Spanish-speaking patients had higher odds (OR 5.7) of signing consent for studies with Spanish consent available at study opening

Description #

When a study had a Spanish consent document available at the time of study opening, patients whose primary language was Spanish had markedly higher odds of signing consent for it than for studies without one: OR 5.7 (95% CI 3.8–8.5, P < 0.001) (Extended Data Table 4). This is the study's most direct signal that up-front availability of a language-matched consent document raises enrollment of speakers of that language.

"Patients with Spanish as their primary language had higher odds of signing consent documents for studies that had Spanish consent at study opening than those without (OR 5.7, 95% CI 3.8 to 8.5, P < 0.001) (Extended Data Table 4)." (Maria, 2023, p. 857)

Methods Context #

What? #ⓘ

The observable: the odds that a Spanish-primary patient signed consent, comparing studies with vs without a Spanish consent document available at study opening.

"Odds for patients with Spanish as their primary language of signing of consent to studies with Spanish translation compared to studies without them." (Maria, 2023, Extended Data Table 4)

How? #ⓘ

GEE logistic regression clustered by patient, restricted to the 12 studies with Spanish consent at study opening (10 industry, 2 non-industry) versus those without.

"Of these 18 studies, 12 had Spanish consent documents at study opening: 10 industry-sponsored and 2 non-industry-sponsored (Extended Data Table 4)." (Maria, 2023, p. 857)

Who? #ⓘ

Spanish-primary patients (the most common non-English primary language, n = 231 patients) across studies at one cancer centre, 2013–2018.

"The most common primary languages other than English were Spanish (40.8%, n = 231)" (Maria, 2023, p. 856)

Other Notes #

The authors caution this association is subject to reverse causation — a study anticipating many Hispanic patients may be more likely to have Spanish consent at opening (see qualifying caveat).

Caveats #

• Spanish-consent-at-opening enrolment association is subject to reverse causation (Maria 2023) The finding that Spanish-primary patients enrolled more in studies with Spanish consent at study opening is vulnerable to reverse causation, so it cannot be read as evidence that providing translation causes higher enrolment. The presence of translated documents at study opening in single studies of less common languages (e.g. Thai) signals that document availability tracks anticipated enrolment; and a study expected to enrol many Hispanic patients would be more likely to have Spanish consent available at opening in the first place. The authors flag this analysis for cautious interpretation and future investigation.