Representation of Older Adults in Cardiovascular Disease Trials Since the Inclusion Across the Lifespan Policy

Abstract

This review compares representation of older adults in interventional cardiovascular disease studies in the US before and after implementation of the National Institutes of Health’s Inclusion Across the Lifespan Policy.

Older adults traditionally have been underrepresented in cardiovascular disease (CVD) trials,^1,2 requiring extrapolation of evidence from younger populations.³ In response, the National Institutes of Health (NIH) established the Inclusion Across the Lifespan Policy in January 2019 to expand representation of older adults and children in clinical research.^4,5 The policy requires NIH studies to include individuals across the lifespan and to justify age-based exclusions. We set out to assess the early impact of this policy for the inclusion of older adults in CVD trials with a variety of funding sources.

Methods

For this review, we identified all phase 3 and phase 4 CVD interventional studies in the US for coronary artery disease, congestive heart failure, atrial fibrillation, peripheral artery disease, and valvular heart disease starting on or after January 25, 2018, and registered on ClinicalTrials.gov as of February 11, 2020. Trials were classified as prepolicy if they commenced enrollment 1 year before the NIH policy (January 25, 2018, to January 24, 2019) and as postpolicy if they commenced enrollment on or after January 25, 2019. We excluded 8 trials that were withdrawn, suspended, or terminated and 4 trials that were targeted at pediatric or pregnant populations (eFigure and eReferences in the Supplement).

Inclusion/exclusion criteria were abstracted, including age limits, trial type, study intervention, location, sponsoring agency, target study size and duration, and the study’s primary and secondary end points. These were compared using the χ² test or Fisher exact test for categorical variables and the Mann-Whitney test for continuous variables. Inclusion of patient-centric end points most relevant to geriatric populations was assessed, including health status, function, quality-of-life measures, mobility assessments, and patient-reported outcomes. Exclusion criteria that may disproportionately affect older adults were also examined, such as physical disability, decreased life expectancy, and cognitive impairment.¹

Results

We identified 97 eligible CVD trials, including 47 prepolicy and 50 postpolicy trials. Aggregate study characteristics for prepolicy and postpolicy trials are presented in the Table. Most trials did not include age limits in the prepolicy or postpolicy eras (14 of 47 [30%] vs 16 of 50 [32%]; P = .81). However, most postpolicy trials lacked geriatric-centric outcomes as first- or second-degree end points (32 of 50 [64%]), and exclusion criteria disproportionately affecting older adults were common both prepolicy (35 of 47 [75%]) and postpolicy (35 of 50 [70%]) (P = .62) (Figure).

Table. Characteristics of US Cardiovascular Disease Trials Before and After Implementation of the Inclusion Across the Lifespan Policy.

