# F867: QAPI/QAA improvement activities

> F867 is where QAPI becomes visible. The home must track medical errors and adverse events, find their causes and set priorities. It must fix problems at the system level and measure whether fixes hold. It must also run at least one improvement project a year.

Source: https://incidentkit.ai/compliance/f-tags/f867 · Updated Oct 5, 2026

## Key facts

- **Tag title:** QAPI/QAA Improvement Activities
- **Regulation:** 42 CFR 483.75(c), (d), (e) and (g)(2)(ii)-(iii). The old F866 was folded into F867
- **Guidance relied on:** Appendix PP Rev. 232 (issued 07-23-25, in use since 04-28-25); F867 section Rev. 229 (04-25-25, implementation 04-28-25)
- **Improvement project:** At least one distinct project a year on a high-risk or problem-prone area found through data
- **Data frequency:** Set by the home. High-risk data is collected more often (daily, weekly or monthly) until performance is satisfactory
- **New in the 2025 guidance:** Health equity. Feedback, sub-population data and factors such as race, language and socioeconomic status in error analysis
- **Severity note:** CMS gives examples at levels 1 through 4

- **Citation:** F867 · 42 CFR 483.75(c), (d), (e) and (g)(2)(ii)-(iii)
- **Authority:** CMS
- **Applies to:** Medicare-certified skilled nursing facilities, Medicaid-certified nursing facilities

## What F867 covers

F867 enforces four parts of 42 CFR 483.75. Paragraph (c) covers policies for feedback, data and adverse event monitoring. Paragraphs (d) and (e) cover analysis, systemic action, priorities and improvement projects. Paragraph (g)(2)(ii)-(iii) covers the QAA committee's duty to act on quality deficiencies and review data, including drug regimen review data.

The home must track medical errors and adverse resident events, find causes and prevent repeats. CMS defines an adverse event as an unwanted and usually unexpected event that causes death or serious injury, or the risk of it. A near miss is a serious error that did not become an adverse event, by chance or interception.

Priorities should follow high-risk, high-volume and problem-prone areas. CMS examples: tracheostomy care, pressure injury prevention and high-risk drugs (anticoagulants, insulin, opioids) are high-risk. Transcribing orders and giving medications are high-volume. Call bell response, staff turnover and lost laundry are problem-prone.

## Events CMS expects you to track

CMS and AHRQ list potentially preventable events. They cite a 2014 HHS Office of Inspector General finding. About one in three Medicare beneficiaries was harmed by an adverse or temporary harm event within 35 days of a nursing home stay. Nearly 60 percent of those events were potentially preventable.

*CMS examples of potentially preventable events*

| Category | Examples |
| --- | --- |
| Medication | Delirium with opiates or psychotropics. Hypoglycemia with antidiabetic drugs. Bleeding with antithrombotics. Drug toxicity (digoxin, phenytoin, lithium). Constipation or impaction with opiates |
| Care | Falls, skin tears and other care-related trauma. Avoidable pressure injuries. Dehydration. Feeding tube complications. Elopement. Abuse, neglect, misappropriation and exploitation |
| Infection | Pneumonia and influenza. Urinary tract infections, including catheter-associated. C. difficile and norovirus. Skin and wound infections |

CMS says an adverse event, such as a cognitively impaired resident eloping, is a high-risk problem needing corrective action.

## What surveyors check

Surveyors use the QAPI and QAA Review pathway at the end of the survey. Repeat deficiencies the committee never identified or prioritized suggest it is not doing its job.

*What to have ready*

| Surveyors ask for | Have ready |
| --- | --- |
| Policies for feedback and data | How staff, resident and representative feedback is gathered and used, plus each department's data |
| Adverse event and error tracking | A log with causes analyzed and preventive action for each, not only counts |
| Performance indicators | Thresholds, goals and how often each is reviewed |
| A systematic method | Root cause analysis, reverse tracker, failure modes review or similar. The cause it found |
| Corrective action plans | Problem definition with contributing causes, measurable goals, step-by-step interventions, tracking plan |
| Results and the annual project | Data showing the fix held, and the year's project with its findings |

## What raises F867 severity

Severity follows what the missing QAPI work allowed. Each CMS example is a problem the committee should have caught.

*CMS examples of F867 severity*

| Level | Example from CMS guidance |
| --- | --- |
| 4: immediate jeopardy | Residents had third-degree burns the month before. Hot water data was collected but never reviewed. The committee did not track how residents' code status reached staff |
| 3: actual harm | Repeat deficiencies on two surveys about discharge needs. A resident left without diabetes education and was rehospitalized. The committee was unaware and did not monitor discharge |
| 2: potential for more than minimal harm | A quality deficiency about inaccurate weight measurement from the prior survey was not corrected or tracked |
| 1: minimal potential for harm | A plan to check monthly for three months. No proof of checks in the second month |

## Clocks and frequency

F867 has no deadline for reporting an event. Collect data on the home's schedule, and more often for high-risk issues. The committee reviews at least quarterly under [F868](https://incidentkit.ai/compliance/f-tags/f868). CMS expects issues likely to cause serious harm, impairment or death to be answered immediately. Abuse and neglect clocks sit in [F609](https://incidentkit.ai/compliance/f-tags/f609).

## Gaps that lead to citations

- Events are logged and counted, but no one analyzes cause.
- The same corrective action, usually an in-service, repeats for every event.
- Goals are not measurable, so success cannot be shown.
- A tracking plan has missing months.
- The improvement project came from a list, not home data.
- The committee does not know about a repeat survey deficiency.
- Data is not split by sub-population, and error analysis skips equity factors.

