Establishing a baseline for multilingual capabilities of medical students at the Michigan State University College of Osteopathic Medicine

Methods

This descriptive, exploratory pilot study was approved as an exempt (category 2i) project after review by the Michigan State University Institutional Review Board (IRB# 0004540). After reviewing the literature and not finding a survey instrument available that fit this study team’s desired aims, the authors developed a pilot survey. The survey (Appendix A) was distributed to all current (as of June 2020) MSUCOM medical students, i.e., osteopathic medical students (OMS) years I–IV. The Qualtrics platform was utilized for this project to build, distribute, and collect data for the survey [6]. Prospective participants were informed both in the invitation email and at the beginning of the survey that they were being asked to participate in a study designed to ascertain information relating to medical students’ ability to speak multiple languages. In both the email and at the beginning of the survey, prospective participants were informed via computer-generated written text of the following: participation was voluntary, they would be able to skip any question and or withdraw from the study at any time, and there was no cost to participate. The final statement reads, “You indicate that you voluntarily agreed to participate in this research study by submitting the survey.”

Demographic information regarding age category, self-identified gender, and current medical student training level, i.e., the current medical school year (I–IV) they were enrolled in, was also collected.

Self-reported language and language fluency was identified by providing a list of the more common languages identified in the 2017 Doximity report regarding language barriers in US healthcare [7]. These languages consisted of English, Spanish, Hindi, French, Persian/Farsi, Chinese, Arabic, German, Russian, Italian, and Hebrew. An “other” option was included for respondents to write in any additional languages beyond those “common languages” identified from the Doximity report [7]. For each language specified, further questions (please see Appendix A survey instrument item #9) were asked to assess self-reported proficiency levels. Selection options derived from the “core scale” from the American Council on the Teaching of Foreign Languages (ACTFL) consisting of the following options were utilized: Novice (“Communicate minimally with formulaic and rote utterances, lists, and phrases”), Intermediate (“Create with language, initiate, maintain, and bring to a close simple conversation by asking and responding to simple questions”), Advanced (“Narrate and describe in the past, present, and future. Deal effectively with an unanticipated complication”), Superior (“Discuss topics extensively, support opinions, hypothesize. Deal with linguistically unfamiliar situations”), and Distinguished (“Ability to tailor language to a specific audience, persuade, negotiate. Deal with nuance and subtlety”) [8].

Surveys were sent out to all 1,226 medical students currently (as of 2020–2021) enrolled at the MSUCOM. The survey was first sent out on July 15, 2020, with reminders sent out at the beginning of both August and September of 2020. The survey was closed for further responses on October 1, 2020.

In addition to the Qualtrics software [6], which automatically supplies survey participant responses in a de-identified form, summary descriptive data are only presented in grouped, de-identified terms, and compiled/stratified by training level (year of medical school enrollment). The survey was distributed via email by the last author, KR, who is the MSUCOM Associate Dean of Admissions and Student Life. Those who chose to participate were offered the chance to win a $50 Amazon gift card. Winners were determined utilizing an online random number generator. Descriptive frequencies and percentages were calculated. Author SJW performed all statistical analyses utilizing SPSS Version 25 software [9].

Results

Of those 1,226 MSUCOM medical students asked to participate, n=698 responded, a 56.9% response rate (Table 1), Although data on age and gender were collected, this data was not utilized for the premise of this pilot study.

The total number of languages that the participants self-reported as being able to speak was 866, which is larger than the sample size because it includes participants who indicated that they spoke more than one language. This meant that on average, among those who spoke more than one language, the average number of languages spoken was the total number of languages (866) divided by the total number who reported that they spoke more than one language (382), thus 866/382=2.3 (Table 2).

Overall, the top three self-reported languages spoken were English with 322 (47.6%), Spanish with 169 (24.2%), and Arabic with 64 (9.2%) (Table 2).

In the “other” category, the top five languages that participants reported knowing included: Telugu (n=12), American Sign Language (ASL, n=11), Korean (n=11), Punjabi (n=11), and Urdu (n=11) (Table 3).

Finally, out of the n=670 (95.9% of total 698 respondents) individuals who answered further questions regarding their exposure to languages/cultures other than their own, 249 (37.2%) said that they had studied abroad (Table 4), 177 (26.4%) said they had lived in another country for more than six months (Table 5), and 228 (34.1%) responded “yes” to the question regarding whether they had been immersed in another culture for at least three months (Appendix B-18).

