here is a link to the abstract and article that discusses implementation and evaluation of a standing order, although it does not give you thesteps and those involved, I hope that it gives you an idea why an MDor other cliniician might think that impleme

Implementation and evaluation of a nursing assessment/standing orders– based inpatient pneumococcal vaccination program Carl Eckrode, MPH, RRT-NPS,b Nancy Church, RN, MT,a and Woodruff J. English III, MDa

Portland and Gresham, Oregon

Background: Pneumococcal vaccination is recommended for patients aged 65 years and greater; inpatient vaccination has been suggested as means to increase vaccination rates is this population. Our hospital implemented an inpatient pneumococcal vacci- nation program, and expanded the population of interest to include patients aged 2 to 64 years with risk factors for pneumococcal bacteremia. We studied the outcomes of this program to determine if the rate of pneumococcal vaccination opportunities and pneumococcal vaccination rate could be significantly increased through the application of an in-hospital pneumococcal vaccina- tion program, based on standing orders and assessment by Registered Nurses, when compared to our previous method of physi- cian assessment and written vaccination order for each patient. Methods: Subjects were inpatients admitted to non-intensive care units of our hospital from August to December of 2004. Cases were aged greater than 65 years, or were greater than 2 years of age with selected risk factors. Patients with previous pneumococcal vaccination with the past five years, in terminal or comfort care, those allergic to vaccine components, patients who received organ or bone marrow transplants in the year prior to the study, and those physicians barred them from the vaccination protocol were excluded. Program effectiveness was evaluated through retrospective evaluation of medical records to determine if subjects had been evaluated for vaccination eligibility, and if subjects were eligible, whether or not they had received pneumococcal vaccination. Results: Overall vaccination opportunity rate after implementation of the standing orders-based program increased form 8.6% to 59.1%, and overall vaccination rates improved form 0% to 15.4%. The study found a statistically significant difference in the rate of pneumococcal vaccination opportunities (x2 = 182.46, p = .00) and the pneumococcal vaccination rate (x2 = 56, p = .00) between the two methods of assessment and vaccination; these results are attributable to the study intervention. Conclusions: The study program contributed to increased overall vaccination opportunity and vaccination rates, when compared to the previous method. The overall rates of vaccination attained by this program were often lower than those reported in the ex- isting literature for other program designs; however, this may be due to an unusually high rate of vaccination refusal. (Am J Infect Control 2007;35:508-15.)

The significance of invasive pneumococcal disease cannot be understated, because disease caused by Streptococcus pneumoniae has been reported to be responsible for an estimated 36% of community- acquired pneumonia, an estimated 50% of nosocomial pneumonias, 50,000 cases of bacteremia, and an esti- mated 3000 to 6000 cases of meningitis each year in the United States.1 Forty thousand deaths have been re- ported each year from pneumococcal infection,2,3 with an estimated 175,000 hospitalizations due to the dis- ease each year in the United States.1 The case-fatality

From the Infection Control Department,a Providence St. Vincent Hos- pital and Medical Center, Portland, Oregon; and Respiratory Care Program,b Mt. Hood Community College, Gresham, Oregon.

Address correspondence to Carl Eckrode, MPH, RRT-NPS, Mt. Hood Community College, 26000 SE Stark Street, Gresham, OR 97222.

0196-6553/$32.00

Copyright ª 2007 by the Association for Professionals in Infection Control and Epidemiology, Inc.

doi:10.1016/j.ajic.2006.08.005

508

can range up to 60% for elderly patients who have pneumococcal bacteremia.1 The comorbid condition of bacteremic pneumococcal disease, pneumococcal pneumonia, has a case-fatality of 5% to 7%;1-6 as one of the leading infectious causes of death in the United States,7 it has killed more persons annually than AIDS, tuberculosis, meningitis, and endocarditis combined.8 This degree of morbidity and mortality from a vaccine-preventable disease is unacceptable; vaccination has been proven to reduce morbidity and mortality significantly.9

The current pneumococcal vaccine for adults, the 23-Valent Pneumococcal Polysaccharide Vaccine (23PPV) (Pneumovax, Pasteur-Merieux MSD, Malvern, PA; Pnu- Imune, Wyeth-Lederle, Pearl River, NY), is well targeted against many of the S pneumoniae serotypes that are responsible for illness. For example, in a study of pneumococcal disease, 78 isolates were recovered from patients who had pneumococcal pneumonia. Seventy-one of those isolates (91%) were remarkable, in that the serotype of the isolated organism was one that was included in the 23-valent pneumococcal

Eckrode, Church, and English October 2007 509

vaccine.10 The pneumococcal vaccine has been found safe, with reports of serious reactions rare,11-13 although they are more likely on revaccination.14 Reactions that do occur to the 23PPV typically are local reactions and low-grade fever, which generally are self-limiting,15

and may be more likely in patients who have coinciden- tal upper respiratory illness.12 One of the rare systemic reactions reported was reversible, and may have been associated with a simultaneous influenza vaccination.16

