Enrollment of Older Patients, Women, and Racial and Ethnic Minorities in Contemporary Heart Failure Clinical Trials: A Systematic Review.

IMPORTANCE

Despite the importance of age, sex, and race/ethnicity representativeness in clinical trials, limited data exist regarding the enrollment trends of these groups in contemporary heart failure (HF) trials.

OBJECTIVE

To characterize the representation of older patients, women, and racial and ethnic minorities in HF trials.

EVIDENCE REVIEW

We performed a systematic search of HF trials enrolling more than 400 participants published between January 2001 and December 2016 using PubMed/Medline and ClinicalTrials.gov. A total of 118 trials enrolling a cumulative 215 508 patients were included. Trial findings were compared with large epidemiologic studies indexed to hospitalization status and ejection fraction.

FINDINGS

Median number of participants per trial was 994 (interquartile range [IQR], 543-1899), enrolled from a median of 82 (IQR, 28-171) study sites. Overall, 94 trials (80%) enrolled patients with HF with reduced ejection fraction (HFrEF) exclusively. Mean (SD) age of trial participants was 65 (11) years (from 64 years in 2001 to 2004 to 65 years in 2013 to 2016), and 58 873 of 215 508 were women (27%; from 26% in 2001 to 2004 to 29% in 2013 to 2016); no significant temporal trends were observed (P ≥ .60 for both). Chronic HF with preserved ejection fraction (HFpEF) trials enrolled older participants (mean [SD] age, 71 [7] years compared with 65 [11] years for HFrEF and 66 [12] years for acute HF [AHF] trials; P = .01). Corresponding mean ages in US epidemiologic studies were 69 years for HFrEF and 73 years for both patients with HFpEF and patients with AHF. The HFpEF trials had a higher proportion of women (n = 4940 of 8845 [56%]) compared with HFrEF (n = 34 397 of 143 538 [24%]) or AHF (n = 11 013 of 34 633 [32%]) (P < .001). Corresponding weighted proportions of women in HFpEF, HFrEF, and AHF trials in epidemiologic studies were 62%, 29%, and 50%, respectively. Distribution of racial/ethnic groups was reported in 55% (47%) of the trials; 22% of the participants were not white (n = 27 463 of 124 980), with significant increase over time from 13% in 2001 to 2004 (n = 5606 of 44 616) to 30% in 2013 to 2016 (8421 of 28 073) (P = .01).

CONCLUSIONS AND RELEVANCE

In contemporary HF trials, older patients and women are consistently underrepresented. Race/ethnicity data are reported in less than half of trials; when reported, such data show that enrollment of nonwhite patients increased over time.