not have these types of HAI. The purpose of the study was to examine the

therapeutic interventions and epidemiologic surveillance techniques that were employed

in acute care hospitals and their effect on CLABSI rates in hospitals with varying bed

capacities in the state. In South Carolina, hospital bed capacities range from 24 bed

critical access hospitals to 845 bed regional medical centers. There are 91 hospitals

within the state, and for the 63 acute care hospitals, the average capacity is 177 beds

(SCHA, 2010).

To clarify terminology in the research questions, acute care hospitals are short-

term facilities that provide patient care for brief but severe episodes of illness, healthcare

conditions that result from disease or trauma, and the post-operative phase of recovery

(SCDHEC, 2009). A brief hospitalization is defined as a hospital stay with duration less

than eleven days (Herz, Endicott, & Gibbon, 1979). Services rendered in an acute care

hospital are provided by a variety of clinical personnel who have abilities that include

diagnosis and clinical management for a wide array of conditions and injuries.

Therapeutic interventions are processes used by healthcare professionals who utilize

technology, pharmaceuticals, and medical supplies to accomplish their goals in the

provision of care in acute care hospitals (SCDHEC, 2009).

Epidemiologic surveillance is the ongoing, systematic collection, analysis,

interpretation, and dissemination of data regarding a health-related event for use in public

health to reduce morbidity and mortality to improve health outcomes (CDC, 2001). There

are essential elements of surveillance that are important to CLABSI clinical management.

Infection control practitioners should focus their efforts in a targeted method or perform

surveillance by objective (CDC, 2006). This allows the practitioners to focus their efforts

on specific events, processes, organisms, or populations of patients. The ICP may also

choose to conduct a comprehensive approach to surveillance that includes continuous

monitoring of all patients and processes performed within the hospital environment;

although the continuous approach is a labor-intensive process. Data to be analyzed when

conducting surveillance includes demographic data, infection onset, site of infection, and

patient location during onset of the HAI. Risk factors are evaluated to determine devices

being used and procedures being conducted. Laboratory data are reviewed to identify

pathogens, serology, pathology and antibiogram identification. Radiology and imaging

records may be used in an effort to determine radiographic identification of infection.

The Centers for Disease Control indicates that surveillance assesses the

population and identifies those who are at greatest risk for the development of CLABSI

(CDC, 2006). This was accomplished through evaluating healthcare-associated

infections as well as monitoring patient care practices focused on the prevention of HAI.

Persons conducting surveillance evaluated the outcomes or processes of care being

provided. Examples of outcomes include HAI, specific organism causing sites of

infection, colonization, pyogenic, or pus forming complications, and infections that are

derived from vascular access devices.

The CDC (2006) stated that processes that require surveillance are surgical

procedures, medication administration errors, and practitioner compliance with protocols

and central line insertion practices (CLIP). Other examples that are not related to either

process or outcome include occurrence of reportable diseases and conditions, and hospital

personnel who have communicable disease(s).

Other elements that are essential to the surveillance process include the infection

control professionals’ (ICP) ability to establish observation time periods, to select

surveillance methodologies, to use standardized definitions for data collection, to collect

denominator or population data in the event that rates of infection are calculated, and to

analyze surveillance data, and the reporting of surveillance information within specific

time constraints (CDC, 2006).

Problem Statement

Health, medical, and financial consequences arise for healthcare institutions when

length of stay was increased after developing CLABSI or other HAIs while utilizing

central lines as an adjunctive procedure. Healthcare executives and medical practitioners

alike recognize the importance of clinical intervention to ameliorate the impact of

financial loss when the length of stay increases following the development of hospital-

acquired central line associated infection. Endemic HAI surveillance should be conducted

by ICPs in an active, patient based method (CDC, 2006). The CDC recommendation

encouraged the ICP to perform this function in a prospective approach that provides

incidence rates that are risk adjusted. The research identified surveillance patterns in

acute care hospitals in South Carolina. These surveillance patterns determined clinically

effective methods that created an improved healthcare outcome as indicated from the data

collected from CLABSI records.

Research Design

In this study, quantitative research was conducted with the goal of determining

the correlation of records from acute care hospitals with varying bed capacities

(independent variables) and the incidence of hospital-acquired infection (dependent

variables) in the population observed who develop central line associated bloodstream

infections (Creswell, 2009). The quantitative research design, which was descriptive,

used a correlational approach to conduct the study. In this research, correlational

research was ideal to use when the researcher desired to identify the linkages or

relationships between variables when a study cannot be conducted. To estimate

accurately the relationship between variables, a descriptive (correlational) study usually

needs a larger sample of subjects. The study compared samples of all reported CLABSI

cases in the population at acute care hospitals in South Carolina with varying bed

capacities of 50 or less, 51 to 200, 201 to 500, and 501 or more beds from January 1,

2008 through December 31, 2008. Since all reported records were used during this span

of time, a power analysis was not required to assure adequate volumes of records are in

the study. The numbers of records studied are in excess of a Confidence Level of 99%

and a Confidence Interval of one. To prevent bias, all samples were assigned from a

database housed at the South Carolina Department of Health and Environmental

Control’s Division of Acute Disease Epidemiology (SCDHEC DADE). The methodology

for the study (correlational) was used to determine the relationship of surveillance and

therapeutic measures that minimize HAI and CLABSI in acute care hospitals in South

Carolina. The research did determine the effect that activities have on incidence and

prevalence of HAI in recipients of central line catheter placement.

Analysis for the study was conducted by using parametric inferential statistics to

identify linkages in data in the population being analyzed. This analysis evaluated those

mutually exclusive and all exclusive hypotheses or null and alternative hypotheses. The

alternative Hypothesis typically acknowledges that differences existing between data are

caused by systematic variances between the groups. Multivariate Analysis of Variance

(MANOVA) was employed since more than one dependent variable existed in the study.

The use of MANOVA allowed research to be conducted on all dependent variables at

once. In this research, all dependent variables were related to each other, and it is

expected that each dependent variable is normally distributed and measured on an

interval (Cronk, 2006). Threats to internal validity for the research include (a) a

possibility that all hospitals do not have CLABSI, (b) data collector bias occurred prior to

submission to South Carolina Department of Health and Environmental Control, (c)

CLABSI may be evaluated as all other HAI, and (d) hospitals do not have a sufficient

number of infection control practitioners dedicated to the monitoring of CLABSI.

Staffing recommendations state that 1.0 to 1.5 full-time equivalent infection control

practitioners to every 100 occupied beds within a hospital setting is ideal (Wright et al.,

2010). It was not expected that external biases would be attributable to selection bias

since all available records was selected from the database. Minimal threat from reactive

effects of experimental arrangements, reactive effects of testing, or multiple treatment

interference is anticipated (Patten, 2009).

The quantitative research design was descriptive and used a correlational

approach to conduct the study. Correlational research provided a different approach

whereby the researcher can investigate fully the independent variable’s relationship to the

dependent variables of the study (Leedy & Ormrod, 2005).

Target Population, Sampling Method, and Related Procedures

Target population

South Carolina Hospital Association (2010) reported that in the State of South

Carolina there are 63 short-term and long-term acute care facilities of differing bed

capacities (24-845 beds). From these medical facilities, the study evaluated data from

only acute care hospitals that provided care requiring central lines as part of adjunctive

care. Rehabilitation, pediatric, behavioral health, and substance abuse facilities were not

included in this study population in an effort to focus research on the needs of acute care

hospitals and to minimize variations in data often seen with long-term care patients. From

the 63 hospitals that were acute care oriented, the research stratified hospitals by bed

capacity (a) 50 or less, (b) 51 to 200, (c) 201 to 500, and (d) 501 or more. Acute care

hospitals must provide complete care for every aspect of injury, from prevention through

rehabilitation. These hospitals must have adequate depths of resources and personnel with

the capability of providing competent healthcare leadership, education, research, and

systems planning.

All records in this study showed evidence of CLABSI during the participant’s

hospitalization and treatment with a central line access device. Only records indicating

CLABSI were evaluated in the study.

Procedures

Sampling Methods

In the non-probability sampling study, a purposive sampling technique was used

as a means of pulling representatives into the study population since all records reported

to SCDHEC DADE in a CLABSI database in a specific time span was used. The

sampling frame includes all records in a specific time span of January 1, 2008 through

December 31, 2008. This selection process was comprehensive so that no biases exist.

Non-probability sampling provided the researcher no opportunity to assure forecasting

that all elements of the population were represented in the sample (Leedy & Ormrod,

2010). Forecasting is defined as using a historical perspective of data to infer future

outcomes (Fildes, Nikolopoulos, Crone, & Syntetos, 2008). Forecasting was not required

since a correlational design was being employed. The comprehensive selection process

was used so that each medical record in the population had an opportunity for evaluation

from the data source. The data came from four strata in the set of data (a) hospitals with

bed capacities 50 or less, (b) 51 to 200, (c) 201 to 500, and (d) 501 or more. In purposive

sampling, the researcher selects records comprehensively from all populations within a

preidentified group that was being sought out for study (Neuman, 2006). This process

guaranteed equal representation in the identified strata (Leedy & Ormrod, 2005).

Sample Size

The target population for the study was all medical record files from all acute care

hospitals in South Carolina reported in the time span of January 1, 2008 through

December 31, 2008. Approximately 700 medical records were available for this study

during the time-span; therefore, a power analysis was not required to assure adequate

volume of record files. Each medical record selected in the sampling represented a central

line associated hospital-acquired infection diagnosis. The volume of records used in this

research assured that a Confidence Level of 99% and a Confidence Interval of one was

apparent.

Sampling Procedures

Data was obtained from the SCDHEC DADE. The data were reported by

mandate to the South Carolina Department of Health and Environmental Control for

clinical data evaluation. An information request was directed to the Division of Acute

Disease Epidemiology with constraints identified as all reporting acute care hospitals

with a bed capacity of 24 beds through 845. All information provided had patient

identifiers removed prior to acceptance to the research study. Data had hospital identifiers

removed as well; only hospital bed capacity was known.

Recruitment

Recruitment was unnecessary since the information being analyzed was

secondary data from a preexisting database provided by the South Carolina Department

of Health and Environmental Control Division of Acute Disease Epidemiology.

Setting

The setting for the study included study participants receiving central line

catheters as part of their adjunctive therapies in all acute care hospitals in the State of

South Carolina. Rehabilitation, pediatric, behavioral health, and substance abuse facilities

were not included in this study population. The research stratified the 63 acute care

hospitals by bed capacity (a) 50 or less, (b) 51 to 200, (c) 201 to 500, and (d) 501 or

more.

Protection of Participants

Ethical issues surrounding research where human beings are the potential sources

of data collection include (a) protection from harm; (b) right to privacy; (c) informed

consent with ethical considerations; (d) and honesty with professional colleagues (Leedy

& Ormrod, 2010). The research was conducted under guidance of the Internal Review

Board and a professional code of ethics of a doctoral student who is also a medical

professional. The Health Insurance Portability and Accountability Act (HIPAA) of 1996

were followed in accordance with federal law. HIPAA is a federal rule that provides

privacy and security safeguards for the protection of the patient’s private, confidential

information (Gostin & Nass, 2009). The data were provided without any identifying

markers that would indicate who was the recipient of care. Only hospital type and bed

capacity was provided and the data used is public domain information through the South

Carolina Department of Health and Environmental Control.

Operationalization of Variables

Validity in quantitative research was described as construct validity (Bordens &

Abbott, 2008). The construct identified in quantitative research was the initial concept, or

hypothesis that determines which data was to be gathered and how it was to be gathered.

Constructs measured included CLABSI data identified in all acute care hospitals and

surveillance and therapeutic measures used to minimize the infectious processes. The

dependent variables in the research were the demographic and clinical factors reported by

each facility on the National Healthcare Safety Network (NHSN) Primary Bloodstream

Infection report and the HIDA Hospital Infection Control Processes Report. Hospitals

report this information monthly to SCDHEC DADE. The independent variables were

hospitals of varying bed capacities. Data was collected to identify each variable through

resources requested from South Carolina Department of Health and Environmental

Control. Values are assigned to each variable because they have mutually exclusive

categories (bed capacity of 50 or less, 51 to 200, 201 to 500, and 501 or more).

To operationalize each of these variables, efforts were made to fully define and

describe each variable to determine how data was collected to create each variable, and

how values were assigned to each variable, and to determine and identify the level of

measurement for each variable. Data derived from the independent variables, and acute

care hospitals, were described by the numbers of patients that are provided care over a

span of time (number of average occupied beds/annually) and the magnitude of care

being provided as seen by number of central line days experienced by each clinical site.

Data was collected for these variables by secondary data sources provided by the

SCDHEC DADE, which is responsible for collecting and storing this information in a

central repository, available through public domain.

Dependent variables included (a) numbers of infection control practitioners, (b)

total hours per week dedicated to infection control practices, (c) persons who provide

oversight to infection control practitioners within the facility, (d) frequency of reviews of

infection control policies, (e) presence or absence of a syndromic surveillance program,

(f) presence or absence of syndromic surveillance programs that are automated with the

laboratory, (g) presence or absence of a defined method and protocol for monitoring

compliance with hand hygiene policies, (h) persons who are monitored for hand hygiene

practices within the facility, (i) method of hand hygiene monitoring, (j) person who

perform hand hygiene compliance monitoring, (k) frequency of hand hygiene

monitoring, (l) total number of hospital wide monthly observations, (m) locations

selected for monitoring, (n) presence or absence of protocols/procedures for CLABSI

infection control processes, (o) processes included in CLABSI infection control

protocols, (p) monitoring /compliance methods used for procedures, (q) frequency of

CLABSI monitoring, and (r) frequency of temporary or permanent central lines.

The data was collected through a database provided by the South Carolina

Department of Health and Environmental Control that contained detail information from

CLABSI records. Each component of data was identified by ordinal point with an

arbitrary record number. To determine and identify the level of each variable, Statistical

Test was the basic scale of measure used. Many statisticians believe that this view is

restrictive, but appropriate with interval or ratio data derived from ordinal data (Bordens

& Abbott, 2008).

Instrumentation

The integrated data in the study was secondary data collected through the

SCDHEC DADE. Data was collected into an Excel spreadsheet and imported into

statistical software for analysis. Microsoft has developed an intact instrument for the

specific purpose of data collection and statistical analysis, PASW Statistics 19 (formerly

SPSS) statistical software, which was used to provide an integrated environment for

analysis and interpretation of data that, may reveal patterns, anomalies, key variables, and

relationships in the data of the research.

Research Questions and Hypotheses

These following research questions were intended to provide clarity to a

confusing area of healthcare and clinical management of CLABSI. These questions

focused on the CLABSI infection rate at acute care hospitals with varying bed capacities.

Surveillance interventions used by these hospitals were measured against results of

CLABSI rates indicated by each hospital. In addition, a research consideration was what

infection control processes were in place for monitoring and treatment after development

of hospital-acquired infection associated with central line placement. In the correlational

design research, a determination of the following facts occurred:

Research Question 1. Is there a significant variance in the rates of CLABSI

reported in hospitals with capacities of 50 or less, 51 to 200, 201 to 500, and 501 beds

and more?

H1 for Research Question 1. There is a significantly higher incidence of CLABSI

observed in hospitals of higher volume than lower bed capacities.

H2 for Research Question 1. There are significant differences between responsive

measures used by hospitals to minimize the effect of CLABSI when two or more ICPs

that are full-time are employed for surveillance purposes.

H0 for Research Question 1. There is a significantly lower incidence of CLABSI

observed in hospitals of higher bed capacities than in hospitals of lower bed capacities.

H0 for Research Question 1: Hospital CLABSI rates are not dependent upon the

number of full time staff and number of hours of Infection Control activities performed

per month.

Research Question 2. Is there a correlation between monitoring and compliance methods

and incidence rates of CLABSI infection?

H1 for Research Question 2: A lower rate of CLABSI is observed when direct

observation is conducted during catheter insertion.

H2 for Research Question 2: When staff members are authorized to stop invasive

procedures when they observe protocols not being followed, poor compliance with

aseptic technique that results in increased incidence of CLABSI.

H3 for Research Question 2: The use of a preestablished procedure checklist

decreases the rate of CLABSI in acute care hospitals in South Carolina.

H4 for Research Question 2: Acute care hospitals with larger bed capacities have

well-defined syndromic surveillance programs automated with their laboratories and have

more full time infection control practitioners on staff and perform more hours of infection

control activities per month.

H5 for Research Question 2. A greater number of person-hours used to identify,

observe technique, evaluate insertion site care, and perform surveillance of hand washing

technique decreases the number of infectious processes.

H0 for Research Question 2: No change in the rate of CLABSI is observed when

direct observation is conducted during catheter insertion.

H0 for Research Question 2: When staff members are authorized to stop invasive

procedures for unacceptable performance of procedure, non-compliance incidence rates

of CLABSI has no appreciable change.

H0 for Research Question 2: The use of preestablished procedure checklist does

not affect the rate of CLABSI in acute care hospitals in South Carolina.

H0 for Research Question 2: Hospital CLABSI rates are not dependent upon the

number of full time staff and number of hours of infection control activities performed

per month.

Research Question 3. Does a relationship exist between an increased rate of CLABSI in

acute care hospitals and integration of prevention processes when included in their

facilities protocols?

H1 for Research Question 3: The daily review of line necessity and prompt

removal of unnecessary lines minimize the rate of CLABSI in South Carolina’s hospitals.

H2 for Research Question 3: Monitoring of optimal catheter site selection

minimizes the incidence of CLABSI infection.

H3 for Research Question 3: Acute care hospitals that use chlorhexidine skin

antiseptics during site preparation and care have lower incidence of CLABSI infection.

H4 for Research Question 3: Acute care hospitals that use maximal barrier

precautions upon insertion observe a lower incidence of CLABSI infection.

H5 for Research Question 3: CLABSI rates are lower when ICPs conduct hand

hygiene monitoring.

H0 for Research Question 3: CLABSI rates are higher when hygiene monitoring

is conducted more frequently.

H0 for Research Question 3. Higher CLABSI rates are seen when the total

number of hospital wide monthly observations are increased.

H0 for Research Question 3: The daily review of line necessity and prompt

removal of unnecessary lines does not influence the rate of CLABSI in South Carolina’s

hospitals.

H0 for Research Question 3: Monitoring of optimal catheter site selection does

not affect the incidence of CLABSI infection.

H0 for Research Question 3: Acute care hospitals that use chlorhexidine skin

antiseptics during site preparation and care do not influence the incidence of CLABSI

infection.

H0 for Research Question 3: Acute care hospitals that use maximal barrier

precautions upon insertion have no affect on the incidence of CLABSI infection.

H0 for Research Question 3: CLABSI rates are higher when hygiene monitoring

is conducted less frequently by ICPs.

Data Collection

The procedure for data collection included request for data through the South

Carolina Department of Health and Environmental Control. A request for data was

routed through the Division of Acute Disease Epidemiology with the use of a permission

letter provided by Capella University. The data request was for files from the DADE

database that contains data from both the NHSN and the HIDA Hospital Infection

Control. Data was requested from January 1, 2008, through December 31, 2008, on

records indicating a diagnosis of CLABSI hospital-acquired infection (HAI). The request

for the data was placed in an Excel spreadsheet format. The Excel spreadsheet format

allows for easy integration into the PASW Version 19 software for data and statistical

analysis.

Field Testing and Pilot Testing

The research conducted neither field nor pilot testing. The data collection

occurred through submission of epidemiological information to the Division of Acute

Disease Epidemiology by hospitals in the State of South Carolina. The nature of the study

was that of a retrospective analysis and does not require the researcher to pilot test the

data collection procedures.

Data Analysis

To meet the purpose of the research and to provide answers to the research

questions, data analysis used appropriate statistical analysis procedures. Data analysis

was conducted by using the software system, PASW Statistical Software Version 19 for

Windows.

