Female Recruitment Into Cardiovascular Disease Trials

Abstract

Women are underrepresented in cardiovascular (CV) clinical trials, which may contribute to disease management and health equity barriers.1Cho L Vest AR O'Donoghue ML Ogunniyi MO Sarma AA Denby KJ Lau ES Poole JE Lindley KJ Mehran R Cardiovascular Disease in Women Committee Leadership CouncilIncreasing participation of women in cardiovascular trials: JACC council perspectives.J Am Coll Cardiol. 2021; 78: 737-751Crossref PubMed Scopus (30) Google Scholar Despite the National Institutes of Health (NIH) and the Food and Drug Administration policies recommending transparency and inclusion in reporting trial enrollment data by biologic gender, female representation in CV trials has not been adequately investigated. (Note: "biologic gender" and "gender identity" are The American Journal of Cardiology's preferred terminology and are used in place of terms such as "sex" and "gender".) The paucity of research addressing equity in trial representation based on gender limits the development, applicability, and adaption of novel CV therapies among women. We sought to characterize and analyze trends in gender representation among NIH-funded CV trials between 2000 and 2019 and identify recruitment strategies associated with female representation in trials with available protocols. The methods for this systematic review have been previously published.2Prasanna A Miller HN Wu Y Peeler A Ogungbe O Plante TB Juraschek SP Recruitment of black adults into cardiovascular disease trials.J Am Heart Assoc. 2021; 10e021108Crossref PubMed Scopus (21) Google Scholar We searched ClinicalTrials.gov for NIH-sponsored adult interventional CV disease trials completed between January 1, 2000 and December 31, 2019, with ≥1 US site, ≥1 arm, and ≥100 participants. A total of 2 reviewers independently extracted data from the results reported on ClinicalTrials.gov, available protocols, and primary publications. A third author adjudicated the discrepancies. We characterized trials by the distribution of female participants (<25%, 25% to <50%, ≥50%, corresponding closely to quartile distributions), study type (coronary artery disease, heart failure, stroke/cerebrovascular, hypertension, hyperlipidemia, arrhythmia, other), and intervention type (drug, procedure, behavioral, other). The participation to prevalence ratio (PPR) was obtained by dividing proportion of women in our sample's studies by the American Heart Association's estimated proportion in the corresponding US disease population.3Virani SS Alonso A Benjamin EJ Bittencourt MS Callaway CW Carson AP Chamberlain AM Chang AR Cheng S Delling FN Djousse L Elkind MSV Ferguson JF Fornage M Khan SS Kissela BM Knutson KL Kwan TW Lackland DT Lewis TT Lichtman JH Longenecker CT Loop MS Lutsey PL Martin SS Matsushita K Moran AE Mussolino ME Perak AM Rosamond WD Roth GA Sampson UKA Satou GM Schroeder EB Shah SH Shay CM Spartano NL Stokes A Tirschwell DL VanWagner LB Tsao CW American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics SubcommitteeHeart disease and stroke Statistics-2020 update: a report from the American Heart Association.Circulation. 2020; 141: E139-E596Crossref PubMed Scopus (4463) Google Scholar A PPR of 0.80 to 1.2 was considered adequate representation, <0.80 an underrepresentation, and >1.2 an overrepresentation.4Jin X Chandramouli C Allocco B Gong E Lam CSP Yan LL Women's participation in cardiovascular clinical trials from 2010 to 2017.Circulation. 2020; 141: 540-548Crossref PubMed Scopus (117) Google Scholar Independent-sample t tests with unequal variances or analysis of variance were used to test the statistical significance, with distribution of female participation as the dependent variable using Stata SE 16.1 (StataCorp, College Station, Texas). Among 156 trials identified on ClinicalTrials.gov, 54 were excluded for ≥1 prespecified eligibility criteria and 2 for starting enrollment before 2000. Of the 62 trials with available public or investigator-requested protocols, the median enrollment size was 305 (interquartile range 159 to 523) and the median age was 64 years (60 to 68). The median enrollment of women was 39% (range 27% to 47%). Overall, 19% (12/62) of trials specified recruitment targets for women in their protocol (Table 1). The median female recruitment target was 41% (30% to 50%), with 67% (8/12) meeting the recruitment targets. Gender-specific