Root Cause Analysis in Healthcare: A Practical Guide with Examples

When something goes wrong in healthcare — a medication error, a patient fall, a wrong-site near-miss — the crucial question is not who but why. Root cause analysis (RCA) is the structured method for answering that question: digging beneath the surface event to find the underlying system failures, so you can fix the real problem rather than the symptom. This guide explains what RCA is, how to run one, and the key tools, with practical examples.

Root cause analysis is a systematic process for identifying the underlying causes of an adverse event or near-miss. Its guiding insight is that most harm in healthcare doesn't come from careless individuals — it comes from flawed systems that set good people up to fail. RCA looks past the immediate mistake ("the nurse gave the wrong dose") to the system factors behind it ("look-alike vials were stored together, the label was ambiguous, and the ward was understaffed that night"). Fix those, and you prevent the next error.

RCA is typically triggered by a serious adverse event or sentinel event, but it's equally valuable for near-misses — the free lessons where harm was narrowly avoided. Accreditation frameworks like JCI, CBAHI and GAHAR expect facilities to investigate serious events with a structured method, making RCA a core skill for quality and safety teams.

A crucial point: the strongest corrective actions change the system (forcing functions, redesigned processes, technology safeguards), while the weakest merely ask people to try harder (reminders, re-education). Good RCA pushes toward the former.

The simplest and most widely used tool: ask "why?" repeatedly — usually about five times — to move from symptom to root cause. For example: a patient received the wrong medication. Why? The label was misread. Why? Two drugs had look-alike packaging. Why? They were stored side by side. Why? There was no separation policy for look-alike drugs. Why? The risk had never been assessed. The root cause isn't "careless nurse" — it's a missing safeguard, which you can now fix.

For more complex events, the fishbone diagram maps contributing factors across categories — commonly people, process, equipment, environment, materials and management. Laid out as a fish skeleton with the problem as the "head", it helps a team see the many factors that combined to cause an event, rather than fixating on one. It pairs naturally with the 5 Whys: fishbone to find the factors, 5 Whys to drill into each.

An RCA is only worth doing if it changes something. The common failure is a thorough analysis followed by a weak action — "remind staff to be careful" — that doesn't touch the system. The best teams close every RCA with specific, strong, assigned and measured actions, and revisit them to confirm they worked. Done well, each analysis makes the next patient safer — which is why RCA sits at the heart of both patient safety and quality improvement. (It's also a core skill in the CPPS and in Lean Six Sigma's Analyze phase.)

Traditional RCA has a well-known weakness: thorough analysis followed by weak, unimplemented actions. The updated approach often called RCA2 ('RCA and Actions') exists to fix this, putting the emphasis on strong, sustainable corrective actions and on prioritising which events to investigate based on risk. Its core lessons are practical: focus your limited RCA capacity on the highest-risk events, insist on system-level actions over 'retrain and remind', assign every action an owner and a date, and measure whether it actually worked.

Not all fixes are equal. The strongest actions force safety and don't depend on human vigilance — physical constraints, forcing functions, automation, and standardisation. Weaker actions rely on people remembering to do the right thing — warnings, policies, training. RCA experts use this hierarchy deliberately: whenever possible, redesign the system so the error becomes impossible or obvious, rather than simply asking staff to be more careful. A good RCA is judged not by the depth of its analysis but by the strength of the actions it produces.