## How to show a good investigation

CMS says a corrective action fixes the cause at the systems level, not just the symptom. Tests of change or Plan-Do-Study-Act cycles are allowed until goals are met.

1. **Capture** Log the event or near miss when it happens.
2. **Analyze** Find the cause with a systematic method.
3. **Act on the system** Set a measurable goal, an owner and a date.
4. **Measure** Check as planned and keep proof for every period.
5. **Report** Bring results to the committee. Close only when the fix held.

## How IncidentKit supports F867

[Incident reporting](https://incidentkit.ai/product/incident-reporting) with [Lauren](https://incidentkit.ai/product/lauren) captures events and near misses, and a person reviews and signs. [Investigations](https://incidentkit.ai/product/investigations) record contributing factors and five whys. [Corrective actions](https://incidentkit.ai/product/corrective-actions) carry an owner, due date, evidence and an effectiveness check. Nothing closes until verified. [Analytics](https://incidentkit.ai/product/analytics) show where events cluster, and [compliance packets](https://incidentkit.ai/product/compliance-packets) include a QAPI summary. Choosing priorities and running the committee remain your team's work.

## What the rule asks for, and how IncidentKit supports it

| Requirement | IncidentKit |
| --- | --- |
| Track medical errors and adverse events and analyze their causes | Incident reporting with Lauren. Investigations with contributing factors and five whys. Analytics by location, shift, equipment and cause. |
| Corrective actions that change the system | Corrective actions carry an owner, due date and evidence, tied back to the investigation. |
| Measure success and show improvements were sustained | Each corrective action has an effectiveness check. Nothing closes until verified. The audit trail keeps the history. |
| Set priorities from data | Analytics show where incidents cluster, so the committee can choose high-risk and problem-prone areas. |
| QAA committee reviews data and acts on it | Compliance packets include a QAPI summary for each meeting. |
| Equity analysis of errors and events | Analytics cluster by location, shift, equipment and cause. Breakdowns by language or race come from your own data and committee review. |

## Frequently asked questions

### What events must a nursing home track under F867?

Medical errors and adverse resident events, plus near misses. The program must track them, find causes and prevent repeats.

### How many performance improvement projects are required?

At least one distinct project a year on a high-risk or problem-prone area found through the home's own data. The number should reflect the home's services and facility assessment.

### What should a corrective action plan contain?

A problem definition with contributing causes, measurable goals, step-by-step interventions and a tracking plan. It should fix the underlying cause at the systems level.

### What changed in the 2025 F867 guidance?

CMS added health equity guidance. Homes should consider equity feedback and monitor outcomes for sub-populations. Error analysis should include factors such as race, sexual orientation, socioeconomic status and preferred language. Surveyors have used it since April 28, 2025.

### Can monitoring gaps alone be cited at F867?

Yes. CMS's level 1 example is a plan for monthly checks for three months, with no proof the second month's checks happened. Higher levels cover unreviewed data or an unmonitored high-risk system.

## Sources

- [eCFR, 42 CFR 483.75 (quality assurance and performance improvement), current through 2026-10-01](https://www.ecfr.gov/current/title-42/section-483.75)
- [CMS State Operations Manual, Appendix PP, Guidance to Surveyors for Long Term Care Facilities (Rev. 232, issued 07-23-25; revised guidance used on surveys since 04-28-25): F867 section (Rev. 229, issued 04-25-25, effective 04-25-25, implementation 04-28-25)](https://www.cms.gov/regulations-and-guidance/guidance/manuals/downloads/som107ap_pp_guidelines_ltcf.pdf)
- [CMS memo QSO-25-14-NH (revised 2025-03-10): Revised Long-Term Care Surveyor Guidance, effective April 28, 2025](https://www.cms.gov/files/document/qso-25-14-nh-revised-2025-03-10.pdf)
- [HHS Office of Inspector General, Adverse Events in Skilled Nursing Facilities: National Incidence Among Medicare Beneficiaries (OEI-06-11-00370, February 27, 2014)](https://oig.hhs.gov/reports/all/2014/adverse-events-in-skilled-nursing-facilities-national-incidence-among-medicare-beneficiaries/)
- [LeadingAge, List of Revised F-tags in New RoPs Guidance 2022 (tag titles as listed for CMS's June 2022 revisions)](https://leadingage.org/sites/default/files/List%20of%20Revised%20F-Tags.pdf)
- [CMS State Operations Manual, Chapter 7, Survey and Enforcement Process for Skilled Nursing Facilities and Nursing Facilities (Rev. 244, issued 06-26-26): scope and severity matrix, immediate jeopardy, substandard quality of care](https://www.cms.gov/regulations-and-guidance/guidance/manuals/downloads/som107c07pdf.pdf)

## Related

- [F865 QAPI program and plan: what surveyors ask for](https://incidentkit.ai/compliance/f-tags/f865)
- [F868 QAA committee: members, meetings and evidence](https://incidentkit.ai/compliance/f-tags/f868)
- [Nursing home QAPI requirements: 42 CFR 483.75 explained](https://incidentkit.ai/compliance/cms-qapi/skilled-nursing-facilities)
- [QAPI Committee Meetings: Agenda, Data and Minutes](https://incidentkit.ai/use-cases/qapi-committee-meetings)
- [Corrective Actions: How to Close Them With Proof](https://incidentkit.ai/use-cases/close-corrective-actions)
- [Root cause analysis and CAPA: methods and strong actions](https://incidentkit.ai/guides/root-cause-analysis-and-capa-guide)
- [Performance improvement project: definition and meaning](https://incidentkit.ai/glossary/performance-improvement-project)