Upon receipt of the data from the South Carolina Department of Health and

Environmental Control’s Division of Acute Disease Epidemiology, the researcher

developed a data file within PASW to store the data. The research employed standard

univariate descriptive statistics to analyze the demographics and CLABSI data as

stratified by the clinical site and the target population. Means, ranges, and standard

deviations were used to summarize the data.

The researcher used Multivariate analysis of variance (MANOVA) tests to

determine differences in variance between the variables. MANOVA examined the

differences between groups when two or more independent variables and two or more

dependent variables existed within the research. An analysis of each research question

determined discriminate analysis and factor analysis. The primary reason for using

MANOVA was to investigate how two or more dependent variables related to group

differences. MANOVA was considered an ANOVA with several dependent variables.

ANOVA typically analyzes the difference in means between two or more groups.

MANOVA measures the difference in two or more groups of means being studied

(Bordens & Abbott, 2008).

MANOVA is typically used to evaluate the relationship between several

dependent variables or outcomes simultaneously. MANOVA has the capability in

research analysis to determine if study components differ along a combination of

dimensions (Field, 2009). MANOVA provided the researcher an opportunity to test

hypotheses regarding the effect of one or more independent variables on two or more

dependent variables. When conducting this test, a p-value is generated to determine

whether the null Hypothesis can be rejected or failed to be rejected. Researchers use

MANOVA to determine if there are differences in the mean values of the dependent

variables between the varying levels of the independent variable. MANOVA can measure

differences in correlations among the dependent variables between the different levels of

the independent variable.

MANOVA may be used when two major data situations exists. The first situation

occurs when there are multiple correlated dependent variables, and the researcher

requires a single, statistical test on this set of variables instead of performing multiple

individual tests as conducted in ANOVA. The second and often the more important

reason is to explore how independent variables influence the development of patterns of

response with the dependent variables. In this occurrence, the researcher used contrasting

codes with the dependent variables to test hypotheses regarding how the independent

variables predict the dependent variables (Field, 2009).

In MANOVA, research tests the null Hypothesis that indicates differences

between the groups. The dependent variables correlated with one another, and a new

dependent variable maximizes group differences. This MANOVA application identified

for this research was a super-factorial design. Records analyzed in this research received

multiple treatment combinations since multiple independent variables existed and are

measured against multiple dependent variables. The interaction between the variables

develops composite dimensions specific to each effect on the research design. In this

research, factorial MANOVA calculated multiple sets of composite variables, and each

set is specific to a particular outcome (Bordens & Abbott, 2008).

Limitations of the Research Design

Limitations of the retrospective descriptive study design existed within the

research. Cause and effect cannot be determined in the research since control group data

does not exist. Conclusions drawn by research were limited and may only be able to

determine relationships and not cause and effect (Sproull, 1995). Concepts associated

with internal and external validity are problematic and are discussed later in this chapter.

Research processes and procedures of initial data collection are not under the control of

the study and thus pose validity concerns (Neuman, 2006).

Internal Validity

The correlational research defined the relationship between CLABSI in

participants and acute care hospitals. In a correlational design researchers observe,

describe, and document a condition or event that occurs, and then they attempt to find

possible patterns (Patten, 2009). Threats to internal validity in the correlational design

focused the research on specific factors that were known to affect or may logically be

expected to affect the variable on which groups are being evaluated. Other threats

focused study efforts on the likelihood of the comparison groups differing on each of

these factors and then evaluation of these threats because of how likely a researcher may

have a plan to control for them. Specific threats to the study included events that would

cause bloodstream infections that were not associated with the placement of central lines

(e.g., sepsis, bacteremia). An additional threat to the study would be associated with a

hospital changing staffing levels; and/or development of a new process of surveillance

assessment or policy change that affects how clinical assessment/monitoring is

performed. To minimize this effect, acute care hospital data was evaluated closely to

assure variances in data integrity are not compromised to minimize this artifact. Other

attributes or threats to internal validity as identified by Patten (2009) which do not exist

include history, maturation, instrumentation, testing, or statistical regression. Intact

groups existed in the study since participants were not assigned to hospitals at random.

External Validity

External validity in the study was the extent to which the findings of a particular

study can be generalized to factors other than those observed in the study. External

validity threats are in the form of interactions. The study does not include treatment

attribute interaction but does include treatment-setting interactions. In the study, the

independent variables (acute care hospitals) may interact with other external factors

(staffing, surveillance processes) or context to result in different effects for different

settings so that the effect cannot be generalized to all settings. These threats were not

repaired since the study utilized ex post facto data.

Expected Findings

The research expected to reveal a relationship between the bed capacity of

hospitals and the rates of CLABSI and HAI resulting form insertion of central lines.

Researcher’s Position Statement

The researcher’s relationship to the study was one of academic interest. In-depth

study of CLABSI, including how prevention and surveillance strategies affect incidence

and prevalence of HAI in the hospital environment, was research worthy. A

determination of effect on CLABSI rates of surveillance, hygiene practices, and treatment

methods was critical to determining if these measures have a positive or negative effect

on incidence and prevalence of infection. This research builds upon previously conducted

research in the area of CLABSI on a national level, but not specifically for South

Carolina since there are no published studies available for this state. There were no

conflicts of interest that existed between the researcher and South Carolina Department of

Health and Environmental Control or mandatory reporting agencies that submit data to

the health agency. The researcher’s views on the topic are strictly academic and based on

a desire to know and understand how organizational policies and processes may or may

not affect clinical outcomes. Strategies to prevent biases from developing during the

course of the research include close interaction between the researcher and his mentor

and committee. A desire to understand the problem and a need to know the research

rationale ensured the elimination of biases within the study.

Ethical Issues in the Study

In the use of secondary data in the study, the ethical concern related to the

research was the issue of privacy and confidentiality. The research addressed these

concerns by only using de-identified data and large sample sizes. The study did not use

human participants in the course of collection of data. The implementation of HIPAA

standards existed with de-identified data to assure compliance with the federal mandate.

No vulnerable populations existed within the study. There were no research assistants or

aides to assists in the conducting of the research. No physical control, coercion, undue

influence, or manipulation of persons or data occurred. Because the data analyzed was

within the public domain and individual interaction does not occur, there was no need for

participant consent forms. The research did not offer benefits, and no risks existed for

organizations that provided data files for anyone whose has information contained within

data files. The research received IRB approval at both Capella University and South

Carolina Department of Health and Environmental Control to assist in the oversight of

the study and to assure that conducted research used proper precautions to prevent

breaches in confidentiality.

Conclusion

The study sought to determine whether a correlational relationship existed in

hospital records reported to SCDHEC DADE with CLABSI diagnosis, acute care

hospitals of varying bed capacities, surveillance techniques, and therapeutic processes

employed by those healthcare facilities. Quantitative research and a correlational design

measured the incidence and prevalence rates of infection, use of therapeutic regimes for

treatment, surveillance processes, and hygiene matters with personnel and performance of

procedures. Independent variables, acute care hospitals with varying bed capacities are

measurable criteria that describe sources of records for the study. MANOVA measured

multiple dependent and independent variables. Relationships between the dependent and

independent variables provided the researcher an opportunity to observe a correlation in

data, which may exist and will aid healthcare providers in their efforts to evaluate the

efficacy of surveillance, hygiene, and treatment options.

CHAPTER 4. DATA COLLECTION AND ANALYSIS

Introduction

This chapter provides data analysis to answer research questions and hypotheses

developed while seeking to determine the significance of central line associated

bloodstream infections (CLABSI) in hospitals in the State of South Carolina over a time-

span of one calendar year (2008). The year 2008 is significant to this study since this is

the first year of data collection and reporting of Hospital Infection Disclosure Act

(HIDA) data to the Division of Acute Disease Epidemiology (DADE) at the South

Carolina Department of Health and Environmental Control (SCDHEC). This data were

mandatorily reported to and comprehensively analyzed by the SCDHEC. For the purpose

of this study, an evaluation focused on CLABSI and bed capacity of responding

hospitals. This study was conducted in an attempt to determine if facility size determines

quality of care by using effective surveillance and therapeutic measures. The National

Healthcare Safety Network (NHSN) transmitted this data to SCDHEC for analysis. The

NHSN provided a secure, internet-based surveillance system for integrated patient and

healthcare worker safety monitoring and reporting.

This chapter described data from participating health care organizations including

demographic information that illustrated the hospitals clinical and surveillance

characteristics and the connection between analysis and outcomes. Then, a statement of

the results from analysis of data was presented. A brief point-by-point summary of the

results and findings of the data analysis was described and organized to address the

research questions and hypotheses identified earlier in the dissertation. Details of the

analysis were described by providing a fully detailed presentation of the data analysis

process and results. These tasks were accomplished by evaluating each research question

and hypothesis in order and describing the analysis by presenting the results for that

question and hypothesis. Finally, this chapter provided a summary of the findings that

transition to Chapter 5.

The purpose of this research was to determine what causal factors determine if

CLABSI will occur and what predictive measures minimize the effect of infection. An

analysis of correlative factors may lay the foundation for additional groundwork for

future research focused on minimizing the risk of central line infection in hospitalized

patients. The curiosity in such research exists because of the focus on chronic disease

epidemiology and the extensive morbidity and mortality associated with hospital-

acquired infections (HAI). A background in clinical medicine provides an understanding

of the hospital, the critical care clinical environment, and recently the role that public

health policy has in population health management. Pursuit of existing data from the

Division of Acute Disease Epidemiology (DADE) provided many opportunities for

clinical research utilizing historical data that was not previously managed. The SCDHEC

DADE provided a living laboratory to apply this information and construct the study in a

way that would be beneficial to healthcare professionals, public health agencies, students

of CLABSI, and to the doctoral dissertation process. The effects this data had on clinical

outcomes were discussed over the course of the past year to determine a need for analysis

and further research with the SCDHEC Division of Acute Disease Epidemiology.

Description of the Sampled Data

The data provided for research from SCDHEC DADE was collected from the

Hospital Acquired Infection Reports (HIDA). Each hospital provided their specific

HIDA reports bi-annually in January and July. Data was reported by hospital bed

capacity and each report contains the hospital name (de-identified for this study), the

number of licensed general beds (excluding psychiatric and substance abuse beds), the

average daily census for general beds, number of surgical procedures performed, number

of infections, surgical site infection (SSI) rates for each reportable surgical procedure and

separated by risk category. This report also includes the number of central line days,

number of central line associated bloodstream infections (CLABSI), and CLABSI

infection rates by defined location within the hospital. In the HIDA report, hospitals are

not compared to other hospitals within the state. Each defined hospital’s infection rate for

the reportable hospital-acquired infection (HAI) is compared with the historical

experience in the standard population of all NHSN users across the United States.

The specific data evaluated in this study was derived directly from the HIDA

Reports that are completed by each hospital. Only CLABSI data was desired and

analyzed for this study. Data obtained reflects nine hospitals from 0 to 50 beds, 30

hospitals from 51 to 200 beds, 19 hospitals from 201 to 500 beds, and five hospitals with

500 or more beds within their facility. Answers to the HIDA questions were submitted to

SCDHEC’s DADE by hospitals in South Carolina over the span January to December

2008 (See Appendix A and Appendix B).

100

Research Methodology and Data Analysis

This chapter provided an opportunity to evaluate research data quantitatively in a

correlational design and by integrating statistical techniques such as correlation, cluster

analysis, and factorial analysis to evaluate objective theories that examined the

relationship between variables in the study. The research provided an opportunity to

analyze the data according to their linear functions as well as the integration of their uses

in clinical analysis and decision-making. Data gathered from SCDHEC DADE and

supporting information from HIDA, NHSN, and the South Carolina Hospital Association

(SCHA) provided clinically significant information for this research.

In the analysis of data, a combined technique of inductive and deductive thematic

analysis was conducted. This integrative process of analysis detailed in this research

demonstrated the ability of using a hybrid approach to thematic analysis by using a large

volume of data for inductive thematic analysis and the ability to utilize deductive

reasoning to determine trends or themes in data. Analysis of data in this chapter provided

an opportunity to observe patterns of central lines associated bloodstream infections and

the role that surveillance processes and therapeutic modalities had in minimizing these

types of hospital-acquired infections (HAI). The analysis of this data also provided an

opportunity to understand the role that various hospital capacities had on infection by

measuring the size of the organization and the volume of infections observed. An

analysis of patterns in these varying paradigms of care provide health care organizations

an opportunity to lay the groundwork for future research focused on reducing rates of

infection by implementing aggressive surveillance and therapeutic procedures and

policies within hospitals in South Carolina.

101

MANOVA was used to determine whether single or multiple categorical variables

significantly explain the variation in a combination of several continuous dependent

variables. If there were significant relationships between the independent and the

dependent variable(s), this would indicate that the independent variable(s) significantly

explain the variation in the dependent variables. In this research, MANOVA was used to

investigate the differences between hospital bed capacities, integrative surveillance

measures, and therapeutic modalities employed to minimize the incidence of CLABSI.

No significant variances occurred from the discussion of the processes or protocols

described in Chapter 3. No problems were identified arising from data collection or

analysis. Three research questions were constructed after thoroughly reviewing the HIDA

document that was provided to hospitals:

1. Is there a significant variance in the rates of CLABSI reported in hospitals

with capacities of (a) 50 or less, b) 51 to 200, (c) 201 to 500, and (d) 501 beds and more?

2. Is there a correlation between monitoring and compliance methods and

incidence rates of CLABSI infection?

3. Does a relationship exist between an increased rate of CLABSI in acute care

hospitals and integration of prevention processes when included in their facilities

protocols?

Hypotheses

H1: There is a significantly higher incidence of CLABSI observed in hospitals of

higher volume than lower bed capacities.

H2: There are significant differences between responsive measures used by

102

hospitals to minimize the effect of CLABSI when two or more full-time Infection Control

Practitioners (ICP) are employed for surveillance purposes.

H0 : There is a significantly lower incidence of CLABSI observed in hospitals of

higher bed capacities than in hospitals of lower bed capacities.

H0 : Hospital CLABSI rates are not dependent upon the number of full-time staff

and number of hours of infection control activities performed per month.

Research Question 2. Is there a correlation between monitoring and compliance

methods and incidence rates of CLABSI infection?

H1 : A lower rate of CLABSI is observed when direct observation is conducted

during catheter insertion.

H2 : When staff members are authorized to stop invasive procedures when they

observe protocols not being followed, poor compliance with aseptic technique that results

in increased incidence of CLABSI.

H3 : The use of a preestablished procedure checklist decreases the rate of

CLABSI in acute care hospitals in South Carolina.

H4 : Acute care hospitals with larger bed capacities have well-defined syndromic

surveillance programs automated with their laboratories in addition have more full-time

infection control practitioners on staff and perform more hours of infection control

activities per month.

H5 : A greater number of person-hours used to identify, observe technique,

evaluate insertion site care, and perform surveillance of hand-washing technique

decreases the number of infectious processes.

H0 : No change in the rate of CLABSI is observed when direct observation is

103

conducted during catheter insertion.

H0 : When staff members are authorized to stop invasive procedures for

unacceptable performance of procedure, non-compliance incidence rates of CLABSI

have no appreciable change.

H0 : The use of preestablished procedure checklist does not affect the rate of

CLABSI in acute care hospitals in South Carolina.

H0 : Hospital CLABSI rates are not dependent upon the number of full-time staff

and number of hours of infection control activities performed per month.

Research Question 3. Does a relationship exist between increased rates of

CLABSI in acute care hospitals and integration of prevention processes when included in

their facilities protocols?

H1 : The daily review of central line necessity and prompt removal of unnecessary

central lines minimize the rate of CLABSI in South Carolina’s hospitals.

H2 : Monitoring of optimal catheter site selection minimizes the incidence of

CLABSI infection.

H3 : Acute care hospitals that use chlorhexidine skin antiseptics during site

preparation and care have lower incidence of CLABSI infection.

H4 : Acute care hospitals that use maximal barrier precautions upon insertion

observe a lower incidence of CLABSI infection.

H5 : CLABSI rates are lower when Infection Control Practitioners conduct hand

hygiene monitoring.

H0 : CLABSI rates are higher when hygiene monitoring is conducted more

frequently.

104

H0 : Higher CLABSI rates are seen when the total number of hospital wide

monthly observations are increased.

H0 : The daily review of central line necessity and prompt removal of

unnecessary central lines does not influence the rate of CLABSI in South Carolina’s

hospitals.

H0 : Monitoring of optimal catheter site selection does not affect the incidence of

CLABSI infection.

H0 : Acute care hospitals that use chlorhexidine skin antiseptics during site

preparation and care does not influence the incidence of CLABSI infection.

H0 : Acute care hospitals that use maximal barrier precautions upon insertion has

no affect on the incidence of CLABSI infection.

H0 : CLABSI rates are higher when hygiene monitoring is conducted less

frequently by Infection Control Practitioners.

Presentation of Data and Results of the Analysis

Sixty-three acute care hospitals are presented in this study. Additionally, these

hospitals were stratified into four categories based upon bed capacity as described in the

HIDA questionnaire reported to SCDHEC DADE. The data was presented in aggregate

form and analyzed by bed-capacity category. Standard descriptive statistical data were

employed to define the demographic characteristics of hospitals with varying bed

capacities. Data analyzed included hospital bed capacity and categorization of these

hospitals into striated groups: 0 to 50, 51 to 200, 201 to 500, and those with more than

500 beds.

105

The following tables describe the statistical observations found in the hospital

populations. Hospitals were provided a random number that served as a means to

identify that same facility throughout the research. The tables indicate the number of

CLABSIs identified during the year 2008, the number of central line days, the infection

rate per 1000 central line days, and the confidence interval, 95%. Lastly, this table

contains the Standardized Infection Ratio (SIR).

The number of central line days is a description of the total number of days a

central line is in place for patients in a hospital unit. The infection rate was obtained

when the total number of central line-associated bloodstream infections was divided by

the number of central line days. That result was then multiplied by 1,000. A confidence

interval of 95% is a range of values around that statistic with the probability that the true

value is contained within that range (Field, 2009). The SIR compares the actual number

of HAIs in a facility with the baseline of infection in the general population with a central

line. A SIR greater than 1.0 indicates that more HAIs were observed than predicted.

Conversely, a SIR of less than 1.0 indicates that fewer HAIs were observed than

predicted.

106

Table 1. Hospital Bed Capacity of 0 to 50 Receiving Central Lines

________________________________________________________________________ Hospital Bed Cap. Number of Number Infection Confidence SIR

CLABSI of Central Rate (per Interval Line Days 1000 (95%) Central Line Days

Hospital 1 25 0 113 0 ----- ----- Hospital 2 25 0 15 0 ----- ----- Hospital 3 25 0 0 0 ----- ----- Hospital 4 25 0 0 0 ----- ----- Hospital 5 25 1 151 6.62 [-6.3,19.5] 1.0004 Hospital 6 32 0 77 0 ----- ----- Hospital 7 41 0 119 0 ----- ----- Hospital 8 43 0 842 0 ----- ----- Hospital 9 48 0 45 0 ----- ----- * NA identifies that category as not applicable.

As observed in Table 1, few hospitals provide central line adjunctive care as

demonstrated by those hospitals with zero or few central line days. These hospitals are

typically critical access facilities in small rural communities. Critical access hospitals

(CAH) must provide 24-hour emergency services, with medical staff on-site; or on-call

and available on-site within 30 minutes, or within 60 minutes if certain frontier area

criteria are met. South Carolina does not have frontier areas and the 60-minute criterion

does not apply. A CAH must develop agreements with an acute care hospital related to

patient referral and transfer, communication, emergency, and non-emergency patient

transportation. CAHs must be located in a rural area and meet one of the following

criteria. There must be a 35-mile distance from another hospital, or 15 miles from

another hospital in mountainous terrain or areas with only secondary roads (Sand, 2010).