Discussion

This is a novel study examining the language proficiencies of medical students at the MSUCOM. Assessing language proficiencies of medical students may become more important as improving access to healthcare and the overall quality of healthcare are a priority [10]. The doctor–patient relationship has long been fundamental to quality care, and verbal communication is one of the keys to that relationship [11].

From census data, we know that areas relevant to the MSUCOM, such as southeast Michigan, have large Spanish- and Arabic-speaking populations [12]. The results of this study show that MSUCOM is currently educating medical students who have such language skills. More specifically, the authors observed that 169 (19.5%) of our respondents stated that they speak Spanish and 64 (7.4%) noted that they speak Arabic. Among the Spanish speakers, the majority, 110/184 (59.8%), noted novice or intermediate proficiency, which suggests that there is room for improving those skills to be effective communicators, particularly if aiming to engage with patients in local communities. Although not investigated, this could improve if students become engaged in medical electives within Spanish-speaking countries offered by the college. Conversely, the majority of the Arabic speakers, 49/68 (72.1%), self-identified as being advanced, superior, or distinguished in proficiency. As such, these MSUCOM students’ language proficiency could allow them to work more fluidly within communities with large Arabic-speaking populations. Several cities including Sterling Heights, located in Macomb County, as well as Dearborn, located in Wayne County, have significant numbers of the population speaking Arabic, and the statewide Arab population is roughly 277,534 [13].

In this study, the authors also evaluated whether or not the students studied abroad and if they had lived in another country outside the United States for more than six months. The authors posited that these questions might help assess cultural experiences rather than just language acquisition [14]. Among those asked, 177 (26.4%) students noted that they had lived outside the United States for more than six months, and 249 (37.2%) of MSUCOM medical students had studied abroad at some point in their education. In future studies, to assess beyond descriptive statistics, it may be beneficial to assess similar “cultural experiences” on a more granular level. For example, studies with a significantly larger sample size may have the statistical power to examine those who have studied abroad and or lived outside of the United States by self-reported, or better, a more objective measure, of second language proficiency.

The survey included 10 languages that respondents could select, and/or they could write in another language that may not have been included in the original 10. In the “other” category, students noted 42 different languages that they speak (Table 3). The top five of the “other” languages included two languages primarily from the Indian subcontinent, Telegu and Punjabi, with 12 and 11 student speakers, respectively. There were 11 speakers of Urdu, a language spoken in both Pakistan and India [15]. Eleven students self-reported that they speak Korean, and 11 students noted that they speak American Sign Language (ASL).

This is the first study that the authors are aware of that examined the language proficiencies of medical students. It is the authors hope that MSUCOM medical students’ experiences could be enhanced with curriculum changes based on language proficiency and placement within diverse communities. Curriculum changes that could be revised within the college include the primary care intensive preceptorships that second-year preclerkship students are required to complete as part of curriculum requirements [16]. By being placed in ambulatory settings in which medical students have similar language capabilities and proficiency to the local population, they may improve their ability to provide quality care.

The goal of the authors is to have the survey instrument utilized validated. In addition, the authors hope that the data obtained during this study could be replicated on a larger scale at other health professional schools not only within the state of Michigan, but also across the United States. This could help to build baseline knowledge on multilingual capabilities of individuals providing care.

Conclusions

Although this descriptive exploratory study has limitations, the authors hope that this information could pave the way for further studies that could become a valuable tool for medical school curriculum planning and for the enhancement of patient care. More specifically, the authors would suggest similar studies replicating/expanding upon examining the language skills of the physicians in graduate medical education.

Although the data for this pilot study were collected solely from students enrolled at MSUCOM, the methods utilized here could be replicated and expanded upon in future studies. For example, similar studies could be conducted through additional graduate medical education consortiums (statewide or otherwise), or at organized health professional schools including physician assistant schools and nursing schools. Such data could be beneficial in determining the placement of healthcare professional students and trainees with multilingual capabilities in improving health outcomes for local communities.

The authors plan to complete studies examining language skills of all of the medical students in a state or region. Osteopathic medical schools across the United States may be contacted by utilizing affiliate organization contact information listed on the American Association of Colleges of Osteopathic Medicine (AACOM) website to try to replicate such studies in diverse student populations across the country [17].

Additionally, the collection of larger study samples going forward could allow for more robust data on additional language proficiencies to be collected, for example, to determine whether the “limited” data for the remaining languages collected in this study would change with a much larger sample (see Appendix Tables B-9–B-17). To that end, the authors are considering conducting a larger study that would include participants from additional US medical schools.