Census data predict that Latinos will constitute the largest minority population in the United States by 2010.1 By 2050, the U.S. Latino population will exceed 100 million and will account for approximately one quarter of the U.S. population.1 From 1990 to 2000, the number of people who spoke Spanish at home in the United States increased from 17.3 to 28.1 million.1
There currently exists an insufficient supply of Spanish-speaking health care clinicians to meet the demands of the Spanish-speaking population in many communities in the United States.2 Patient satisfaction, adherence to medical advice, and cost-effective use of ambulatory services are highest when there is language concordance between clinician and patient.3–8 In situations where language discordance exists, the most effective means of communication is through professional interpretation.9–12 In 2003, the Institutes of Medicine published Unequal Treatment: Confronting Racial and Ethnic Disparities in Healthcare that called for an increase in the availability of interpretation services to overcome language barriers that could negatively affect the quality of health care.13 It is clear that the use of family members, ancillary staff, or other nonprofessional interpretation services have higher misinterpretation rates than those that occur with professional interpretation.14–16 Furthermore, language barriers between clinicians and patients may alter clinical management and result in worse clinical outcomes.17–22 Professional interpretation services, however, may often be underutilized due to cost, time inefficiency, use of nonprofessional interpretation, reliance on suboptimal Spanish language skills of the clinician, or reliance on suboptimal English skills of the patient. A recent publication reported that 53% of pediatrics residents in one residency program used their “nonproficient” Spanish language skills “often” or “every day” rather than using professional interpretation services, despite their self-reported inability to communicate effectively in Spanish.23
We carried out this study to add to this literature by assessing the interactions that medical students and residents from multiple specialties had with Spanish-speaking patients at one university hospital. Furthermore, we examined trainees' willingness to have their language skills formally evaluated and to participate in further language training.
In the fall of 2004, we surveyed all fourth-year medical students and non-first-year residents in family practice, pediatrics, medicine–pediatrics, internal medicine, obstetrics–gynecology, and emergency medicine at the University of Rochester School of Medicine and Dentistry. First-, second-, and third-year medical students, along with first-year residents, were excluded from the study to ensure that all respondents had at least one year of clinical experience. The focus was on trainees who had recent outpatient experiences, and thus surgery residency programs were not included. The sample size was calculated to test the hypothesis that no more than 20% of medical trainees who report having “no or rudimentary” language skills had used those skills to take a history and/or give medical advice without the assistance of professional interpretation services. This figure was based on the assumption that if greater than one out of five trainees reported this, it could be considered a common situation. Based on a two-sided alpha of .05, a power of 90%, an estimated response rate of 60%, and an estimate that 75% of respondents would report having “no or rudimentary” language skills, we calculated that a sample size of at least 160 was needed for findings to be meaningful.
The total number of medical students and residents who qualified to enter the study was 275. Of these potential 275 participants, 12 were excluded because they were unable to gain access to their hospital mail or e-mail accounts during the nine weeks of study enrollment. Therefore, 263 medical trainees were eligible for this cross-sectional study: 97 medical students and 166 residents. The entire population of 263 participants was surveyed in order to increase the likelihood of obtaining a sufficient number of responses for analysis. The University of Rochester School of Medicine and Dentistry was chosen for study because it is located in a city that has experienced a 47% rise in its Latino population from 1990 to 2000, representing 13% of the overall population.1 Professional in-person interpretation and phone-service Spanish interpretation are available at all times at the university.
The survey instrument was designed, pretested, and pilot tested in the spring of 2004. The final draft of the questionnaire consisted of 13 items (see the Appendix). Participation in the study was voluntary and was approved by the University of Rochester Institutional Review Board in August 2004. Questionnaires were distributed by hospital mail service at three-week intervals for a total of three possible mailings from September to November of 2004. Code numbers were assigned to each questionnaire to ensure confidentiality. Survey packets consisted of a brief cover letter, the one-page questionnaire, an addressed return envelope, and a one-time $2.50 incentive coupon to be used at the medical center coffee shop. E-mail reminders were sent out at the time of each mailing. An electronic version of the questionnaire was distributed at the time of the third mailing for those who had not responded to the first two mailings. This was done to facilitate the return of questionnaires by those who did not have access to their university mailboxes during the time of the study.
All data were entered into an electronic database within 72 hours of receipt of a completed questionnaire. Each questionnaire's responses were verified against the database at the conclusion of the study period to decrease the likelihood of error. Interim analysis was performed only to generate a list of nonresponders for future mailings. Standard statistical software was utilized to perform all statistical analyses.
