Improving diversity in research and trial participation: the challenges of language

Abstract

The COVID-19 pandemic has brought into focus issues of health inequality and the disproportionate disease burden experienced by ethnic minority communities, compared with the population as a whole.1Khunti K Singh AK Pareek M Hanif W Is ethnicity linked to incidence or outcomes of COVID-19?.BMJ. 2020; 369m1548Crossref PubMed Scopus (253) Google Scholar High-quality research is the basis for an evidence-based approach to the delivery of health care. It is widely accepted that some subgroups of patients might respond differently to interventions due to genetic, social, and cultural factors.2Conforti F Pala L Bagnardi V et al.Cancer immunotherapy efficacy and patients' sex: a systematic review and meta-analysis.Lancet Oncol. 2018; 19: 737-746Summary Full Text Full Text PDF PubMed Scopus (313) Google Scholar Therefore, diversity within study cohorts is crucial to ensure research is generalisable,3Oh SS Galanter J Thakur N et al.Diversity in clinical and biomedical research: a promise yet to be fulfilled.PLoS Med. 2015; 12e1001918Crossref PubMed Scopus (228) Google Scholar benefits all of society, and does not perpetuate existing inequalities. In some demographically heterogeneous countries, such as the UK, national funding bodies have issued guidance that aims to promote equality within research.4Wellcome TrustClinical trials policy.https://wellcome.org/grant-funding/guidance/clinical-trials-policyDate: 2021Date accessed: April 18, 2021Google Scholar The aim of improving representation in research involvement is to improve on the current status quo, and is scientifically justified in light of differential disease risks and ethically justified given differences in access to care. However, issues of feasibility and knowledge gaps hamper efforts to achieve this improved representation. An important under-researched area in the context of diversity is methods to accommodate the language needs of under-served ethnic minorities, some of whom might have low first language literacy, health literacy, or English language proficiency (in majority English-speaking societies, such as the UK). Engaging a substantial proportion of such groups to make the recruited sample more representative of the target population is unlikely to be easy or cheap. Simply translating a document into a heritage language is unlikely to be an effective solution for this complex challenge. Such an approach is impervious to working with communities to facilitate sustained understanding and enhance trust. Defining the scale of this issue is difficult. A review on telehealth interventions for type 2 diabetes, a condition for which ethnic minority groups are over-represented and experience poorer outcomes, found that only two-thirds of included trials reported the ethnic composition of participants.5Isaacs T Hunt D Ward D Rooshenas L Edwards L The inclusion of ethnic minority patients and the role of language in telehealth trials for type 2 diabetes: a systematic review.J Med Internet Res. 2016; 18: e256Crossref PubMed Scopus (12) Google Scholar Language and socioeconomic variables were not routinely described, making it difficult to compare characteristics of the recruited sample with the target population and to examine within-group and between-group differences (where language barriers are present for some but not all ethnic groups). These findings suggests the need to improve reporting practices to disambiguate language barriers from cultural factors. In establishing a need for translation, and other related support in the planning stage of a trial, it is important to distinguish trial contexts where language is required solely as part of the informed consent process (an ethical imperative for all research participants) and trial contexts where language is part of the intervention. Use of the same word or the direct translated equivalent word might convey differing concepts.6Gregg J Saha S Communicative competence: a framework for understanding language barriers in health care.J Gen Intern Med. 2007; 22: 368-370Crossref PubMed Scopus (26) Google Scholar Communicating symptom experiences such as pain can differ across languages, and if these form part of the intervention or outcome assessment, comparisons are likely to be affected. Despite the ethical issues, there are some contexts where there could be meaningful grounds for exclusion based on language proficiency (eg, the nature of the intervention is different when offered in different languages), but this would need to be justified. There are gaps in current knowledge of the most effective methods to inform best practice guidance aimed at more inclusive recruitment to clinical trials centring on language. To address these gaps, the following points are proposed for researchers considering language translation in clinical trials. First, language-specific information for the target population (eg, proficiency in English and other languages) alongside ethnicity information should be considered at the trial conceptualisation phase.7Treweek S, Banister K, Bower P, et al. Developing the INCLUDE ethnicity framework—a tool to help trialists design trials that better reflect the communities they serve. Trials (in press).Google Scholar More accurate and detailed data on first language literacy in diverse populations can better assess demand and lead to a more accurate cost-benefit analysis of providing language translation as part of clinical trials. Second, matters of inclusion need to be considered at all stages of the research process, from initial research focus through to implementation, addressing key barriers from previous literature that limit participation.8Witham MD Anderson E Carroll C et al.Developing a roadmap to improve trial delivery for under-served groups: results from a UK multi-stakeholder process.Trials. 2020; 21: 694Crossref PubMed Scopus (17) Google Scholar Third, robust translation methods should be employed (eg, forwards and backwards translation with input from multiple translators and clinical reviewers).9Brislin RW Back-translation for cross-cultural research.J Cross Cult Psychol. 1970; 1: 185-216Crossref Scopus (6413) Google Scholar Where possible, in the written mode, emphasis should be placed on achieving conceptual equivalence, with appropriate readability levels, assessed via end-user testing.10Isaacs T Murdoch J Demjén Z Stevenson F Examining the language demands of informed consent documents in patient recruitment to cancer trials using tools from corpus and computational linguistics.Health (London). 2020; (published online Oct 13.)https://doi.org/10.1177/1363459320963431Crossref PubMed Scopus (1) Google Scholar Improving access and promoting diversity in research is a priority for governments and funders seeking to reduce health and socioeconomic disparities.7Treweek S, Banister K, Bower P, et al. Developing the INCLUDE ethnicity framework—a tool to help trialists design trials that better reflect the communities they serve. Trials (in press).Google Scholar However, strategies promoting inclusion cannot be overburdensome and limit the feasibility of carrying out research. Pragmatic approaches are needed to ensure that research is open to a broader participant base, without undue participant exclusion decisions being made due to the non-clinical criterion of language barriers and with additional resources available to support the inclusion of ethnic (including linguistic) minorities. Once an intervention or therapy has shown efficacy at a population level, targeted approaches, where there is greater scope for tailoring of interventions, could be successful. KK is a director of the University of Leicester Centre for Black Minority Ethnic Health, trustee of the South Asian Health Foundation, chair of the Ethnicity Subgroup of the Scientific Advisory Group for Emergencies (SAGE), and member of Independent SAGE. AW and TI declare no competing interests.

Topics & Concepts