subgroup analyses were reported in 13% (8/62) of trials. Active recruitment strategies were used in 82% (51/62), passive recruitment strategies were used in 2% (1/62), and both active and passive recruitment strategies were used in 13% (8/62) of the trials. Community-based recruitment occurred in 16% (10/62) of trials, with 6.0% (4/62) specifying community input during the study design stages and 3.0% (2/62) listing community members as co-authors. Approximately 21% (13/62) of trials specified participant compensation beyond travel. There were no significant differences in recruitment strategies by female participation level.Table 1Recruitment characteristics of trials with protocols by distribution of female participants enrolledOverallDistribution of female participants<25%25%–<50%≥50%n=62n=12n=37n=13p ValueGender recruitment target defined?0.49 Yes12/62 (19%)2/12 (17%)8/37 (22%)2/13 (15%) No or not reported50/62 (81%)10/12 (83%)29/37 (78%)11/13 (85%)Defined gender recruitment target met?0.13 Yes8/12 (67%)0/2 (0%)6/8 (75%)2/2 (100%) No4/12 (33%)2/2 (100%)2/8 (25%)0/2 (0%)Active and/or passive recruitment?0.66 Active51/62 (82%)9/12 (75%)31/37 (84%)11/13 (85%) Passive1/62 (2%)0/12 (0%)1/37 (3%)0/13 (0%) Active and passive8/62 (13%)2/12 (17%)4/37 (11%)2/13 (15%) Missing2/62 (3%)1/12 (8%)1/37 (3%)0/13 (0%)EMR-based recruitment?0.14 Yes32/62 (52%)4/12 (33%)19/37 (51%)9/13 (69%) No or not reported30/62 (48%)8/12(67%)18/37 (49%)4/13(31%)Recruitment by referring healthcare provider?0.97 Yes38/62 (61%)7/12 (58%)23/37 (62%)8/13 (62%) No or not reported24/62 (39%)5/12 (42%)14/37 (38%)5/13 (38%)Community-based recruitment?0.11 Yes10/62 (16%)2/12 (17%)4/37 (11%)4/13 (31%) No or not reported52/62 (84%)10/12 (83%)33/37 (89%)9/13 (69%)Community input for study design?0.76 Yes4/62 (6%)1/12 (8%)2/37 (5%)1/13 (8%) No or not reported58/62 (94%)11/12 (92%)35/37 (95%)12/13 (92%)Community members included as co-authors?0.82 Yes2/62 (3%)0/12 (0%)2/37 (5%)0/13 (0%) No or not reported60/62 (97%)12/12 (100%)35/37 (95%)13/13 (100%)Participant compensation (other than for travel)?0.41 Yes13/62 (21%)2/12 (17%)7/37 (19%)4/13 (31%) No or not reported49/62 (79%)10/12 (83%)30/37 (81%)9/13 (69%) Open table in a new tab There were no significant differences in the proportion of women recruited between 2002 and 2017 (p = 0.82; Figure 1). The proportion of female participants was higher in hypertension versus nonhypertension trials (p = 0.006) and lower in heart failure versus nonheart failure trials (p = 0.037; Figure 1). The trials for most study types demonstrated <0.8 PPR compared with the US prevalence, except for hypertension trials (PPR = 1.14) and hyperlipidemia trials (PPR = 0.81). The proportion of female participants was higher in behavioral versus nonbehavioral intervention trials (p = 0.027), lower in drug versus nondrug trials (p = 0.005), and lower in procedure versus nonprocedure trials (p = 0.047; Figure 1). In this systematic review of NIH-funded CV disease trials, we found an underrepresentation of females, low prespecification of recruitment targets, low utilization of community-based and passive recruitment strategies, and no significant changes in recruitment trends throughout the past 2 decades. Similar to previous studies, hypertension-related trials and behavioral intervention trials had a higher proportion of female participants, whereas heart failure trials and drug/procedure intervention trials had a lower proportion.5Melloni C Berger JS Wang TY Gunes F Stebbins A Pieper KS Dolor RJ Douglas PS Mark DB Newby LK Representation of women in randomized clinical trials of cardiovascular disease prevention.Circ Cardiovasc Qual Outcomes. 2010; 3: 135-142Crossref PubMed Scopus (330) Google Scholar,6Tsang W Alter DA Wijeysundera HC Zhang T Ko DT The impact of cardiovascular disease prevalence on women's enrollment in landmark randomized cardiovascular trials: a systematic review.J Gen Intern Med. 2012; 27: 93-98Crossref PubMed Scopus (36) Google Scholar These findings indicate a possible pattern of prevention trials involving behavioral interventions, exhibiting more female participation than trials regarding more advanced CV diseases and invasive interventions. Some studies have observed that women appear more risk averse than men when evaluated on their willingness to participate in clinical trials.7Peterson ED Lytle BL Biswas MS Coombs L Willingness to participate in cardiac trials.Am J Geriatr Cardiol. 