These smaller hospitals typically stabilize a patient and transport rather than

maintaining the patient on site especially when more advance levels of care are required.

As observed in the number of central line days few patients were provided this form of

107

treatment or less critical patients are provided central lines but are maintained for a much

shorter span of time.

The data contained with Table 1 reflects one central line infection over the course

of the year 2008. A range of 0 to 842 days is observed for central line days yielding an

infection rate of 6.62. This fell within the range of -6.3 to 19.5 that are reflective of a

95% confidence interval. A SIR score for that one hospital reflected 1.0004.

108

Table 2. Hospital Bed Capacity of 51 to 200 Receiving Central Lines

Hospital Bed Cap. Number of Number Infection Confidence SIR CLABSI of Central Rate (per Interval Line Days 1000 (95%) Central Line Days

Hospital 10 59 0 11 0 ----- ----- Hospital 11 53 0 6 0 ----- ----- Hospital 12 197 2 1449 1.38 [-0.5, 3.3] 1.905 Hospital 13 55 1 242 4.13 [-4.0, 12.2] 1.707 Hospital 14 116 5 1280 3.9 [0.5, 7.3] 1.523 Hospital 15 82 0 333 0 ----- ----- Hospital 16 59 0 111 0 ----- ----- Hospital 17 130 7 2572 2.72 [0.7, 4.7] 7.403 Hospital 18 131 5 1248 4 [0.5, 7.5] 1.603 Hospital 19 130 7 1041 6.72 [1.7, 11.7] 4.519 Hospital 20 82 0 298 0 ----- ----- Hospital 21 93 6 1607 3.73 [0.7, 6.7] 1.393 Hospital 22 121 6 1545 3.88 [0.8, 7.0] 1.507 Hospital 23 76 3 763 3.93 [-0.5, 8.3] 1.545 Hospital 24 105 0 1061 0 ----- ----- Hospital 25 124 3 352 8.52 [-1.1, 18.1] 7.261 Hospital 26 102 0 198 0 ----- ----- Hospital 27 72 0 186 0 ----- ----- Hospital 28 109 0 162 0 ----- ----- Hospital 29 90 0 71 0 ----- ----- Hospital 30 169 0 464 0 ----- ----- Hospital 31 109 4 1073 3.73 [0.1, 7.3] 1.391 Hospital 32 72 0 8 0 ----- ----- Hospital 33 56 0 609 0 ----- ----- Hospital 34 97 4 917 4.36 [0.1, 8.7] 1.902 Hospital 35 93 0 0 0 ----- ----- Hospital 36 94 2 2786 0.72 [-0.3, 1.7] 5.169 Hospital 37 125 1 600 1.67 [-1.6, 5.0] 2.783 Hospital 38 167 6 1077 5.57 [1.1, 10.1] 3.103 Hospital 39 143 0 286 0 ----- ----- * NA identifies that category as not applicable.

In the 51 to 200 bed capacity range, significantly higher population hospitals that

use central line were seen in the statistical evaluation. Additional line usage was

reflected since there are many of South Carolina’s hospitals within this bed range. The

data represents 30 hospitals with significantly more central line days and an increase

109

number of CLABSIs over the same span of time as observed in the 0 to 50 bed-category.

This data indicated that 51 to 200 bed facilities observed a higher acuity patient requiring

increased sustentative care.

Table 3. Hospital Bed Capacity of 201 to 500 Receiving Central Lines

________________________________________________________________________ Hospital Bed Cap. Number of Number Infection Confidence SIR

CLABSI of Central Rate (per Interval Line Days 1000 (95%) Central Line Days

Hospital 40 230 6 4836 1.24 [0.2, 2.2] 1.001 Hospital 41 461 6 3830 1.57 [0.3, 2.9] 0.9978 Hospital 42 204 2 2064 0.97 [-0.4, 2.4] 0.9990 Hospital 43 310 16 3583 4.47 [2.3, 6.7] 0.9990 Hospital 44 210 0 841 0 ----- ----- Hospital 45 219 8 2530 3.16 [1.0, 5.4] 1.0001 Hospital 46 414 6 4765 1.3 [0.3, 2.3] 0.9686 Hospital 47 209 1 2044 0.49 [-0.5, 1.5] 0.9984 Hospital 48 453 13 6738 1.93 [0.9, 2.9] 0.9997 Hospital 49 467 3 745 4.03 [-0.5, 8.5] 0.9992 Hospital 50 288 10 4896 2.04 [0.7, 3.3] 1.0012 Hospital 51 258 0 2624 0 ----- ----- Hospital 52 286 13 3271 3.97 [1.8, 6.2] 1.001 Hospital 53 368 10 4362 2.29 [0.9, 3.7] 1.001 Hospital 54 414 6 1318 4.55 [0.9, 8.3] 1.005 Hospital 55 217 0 926 0 ----- ----- Hospital 56 245 0 4505 0 ----- ----- Hospital 57 296 24 7411 3.24 [1.9, 4.5] 0.9995 Hospital 58 283 2 2018 0.99 [-0.4, 2.4] 1.0010 * NA identifies that category as not applicable.

In the 201 to 500 bed capacity category, a significantly higher population of their

data is seen in the statistical evaluation. This is reflective since many of South Carolina’s

hospitals that provide a much higher level of care and provide care for critically ill

patients are within this bed range. Facilities in South Carolina in this bed range are

typically found in metropolitan environments. These hospitals may also serve as referral

110

facilities for much smaller communities. The data represents 19 health care facilities with

significantly more central line days and an increase number of CLABSIs over the same

span of time as observed in the 0 to 50 and 51 to 200 bed categories. This data supports

that 201 to 500 bed facilities observed a higher acuity patient over a longer span of time.

Table 4. Hospital Bed Capacity of 500 Beds or More Receiving Central Lines

________________________________________________________________________ Hospital Bed Cap. Number of Number Infection Confidence SIR

CLABSI of Central Rate (per Interval Line Days 1000 (95%) Central Line Days

Hospital 59 845 41 6824 6.01 [4.2, 7.8] 0.999 Hospital 60 709 14 5003 2.8 [1.3, 4.3] 0.999 Hospital 61 649 4 1575 2.54 [0, 5.0] 0.999 Hospital 62 649 26 3750 6.93 [4.3, 9.5] 1.001 Hospital 63 540 11 3956 2.78 [1.2, 4.4] 1.000

In the 501 and more bed capacity category, a significantly lower population of

their data was seen in the statistical evaluation but a significantly higher number of

central line days per hospital. This was observed since hospitals of this bed capacity in

South Carolina are major referral centers and typically have trauma centers which results

in higher patient acuity. These centers are often teaching hospitals that provide a higher

level of care and provide care for critically ill patients over a longer span of time.

Facilities in South Carolina in this bed range are typically seen in larger communities and

metropolitan service areas. These hospitals may also serve as referral facilities for much

smaller communities. These regional medical centers are often linked with physician

residency programs and other colleges and universities use them as educational host for

clinical students. The data represented 19 health care facilities with significantly more

111

central line days and an increased number of CLABSIs over the same span of time as

observed in the 0 to 50, 51 to 200, or 201 to 500 bed categories. This data supported

evidence that 500 and more bed facilities observe an increase in patient acuity over a

longer span of time.

Data analysis response to Hypothesis 1.

Table 5. Case Summary Bed Category/Infection Rate per 1000 Central Line Days

Bed Category Infection Rate (per 1000 Central Line Days)

Composite SIR rate for each category

0-50 Beds 6.62 1.0004

51-200 Beds 3.93 2.981

201-500 Beds 2.55 .9981

> 500 Beds 4.21 .9996

Table 5 depicts the infection rate among the hospital categories and describes

composite infection rates per 1000 central line days. It was observed that hospitals with

the highest acuity rates and the lowest opportunities to provide care have the highest

infection rates as indicated by the increased infection rate per 1000 central line days and

the composite SIR ratio greater than 1. Table 5 describes that even with an infection rate

of 6.62, the 0 to 50 bed facilities are close to a normal tendency for that size organization.

The 51 to 200 bed category has nearly three times the expected rate of hospital-

acquired infection at 2.98 and the lowest SIR of .9981 in the 201 to 500 bed facilities.

Mark, Harless, McCue, and Xu (2004) suggested that hospitals in the 500 and more

112

category have much higher acuity patients and reflects a SIR less than expected for that

size facility (.996).

Table 6. Multivariate Test to Define Hospital Bed Capacity and Infection Rate/1000 Central Line Days

Effect Value F Hypothesis df Error

df Sig.

Intercept Wilks’Lambda Roy’s Largest Root Pillai’s Trace

.089 10.297 .911

15.446 15.446 15.446

2.000 2.000 2.000

3.000 3.000 3.000

.026

.026 Bed cap. Wilks’Lambda Roy’s Largest Root Pillai’s Trace

.021 11.058 1.658

.499 1.264 .555

70.000 35.000 70.000

6.000 4.000 8.000

.923

.461

.909

Table 6 provides a summary of the Multivariate Analysis of Variance

(MANOVA) used to identify the differences in the hospital bed capacity and the infection

rate per 1000 central line days. This table shows the F ratios (Wilk’s Lambda and Roy’s

Greatest Root) were calculated with the use of Predictive Analytics Software also known

as PASW 19. Use of Statistical Decision Tree ™ (Andrews, et al., 1981) suggested that

analysis of the data be conducted utilizing the multivariate test of Wilks’ Lambda, Roy’s

Largest Root, and Pillai’s Trace because the increase of variance in both dependent and

independent variables. A decision tree, shown in Figure 1, describes the assumption

process used when defining the multivariate test used. Wilks’ Lambda was used since the

data is a probability distribution and most often employed in multivariate hypothesis

testing, especially with regard to the Likelihood-Ratio test. Field (2009) indicated this

could be interpreted as the proportion of the variance in the outcomes that is not

explained by an effect. Roy’s Largest Root was utilized because it behaves differently

113

from the other three test tools. In instances where the other three are not significant and

Roy's is significant, the effect should be considered not significant. Pillai’s Trace is the

sum of the proportion of unexplained variance on the discriminant function variants of

the data (Field, 2009). As described in Table 6, infection rate F (15.446) and significance

(.026) was found to have a statistically significant relationship to the bed capacities of the

study. A review of the analysis of the quantitative data suggest that hospitals that provide

higher levels of care, also see higher acuity patients, and increased volumes of patients

than are customarily seen in smaller rural facilities. Because of this, smaller rural

hospitals are less likely to insert invasive central lines on a routine basis and may not be

as proficient as larger facilities that insert more frequently and to a more diverse patient

population in terms of ethnicity, income, and health status.

Figure 1. Statistical Decision Tree

114

Table 7. Statistical Analysis of Hypothesis 1 Dependent Variables

Source Dependent Variable

Type III Sum of Squares

df Mean Square

F Sig.

Corrected Model

Infection Rate/1000 Central Line Days No. of CLABSI

131.575 2343.100

35 35

3.759 66.946

.335 1.104

.968 .529

Intercept Infection Rate/1000 Central Line Days No. of CLABSI

334.850 1920.007

1 1

334.850 1920.007

129.878 31.670

.005 .005

Bed capacity

Infection Rate/1000 Central Line Days No. of CLABSI

131.575 2343.100

35 35

3.759 66.946

.335 1.104

.968 .529

Error Infection Rate/1000 Central Line Days No. of CLABSI

44.829 242.500

4 4

11.207 60.625

------

----- -----

Total Infection Rate/1000 Central Line Days No. of CLABSI

550.887 4602.000

40 40

--------

-------

------

-----

------

Corrected Total

Infection Rate/1000 Central Line Days No. of CLABSI

176.404 2585.500

39 39

-------

------

115

Table 7 analyzes the statistical processes provided by PASW 19 to evaluate the

Multivariate Tests, Between-Subject Effects. The Between-Subjects test component of

the results section contains two elements of information. The “Between-Subjects”

components indicate that the independent variable was manipulated between two or more

differing groups. One group served as the experimental group, and a different group

served as the control group. There are primary distinctions between this type of

manipulation and one that manipulates the independent variable within-subjects. The

within manipulation of the independent variable uses the same group of participants in

both conditions. These two distinctive designs require two distinctive analyses as well as

two distinctive analysis strategies. This MANOVA test identifies whether different

levels of the independent variables have a significant effect on a linear combination of

each of the dependent variables. There are interactions between the independent variable

and a linear combination of the dependent variables. A significant univariate effect

existed for each of the dependent variables. Other important factors within this analysis

are the relevance of significance or Sig. If Sig. is greater than 0.05 the researcher must

fail to reject the null Hypothesis. If 0.05 ≥ Sig. > 0.01, the researcher must reject the null

Hypothesis and accept the alternative Hypothesis and it is considered statistically

significant (Confidence Level 95 %). If Sig. is less than or equal to 0.01, the researcher

must reject the null Hypothesis accept the alternative Hypothesis and the study is

considered statistically significant (Confidence Level 99 %).

In the assessment of Tables 6 and 7, the results determined that Wilks’ Lambda is

significant. This indicates a direct measure of the proportion of variance in the

combination of dependent variables and degrees of freedom statistically. One-way

116

MANOVA was calculated examining the effect of the number of CLABSI infection,

infection rate, and bed capacity. A significant effect was found (Lambda (70, 6) = .021,

Sig. = .923) indicating a high correlation of activity between all variables in this analysis.

Table 8. Data Correlation of the Infection Rate to Hospital Bed Capacity

Infection Rate/1000

Central Line Days

Bed Capacity

Infection Rate/1000 Central Line Days

Pearson Correlation Sig. (2-tailed) N

1 -- 40

.153

.345 40

Bed Capacity Pearson Correlation Sig. (2-tailed) N

.153

.345 40

1 -- 63

In Table 8, the data was evaluated for straight-line relationships between two sets

of variables. Correlations can range between -1 (perfect negative) and +1 (perfect

positive), with 0 indicating no straight-line relationship. Correlation measures the extent

that variables co-vary, and the dependence of one variable on another. Correlation also

describes how effectively values of one variable can predict values of another variable.

Essentially, Pearson describes the strength of the linear relationship. A Pearson

correlation coefficient was calculated for the relationship between subjects’ bed-capacity

and infection rate per 1000 central line days. A weak positive correlation was found (r

(38) =.153), indicating a weak linear relationship between the two variables. Essentially,

even though a weak correlation indicated a relationship between bed capacity and

infection rate.

117

Data analysis response to Hypothesis 2.

Table 9. Multivariate Test to Define Correlations Between Monitoring (Surveillance) Methods, Bed Capacity, and CLABSI

Effect Value F Hypothesis df Error df

Sig.

Intercept Wilks’Lambda Roy’s Largest Root Pillai’s Trace

.001 700.783 .999

3503.915 3503.915 3503.915

2.000 2.000 2.000

10.000 10.000 10.000

.000

Bed cap. Wilks’Lambda Roy’s Largest Root Pillai’s Trace

.000 330.301 1.941

16.215 77.304 7.754

94.000 47.000 94.000

20.000 11.000 22.000

.000

Multivariate test help define statistical parameters identified when surveillance

activities are conducted. Surveillance actions were evaluated in all hospitals. Wilks’

Lambda, Roy’s Largest Root, and Pillai’s Trace were used to obtain different views on

the data set. This helped determine if random assignment of data to levels of one or more

nominal independent variables were measurable. In this study, all hospitals are measured

on several continuous dependent variables. This study uses multivariate testing to

determine if the data form linear combinations of the dependent variables, this determines

if the data are discriminate among the groups in the particular experimental design (Field,

2009). Simply stated a linear combination is the sum of some set of ordered pairs in an

analysis in this example, various surveillance measures including hours of surveillance,

bed capacity, integration of infection control practitioners and CLABSI. One-way

MANOVA was calculated examining the effect of hours of surveillance and Infection

Control Practitioners and bed-capacity. A significant effect was found (Lambda (94, 20)

118

= .000, Sig. = .000) indicating a positive correlation or relationship exist between these

variables.

Table 10. Comparative Data to Evaluate Effectiveness of Increased Numbers of Infection Control Practitioners and Hours of Surveillance Activities Source Dependent

Variable Type III Sum of Squares

df Mean Square

F Sig.

Corrected Model

ICPs Hours of Surveillance

58.323 47592.712

47 47

1.241 1012.611

75.834 15.664

.000 .000

Intercept ICPs Hours of Surveillance

125.929 46196.121

1 1

125.929 46196.121

7695.676 713.704

.000 .000

Bed capacity

ICPs Hours of Surveillance

58.323 47592.712

47 47

1.241 1012.611

75.834 15.664

.000 .000

Error ICPs Hours of Surveillance

.180 712.000

11 11

.016 64.727

-----

-------

-----

------

Total ICPs Hours of Surveillance

188.270 93006.000

59 59

-------- -------

-------- ---------

------ ------

Corrected Total

ICPs Hours of Surveillance

58.503 48304.712

58 58

-------

---------

--------

------

-------

Comparison data assists the researcher to evaluate effectiveness of measurable

data. In the study, the number of infection control practitioners and the number of hours

of surveillance that was conducted in the various hospital groups was compared to the

incidence and prevalence of central line infections. It may be determined that comparison

studies find unexpected effects in the experiment (Leedy & Ormrod, 2010). Follow-up

119

univariate ANOVAs indicated that rates were improved with more hours of surveillance

(F (47, 56) =75.835, Sig. = .000).

Table 11. Data Correlation Between Infection Rate, Bed Capacity, and Infection Control Practitioners (ICP)

Infection

Rate/1000 Central Line Days

Bed Capacity ICPs

Infection Rate/1000 Central Line Days

Pearson Correlation Sig. (2-tailed) N

1 40

.153

.345 40

.054

.751 37

Bed Capacity Pearson Correlation Sig. (2-tailed) N

.153 .345 40

1 63

.747 .000 60

ICPs Pearson Correlation Sig. (2-tailed) N

.054 .751 37

.747 .000 60

1 ---- 60

A Pearson’s correlation was calculated examining the relationship between

infection rate per 1000 central line days, bed capacity, and infection control practitioners.

A positive correlation was found (r (35) = .153, Sig. <.001) among all variables

indicating a link between infection rate, bed capacity and the role of infection control

practitioners in minimizing infection rate.

120

Table 12. Multivariate Test to Determine Infection Control Practitioners and Bed Capacity

Effect Value F Hypothesis df Error df

Sig.

Intercept Wilks’Lambda Roy’s Largest Root Pillai’s Trace

.9541 20.730 .954

114.015 114.015 114.015

2.000 2.000 2.000

11.000 11.000 11.000

.000

.000 Bed cap. Wilks’Lambda Roy’s Largest Root

.043 4.284 1.583

.840 1.028 .912

100.000 50.000 100.000

22.000 12.000 24.000

.726

.513

.639

An analysis of bed-capacity and integration of infection control practitioners were

evaluated. One-way ANOVA was calculated examining the effect of the infection

control practitioners on bed-capacity. A significant effect was found (Lambda (100, 22)

= .043, Sig. = .726) indicating a positive correlation between integration of infection

control practitioners to bed capacity.

Table 13. Comparative Data to Evaluate Effectiveness of Surveillance and Therapeutic Activities to CLABSI Events

Source Dependent Variable Type III Sum of Squares

df Mean Square F Sig.

Corrected Model

Direct Observation of Procedure Staff Authorized to Stop a Procedure for Non-Compliance

8.413 13.629

50 50

.168 .273

1.010 .861

.529 .665

Intercept Direct Observation of Procedure Staff Authorized to Stop a Procedure for Non-Compliance

40.022 10.287

1 1

40.022 10.287

240.130 32.484

.000 .000

121

Table 13. continued Comparative Data to Evaluate Effectiveness of Surveillance and Therapeutic Activities to CLABSI Events

Source Dependent Variable Type III

Sum of Squares

df Mean Square F Sig.

