In a recent BMJ article entitled ‘Our current approach to root cause analysis: is it contributing to our failure to improve patient safety?‘, the authors’ state “Although RCA’s have been proposed as a mechanism for change, safety scientists believe the lack of improvement of adverse event rates in healthcare is largely because our methods of approaching change are ineffective”.
The abstract states the study aims to build on prior work by examining the types of solutions proposed in response to RCA’s conducted. So the focus is on the type of solutions proposed, and the potential reasons for their ineffectiveness over the past 15 years. Understanding this study is coming from researchers and academia, I would like to reach out to the front lines, the analysts who actually conduct the RCA’s.
If we are proposing RCA solutions that are ineffective, we have to understand 1) were the identified root causes inadequate, insufficient or wrong and/or 2) were the root causes identified adequate/sufficient and our recommendations not appropriate to effectively address them?
Having been in the Reliability and RCA business for 30+ years, primarily as a practitioner, I find that it is our ‘Shallow Cause Analysis (SCA)’ approaches that lead to the identification of inadequate/insufficient/wrong root causes. For an example of this, please read ‘Are All RCA Approaches Created Equal?‘. This is because often our RCA approaches themselves lack depth (do not drill down past the decision-makers and uncover system flaws adversely influencing decision-making) and/or lack comprehensiveness (focus on asking ‘Why’ [linear thinking] versus ‘How Could’ to uncover parallel actions taking place that led to bad outcomes).
The referenced BMJ research paper only addresses RCA solutions, and does not focus on ineffective RCA approaches. So our recommendations can only be as good as our identified root causes.
Root Cause Analysis (RCA) has a public relations problem! Whenever an organization seems to adapt ‘RCA’; when the organization does not produce overall results, they attribute it to RCA not working. Seems logical right? Wrong.
There over around 6,000 hospitals in the U.S. and 95%+ are accredited via various accrediting agencies. We know this because they are proud of this accreditation and put up their certification in the front lobby to let the patients and families see and gain confidence in that accreditation. Part of this accreditation process involves accrediting their RCA programs and their ability to meet the agency’s requirements.
So if the majority of hospitals in the U.S. have accredited RCA efforts, why do the overall patient safety numbers continue to slide? Preventable medical errors persist as the No. 3 killer in the U.S. – third only to heart disease and cancer – claiming the lives of some 400,000 people each year.
Is it a logical deduction that RCA compliance does not equal patient safety? Shouldn’t compliance equate to improved patient safety? We should do a valid RCA on this paradox:-)
So why isn’t RCA working effectively in HC, when it is has been field-proven in other industries for decades?
Whenever anyone is time pressured, they will likely opt for the path of least resistance. It’s human nature. Caregivers no doubt are very busy people and we should respect that because their ‘product’ is the quality of our lives. Most of us work in businesses that make widgets or provide other types of services, but we are not responsible for quality of life like those in hospitals.
How does this time pressure affect RCA? It is very likely the overwhelming majority of caregivers did not go to school to learn how to do RCA. It was likely not included in their curriculum at all. So when faced with a regulatory/standards requirement to do it, this is often viewed as another burden on my already full plate.
So I think to myself, how do I quickly satisfy this compliance requirement with the least impact on my ‘real job’? This is the thinking that leads us to the path of “Shallow Cause Analysis” (SCA). Success to me, at this point, is compliance. Actual patient safety is a pipe dream. This is because the time to do an effective RCA is taking me away from my patient time…isn’t this contrary to my purpose here in the hospital?
This reminds me of a saying:
“We never seem to have the time and budget to do things right, but we always seem to have the time and budget to do them again”
This pushes us to take shortcuts in our analyses in the interest of providing more care to the patient.
What do such short cuts look like and how do they impact an RCA?
1. We tend to accept hearsay as a fact as we do not have the time to go out and collect hard evidence to back up assumptions/opinions.
2. We tend to ask only “Why” and not “How Could”. While this seems trivial on the surface, think about the difference in the population of answers. I do not consider 5-Whys a valid RCA approach because it asserts A) there is only a linear path to failure, B) opinion serves as fact and 3) there is only one root cause to any event. These are simply not true. The fact is that things happen in parallel (asking ‘How Could’ connotes more than one answer) to cause undesirable outcomes. Unfortunately there are more often than not, multiple root causes and contributing factors to bad outcomes. I don’t know why they stop at ‘5’ whys…why not 3 or 8? True RCA will seek to find the multiple paths of failure that combined to cause the bad outcome and use solid evidence to back it up.
3. Time pressure can force us to use ‘pick list’ RCA. This means there are a series of drop down menus where we select the option closest to what we ‘think’ happened. The potential issue with this route is that all the possibilities of what could have happened, will never ALL be in our drop downs. Therefore we will pick what is closest. Eventually this type of RCA drives us to a predetermined conclusion. However, it makes life easier for those compiling the RCA statistics because it is easier to trend across the RCA database. The problem is we are making it easier to trend potentially bad information:-)
4. We tend to use regulatory/standards RCA forms because we believe that by using that form, there is a greater chance the auditor will accept my submittal. On the surface, this is probably true (unfortunately). Most regulatory and standards’ RCA forms are very similar in format. They are a ‘one-size-fits all’ solution to RCA. We ask the same questions no matter the event. This is called ‘categorical’ RCA in our business. We ask about the standard cause categories and what we felt could have gone wrong in them. Then we fill in the blanks with what we ‘think’ went wrong, because we don’t have the time to collect the real evidence. Such forms are desirable for the receiving agencies for the same reasons as described in the Pick List item (3), it provides consistency of how information is submitted so it is easier to compile statistics. This is fine, but where are the correlations of accepted RCA’s to bottom-line improvements in patient safety?
Why Many RCA Efforts fail (Video: Why RCA Efforts Are Not Meeting Expectations):
1. Little to no corporate expectations for the effectiveness of the RCA effort (no tie to patient safety). Who knows their ROI’s for their RCA effort?
2. Inadequate management support and lack of proper infrastructure to support RCA analysts.
3. Inadequate training in effective RCA methods and tools. We do the best with what we have and cannot afford tools that make it easier to perform the analyses and compile the information.
4. Lack of accountability for implementing corrective actions and tying them to a quantifiable patient safety metric (compliance alone is often the end game). Is anyone held accountable for an RCA effort to close the gap in unacceptable patient safety metrics?
5. Lack of adequate resources provided to the analysts to make conducting the analyses easier and more comprehensive (e.g. – technology solutions, conference rooms with proper A/V, etc.)
6. Lack of adequate policies and procedures to support RCA (they are written for compliance and not directed at patient safety). When I try to get the docs to participate, this is viewed beneath them. I cannot cite a policy that requires them to participate. If there is no negative consequence for not participating, why should they?
7. Lack of leadership’s to accept that they may be a contributor to the undesirable outcome. RCA is about uncovering the truth and oftentimes, leadership declines to look in the mirrorand accept that truth. It is difficult for us to admit we may be part of the problem.
For those on the RCA front lines, what are the reasons you believe that RCA’s are being viewed as ineffective in HC (as opposed to other industries where RCA has been field-proven for decades to be an invaluable process)? What are the hurdles you face to doing RCA the ‘right way’? In your organization does RCA compliance correlate to measurable, bottom-line improvements in patient safety (if so, please cite some of your successes)? If not, why not?
Note: I revived this blog from several years ago to reflect and see if we have made any headway on this most important topic, and can we prove patient’s are safer as a result of our efforts?