Research Derived From Medicare’s Coverage With Evidence Development Program

TL;DR

This study examines research on medical treatments approved through Medicare's Coverage With Evidence Development program to understand their scientific impact.

Contribution

The study characterizes the scientific impact of research on medical items and services approved by Medicare's Coverage With Evidence Development program.

Findings

Published research articles were analyzed to assess their scientific impact.

The study provides insights into the types of medical items and services evaluated through the program.

Abstract

This cross-sectional study assesses published research articles about medical items and services approved by the Coverage With Evidence Development program to characterize their scientific impact.

Tables

Table 1.. Medicare-Approved Studies for Items and Services With Coverage With Evidence Development Requirements and Associated Research Publications, Overall and Stratified by Therapeutic Area, Study Type, and Study Completion Status

Characteristic	Studies, No. (%)		Research publications, No. (%)	Research publications per study, median (IQR)
	CMS-approved studies	CMS-approved studies with at least 1 research publication
Overall	116 (100)	53 (45.7)	556 (100)	0 (0-3)
Therapeutic area
Cardiology	68 (58.6)	30 (44.1)	454 (81.7)	0 (0-3.8)
Hematology	20 (17.2)	5 (25.0)	14 (2.5)	0 (0-0.5)
Musculoskeletal	0	0	0	0
Neurology	10 (8.6)	5 (50.0)	20 (3.6)	0 (0-3)
Oncology	10 (8.6)	8 (80.0)	58 (10.4)	1 (0.75-12)
Otolaryngology	5 (4.3)	2 (40.0)	2 (0.4)	0 (0-1)
Pulmonology	2 (1.7)	2 (100)	7 (1.3)	3.5 (2-5)
Psychiatry	1 (0.9)	1 (100)	1 (0.2)	1 (1-1)
Study type
Clinical trial	86 (74.1)	43 (50.0)	267 (48.2)	0 (0-2)
Registry	23 (19.8)	8 (34.8)	281 (50.4)	0 (0-4)
Claims-based study	4 (3.4)	1 (25.0)	4 (0.7)	0 (0-3)
Expanded access	3 (2.6)	1 (33.3)	4 (0.7)	0 (0-4)
Study completion statusa
Complete	32 (27.6)	21 (65.6)	122 (21.9)	1 (0-4.8)
Active, not recruiting	26 (22.4)	17 (65.4)	143 (25.7)	2 (0-5.8)
Recruiting	26 (22.4)	9 (34.6)	262 (47.1)	0 (0-2.5)
Not yet recruiting	1 (0.9)	0	0	0
Suspended	1 (0.9)	0	0	0
Terminated	14 (12.1)	4 (28.6)	9 (1.6)	0 (0-1)
Unknown status	16 (13.8)	2 (12.5)	20 (3.6)	0 (0-0)

Table 2.. Characteristics of Research Publications Associated With Items and Services With Coverage With Evidence Development Requirements, Stratified by Medicare-Approved Study Type

Characteristic	Research publications, No. (%)
	Derived from clinical trial studies (n = 267)	Derived from registry studies (n = 281)	Derived from claims-based or expanded access studies (n = 8)
Therapeutic area
Cardiology	183 (68.5)	263 (93.6)	8 (100)
Hematology	12 (4.5)	2 (0.7)	0
Musculoskeletal	0	0	0
Neurology	16 (6.0)	4 (1.4)	0
Oncology	46 (17.2)	12 (4.3)	0
Otolaryngology	2 (0.7)	0	0
Pulmonology	7 (2.6)	0	0
Psychiatry	1 (0.4)	0	0
Study completion statusa
Complete	118 (44.2)	4 (1.4)	0
Active, not recruiting	123 (46.1)	12 (4.3)	8 (100)
Recruiting	9 (3.4)	253 (90.0)	0
Not yet recruiting	0	0	0
Suspended	0	0	0
Terminated	9 (3.4)	0	0
Unknown status	8 (3.0)	12 (4.3)	0
Cited by FDA	0	1 (0.4)	0
Cited by clinical practice guideline in Scopus	76 (28.5)	117 (41.6)	1 (12.5)
Citation count, median (IQR)	22 (6-61)	25 (10-65.5)	23 (6-45)
Time since publication to July 1, 2023, median (IQR), mo	36 (18-67)	64 (39-103)	21 (15-54)