Bed capacity

Direct Observation of Procedure Staff Authorized to Stop a Procedure for Non-Compliance

8.413 13.629

50 50

.168 .273

1.010 .861

.529 .665

Error Direct Observation of Procedure Staff Authorized to Stop a Procedure for Non-Compliance

2.000 3.800

12 12

-----

Total Direct Observation of Procedure Staff Authorized to Stop a Procedure for Non-Compliance

55.000 29.000

63 63

-----

----

-----

Corrected Total

Direct Observation of Procedure Staff Authorized to Stop a Procedure for Non-Compliance

10.413 17.429

62 62

----

-----

Univariate ANOVAs indicated that infection rate was improved when enhanced

surveillance and therapeutic measures were incorporated into the treatment paradigm.

Direct observation of procedure = (F (50, 60) = 1.010, Sig. = .000). Staff Authorized to

Stop a Procedure for Non-Compliance = (F (1, 61) = 32.484, Sig. = .000). Evidence

suggests a positive correlation exists between the staff’s ability to halt a procedure if non-

compliance with aseptic techniques was observed, the direct observation of procedures to

bed capacity.

122

Table 14. Data Correlation to Evaluate Effectiveness of Surveillance and Therapeutic Activities to CLABSI Events

Infection

Rate/100 0 Central Line Days

Bed Cap.

Direct Observatio n

Staff is Authorized to Stop Procedure for Non- Compliance

Use Check List for Insertion

Syndromic Surveillance System

Infection Rate/1000 Central Line Days

Pearson Correlation Sig. (2-tailed) N

1 40 40

.153 .345 40

-.025 .876 40

-.117 .473 40

--------- --------- ---------

--------

------- -------

Bed Capacity

Pearson Correlation Sig. (2-tailed) N

.153 .345 40

1 63 ----

.112 .383 63

.208 .102 63

--------- --------- ---------

-------

------- -------

Direct Observation of Procedure

Pearson Correlation Sig. (2-tailed) N

-----

----- -----

-----

------ ------

1 63

.244 .054 63

--------- --------- ---------

-------

-------- --------

Staff is Authorized to Stop Procedure for Non- Compliance

Pearson Correlation Sig. (2-tailed) N

----- ------ -----

------ ------ -----

.244 .054 63

1 ----- 63

--------- ---------

--------

------

-------

Use Check List for Insertion

Pearson Correlation Sig. (2-tailed) N

-.005 .977 40

.190 .135 63

---- ----- -----

----

---- -----

1 --- 63

.041 .748 63

Syndromic Surveillance System

Pearson Correlation Sig. (2-tailed) N

.054 .738 40

-.297 .018 63

------

--------- --------

------

-------

.041 .748 63

1 ---- 63

A Pearson’s correlation was calculated examining the relationship between

infection rate per 1000 central line days, bed capacity, direct observation of the

procedure, the staff’s ability to stop the procedure for non-compliance with hygienic

measures and the use of checklist during the insertion process. In the correlation between

123

infection rate and direct observation of the procedure, a weak negative correlation was

found (r (38) = -.025. Sig. >.05.) and Sig. (2-tailed) being greater than .05. The

indications are that minimal significance between the correlations existed between these

two variables. In the correlation between infection rate and the staff’s ability to stop a

procedure due to failure of compliance, a weak negative correlation was found (r (38) = -

.117. Sig. >.05.) and Sig. (2-tailed) being greater than .05 indications are that, no

statistical significance between the correlations existed between these two variables. Bed

capacity correlation was significant with direct observation of therapy a positive

correlation existed when (r (61) = .112. Sig. >, 05 and Sig. (2 tailed) being greater than

.05 indicates no correlation and significance. No correlation existed between bed capacity

and staff authorized to stop a procedure when (r (38) = .208, Sig. < .05 and Sig. (2 tailed)

being greater than .05. When evaluating the correlation between checklist use and

infection rate a (r (38) = -.005 and a Sig. (2-tailed) = .977. This indicated that there is no

statistically significant correlation between the two variables. That means, increases or

decreases in one variable does not significantly relate to increases or decreases in the

second variable. Regarding syndromic surveillance and infection rate, (r (38) = .054. Sig.

< .05 and Sig. (2 tailed) being less than or equal to .05 indicates a statistically significant

correlation and significance. Syndromic surveillance and bed capacity indicated (r (61) =

.018. Sig. < .05 and Sig. (2 tailed) being greater than.05 indicates a statistically

insignificant correlation and significance. Analysis indicates that dependent on the size of

the hospital impacts the number of personnel involved with various surveillance and

therapeutic measures. Minimal or weak levels of significance indicated some positive

correlation in decreasing CLABSI infection.

124

Table 15. Multivariate Test to Determine Effectiveness of Surveillance and Therapeutic Activities to CLABSI Events

Effect Value F Hypothesis df Error

df Sig.

Intercept Wilks’Lambda Roy’s Largest Root Pillai’s Trace

.004 ----- .996

165.941 --------- 165.941

3.000 3.000 3.000

2.000 2.000 2.000

.006

.006 Bed Cap. Wilks’Lambda Roy’s Largest Root Pillai’s Trace

.002 27.342 2.457

.451 3.125 .517

105.000 35.000 105.000

6.894 4.000 12.000

.961

.137

.962

An analysis of bed-capacity and methods of effective surveillance and therapeutic

activities were evaluated. In this category of surveillance and therapeutic measures,

direct observation of procedures, the ability for the staff to stop a procedure for non-

compliance, utilization of check list during insertion, and the integration of an enhance

surveillance system connected to a laboratory was analyzed. One-way ANOVA was

calculated examining the effect of the integration of surveillance and therapeutic

measures on bed-capacity. A significant effect was found (Lambda (100, 22) = .043, Sig.

= .726) indicating a positive correlation in the decrease of infection with various

surveillance and therapeutic measures in place.

125

Table 16. Comparison Data to Determine Effectiveness of Surveillance and Therapeutic Activities to CLABSI Events

Source Dependent Variable Type III Sum of Squares

df Mean Square F Sig.

Corrected Model

Direct Observation of Procedure. Staff Authorized to Stop a Procedure for Non-Compliance. Infection Rate/1000 Central Line Days. Use Check List. Syndromic Surveillance.

3.100 9.900 131.575 9.783 8.533

35 35 35 47 47

.089 .283 3.759 .208 .182

.709 .566 .335 .781 1.210

.749 .843 .968 .739 .377

Intercept Direct Observation of Procedure. Staff Authorized to Stop a Procedure for Non-Compliance. Infection Rate/1000 Central Line Days. Use Check List. Syndromic Surveillance.

31.066 11.765 334.850 27.712 35.570

1 1 1 1 1

31.066 11.765 334.850 27.712 35.570

248.529 23.529 29.878 103.920 237.136

.000 .008 .005 .000 .000

Bed Cap. Direct Observation of Procedure. Staff Authorized to Stop a Procedure for Non-Compliance. Infection Rate/1000 Central Line Days. Use Check List. Syndromic Surveillance.

3.100 9.900 131.575 9.783 8.553 -------

------

35 35 35 47 47 ----

.089 .283 3.759 .208 .182 ------

------

.709 .566 .335 .781 1.210 ------

-------

.749 .743 .968 .739 .377 ----- ------

126

Table 16. continued Comparison Data to Determine Effectiveness of Surveillance and Therapeutic Activities to CLABSI Events Source Dependent Variable Type III

Sum of Squares

df Mean Square F Sig.

Error Direct Observation of Procedure. Staff Authorized to Stop a Procedure for Non-Compliance. Infection Rate/1000 Central Line Days. Use Check List. Syndromic Surveillance.

.500 2.000 44.829 3.200

4 4 4 12 12

.125 .500 11.207 .267 .150

-------

------- ------- ------ ------

------

------- ------ ------

Total Direct Observation of Procedure. Staff Authorized to Stop a Procedure for Non-Compliance. Infection Rate/1000 Central Line Days. Use Check List. Syndromic Surveillance

36.000 24.000 550.887 41.000 52.000

40 40 40 60 60

------ ------ ------ ------ ------

------

------ ------ ------ ------

-----

------ ------ ------ ------

Corrected Total

Direct Observation of Procedure. Staff Authorized to Stop a Procedure for Non-Compliance. Infection Rate/1000 Central Line Days. Use Check List. Syndromic Surveillance.

3.600 11.900 176.404 12.983 10.333

39 39 39 59 59

------

----- ----- -----

------

------ ------- ------- -------

------

------ ------ ------ ------

127

Comparison data assists the researcher when desiring to evaluate effectiveness of

measurable data. In the study, direct observation of procedures, the ability for staff

members to stop a procedure for non-compliance, the use of check lists and integration of

surveillance systems conducted in the various hospital bed capacity groups was compared

to the incidence and prevalence of central line infections. It may be determined that

comparison studies find unexpected effects in the experiment (Leedy & Ormrod, 2010).

Follow-up univariate ANOVAs indicated that rates were improved with more hours of

surveillance, use of direct observation techniques, use of check lists, and the ability to

stop therapies due to non-compliance (F(47,56)=75.835, Sig. = .000).

Table 17. Methodological Processes to Evaluate CLABSI, Clinical Interventions and Hospital Bed Capacity

Effect Value F Hypothesis df Error df

Sig.

Intercept Wilks’Lambda Roy’s Largest Root Pillai’s Trace

.001 729.979 .999

2433.265 2433.265 2433.265

3.000 3.000 3.000

10.000 10.000 10.000

.000

.000 Bed Cap. Wilks’Lambda Roy’s Largest Root Pillai’s Trace

.000 325.525 2.563

4.032 83.113 1.498

141.000 47.000 141.000

30.899 12.000 36.000

.000

An analysis of bed-capacity and methods of effective surveillance and therapeutic

processes were evaluated. In this category, surveillance and therapeutic measures, direct

observation of procedures, the ability for the staff to stop a procedure for non-

compliance, utilization of check lists during insertion, and the integration of an enhance

surveillance system connected to a laboratory was analyzed and the correlation to central

line infection. One-way ANOVA was calculated examining the effect of the integration

128

of surveillance and therapeutic measures on bed-capacity. A significant effect was found

(Lambda (141, 30.899), Sig. = .000). Indicating a positive correlation between these

therapeutic and surveillance measures, bed-capacity and decreased numbers of central

line infections.

Table 18. Methodological Processes to Evaluate Intercepts, Infection rate, and Bed Capacity and their Effects on Decreased Rates of Central Line Infection

Effect Value F Hypothesis df Error df

Sig.

Intercept Wilks’Lambda Roy’s Largest Root Pillai’s Trace

------ ------ ------

----- ----- -----

5.000 5.000 5.000

---- ---- ----

----- ----- -----

Infection Wilks’Lambda Rate/1000 Roy’s Largest Root Days Pillai’s Trace

------ ------ ------

----- ----- -----

5.000 5.000 5.000

---- ---- ----

----- ----- -----

Bed Cap. Wilks’Lambda Roy’s Largest Root Pillai’s Trace

.000 4.412 2.000

1.930 4.136 .063

160.000 160.000 160.000

.224 3.000 15.000

.075

.000 1.000

Analyses of bed-capacity and infection rates were evaluated in their relationship

to central line. One-way ANOVA was calculated examining the effect of the integration

of infection rates on bed-capacity. A significant effect was found (Lambda (160, 3) =

4.412, Sig. = .000). This analysis revealed a direct correlation of this variable and

statistical significance of infection rate to bed capacity.

129

Table 19. Statistical Analysis between CLABSI and Hand Hygienic Measures

Source Dependent Variable Type III Sum of Squares

df Mean Square

F Sig.

Corrected Model

Hours Monitors Optimal catheter site. ICP Hand Hygiene Hand Hygiene Daily Hand Hygiene Weekly Hand Hygiene Monthly Hand Hygiene Quarterly Hand Hygiene Unknown

41221.563 4.876 7.742 5.027 7.273 8.802 3.312 4.324

33 33 33 33 33 33 33 33

1249.138 .148 .235 .152 .220 .267 .100 .131

18.063 .158 .354 5.800 1.448 1.814 1.178 3.223

.017

.998 .946 .000 .436 .349 .523 .000

Intercept Hours Monitors Optimal catheter site. ICP Hand Hygiene Hand Hygiene Daily Hand Hygiene Weekly Hand Hygiene Monthly Hand Hygiene Quarterly Hand Hygiene Unknown

5941.295 4.357 2.323 .137 1.557 .626 .062 .095

1 1 1 1 1 1 1 1

5941.295 4.357 2.323 .137 1.557 .626 .062 .095

85.913 4.668 3.505 5.226 10.228 4.255 .722 2.345

.003

.120 .158 .000 .049 .131 .458 .000

Bed Cap. Hours Monitors Optimal catheter site. ICP Hand Hygiene Hand Hygiene Daily Hand Hygiene Weekly Hand Hygiene Monthly Hand Hygiene Quarterly Hand Hygiene Unknown

40827.974 4.820 7.525 5.022 7.056 8.132 2.993 4.303

32 32 32 32 32 32 32 32

1275.874 .151 .235 .157 .220 .254 .094 .134

18.450 .161 .355 5.975 1.448 1.728 1.098 3.308

.017

.998 .945 .000 .435 .367 .553 .000

Infection Rate/ 1000 Central Line Days

Hours Monitors Optimal catheter site. ICP Hand Hygiene Hand Hygiene Daily Hand Hygiene Weekly Hand Hygiene Monthly Hand Hygiene Quarterly Hand Hygiene Unknown

17.036 .200 .012 .000 .543 .059 .244 .000

1 1 1 1 1 1 1 1

17.036 .200 .012 .000 .543 .059 .244 .000

.246

.214 .018 .000 3.568 .401 2.868 .000

.654

.675 .901 1.00 .165 .572 .189 1.00

130

Table 19. continued Statistical Analysis between CLABSI and Hand Hygienic Measures

Source Dependent Variable Type III Sum of Squares

df Mean Square

F Sig.

Error Hours Monitors Optimal catheter site. ICP Hand Hygiene Hand Hygiene Daily Hand Hygiene Weekly Hand Hygiene Monthly Hand Hygiene Quarterly Hand Hygiene Unknown

207 464 2.8 1.988 7.879 .457 .441 .256 1.220

3 3 3 3 3 3 3 3

69.155 .933 .663 2.626 .152 .147 .085 4.065

----- ----- ---- ---- ---- ---- ---- ----

---- ---- ---- ---- ---- ---- ---- ----

Total Hours Monitors Optimal catheter site. ICP Hand Hygiene Hand Hygiene Daily Hand Hygiene Weekly Hand Hygiene Monthly Hand Hygiene Quarterly Hand Hygiene Unknown

86335.000 32.000 28.000 6.000 11.000 19.000 4.000 5.000

37 37 37 37 37 37 37 37

------ ------ ----- ----- ---- ----

---- ---- ---- ---- ---- ----

Corrected Total

Hours Monitors Optimal catheter site. ICP Hand Hygiene Hand Hygiene Daily Hand Hygiene Weekly Hand Hygiene Monthly Hand Hygiene Quarterly Hand Hygiene Unknown

41429.027 7.676 9.730 5.027 7.730 9.243 3.568 4.324

36 36 36 36 36 36 36 36

--- --- --- --- --- --- --- ---

Comparison data assisted the researcher when evaluating effectiveness of

measurable data. In the study, the number of infection control hours of surveillance, the

integration of optimal catheter site placement, use of infection control practitioners to

monitor hand hygiene on a daily, weekly, monthly, and quarterly basis was conducted in

the various hospital groups and compared to the incidence and prevalence of central line

infections. Univariate ANOVAs indicated that infection rates were improved with more

hours of surveillance (F (1, 3) =.246, Sig. = .654) and daily surveillance of hand hygiene

131

measures (F (1, 3) =3.568, Sig. = .165). Hand hygiene and daily surveillance

methodologies were identified as correlations in the data. Increased hours of surveillance

(F (32, 18.45) =. 40827.974, Sig. = .017 and (F (32, 5.975) = 5.022, Sig. = .000) for daily

assessment of hand hygiene. These statistics indicated a strong correlation of the

variables and statistical significance that indicated that increased activity by an infection

control practitioner may decrease the number of central line infection by employing

recognition strategies to minimize the risks associated with insertion, management, and

care of the line site as well as enhanced hand washing techniques.

Table 20. Correlation between Rates, Surveillance, and Review

Infection Rate

Bed Cap.

Hours of Surveillance

Daily Review of Line Necessity

Monitor Optimal Catheter Selection Site

Infection Rate Pearson Correlation

1 .153 -.097 .058 -.057

Sig. (2- tailed)

---- .345 .566 .720 .726

N 40 40 40 40 40

Bed Cap. Pearson Correlation

.153 1 .588 .256 .048

Sig. (2- tailed)

.345 63 .000 .043 .707

N 40 59 63 63

Hours of Surveillance

Pearson Correlation

-.097 .588 1 .125 -.068

Sig. (2- tailed)

.566 .000 ---- .344 .609

N 37 59 59 59 59

132

Table 20. continued Correlation between Rates, Surveillance, and Review

Infection Rate

Bed Cap.

Hours of Surveillance

Daily Review of Line Necessity

Monitor Optimal Catheter Selection Site

Daily Review of Line Necessity

Pearson Correlation

.058 .256 .125 1 .258

Sig. (2- tailed)

.726 .707 .609 .041 ----

N 40 63 59 63 63

Monitor Optimal Catheter Selection Site

Pearson Correlation

-.057 .048 -.068 .258 1

Sig. (2- tailed)

.726 .707 .609 .041 ----

N 40 63 59 63 63

A Pearson’s correlation was calculated examining the relationship between

infection rate per 1000 central line days, bed capacity, hours of surveillance, daily review

of line necessity, and monitoring of optimal catheter selection site during the insertion

process. In the correlation between infection rate and hours of surveillance, a weak

negative correlation was found (r (35) = -.097). With Sig. (2-tailed), being greater than

.05 indicates that minimal significance between the correlations existed between these

two variables. In the correlation between infection rate and the daily review of central

line necessity, a weak negative correlation was found (r (38) = .058). With Sig. (2-tailed)

being greater than .05 indicates are that, no statistical significance between the

correlations existed between these two variables. Infection rate and monitoring of

133

catheter selection site revealed a (r (61) = .112) and Sig. (2 tailed) being greater than .05

indicates no correlation and significance. Bed capacity and hours of surveillance

indicated (r (57) = .588) and Sig. (2 tailed) being equal to .000. This demonstrated a

statistically significant correlation between these two variables. When evaluating the

correlation between bed capacity and daily review of central line necessity a (r (61) =

.256) and a Sig. (2-tailed) = .043. This indicated that there was a statistically significant

correlation between the two variables. Bed capacity and monitoring optimal catheter

selection site identified a (r (61) = .048) and a Sig. (2-tailed) = .707 indicated that there

was no significance.

Table 21. Correlation between Infection Rates, and Hand Hygiene Evaluation

Infection Rate

Bed Cap.

Hand Hygiene Daily

Hand Hygiene Weekly

Hand Hygien e Month- ly

Hand Hygiene Quarter- ly

Infection Rate

Pearson Correlation

1 .153 .054 -.184 .305 .219

Sig. (2- tailed)

---- .345 .054 -.184 .305 .219

N 40 40 40 40 40 40

Bed Cap. Pearson Correlation

.153 1 .297 .324 -.345 -.127

Sig. (2- tailed)

.345 ---- .018 .010 .006 .322

N 40 63 63 63 63 63

134

Table 21. continued Correlation between Infection Rates, and Hand Hygiene Evaluation

Infection Rate

Bed Cap.

