Lesson learned 6

Root cause analysis: An effective Ql tool

K aren Z u lk o w s k i

DNS, RN Email drkarenz@aol.com

ABSTRACT

The purpose o f a root cause analysis (RCA) is to clarify exactly which adverse event has happened, determine why it happened and help prevent it occurring again. It is not punitive to any individual; rather the goal o f an RC A is to prevent fu tu re harm by eliminating A LL errors or issues that underlie adverse events. It could also be used to determine ivhy something worked well. A n R C A looks at all the reasons something happened that are not immediately obvious, rather than focusing on a single cause.

A n RC A helps identify these underlying problems using a systems approach to identify active errors (errors occurring at the point of contact between humans and a complex health care system) and latent errors (hidden problems w ith in health care system s that contribute to adverse events). It looks at the "big" picture1. The problems that are uncovered may result in changes in reporting, record keeping or process. It is im portant to keep in m ind that m u ltip le errors and system fla w s m u st intersect fo r a critical incident to reach the patient. Labelling one or even several o f these fa cto rs as the cause, w ith o u t consideration o f all factors, may obscure the true picture o f patient care.

USING ROOT CAUSE ANALYSIS FOR ADVERSE EVENTS

U nfortunately, adverse events h ap p en d u rin g m edical care. Som e m ay be life-th reaten in g , m ajor erro rs, w h ile o th ers m ay be problematic b u t not life-threatening2. U nderstanding w h y p ro b le m s o ccu rred can h e lp c h an g e care p ro cesses an d im prove p atien t outcom es. Root cause analysis (RCA) can be especially h elp fu l w h en ex am in in g w h y a p a tie n t developed a pressure injury. Often both patients and families attribute the developm ent of pressure injuries to an event that should never occur. However, patient factors such as medical conditions, m edications, as well as facility issues are often the actual "cause".

The aim of an RCA is to clarify w h ich ad v erse ev en t has occurred, determ ine the causes and help prevent it happening again3. It is n ot in ten d ed to be p u n itiv e to any individual; ra th e r th e goal of an RCA is to p re v e n t fu tu re h a rm by elim inating any errors or issues th at m ay u n d erlie adverse

events. It co u ld also be used to assess a process th a t has w orked well. It looks at all the causes rather than focusing on a single cause.

A n RCA uses a system s approach to identify active errors (errors occurring at the point of contact betw een hum ans and a complex system) and latent errors (hidden problem s w ithin health care systems that contribute to adverse events). It looks at the "big" picture. The problem s th at are uncovered m ay require a change in reporting, record keeping or process. It is im p o rta n t to rem em ber th a t m u ltip le errors an d system flaws m ust intersect for a critical incident to reach the patient. Labelling one or even several of these factors as the cause, w ith o u t co n sideration of all factors, m ay obscure the true picture of caring for a p atient's w o u n d 4.

P erform ing an RCA m u st be an in terd iscip lin ary process. It n e e d s to in v o lv e th o s e p e r s o n s m o s t fa m ilia r w ith w h a t h a p p e n e d , b u t also th e p e o p le fam ilia r w ith o th e r d e p a r t m e n t s th a t in f lu e n c e d o r c o n t r i b u t e d to w h a t h a p p e n e d . C o n d u c tin g an RCA m ean s g o in g d e e p e r by asking the question w hy at each level of cause an d effect. The goal is to identify changes for the system and, as such, it should be conducted in a m an n er th at is as im partial as possible.

To be thorough, an RCA m ust include: exam ination of hum an an d system factors; analy sis of the u n d e rly in g cause and effect through a series of "w hy" questions; identification of risks th a t occu rred an d th eir p o te n tia l co n trib u tio n s; and d e te rm in a tio n of w hich im p ro v e m e n ts cou ld be m a d e to the processes w ith in the system to p re v e n t this o ccurring again. To be cred ib le, an RCA m ust: in v ite p a rtic ip a tio n by o rganisational lead ersh ip as w ell as those m ost closely in v o lv e d in th e p ro c e s s e s a n d s y ste m s; be in te r n a lly consistent; include consideration of relev an t literatu re and examine the basis for best practice4.

