The Department of Health and Human Services Office of Inspector General reported on 18 June 2026 that the Centers for Medicare and Medicaid Services does not have effective processes to ensure the accuracy of the staffing information nursing homes submit to the Payroll-Based Journal, the federal database that records the hours each direct care worker was paid to work. The audit, numbered A-09-24-02005 and issued 17 June 2026, examined registered nurse hours reported for March 2024. For 45 of 100 sampled items, homes reported a net 748.5 hours that were not supported in accordance with federal requirements. For two further items, 336 hours could not be verified at all because the homes produced no supporting documentation. Projecting from the sample, the auditors estimated that for that one month homes reported roughly 938,000 unsupported hours, about 5 percent of hours in the sampling frame, involving about 53,000 registered nurses, or 42 percent of those in the frame.

The timing matters for families comparing homes. Payroll-Based Journal data is the source CMS uses to calculate the staffing measure and the staffing star rating on Care Compare, a transition CMS set out in its 2018 guidance to states, and staffing is one of three domains behind the overall star rating that families weigh when choosing a home for a relative with Alzheimer's disease, vascular dementia, Lewy body dementia, frontotemporal dementia or a mixed picture. That reporting system is now the principal federal instrument on nurse staffing. CMS published an interim final rule with comment period on 3 December 2025 repealing the minimum staffing standards for long-term care facilities, including the 24 hour registered nurse presence and the hours per resident day floors, effective 2 February 2026, after Congress barred enforcement of those provisions until 30 September 2034.

The audit does not describe deliberate inflation. The 748.5 hour figure is a net one, and the trade publication McKnight's Long-Term Care News, reporting on the same audit, said errors ran in both directions, from one sample item that underreported 60 hours to another that overreported 192.5 hours. OIG wrote that most of the hours it reviewed were supported, and attributed many errors to homes misunderstanding or misinterpreting the requirements; CMS had issued updated Payroll-Based Journal guidance and frequently asked questions in June 2025 while the audit was under way. OIG made four recommendations, including that CMS educate homes on the updated guidance, communicate audit trends, weigh the audit results when selecting homes for follow-up audits, and require its contractor to check whether homes fixed problems found in earlier audits. CMS concurred with two, did not concur with one, and took no position on the fourth.

Existing policy already ties data quality to the public rating: under the five star system, homes that fail to respond to a CMS staffing audit, or whose audit shows significant discrepancies between hours reported and hours verified, receive a one star staffing rating for three months. The unresolved question the audit leaves is how often that check is applied, and whether it is enough for a consumer-facing rating that is now carrying more weight than it did when a staffing floor still stood behind it.