More than 68,000 COVID-19 cases and 16,000 related deaths in US nursing homes may have gone uncounted because they occurred before federal guidelines required facilities to report case and death data in late May 2020, suggests a study yesterday in JAMA Network Open.
Led by a Harvard University researcher, the study compared the number of COVID-19 cases and deaths reported by 15,415 nursing homes in 20 states to the Centers for Disease Control and Prevention (CDC) National Healthcare Safety Network (NHSN) with those reported to state departments of health by May 24, 2020. Analysis took place from December 2020 to May 2021.
Until May 24, more than 3 months after the first US COVID-19 nursing home outbreak at the Life Care Center in Kirkland, Washington, the federal government didn’t mandate nursing home reporting of COVID-19 cases and deaths. Facilities could have voluntarily reported retrospective data, but many didn’t.
For instance, the researchers pointed out that Life Care Center reported no coronavirus cases in the May 24 NHSN submission, even though a March 2020 CDC investigation had identified 81 cases and 23 deaths of residents in that facility.
“The delay in required reporting means that the NHSN data miss a significant period of the pandemic, in which cases and deaths were increasing more rapidly than any other point in 2020 except during the wave in the final months of the year,” they said.
Federal data were misleading
Of all nursing homes in the sample, 4,599 facilities were in 12 states that mandated case reporting, while 7,405 were in 19 states that mandated death reporting. On average, 43.7% of COVID-19 cases and 40.0% of related deaths before May 24 in sample states weren’t reported in the NHSN submission, suggesting the omission of 68,613 cases and 16,623 deaths, representing 11.6% of cases and 14.0% of deaths in nursing home residents in 2020.
Nursing homes in both analytic samples had significantly more COVID-19 cases and deaths using the NHSN data than facilities in nonsample states (average cases per facility, 8.1 vs 2.4; average deaths, 2.2 vs 1.4). There were no differences in nonreporting before May 24 by nursing home characteristics such as region, ownership, chain affiliation, or star rating.
From 40% to 60% of COVID-19 cases and 30% to 50% of deaths before May 24 went unreported. In the last 2020 NHSN submission on Dec 27, the percentages of unreported COVID-19 cases and deaths had fallen (average, 13.9% of cases, 18.7% of deaths), which the researchers said reflected the continued toll of the virus on facilities after mandatory reporting began.
After May 24, California, Colorado, Georgia, Kentucky, and Pennsylvania’s state and federal case and death data were aligned, while Connecticut, Florida, Massachusetts, New Jersey, and Rhode Island had higher cases and deaths in the state data than the federal data. In contrast, state reports of COVID-19 cases and deaths were lower than those in the federal data in New Hampshire, Tennessee, and New York.
The researchers said that academic and policymakers’ analyses of nursing home-level determinants of COVID-19 cases and deaths have been compromised by their reliance on federal data.
“Consistent with the fact that states in the Northeast were hit hardest in the early months of the pandemic but generally experienced lower case and death rates in later months, we found that unreported cases and deaths represented a significantly larger share of year-end totals in the Northeast than in the South and West, where most cases and deaths occurred later,” the authors wrote.
Federal underreporting of nursing home cases and deaths “may lead to misleading conclusions about the role of different facility characteristics and state or federal policies in explaining COVID outbreaks,” the researchers said.
True tallies may never be known
In a commentary in the same journal, Elizabeth White, PhD, APRN, of Brown University School of Public Health, said that future research using the NHSN data needs to account for the significant data limitation uncovered by the study.
“A large body of research has already developed examining various facility- and community-level factors associated with COVID-19 outcomes in nursing homes using the NHSN data,” she said. “The study by Shen and colleagues provides important context for evaluating the rigor of these studies.”
White noted that only half of US states collected and publicly released nursing home coronavirus data in early 2020 and that the data limitations were compounded by scarce testing resources that limited timely, accurate diagnosis of COVID-19.
The estimates in the study “help to quantify the amount of underreporting in the federal NHSN data but they are just that: estimates,” she wrote. “As such, although current measures of COVID-19 prevalence and mortality reflect a devastating public health crisis for the nursing home population, it is likely that even more lives were impacted that will never be fully captured in existing data.”