The process also in v olves rep eated ly asking the q u estio n why, sometimes called the rule of five whys. The event that happened is the outcome. For example, a patient developed

w w w . w c e t n . o r g

a pressure injury, which is an adverse outcome. Why did it develop? The five whys were:

1. He was immobile.

2. He refused to move because of pain.

3. He was not offered pain medication before an attempt to turn him.

4. The certified nursing assistant (CNA) did not know to ask the nurse.

5. Lack of education, and so on.

In this example, the direct cause is lack of movement, but the indirect causes are the most significant and require changes in the system.

THE "HOW" OF RCA:

1. An RCA begins with data collection and reconstruction of the event in question through record review and participant interviews. It is im portant to stress to staff that this process is not designed to blame any individual(s).

2. The interdisciplinary RCA team analyses the sequence of events leading to the event or error, with the goal of identifying how the event occurred (through identification of active errors) and why the event occurred (through system atic identification and analysis of latent errors). It looks at all departments and processes within the facility. This is the five whys stage.

3. The RCA process then identifies where changes/ improvements should be made.

4. The changes are trialled or pilot-tested.

5. Outcomes are tracked and examined.

6. The process to be changed is refined.

7. The change is made facility-wide.

8. Re-evaluation is ongoing.

In other words, exactly what happened, why did it happen and where did it happen? Problems are identified in:

1. Policy/procedures.

2. Safeguards (barriers and controls).

3. Environment.

4. Equipment.

5. Information technology.

6. Fatigue in scheduling.

7. Training.

8. Communication.

9. Other issues.

Once the issues have been identified then required changes need to be planned and im plem ented. This includes addressing how this problem can be corrected so it does not reoccur. Other considerations include deciding which changes will be implemented and who is responsible for the implementation. The process must be tested and evaluated, then redesigned or modified, as required. This process is often represented as a fishbone diagram (see Figure 1). It considers: equipment/supply factors; environmental factors (location, physical layout, and safety); policy/procedures/ rules factors (standards or compliance with standards, and documentation); and people/staff factors (ability/ supervision/staffing, lack of knowledge or information, scheduling, and communication). One of the best ways to learn how to follow this process is with an example. The results are often conceptualised in a fishbone or Ishikawa diagram to help see the bigger picture5-6. However, this is just a guide to the question how and is often rather intimidating to people using it for the first time. The important point is to remember this is just a tool to help look at quality improvement in a bigger picture7. An RCA is effective when looking at the big picture with pressure injury8-10.

CASE STUDY: AMANDA'S STORY — HOW DID THIS WOUND HAPPEN?

Day 1: Amanda is brought to the emergency room (ER) via ambulance and on a backboard at 1930 on Saturday. Her daughter arrives shortly thereafter. Amanda is a 78-year- old widow, with a history of arteriosclerotic heart disease (ASHD), diabetes mellitus (DM) and hypertension. She had a coronary artery bypass graft (CABG) six years ago. Currently, she is complaining of nausea, blurred vision and pain in her left hip. The ride in the ambulance was 45 minutes. When Amanda arrives, the ER is very busy, with patients from a traffic accident. She is put in a trauma bay and her vital signs are taken right away. The ER is short-staffed by two registered nurses (RNs) due to a local flu outbreak. Therefore, her history isn't taken for another 30 minutes (2000) and the physician arrives at 2015.

Amanda appears to be alert and oriented to time and place but is a poor historian. However, her daughter Ginger tells you her mother has had periods of confusion over the past several days and fell at home yesterday and again today, although she is sure her mother was "only on the floor for 15 minutes before we found her". She goes on to say that she recently moved her mother into her home from out of town, because she wasn't looking after herself or taking her medications. In the last six months, she has lost 30 lb. As a result of the long-distance move, no previous medical records are immediately available.

