Diversity and inclusion in chronic hepatitis B randomised controlled trials: A systematic meta-epidemiological review

Abstract

Background & aims: Chronic hepatitis B (CHB) affects global populations unequally, with variable prevalence and pathophysiology. Clinical trials must balance efficiency with adequate representation of the populations most likely to benefit from the interventions they test. We aimed to investigate diversity and inclusion in CHB trials.

Methods: We performed a meta-epidemiological study of randomised controlled trials recruiting people with CHB published in MEDLINE and Embase, January 2010 to July 2023. We extracted participant age, sex, country of recruitment, race and ethnicity, and hepatitis B genotype. We calculated proportions of trials reporting participant demographics and results by demographics (transparency). We compared participants proportionately to global populations affected by CHB of different demographics (representation), and examined demographic-based trial exclusion criteria.

Results: Among 287 trials (81.9% single-country, 18.1% multinational) with 57,503 participants (median size: 102, IQR: 60-185), 97.9% tested drug interventions. Most trials reported participants' age distribution (285, 99.3%) and sex (278, 96.9%). However, only 37.3% (107) trials reported race and ethnicity, 84 (29.3%) reported genotype, and, among multinational trials, only 19 (36.5%) reported recruitment numbers by country. Less than 3% trials reported demography-stratified results. Female sex was under-represented (42.0% people with CHB, 28.7% trial participants). Geographic disparities between those affected by CHB and trial participation were marked for Africa (31.0% vs. 0.01%; under-representation), and Americas or Europe (5.7% vs. 14.0%; over-representation). Many trials had exclusion criteria based on age (71.4% children, 41.5% older adults) or sex-related (157, 54.7%), mostly excluding women who were pregnant, breastfeeding, or of reproduction potential.

Conclusions: Clinical trials for CHB are not inclusive of women and people in Africa. Researchers, funders, and publishers should actively consider diversity and inclusion of trials.

Impact and implications: Clinical trials must balance the need to recruit homogenous participants to efficiently measure an intervention's effectiveness, with the need to produce evidence that can be applied to the whole population affected by a disease. We found chronic hepatitis B (CHB) clinical trials often failed to report basic demographic characteristics of participants, and had under-representation of women and people living in Africa. Given varied disease pathophysiology and treatment needs among different groups, this suggest a mismatch of evidence generation compared with the populations needing treatment, whereby the benefits and harms of different interventions across populations are not being adequately studied. We suggest relevant stakeholders, including researchers, funders, and publishers of CHB clinical trials, should actively recruit under-represented populations, target interventions to those most at need, and either consider demographic factors in results reporting and analysis, or make data easily available for interrogation.

Keywords: Chronic hepatitis B; Diversity; Equity; Hepatitis B virus; Inclusion; Meta-epidemiology; Representation.

Graphical abstract

Fig. 1

PRISMA flowchart of search and record inclusion.

Fig. 2

Tornado plot demonstrating the proportions of the global population with CHB to those of participants in CHB clinical trials. (A) By sex∗ and hepatitis B virus genotype^† and (B) by geographic regions of recruitment^‡. Representation ratio defined as the proportion of people in CHB trials divided the proportion of people globally with CHB. ∗Included participants with reported sex from 278 trials (n = 56,111). ^†Included participants with reported HBV genotypes from 84 trials (n = 14,060). Genotypes with a global distribution less than 1.0% were not included. Global data adapted from: Velkov S, Ott JJ, Protzer U, Michler T. The global hepatitis B virus genotype distribution approximated from available genotyping data; Genes (Basel), 2018;9:495. ^‡Included participants from 235 single-country trials and 19 multinational trials with reported recruitment locations (n = 51,841). Geographic regions were classified per United Nations geographic regions and World Bank income classification. CHB, chronic hepatitis B.

Fig. 3

Geographic distribution. (A) People living with chronic hepatitis B in 2021∗ and (B) total weighted number of participants included in trials (n = 51,841)^†. ∗Data extracted from Global Burden of Disease Study 2021. Countries with no available prevalence data are shown in white. ^†Trials with no geographic information are not shown (n = 33).

Fig. 4

Exclusion criteria in CHB clinical trials (N = 287) by publication year. The dashed line indicates 50%.