Hand Hygiene Daily

Hand Hygiene Weekly

Hand Hygien e Month- ly

Hand Hygiene Quarter- ly

Hand Hygiene Weekly

Pearson Correlation

.054 .297 1 .146 -.233 .193

Sig. (2- tailed)

.742 .018 ---- .254 .078 .130

N 40 63 63 63 63 63

Hand Hygiene Monthly

Pearson Correlation

.305 -.345 -.223 -.285 1 -.003

Sig. (2- tailed)

.055 .006 .078 .023 ---- .978

N 40 63 63 63 63 63 Hand Hygiene Quarterly

Pearson Correlation

.219 -.127 -.035 -.003 -.043 -.043

Sig. (2- tailed)

.174 .322 .784 .978 .740 .740

N 40 63 63 63 63 63

A Pearson’s correlation was calculated examining the relationship between

infection rate per 1000 central line days, bed capacity, and hand hygiene at varying

intervals was examined in this analysis. In the correlation between infection rate and

daily hand hygiene, a positive correlation was found (r (38) = .054). With Sig. (2-tailed)

of .054 being slightly greater than .05, minimal significance between the correlations

existed between these two variables. In the correlation between infection rate and

monthly hand hygiene, no statistically significant correlation was found (r (38) = .305).

135

With Sig. (2-tailed) being greater than .05 indications are that, no statistical significance

between the correlations existed between these two variables. Infection rate and hand

hygiene quarterly revealed a (r (38) = -.127) and Sig. (2 tailed) being greater than .05

indicates no correlation and no significance. Bed capacity and daily hand hygiene (r (61)

= .297), and Sig. (2 tailed) being equal to .018. This found a statistically significant

correlation between these two variables. When evaluating the correlation between bed

capacity and weekly review of hand hygiene a (r (61) = .324 and a Sig. (2-tailed) = .010.

This indicated that there is statistically significant correlation between the two variables.

Bed capacity and monitoring of monthly hand hygiene identified a (r (61) = -.345 and a

Sig. (2-tailed) = .006 indicated a statistical significance between the two variable. In the

bed capacity and hand hygiene monitoring on a monthly basis identified a (r (61) = -.127

and a Sig. (2-tailed) = .322. These statistics indicate that there is no statistical

significance between these two variables.

Data analysis response to Hypothesis 3.

Table 22. Statistical Analysis between CLABSI Rate and Clinical Interventions

Source Dependent Variable

Type III Sum of Square s

df Mean Square s

F Sig.

Corrected Model

Daily Review of Line Necessity

7.975 36 .222 8.388 .000

Monitors Optimal Site Selection

5.600 36 .156 .167 .998

Chlorhexadine Skin Asepsis

3.173 36 .088 .619 .797

Max. Barrier Protection upon Insertion

2.402 36 .067 .134 .999

136

Table 22. continued Statistical Analysis between CLABSI Rate and Clinical Interventions Source Dependent

Variable Type III Sum of Squares

df Mean Squares

F Sig.

Intercept Daily Review of Line Necessity

.347 1 .347 1.315 .000

Monitors Optimal Site Selection

4.533 1 4.533 4.857 .115

Chlorhexadine Skin Asepsis

1.003 1 1.003 7.043 .077

Max. Barrier Protection upon Insertion

3.956 1 3.956 7.923 .067

Infection Rate/1000 Central Line Days

Daily Review of Line Necessity

.000 1 .000 .000 1.000

Monitors Optimal Site Selection

.200 1 .200 .214 .675

Chlorhexadine Skin Asepsis

1.073 1 1.073 7.534 .071

Max. Barrier Protection upon Insertion

.002 1 .002 .004 .951

Bed Capacity

Daily Review of Line Necessity

7.948 35 .277 8.598 .000

Monitors Optimal Site Selection

5.573 35 .159 .171 .998

Chlorhexadine Skin Asepsis

2.877 35 .082 .577 .882

Max. Barrier Protection upon Insertion

2.395 35 .068 .137 .999

137

Table 22. continued Statistical Analysis between CLABSI Rate and Clinical Interventions Source Dependent

Variable Type III Sum of Squares

df Mean Squares

F Sig.

Error Daily Review of Line Necessity

7.923 3 2.641 --- ---

Monitors Optimal Site Selection

2.800 3 .933 --- ---

Chlorhexadine Skin Asepsis

.427 3 .142 --- ---

Max. Barrier Protection upon Insertion

1.498 3 .499 --- ---

Total Daily Review of Line Necessity

11.000 40 --- --- ---

Monitors Optimal Site Selection

34.000 40 --- --- ---

Chlorhexadine Skin Asepsis

36.000 40 --- --- ---

Max. Barrier Protection upon Insertion

40.000 40 --- --- ---

Corrected Total

Daily Review of Line Necessity

7.975 39 --- --- ---

Monitors Optimal Site Selection

8.400 39 --- --- ---

Chlorhexadine Skin Asepsis

3.600 39 --- --- ---

Max. Barrier Protection upon Insertion

3.900 39 --- --- ---

Comparison data assists the researcher when evaluating effectiveness of

measurable data. In the study, the CLABSI rate and therapeutic measures that were

conducted in the various hospital groups was compared to the incidence and prevalence

of central line infections. It may be determined that comparison studies find unexpected

138

effects in the experiment (Leedy & Ormrod, 2010). Univariate ANOVAs indicated that

rates were improved with daily review of central line necessity (F (35, .277) =7.988).

Table 23. Correlation between CLABSI rate and Clinical Interventions

Bed Cap.

Infection Rate/1000 CLD

Chlorhexidine Skin Asepsis

Max. Barrier Precautions Upon Insertion

Bed Cap. Pearson Correlation

Sig. (2 Tailed)

.153

.345

.074

.564

.039

.763

63 Infection Rate Pearson

Correlation

Sig. (2 Tailed)

.153

.345

---

.287

.073

.042

.796

40 Chlorhexidine Pearson

Correlation Sig. (2 Tailed)

.074

.564

.287

.073

---

.643

.000

63 Maximal Barrier

Precautions upon Insertion

Pearson Correlation

Sig. (2 Tailed)

.039

.763

.042

.796

.643

.000

---

A Pearson’s correlation was calculated that examined the relationship between

infection rate per 1000 central line days, use of chlorhexidine and maximal barrier

precautions upon insertion in this analysis. In the correlation between infection rate and

chlorhexidine use, a positive correlation was found (r (38) = .287). With Sig. (2-tailed) of

.073 being slightly greater than .05. An indication that minimal significance between the

correlations existed between these two variables. In the correlation between infection

139

rate and maximal barrier precautions upon insertion, no statistically significant

correlation was found (r (61) = .039). With Sig. (2-tailed) of .039 being greater than .05,

indications are that no statistical significance between the correlations existed between

these two variables. Bed capacity and chlorhexidine use indicates a (r (61) = .039) and

Sig. (2 tailed) being greater than .05 indicates no correlation and significance. Bed

capacity and maximal barrier precaution usage finds (r (61) = .297) and Sig. (2-tailed)

being equal to .763. No statistically significant correlation existed between these two

variables.

Hypotheses Data Analysis

Central line associated bloodstream infections and correlations to bed capacity

and categories were calculated and stratified with the use of MANOVA. The study

employed the use of multivariate statistics, correlational evaluation, and the use of

Wilks’Lambda, Roy’s Largest Root, and Pillai’s Trace as recommended by Statistical

Decision Tree Software™ (Andrews, et. al., 1981). The use of Factorial MANOVA was

selected since this MANOVA application is super-factorial. This is appropriate since the

data are integrated under multiple treatment combinations or multiple independent

variables and are measured using several continuous dependent variables. When

MANOVA is applied in this manner, it maintains the advantages that factorial designs

have over simple one-way designs and multiple composite dimensions of analysis may be

formed specific to each effect in the design, which optimally separate the groups being

evaluated for each effect (Field, 2009). In other words, factorial MANOVA involves the

calculation of several sets of composite variables and each set is specific to a particular

140

effect. In this study, hospital participants are nested under multiple treatment

combinations since multiple independent variables are measured using multiple

continuous dependent variables. This study design provided advantages that factorial

designs have over simple one-way designs with the high volume of variables. These

multiple variable developed a composite dimensions specific to each desired effect. The

design optimally separates the groups being evaluated for each effect.

Use of Statistical Decision Tree™ (Andrews, et al., 1981) suggested that best

analysis of the data be conducted utilizing the multivariate test of Wilks’ Lambda, Roy’s

Largest Root, and Pillai’s Trace because the variation in both dependent and independent

variables. Wilks’ Lambda was used since the data was a probability distribution and is

commonly used in multivariate hypothesis testing, especially with regard to the

likelihood-ratio test. Field (2009) indicates this can be interpreted as the proportion of the

variance in the outcomes that is not explained by an effect. In this study, Roy’s Largest

Root is utilized because it behaves differently from the other three test statistics. In

instances where the other three are not significant and Roy's is significant, the effect

should be considered not significant. Pillai’s Trace is the sum of the proportion of

unexplained variance on the discriminant function variates of the data (Field, 2009).

In this study, Tests of Between-Subject Effects was employed. This MANOVA

procedure identified whether different levels of the independent variables have a

significant effect on a linear combination of each of the dependent variables. There are

interactions between the independent variable and a linear combination of the dependent

variables. Significant univariate effects existed for each of the dependent variables.

141

Summary

This study determined how the sentinel year of HIDA documentation of CLABSI

in hospitals in the State of South Carolina correlated surveillance and therapeutic

measures and decrease of infection by hospital bed capacity. The following findings

emerged from the results of the data analysis. Significant variance existed in the rates of

CLABSI reported in hospitals with capacities of 50 or less, 51 to 200, 201 to 500, and

501 beds and more and the data collected provides an opportunity to identify correlative

patterns within the data. Correlation between monitoring and compliance methods and

incidence rates of CLABSI infection also existed and this research provided insight to

these therapeutic modalities and clinical surveillance measures identified by HIDA as

potential sources of processes that may ameliorate the significance and impact of

CLABSI in hospitals. Relationships existed between an increased rate of CLABSI in

acute care hospitals and integration of prevention processes when included in their

facilities protocols. This study provided an opportunity to analyze these relationships in

the context of how and what measures may be employed to evaluate the efficacy of these

relationships. Data was statistically evaluated by a variety of measures as suggested by

Statistical Decision Tree™ and the text Discovering Statistics Using SPSS.

This chapter provided an opportunity for data analysis to answer research

questions and hypotheses developed while seeking to determine the significance of

central line associated bloodstream infection in hospitals in the State of South Carolina

over a span of one calendar year (2008). The year 2008 was significant to this study

since this is the first year of data collection and reporting of Hospital Infection Disclosure

Act (HIDA) data to the Division of Acute Disease Epidemiology (DADE) at the South

142

Carolina Department of Health and Environmental Control (SCDHEC). This data was

mandatorily reported and comprehensively analyzed. An emphasis of evaluation focused

on bed capacity of the responding hospitals to determine if facility size determines

quality of care of surveillance and therapeutic measures.

143

CHAPTER 5. RESULTS, CONCLUSIONS, AND RECOMMENDATIONS

Introduction

This chapter provides an opportunity to define problems that precipitated the

study and the ability to identify salient concepts that surround them and utilize scholarly

research questions to find solutions. The opportunity to interpret answers for these

research questions will provide health care executives, physicians, and infection control

practitioners’ with information to guide their processes in a desire to develop a better

informed health care team that can provide integrative solutions to surveillance and

treatment of CLABSI patients. Additional questions should be identified that will

provide more knowledge during this research process.

This study focused on the following research questions to determine if there was a

significant variance in the rates of central line associated bloodstream infections

(CLABSI) reported in hospitals with bed capacities of 50 or less, 51 to 200, 201 to 500,

and 501 beds and more. The research questions asked (1) is there a correlation between

monitoring and compliance methods and incidence rates of CLABSI infection? (2) Does

a relationship exist between an increased rate of CLABSI in acute care hospitals and

integration of prevention processes when included in their facilities’ protocols?

MANOVA was used to evaluate both dependent and independent variables. Categorical

predictors were anticipated and the data met assumptions for parametric test due to

normal distribution of data points, homogeneity of variance, interval, and independence.

The information used for this research was collected from 63 acute care hospitals in

South Carolina that mandatorily reported Hospital Information Disclosure Act (HIDA)

data via the National Healthcare Safety Network (NHSN) network for the purpose of data

144

collection and analysis. Data points collected were stratified to provide a true

representation of rural and metropolitan hospitals across the State of South Carolina.

While South Carolina has many health care facilities, this study focused on only

inpatient, acute care facilities and did not include rehabilitation hospitals, pediatrics

facilities, or behavioral health hospitals since the focus was specifically oriented to acute

care adult patients. This study utilized a quantitative research methodology and used a

correlational approach to determine the role that surveillance and therapeutic measures

had on central line associated bloodstream infections in adult patients in acute care

hospitals between January 1, 2008 and December 31, 2008. This data was collected and

analyzed by the use of Microsoft Excel and PASW 19 software to efficiently and

accurately evaluate the variables in question.

The conceptual basis for the proposed research was developed from hospital data

and from hospitals that indicate that hospitals with greater bed capacities provided care

differently than smaller healthcare organizations that might not have the opportunity to

provide advanced techniques. Infection control programs are important in the prevention

of these illnesses by developing clinical strategies that modify the risk factors for

hospital-acquired infection.

Summary of Results

Demographic Analysis

Sixty-three acute care hospitals provided HIDA data through the NHSN, of whom

nine (14.2%) were in the zero to 50-beds category, 30 (47.6%) were 51-200 beds, 19

(30.1%) were 201-500 beds and five (7.9%) fell into the greater than 500 bed category.

145

These health care facilities were equally distributed across the State of South Carolina

and represent community based acute care hospitals because they provide a wide array of

clinical services to their communities. These hospitals were selected because they

predominantly provide adult acute and chronic care and are not focused specifically on

pediatrics, rehabilitation of behavioral health that may not customarily provide care with

the use of central lines.

Hospital Population Data

Multiple factors were identified as valuable in determining rates of CLABSI on

the facility size and the multiple co-variate factors surrounding positive or negative effect

of surveillance and therapeutic measures. Central line days described the total number of

days a central line is in place for patients in critical care units. CLABSI Infection rate was

determined by dividing the total number of central line associated bloodstream infections

by the number of central line days for that individual facility. The Central Line

Utilization Ratio was derived by dividing the number of central line days by the number

of patient days identified by the health care facility. Standardized Infection Ratio (SIR)

was found to be useful by infection control personnel since this was a summary measure

used to compare the CLABSI experience among hospitals or in sub-groups. SIR was

calculated by dividing the number of CLABSI infections by the expected number of

CLABSI infections for that facility. A SIR of 1.0 means the observed number of

infections equals the calculated expectations of that facility for CLABSI. SIR of greater

than 1.0 means the infection rate was higher than was expected within that population of

patients. The difference above 1.0 was the percentage that infection rate exceeds the

146

population standards. Conversely, a SIR less than 1.0 indicates infection rates lower than

expected for that population (CDC, 2010).

In the 0-50 bed population group, only one hospital reported having a CLABSI

infection during the span of time January 1, 2008 to December 31, 2008. All hospitals in

that category reflected one infection with an Infection Rate of 6.62 and a SIR of 1.004.

The Confidence Interval was -6.3 to 19.5 at 95%. The number of central line days of all

hospitals in this grouping was very small with a Mean of 145.56 days, Median of 61

days, and Mode of zero. For this population of hospitals, a SIR ratio of 1.004 was

slightly higher than what was expected for patients in that hospital bed category.

In the 51-200 bed category, the average number of CLABSI infections for all

hospitals was 2.07, Median of 0.5 and Mode of 0. Regarding central lines days, a Mean

of 745.2 days and a Median of 532 days and no Mode was revealed. The infection rate of

this patient population was 1.965 and a Median of 0.36 and no Mode. SIR acknowledged

an average SIR of 1.902 and a Mean of 2.981. In this population grouping, the SIR of

1.92 was higher than expected for hospitals in this category.

The 201-500 bed category revealed an average number of CLABSIs for this

population of 6.63, a Median of 6.0 and a Mode of 6. Central line days reflected a Mean

of 3,331.9 days and a Median of 3,271 days, and no Mode is available. The Infection rate

in this population was 1.907 and a Median of 1.57, and a SIR average of 0.998 with a

Median of .9997 and a Mode of 1.001. SIR in the 201-500 bed population was 1.001 and

approximates what would be expected in this grouping of hospitals.

In the greater than 500 bed category the average number of CLABSI infections

rose to 19.2, a Median of 14 and no Mode was available. Central line days increased on

147

average to 4,221.6 days with a Median of 3,956 days and no Mode. The Infection rate

also increased with an average of 4.212, the Median number increased to 2.8. Lastly, the

SIR for this population of hospitals realized an average of 0.9996 with a Median of 0.999

and a Mode of 0.999. Even with a greater number of central line infections and more

central line days, a SIR ratio of 0.9996 was slightly lower than general expectations of

this population of hospitals.

Research Question 1 states there was a significant variance in the rates of

CLABSI reported in hospitals with capacities of 50 or less, 51 to 200, 201 to 500, and

501 beds and more. When utilizing Multivariate Test in PASW 19, a statistical analysis

of hospital categories and infection rate suggested that in an analysis of the variance that

F was 15.446 and was considered large while p value as identified by Significance is

relatively low at .026 and supports the decision to reject the null Hypothesis and accept

the alternate Hypothesis.

Hypotheses for Research Question 1 include:

H1: There is a significantly higher incidence of CLABSI observed in hospitals of

higher volume than lower bed capacities.

H2: There are significant differences between responsive measures used by

hospitals to minimize the effect of CLABSI when two or more full-time Infection Control

Practitioners (ICP) are employed for surveillance purposes.

H0 : There is a significantly lower incidence of CLABSI observed in hospitals of

higher bed capacities than in hospitals of lower bed capacities.

H0 : Hospital CLABSI rates are not dependent upon the number of full-time staff

and number of hours of infection control activities performed per month.

148

Hypothesis df at 2.000 reflects the degrees of freedom for the sample. Wilks’

Lambda, Roy’s Largest Root, and Pillai’s Trace were calculated for this data. An

understanding of the significance of these three tests was important to understand the

values. Field (2009) indicated that Pillai’s Trace was the sum of the proportion of

explained variances on the discriminant functions. This value approximates can be turned

into a value that has an approximate F-distribution. Wilks’ Lambda was the product of

the unexplained variance on each side of the variate. This tool represents the ratio of error

variance to total variance for each variate. When a Wilk’s Lambda value was small, the

test was considered statistically significant. Roy’s Largest Root represents the proportion

of explained variance to unexplained variance. Roy’s Largest Root represented the

maximum possible Between-Group differences for the data collected. In Table 5, all

factors indicated a degree of significance between bed capacity and infection rate of these

hospitals. In the evaluation of relationship between the response and predictor variables

as described by F value of 31.670 and Sig. of .005 a relationship exist among variances.

Table 7 confirmed this relationship as described by the Pearson Correlation with a scale

of -1.0 to 1.0 on the Pearson scale, the data represented indicated a positive relationship

with central line days and bed capacity of .153 and .345 respectively. The Sig. (2-tailed)

value signifies a statistically significant correlation when the value was greater than .05.

The value in this analysis was .345. The final analyses was indicative of bed capacity

and CLABSI rate that was determined by the number of CLABSIs as compared to the

number of central line days and resulting infection rate within the scope of a 95%

Confidence Interval and the Standardized Infection Rate (SIR).