Although the efficacy of the 23PPV for prevention of pneumonia is still subject to discussion and study, an overwhelming amount of evidence and expert opinion support the contention that it is efficacious and cost-effective in immunocompetent patients and those whose diminished immune responses place them at risk for bacteremic pneumococcal dis- ease.3,4,11,17-26 The body of evidence clearly supports the vaccine’s efficacy; however, historically the pneu- mococcal polysaccharide vaccine has been underutil- ized.11,27-29 In response, the Centers for Disease Control and Prevention (CDC) Healthy People 2010 ob- jective for pneumococcal vaccination in people greater than 65 years of age was set at 90% of elderly ever vac- cinated, and objectives were set for vaccination targets of 60% coverage with pneumococcal vaccines among high-risk adults aged 16 to 64 years.30 Nonetheless, by 2002, only 62% of adults older than 65 years had received the pneumococcal vaccine.8 A recent report shows only slight progress in 2003, to a median of 64.2%, still well below the national health objectives for 2010 of 90%.31 These low rates may be due to a lack of formal programs to ensure proper screening and vaccine administration,32 various beliefs on vacci- nation held by the targeted populations,33 or a combi- nation of factors. Whatever the reasons, these low vaccination rates can have serious negative sequelae. For instance, two thirds of persons who had serious pneumococcal disease had been hospitalized within the previous 4 years before their pneumococcal illness, yet few had received pneumococcal vaccine.2,7,9,10

Many strategies for increasing pneumococcal vacci- nation rates have been examined. Among the many methods for increasing vaccination opportunity, hospi- tal-based vaccination programs are appealing, given that inpatients often are within the populations that are at high risk for invasive pneumococcal disease.26,34

Support for inpatient pneumococcal vaccination dates back to the early 1990s.11,20 Thus, the Advisory Com- mittee on Immunization Practices (ACIP)35 and the Task Force on Community Preventive Services have recom- mended administration of pneumococcal polysaccharide vaccines to inpatients.11,34,36 Hospital-based vaccination also may be more cost-effective than a mass vaccination effort, because of the selective targeting of the at-risk population. To take advantage of this selective targeting,

a variety of approaches to inpatient pneumococcal vacci- nation has been attempted, such as the use of nurses, pharmacists, and other nonphysician providers for screening and identification of eligible patients.1,26,37,38

Programs also have relied on written physician orders or patient education to increase vaccination rates,37,38

whereas others attempted to combine pneumococcal vaccination into a clinical pathway for pneumonia treat- ment.39 The degree of success of these programs varied widely.

Because of the clear value of hospital-based pneu- mococcal vaccination programs, the CDC and ACIP have promoted proactive approaches, such as the use of standing orders to simplify the vaccination process and ensure that every patient who enters the health care system is immunized.8,40,41 To aid institutions in implementing a standing orders–based program, the Centers for Medicare and Medicaid Services (CMS) promulgated a significant administrative rule change. Relieving one of the major barriers to implementa- tion of inpatient vaccination programs, CMS standards changed to allow the use of standing orders for pneu- mococcal and influenza vaccination.42 Additionally, CMS levied the requirement to provide pneumococcal vaccination to inpatients aged 65 years or greater. The CMS administrative rule change on standing orders enabled institutions to devise programs that permitted nonphysician providers to assess patients and adminis- ter medications, including vaccines, according to an institution-approved protocol, without a physician’s examination or individual, patient-specific orders.

Our program was designed in response to an unac- ceptably low pneumococcal vaccination rate. For the year prior to program inception (ie, 2003), the inpatient pneumococcal immunization rate in our hospital was only 13%. Given the scope of our problem, it was nec- essary to plan and implement a vaccination program carefully, but quickly. This was an institutional priority, facilitated by the new CMS positions on pneumococcal vaccination and standing orders. Thus, the organization elected to use Borg and Gall’s Research and Develop- ment Cycle, as applied to creating a health promotion program;43 the use of this model allowed for a rapid pro- gram development. We elected to use the registered nurse in the role of assessor and for vaccination deliv- ery. This study assessed the success of this program for the first 90 days after program implementation.

METHODS

The research method chosen for our study was the field experiment; the study design chosen was the Comparison Group Posttest-Only Design.44 The com- parison group was selected from the population of inpa- tients who met the program criteria for pneumococcal

510 Vol. 35 No. 8 Eckrode, Church, and English

vaccination during the 90 days prior to program imple- mentation. The study group was selected from the pop- ulation who met the program criteria for vaccination during the 90 days after program implementation. Ran- dom assignment within the groups was performed. The outcomes of interest—the percentage of eligible subjects provided with vaccination opportunity and vaccine delivery—can be measured accurately, despite variations in and between the sample groups.

Data collection

A computer-generated list of patients who were admitted to the inpatient units was provided to the Providence St. Vincent Hospital and Medical Center Infection Control Department, and medical chart docu- ments were obtained electronically. These documents are collected routinely, and these records were used to obtain the information for evaluation of the pro- gram. The use of this existing data did not require the development or implementation of a survey instru- ment or questionnaire.

