Marketplace Pulse: COVID Testing and Commercial Insurance

The Marketplace Pulse series provides expert insights on timely policy topics related to the health insurance marketplaces. The series, authored by RWJF Senior Policy Adviser Katherine Hempstead, analyzes changes in the individual market; shifting carrier trends; nationwide insurance data; and more to help states, researchers, and policymakers better understand the pulse of the marketplace.

The Centers for Medicare & Medicaid Services (CMS) recently made a much anticipated clarification of guidance regarding insurer responsibility for COVID testing. Going forward, insurance plans will no longer be able to use medical necessity as a standard for coverage and will instead be required to cover all diagnostic COVID tests with no cost sharing for enrollees. Whether symptomatic or asymptomatic, enrollees seeking testing are deemed to be making an “individualized clinical assessment” of their need, and their test must be covered. This guidance applies to all kinds of approved COVID tests, including rapid response and presumably home tests, as well as tests conducted as part of state or local public testing availabilities such as drive-through locations. Coverage is not required, but it is permitted for tests that are conducted as part of workplace or public health surveillance.

This guidance accompanies other new efforts to expand testing, including the establishment of a National Pandemic Testing Board, plans to increase testing in schools and other priority locations, and investments in rapid response tests and the domestic supply chain in general. The new guidance to insurers is an effort to increase the demand for testing by removing a financial barrier. Many of the Administration’s other efforts target supply. Together, these changes should have significant implications for the volume of COVID testing, particularly that which is covered by insurance. What do we know about current patterns?

Levels

New data on COVID testing in a commercially insured population suggest that the level of testing has risen over time but is far below optimal levels. An analysis of 2020 data for nearly 11 million members shows that the share getting tested each month rose from nearly zero in March to more than seven percent in September (Figure 1). During this period, most members getting tests got tested once—approximately 1.1 million members received 1.6 million tests. But almost 90 percent of members got no test at all, at least none of which resulted in a claim. While little demographic information is available, those getting tested were similar in terms of age to the overall population of enrollees, with a median age of 38 (versus 37 in the overall population), and were somewhat more likely to be female (58 percent of members who got tested versus 52 percent overall).

Types

There are two major kinds of COVID diagnostic tests. Molecular tests, usually polymerase chain reaction (PCR) tests, detect genetic material, while antigen tests test for proteins from a virus particle. Antigen tests can provide results more quickly, but have lower levels of specificity, meaning that they are more likely than PCR tests to result in a false negative. PCR tests usually do not provide results for several days, although there are some rapid PCR tests. Among this group of commercially insured enrollees, the great majority of COVID tests were PCR, but the share which were antigen tests rose considerably in the last three months, from zero percent to about eight percent between June and September. This is probably not representative of the broader universe of testing, since some states which track all tests by type report a higher share of antigen tests, and many of the testing sites that do not accept insurance specialize in rapid response or antigen testing.

Prices

There was pretty much no cost sharing to speak of for this population of commercially insured patients. For antigen tests, the median copay was $51 in the top one half of a percent of tests, and zero for all points below. For PCR tests, there was less cost sharing than that. Yet the allowed amounts increased quite a bit over the time period studied (Figure 2). For PCR tests, the price bumped up between March and May from $62 to $91 and rose by another nearly 10 percent by September. Much of this increase is attributable to the use of a new laboratory billing code (U0003). In April, CMS changed its payment policy for COVID testing for labs using high throughput technologies. The payment rate was raised to $100 and the new code (U0003) was released. The percent of PCR tests that were reimbursed with the high throughput code rose to more than 60 percent by May, and roughly 80 percent by September. The allowed amount in commercial insurance has stayed quite close to the $100 allowed by Medicare. In an effort to reward speed, beginning in January 2021, CMS dropped the payment for high throughput U0003 to $75. However, they added a new code—U0005—which allow labs to bill an additional $25 if they can provide results within two calendar days. Allowed amounts for antigen tests are quite a bit lower ($54 in September versus $104 for PCR tests), but their median price floated up about 10 percent between June and September for reasons that are not clear.

Likely Changes

The new guidance is likely to accelerate demand for rapid testing for people who are asymptomatic. In particular, this may increase testing among young adults, who are more likely to be asymptomatic and are responsible for much of the COVID transmission. This should increase the demand for antigen testing, since rapid PCR tests are relatively scarce. Even without the new guidance, the share of antigen tests rose fairly rapidly between June and September. Antigen tests can currently cost as much as $130 at locations that don't accept insurance. Providers who have benefited from this temporary cash market will have to adjust their business model when patients realize they no longer have pay out of pocket. While this could cause some upward pressure on allowed amounts for antigen tests, the new guidance and the accompanying investments in supply may also speed the development of cheaper rapid tests, including home tests and rapid PCR tests. The new guidance may also have implications for where testing happens. In this population of commercial enrollees, PCR and antigen testing took place in quite different locations. Eighty percent of PCR testing occurred in hospital outpatient or laboratory settings, much of the latter probably being retail pharmacy. Eighty percent of antigen tests took place in urgent care settings and physicians’ offices. Given the changes in reimbursement, we will likely see broader availability of rapid testing in all settings.

Pump Up the Volume

If the U.S. population as a whole had the same basic rate of testing as this population of commercial enrollees, a rough estimate is that approximately 25 million people would have been tested in September 2020. The actual number tested was about 28 million. While this estimate is probably a bit of a lowball, since it under-measures antigen testing, which is less consistently reported, and may also underestimate the potentially higher rate of testing among older adults, it does suggest that the “medical necessity” screening criterion has constrained our level of testing. Enrollees got relatively few tests, and it appears that not much testing happened outside of insurance. That which did occurred largely among those who were willing to pay out of pocket, not necessarily those for whom testing would be most beneficial.

We have consistently undertested. A group of experts convened by the Rockefeller Foundation recently suggested that 70 million tests each week would be a reasonable goal for January. The actual total was closer to 60 million for the entire month. In the first two months of this year, testing levels declined, a trend which public health experts decry. The medical screening standard is not the only reason for our inadequate testing, but it was a preventable error which capped testing at suboptimal levels, likely inhibited the development of rapid tests, and has surely contributed to the number of infections and deaths. This new guidance, along with the release of our first national testing strategy, are overdue but welcome, and should give a badly needed boost to both the quantity and speed of testing.

The exhibits in this article are provided by Wakely Consulting Group. Please download the appendix for important details regarding the methodology, assumptions and key limitations with this information. Wakely does not intend to create a reliance to other parties receiving this work, who should retain their own experts in interpreting this information.