Analyses were based upon each respondent's reported level of Spanish language fluency (question 9). Respondents characterized their Spanish as either none, rudimentary, conversational, or fluent. Rudimentary was defined as being “able to ask a few basic questions and understand simple responses.” Conversational was defined as being “able to have a two-way medical conversation with a patient (take a history and give medical advice without the assistance of an interpreter).”
In addition to language fluency, communication skills were assessed by responses to questions 1 and 6. In these questions, respondents were asked to verify that if they had categorized themselves as either none or rudimentary Spanish speakers, they also acknowledged that they would be unable to communicate effectively with a Spanish-speaking patient. This was done to generate a subpopulation of respondents who should use interpretation to conduct a visit with a Spanish-speaking patient. Analyses were then performed to determine if trainees who should be relying on interpreters were consistently doing so. Furthermore, in circumstances when interpretation was used, questions were aimed at determining if professional interpretation was used more frequently that other nonprofessional forms of interpretation. Multiple questions were designed to describe trainees' willingness for language evaluation and training. Last, one question was designed to explore whether trainees believed that Spanish-speaking patients received a lower quality of care compared with English-speaking patients. Descriptive and chi-square analyses were performed. Comparisons were made by training category (medical student versus resident) and level of Spanish fluency, when appropriate.
Two hundred and forty-one of the 263 participants (92%) returned the questionnaire. Respondents by training category ranged from 18 of 21 (86%) in the obstetrics–gynecology residency program to 24 of 24 (100%) in the medicine–pediatrics program. The medical student response rate of 89 of 97 (92%) was the same as the pooled responses from all residency programs (92%, p > .1). Of the two hundred and forty-one respondents, all but one reported that they had seen patients who were able to communicate only in Spanish during their training at the university, and 165 of 241 (68%) reported seeing such patients at least monthly.
Three of the questions were created to identify which respondents would require an interpreter to communicate effectively with a Spanish-speaking patient (see the Appendix, questions 9, 1, and 6). In question 9, respondents assessed their degree of Spanish language fluency based on the defined categories of none (77 [32%]), rudimentary (122 [51%]), conversational (31 [13%]), and fluent (11 [5%]). Responses by training category are shown in Figure 1. The respondents with “no” or “rudimentary” Spanish language skills (199/241 [83%]) were analyzed together, as they reported “less than conversational” language skills. Of these 199 respondents who reported less than conversational Spanish language skills, 174 (87%) stated that they were “not” able to, and 21 (11%) could only “occasionally,” conduct a medical visit effectively in Spanish (question 1).
Trainees were also asked how often they could understand Spanish-speaking patients (question 6). The same 199 respondents with less than conversational language skills stated never (31 [16%]), rarely (64 [32%]), sometimes (84 [42%]), or most of the time (20 [10%]). Therefore, nearly all of the 199 respondents with “none” or “rudimentary” Spanish language skills acknowledged that they could not consistently conduct a medical visit in Spanish, and therefore should be using interpretation services.
Trainees were asked how often they used interpretation services rather than relying on their own Spanish language skills (Table 1). Of the 199 who reported having less than conversational Spanish skills, 105 (53%) noted that that they did not always use an interpreter when seeing patients who only speak Spanish. Respondents with less than conversational language skills were more likely to always use an interpreter than those with conversational/fluent language skills (94/199, or 47% versus 2/42, or 5%; p < .001). The trend was that medical students were more likely than residents to always use an interpreter, but this did not reach statistical significance (p > .2).
Further questions were asked regarding the use of interpretation services at the university. Two hundred and twenty-one (92%) respondents reported using Spanish interpreters at some point during their training. Each of these respondents was asked to estimate the percentage of time that they used various sources of interpretation (question 4). Responses were averaged together and showed that trainees called upon nonprofessional interpretation services, such as family or staff members, more often than they called on professional services (58% versus 42%, p < .05). Reasons for utilizing nonprofessional services were recorded (question 8). Respondents were instructed to check all responses that apply, therefore percentages totaled more than 100%. One hundred and sixty-two (73%) noted the long waiting time for an interpreter, 129 (58%) stated that the use of family/friends/staff provides adequate interpretation, 84 (38%) noted family preference, and 41 (19%) cited other reasons. Only two respondents reported that nonprofessional interpretation services were used because of an emergency situation.
