Year : 2007 | Volume
: 20 | Issue : 3 | Page : 82-
Exploring Language Barriers to Evidence-based Health Care (EBHC) in Post-graduate Medical Students: A Randomised Trial
LM Letelier1, N Zamarin1, M Andrade1, L Gabrielli1, G Caiozzi2, P Viviani1, A Riquelme1,
1 Pontificia Universidad Catolica de Chile, Santiago, Chile
2 Hospital de Urgencias Asistencia Publica, Santiago, Chile
L M Letelier
Lira 44, Santiago
Background: Understanding the written English language might be a barrier when teaching Evidence-based Health Care (EBHC) to Spanish-speaking physicians.
Aim: To quantify the magnitude of this potential barrier.
Method: Cochrane Review abstracts in English or in Spanish were randomly distributed among first-year residents at the Pontificia Universidad Catolica of Chile. Residents answered investigator-designed questionnaires to measure their comprehension while the time needed to complete the task was recorded.
Results: Groups were similar at baseline. Mean score for those reading in Spanish was 11.9 ± 2.8 (range 5 to 18) compared to 10.5 ± 3.8 (range 1 to 17) for those reading in English (p=0.04). Low scores (£ 9) were twice as frequent for the English group than for the Spanish group (16.7% vs 34.7%; p=0.042). The time to complete the task was also longer for the group reading in English.
Conclusion: Language should be taken into account when teaching EBHC to Spanish-speaking physicians.
|How to cite this article:|
Letelier L M, Zamarin N, Andrade M, Gabrielli L, Caiozzi G, Viviani P, Riquelme A. Exploring Language Barriers to Evidence-based Health Care (EBHC) in Post-graduate Medical Students: A Randomised Trial.Educ Health 2007;20:82-82
|How to cite this URL:|
Letelier L M, Zamarin N, Andrade M, Gabrielli L, Caiozzi G, Viviani P, Riquelme A. Exploring Language Barriers to Evidence-based Health Care (EBHC) in Post-graduate Medical Students: A Randomised Trial. Educ Health [serial online] 2007 [cited 2020 Sep 19 ];20:82-82
Available from: http://www.educationforhealth.net/text.asp?2007/20/3/82/101604
Clinicians are expected to use Evidence-based Health Care (EBHC) tools such as framing clinical questions, searching the literature and critically appraising the evidence in order to deliver the best care to patients (Oxman et al., 1993; Letelier & Moore, 2003). The medical literature is constantly growing and most of it is published in English. Hence, to practice EBHC, clinicians need skills to conduct literature searches, critical appraisals and also to understand the written English language.
Several reports address language barriers between physicians and their non-English speaking patients (Leyva et al., 2005; Moss & Roberts, 2005), but few examine the difficulties experienced by non-English speaking clinicians while trying to understand the medical literature in English. One trial (Gulbrandsen et al., 2002) found that physicians reading in their mother tongue had better comprehension than those reading in English. To our knowledge, the magnitude of language barriers has not been quantified in Spanish-speaking countries. Chile is a Spanish-speaking country where several medical schools promote EBHC. When teaching EBHC, language becomes a new barrier to the process, so it seemed important to start to look at the magnitude of this potential barrier.
Study objectives were: (1) determine physicians’ comprehension of the medical literature in English versus Spanish and (2) measure the time needed to achieve understanding.
We selected a Cochrane Review abstract on smoking cessation for prevention of coronary disease (Critchley & Capewell, 2003). Two questionnaires were created by the authors and validated by a group of faculty and students from the EBHC Unit, as well as by an expert on evaluation from the Medical Education Office at our university. The first questionnaire gathered the following information on baseline characteristics of participants: previous contact with Cochrane Reviews, self-reported English skills and formal experiences with the English language. The second questionnaire consisted of five open-ended questions regarding 19 key elements of the abstract to assess comprehension. This questionnaire was piloted on a group of undergraduate, postgraduate medical students and faculty associated with the EBHC unit at the Pontificia Universidad Catolica de Chile (PUC). After piloting, the questionnaire was refined by a medical educator, validating its appropriateness for the target population to ensure it could be answered by reading the abstract and not by previous knowledge on the topic. Both questionnaires were available in Spanish.
The instrument was organised into a set of three pages: first, the general information questionnaire; second, the Cochrane review abstract (English or Spanish), both original from the Cochrane Library and the Cochrane Library Plus (Spanish), but edited in the same format; and third, the comprehension questionnaire.
Randomization and sampling process
One investigator stapled and numbered each set using a computer-generated random sequence. The first and third pages were identical, so allocation of participants to either language was concealed by using numbered sets that did not allow investigators or participants to see the language on the abstract.
During an introductory session for physicians starting their Residency at PUC, attendants were asked to participate. After instructions, the sets were randomly distributed. All participants completed the first questionnaire simultaneously. Once completed, all participants unstapled the set at the same time and started reading the abstract while answering the comprehension questionnaire. They were encouraged to answer according to the abstract, not on previous knowledge, and to return the set immediately after completion. The time to complete the questionnaire was recorded for each participant.
Two independent reviewers, blinded to the language of the abstract, scored the comprehension questionnaires. A third reviewer solved differences in score by discussion. Disagreement in scoring occurred in less than 10% of the questionnaires.