149

In the 0 to 50 bed category, with an average number of central line days of

145.555 only one central line infection was observed. This revealed an infection rate of

6.62 and a SIR of 1.004. This represents a slight increase in infection rate for what was

expected for a health care institution of this size. In the 51 to 200 hospital bed category,

an average of 745.2 central line days revealed an average number of CLABSI infections

of 2.066. This resulted in an infection rate of 3.93 and an average SIR of 2.98. A SIR of

2.98 is approximately three times the normal level of central line infections that would be

observed in a hospital population of this category. Hospitals in the 201 to 500 bed

category, an average 3,331.947 central line days were observed resulting in an average of

6.632 CLABSI infections. An infection rate of 1.907 reflected a SIR of .998. This

population of hospitals reflects a slightly lower incidence of CLABSI as compared to

other hospitals within that same category nationally. For hospitals with greater than 501

beds within their health care facility, a substantial increase of central line days was

acknowledged with an average of 4,221.6 line days in the year 2008. An average of 19.2

CLABSI infections was realized in that same time span. This reflected an average

infection rate of 4.212 and a SIR of .9996. This reflects an overall lower incidence than

would be expected nationally but only a marginal decrease. These results indicated that

the Hypothesis, H1 indicated that there was a significantly higher incidence of CLABSI

observed in hospitals of higher volume than lower bed capacities; H1 is a true Hypothesis.

H2 indicates that the significant differences between responsive measures used by

hospitals to minimize the effect of CLABSI when two or more full-time infection control

practitioners are employed for surveillance purposes is true. Hospitals with 0 to 50 beds

that have high incidence rates have a Mean (1), Median (1), and Mode (1) of infection

150

control practitioners on staff. These practitioners are conducting an average of 8.88

hours of surveillance, Median (9), and Mode (10). Hospitals in the 51 to 200 bed category

have an average of .96 ICPs, Median (1) and Mode (1) and conduct an average of 14.8

hours of infection control weekly, Median (11 hours), and Mode (10 hours). Hospital in

the 201 to 500 bed category have an average of 1.71 ICPs staffed, Median (2), Mode (1)

and conduct and average of 36.89 hours of infection control weekly, Median (30 hours),

and Mode (30 hours). Hospital in the 501 beds and more category have an average of 4.2

ICPs, Median (4), and Mode (5) conducting an average of 89.4 hours of infection control

weekly, Median (70 hours), and Mode (na). It was observed that hospitals with greater

capacity and more staff dedicated to conducting more hours of surveillance have lower

SIR ratios (.999 and .996 respectively). H0. There was a significantly lower incidence of

CLABSI observed in hospitals of higher bed capacities than in hospitals of lower bed

capacities were true. Smaller hospitals by comparison have a significantly higher rate of

CLABSI than hospitals with a greater bed capacity. This null Hypothesis was rejected as

false. H0. stated that hospital CLABSI rates are not dependent upon the number of full-

time staff and number of hours of infection control activities performed per month was

rejected as indicated with higher bed capacity hospitals employing more ICPs and

conducting more hours of surveillance have a lower SIR (.999 and .996 respectively).

Research Question 2 asked if there is a correlation between monitoring and

compliance methods and incidence rates of CLABSI infection. Hypotheses for Question

2 include:

H1 : A lower rate of CLABSI is observed when direct observation is conducted

during catheter insertion.

151

H2 : When staff members are authorized to stop invasive procedures when they

observe protocols not being followed, and poor compliance with aseptic technique that

results in increased incidence of CLABSI.

H3 : The use of a preestablished procedure checklist decreases the rate of

CLABSI in acute care hospitals in South Carolina.

H4 : Acute care hospitals with larger bed capacities have well-defined syndromic

surveillance programs automated with their laboratories in addition have more full-time

infection control practitioners on staff and perform more hours of infection control

activities per month.

H5 : A greater number of person-hours used to identify, observe technique,

evaluate insertion site care, and perform surveillance of hand-washing technique

decreases the number of infectious processes.

H0 : No change in the rate of CLABSI is observed when direct observation is

conducted during catheter insertion.

H0 : When staff members are authorized to stop invasive procedures for

unacceptable performance of procedure, non-compliance incidence rates of CLABSI

have no appreciable change.

H0 : The use of preestablished procedure checklist does not affect the rate of

CLABSI in acute care hospitals in South Carolina.

H0 : Hospital CLABSI rates are not dependent upon the number of full-time staff

and number of hours of infection control activities performed per month.

Statistically, larger capacity hospitals are conducting a significantly higher

volume of monitoring and observation processes than hospitals with fewer beds.

152

H1 stated that a lower rate of CLABSI was observed when direct observation was

conducted during catheter insertion; this is a true hypothesis. During the catheterization

process, hospitals with 0 to 50 beds are compliant 88.9% of the time, 51 to 200 bed

hospitals are compliant at a rate of 53.3%, 201 to 500 bed hospitals have a compliance

rating of 42.11%, and 100% of hospital with greater than 501 beds indicate compliance.

H2 stated that when staff members are authorized to stop invasive procedures when they

observe protocols not being followed or poor compliance with aseptic technique that

results in increased incidence of CLABSI. This Hypothesis is considered false. Hospitals

in the 0 to 50 bed range indicate compliance 55.56%, 51 to 200 beds are compliant

26.67%, 201 to 500 beds facilities are 36.84%, and 501 and greater are 100% compliant.

H3 indicates that the use of a preestablished procedure checklist decreases the rate

of CLABSI in acute care hospitals in South Carolina. This Hypothesis is considered true.

Hospitals in the 0 to 50 bed range indicate compliance 66.67%, 51 to 200 beds are

compliant 60%, 201 to 500 beds facilities are 73.68%, and 501 and greater are 100%

compliant. H4 indicates that acute care hospitals with larger bed capacities have well-

defined syndromic surveillance programs automated with their laboratories in addition,

have more full-time infection control practitioners on staff and perform more hours of

infection control activities per month. This is considered a true hypothesis. Hospitals in

the 0 to 50 bed range indicate compliance 33.33%, 51 to 200 beds are compliant 50%,

201 to 500 beds facilities are 31.58%, and 501 and greater are 60% compliant with

having laboratory interface with the infection control system. The question regarding

hours of surveillance was previously answered. H5 inquires whether a greater number of

person-hours used to identify, observe technique, evaluate insertion site care, and perform

153

surveillance of hand hygiene practices decreased the number of infectious processes. It

was observed in the data that when more techniques were employed for surveillance that

infections actually decreased. This Hypothesis is true. Hospitals in the 0 to 50 bed range

indicate compliance 88.89%, 51 to 200 beds are compliant 76.67%, 201 to 500 beds

facilities are 84.21%, and 501 and greater are 100% compliant. H0 indicates that no

change in the rate of CLABSI was observed when direct observation was conducted

during catheter insertion. This null Hypothesis was rejected. H0 states that when staff

members are authorized to stop invasive procedures for unacceptable performance of

procedure, non-compliance incidence rates of CLABSI have no appreciable change. This

null Hypothesis was rejected. H0 states that the use of preestablished procedure checklist

does not affect the rate of CLABSI in acute care hospitals in South Carolina. This null

Hypothesis was rejected. H0 states that hospital CLABSI rates are not dependent upon the

number of full-time staff and number of hours of infection control activities performed

per month. This null Hypothesis was rejected.

Research Question 3 asked whether a relationship exist between an increased

rate of CLABSI in acute care hospitals and integration of prevention processes when

included in their facilities protocols. H1 states that the daily review of line necessity and

prompt removal of unnecessary lines minimize the rate of CLABSI in South Carolina’s

hospitals. Hypotheses for Research Question 3 included:

H1 : The daily review of central line necessity and prompt removal of unnecessary

central lines minimize the rate of CLABSI in South Carolina’s hospitals.

H2 : Monitoring of optimal catheter site selection minimizes the incidence of

CLABSI infection.

154

H3 : Acute care hospitals that use chlorhexidine skin antiseptics during site

preparation and care have lower incidence of CLABSI infection.

H4 : Acute care hospitals that use maximal barrier precautions upon insertion

observe a lower incidence of CLABSI infection.

H5 : CLABSI rates are lower when Infection Control Practitioners conduct hand

hygiene monitoring.

H0 : CLABSI rates are higher when hygiene monitoring is conducted more

frequently.

H0 : Higher CLABSI rates are seen when the total number of hospital wide

monthly observations are increased.

H0 : The daily review of central line necessity and prompt removal of unnecessary

central lines does not influence the rate of CLABSI in South Carolina’s hospitals.

H0 : Monitoring of optimal catheter site selection does not affect the incidence of

CLABSI infection.

H0 : Acute care hospitals that use chlorhexidine skin antiseptics during site

preparation and care does not influence the incidence of CLABSI infection.

H0 : Acute care hospitals that use maximal barrier precautions upon insertion has

no affect on the incidence of CLABSI infection.

H0 : CLABSI rates are higher when hygiene monitoring is conducted less

frequently by Infection Control Practitioners.

Hospitals in the 0 to 50 bed range indicate compliance 77.78%, 51 to 200 beds are

compliant 56.67%, 201 to 500 beds facilities are 57.89%, and 501 and greater are 80%

compliant. H2 states that monitoring of optimal catheter site selection minimizes the

155

incidence of CLABSI infection. Hospitals in the 0 to 50 bed range indicate compliance

77.78%, 51 to 200 beds are compliant 73.33%, 201 to 500 beds facilities are 68.42%, and

501 and greater are 100% compliant. This hypothesis was accepted. H3 states that acute

care hospitals that use chlorhexidine skin antiseptics during site preparation and care have

lower incidence of CLABSI infection. Hospitals in the 0 to 50 bed range indicate

compliance 77.78%, 51 to 200 beds are compliant 86.67%, 201 to 500 beds facilities are

89.47%, and 501 and greater are 100% compliant. This hypothesis was accepted. H4

states that acute care hospitals that use maximal barrier precautions upon insertion

observe a lower incidence of CLABSI infection. Hospitals in the 0 to 50 bed range

indicate compliance 88.89%, 51 to 200 beds are compliant 80%, 201 to 500 beds

facilities are 84.21%, and 501 and greater are 100% compliant. This hypothesis was

accepted. H5 states that CLABSI rates are lower when ICPs conduct hand hygiene

monitoring. Hospitals in the 0 to 50 bed range indicate compliance 100%, 51 to 200 beds

are compliant 70%, 201 to 500 beds facilities are 73.68%, and 501 and greater are 40%

compliant. This hypothesis was rejected. H0 states that CLABSI rates are higher when

hygiene monitoring is conducted more frequently. Data indicates that there was 0%

compliance with daily monitoring. Larger hospitals, 500 beds and more are compliant at

60% with the majority of all hospitals conducting evaluation measures either monthly or

quarterly. H0 states that higher CLABSI rates are seen when the total number of hospital

wide monthly observations are increased. This null hypothesis was rejected due to

defined data described above. H0 states that daily review of line necessity and prompt

removal of unnecessary lines does not influence the rate of CLABSI in South Carolina’s

hospitals. This null hypothesis was rejected due to defined data described above. H0

156

indicates that monitoring of optimal catheter site selection does not affect the incidence

of CLABSI infection. This null hypothesis was rejected due to defined data described

above. H0 indicates that acute care hospitals that use chlorhexidine skin antiseptics during

site preparation and care do not affect the incidence of CLABSI infection. This null

hypothesis was rejected due to defined data described above. H0 stated that acute care

hospitals that use maximal barrier precautions upon insertion has no affect on the

incidence of CLABSI infection. This null hypothesis was rejected due to defined data

described above. H0 stated that CLABSI rates are higher when hygiene monitoring is

conducted less frequently by ICPs. There is a failure to reject the null hypothesis.

Discussion of the Conclusions in Relation to the Literature The costs of CLABSI for health care organizations, payers, as well as the physical

and emotional toll their infections have is significant due to the high cost of medical care,

utilization of medical and human resources, and the financial impact on the patient. With

the rising cost of healthcare, a hospital’s ability to evaluate and provide positive

intervention in clinical care through effective therapeutic measures and early recognition

of illness through active surveillance programs could provide both the health care

organization and the patient with much needed economic and human impact relief.

CLABSIs are responsible for a significant number of health-care–acquired conditions that

are dangerous to patients receiving medical care. The CDC (2011) report that

approximately one in 20 patients receiving hospital care in the United States acquires an

HAI annually. CLABSIs are acknowledged as one of the most deadly types of HAI. A

mortality rate of 12%–25% is seen among this group of patients.

157

This study reveals that health care organizations worked toward improving health

care and clinical processes aimed to ameliorate the deadly effects of central line

infections by implementing effective surveillance and therapeutic measures. These

findings are similar to the findings of other studies (Pronovost, Needham, and Berenholz,

2006) in which they identified the role that individual interventions in the health care

environment has on determining therapeutic and surveillance measures and if these

components of care are cost-effective and clinically meaningful. An example of these

interventions include the use of chlorhexidine as a skin antiseptic and how the addition of

a chlorhexidine-impregnated sponge dressings contributes to the overall reduction in

infection rates.

In an effort to close the gap between this research and current literature, it was

widely accepted in the study that central line-associated bloodstream infections are

considered a significant cause of morbidity and mortality in hospitalized patients. In the

perspective of this research, recommendations are strong to empirically use processes

such as bundling of therapeutic procedures, good hand hygiene techniques, line

management and other preventative measures; but general research has not been

conducted to evaluate all perspectives covered by the HIDA document. Therefore, the

appropriate interventions to prevent, control, and reduce CLABSIs have yet to be

analyzed thoroughly. This research indicates that the CLABSI rate ranged from 1.6 to

44.6 cases per 1000 central line days in critical care units and was associated with

significant extra mortality, with an odds ratio ranging from 2.8 to 9.5. The results of six

interventional studies indicated that hand hygiene and educational programs were related

to a significant reduction in CLABSI rates. Few studies show successful interventions for

158

a reduction in CLABSI. Research indicates that additional epidemiological studies are

required to understand all of the effects of CLABSI and to develop more-definitive

approaches for CLABSI prevention in the form of practical, low-cost, low technology

measures that will improve outcome and keep costs associated with CLABSI patients

low.

Recommendations for Further Study

Recommendation for further research would include the use of multiple years of

data to eliminate variances in reported data provided to SCDHEC DADE. Additional

recommendations include the use of national and international collaborative studies

designed to guide research utilizing evidence-based clinical practice guidelines,

incorporate change methodologies, and reliance on clinical and process improvement

processes over a continuum to determine efficacy of these practices and processes.

Collaborative approaches to improve and influence healthcare appeals to

clinicians and public health professionals but lacks definitive evidence of its

effectiveness. Research opportunities exist in the appropriate use of perioperative

antibiotics to include appropriate timing, selection, and duration of antibiotics to prevent

or minimize the risk of developing a CLABSI. Research surrounding the environment of

provider awareness and implementation of strategies to enhance organizational change

provides an opportunity for the development of multidisciplinary teams, promotes adding

staff to existing teams, monitoring and promoting changes in infection control practices,

implementation of audit and feedback forums as well as integration of consensus building

159

sessions. These attributes provide a rich opportunity to minimize the impact CLABSI has

on both academic and clinical settings.

Limitations, Delimitations, and Recommendations The following limitations existed in this study. The data supplied by SCDHEC

was limited in categories by the number of smaller hospitals that report fewer CLABSI or

use of central lines. This provided few opportunities to evaluate the effectiveness of

central line therapies in smaller organizations. Limitations indicate that the collection of

data in the initial year may not be comprehensive even though data were mandatorily

reported. This limitation relates to the self-report survey process. The nature of the

research relies on self-report data in which the data collectors’ responses can be

inaccurate because of the inherent nature of self-report. Not all facilities included all data

requested leaving unknown information on variables required to effectively evaluate the

data. Clarification of data request to assure that all hospital are truly reporting the same

types of information and remove personal biases from the data collection process by

providing inservice education to all data collectors would minimize artifact in the

collection process.

Recommendations Based on Delimitations

The delimitations of this study are those characteristics that limited the scope or

defined finite boundaries of inquiry that was determined by consciously excluding or

including decisions that were made throughout the development of the proposal or

subsequent research. Among these delimitations was the choice of objectives, research

160

questions, variables of interest, and alternative theoretical perspectives that could have

been included into this study. In retrospect, the first limiting concept was the choice of

the research problem; additional preliminary research would have provided guidance into

specific areas of CLABSI and focus efforts into more concrete and less abstract

hypotheses and research questions. The major concern for this research project included

the use of too many variables. A better option would have been to limit these variable to

a fewer number and the use of ANOVA versus MANOVA in the research design.

No additional research recommendations were required to satisfy investigational

issues not supported by the data but relevant to the research problem. Any

recommendations of this nature would be included in additional research request.

Conclusion

The impact that CLABSI has on health care organizations, practitioners, insurers

and the patient population was significant. The direct and indirect cost of care for

patients was substantial. The CDC reported on March 1, 2011, that in 2009, an estimated

25,000 fewer CLABSIs occurred among patients in ICUs in the United States than in

2001. This accounted for a 58% reduction in CLABSI cases in U.S. based hospitals. The

CDC determined that the cumulative number of CLABSIs prevented since 2001 was

substantially higher because reductions have occurred annually for the past decade.

Reductions in CLABSI represent an estimated 3,000-6,000 lives saved and estimated

healthcare savings of $414 million in 2009. With an anticipated cost of $16,550 and a

mortality rate of up to 25%, preventing CLABSI in health care facilities could save $1.8

billion, and the impact on lives saved could be as high as 27,000 annually. The majority

161

of CLABSIs are now occurring outside the walls of critical care units and hospitals.

Many are currently seen in outpatient facilities, especially in outpatient dialysis clinics.

Reductions in CLABSIs in critical care units are likely because of coordinated

effort by governmental agencies, professional groups, and health-care personnel who

implement best practices provided by academic and clinical research (CDC, 2011). This

study provided much information on how hospitals in South Carolina responded to the

2008 data request on their clinical behaviors used to minimize the impact of CLABSI

within their facilities.

In summary, this research was consistent with infection control theories integral

to hospital care and public health concerns. These theories indicate an increasing number

of HAIs are preventable by consistent adherence to evidence-based strategies and

evidence that prevention strategies can be successful. This research is also consistent

with the ideals of the Association for Professionals in Infection Control and

Epidemiology (APIC), who believes that continued use of monitoring and integration of

bundling strategies will result in an ongoing reduction in CLABSI both inside and outside

the critical care unit. The scientific community provides support for the CLABSI

prevention "bundle" that is widely supported by evidenced-based recommendations

issued by the Centers for Disease Control & Prevention (CDC)’s Healthcare Infection

Control Practices Advisory Committee (HICPAC). This doctoral dissertation as well as

other research discovered in literature suggests that each component of a CLABSI bundle

is important but collectively all strategies observed in the bundling process provide the

efficacy of synergistic prevention interventions.

162

REFERENCES

Agency for Healthcare Research and Quality. (2007). Closing the quality gap: A critical analysis of quality improvement strategies. Retrieved August 19, 2010, from http://www.ncbi.nlm.nih.gov/bookshelf/br.fcgi?book=techrev9v6

America's Health Rankings. (2010). Acknowledgement. Retrieved October, 23, 2010, from http://americashealthrankings.org/2009/acknowledgement.aspx

Andrews, F.M., Klem, L., Davidson, T.N., O’Malley, P., & Rodgers, W.L. (1981). A guide for selecting statistical techniques for analyzing social science data (2nd ed.). The University of Michigan.

Arozullah, A. M., Khuri, S. F., Henderson, W. G., & Daley, J. (2001). Development and validation of a multifactorial risk index for predicting postoperative pneumonia after major noncardiac surgery. Annals of Internal Medicine, 135(10), 847-857. Retrieved from www.annals.org/content/135/10/847.full.pdf

Ash, S. R. (2008). Advances in tunneled central venous catheters for dialysis: Design and performance. Seminars in Dialysis, 1-12. doi:10.1111/j.1525-139x.2008.00494.x

Baker, J., Lambert, R., Poulos, K., & Feldman, J. (2000). Managing the cost of care: A predictive study to identify critical care patients at risk for nosocomial pneumonia. Journal of Healthcare Finance, 26(3), 73-82.