The records for the sampled patients were evaluated to determine if the patient was offered vaccination, and, if the patient consented, received the pneumococ- cal vaccination. The records evaluated included the medication administration record, signed pneumococ- cal vaccine orders, or signed vaccine consent forms (a local requirement) with the date, dose, and vaccine lot number entered. This data was converted to, and recorded as, categorical (nominal) data (eg, the patient was offered vaccination or the patient was not offered vaccination, and the patient received the vaccination or the patient did not receive the vaccination). The data were entered into a coding sheet, using the Microsoft Excel 2002 software package (Microsoft, Redmond, WA). After conversion and coding, the percentage of eligible randomly sampled patients who were or were not offered vaccination, and who did or did not receive the vaccination was calculated and compared with the number of randomly sampled patients who were or were not offered vaccination and who did or did not receive vaccination prior to program implementation.

Security and confidentiality of data were maintained throughout the process, through multiple security mea- sures (eg, use of passwords on computers, ‘‘sanitizing’’ records to remove protected health information, secur- ing records in locked storage when not in use, and the use of strong encryption (PGP 7.03, Network Associates, Dallas, TX) for data stored in portable media). Patient identifiers, such as name or medical record number, were not recorded on coding forms or in newly created databases. Documents containing protected health infor- mation were destroyed by shredding subsequent to use.

Human/animal subjects

This was a study of a system for ensuring vaccina- tion opportunity and administration, and no human or animal experimentation was performed. The pneu- mococcal polysaccharide vaccine is approved for use in the study population, and no deviation from the pur- pose of the approved use (‘‘off-label’’ use) occurred. Thus, Institutional Review Board approval was not re- quired. All patients or their representatives were pro- vided with language-appropriate vaccine information statements as part of the vaccination protocol. This reflected local hospital policy.

Study sample

Study subjects included the population of Provi- dence St. Vincent Hospital and Medical Center inpa- tients during the months of September, October, and November, aged 65 years or greater, and inpatients be- tween 2 and 64 years of age with the presence of one or more risk factors for pneumococcal disease. For this study, high-risk patients were defined as those with medical conditions that significantly increase a patient’s vulnerability to pneumococcal pneumonia or invasive pneumococcal disease, making them eligible for pneu- mococcal vaccination if aged between 2 and 64 years. These factors included cardiovascular disease (IDC- 9-CM 331-336, 340-345, 347, 358, 359, 393-429, 440.9), cerebrovascular disease (ICD-9-CM 430-438), lower and chronic respiratory diseases (ICD-9-CM 466, 480- 487, 490-496, 500-505, 506.4, 508, 510-519), renal disease (ICD-9-CM 580-586), diabetes mellitus (ICD- 9-CM 250), chronic alcoholism and its complications (ICD-9-CM 291, 303, 571, 577), sickle cell disease (ICD- 9-CM 282.5), asplenia (functional or anatomic) (ICD-9- CM 45.1), and malignant neoplasms (ICD-9-CM 140-233). These risk factors were selected by physicians and other practitioners during design of the Providence St. Vin- cent program, reflecting ACIP-identified risk factors. Pa- tients who were excluded from the study were those in terminal or comfort care; those allergic to vaccine com- ponents; patients who received organ or bone marrow transplants in the year prior to the study; anyone who received pneumococcal vaccination in the 5 years prior to the study; patients in critical care, short-stay, emer- gency, and obstetrics units; and those whose physicians have excluded them from the vaccination protocol.

Statistical analysis

The intervention factor (independent variable) in this study was the implementation of a program for administration of the pneumococcal vaccine prior to discharge. Measurements included evaluation of medical record documentation of assessment for

Eckrode, Church, and English October 2007 511

vaccination eligibility (opportunity), actual vaccina- tion, patient age, and presence of ACIP-identified risk factors in the primary diagnosis (by ICD-9-CM code). The outcome (dependent) variables of interest were as- sessment for vaccination eligibility (provision of vacci- nation opportunity) and actual vaccinations given.

All statistical analyses were accomplished using the SPSS 10.1 for Windows (SPSS, Chicago, IL) software package. The x2 (goodness of fit/likelihood ratio tests) statistics were used for analysis of categorical variables, to determine the strength of association between the program implementation and rates of opportunity and vaccination, using the uncorrected x2 statistic for 2 3 2 contingency tables.45,46 Statistical significance was considered at P , .05.

RESULTS

Data gathered for the first 90 days after program implementation revealed that, of 5072 inpatients in the affected units, 1106 (28%) met the program criteria for pneumococcal vaccination. Data for the 90 days prior to program implementation revealed that, of 5543 inpatients in the affected units, 2874 (52%) met the program criteria for pneumococcal vaccination. Based on these data, a random sample of patients was drawn for further evaluation. Actual calculation of sample size was performed by use of the Raosoft Sample Size Calculator (Raosoft 2004, Seattle, WA). For the population eligible for pneumococcal vaccina- tion during the 90-day period after program interven- tion (N = 1106), the sample size was 286 (Table 1). Two hundred and eighty-six subjects were randomly selected from the postintervention population, based on a value obtained from a random number table. For the population eligible for pneumococcal vaccination during the comparison period (N = 2874), the sample