Of all respondents, 226 (94%) expressed a desire to improve their Spanish language skills, 194 (80%) reported they would probably or definitely participate in individual language training, and 168 (70%) noted that it is at least possible that they will use their language skills as an attending physician. The analysis was performed again with the removal of those who have “no” Spanish language skills. This analysis was performed in order to identify if there exists a subset of trainees with at least rudimentary language skills who have interest in learning and using Spanish. The responses of the 164 (68%) who reported having at least rudimentary language skills are noted in Table 2. One hundred and thirty-nine (85%) stated that they probably or definitely would be willing to participate in further language instruction, and 132 (80%) stated that it is at least possible that they will use their Spanish language skills as an attending physician. Furthermore, 115 (70%) reported that they would at least be “possibly” willing to have their language skills formally evaluated.
Of all 241 respondents, 122 (51%) reported that Spanish-speaking patients probably or definitely receive a lower quality of health care than do English-speaking patients. This did not differ significantly between medical students and residents (52/89, or 58% versus 70/152, or 46%; p > .05). There was, however, a significant difference in responses about this issue between those with at least conversational language skills and those with less than conversational language skills (30/42, or 71% versus 92/199, or 46%; p < .01).
At the time of our study, virtually all medical trainees at this one university hospital had experienced clinical encounters with Spanish-speaking patients. Trainees with self-assessed poor Spanish proficiency frequently reported communicating with Spanish-speaking patients without the assistance of an interpreter. If an interpreter was utilized, it was most likely that nonprofessionals were used. Trainees cited many reasons for not utilizing professional interpretation services, most commonly because of time constraints. Most trainees predicted that they will use their Spanish as attending physicians, and the majority would welcome further training and formal evaluation of their language skills. They also perceived that Spanish-speaking patients receive a lower quality of care compared to English-speaking patients.
We found only one previous peer-reviewed publication about the use of Spanish by medical trainees.23 It demonstrated that pediatrics residents in one training program used their inadequate Spanish to communicate directly with Spanish-speaking patients. The current study was conducted in a different geographic area from that of the study just mentioned, and included both medical students and residents from multiple specialties. The results of this study support the previously published data by demonstrating that trainees underutilize available interpretation services. These findings add to the literature by showing that this underutilization may generalize to medical students and residents in multiple specialties. Based on these results each medical educational facility should investigate how their trainees communicate with patients who do not speak English. Those responsible for training medical students and residents should emphasize that ensuring adequate clinician–patient communication is the clinician's responsibility. It should also be emphasized that time inefficiencies or other barriers should not become reasons to carry out inadequate communication. To facilitate the use of interpreters, each institution should implement policies that aim to minimize inefficiencies and other barriers associated with interpreter use.
There also needs to be a mechanism to ensure that those who are communicating in Spanish without the assistance of an interpreter possess adequate language skills. One of the general competencies outlined by the Accreditation Council for Graduate Medical Education states that a resident should possess “interpersonal and communication skills that result in effective information exchange and teaming with patients, their families, and other health professionals.”24 For those trainees who choose to use Spanish, a formal evaluation of their skills should be implemented to ensure that “effective information exchange” between clinician and patient exists. Use of a second language could be viewed as a clinical skill that is effective only if done properly, recognizing that improper use could have clinical consequences.18
If an evaluation process was implemented it would serve as a reminder to those who score poorly that they should be using an interpreter. This may motivate these trainees to matriculate into language training courses, which could eventually decrease the need for interpretation services and improve patient care.25–26 For those who score well it would add to their repertoire of credentialed clinical skills. Research should focus on validating a standardized Spanish language evaluation tool that could be used for medical professionals.
There are a number of limitations to this study. The most important of these is that trainees reported their own levels of Spanish fluency. Analyses were performed with the assumption that trainees were not misclassifying themselves regarding their language fluency. No data exist on the ability of medical trainees to self-assess second language fluency, so it is reasonable to question the validity of our respondents' self-assessments. It is noteworthy, however, that the use of other questions to verify a respondent's inability to communicate with Spanish-speaking patients suggests that misclassification was minimal.
Second, responses to the survey were based upon clinical interactions with Spanish-speaking patients living in Rochester, New York. U.S. census data show that Puerto Ricans account for over 75% of the Latino population of Rochester, but represent only 10% of all Latinos nationally.1 Given that substantial linguistic differences exist among Latino populations, it is conceivable that medical trainees may perceive and characterize their ability to communicate in Spanish differently with Puerto Rican patients than with patients from other Latino subgroups.27 Therefore, these results obtained in Rochester may not generalize to other Spanish-speaking populations in the United States.
