Many Why’s Of A Failing Root Cause Analysis In Healthcare

Many Why’s Of A Failing Root Cause Analysis In Healthcare
By Admin | Posted on September 20, 2020 | Comments: 5

Healthcare professionals regularly come across a situation of ‘incident investigation’ in their workplace. When the incidents are categorized as sentinel or serious events, it requires in-depth analysis using approved methodologies like RCA (Root Cause Analysis). Cause analysis tools such as '5 Why's are used for conducting a root cause analysis. From my professional engagement with several organizations to review RCA results, more often I encounter an interesting fact that, most of those exercises didn’t derive root causes, but concluded with obvious causes. Here I am sharing few of most frequent reasons among many, to my colleagues out there in Healthcare quality & improvement arena.

Approach biases: If you start an investigation with an objective to uncover something that has gone wrong, it will ruin your investigation. Outcome bias is the (common) tendency to judge the action, based on outcome of an incident. It will lead to errors in interpretation of evidence, you will not conclude a fair investigation and, most importantly, what you discover may not be relevant. So do not ask what went wrong – ask, ‘What happened?’.

Hindsight bias is the possibility to judge the actions in an incident based upon information not available at the time of the incident. It is very easy to notice ‘clues’ when investigating serious incidents and to over-interpret their value. It is important that investigators do not confuse a clue that has ‘foresight’ value with one that only has value in hindsight.

Lack of Training & Quality of Investigation: Skill set of investigator/team is often overlooked, it is important that investigator is knowledgeable to follow the chosen RCA framework. Incidents have to be reconstructed from different sources of varying degrees of reliability, usefulness and accessibility; from patient records, staff interviews and records of duty rota etc. Quality investigation requires a team that is proficient in systems thinking, human factors, cognitive interviewing, and data analysis.

Missing the underlying patient factors: Most of the RCA tools currently used are historically derived from industries like aviation. These tools were designed to find the root cause of a failure in a process that should operate perfectly every time; a series of defined processes should be expected to produce a perfect outcome. In medicine, a patient presents to us with a pathology that may be very serious and even with our best efforts the outcome for the patient may not be good. I came across incidents that had rare patient factors, not formerly identified until an incident of patient mortality and that was vital to the investigation. Moreover, patient social and psychological factors also could contribute to a situation how it evolved around the time of incident.

Failure to learn or communicate learning: Process of conducting a RCA itself is a series of learning. Situation analysis, patient factors and other ‘system factors’ contributing to the ‘incident’ are all providing new set of learning to those involved in it. Most of the time I have observed that the identified “root cause’ is communicated to a close group of people involved in it. However, learning opportunity from other number of factors mentioned above are not conveyed to hospital staff through a planned training activity.

Lack of objective evidence: often abovementioned biases will lead to conclusions or assumptions without demonstrating objectivity in your analysis and interpretation of what happened. Use of standardized evidence sources or templates moves the spotlight away from the opinions or intuitions of individual investigators or ‘experts’ in the field and enables an investigation to be more objective.

Lack of system Focus: When analyzing why something happened it is important to identify how a given system interacts with the individual within the context of that particular event. Did the system help? Did it hinder? Or was it neutral?

A solid understanding of why humans make mistakes is also necessary when hypothesizing about human involvement in an incident. Human error is manageable when the human error drivers are understood. Identifying deficient management systems or latent root causes will uncover human error drivers.

Lack of timely & responsive Investigation: Health authorities and accreditation bodies set a timeframe for reporting and submitting formal investigation report. It is important to understand that these timelines are only guidelines as maximum time to report/submit. An investigation started at the early hours of incident will benefit from fresh memory of staff involved, availability of evidences and immediate responses from patient, family, etc. Responsive investigation to consider all such responses and not to leave any questions unanswered, using a standardized tool may condition the investigator to set of pre-defined questions, overlooking relevant questions that should arise in each situation.

Hierarchical Hijack: Last but not least, often investigators may ends their analysis once they have reached a cause of mutual convenience, as most of the system error identified deemed to be beyond the capability or willingness of the organization to resolve or cause deeper threat to organizational and sociopolitical dynamics or to preserve interpersonal relationships.

Lack of data and Disaggregated analysis: The current RCA approach limited with analysis of individual incidents in isolation and within bounded historical data, often ignoring the factual data of contributing factors over isolated incidents, leads to ‘Failure to test theories’ with supporting technical data.

Lack of data and historical information also frustrates the organization’s ability to assess its vulnerability to recurring events. Organizations’ averting specific very rare events rather than addressing the conditions that allowed the event to occur. Mechanisms for aggregating learning from incidents and creating alerts do exist in technology supported data base like “AccreHealth Technology’. An innovating software solution ( offer digital workflows for Incident analysis and built-in RCA tools ,solutions that can transform your Incident Management system with dashboards that provides 360° data and trends.

Author is an accreditation consultant with an experience of leading 75+ hospitals across Middle East and Asia on their journey to international accreditation. Mr. Thomas is a change agent, who advocates innovation and technology to facelift complex workflows in a Healthcare environment. Write to (