Table 1. Demographic characteristics of sampled patients, first 90 days postintervention (study group)

Characteristics N = 286 (%)

Patients

Male 99 (34.6%)

Age in years (mean) 70.91

Risk factors (n = 137)

Cardiac disease 64 (22.4%)

Cancer 51 (19%)

Pulmonary disease 12 (4.2%)

Renal disease 4 (1.4%)

Cerebrovascular disease 4 (1.4%)

Diabetes 1 (0.3%)

Chronic alcoholism 1 (0.3%)

Asplenia 0 (0%)

Sickle cell disease 0 (0%)

size was 338 (Table 2). A sample of 338 subjects from the population of vaccination-eligible inpatients in the 90-day period prior to program selection was se- lected randomly, using the same calculator and in the same random fashion as the study group. Of note, ex- amination of the demographics of the samples re- vealed anomalies, such as only one sampled patient with a diagnosis of diabetes, and smaller than expected numbers of patients whose sole risk factor was age greater than 65 years. These anomalies were noted, and were not considered significant when analyzing the data for the outcomes of interest.

For the 90 days prior to program implementation, 29 (8.6%) of the 338 sampled patients (comparison group) were provided vaccination opportunity. In this comparison group, none of the 338 vaccination-eligible patients received the pneumococcal vaccine. Of note, the demographics of the comparison group did not match those of the study group exactly. We did not consider these variations in demographics to be sig- nificant. Still, seasonal variations in disease patterns and aperiodic, nonseasonal changes in patient demo- graphics must be acknowledged.

For the 90 days after implementation of the vaccina- tion program, where registered nurses assessed the patients for vaccination eligibility and provided vaccina- tion under standing orders, 169 (59%) of the 286 sam- pled patients were provided vaccination opportunity, and 44 (15%) of the 286 sampled patients received pneumococcal vaccination. The vaccination rate for those subjects assessed under the standing order pro- gram was 26% (44 of 169). Again, the demographics of the comparison group did not match those of the study group exactly, but the variations were not considered to be significant. There are several plausible reasons for missed vaccination opportunities and the low vaccina- tion rate, and these are addressed in the Discussion.

Table 2. Demographic characteristics of sampled patients, 90 days preintervention (comparison group)

Characteristics N = 338 (%)

Patients

Male 119 (35.2%)

Age in years (mean) 72.92

Risk factors (n = 148)

Cardiac disease 73 (21.6%)

Cancer 42 (12.4%)

Pulmonary disease 15 (4.4%)

Cerebrovascular disease 13 (3.8%)

Renal disease 4 (1%)

Chronic alcoholism 1 (0.3%)

Diabetes 0 (0%)

Asplenia 0 (0%)

Sickle cell disease 0 (0%)

512 Vol. 35 No. 8 Eckrode, Church, and English

Further examination of the data revealed differences in vaccination opportunity between the population aged 65 years and older and the population aged be- tween 2 and 64 years of age. In the comparison group, vaccination opportunity was provided to 17 (26%) of the 340 patients aged 65 years and older; in the subpopula- tion aged between 2 and 64 years of age, vaccination op- portunity was provided to 12 (3.6%) of 334 subjects. In the study group, vaccination opportunity was provided to 140 (73.3%) of the 191 patients aged 65 years and older; in subpopulation aged between 2 and 64 years, vaccination opportunity was provided to 29 (30.5%) of 95 subjects. Again, the results revealed missed vaccina- tion opportunities that must be addressed.

Differences in actual vaccination rate also were identified within the study group, when stratified by age. The vaccination rate for the subpopulation of pa- tients aged 65 years and older was 24.3%. In the sub- population aged between 2 and 64 years, the actual vaccination rate was 52.6% (Table 3). The study pro- gram achieved an overall vaccination rate of 15.4%.

Following the descriptive analysis, the study data were examined and subjected to inferential analysis to determine whether the rates of opportunity and vaccination were influenced by the standing orders– based pneumococcal vaccination program. Using the x

2 (goodness of fit/likelihood ratio) test, the study found a significant statistical difference in the rate of pneumococcal vaccination opportunities (x2 = 182.46; P = .00) and the pneumococcal vaccination rate (x2 = 56; P = .00).

DISCUSSION

The field experiment method that was chosen for this study has a high external validity, and is well suited

Table 3. Results of pneumococcal vaccination program

Comparison

group

Study

group

Total number, (%) provided

vaccination opportunity

29 (8.6) 169 (59.1)

Total number, (%) given

pneumococcal vaccination

0 (0) 44 (15.4)

Subset analysis

Number, (%) provided

vaccination opportunity,

patients aged $ 65 years

17 (5.0) 140 (73.3)

Number, (%) patients given

pneumococcal vaccination,

aged $ 65 years

0 (0) 34 (24.3)

Number, (%) provided vaccination

opportunity, aged 2-64 years

with risk factors

12 (3.6) 29 (30.5)

Number, (%) patients given

pneumococcal vaccination,

aged 2-64 years with risk factors

0 (0) 10 (52.6)

for applied research that is focused on problem solving. This design was chosen as the alternative to dividing the postimplementation population into comparison and study groups, with one group eligible for evalua- tion and vaccination by the registered nurse and the other group reliant on the previous procedure of physi- cian evaluation and order (which may not have provided an opportunity for vaccination, perhaps be- cause many physicians and nurses did not actively encourage pneumococcal polysaccharide vaccination). There is a weakness associated with this study method, however. The weakness that directly affects this study is a result of the sample selection process, which did not draw sample subjects for the comparison and study groups from the same population. This was due to ethical considerations; we believed that patients could not be denied an efficacious vaccine to determine the success of a newly designed vaccination program.

