With national attention focused on the unacceptable prevalence of medical errors, the healthcare sector has turned to highly reliable industries, including aviation and nuclear power, to learn about their approach to error investigation. These industries have been using root-cause analysis (RCA) for decades to uncover latent system issues that contribute to performance variations.

During the mid-1990s, the U. S. Department of Veteran Affairs and The Joint Commission adopted standards that require healthcare organizations to complete and submit RCA reports on serious events. Since then, thousands of RCA reports have been submitted. A review of these reports suggests RCAs have helped shift blame from individuals to systemic causes.

But have RCAs been effective in reducing errors? No robust evidence suggests they’ve improved patient safety. What’s more, with each RCA taking 20 to 90 person-hours to complete, cost-effectiveness is questionable.

So why has a process that has proven to be effective in other industries failed to yield similar success in health care? Perhaps the answer lies in its execution.

Understanding the RCA process

At its core, the RCA process poses three questions: What happened? Why did it happen? What can be done to prevent it from happening again? Several qualitative studies of healthcare RCAs have identified deficits in each stage of this process. Common themes of study findings include:

• lack of content expertise among investigative team members
• challenges in getting all relevant parties involved in the investigation
• focusing on the “when” and the timeline rather than the “why”
• participants’ painful emotions and the facilitator’s inadequate handling of them
• more focus on completing the report than identifying the true root cause
• inadequate knowledge of human-factors engineering concepts
• failure to aggregate events to identify trends
• formulating action plans based on what’s possible immediately rather than what’s needed to prevent recurrences (for instance, focusing on education and policy changes rather than systems or engineering changes)
• lack of a formal process for grading the strength of each action plan
• insufficient communication of RCA findings and action plans within the organization
• no reliable way to determine if the recurrence risk has been reduced.

So how can we make the RCA process more effective? First, recognize that it’s effective only when done in the spirit of inquiry, where a fair and just culture has encouraged staff and clinicians to feel accountable for their own errors and error-prone systems and to feel comfortable reporting and discussing these errors openly.

Preparing for the RCA

Getting the right people to the discussion table is vital. To anticipate the key individuals needed for the RCA, the RCA facilitator must delve deeply into the facts before scheduling the meeting, examining the medical record and interviewing involved staff and clinicians. Staff and clinicians at the “sharp end” of patient care must attend the meeting to ensure that everyone involved understands the details of the event, as well as to brainstorm innovative solutions. Managers and executives can provide a broad view of the process under examination, along with institutional support to drive process improvement.

The electronic health record (EHR) liaison should be invited to every RCA meeting. Many hospital procedures are conducted within the EHR and can be improved simply by modifying the system. Similarly, to build credibility and accountability into the process, clinical leaders and expert staff should be involved in preparing and presenting the event timeline, evidence-based literature, and hospital policies to review. Equipment that was involved in the event should be brought to the table (if possible) for examination by the RCA team.

Facilitating the RCA

The RCA process can be cumbersome if participants rely on paper flipcharts. To streamline the process, the RCA document should be projected electronically, and the search for root causes should be documented live during the meeting.

The facilitator should focus the investigation by asking the team to compose a short sentence describing the undesirable outcome under investigation. From this single outcome, the team generates a list of possible causes by category. Categories may include:

• device-related factors (known limitations of the device or unexpected device failures)
• environmental factors
• patient-related factors
• communication factors
• situational-awareness factors (such as failure to perceive, comprehend, or foresee a risk)
• human errors (slips, lapses, mistakes, and misjudgments)
• procedural violations (failure to recognize the risks of violating a rule or disregard for the consequences)
• lack of proficiency or skill
• training factors
• staffing insufficiencies
• leadership concerns.

