Big gaps in long COVID data: How employers can make health coverage decisions

You might expect that data tells a story that’s pretty straightforward. With health claims data, for example, the number of sick people with a condition should more or less line up with the number of members with benefits claims for that condition. Traditionally, organizations make decisions related to their benefits offerings based on this data. But if the last few years have taught us anything, it’s to expect the unexpected.

Take post-acute sequelae of COVID-19 — better known as long COVID — as an example. Estimates from the Brookings Institution and the CDC suggest that around 16 million people in the U.S. between the ages of 18 and 65 have long COVID.

The problem is that health claims data doesn’t match the number of people suffering from this condition, and that spells trouble for employers who need to assess and adjust their benefits to ensure that what they offer meets people’s needs.

Here is why long COVID presents data reporting challenges — and what employers need to know in order to respond.

Long COVID claims data is incomplete

Analysis of 2021 and 2022 health claims data by Springbuk shows a surprisingly low rate of long COVID cases compared with reporting on the general population. The analysis of hundreds of thousands of employee health claims found that less than 2% of members who were diagnosed with COVID-19 in August 2021 or later were also diagnosed with long COVID. This is in stark contrast to CDC data, which suggests the figure could be around 20%.

The disparity in data is a problem because it could hurt employers’ ability to plan future health benefits coverage. It can also lead to under-covering employee health needs or result in unplanned cost overruns for employer-provided health care plans, not to mention turnover resulting from employees feeling dissatisfied with their benefits.

Employers depend on data to make decisions that will ensure employees receive the care they need while ensuring that spending remains efficient. But discrepancies in data make it difficult to make informed decisions.

Employers must remain flexible

Why is there a discrepancy between commercial claims data — indicating that less than 2% of people who contracted COVID-19 also experienced long COVID — and studies that suggest this figure is closer to 20%? There are several reasons why this occurred. Understanding them can help organizations make informed decisions about their benefits offerings in the coming year.

• There’s a lack of alignment on how long COVID should be defined: One problem with long COVID is that it’s still considered a new condition, and there are no specific diagnostic tests to identify it. Because symptoms are often vague or nonspecific (e.g., cough, fatigue, headache), long COVID can be easily mistaken for other diseases. As a result, when people are sick and visit their doctor, it’s possible that long COVID is misdiagnosed as something else, like the common cold or bronchitis, further depressing the number of reported long COVID cases. What’s more, experts disagree on a definition for the condition, with the CDC saying long COVID patients experience symptoms for “more than four weeks,” and the WHO saying symptoms must last for at least two months. With so much uncertainty around how to define this disease, many doctors may hesitate to make a diagnosis, leading to the condition being underreported.
• Long COVID didn’t appear in health claims data until October 2021: For long COVID cases to appear in claims data, physicians must diagnose patients. But because long COVID is such a new disease, a health care reporting code for this condition didn’t even exist until October 2021. In other words, individuals who had COVID-19 in 2020 and early 2021 may have had higher rates of long COVID. We can’t know for sure, though, because there was no reporting code for this condition.
• COVID-19 variants are unpredictable: Variants of COVID-19 also seem to play a role in long COVID rates, making it difficult to project the number of future cases. For example, a 2022 study published in The Lancet suggests that rates of long COVID may be up to 50% lower with patients who contract the omicron variant compared with those with the delta variant. Because we can’t predict the nature of future COVID-19 variants, it will, in turn, be a struggle to predict the number of future long COVID cases.
• The role of vaccines is becoming clearer: While researchers are still learning about long COVID, several studies have shed light on the role of vaccines. The UK Health Security Agency found that among people who contracted COVID-19, those who are vaccinated are less likely to suffer from long COVID than those who are unvaccinated. Meanwhile, analysis of data from the U.S. suggests that vaccinated patients are up to 10 times less likely to experience two or more long COVID symptoms than those who are unvaccinated.

Employers need to recognize the threat of long COVID

Benefits claims data may also be skewed by employees who have been forced to drop out of the workforce due to long COVID. If an employee finds they can no longer work because their symptoms prevent them from being able to perform, they may lose their health insurance and their diagnosis, in turn, won’t appear in claims data.

The effects of long COVID on the labor force are far-reaching. In fact, about 1.1 million people are out of work right now due to long COVID, according to the Brookings Institution. What’s more, an additional 46% of long COVID patients — or 2.1 million people — reduced their work hours instead of taking time off, according to a study published in The Lancet.

These figures are shocking when you consider the severe labor shortage in the U.S. There are currently over 11 million open jobs in the country, but only about 6 million unemployed workers to fill them, according to the BLS. While there are myriad reasons for this labor shortage, long COVID is contributing greatly to the problem. Research from the Brookings Institution cited above estimates that long COVID alone is responsible for 15% of all currently unfilled positions.

Related: 3 things that could make your health data unreliable

Given long COVID’s outsized influence on the workforce, it is key for employers to remain flexible. It’s critical to recognize that we are still in the midst of a pandemic and that many people are dealing with health conditions that we do not fully understand. By providing flexible work arrangements, generous sick leave, and by making changes to benefits plans that fit employees’ ever-changing needs, organizations can respond to these challenges without falling victim to decreased productivity or increased turnover.

To understand long COVID, let data be your guide

Employers have a great responsibility to provide benefits that will ensure that their people get the care they need — and a thorough understanding of data plays a big part in fulfilling it. To understand the impact of long COVID, leaders must look at downstream costs and utilization patterns for individuals who have had COVID-19. By doing so, organizations can get a clearer picture of how long COVID is affecting their population’s wellbeing and productivity.

My advice: Don’t underestimate the impact of long COVID on your people.

Dr. Janet Young serves as lead clinical scientist on Springbuk’s Data Science and Methods team.