Bishop, L., Dougherty, L., Bodenham, A., Mansi, J., Crowe, P., Kibbler, C., …Treleaven, J. (2007). Guidelines on the insertion and management of central venous access devices in adults. International Journal of Laboratory Hematology, 29(4), 261- 278. doi:10.1111/J.1751-553X.2007.00931.X

Bleasdale, S. C., Trick, W. E., Gonzalez, I. M., Lyles, R. D., Hayden, M. K., & Weinstein, R. A. (2007). Effectiveness of chlorhexidine bathing to reduce catheter-associated bloodstream infections in medical intensive care unit patients. Archives of Internal Medicine, 167(19), 2073-2079. doi:10.1001/archinte.167.19.2073

Bookstaver, P. B., Gerrald, K. R., & Moran, R. R. (2010). Clinical outcomes of antimicrobial lock solutions used in a treatment modality: A retrospective case series analysis. Clinical Pharmacology: Advances and Applications, 2, 123-130.

163

Bootsma, M., Bonten, M., Pelupessy, I., Hoepelman, A., & Diekmann, O. (2002, May). Nosocomial infection (NI) and length of stay (LOS) in intensive care: A simple observation on cause and effect. Paper presented at the Interscience Conference on Antimicrobial Agents and Chemotherapy, Utrecht, Netherlands. 42(Abstract no. K-656) 1. Retrieved from http://gateway.nlm.nih.gov/MeetingAbstracts/ma? f=102267946.html

Bordens, K. S., & Abbott, B. B. (2008). Research design and methods: A process approach (7th ed.). Boston: McGraw Hill.

Burton, D. C., Edwards, J. R., Horan, T. C., & Fridkin, S. K. (2008, April). Trends in central line associated bloodstream infections in intensive care units-united states, 1997-2007. Paper presented at the SHEA 18th Annual Scientific Meeting, Orlando, Florida. Retrieved from http://www.cdc.gov/ncidod/dhqp/SHEA_Abstract2.html

Burton, D. C., Edwards, J. R., Horan, T. C., Jernigan, J. A., & Fridkin, S. K. (2009). Methicillin-resistant staphylococcus aureus central line associated bloodstream infections in US intensive care units, 1997-2007. JAMA: The Journal of the American Medical Association, 301(7), 727-736. doi:10.1001/jama.2009.153

Burton, D. C., Edwards, J. R., Horan, T. C., Jernigan, J. A., & Fridkin, S. K. (2009). Methicillin-resistant staphylococcus aureus central line-associated bloodstream infections in US intensive care units, 1997-2007. JAMA: The Journal of the American Medical Association, 301(7), 727-736. doi:10.1001/jama.2009.153

Calfee, D., Salgado, C., Classen, D., Arias, K., Podgorny, K., Anderson, D.,…Yokoe, D. (2008). Strategies to prevent transmission of Methicillin Resistant staphylococcus aureus in acute care hospitals. Infection Control and Hospital Epidemiology, 29, S62-S80. doi:10.1086/591061

Camps Serra, M., Cervera, C., Pumarola, T., Moreno, A., Perelio, R., Torres, A., … Marcos, M. A. (2008). Virological diagnosis in community-acquired pneumonia in immunocompromised patients. European Respiratory Journal, 31, 618-624. doi:10.1183/09031936.00073807

Centers for Disease Control (CDC). (2005). Reduction in central line associated bloodstream infections among patients in intensive care units-Pennsylvania, April 2001-March 2005. MMWR-Morbidity and Mortality Weekly Review, 54(40), 1013-1016. Retrieved from http://www.cdc.gov/mmwr/preview/mmwrhtml/ mm5440a2.htm

Centers for Disease Control (CDC). (2006). Outline for healthcare associated infections surveillance. Retrieved July 29, 2010, from www.cdc.gov/nhsn/PDFS/ outlineforhaisurveillance.pdf

164

Centers for Disease Control (CDC). (2009a). About HICPAC. Retrieved August 4, 2010, from http://www.cdc.gov/hicpac/about.html

Centers for Disease Control (CDC). (2009b). Catheter associated bloodstream infections. Retrieved August 1, 2010, from http://www.cdc.gov/ncidod/dhqp/pdf/ guidelines/ BSI_largertext.pdf

Centers for Disease Control (CDC). (2009c). Questions and answers about healthcare associated infections. Retrieved July/15, 2010, from http://www.cdc.gov/ ncidod/dhqp/hai_qa.html

Centers for Disease Control (CDC). (2010). CDC organization. Retrieved September 13, 2010, from http://www.cdc.gov/about/organization/cio.htm

Centers for Disease Control (CDC). (2010). CLABSI baseline prevention practices assessment tool for states establishing HAI prevention collaboratives using AARA funds. Retrieved July 18, 2010, from http://www.cdc.gov/HAI/recoveryact/pdf/ CLABSI_EvalQuestions_Final.pdf

Centers for Disease Control (CDC). (2011). Vital signs: central line associated bloodstream infections. United States 2001, 2008, 2009. MMWR-Morbidity and Mortality Weekly Review, (60), Early release. Retrieved March 1, 2011.

Chen, Y. Y., Chou, Y. C., & Chou, P. (2005). Impact of nosocomial infection on cost of illness and length of stay in intensive care units. Infection Control and Hospital Epidemiology, 26(3), 281-287. Retrieved from http://ir.ym.edu.tw/ ir/handle/ 987654321/15074

Chittick, P., & Sherertz, R. (2010). Recognition and prevention of nosocomial vascular device and related bloodstream infections in the intensive care unit. Critical Care Medicine, 38, S363-S372. doi:10.1097?CCM.obo13e3181e6cdca

Cohen, A.L., Calfee, D., Fridkin, S.K., Huang, S.S., Jernigan, J.A., Lautenbach, E., Oriola, S., ... Weinstein, R.A. (2008). Recommendations for metrics for multidrug-resistant organisms in healthcare settings: SHEA/HICPAC position paper. Infection Control and Hospital Epidemiology, 29(10), 901-913. doi:10.1086/591741

Coia, J. E., Duckworth, G. J., Edwards, D. I., Farrington, M., Fry, C., Humphreys, H., … Tucker, D. R. (2006). Erratum to "guidelines for the control and prevention of methicillin-resistant staphylococcus aureus (MRSA) in healthcare facilities. Journal of Hospital Infection, 64(1), 97-98. doi:10.1016/j.jhin.2006.06.001

Cresswell, J. W. (2009). Research design: Qualitative, quantitative, and mixed methods approaches (3rd ed.). Los Angeles: Sage.

165

Cronk, B. C. (2006). How to use SPSS (4th ed.). Glendale, CA: Pyrczak.

De Angelis, G., Murthy, A., Beyersmann, J., & Harbarth, S. (2010). Estimating the impact of healthcare-associated infections on length of stay. Clinical Microbiology and Infection, , September 25, 2010. doi:10.1111/j.1469- 0691.2010.03332.x

Douglas, L., Aspin, A., Jimmeson, N., & Lawrance, V. (2009). Central venous access devices: Review of practice. Paediatric Nursing, 21(5), 19-22. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&db=c8h&AN=2010340977& site=ehost-live&scope=site

Duffy, J. R. (2002). Nosocomial infections: Important acute care nursing sensitive outcomes indicators. AACN Clinical Issues: Advance Practice in Acute and Critical Care, 13(3), 358-366.

Dugsdale, D. C., Vyas, J. M., & Zieve, D. (2009). Hospital acquired pneumonia. Medline Plus, U.S. National Library of Medicine and the National Institute of Health, Retrieved from http://www.nlm.nih.gov/medlineplus/ency/article/000146.htm

Edwards, J. R., Peterson, K. D., Andrus, M. L., Dudeck, M. A., Pollock, D. A., & Horan, T. C. (2008). National Healthcare Safety Network (NHSN) report, data summary for 2006 through 2007, issued November 2008. American Journal of Infection Control, 36, 609-626. doi:10.1016/j.ajic.2008.08.001

Edwards, J. R., Peterson, K. D., Andrus, M. L., Tolson, J. S., Goulding, J. S., Dudeck, M. A., … Horan, T. C. (2007). National Healthcare Safety Network (NHSN) report, data summary for 2006, issued June 2007. American Journal of Infection Control, 35(5), 290-301.

Edwards, J., Peterson, K., D., Mu, Y., Bangerjee, S., Allen-Bridson, K., Morrell, G., … Horan, T. C. (2009). National Healthcare Safety Network (NHSN) report: Data summary for 2006 through 2008, issued December 2009. American Journal of Infection Control, 37, 783-805. doi:10.1016/j.ajic.2009.10.001

Elixhauser, A., & Steiner, C. (2007). Infections with methicillin-resistant staphylococcus aureus (MRSA) in U.S. hospitals, 1993-2005. AHRQ Healthcare Cost and Utilization Project Statistical Brief, 35(35), 1-10. Retrieved from http://www.hcup-usahrq.gov/reports/statbriefs/sb35.pdf

Farley, J. E. (2008). Epidemiology, clinical manifestations, and treatment options for skin and soft tissue infection caused by community-acquired methicillin-resistant staphylococcus aureus. Journal of the American Academy of Nurse Practitioners, 20(2), 85. doi:10.1111/j.1745-7599.2007.00290.x

166

Fernandez-Hidalgo, N., Almirante, B., Calleja, R., Ruiz, I., Planes, A. M., Rodriguez, D., … Pahissa, A. (2006). Antibiotic-lock therapy for long-term intravascular catheter-related bacteraemia: Results of an open, non-comparative study. Journal of Antimicrobial Chemotherapy, 57(6), 1172-1180. doi:10.1093/jac/dkl103

Field, A. (2009). Discovering statistics using SPSS (3rd ed.). Thousand Oaks, CA: Sage Publications.

Fildes, R., Nikolopoulos, K., Crone, S. F., & Syntetos, A. A. (2008). Forecasting and operational research: A review. Journal of Operational Research Society, 59, 1150-1172. doi:10:1057/palgrave.jors.2602597

Furuno, J. P., McGregor, J. C., Harris, A. D., Johnson, J. A., Johnson, J. K., Langenberg, P., … Perencevich, E. N. (2006). Identifying groups at high risk for carriage of antibiotic-resistant bacteria. Archives of Internal Medicine, 166(5), 580-585. doi:10.1001/archinte.166.5.580

García-Teresa, M. Á., Casado-Flores, J., Domínguez, M. Á. D., Roqueta-Mas, J., Cambra-Lasaosa, F., Concha-Torre, A., & Fernández-Pérez, C. (2007). Infectious complications of percutaneous central venous catheterization in pediatric patients. Intensive Care Medicine, 33(3), 466. Retrieved from http://proquest.umi. com.library.capella.edu/pqdweb?did=1225923951&Fmt=7&clientId=62763&RQ T=309&VName=PQD

Gay, L. R. (1996). Educational research: Competencies for analysis and application (5th ed.). Englewood Cliffs, NJ: Prentice Hall.

Gostin, L. O., & Nass, S. (2009). Reforming the HIPAA privacy rule: Safeguarding privacy and promoting research. JAMA: The Journal of the American Medical Association, 301(13), 1373-1375. doi:10.1001/jama.2009.424

Griffith, J. R., & White, K. R. (2007). The well-managed healthcare organization (6th ed.). Chicago: Health Adminstration Press.

Groeneveld, A. B., & Hack, C. E. (2008). The role of inate immune response in hospital- versus community acquired infection in febrile medical patients. International Journal of Infectious Disease, 12(6), 660-670. Retrieved from http://cat.inist. fr/?aModele=afficheN&cpsidt=20862133

Guilbert, M. C., Elkouri, S., Bracco, D., Corriveau, M. M., Beaudoin, N., Dubois, M. J.,… Blair, J. F. (2008). Arterial trauma during central venous catheter insertion: Case series, review and proposed algorithm. Journal of Vascular Surgery, 48(4), 918-925. doi:10.1016/j.jvs.2008.04.046

167

Hageman, J. C., Liedtke, L. A., Sunenshine, R. H., Strausbaugh, L. J., McDonald, L. C., & Tenover, F. C. (2006). Management of persistent bacteremia caused by methicillin-resistant staphylococcus aureus: A survey of infectious diseases consultants. Clinical Infectious Diseases, 43(5), e42-e45. Retrieved from http://ezproxy.library.capella.edu/login?url=http://search.ebscohost.com/login.asp x?direct=true&db=aph&AN=22231479&site=ehost-live&scope=site

Haley, R. W., Quade, D., Freeman, H. E., & Bennett, J. V. (1980). Study on the efficacy of nosocomial infection control (SENIC project): Summary of study design. American Journal of Epidemiology, 111(5), 472-485. Retrieved from http://aje. oxfordjournals.org/cgi/content/abstract/111/5/472

Hall, K. K., & Lyman, J. A. (2006). Updated review of blood culture contamination. Clinical Microbiology Reviews, 19(4), 788-802. doi:10.1128/CMR.00062-05

Hansen, S. K., Rainey, P. B., Haagensen, J. A., & Molin, S. (2007). Evolution of species interaction in a biofilm community. Nature, (February), September 26, 2010-533- 536. Retrieved from http://web.me.com/redifiori/Russell_Di_Fiori/ Prokaryotes_ files /biofilmevolutionsymbiosis.pdf

Harris, A. D., McGregor, J. C., & Furuno, J. P. (2006). What infection control interventions should be undertaken to control multidrug-resistant gram-negative bacteria? Clinical Infectious Diseases, 43, S57-S61. doi:1058-48381/2006 /4305S2-000

Herz, M. I., Endicott, J., & Gibbon, M. (1979). Brief hospitalization: Two-year follow- up. Archives of General Psychiatry, 36(6), 701-705. Retrieved from http://archpsyc.ama-assn.org/cgi/content/abstract/36/6/701

Herzer, K., & Seshamani, M. (2010). A success story in American healthcare: Eliminating infections and saving lives in Michigan (Quality Report Washington, DC: U.S. Department of Health and Human Services. Retrieved from http://healthreform.gov/reports/success/success_story.pdf. (The quality and safety of healthcare in the United States)

Heyland, D. K., Cook, D. J., & Griffith, L. (1999). The attributal morbidity and mortality of ventilator associated pneumonia in the critically ill patient. American Journal of Respiratory Critical Care Medicine, 159, 1249-1256.

HICPAC. (2009). About HICPAC. Retrieved August 4, 2010, from http://www.cdc. gov/hicpac/about.html

168

Hidron, A. I., Edwards, J. R., Patel, J., Horan, T. C., Sievert, D. M., Pollock, D. A., Fridkin, S. K., National Healthcare Safety Network Team, & Participating National Healthcare Safety Network Facilities. (2008). NHSN annual update: Antimicrobial-resistant pathogens associated with healthcare-associated infections: Annual summary of data reported to the national healthcare safety network at the centers for disease control and prevention, 2006-2007. Infection Control and Hospital Epidemiology : The Official Journal of the Society of Hospital Epidemiologists of America, 29(11), 996-1011. doi:10.1086/591861

Howman, S. F. (1999). Mechanical ventilation: A review and update for clinicians. Hospital Physician, 35(12), 26-36. Retrieved from http://www.turner- white.com/pdf/hp_dec99-vent.pdf

Hu, K. K., Veenstra, D.L., Lipsky, B.A., Saint, S. (2004). Use of maximal sterile barriers during central venous catheter insertion: Clinical and economic outcomes. Clinical Infectious Diseases,39:1441-5.

Huang, S. S., Yokoe, D. S., Hinrichsen, V. L., Spurchise, L. S., Datta, R., Miroshnik, I., & Platt, R. (2006). Impact of routine intensive care unit surveillance cultures and resultant barrier precautions on hospital-wide methicillin-resistant staphylococcus aureus bacteremia. Clinical Infectious Diseases, 43(8), 971-978. doi:1058- 4838/2006-0004

Institute for Healthcare Improvement.Mentor hospital registry: Central line infection. Online, July 14, 2010. Retrieved from http://www.ihi.org/IHI/ programs/ campaign/mentor_registry_cli.htm

Institute of Medicine, & Committee on injury prevention and control. (1999). Reducing the burden of injury: Advancing prevention and treatment. Washington, DC: National Academy Press.

Jaeger, K., Zenz, S., Juttner, B., Ruschulte, H., Kuse, E., Heine, J., … Karthaus, M. (2005). Reduction of catheter related infections in neutropenic patients: A prospective controlled randomized trial using a chlorhexidine and silver sulfadiazine impregnated central venous catheter. Annals of Hematology, 84(4), 258-262. doi:10.1007/s00277-004.0972-6

Jarvis, W. R. (1996). CIPI: 4th International Conference on the Prevention of Infections. Infection Control and Hospital Epidemiology, 17(8), 552-557. Retrieved from http://www.jstor.org/stable/30141291

169

Jarvis, W. R., Murphy, C., Hall, K., Fogle, P. J., Karchmer, T. B., Harrington, G., … Sherertz, R. J. (2009). Healthcare associated bloodstream infections associated with negative or positive pressure or displacement mechanical valve needleless connectors. Clinical Infectious Diseases, 49(Special Article), 000-000. doi:10.1086/648418

Johanson, W. G., & Dever, L. L. (2003). Nosocomial pneumonia. Intensive Care Medicine, 29(1), 23-29. Retrieved from www.ncbi.nlm. nih.gov/ gov/ pubmed/12528018

Kaiser Family Foundation. (2010). State health facts. Retrieved October, 20, 2010, from http://www.statehealthfacts.org/profileglance.jsp?rgn=42&print=1

Khan, J. M., Goss, C. H., Heagerty, P. J., Kramer, A. A., O'Brien, C. O., & Rubenfeld, G. D. (2006). Hospital volume and the outcomes of mechanical ventilation. New England Journal of Medicine, 355(Special Article), 41-50.

Kleinpell, R. M. (2004). Working out the complexities of severe sepsis. Nursing Management, 35(5), 48A. Retrieved from http://proquest.umi.com. library. capella.edu/pqdweb?did=637495761&Fmt=7&clientId=62763&RQT=309&VNa me=PQD

Klevens, R. M., Morrison, M. A., Nadle, J., Petit, S., Gershman, K., Ray, S., Harrison, L. H., … Fridkin, S. K. (2007). Invasive methicillin-resistant Staphylococcus aureus infections in the United States. JAMA: The Journal of the American Medical Association, 298(15), 1763-1771. doi:10.1001/jama.298.15.1763

Klevins, R. M., Edwards, J. R., Richards, C. L., Horan, T. C., Gaynes, R. P., Pollock, D. A., & Cardo, D. M. (2007). Estimating healthcare associated infections and deaths in U.S. hospitals, 2002. Public Health Reports, 122(March-April), 160-166. Retrieved from www.cdc.gov/ncidod/dqp/pdf/hicpac/infections_death.pdf

Laux, L., Dysert, K., Kiely, S., & Weimerskirch, J. (2010). Trauma VAP SWAT team: A rapid response to infection prevention. Critical Care Nursing Quarterly, 33(2), 126-131. doi:10.1097/CNQ.obo13e3181d911bb

Leedy, P. D., & Ormrod, J. E. (2005). Practical research: Planning and design (8th ed.). Upper Saddle River, NJ: Pearson-Merrill Prentice Hall.