Outcomes were: comprehension, defined as the score obtained in the questionnaire and the time needed to answer the comprehension questionnaire.
Baseline characteristics were compared using chi-square or Fischer’s exact tests. To compare primary outcomes between the groups and the association of outcomes with baseline characteristics, we used Student t-test for independent samples. To explain differences, a multivariate regression analysis was performed using the Statistical Software Minitab Release 14.
The study was approved by the Board of Postgraduate School of Medicine of the PUC and participants were asked for their consent to participate in the study.
Ninety-seven physicians agreed to participate (100% of attendants). Baseline characteristics were similar for the two groups (Table 1).
Table 1: Baseline characteristics of study groups
Mean score was 11.2 ± 3.4 points (range 1 to 18) and mean time to complete the questionnaire was 12.2 ± 2 minutes (range 7.7 to 17.7). When comparing the group reading in English with the group reading in Spanish, we found a statistically significant difference between the groups for both outcomes (Table 2). Mean score for those reading in Spanish was 11.9 ± 2.8 (range 5 to 18) compared to 10.5 ± 3.8 (range 1 to 17) for those reading in English (p=0.04). Considering the highest score was 18, scores £ 9 were defined as “low scores”; in the Spanish group 16.7% scored £ 9 whereas, 34.7% scored £ 9 in the English group (p=0.042).
Time to complete the questionnaire was 11.8 ± 2 minutes (range 7.7 to 17.2) for the Spanish group and 12.6 ± 2 minutes (range 8.8 to 17.7) for the English group (p=0.047).
Table 2: Results of comprehension and time needed to complete the task in the two groups
No correlation was found between high scores and the time to complete the task. In the Spanish group, participants acknowledging previous experience with Cochrane Reviews scored better (12.8 ± 2.4) than those who had not read a Cochrane Review (9.9 ± 2.5; p = 0.001). For the English group, scores were similar for those who had or had not read a Cochrane Review before (10.4 vs. 10.8).
Participants who reported to have confidence in their skills to read English scored better than those reporting little or no skills. There was no difference between those who had formal previous experiences with the English language (as defined in Table 1).
Multivariate regression analysis of baseline characteristics, showed that confident self-perception of English skills (pDiscussion
Our trial shows less comprehension and longer time to complete the task in the group reading in English, than in the one reading in Spanish. “Low scores” were twice as frequent when reading in English. Results confirm that language is a barrier to consider when teaching EBHC to Spanish-speaking students.
Although differences between the groups were statistically significant, the absolute differences were small and our results may underestimate the real difference. First, this may be due to a selected population of residents, who may not be representative of most physicians. Second, there might be an effect of using an abstract only, which is more structured than full-text articles where understanding might be easier. Differences might increase accordingly in relation to length and complexity of the text. Third, it was impossible to calculate optimal sample size, since we had no previous measures of the magnitude of the problem. A larger sample might reveal bigger differences.
Participants acquainted with Cochrane Reviews scored higher in the Spanish group, but not in the English group, pointing out that language may be a barrier for comprehension. Since we did not formally assess previous content knowledge or English skills, using concealed randomization, this should allow these variables to distribute randomly between groups.
Overall, a weakness of our study is not having a valid instrument. The questionnaire was validated for this particular abstract at our institution but it might not be applicable in other settings.
Medical students who are not confident in their English skills should be encouraged to learn English early during their training, and medical schools should consider English support programmes for this specific group. Therefore, it seems crucial to have local assessments regarding the magnitude of the problem in different settings.
Improving English skills seems to be the optimal solution. Meanwhile, organizations devoted to “evidence dissemination” should be encouraged to maintain their efforts in translating evidence into different languages.
Spanish-speaking physicians understand less and with a bigger effort when reading in English. In Spanish-speaking countries, English language appears to be a barrier to EBHC.
Source of funding: No external funding.
Conflict of interest: None.
The authors thank Sue Roff from the Centre for Medical Education, University of Dundee, for her help in reviewing this study.
CRITCHLEY, J. & CAPEWELL, S. (2003). Smoking cessation for the secondary prevention of coronary heart disease. Cochrane Database of Systematic Reviews, Issue 4. Art. No.: CD003041. DOI: 10.1002/14651858.CD003041.pub2. Available at: http://www.mrw.interscience.wiley.com/cochrane/clsysrev/articles/CD003041/frame.html
GULBRANDSEN, P., SCHROEDER, T., MILERAD, J. & NYLENNA, M. (2002). Paper or screen, mother tongue or English: which is better? Journal of the American Medical Association, 287, 2851-2853.
LETELIER, L.M. & MOORE, P. (2003). Evidence-based medicine: a view after a decade. Revista Medica de Chile, 131, 939-946.
LEYVA, M., SHARIF, I. & OZUAH, P.O. (2005). Health literacy among Spanish speaking Latino parents with limited English proficiency. Ambulatory Pediatrics, 5, 56-59.
MOSS, B. & ROBERTS, C. (2005). Explanations, explanations, explanations: how do patients with limited English construct narrative accounts in multi-lingual, multi-ethnic settings, and how can GPs interpret them? Family Practice, 22, 412-418.
OXMAN, A.D., SACKETT, D.L. & GUYATT, G.H. (1993). Users' guides to the medical literature. I. How to get started. The Evidence-Based Medicine Working Group. Journal of the American Medical Association, 270, 2093-2095.