It is Monday morning and the health services administrator of a small rural jail is investigating an unexpected death over the weekend. The patient had just been seen on Friday afternoon by the medical director for heart palpitations and lethargy. He said he was thirsty but didn’t like the taste of the water in his cell. After a physical assessment and EKG, labs were drawn before he was sent back to his cell with instructions to drink more. By Sunday the patient was found unresponsive in his cell. Resuscitation efforts were unsuccessful.
Analysis of adverse events, whether near misses or actual clinical errors that cause patient harm, is an important part of a patient safety system. A primary mode of adverse event evaluation is root cause analysis, which reconstructs the events and trajectory of an incident to determine causative factors and safety failure modes. The strength of root cause analysis is the ability to thoroughly evaluate all possible contributing causes for the event, rather than simply relying on the first or most obvious cause discovered in an investigation.
By fully evaluating an adverse event, both active and latent causes of clinical error can be identified. Active causes are readily apparent at the point of care and often include actions of staff members. Latent causes are more insidious and involve less obvious system design issues or process failures.
The root cause analysis process involves asking three primary questions about the event:
- What happened?
- Why did it happen?
- What can be done to prevent it from happening again?
Systematically answering these questions can reveal latent and active components of an incident so that evaluators are able to develop a more accurate picture of the context of a clinical error.
What Happened
The starting point of any good investigation is an understanding of what happened. Often a written incident report or error report is submitted. This can form the basis of an inquiry but is only a starting point. A full understanding of an adverse event requires interviews with involved parties, review of medical documentation and pertinent policies and procedures. It is important to keep an unbiased mental framework during this phase. Get all the facts of the situation out in the open before beginning to determine causality.
Why Did It Happen
The “Why” question should be asked multiple times to dig deeper into causality. For example, failure to notify the provider about a critical lab value which resulted in a patient injury may, at first, seem to be a communication error on the part of the laboratory service.
Asking the “Why” question successively, it is found that the laboratory automatically faxes critical values to the site. The fax machine is located in the health service administrator’s office, which is locked and unattended on Saturday. Per protocol, the laboratory service also calls the phone of the ordering physician. The physician in this example does not carry his work cell phone when not on-call, and the on-call physician for that weekend uses a different cell phone number posted for weekend staff but not available to the laboratory service. The patient’s serum glucose was over 800 and the BUN was high, as well [A later mortality review would suggest ketosis-prone undiagnosed diabetes (KPD) as the cause].
Developing an effective root cause analysis requires a determination of all possible factors contributing to adverse events in the clinical setting. Henriksen et al. (2008) provide a thorough framework for determining contributing factors, from latent conditions to active errors (See Figure). This five-tier framework is appropriate for correctional health care programs to use in evaluating adverse events. When using the framework to evaluate causation, begin with the first tier (closest to the patient) and move outward, considering the contribution of each tier to the resulting event. Latent conditions that may be hidden from view in the initial evaluation become visible. System change is necessary when a latent issue emerges regularly in the root cause analysis of individual adverse events. For example, if staff fatigue is implicated in the first-tier evaluation of several events, additional consideration can be given to patient load and staffing conditions in the fourth-tier evaluation.
“How” and “Why” questions guide a root cause analysis through the various tiers of contributing factors. In the initial round of evaluation, this is an educated guess on the part of the investigation team. Validation of the resulting hypotheses then determines the next course of action. The team may find that they must search further back in the sequence of events for latent causes not apparent in the initial resulting error evaluation. A rich field of inquiry is produced from a thorough root cause analysis of a single significant adverse event or multiple similar events that might suggest a system failure.
Mental tendencies and personal preferences can invade the adverse event analysis and require vigilance to prevent or eliminate. Read more about common cognitive biases here.
How to Prevent it from Happening Again
The investigators must determine all possible causes of a clinical error in order to create a plan of action. Active causes are easier to correct than latent causes, and, therefore, it can be tempting to primarily focus on correction of active causes. Latent cause correction, however, will result in more significant organization-wide harm reduction because of these causes, at the blunt end of care delivery, affect multiple aspects of care.
Adverse event analysis by determining root causes is a primary function of an effective patient safety program. A structure that supports patient safety principles in reporting and analyzing adverse events will also support the reduction of patient harm and liability in the correctional setting.
Have you used root cause analysis to investigate a patient care issue? Share your experiences in the comments section of this post.
Originally published in the Spring 2016 issue of CorrectCare™, the quarterly magazine of the National Commission on Correctional Health Care.
Learn more about all aspects of Correctional Health Care Patient Safety in the book “Correctional Health Care Patient Safety Handbook“.
Helena Parker says
These are excellent Tips, and applicable to all form of investigations. You are right that in any investigation, “How” and “Why” are the basis for reaching conclusions. I am running an LPN/LVN training institute, and I am going to share your post with my trainees because tomorrow, they may also be a part of the investigation, and they will remember the value of your Tips. Thank you for sharing.