Lin, M. Y., & Hayden, M. K. (2010). Methicillin-resistant Staphylococcus aureus and Vancomycin resistant Enterococcus: Recognition and prevention in intensive care units. Critical Care Medicine, 38, S335-S344. doi:10.1097/ CCM.obo13e3181 e6ab12

170

Linenberger, M. L. (2006). Catheter-related thrombosis: Risks, diagnosis, and management. Journal of the National Comprehensive Cancer Network, 4(9), 889- 901. Retrieved from http://www.jnccn.org/content/4/9/889.abstract

Liu, J. H., Zingmond, D. S., & McGory, M. L. (2006). Disparities in the utilization of high volume hospitals for complex surgery. Journal of the American Medical Association, 296(16), 1973-1980. doi:10.1001/jama.296.16.1973

Lyles, A., Fanikos, J., & Jewell, K. (2009). Aligning pharmacy and health-system objectives to eliminate central-line-associated bacteremias. American Journal of Health-System Pharmacy, 66(24), 2189-2197. doi:10.2146/ajhp080679

Maine Center for Disease Control and Prevention. (2010). State of Maine Healthcare Associated Infection Plan Augusta, ME: Maine Center for Disease Control and Prevention. Retrieved from http://www.cdc.gov/HAI/pdfs/stateplans/ma.pdf. (Healthcare Associated Infection Plan)

Maki, D. G., Kluger, D. M., & Crnich, C. J. (2006). The risk of bloodstream infection in adults with different intravascular devices: A systematic review of 200 published prospective studies. Mayo Clinic Proceedings, 81(9), 1159-1171. doi:10.4065/ 81.9.1159

Marik, P. E. (2001). Catheter related bloodstream infection. In Handbook of evidence based critical care (1st ed., pp. 117-128). New York: Springer Science Business Media LLC. doi:10.1007/978-1-4419-5923-2_11

Mark, B.A., Harless, D.W., McCue, M., Xu, Y. (2004). A longitudinal examination of hospital registered nurse staffing and quality of care. Health Services Research,39(2), 279-300.

Marschall, J., Mermel, L. A., Classen, D., Arias, K. M., Podgorny, K., Anderson, D. J., …Yokoe, D. S. (2008). Strategies to prevent central line-associated bloodstream infections in acute care hospitals. Infection Control and Hospital Epidemiology, 29(Supplement 1), S22-S30. doi:10.1086/591059

McGee, G., Shelton, W., Gruenberg, D. A., Rose, S. L., Rutter, A. E., & Socaris, S. (2006). Factors influencing length of stay in the intensive care unit. American Journal of Critical Care, 15, 502-509.

McKibben, L., Horan, T., Tokars, J. I., Fowler, G., Cardo, D. M., Pearson, M. L., & Brennan, P. J. (2005). Guidance on public reporting of healthcare-associated infections: Recommendations of the healthcare infection control practices advisory committee. American Journal of Infection Control, 33, 217-226. doi:10.1016/ajic.2005.04.001

171

McKinney, M. (2010). Focusing on infections. Modern Healthcare, 40(16), 12-12. Retrieved from http://ezproxy.library.capella.edu/login ?url=http://search. ebscohost.com/login.aspx?direct=true&db=aph&AN=50074532&site=ehost- live&scope=site

Mehta, A., Rosenthal, V. D., Mehta, Y., Chakravarthy, M., Todi, S. K., Sen, N., … Damanj, N. (2007). Device associated nosocomial infection rates in intensive care units of seven Indian cities. Findings of the international nosocomial infection control consortium (INICC). The Journal of Hospital Infection, 67(2), 168-174. doi:10.1016/j.jhin.2007.07.008

Mimoz, O., Villeminey, S., Ragot, S., Dahyot-Fizelier, C., Laksiri, L., Petitpas, F., & Debaene, B. (2007). Chlorhexidine-based antiseptic solution vs alcohol-based povidone-iodine for central venous catheter care. Archives of Internal Medicine, 167(19), 2066-2072. doi:10.1001/archinte.167.19.2066

Moreno, C. A., Rosenthal, V. D., Olarte, N., Gomez, W. V., Sussman, O., Agudelo, J. G., … Henriquez, D. (2006). Device associated infection rate and mortality in intensive care units of 9 Columbian hospitals: Findings of the international nosocomial infection control consortium. Infection Control and Hospital Epidemiology, 27(4), 000-000. doi:0899-823x/2006/2714-00xx

Murphy, G. J., Reeves, B. C., Rogers, C. A., Rizvi, S. I. A., Culliford, L., & Angelini, G. D. (2007). Increased mortality, postoperative morbidity, and cost after red blood cell transfusion in patients having cardiac surgery. Circulation, 116(22), 2544- 2552. doi:10.1161/CIRCULATIONAHA.107.698977

Muto, C., Herbert, C., Harrison, E., Edwards, J. R., Horan, T., Andrus, M., … Kutty, P. K. (2005). Reduction in central line-associated bloodstream infections among patients in intensive care units-Pennsylvania, April 2001- March 2005. MMWR Weekly, 54(40), 1013-1016. Retrieved from http://proquest.umi.com.library. capella.edu/pqdweb?RQT=568&VInst=PROD&VName=PQD&VType=PQD&F mt=6&did=917860111&TS=1283223691&fromjs=1

Netstate. (2009). South Carolina economy. Retrieved October, 18, 2010, from http://www.netstate.com/economy/sc_economy.htm

Neuman, W. L. (2006). Social research methods: Qualitative and quantitative approaches (6th ed.). Boston, MA: Allyn & Bacon.

Odds, F. C., Hanson, M. F., Davidson, A. D., Jacobsen, M. D., Wright, P., Whyte, J. A., … Jones, B. L. (2007). One-year prospective survey of candida bloodstream infections in scotland. Journal of Medical Microbiology, 56, 1066-1075. doi:10.1099/jmm.0.47239-0

172

Osih, R. B., McGregor, J. C., Rich, S. E., Moore, A. C., Furuno, J. P., Perencevich, E. N., & Harris, A. D. (2007). Impact of empiric antibiotic therapy on outcomes in patients with Pseudomonas aeruginosa bacteremia. Antimicrobial Agents and Chemotherapy, 51(3), 839-844. doi:10.1128/AAC.00901-06

Paterson, D. L. (2007). Serious infections in the intensive care unit: Pseudomonas aeruginosa and Acinetobacter baumannii. Clinical Infectious Diseases, 43(Suppl 2), S41-S42. doi:1058-4838/2006/4305s2-0001

Paterson, D. L. (2006). The epidemiological profile of infections with multidrug-resistant Pseudomonas aeruginosa and Acinetobacter species. Clinical Infectious Diseases, 43, S43-S48. doi:1058-4838/2006/4305s2-0002

Patten, M. L. (2009). Understanding research methods: An overview of the essentials (7th ed.). Glendale, CA: Pyrczak.

Pennsylvania Patient Safety Advisory Committee. (2010). Beyond the bundle: Reducing the risk of central line-associated bloodstream infections. Pennsylvania Patient Safety Advisory, 18(7(Supplemental 1)), 1-9.

Pierce, G. E. (2005). Pseudomonas aeruginosa, Candida albicans, and device related nosocomial infections: Implications, trends, and potential approaches for control. Journal of Industrial Microbiology and Biotechnology, 32, 309-318. doi:10.1007 /s10295-005-0225-2

Pittet, D., Allegranzi, B., & Boyce, J. (2009). The world health organization guidelines on hand hygiene in healthcare and their consensus recommendations. Infection Control and Hospital Epidemiology, 30(7), 611-622. doi:10.1086/600379

Pittiruti, M., Hamilton, H., Biffi, R., MacFie, J., & Pertkiewicz, M. (2009). ESPEN guidelines on parenteral nutrition: Central venous catheters (access, care, diagnosis and therapy of complications). Clinical Nutrition, 28(4), 365-377. doi:DOI: 10.1016/j.clnu.2009.03.015

Pronovost, P., Needham, D., Berenholtz, S., Sinopoli, D., Chu, H., Cosgrove, S., … Goeschel, C. (2006). An intervention to decrease catheter-related bloodstream infections in the ICU. New England Journal of Medicine, 355(26), 2725-2732. Retrieved from http://dx.doi.org/10.1056/NEJMoa061115

Ranji, S. R., Shetty, K., Posley, K. A., Lewis, R., Sundaram, V., Galvin, C. M., & Winston, L. G. (2007). Closing the quality gap: A critical analysis of quality improvement strategies (AHRQ No. 04(07)-0051-6). Stanford, CA: Stanford University-UCSF Evidence based Practice Center.

173

Richards, C., Brennan, P. J., & Straube, B. (2009). Toward the elimination of healthcare associated infections. Centers for Disease Control, Atlanta, GA.

Robicsek, A., Beaumont, J. L., Paule, S. M., Hacek, D. M., Thomson, J., Richard B., … Peterson, L. R. (2008). Universal surveillance for methicillin-resistant Staphylococcus aureus in 3 affiliated hospitals. Annals of Internal Medicine, 148(6), 409-W87. Retrieved from http://ezproxy.library. capella.edu/login? url=http://search.ebscohost.com/login.aspx?direct=true&db=aph&AN=31322761 &site=ehost-live&scope=site

Rosenthal, V. D., Maki, D. G., Salomao, R., Moreno, C. Á., Mehta, Y., Higuera, F., … International Nosocomial Infection Control Consortium. (2006). Device- associated nosocomial infections in 55 intensive care units of 8 developing countries. Annals of Internal Medicine, 145(8), 582-591.

Scales, K. (2010). Central venous access devices part 1: Devices for acute care. British Journal of Nursing (BJN), 19(2), 88-92. Retrieved from http://ezproxy.library. capella.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=c8 h&AN=2010544298&site=ehost-live&scope=site

Scales, K. (2010). Central venous access devices: Part 2 for intermediate and long-term use. British Journal of Nursing (BJN), 19(5), S20-5. Retrieved from http://ezproxy.library.capella.edu/login?url=http://search.ebscohost.com/login.asp x?direct=true&db=c8h&AN=2010626206&site=ehost-live&scope=site

SCDHEC. (2010). Retrieved September/22, 2010, from http://www.scdhec.gov/ health/ disease/hai/reports.htm/#reports

Schmittgen, T. D., & Livak, K. J. (2008). Analyzing real-time PCR data by the comparative CT method. Nature Protocols, 3, 1101-1108. doi:10.1038/ nprot. 2008.73

Schwaber, M. J., Navon-Venezia, S., Kaye, K. S., Ben-Ami, R., Schwartz, D., & Carmeli, Y. (2006). Clinical and economic impact of bacteremia with extended- spectrum-{beta}-lactamase-producing enterobacteriaceae. Antimicrobial Agents and Chemotherapy, 50(4), 1257-1262. doi:10.1128/AAC.50.4.1257-1262.2006

Scott, R. D. (2009). The direct medical costs of healthcare associated infections in U.S. hospitals and the benefits of prevention (Technical No. CS200891-A). Atlanta, Georgia: Division of Healthcare Quality Promotion National Center for Preparedness, Detection, and Control of Infectious Diseases. Retrieved from http://www.cdc.gov/ncidod/dhqp/pdf/Scott_CostPaper.pdf.

174

Shannon, R. P., Frndak, D., Grunden, N., Lloyd, J. C., Herbert, C., Patel, B., … Spear, S. J. (2006). Using real-time problems solving to eliminate central line infections. Joint Commission Journal on Quality and Patient Safety, 32(9), 479-487. Retrieved from www.naidagrundin.com/CLABpaper.pdf

Sherertz, R. J., Boger, M. S., Collins, C. A., Mason, L., & Raad, I. I. (2006). Comparative in vitro efficacies of various catheter lock solutions. Antimicrobial Agents and Chemotherapy, 50(5), August 9, 2010-1865. doi:10.1128/AAC.50.5.1865- 1868.2006

Shi, L., & Singh, D. A. (2008). Delivering healthcare in America: A systems approach (4th ed.). Sudbury, MA: Jones and Bartlett Publishers.

Shulgen, G., Kropec, A., Kappstein, I., Daschner, F., & Schumacher, M. (2000). Estimation of extra hospital stay attributable to nosocomial infections heterogeneity and timing of events. Journal of Clinical Epidemiology, 54(4), 409- 417. Retrieved from http://ingentaconnect.com/ content/els/08954356/ 2000/00000053/00000004/art0018

Siegel, J. D., Rhinehart, E., Jackson, M., & Chiarello, L. (2007). Management of multidrug-resistant organisms in healthcare settings, 2006. American Journal of Infection Control, 35(10 Suppl 2), S165-S193. Retrieved from http://www.ncbi. nlm.nih.gov/pubmed/18068814

Siempos, I. I., Kopterides, P., Tsangaris, I., Dimopoulou, I., & Armaganidis, A. (2009). Impact of catheter related bloodstream infections on the mortality of critically ill patients: A Meta analysis. Critical Care Medicine, 37(7), 2283-2289. doi:10. 1097/CCMobo13e318ao2a67

Sligl, W., Taylor, G., & Brindley, P. G. (2006). Five years of nosocomial gram-negative bacteremia in a general intensive care unit: Epidemiology, antimicrobial susceptibility patterns, and outcomes. International Journal of Infectious Diseases, 10, 320-325. doi:10.1016/j.ijid.2005.07.003

South Carolina Department of Health and Environmental Control. (2009). Hospital acquired infections (HAI) in South Carolina. Retrieved July/29, 2010, from http://www.scdhec.gov/health/disease/hai/terms.htm

South Carolina Department of Parks, Recreation & Tourism. (2010). South Carolina facts and figures. Retrieved October, 23, 2010, from http://www.scprt.com/facts- figures/helpwithhomework/fastfacts.aspx

South Carolina Department of Tourism. (2010). Economic contribution of tourism in South Carolina. Retrieved October, 22, 2010, from http://www.scprt.com/facts- figures/helpwithhomework/fastfacts.aspx

175

South Carolina Hospital Association. (2009). Connecting people for a healthier South Carolina. Retrieved October, 20, 2010, from http://accesshealthsc.net/about.php

South Carolina Hospital Association. (2010). Stop BSI. Retrieved July 14, 2010, from http://www.everypatient.net/stopBSI/7-scha-initiatives/56-stop-bsi-project

South Carolina Hospital Association (SCHA). (2010). Stop BSI. Retrieved August 4, 2010, from http://www.everypatient.net/images/stories/docs/stopbsi/01- STOP%20BSI%20fact%20sheet.pdf

Sproull, N. L. (1995). Handbook of research methods: A guide for practitioners and students in the social sciences (2nd ed.). Lanham, MD: Scarecrow Press.

South Carolina Code of Laws (Unannotated) Current through the End of the 2009 Session, HealthU.S.C. Section 44-7-2430(A)(2) (1976). Retrieved from http://www.scstatehouse.gov/code/t44c007.htm

Statistical Consulting Group. (2007). Introduction to SAS. UCLA: Academic Technology Services. Retrieved form http://www.ats.ucla.edu/stat/sas/notes2/

Stone, S. P., Cooper, B. S., Kibbler, C. C., Cookson, B. D., Roberts, J. A., Medley, G. F., … Davey, P. G. (2007). The ORION statement: Guidelines for transparent reporting of outbreak reports and intervention studies of nosocomial infection. Journal of Antimicrobial Chemotherapy, 59(5), 833-840. doi:10.1093/jac/dkm055

Sultz, H.A., & Young, K.M. (2006). Healthcare USA: Understanding its organization and delivery (5th ed.). Sudbury, MA: Jones and Bartlett Publishers.

Tacconneli, E., De Angelis, G., Cataldo, M. A., Pozzi, E., & Cauda, R. (2008). Does antibiotic exposure increase the risk of methicillin-resistant Staphylococcus aureus (MRSA) isolation? A systematic review and Meta analysis. Journal of Antimicrobial Chemotherapy, 61, September 17, 2010-26-38. doi:10.1093/jac/ dkm416

Taheri, P. A., Butz, D. A., & Greenfield, L. J. (2000). Length of stay minimal impact on the cost of hospital admission. Journal of the American College of Surgery, 19(2), 123-130. Retrieved from http://www.lexisnexis.com.library. capella.edu/us/ lnacademic/results/docview/docview.do?docLinkInd=true&risb=21_T861414600 8&format=GNBFI&SORT=BOOLEAN&startDocNo=1&resultsUrlKey=29_T86 14144213&cisb=22_T8614146011&treeMax=true&treeWidth=0&csi=6532&doc No=1

176

Tarricone, R., Torbica, A., Franzetti, F., & Rosenthal, V. (2010). Hospital cost of central line associated bloodstream infections and cost effectiveness of closed vs. open infusion containers. The case of intensive care units in Italy. Cost Effectiveness and Resource Allocation, 8(8), October 5, 2010-1-13. doi:10.1186/1478-7547-8-8

Trust for America's Health. (2010). South Carolina public health data. Retrieved October, 23, 2010, from http://healthyamericans.org/states/?stateid=SC

Turcotte, S., Dubé, S., & Beauchamp, G. (2006). Peripherally inserted central venous catheters are not superior to central venous catheters in the acute care of surgical patients on the ward. World Journal of Surgery, 30(8), 1605. doi:10.1007/s00268- 005-0174-y

U.S. Bureau of Labor Statistics. (2010). Economy at a glance-South Carolina. Retrieved October, 18, 2010, from http://data.bls.gov/cgi-bin/print.pl/eag/eag.sc.htm

U.S. Census Bureau. (2010). Median household income (in 2009 inflation adjusted dollars). Retrieved October, 23, 2010, from http://www.census.gov /hhes/ www/income/statemedfaminc.html

Umscheid, C. A., Mitchell, M. D., Agarwal, R., Williams, K. & Brennan, P. J. (2010). Mortality from reasonably preventable, hospital-acquired infections. Retrieved August/2010, 2010, from http://www.sheaonline.org/ Assets/files/0408_ Penn_ Study.pdf

Valderas, J. M., Starfield, B., Sibbald, B., Salisbury, C., & Roland, M. (2009). Defining comorbidity: Implications for understanding health and health services. Annals of Family Medicine, 7(4), 357-363. doi:10.1370/afm.983

Veenstra, D. L., Saint, S., Saha, S., Lumley, T., & Sullivan, S. D. (1999). Efficacy of antiseptic-impregnated central venous catheters in preventing catheter-related bloodstream infection: A Meta-analysis. JAMA: The Journal of the American Medical Association, 281(3), 261-267. doi:10.1001/jama.281.3.261

von Eiff, C., Jansen, B., Kohnen, W., & Becker, K. (2005). Infections associated with medical devices: Pathogenesis management and prophylaxis. Drugs, 65(2), 179- 214.

Warren, D., Cosgrove, S., Diekema, D., Zuccotti, G., Climo, M., Bolon, M., … Fraser, V. (2006). A multicenter intervention to prevent catheter associated bloodstream infections. Infection Control and Hospital Epidemiology, 27(7), 662-669. Retrieved from http://dx.doi.org/10.1086/506184

177

Weinstein, R. A. (2009). An evidence bases approach to the prevention of central line associated bloodstream infections in hospitalized patients. Paper presented at the Medscape CME. Retrieved from http://cme.medscape.com/viewarticle/705882

Wright, S. B., Ostrowsky, B., Fishman, N., Deloney, V. M., Mermel, L., & Perl, T. M. (2010). Expanding roles of healthcare epidemiology and infection control in spite of limited resources and compensation. Infection Control and Hospital Epidemiology, 31(2), 127-132. doi:10.1086/650199

Yang, Y., McBride, M. V., Rodvold, K. A., Tverdek, F., Trese, A. M., Hennenfent, J., … Schumock, G. T. (2010). Hospital policies and practices on prevention and treatment of infections caused by methicillin-resistant Staphylococcus aureus. American Journal of Health-System Pharmacy, 67(12), 1017-1024. doi:10.2146/ajhp090563

178

APPENDIX A. HIDA Hospital Infection Control Processes Report

179

180

181

182

183

184

185

186

187

188

189

190

191

APPENDIX B. NHSN BSI Form

192

193

194