SafetyInsights.org

This study assessed the strength of root cause analysis (RCA) recommendations and their perceived effectiveness and sustainability to longer-term change.

All sentinel event RCAs between 2010-2015 in the Victorian public health system were analysed. An expert panel assessed the strength of the recommendations.

The recommendations ranking categories were:

• ‘strong’ – more likely to be effective and sustainable
• ‘medium’ possibly effective and sustainable
• ‘weak’ – less likely to be effective and sustainable

Although the coding scheme is specific to healthcare, it’s ranking is comparable to the hierarchy of control, e.g. elimination and engineering are ranked highly whereas administrative and document change are weaker.

Results:

227 RCAs and 1137 corrective recommendations were part of the sample. Of these, only 8% were coded as strong, 44% medium and 48% weak.

In 11% of cases there was five or more weak recommendations and in 72% of cases no strong recommendations were made.

The most common recommendations were reviewing or enhancing a policy/guideline/documentation and also training and education. Such weak recommendations are unlikely to lead to effective and sustainable change to reduce the chance of repeating similar events.

The authors found that just a small proportion of recommendations from RCAs have a ‘strong’ nature, e.g. higher up the hierarchy. They state that “insights from the majority of RCAs are not likely to inform practice or process improvements“ (p124).

Given the time and resource investment into RCAs it’s argued that their potential for improving and effecting change remain under-realised.

They also discuss how the majority of RCAs revolved around interviewing the involved staff and reviewing documentation. It’s noted that this focus may weaken the impact of the RCA because interviews are susceptible to recall bias and documents don’t reflect actual work. They argue that more focus should be placed on in-situ observational techniques and low-fidelity simulation may help identify ‘work-arounds’ and organizational structures or processes that influence or constrain behaviours or actions by individuals (p129).

It’s argued that relying on an “arbitrary list of ‘Sentinel Event’” classifications to determine when RCAs should be undertaken is questionable – indicating that the capacity for learning shouldn’t be driven primarily by arbitrary lists but rather on the context of the situation.

They also note that the use of relying on a list of sentinel events for indicating when RCAs to be undertaken is not in their view evidence based. In my view these findings also apply to situations in other industries with critical risks, cardinal rules etc. or relying on actual or potential severity as a trigger for RCA.

Finally, authors believe more human factors expertise is needed in organisations and investigations, more independence during the investigation and more frequent use of observational and simulation techniques for identifying underlying systems factors.

It’s argued that “Time spent in repeatedly investigating similar incidents may be better spent aggregating and thematically analysing the full range of existing sources of information about patient safety (p130).

Authors: Peter D. Hibbert, Matthew J. W. Thomas, Anita Deakin, William B. Runciman, Jeffrey Braithwaite, Stephanie Lomax, Jonathan Prescott, Glenda Gorrie, Amy Szczygielski, Tanja Surwald, Catherine Fraser, 2018, International Journal for Quality in Health Care

Study link: https://doi.org/10.1093/intqhc/mzx181

Link to the LinkedIn article: https://www.linkedin.com/pulse/root-cause-analyses-recommendations-effective-study-ben-hutchinson