How we counted senior-care complaints and inspection findings

The two reports use different public datasets. We keep their care settings, time periods and units separate. Neither report estimates a resident’s likelihood of harm or ranks states by care quality.

Residential-care complaints: NORS, federal fiscal year 2025

The National Ombudsman Reporting System (NORS) summarizes ombudsman program casework. Our report uses the FFY 2025 reports linked by the National Ombudsman Resource Center: the Annual Roll-up, Table A selected state and region information, and Complaint Analysis by Complaint Code.

The reporting period is October 1, 2024 through September 30, 2025. The published tables cover 50 states, the District of Columbia and Puerto Rico. We retain all 52 reported jurisdictions. This describes the reporting tables used, not every place where an ombudsman program operates.

NORS reports cases and complaints closed during the fiscal year. A case can include multiple complaints, and complaint verification is recorded separately. Counts are not unique residents, proven violations or injuries. The case and setting definitions explain those distinctions.

Residential care communities include assisted living and other qualifying residential settings. Nursing facilities and other settings are separate columns. We do not label the all-setting total as assisted-living complaints.

For each jurisdiction, other-setting complaints equal the all-setting total minus nursing-facility and residential-care-community complaints. We reconcile those columns to national totals: 55,660 residential-care complaints, 143,252 nursing-facility complaints and 2,903 other-setting complaints, totaling 201,815.

The share for each residential-care complaint category equals that category’s count divided by 55,660, multiplied by 100. Nursing-facility category shares instead divide by 143,252. Calculations use the unrounded counts and round the final percentage to one decimal place. A category’s share within a setting differs from that setting’s share of the category across settings.

The chart shows all 12 residential-care categories. The table and downloadable CSV retain the other settings as context. The alphabetical jurisdiction table is not a ranking. Differences can reflect multiple factors, including reporting and program activity; these totals do not identify which factors explain a state’s result. We make no year-to-year trend claim.

Nursing-home inspection findings: CMS, August 26, 2026 release

The CMS health-deficiencies dataset contains citation rows. We use the retained August 26, 2026 release and select rows where Complaint Deficiency is Y. We count rows without merging citations from the same inspection into one complaint or resident.

The selection contains 131,315 rows. Standard Deficiency is also Y on 29,324 of those rows; a complaint flag does not exclude an association with a standard inspection.

We group scope/severity codes G through L. G–I mean actual harm below immediate jeopardy. J–L mean immediate jeopardy, which can include serious harm that is likely to occur as well as harm that has occurred. The combined label is therefore “actual harm or immediate jeopardy.” It is not a count of confirmed injuries.

The national percentage is 16,610 divided by 131,315, multiplied by 100: 12.6% after rounding. A jurisdiction’s percentage uses its own complaint-associated citations as the denominator. A jurisdiction with zero selected rows retains its zero count and has no calculated percentage. The table retains the 50 states, D.C., Guam and Puerto Rico present in the source.

The source file contains survey dates from March 23, 2017 through July 23, 2026. Its Processing Date field is August 1, 2026; that differs from its catalog release date. This is a multi-year source snapshot, not an annual complaint rate. The table describes the severity mix of selected citations, not a complaint substantiation rate or a state care-quality score.

Reuse and corrections

Each report links to its source and downloadable jurisdiction table. Keep the dataset name, counting unit and period with a quoted result. Do not add the NORS and CMS counts together or compare their percentages as equivalent rates.

A later source release may differ. The downloads accompanying these reports describe the retained snapshots; a source refresh requires recalculating the affected report. For questions about our calculations or a possible correction, contact SeniorCareHomes.com.