Researcher Perceptions of Inclusion of Study Participants Who Use Languages Other Than English.

Importance: Despite constituting 8% of the US population, people who speak languages other than English (LOE) are consistently underrepresented in health-focused research. To make research more inclusive and generalizable, it is crucial to better understand researcher perspectives on barriers to inclusion of participants using LOE and solutions to promote language justice. Objective: To assess researcher perspectives on barriers to and best practices for inclusion of participants using LOE and to generate strategies to make research more inclusive. Design, Setting, and Participants: This exploratory cross-sectional survey study used an anonymous digital Qualtrics survey distributed between March 1 and June 30, 2023. The study was conducted among principal investigators (PIs) and research coordinators at the University of Pittsburgh, Pennsylvania. Any PIs or research coordinators who conducted human participant research in the past 5 years were eligible to participate. A convenience sample, using department emails, research electronic mailing lists, a recruitment repository, and word of mouth, was recruited. Main Outcomes and Measures: Survey questions, drafted with input from community partners, assessed researcher demographics, experience working with participants using LOE, barriers to inclusion, and recommendations to increase research inclusivity. All variables were analyzed using descriptive statistics to calculate means, SDs, and frequencies. Results: There were 339 respondents who completed the survey (260 cisgender females or women [76.7%]). Of these respondents, 127 (37.5%) were PIs and 212 (62.5%) were research coordinators. In terms of race and ethnicity, 8.8% were Asian; 3.8% were Black, African, or African American; 2.4% were Hispanic, Latino, Latina, Latine, or Latinx; 0.9% were Middle Eastern or North African; 70.5% were White; 6.5% were of other race or ethnicity or were multiracial; 2.7% preferred not to say; and 4.4% had missing data. Most respondents primarily conducted clinical research (239 [70.5%]), and 170 (50.1%) worked with participants using LOE in the prior 5 years. In 188 reported cases in which inclusion occurred, 78 respondents (41.5%) did not proactively include participants using LOE but instead reactively included interested participants after studies had begun. Respondents listed lack of training, time and scheduling challenges around interpreter services, and budget constraints as barriers to inclusion. Recommendations to improve inclusion were made by 272 respondents (80.2%), 265 (78.2%) of whom suggested access to low-cost interpreters and translators, 249 (73.5%) of whom suggested training, and 272 (80.2%) of whom suggested availability of validated measures in different languages. Conclusions and Relevance: In this survey study of health PIs and research coordinators affiliated with 1 academic institution, respondents faced multiple barriers to including participants who use LOE in their studies. Because a lack of language representation may compromise the quality and applicability of research, purposeful individual and institutional investments are needed to overcome these barriers.