Another area of concern was the accuracy of the existing data. The results of this study could be biased by errors in the original documents, because docu- mentation errors could lead to error in measurement. To avoid counting a study subject erroneously as not immunized or not provided with the opportunity for immunization, multiple documents were evaluated to determine if a lack of opportunity or nonadministra- tion of vaccine was in appearance only, due to poor or erroneous documentation. The evaluation of multi- ple documents was undertaken to avoid a potential source of bias identified in a previous study. In the pre- vious study, approximately 18% of vaccines ordered were not documented on the medication administra- tion record.32 Thus, multiple documents were evalu- ated to determine if opportunity for, and delivery of, the vaccine did, in fact, occur.

A limitation of this study that may impact on its ex- ternal validity is its timing. This study collected data that addressed only the 90 days immediately before and immediately after program implementation. The first 90 days after program implementation may rep- resent a period of time when interest in, and training on, the program are high. Thus, the results of this study may not predict program performance outside of the initial 90-day period, when the novelty may have worn off, training is in the past, and supervisory attention may be directed elsewhere. The Hawthorne Effect, whereby a new program brings about a short- term improvement in performance, cannot be dis- counted as an explanation for the results of this study, as well. Supporting this premise was an observed de- cline in assessment rates for the month immediately following this study, although assessment rates have improved since. Influenza vaccination was not offered during this study period, and as such, had no influ- ence on the results.

Eckrode, Church, and English October 2007 513

Many of the findings of this study are consistent with previous investigations. For instance, the initial vaccination rate for this study was 0%, as determined by analysis of the comparison group data. This is the same as the preprogram rate reported by Vondracek and colleagues.26 Scarbrough and Landis38 reported a vaccination rate of 1.2% when physicians managed the vaccinations, and a rate of 0.65% was reported un- der similar circumstances by Bloom and colleagues.2

These low vaccination rates are not surprising, and they typically provide much of the impetus for pro- gram design and implementation. An example of a pro- gram that exhibited these kinds of results was the previous Providence St. Vincent program, which relied on physician management of vaccination (without chart reminders), to little success. Others have had pos- itive results using chart reminders; one program used simple chart reminders to increase vaccination rates, and increased pneumococcal vaccination rates from 0% to 53% for patients on a general medical service.26

The outcomes of the study program are comparable to the outcomes reported for programs that were not based on standing orders (eg, a program that relied on family nurse practitioners for assessments, orders, and vaccinations). That program achieved a vaccina- tion rate of 16%, compared with the preimplementa- tion rate of 1.2%.38 As another example, Skledar et al,32 studying the outcomes of a hospital-based pneumococcal vaccination program that was depen- dent on a collaborative effort between physicians and pharmacists, reported vaccination rates of 11% and 31% for newly admitted patients and high-risk pa- tients, respectively. The overall rate of vaccination in the Providence St. Vincent program was higher than the overall rate reported by Skledar et al, but more sig- nificant is the report of a difference in vaccination rates between the total population and the high-risk subpop- ulation. This difference in vaccination rates between all patients and the subpopulation of high-risk patients occurred in the study program as well, with 52.6% of high-risk patients accepting and receiving pneumococ- cal vaccination.

When compared with other standing orders–based programs, whether nursing or pharmacy-driven, the overall vaccination rates of the Providence St. Vincent program often were lower than those reported in the literature. For example, in a pharmacy-based program, assessment rates of 90% and vaccination rates of 29.3% were achieved.47 Significantly, this program was based on a standing order that directed pharma- cists to do the patient assessment. In contrast, also following standing orders, the Providence St. Vincent registered nurses achieved an assessment rate of 59.1%, yielding 34.33% lower rates of vaccination op- portunity than did the pharmacy-based program. A

comparison with nursing-driven programs reveals sim- ilar outcomes. Rhew et al48 reported on a comparison between patient and provider reminders, a nurse stand- ing orders system, and a combination of patient and provider reminders and a nurse standing order system. The nurse standing order system achieved an opportu- nity rate (assessment for vaccination) of 39% and a vaccination rate of 22%. The St. Vincent program com- pares favorably with its assessment rate of 59.1%, but not the overall vaccination rate.

