The good news is that, at least at this point, the percentage of unemployed Americans has dropped from nearly 15% in April to 10.2% according to the most recent figures released by the Bureau of Labor Statistics (BLS).

That's below the number of workers projected to lose their jobs in all four studies, which range from about 14% in the Families USA study to nearly 20% for the KFF report.       

The RWJF analysis found a number of reasons for the disparities among the studies, noting that they did not all use the same time frames to calculate when the jobs were lost, and that one focused on the number of workers who lost coverage but did not account for family members of dependants who may have also been covered under an employer's plan.

The differences in methodology and array of results reflect the uncertainty and shifting terrain of COVID-19 America, wrote the report's authors, Jessica Banthin and John Holahan, both fellows at the Urban Institute's Health Policy Center. 

"A key reason for widely varying projections is that not all of these studies fully address the two main policy questions: how many people will lose employer-sponsored insurance and how many people will become uninsured as a result of the COVID-19 recession," they write. 

While there have been many such studies in recent months, the report said, the four cited were chosen "because they came out early, received some media attention, or represent a certain methodological approach."

Despite their variations, "early projections that provide partial answers to the main questions can still be valuable," it said.

The earliest of the studies, "How the Covid-19 Recession Could Affect Health Insurance Coverage," was released by the Urban Institute on May 4, and projected that if 15% of the workforce lost employment about 30 million people would lose employer-provided health insurance. The results was an estimated 5.1 or 8.5 million people losing insurance due to a job loss. 

That study used econometric projections based on past recessions and historical data comparing employer-based coverage and the unemployment rate, and resulted in a higher estimation than the Urban Institute's second study, released in July. 

"This is most likely because past recessions affected a wider range of workers and more workers with employment-based coverage than the current recession has thus far," the report said.  

The evidence thus far indicates that the current recession is "disproportionately affecting certain sectors of the economy while other sectors are relatively unchanged," it said.

"We anticipate that many workers who lose their jobs and employer-sponsored insurance will find other sources of coverage, so in addition to those two key questions, policymakers also want to know how many will enroll in subsidized coverage through Medicaid or the marketplaces."

The KFF study "Eligibility for ACA Health Coverage Following Job Loss," released in mid-May, found that almost 78 million people lived in a household in which someone lost their job between February and May, and estimated that almost 27 million people would lose health insurance in May because of that unemployment. 

That study estimated that nearly 80 percent of those 27 million people would be eligible for subsidized insurance Medicare or the marketplace. But the authors did not go so far as to estimate how many people would apply for coverage. "Research has shown that people eligible for subsidized coverage do not always take it up," the RWJF study noted. "Take-up varies by program and age of the enrollee, and it is always less than 100 percent."

The second Urban Institute report, "Changes in Health Insurance Coverage Due to the COVID-19 Recession" released July 13, used BLS statistics sorting job losses by industry, occupation, and demographic characteristics through May, and projected that during the last three quarters of 2020, 48 million nonelderly people will be in a family where someone loses their job due to the pandemic. 

That study also found that the current recession, unlike those in prior years, "is disproportionately affecting workers paid low wages," and that many of these workers did not have employer-provided health insurance to begin with. 

It estimated that 10.1 million people will lose employer-based coverage, but about 32% of those will be able to access employer-based insurance through another family member and that 28% would enroll in Medicaid.

Nonetheless, it still estimated that 3.5 million people would become uninsured.

The Families USA study released July 13 found that uninsured workers increased by 5.4 million between February and May, after 21.9 million workers lost their jobs. 

"Unlike the other three studies discussed in this report, the Families USA study assumes that coverage losses have already occurred," it said. "Also, it does not incorporate the dependents of workers who were covered by the employer-sponsored insurance. Incorporating dependents would likely increase the estimated effects by an additional number of people, somewhere between 50 to 100 percent of the number of workers."

Families USA's bluntly titled "The Covid-19 Pandemic and Resulting Economic Crash Have Caused the Greatest health Insurance Losses in American History" used BLS data on changes in employment and information from the U.S. Census Bureau's American Community Survey in a "straightforward" analytical approach, the report said. 

The study found that between 2014 and 2018, 29.8% of employed people had no insurance while 17.5% of those not in the labor force were uninsured. 

"The Families USA study reflects how the consequences of losing a job have changed on average since implementation of the ACA," the report notes. "But the analysis does not account for the unusual nature of the current recession and its disproportionate effects on workers with low wages, people who are less likely to have held employer-provided insurance before their job loss."

The report concluded that, while the pandemic and employment horizons remain unclear, recent findings indicate that the impact on the number of uninsured Americans may not be as grave as first feared. 

"We do not yet know how many people will lose both their jobs and health insurance coverage during the COVID-19 recession, and definitive data will not be available until next year," it said.

"Though employment rebounded somewhat after the huge job losses in March and April 2020, the economic recovery is uncertain and depends on the course of the coronavirus and efforts to mitigate its spread. Recent household surveys with smaller sample sizes than federal surveys have, however, indicated that net changes in insurance coverage thus far have been small. Without definitive data on how health insurance coverage is currently changing and will change in the coming months, models that predict the effects of widespread employment losses on coverage play an important role in alerting policymakers to potential outcomes."

Read more: 

• States ease health plan eligibility requirements for furloughed workers
• ACA exchange systems keep growing in core states
• Insurers loosen regulations to help curb coronavirus outbreak