Letter to the Editor Regarding Bernacki et al May 2021.

Abstract

Readers are invited to submit letters for publication in this department. Submit letters online at http://joem.edmgr.com. Choose “Submit New Manuscript.” A signed copyright assignment and financial disclosure form must be submitted with the letter. Form available at www.joem.org under Author and Reviewer information. Dear Dr. Brandt-Rauf: In the May issue of JOEM, Bernacki et al1 reported on 21,336 workers’ compensation filings, including 1898 claims for COVID-19, from the records of a single, private insurance company covering 11 states in the Midwest. Authors conclude that 1) healthcare workers are at greater risk of contracting COVID-19 than other sectors; 2) the “risk” of filing a claim for COVID-19 is lower than that of filing a claim for musculoskeletal injuries; 3) OSHA's risk stratification in healthcare is accurate; 4) submitting claims for COVID-19 among healthcare workers is indicative of risk. There are several limitations to the study's assumptions and methodology that challenge these conclusions. The sample comes from only one insurance company and the analysis uses only accepted claims. The absence of a count of those eligible to file workers’ compensation—by number of employees per industry—prohibits rate calculation and generalizability. The public sector, which includes around 15% of workers, is self-insured. The healthcare sector is often self-insured, as well, and larger healthcare organizations may not be included in this data set. Also, the insurance company may intentionally exclude certain employers or sectors from coverage because they are high risk. The absence of individual employers and sectors would certainly affect findings. Studies that use claims from state work comp agencies still identify health care as the most common industry for workers’ compensation claims but provide a different picture of the industries with the highest number of claims for COVID-19.2,3 The statistical methods are confusing. Table 2 is described as showing rate ratios, but there are no denominators, so it is not possible to calculate a rate or a ratio or even prevalence. We are wondering if investigators could have matched insurance policies to NAICS codes and come up with a payroll denominator. The case definition of COVID-19 is not clear. Did all claimants have positive COVID tests? Early in the pandemic, a covid test was not available to most people outside of healthcare. This may have influenced the results. States assumed to enact legislation or executive order of rebuttable presumption of eligibility for that state's workers’ compensation benefits were included as a variable (yes/no) in the regression models. On careful examination, these states vary in their presumptions: Illinois includes all essential workers, Michigan and Minnesota include only healthcare workers and first responders; Wisconsin and Missouri include only first responders and Wisconsin's law only covers claims from 3/12-6/10/2020. In other states (eg, Nebraska) the status is pending. This “apples and oranges” variable has questionable meaning. On the flip side, there are likely to be other critical issues/variables that cannot be captured in the database used (eg, provision of general health insurance and paid sick leave; hourly or temporary employment). Submitting workers’ compensation claims for COVID-19 is not necessarily indicative of risk of developing COVID-19. The authors state: “ What may be more important to employees and employers outside the healthcare industry, is our finding that risk of exposure and submitting a job related COVID-19 WC claim is lower than the mostly physical hazards of the industry itself of submitting a non-COVID-19 occupational injury or illness WC claim.” The paper presents no data on the “risk of exposure,” only on accepted claims. Many health systems provided time off for quarantine or isolation. Given that there were no treatments for COVID-19 that did not require hospitalization, there would be no need for hospital workers to file a claim for healthcare coverage. Furthermore, studies across states show that up to 50% of individuals with work-related injuries never file a workers’ compensation claim.4 The percent filing a claim for a work-related illness is even lower.5 Demonstrating the relationship between work and COVID-19 is important to convince the public health infectious disease community of the critical importance of considering “work” and “workplaces” in infectious disease surveillance and research investigations. It is critical that a canon of scientific literature be developed that enhances our understanding of the relationship between work and health.