In an outpatient program that was reliant on stand- ing orders for nurses, the pneumococcal vaccination rate increased from a starting point of 34% to an end point of 63% overall.37 The outcomes for the outpa- tient study were for a 10-year period, however; the out- comes for the Providence St. Vincent study were for the first 90-day period. It could be argued that 10 years of program operation may have resulted in the incorpora- tion of the program into the organizational culture, and a habit of assessing and vaccinating may have been gained over this time period. Another result of this study warrants discussion and comparison. The pneumococ- cal vaccination rate for the high-risk population in the outpatient program (52%) was lower than that for the overall population.37 This is in direct contrast to the results of our study; the high-risk population was vaccinated at a higher rate (52.6%) than was the overall population (15.4%), despite lower rates of vaccination opportunity (30.5% versus 73.3% for the high-risk and 65 years and older populations, respectively).

Unfortunately, vaccination opportunities were missed frequently in the study program. Several bar- riers were identified as obstacles to successful imple- mentation of standing orders–based programs by Middleton et al,49 and provide plausible reasons for the missed opportunities in the study program. Among the provider-related barriers that were delineated by Middleton et al, several were identified within the study program. Concerns for additional workload were ex- pressed during pilot testing of the program; these con- cerns may have been carried forward to the program during the study period. Additionally, some nurses ex- pressed reluctance to administer a vaccination based on a standing order; they desired individual physician orders for each vaccination. Process breakdowns may have contributed to the missed vaccination opportuni- ties, as well; simple failure to assess newly admitted patients and documentation omissions or errors could account for a portion of the missed opportunities. The Providence St. Vincent program addressed patients on multiple services (eg, medical, cardiac, oncology, pul- monology); however, because the registered nurses were responsible for performing the assessment under standing orders that were not service-specific (and no service exempted its patients from vaccination), this

514 Vol. 35 No. 8 Eckrode, Church, and English

should not have had any influence on the overall outcome.

For the population of primary interest to most governmental and nongovernmental agencies (ie, pa- tients aged 65 years and older), the results of the study program show that vaccination opportunity increased significantly, from 5% to 73%. Although an increase of this degree was welcome, it was not as significant an outcome as it appears. Despite this increase in vac- cination opportunity, a considerable share of the sam- pled patients who were provided with that opportunity (74%) declined vaccination. This is more than twice the rate that was reported by Scarbrough and Landis,38

who documented a 35% refusal rate. The rate of decli- nation was slightly more than 2 times higher in the sub- population that was aged 65 years and older (75.7%) than it was in the high-risk subpopulation (34.5%). This negated any advantage that the 65 years and older group may have gained from increased vaccination op- portunity. Reasons for vaccination declination in this study program were documented rarely, which pre- cluded determination of the underlying factors without direct patient interview. Although determination of the factors that drive these declinations are outside the scope of this study, this characteristic also was identi- fied during pilot testing of the Providence St. Vincent program, and has a deleterious effect on vaccination rates. Internal discussion and further review of the lit- erature suggest that lack of direct physician recom- mendation for vaccination, and the fact that some patients were unclear on, or could not recall, their cur- rent vaccination status may have been reasons for dec- lination of the vaccine; however, further study is needed to support this hypothesis.33 Patient or patient representative concerns about the vaccine, concerns that vaccination may not reflect the wishes of the patient’s physician, and resistance to vaccination were suggested as patient barriers to vaccination by Middleton et al.49 It does not seem unreasonable to pro- pose that these barriers existed in the study program, as well. Regardless of the reason, the refusal of vaccina- tion by patients after they have been provided with vaccination opportunity, and in some cases, the appro- priate vaccine information statement and encourage- ment by health care practitioners, is of concern. The reasons for these refusals need examination if vaccina- tion rates are to increase much beyond current levels.

Our inpatient pneumococcal vaccination program, targeted to all patients 65 years and older and those patients ranging from 2 to 64 years with risk factors for pneumococcal bacteremia, resulted in an overall increase in vaccination opportunity and vaccination rates, when compared with our previous method. The overall rates of vaccination that were attained by this program often were lower than those reported in the

existing literature for other program designs; however, this may have been due to an unusually high rate of vaccination refusal.

The program that was evaluated in this study seems to generate sustainable gains over time. Examination of the data 1 year after the end of this study revealed an overall vaccination rate of 39%, compared with the overall vaccination rate of 15.4% during the study pe- riod. Still, this rate was less than our institutional goal of 80% vaccination, and the program continues to re- ceive strong support and emphasis.

Based upon the results of this study, we believe that a standing orders–based, nursing-driven pneumococ- cal vaccination program can increase vaccination rates in the inpatient setting. Barriers to program function, and particularly vaccination receipt, must be examined and overcome if the program is to generate acceptable levels of vaccination opportunity and administration. Given the significance of pneumococcal bacteremia to morbidity and mortality, the increasing prevalence of drug-resistant strains of S pneumoniae, and the positive outcomes of our program, we believe that inpatient pneumococcal vaccination programs have usefulness and merit and can improve public health. We strongly encourage their implementation.

References

1. Centers for Disease Control and Prevention. Pneumococcal disease.

In: Epidemiology and prevention of vaccine-preventable diseases. 9th

ed. Atlanta (GA): Centers for Disease Control and Prevention;

2006. p. 256-7.