Full text

Introduction

Medicare’s Coverage with Evidence Development (CED) program requires beneficiary participation in clinical studies as a condition of coverage for an item or service.^1^ Since CED’s initiation in 2005, concerns have been raised about coverage requirements and associated costs.^2,3^ However, CED offers a trade-off: beneficiary access to novel technologies and manufacturer access to the Medicare market in exchange for research addressing product benefits and harms.^4^ CED may also have positive externalities, generating evidence to inform care decisions beyond Medicare policy, including for the Food and Drug Administration (FDA).^4,5^ In this study, our objective was to identify how many published research articles were derived from CED-approved studies and characterize their scientific impact.

Methods

In accordance with the Common Rule, this study was exempt from ethics review and informed consent because it was not human participant research. We followed the STROBE reporting guideline.

We used the Centers for Medicare & Medicaid Services (CMS) website to identify all items and services with CED program requirements as of July 1, 2023.^6^ For each, we determined therapeutic area and all CED-approved studies, including National Clinical Trial numbers. For each study, we extracted study type (registry vs nonregistry) and completion status from ClinicalTrials.gov. We used multiple search strategies to identify all published English-language original research and systematic reviews that reported results from these CED-approved studies as of July 1, 2023, excluding commentaries and editorials; ClinicalTrials.gov, PubMed, and Google Scholar were searched. For each research publication, we used Scopus to determine number of times cited and proportion cited by clinical practice guidelines as of February 2024. We reviewed FDA product labels to ascertain whether research from CED-approved studies was cited as of March 2024.

Additional study details are provided in eMethods in Supplement 1. Descriptive statistics were used to summarize key characteristics, analyzed in Excel, version 16.63.1 (Microsoft Corp).

Results

As part of the CED program, CMS approved 116 unique studies for 26 items and services, including 19 medical devices (73.1%), 5 blood- or tissue-derived products (19.2%), and 2 drugs (7.7%). As of July 2023, 32 CED-approved studies (27.6%) were complete and 26 (22.4%) were active but no longer recruiting; 68 (58.6%) were in cardiology, while 86 (74.1%) were clinical trials, 23 (19.8%) registries, 4 (3.4%) claims-based studies, and 3 (2.6%) expanded access studies (Table 1).

Results from 57 CED-approved studies (49.1%) had been reported as of July 2023, including 53 (45.7%) with 1 or more research publications and 4 (3.4%) reporting results only on ClinicalTrials.gov, with higher rates among CED-approved studies that were completed (21 [65.6%]) and active but no longer recruiting (17 [65.4%]) (Table 1). In total, there were 556 unique research publications, including 267 (48.2%) from CED-approved clinical trials and 281 (50.4%) from CED-approved registries (Table 2). Clinical trial publications were cited a median (IQR) of 22 (6-61) times, and 76 (28.5%) were cited by clinical practice guidelines. Registry publications were cited a median (IQR) of 25 (10-65.5) times, with 123 (43.8%) cited by clinical practice guidelines and 1 (0.4%) by the FDA.

Discussion

In this cross-sectional analysis of all 26 items and services with CED program requirements, approximately half of 116 CED-approved studies had reported results as of July 2023, totaling 556 unique publications, even though fewer than one-third had been completed. These publications were widely cited, including by clinical practice guidelines. We cannot conclude with certainty that these studies would have been conducted without CED, and CED-approved study completion and results reporting are suboptimal. Nevertheless, the findings suggest that CED contributes not only to CMS decision-making but also the medical research enterprise.

Study limitations include possible completion of some CED-approved studies and publication of additional articles since our July 2023 search. Moreover, despite using multiple publication search strategies, we may have missed some studies, and some might have been used by clinical practice guidelines or the FDA but not cited, rendering our estimates conservative. Publication and citation numbers are imperfect measures of scientific impact.