2004; 13: 11-15Crossref PubMed Google Scholar Therefore, our results support the importance of alliance building with women through more robust recruitment strategies, especially for trials with more advanced diseases and higher risk interventions. This systematic review demonstrates an urgency for diversifying the recruitment strategies that target women to ensure greater representation in CV disease trials. Although active recruitment methods, such as electronic medical record-based recruitment were most common in this review, previous research suggests that passive methods, such as mass mailing, radio advertisements, and community-based workshops, may reach more women, especially within racial/ethnic minority groups.8Lee RE McGinnis KA Sallis JF Castro CM Chen AH Hickmann SA Active vs. passive methods of recruiting ethnic minority women to a health promotion program.Ann Behav Med. 1997; 19: 378-384Crossref PubMed Google Scholar In addition, cultural competency and community support have been noted as important factors in increasing trial participation among women, especially in culturally diverse populations.9National Institutes of Health. Review of the literature: primary barriers and facilitators to participation in clinical research. Available at:https://orwh.od.nih.gov/sites/orwh/files/docs/orwh_outreach_toolkit_litreview.pdf. Accessed on April 11, 2022.Google Scholar,10Johnson DA Joosten YA Wilkins CH Shibao CA Case study: community engagement and clinical trial success: outreach to African American women.Clin Transl Sci. 2015; 8: 388-390Crossref PubMed Scopus (28) Google Scholar Participant-centered and community-engaged approaches, from recruiting female investigators who are part of the community to partnering with community advisory boards, are more likely to facilitate trust building, identify culturally relevant recruitment needs, and develop novel solutions for recruiting specific female populations.9National Institutes of Health. Review of the literature: primary barriers and facilitators to participation in clinical research. Available at:https://orwh.od.nih.gov/sites/orwh/files/docs/orwh_outreach_toolkit_litreview.pdf. Accessed on April 11, 2022.Google Scholar,10Johnson DA Joosten YA Wilkins CH Shibao CA Case study: community engagement and clinical trial success: outreach to African American women.Clin Transl Sci. 2015; 8: 388-390Crossref PubMed Scopus (28) Google Scholar Furthermore, gender and racial/ethnic minority groups are associated with higher cost barriers to research participation, including but not limited to money and time.11Bierer BE White SA Gelinas L Strauss DH Fair payment and just benefits to enhance diversity in clinical research.J Clin Transl Sci. 2021; 5: 159-160Crossref Scopus (8) Google Scholar Because only 21% of trials in our review indicated nontravel compensation, it may be important to consider higher payments to overcome the disproportionate costs and barriers to clinical trial participation for gender and racial/ethnic minority groups. Lastly, this systematic review demonstrates a need for clearer delineation between biologic gender (e.g., female) and gender identity (e.g., women) in CV clinical trials. Despite the growing push to clarify biologic gender-gender identity differences in clinical trials,12Franconi F Campesi I Colombo D Antonini P Sex-gender variable: methodological recommendations for increasing scientific value of clinical studies.Cells. 2019; 8: 476Crossref PubMed Scopus (48) Google Scholar major reporting institutions continue to require and/or allow only biologic gender data. Assessing and reporting biologic gender and gender identity differences are vital for trials because biologic gender and environmental factors, such as hormonal exposures, may affect CV disease course. Our systematic review is limited by the inclusion of only NIH-sponsored trials with available protocols. The strengths of this study include a prespecified, systematic approach to reviewing large-scale CV disease trials spanning nearly 2 decades. Our findings highlight areas of improvement for female enrollment during the planning and recruitment stages of CV clinical trials, including establishing concrete recruitment targets and applying multipronged recruitment strategies that involve community members and organizations. The authors have no conflicts of interest to declare.