2. Bloom HG, Wheeler DA, Lynn J. A managed care organization’s attempt

to increase influenza and pneumococcal immunizations for older adults

in an acute care setting. J Am Geriatr Soc 1999;47(1):106-10.

3. McDaniel LS, Swialto E. Pneumococcal disease: pathogenesis, treat-

ment, and prevention. Infect Dis Clin Pract 2004;12(2):93-8.

4. Jackson LA, Neuzil KM, Yu O, Benson P, Barlow WE, Adams AL, et al.

Effectiveness of pneumococcal polysaccharide vaccine in older adults.

N Engl J Med 2003;348:1747-55.

5. Nichol KL, Baken L, Wourenma J, Nelson A. The health and eco-

nomic benefits associated with pneumococcal vaccination of elderly

persons with chronic lung disease. Arch Int Med 1999;159(20):

2437-42.

6. Torres JM, Cardenas O, Vasquez A, Schlossberg D. Streptococcus pneu-

moniae bacteremia in a community hospital. Chest 1998;113(2):387-90.

7. Williams R. Pneumococcal vaccination. Lippincotts Prim Care Pract

1998;2:625-33.

8. Maki DG. Pneumococcal bacteremia: lessons learned, yet more to

learn. Mayo Clinic Proc 2004;79(5):599-603.

9. Treanor JJ, Hall CB. Influenza and infections of the trachea, bronchi, and

bronchioles. In: Betts RF, Chapman SW, Penn RL, editors. Reese and

Betts’ a practical approach to infectious disease. 5th ed. Philadelphia:

Lippincott Williams and Wilkins; 2003. p. 278-94.

10. Musher DM, Alexandri I, Gravis EA, Yanbeiy N, Eid A, Inderias L, et al.

Bacteremic and nonbacteremic pneumococcal pneumonia: a prospec-

tive study. Medicine 2000;79(4):210-21.

11. Bratzler DW, Houck PM, Jang H, Nsa W, Shook C, Moore L, et al.

Failure to vaccinate Medicare inpatients: a missed opportunity. Arch

Int Med 2002;162:2349-56.

Eckrode, Church, and English October 2007 515

12. Holdiness MR. Rare systemic dermatologic reaction after pneumococ-

cal vaccine administration. South Med J 2003;96(1):64-5.

13. Nichol KL, MacDonald R, Hauge M. Side effects associated with pneu-

mococcal vaccination. Am J Infect Control 1997;25(3):223-8.

14. Jackson LA, Benson P, Sneller VP, Butler JC, Thompson RS, Holder P,

et al. Safety of revaccination with pneumococcal polysaccharide vac-

cine. JAMA 1999;281(3):243-8.

15. Lackner TL, Hamilton RG, Hill JJ, Davey C, Guay DRP. Pneumococcal

polysaccharide revaccination: immunoglobulin G seroconversion, per-

sistence, and safety in frail, chronically ill older subjects. J Am Geriatr

Soc 2003;51:240-5.

16. Fox BC. Leukocyctoclastic vasculitis after pneumococcal vaccination

[letter]. AJIC 1998;26(3):365-6.

17. American Public Health Association. Pneumococcal pneumonia. In:

Chin J, editor. Control of communicable diseases manual. 17th ed.

Washington, DC: American Public Health Association; 2000. p. 388.

18. Christenson B, Lundbergh P, Hedlund J, Örtqvist Ä. Effects of a large-scale

intervention with influenza and 23-valent pneumococcal vaccines in adults

aged 65 years or older: a prospective study. Lancet 2001;357:1008-11.

19. Dworkin MS, Ward JW, Hanson DL, Jones JL, Kaplan JE. Pneumococ-

cal disease among human immunodeficiency virus-infected persons:

incidence, risk factors, and impact of vaccination. Clin Infect Dis

2001;32:794-800.

20. Fedson DS, Harward MP, Reid RA, Kaiser DL. Hospital-based pneumo-

coccal immunization: epidemiological rationale from the Shenandoah

Study. JAMA 1990;264:1117-22.

21. Harrison LH, Dwyer DM, Billman L, Kolczak MS, Schuchat A. Invasive

pneumococcal infection in Baltimore, MD: implications for immunization

policy. Arch Int Med 2000;160:89-94.

22. Lund BC, Ernst EJ, Klepser ME. Strategies in the treatment of penicillin-

resistant Streptococcus pneumoniae. Am J Health Sys Pharm 1998;55:

1987-94.

23. Penn RL, Betts RF. Lower respiratory tract infections (including tuber-

culosis). In: Betts RF, Chapman SW, Penn RL, editors. Reese and Betts’

a practical approach to infectious diseases. 5th ed. Philadelphia: Lippin-

cott Williams & Wilkins; 2003. p. 295-312.

24. Sisk JE, Whang W, Butler JC, Sneller VP, Whitney G. Cost-effectiveness

of vaccination against invasive pneumococcal disease among people

50 through 64 years of age: role of comorbid conditions and race.

Ann Int Med 2003;138:960-8.

