Near-miss reporting and safety culture: a practical guide

Near misses expose weaknesses before anyone is hurt

A near miss is an error or hazard that was caught, or never reached anyone, before it caused harm. The Joint Commission calls these close calls or good catches. OSHA and EPA say near miss.

OSHA's process safety rule covers incidents that "could reasonably have resulted" in a catastrophic release. See near miss.

The cause is the same whether or not someone was hurt. CMS gives a surgery center example. Records for two patients booked for the same foot procedure, on opposite feet, get mixed up. The time-out catches it.

CMS says centers should find such errors, because they can cause adverse events.

Surveyors read near misses as proof of how a system behaves. CMS hospital guidance describes a year with three wrong-site surgeries and five near misses, and no action. CMS says that suggests noncompliance with the QAPI condition.

Most harm events go unreported

In a 2012 HHS Office of Inspector General study, hospital incident systems captured an estimated 14 percent of the harm events Medicare patients experienced. Administrators said staff did not see about 61 percent of harm events as reportable.

Harm is common. A 2022 OIG study found 25 percent of Medicare patients had harm during hospital stays in one month: 12 percent adverse events and 13 percent temporary harm. Physician reviewers judged 43 percent of those events preventable.

Near misses are harder to see, because nothing visible happens. AHRQ's PSNet notes that reports do not show how many near misses occurred. Physicians generally do not use voluntary systems. No feedback is a commonly cited reason people stop.

Just culture holds people accountable without blame

A just culture keeps people accountable for choices and treats most errors as symptoms of the system. AHRQ's PSNet describes a culture of safety as one where people report errors and near misses without fear of reprimand. See just culture.

A simple test, following RCA2: if a well-trained person in typical conditions made the error, others could too. Look for the system factor. Discipline would change one person and leave the next exposed.

RCA2 says to define blameworthy events in advance, such as criminal or deliberately unsafe acts, and handle them through HR. That protects the analysis and the reporter.

Six barriers to reporting, and what fixes them

Eight steps to raise near-miss reporting

1. Define a near miss in plain wordsGive five examples per unit, drawn from your own events.
2. Make the first report take two minutesAllow a phone, QR code, link or email, in the person's own words.
3. Accept reports from everyoneContractors, volunteers and visitors see what employees miss. Skip the login.
4. Acknowledge within a dayEven an automatic message shows someone received it.
5. Triage by riskRate severity and likelihood, not who filed. See the root cause analysis and CAPA guide.
6. Fix cheap things fastQuick wins show reports work. Post what changed where staff will see it.
7. Recognize good catchesThank the person and explain what the catch taught the team.
8. Review monthly by unitLook at trends, repeat near misses and open actions. Keep it short.

See how to get staff to report near misses. The near-miss report template is a short form to start with.

Healthcare examples show the work after the catch

Surgery center. The time-out caught the mixed-up records. The useful work comes next. Why were two same-day cases with opposite sides scheduled so they could be confused? Is the catching check reliable on a busy day?

Nursing home (illustrative scenario). A nursing assistant finds a bed-exit alarm unplugged during rounds. The resident is fine. The report asks why, and finds a shared charger across rooms. A second outlet and a charging check fix it.

Nobody would have found that after a fall.

Lessons from industry and aviation

Process safety. OSHA's standard requires covered employers to investigate any incident that did, or could reasonably have, caused a catastrophic release. Start within 48 hours. A near miss can create a legal duty.

In OSHA and EPA's example, earlier non-lethal releases were blamed on operator error, when funding cuts had weakened mechanical integrity.

BP Texas City. The 2005 refinery explosion killed 15 workers and injured 180. The U.S. Chemical Safety Board found safety deficiencies at all levels. BP had put personal injury measures ahead of process safety indicators.

The board urged reporting of incidents and near misses without fear of retaliation. A low injury rate does not prove a safe process. See TRIR and DART rates.

Aviation. NASA's Aviation Safety Reporting System takes voluntary reports through NASA, not the regulator. Reports are de-identified. The FAA will not use them in enforcement except for criminal offenses and accidents.

A reporter who files within 10 days of an inadvertent violation may qualify for a penalty waiver. The lesson: separate reporting from discipline, and publish what you learn.

Feedback keeps reporting alive

A feedback loop runs from report to acknowledgment, triage, action, verification and a message back to the reporter. Without the last step, reporting decays. PSNet lists missing feedback as a common barrier.

The VA triage rules in RCA2 require timely feedback to every known reporter, whatever the risk score.

A "you said, we did" board or monthly message closes the loop for everyone, not only the person who filed.

Measure trust, not report volume

The Joint Commission's hospital standards expect leaders to evaluate safety culture regularly with valid, reliable tools. Its ambulatory survey guide lists culture of safety data among documents surveyors expect. AHRQ's Surveys on Patient Safety Culture cover hospitals, medical offices, nursing homes, pharmacies and surgery centers.

Also watch for signs of trust:

• Near-miss share of reports, as a trend, not a quota.
• Hours from event to report; days from report to feedback.
• Share of reports that led to a verified action.
• Repeat near misses by location.

PSNet cautions that some organizations celebrate more reports and others fewer, and both can be wrong. A count needs a denominator and an outcome.

How IncidentKit supports near-miss reporting

Staff can report through a QR quick report, email-to-incident or a web form, and describe what happened by text. Voice is rolling out. Lauren asks the follow-up questions, and a person reviews and signs.

IncidentKit has no seats, so adding every employee, contractor and visitor costs nothing extra. Non-patient incidents are free on the Open plan. See near-miss reporting and pricing.

Frequently asked questions

Sources

• CMS State Operations Manual, Appendix L: Ambulatory Surgical Centers (QAPI near-miss example)
• CMS State Operations Manual, Appendix A: Hospitals (A-0263)
• HHS OIG, Hospital Incident Reporting Systems Do Not Capture Most Patient Harm (OEI-06-09-00091, 2012)
• HHS OIG, Adverse Events in Hospitals: A Quarter of Medicare Patients Experienced Harm in October 2018 (2022)
• AHRQ PSNet, Patient Safety Event Reporting primer
• AHRQ PSNet, Culture of Safety primer
• Joint Commission, National Performance Goals effective January 2026, Hospital Program (NPG.02.03.01)
• Joint Commission, Ambulatory Care Accreditation Organization Survey Activity Guide (2026)
• IHI, Action Hierarchy Tool (RCA2 and blameworthy events)
• IHI and NPSF, RCA2: Improving Root Cause Analyses and Actions to Prevent Harm
• 29 CFR Part 1904, including 1904.35 employee involvement (eCFR)
• Pennsylvania MCARE Act (Act 13 of 2002), section 308
• OSHA and EPA, The Importance of Root Cause Analysis During Incident Investigation (fact sheet)
• 29 CFR 1910.119, Process safety management of highly hazardous chemicals (eCFR)
• U.S. Chemical Safety Board, BP America Refinery Explosion
• NASA Aviation Safety Reporting System, Immunity Policies (FAA Advisory Circular 00-46F)

Reviewed against the sources above on Oct 5, 2026. Rules change: confirm current requirements with the issuing body or your counsel before relying on any summary.