Characteristic	No. (%)		P value
	Prepolicy (n = 47)	Postpolicy (n = 50)
Age limit listed	14 (30)	16 (32)	.81
Upper age limit, y
≤70	1 (2)	1 (2)	>.99
≤75	3 (6)	3 (6)	>.99
≤80	5 (11)	9 (18)	.30
≤85	8 (17)	11 (22)	.54
≥90	6 (13)	5 (10)	.67
No age limit	33 (70)	34 (68)	.81
Cardiovascular disease category
Coronary artery disease	15 (32)	13 (26)	.52
Congestive heart failure/cardiomyopathy	21 (45)	18 (36)	.38
Atrial fibrillation/arrhythmias	7 (15)	14 (28)	.12
Peripheral artery disease	2 (4)	4 (8)	.68
Valvular heart disease	2 (4)	1 (2)	.61
Study intervention
Drug	36 (77)	36 (72)	.61
Device	6 (13)	2 (4)	.15
Diagnostic	1 (2)	2 (4)	>.99
Other	4 (9)	10 (20)	.15
Multicenter	24 (51)	27 (54)	.77
Location
US only	30 (64)	42 (84)	.02
International	17 (36)	8 (16)	.02
Sponsoring agencies
Public	10 (21)	5 (10)	.12
Industry	26 (55)	22 (44)	.27
Philanthropy	1 (2)	1 (2)	>.99
Institution	10 (21)	22 (44)	.02
Study characteristics
Target study size, median (IQR)	160 (52-416)	183 (60-694)	.53
Target study duration, median (IQR), mo	27 (19-43)	31 (16-48)	.89
Components of primary and secondary end points
MACE/mortality in primary end point	7 (15)	10 (20)	.51
Health status, function, QOL measures	16 (34)	16 (32)	.83
Gait-speed/mobility assessment	9 (19)	9 (18)	.88
Patient-reported outcomes	18 (38)	17 (34)	.66
Exclusions disproportionately affecting older adults	35 (75)	35 (70)	.62
Physical disability or functional limitations	6 (13)	8 (16)	.65
Decreased life expectancy	13 (28)	10 (20)	.38
Inability to give informed consent	8 (17)	10 (20)	.71
Cognitive impairment	8 (17)	6 (12)	.48
Preexisting medical conditions/comorbidities	19 (40)	26 (52)	.25
Clinician judgment on the basis of patient or treatment characteristics	7 (15)	12 (24)	.26
Conditions that may interfere with optimal data collection or interpretation	12 (26)	12 (24)	.86
Poor compliance with prescribed therapies	8 (17)	10 (20)	.71

Abbreviations: IQR, interquartile range; MACE, major adverse cardiovascular events; QOL, quality of life.

Figure. Barriers to Representation of Older Adults in Cardiovascular Disease Trials Before and After the Inclusion Across the Lifespan Policy.

Trial sponsorship was similar in the prepolicy vs postpolicy periods, with a trend toward fewer publicly funded studies (Table). The proportion of studies that included age limits was similar among industry-funded and publicly funded studies (18 of 48 [38%] vs 4 of 15 [27%]; P = .54), and it was also similar prepolicy and postpolicy in both publicly funded studies (3 of 10 [30%] prepolicy vs 1 of 5 [20%] postpolicy; P > .99) and industry-funded studies (9 of 26 [35%] prepolicy vs 9 of 22 [41%] postpolicy; P = .65). When the small number of NIH-funded trials was compared with all others, the proportion with age limits was similar (1 of 5 [20%] vs 29 of 92 [32%]; P > .99).

Inclusion of measures of health status, function, or quality of life (16 of 47 [34%] prepolicy vs 16 of 50 [32%] postpolicy; P = .83), gait-speed and mobility assessments (9 of 47 [19%] prepolicy vs 9 of 50 [18%] postpolicy; P = .88), and patient-reported outcomes (18 of 47 [38%] prepolicy vs 17 of 50 [34%] postpolicy; P = .66) was similar in all studies before and after the enactment of the policy.

Discussion

One year after adoption of the Inclusion Across the Lifespan Policy, challenges remained in ensuring adequate representation of older adults in CVD trials. One-third of trials included explicit age limits, with no difference before or after the policy, and two-thirds of trials used exclusion criteria that disproportionately affect older adults. In addition, only a minority of studies included patient- and geriatric-centric clinical trial end points after adoption of the policy.

Although this study highlighted early opportunities for improvement in representation of older adults in CVD trials, it was limited by the data available on ClinicalTrials.gov. Moreover, a given clinical trial might not explicitly exclude participants based on chronological age alone, but that does not necessarily translate into adequate enrollment of older participants. Further tracking of actual enrollment by age group will be critical to monitoring the success of the policy. Future policy changes that extend beyond just NIH-funded studies may also be necessary to improve the collection of evidence among older adults with CVD.

Supplement.

eFigure. Cardiovascular Clinical Trial Selection Process on ClinicalTrials.gov

eReferences. Cardiovascular Disease Trials Included in This Analysis