Data on COVID-19 in nursing homes should be handled with caution l Expert Opinion

For the first time, the Centers for Medicare and Medicaid Services (CMS) is releasing data on COVID-19 in individual nursing homes, including case counts, death, and staffing information.

The question is, how useful will this be to patients and their families when trying to choose a nursing home? Our short answer is not very. Based on our experience in creating, disseminating, and studying similar nursing home report cards, we believe the proposed new data on COVID-19 in nursing homes will have limited utility for consumers.

While new COVID-19 cases in the past week can give an indication of current activity in the facility, this information should be interpreted with caution. The main driver of this metric is where a nursing home is located. Our own analyses of early data from state health department websites found that the rate of COVID-19 in the nursing home’s county was one of the strongest predictors of whether a nursing home reported having COVID-19 — stronger than information about the facility itself. Comparisons, if they can be made at all, should be limited to facilities located in the same region.

Other caveats also apply. Consumers should be aware that this information is self-reported by the nursing home. Some nursing homes may monitor patients for signs and symptoms of COVID-19 more aggressively than others. And nursing homes that routinely test their residents are more likely to detect COVID-19. Nursing homes that report fewer COVID-19 cases may be doing less to protect their residents against COVID-19, not more.

The staggering number of deaths in nursing homes is a tragedy, but knowing the number of deaths per facility may not be helpful for consumers. The risk of death from COVID-19 increases with age and underlying conditions. In our analysis of more than 63,000 nursing home residents, residents 85 years and older constituted less than a third of total COVID-19 cases but accounted for 50% of COVID-19 deaths. Residents with dementia made up 52% of COVID-19 cases but 72% of deaths. The takeaway: More deaths do not necessarily equate to worse care. A high death count may simply suggest a nursing home cares for residents who are older and have more underlying health conditions.

» READ MORE: Thousands of deaths later, Pa. finally mandates universal testing in nursing homes, but not for another month

Nursing homes also are required to report shortages of medical supplies and staffing, both of which are essential to protecting residents from infection. Throughout the pandemic, there have been widespread reports of medical supplies shortages at nursing homes, including personal protective equipment. But such shortages are largely driven by market failures and lack of federal response, both of which are beyond the control of individual nursing homes.

Reports of staffing shortages in nursing homes have also been common and are driven in part by COVID-19 infections among the staff (often contracted outside the facility, particularly where community rates are high). Like PPE, staffing shortages can be addressed with significant financial resources, but these shortages are more likely impacted by location. Since most patients choose nursing homes close to home and loved ones, they are unlikely to be able to act on location-dependent attributes such as staffing shortages.

And, as with the data on COVID-19 cases, data on supply and staff shortages are self-reported by nursing homes, which is known to be inaccurate. As a result, since 2018, CMS has relied on payroll-based staffing data to more accurately measure and report staffing on its website, Nursing Home Compare.

Information on COVID-19 case rates would be most reliable and useful with mandated, widespread, and regular testing of both residents and staff. Case rates should be reported separately for those in a nursing home for a short, posthospital, rehabilitation stay (who might enter the nursing home COVID-19 positive) and those in a nursing home for a longer-term residential stay (who are more likely to acquire the infection in the nursing home). Death rates should be risk adjusted to account for differences in age and other health conditions. And all reports should be based on audited and verified data to prevent inaccuracies, and truly reflect a nursing home’s ability to prevent, manage, and treat COVID-19 infections.

» HELP US REPORT: Are you a health care worker, medical provider, government worker, patient, frontline worker or other expert? We want to hear from you.

In the meantime, consumers can look to Nursing Home Compare, state surveys, consumer reviews, and local hospitals, which together help consumers make more informed decisions about nursing homes. But they must know that data tells only part of the story of COVID-19 in nursing homes. Patients and their families should speak to nursing homes to understand the facility’s infection control plans and the ways families can maintain contact with their loved ones inside of the facility. Ask how COVID-19 response activities extend to nursing home staff, for example paying staff sick leave so they can stay home when sick and using a robust system for contact tracing. These are some of the steps that consumers can take now while awaiting and advocating for accurate and reliable data on COVID-19 in nursing homes.

Lissy Hu is CEO of Allscripts subsidiary CarePort Health; Rachel Werner is executive director of the Leonard Davis Institute of Health Economics and professor of medicine at the University of Pennsylvania.