Study population and setting
COVID-19 surveillance was conducted by the West Virginia Department of Health and Human Resources (DHHR) from March 14 to June 11, 2020 across 123 nursing homes. An outbreak was defined as 2 or more confirmed cases within 14 days, where at least one case was a resident of the nursing home. The study also used quality ratings (ranging from 1 to 5 stars) from the Centers for Medicare and Medicaid Services (CMS) health inspection results; CMS conducts random health inspections which are summarized into an overall star rating, as well as noting nursing hours per resident as a measure of staffing, and hospital readmissions as a measure of quality of care. Ratings from December 13, 2018 to February 26, 2020 were used. COVID-19 incidence was also measured at the county level for the cases per 100,000 residents.
Summary of Main Findings
The West Virginia DHHR identified 14 nursing homes (11% of 123) with an outbreak. The median number of cases per nursing home ranged from 1 to 71 cases in a given home, with a total 226 cases. There were 140 total cases among staff members, which ranged from 0 to 39 per nursing home (median: 4 cases/home). The average number of beds in homes with an outbreak was 104.1, compared to 84.6 among non-outbreak homes (p 0.05), and the counties in which they occurred had higher incidence per 100,000 people (177.8 among outbreak homes vs. 105.1 among non-outbreak). The mean number of nurse aid hours per resident in outbreak facilities was lower than that of non-outbreak facilities (1.9 vs. 2.2 hours/resident/day). 50% of the outbreak facilities (N=7) had 1-star ratings, and after adjusting for county-level COVID-19 incidence, the odds of an outbreak among a 1-star facility, compared to 2 to 3-stars, was 7 times higher. And compared to 4 to 5-star facilities, the odds were 17 times higher of an outbreak among 1-star facilities.
The study used surveillance that covered all of West Virginia nursing homes, which means it is generalizable to the state and likely similar states in the US. West Virginia DHHR mandated that all residents and staff are tested in nursing homes as of April 17 2020, which increased the likelihood they were able to identify cases in a timely manner, regardless of symptomatic or asymptomatic cases.
Data from before COVID-19 was used to assess quality and staffing levels were based on annual averages and the most recent CMS inspections, which happen to be conducted 2 weeks before the study period. This means that the data may not reflect the actual staffing and quality levels during the pandemic, which may be higher or lower (e.g., underestimate or overestimate) the association depending on facilities’ responses. They also only controlled for county-level incidence, and there may be other confounding factors not accounted for that may impact the association between star ratings and odds of an outbreak, such as sociodemographic factors or the number of Medicaid-funded residents.
This is a state-wide assessment of outbreaks among nursing home residents and staff using universal surveillance representative of the whole state and similar states, and helps to identify nursing homes that may be at increased risk of an outbreak.
This review was posted on: 3 October 2020