The team discovers root causes by listing preceding causes (events leading up to each cause) until the question “Why?” can no longer be answered. For instance, suppose a nurse who gave a medication to the wrong patient followed the two patient-identifier rule but made an error when comparing the medical record number on the patient’s armband to the medical record number in the medication administration record. In this scenario, the RCA team would compare the seven-digit medical record numbers of two different patients; they discover the numbers differ by just one digit. The team asks why the numbers are so similar and learns that the two patients have identical first and last names and were registered as new patients in the emergency department within 15 minutes of each other. Expecting staff to identify a one-digit discrepancy when the rest of the comparison information is the same is unreasonable. One could easily understand how this human error occurred. (See Sample root cause analysis: Fentanyl overdose by clicking the PDF icon above.)

As the facilitator, you’d lead the RCA team to focus only on causes related to the undesirable outcome, avoiding individual performance issues (which are more appropriate for peer-review settings). Explain the importance of getting to the root causes of human errors so as to help design systems that prevent such errors in the future. Then encourage the team to search for the best risk-reduction strategies. The goal of the RCA is to develop action plans that achieve the most effective harm-prevention strategies. Assign each action plan to a responsible individual, and continue to send reminders to responsible parties until each action plan is completed. (See Effective risk-reduction strategies by clicking the PDF icon above.)

Anchoring improvements

Once the RCA documentation is completed, seek executive oversight and approval of action plans. A patient-safety committee, which includes directors and executives from both clinical and operations departments, should review RCAs before action-plan implementation. The patient-safety committee should provide insight into whether an action plan can be supported and will be effective—or it might suggest revising the plan to avoid jeopardizing another process or department.

To ensure sustainability, the success of each major process change should be measured and the data reported to a governing body. Electronic methods for capturing and reporting RCA data to a designated patient safety organization add to the patient safety database. The facility’s patient-safety committee should trend and analyze RCAs and share stories at the “sharp end” with staff and clinicians, so staff can articulate how the RCA process has improved patient safety. Sharing these stories creates an organizational memory of past harmful events, along with the desire to learn from them to avoid repeating mistakes. The RCA process serves as a bridge to a final solution, which may evolve over time as the problem is studied further and the solution is refined.

Some events occur across hospitals. For example, more than 1,000 instances of epidural-to-I.V. misconnections have been reported to the U.S. Pharmacopeia. Altering tubing design to prevent such misconnections (called a forcing function) is the most effective strategy to prevent recurrences. To implement this change, hospital systems must join with manufacturers to collaborate on equipment redesign. While the scope of this initiative may seem unachievable today, such a bold approach is needed to implement meaningful changes that have a measurable effect in preventing patient harm.

Selected references

Morse RB, Pollack MM. Root cause analyses performed in a children’s hospital: events, action plan strength, and implementation rates. J Healthc Qual. 2012;34(1):55-61.

Nicolini D, Waring J, Mengis J. Policy and practice in the use of root cause analysis to investigate clinical adverse events: mind the gap. Soc Sci Med. 2011;73(2):217-25.

Percarpio KB, Watts BV. A cross-sectional study on the relationship between utilization of root cause analysis and patient safety at 139 Department of Veterans Affairs medical centers. Jt Comm J Qual Patient Saf. 2013;39(1):32-7.

Percarpio KB, Watts BV, Weeks WB. The effectiveness of root cause analysis: what does the literature tell us? Jt Comm J Qual Patient Saf. 2008;34(7):391-8.

Taitz J, Genn K, Brooks V, et al.; NSW RCA Review Committee. System-wide learning from root cause analysis: a report from the New South Wales Root Cause Analysis Review Committee. Qual Saf Health Care. 2010;19(6):e63.

Wu AW, Lipshutz AK, Pronovost PJ. Effectiveness and efficiency of root cause analysis in medicine. JAMA. 2008;299(6):685-7.

Susan Keating is the director of Clinical Risk Management at Shands Hospital at the University of Florida in Gainesville. Susan Tocco is the director of Patient Safety and Transformation at Orlando Health in Orlando, Florida.