Ginger tells you that her mother had been receiving medical care for episodes of atrial fibrillation and flutter prior to the move. Currently she is on 0.125 mg of digoxin daily, and quinidine sulphate and Catapres, but she can't remember the

WCET Journal Volume 38 Number 1 - Januaty/March 2018

E q u ip m e n t/S u p p ly Factors

\ Lack o f proper

E n v iro n m e n ta l Factors t | R o o t C a u s e A n a l y s i s | , J g | §

\L a y o u t

\ equipment V is ib ility \ Lack of oversight of \ equipm ent \ \ Space

1C Cl 1C ■ I w O l I C

D o c u m e n ta tio n / /

S t a f f i n g / back o f k n o w le d g e o r issues / / tra in in g/ Su pervisio n / C o m m u n ic a tio n issues

/ O u td a te d

/ p o lic ie s /p ro c e d u re s / S ch edu lin g

P o lic y /p ro c e d u re s /r u le s

/ P e o p le /s ta ff

dose or times for these. Amanda has rales in her lung sounds and 3+ pitting oedema in her lower extremities. She is started on 4 L of oxygen ( 0 2) via nasal cannula and intravenous (IV) Lactated Ringers at 100 m l/hr. She is transferred to the cardiac intensive care unit (ICU) at 0145. The time on the backboard is 90 minutes. Her stay in ER is 4.5 hours, where a hip and chest X-ray are taken. The chest X-ray showed fluid and presence of pneumonia in both lower lobes of her lungs, but the hip results were not received until after her transfer to the ICU.

Day 2 in the ICU: the cardiac monitor shows atrial fibrillation with a ventricular rate of 189 beats per minute (BPM). Her blood pressure (BP) is 120/82. She has pulse oximetry (pulse ox), and a 12-lead electrocardiogram (ECG), portable chest and hip X-ray and serum digoxin level are ordered. Pulse ox is 80% and Oz is increased to 6 L. She is started on Lasix. The hip X-ray shows a left hip fracture (Fx). The orthopaedic department is consulted, but surgery is deemed not possible until she is medically stable; they order a morphine pump to control the pain, and no turning to her left side.

After being in ICU for four hours, Amanda is more confused b u t her 0 2 statistics are now 90% (0545). She becomes incontinent of urine and tries to climb out of bed several tim es, pulling off her 0 2 and p u llin g out her IV. She is restrained for the next eight hours, but documentation shows she was turned only once (1345). Unfortunately, when she was placed on her right side her 0 2 statistics dropped and she

has to be maintained on her back with the head of her bed elevated. An adult diaper is used to manage the incontinence when excessive due to the Lasix given. Her Braden scale score was 14 on day 2 at 1130.

A fter another eight h ours (2145) h er m edical condition stabilises enough that the restraints can be removed. She is bathed and at this time a stage III pressure injury and a friction injury are noted on her coccyx.

Think about Amanda to answer these questions but imagine if this case happened in your hospital. The idea is to look at the big picture. No one departm ent or person is to blame. Rather there were multiple failures, both in the care provided and the system. The goal is to improve care by addressing the large issues.

So, who is at fault? The answer is no one. Go through the case step by step. Root cause does not blame — it improves and is a positive.

So, the problems are multifactorial:

1. Equipment: H ad not been evaluated for condition, so pressure redistribution was not being adequately provided.

2. Rules/policy/procedure: Training was not adequate for w o u n d p re v e n tio n . There w as no p o licy on communication between departm ents or oversight of cumulative time spent on hard surfaces. There was poor

___________________________________________________________________________w w w . w c e t n . o r g

Look at the p atien t and the event

A m anda's prim ary diagnosis (Dx) and secondary Dx ASHD, DM and hypertension, confusion, decreased O,, w eight loss

W hich of these increased pressure injury risk?

W hich increased m oisture dam age risk?

Decreased micro and macro vascularisation, potential decreased nutrition, decreased oxygenation

Incontinence, climbing o ut of bed increased friction

W hen and w here w ere the w o unds first noted? In ICU 2145

W hen w as the medical provider notified? N ot docum ented

Was family notified of the w o u n d /w h e n ? N ot docum ented

Was skin team consulted? Y /N N ot docum ented

Was a dietary consultation ordered? N ot docum ented

1. Was a skin assessm ent docum ented and where?

Am bulance No

ER No

ICU N ot until w ound w as noted

2. Was a pressure injury risk assessm ent scale completed on ICU admission?

Day 2 at 1130

3. W hat pressure injury and w o u n d risk factors were problem atic and should have been addressed in the care plan?

Immobility-related pain and hip Fx and restraints

Decreased turning sites available

Incontinence / m oisture

N utrition

Decreased tissue perfusion

4. W hat medical conditions and changes in condition increased A m anda's pressure injury risk in the ICU?

H ip Fx

Decreased oxygenation

Incontinence

5. Was skin reassessed and docum ented every 4 or 8 hours in ICU?

No

6. Was m oisture adequately addressed? No and it resulted in m oisture-associated skin dam age

Think about w h at else could have been c au sin g / contributing to increased confusion and incontinence. Were appropriate tests conducted?

