OBJECTIVES:

To examine the number and type of published cost-effectiveness analyses (CEAs) on health services and procedures, and the data sources used to inform them.

METHODS:

We analyzed trends in CEAs published between 1976 through 2021 using data from the Tufts Medical Center CEA Registry and summarized both the total volume of studies and types of interventions under investigation. We also examined in detail the proportions of studies targeting pharmaceuticals and health services/procedures (which included care delivery interventions, surgical procedures, medical procedures, screening, health education or behavior, and maternal/neonatal/reproductive care). For CEAs published in 2021, we examined in detail the primary sources of data on effectiveness (e.g., randomized controlled trials [RCTs], systematic reviews, etc.).

RESULTS:

While the number of CEAs has increased consistently over the years, the proportion of studies devoted to pharmaceutical and service/procedure CEAs has remained relatively constant. In total, pharmaceutical, and service/procedure interventions represented 43% and 42% of CEAs respectively, while pharmaceuticals comprise roughly 15% of health spending. Of a total of 621 CEAs in 2021, approximately 70% of pharmaceutical CEAs (n=325) derived data directly or indirectly from RCTs, including direct RCTs (39.1%), extrapolated from RCTs (19.7%) or used systematic reviews of RCTs (14.5%). In contrast, only 40% of service/procedure CEAs (n=296) used these sources. Service/procedure CEAs instead relied more heavily on retrospective observational data, including previously published data (20.3 %) or literature review (11.8%); about 10% used prospective studies with primary data collection. Differences in sources of effectiveness data between pharmaceutical and service/procedure CEAs were statistically significant (Chi-square test, p=0.009).

CONCLUSIONS:

While the proportion of pharmaceutical and service/procedure CEAs has remained relatively constant over time, the type and quality of data sources informing these evaluations differs significantly. Researchers should explore opportunities to promote CEAs of services/procedures, including through prospective clinical trials and rigorous observational studies for effectiveness data.