25. Whitney CG, Schaffner W, Butler JC. Rethinking recommendations for

use of pneumococcal vaccine in adults. Clin Infect Dis 2001;33:662-75.

26. Vondracek TG, Pham TP, Huycke MM. A hospital-based pharmacy

intervention program for pneumococcal vaccination. Arch Int Med

1998;158:1543-7.

27. Centers for Disease Control and Prevention. Facilitating influenza and

pneumococcal vaccination through standing orders programs. MMWR

Morb Mortal Wkly Rep 2003;52(4):68-9.

28. Nichol KL, Zimmerman R. Generalist and subspecialist physicians’

knowledge, attitudes, and practices regarding influenza and pneumo-

coccal vaccinations for elderly and other high-risk patients. Arch Int

Med 2001;161:2702-8.

29. Peterson RL, Saag K, Wallace RB, Doebbling BN. Influenza and

pneumococcal vaccine receipt in older persons with chronic disease:

a population-based study. Med Care 1999;37:502-9.

30. Poland GA, Shefer AM, McCauley M, Webster PS, Whitley-Williams

PN, Peter G, et al. Standards for adult immunization practices. Am J

Prev Med 2003;25(2):144-50.

31. Centers for Disease Control and Prevention. Influenza and pneumo-

coccal vaccination coverage among persons aged $65 years and per-

sons aged 18-64 years with diabetes or asthma-United States, 2003.

MMWR Morb Mortal Wkly Rep 2004;53:1007-12.

32. Skledar SJ, Hess MM, Ervin KA, Gross RR, Nowalk MP, Carter H,

et al. Designing a hospital-based pneumococcal vaccination program.

Am J Health System Pharm 2003;60:1471-6.

33. Santibanez TA, Nowalk MP, Zimmerman RK, Jewell IK, Bardella IJ,

Wislon SA, et al. Knowledge and beliefs about influenza, pneumo-

coccal disease, and immunizations among older people. J Am Geriatr

Soc 2002;50:1711-6.

34. Robinson KA, Baughman W, Rothrock G, Barrett NL, Pass M, Lexau

C, et al. Epidemiology of invasive Streptococcus pneumoniae infections

in the United States, 1995-1998: opportunities for prevention in the

conjugate vaccine era. JAMA 2001;285:1729-35.

35. Centers for Disease Control and Prevention. Prevention of pneumo-

coccal disease: Recommendations of the Advisory Committee on

Immunization Practices (ACIP). MMWR Recomm Rep 1997;46(RR-8):

1-24.

36. Task Force on Community Preventive Services. Recommendations

regarding interventions to improve vaccination coverage in children,

adolescents, and adults. Am J Prevent Med 2000;18(15):92-140.

37. Nichol KL. Ten-year durability and success of an organized program

to increase influenza and pneumococcal vaccination rates among

high risk adults. Am J Med 1998;105(5):385-92.

38. Scarbrough ML, Landis SE. A pilot study for the development of a hos-

pital-based immunization program. Clin Nurse Spec 1997;11(2):70-5.

39. Metersky ML, Fine JM, Tu GS, Mathur D, Weingarten S, Petrillo MK,

et al. Lack of effect of a pneumonia clinical pathway on hospital-based

pneumococcal vaccination rates. Am J Med 2001;110(2):141-3.

40. Centers for Disease Control and Prevention. Advisory Committee on

Immunization Practices (2004, September 24). Available from: http://

www.cdc.gov/nip/ACIP/default.htm. Accessed October 24, 2004.

41. Centers for Disease Control and Prevention. Facilitating influenza and

pneumococcal vaccination through standing orders programs. MMWR

Morb Mortal Wkly Rep 2003;52(4):68-9.

42. Department of Health and Human Services. Federal Register 61808.

Baltimore, MD. 2002.

43. McKenzie JF, Smeltzer JL. Implementation: strategies and associated

concerns. In: Planning, implementing, and evaluating health promotion

programs: a primer. 3rd ed. Boston: Allyn and Bacon; 2001. p. 258-9.

44. Shi L. Experimental research. In: Health services research methods.

Albany (NY): Delmar Publishers; 1997. p. 143-64.

45. Motulksy H. The design of clinical trials. In: Intuitive biostatistics. New

York: Oxford University Press; 1995. p. 183-91.

46. Rosner B. Hypothesis testing: categorical data. In: Study guide for

fundamentals of biostatistics. 4th ed. Belmont (CA): Wadsworth Pub-

lishing; 1995. p. 135-72.

47. Noped JC, Schomberg R. Implementing an inpatient pharmacy-based

pneumococcal vaccination program. Am J Health System Pharm 2001;

58:1852-5.

48. Rhew DC, Glassman PA, Goetz MB. Improving pneumococcal vaccine

rates-nurse protocols versus clinical reminders. J Gen Intern Med

1999;14:351-6.

49. Middleton DB, Fox DE, Nowalk MP, Skledar SJ, Sokos DR, Zimmerman

RK, et al. Overcoming barriers to establishing an inpatient vaccination

program for pneumococcus using standing orders. Infect Control

Hosp Epidemiol 2005;26:874-81.