C ontributing factor: Lasix for oedem a increased urine o u tp u t w ith no toileting plan

Possible urinary tract infection (UTI)

7. W hen did staff last have training on pressure ulcer prevention?

One year ago, at skills day

U nknow n for staff hired to ICU in past year

W hat d e p a rtm en ts/p eo p le w ere connected to the event? Think broadly (and not just the obvious ones)

Family Am bulance ER ICU X-ray Education N ursing Medicine

WCET Journal Volum e 38 N um b er 1 - janu ary/M arch 20 18

►► Continued.

Staffing ER was inadequately staffed as 2 RNs were ill and not replaced

Surface/transportation factors Time spent - on backboard - in ER - in restraints

Surface evaluation for each department was not done for the past 3-4 years

Intra- and interdepartmental communication Time spent in ambulance, on backboard, and in X-ray was not communicated between departments

documentation of risk and skin condition. The required consultations were not carried out. Care planning was not adequate for risk. Wound policy and procedures had not been updated for four years and were no longer based on current evidence.

3. People: The ER was short-staffed, which impacted time to treatment and skin examination. The patient was a poor historian.

4. Environment: Facility building was not at fault.

You need to decide: Could this pressure injury have been prevented? What should this facility do next?

CONCLUSION

Multiple issues cause adverse events. A pressure injury is not the fault of one person; rather it can be caused by a series of failures that occur. Yet nurses often feel that an RCA will blame them for the pressure injury and they will be penalised. The purpose of an RCA is to improve patient care by examining and correcting all the contributing factors. This is part of an ongoing quality improvement process.

In the above example, m ultiple failures occurred from before admission, to unit and system issues. Unfortunately, the patient is the one that suffered the consequences and developed a pressure injury. If everything was done correctly and an RCA found no errors, the pressure injury would be unavoidable. Everything was done that could be done. An RCA can highlight good care as well as any deficiencies.

REFERENCES 1. Black NH & Vernetti B. Root-cause analysis. Prof Saf 2015; 60-62. 2. Makary MA & Daniel M. Medical error — the third leading cause

of death in the US. BM] [Online] 2016; 353. 3. Uberoi R el al. Root cause analysis in healthcare. Apollo Medicine

2007; 4(l):72-75.

4. AHRQ. Root cause analysis. 2017 [cited 2017 November 20]. Available from: h ttp s ://p s n e t.a h rq .g o v /p rim e rs /p rim e r/1 0 / root-cause-analysis.

5. AHRQ. How to use the fishbone design, [cited N ovem ber 20, 2017], Available from: h ttp s ://w w w .c m s.g o v /m e d ic a re / p r o v id e r - e n r o llm e n t- a n d - c e r tif ic a tio n /q a p i/d o w n lo a d s / fishbonerevised.pdf.

6. ASQ. Fishbone (Ishikaw a) d iagram . N ovem ber 20, 2017]; Available from: h ttp ://a s q .o r g /le a r n -a b o u t- q u a lity /c a u s e - analysis-tools/overview/fishbone.html.

7. Kader Ali NN, Wilson P & Mohammad 1Y. Symptoms Versus Problems Framework (SVP): an innovative root cause analysis tool. International Journal of Organizational Innovation 2014; 7.

8. Dolansky MA et al. N ursing student medication errors: a case study using root cause analysis. J Prof Nurs 2013; 29(2):102-108.

9. Nixon J et al. Severe pressure ulcer study. 2015. 10. Samuriwo R. Pressure ulcer-related harm: beyond root cause

analysis. J Wound Care 2015; 24(8):331—331.

www.wcetn.org

Copyright of World Council of Enterostomal Therapists Journal is the property of Cambridge Publishing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.