According to Chambers et al. (2019), the answer varies widely depending on the health plan. The authors use a data from the Tufts Medical Center Specialty Drug Evidence and Coverage (SPEC) database. 203 specialty drugs issued by 17 of the 20 largest US commercial health plans are included in the database. This study examines nearly 5000 coverage policies across 395 drug-indication pairs.
The authors found that health plans often cited different sources in their policies for a given drug-indication coverage policy. For instance, although multiple health plans issued policies for all drug-indication pairs in the SPEC database, only 38% of the studies were cited by more than one plan in coverage policies for the same drug-indication pair.
The authors also found that there was wide variation in the number of studies cited across plans, ranging from 4.2 citations per coverage policy to 64.4 citations per coverage policy across the 17 plans of interest.
What types of evidence are used? This information is also highly variable.
Health plans cited RCTs with the greatest consistency (18 percent of plans) and economic evaluations with the least consistency (8 percent of plans)….RCTs accounted for a considerably larger percentage of the evidence cited by some health plans than others, ranging from 1 percent to 34 percent. Eight plans did not cite any economic evaluations, and one plan did not cite any evidence syntheses.
One limitation of this study is that it is unclear whether plans cited all of the evidence they reviewed when they determined their coverage policies. Also note that the analysis is only for commercial health plans, not those in the Medicare Advantage, Medicaid, or health insurance exchange (i.e., Obamacare) marketplaces.
Some may view this heterogeneity in evidence used as a bad thing, saying health insurers are not consistent. On the other hand, the diversity of evidence used leads to less group-think and different health plans can make their own decisions on the importance of different sources of evidence. Thus, while this finding of variability in evidence used is interesting, it is not obvious a priori whether this is welfare improving or destroying.