A third potential limitation to this study is that the word “translator” was used in place of “interpreter” in the survey. It is recognized that translation refers to written material and interpretation refers to dialogue. The two terms are frequently used interchangeably by medical trainees at the institution, and therefore this substitution most likely did not affect the validity of the responses.
Last, only linguistic aspects of communication were ascertained. The ability to communicate effectively does not guarantee culturally sensitive medical care.28 Given the diversity of backgrounds of people who speak Spanish, fully understanding all cultural components relevant to patient care is a formidable challenge, but critically important.
In conclusion, these data show that at the time of our study, medical trainees across disciplines at this one teaching hospital were using poor Spanish skills directly with patients. Trainees acknowledged that this occurs frequently but are hesitant to utilize available interpretation services. These findings (1) suggest that efforts should be made to teach trainees the appropriate use of interpretation services, (2) indicate that development of a validated Spanish language evaluation tool is needed for trainees to regularly use their Spanish, and (3) should alert medical educators that similar problems may exist in their institutions.
1 United States census data 〈http://www.census.gov
〉. Accessed 27 January 2006. United States Census Bureau, Washington, DC, 2005.
2 Timmins C. The impact of language barriers on the health care of Latinos in the United States: a review of the literature and guidelines for practice. J Midwifery Womens Health. 2002;47:80–96.
3 Flores G, Fuentes-Afflick E, Barbot O, et al. The health of Latino children: urgent priorities, unanswered questions, and a research agenda. JAMA. 2002;288:82–90.
4 Weinick RM, Krauss NA. Racial/ethnic differences in children's access to care. Am J Public Health. 2000;90:1771–74.
5 Manson A. Language concordance as a determinant of patient compliance and emergency room use in patients with asthma. Med Care. 1988;26:1119–28.
6 Morales LS, Cunningham WE, Brown JA, Liu H, Hays RD. Are Latinos less satisfied with communication by health care providers? J Gen Intern Med. 1999;14:409–17.
7 David RA, Rhee M. The impact of language as a barrier to effective health care in an underserved urban Hispanic community. Mt Sinai J Med. 1998;65:393–97.
8 Mosen DM, Carlson MJ, Morales LS, Hanes PP. Satisfaction with provider communication among Spanish-speaking Medicaid enrollees. Ambul Pediatr. 2004;4:500–4.
9 Kuo D, Fagan MJ. Satisfaction with methods of Spanish interpretation in an ambulatory care clinic. J Gen Intern Med. 1999;14:547–55.
10 Lee LJ, Batal HA, Maselli JH, Kutner JS. Effect of Spanish interpretation method on patient satisfaction in an urban walk-in clinic. J Gen Intern Med. 2002;17:641–45.
11 Haffner L. Translation is not enough. Interpreting in a medical setting. West J Med. 1992;157:255–59.
12 Jacobs EA, Shapard DS, Suaya JA, Stone EL. Overcoming language barriers in health care: costs and benefits of interpreter services. Am J Public Health. 2004;94:866–69.
13 Smedley B, Stith A, Nelson A. Institute of Medicine. Unequal Treatment: Confronting Racial and Ethnic disparities in Healthcare. Washington DC: The National Academies Press, 2003.
14 Elderkin-Thompson V, Silver RC, Waitzkin H. When nurses double as interpreters: a study of Spanish-speaking patients in a US primary care setting. Soc Sci Med. 2001;52:1343–58.
15 Laws MB, Heckscher R, Mayo SJ, Li W, Wilson IB. A new method for evaluating the quality of medical interpretation. Med Care. 2004;42:71–80.
16 Baker DW, Parker RM, Williams MV, Coates WC, Pitkin K. Use and effectiveness of interpreters in an emergency department. JAMA. 1996;275:783–88.
17 Flores G, Laws MB, Mayo SJ, Zuckerman B, Abreu M, Medina L, et al. Errors in medical interpretation and their potential clinical consequences in pediatric encounters. Pediatrics. 2003;111:6–14.
18 Flores G, Rabke-Verani J, Pine W, Sabharwal A. The importance of cultural and linguistic Issues in the emergency care of children Pediatr Emerg Care. August 2002;18:271–84.
19 Flores G, Abreu M, Olivar MA, Kastner B. Access barriers to health care for Latino children. Arch Pediatr Adolesc Med. 1998;152:1119–25.
20 Rogers A, Delgado C, Simon H. The effect of limited English proficiency on admission rates from a pediatric ED: stratification by triage acuity. Am J Emerg Med. 2004;22:534–36.
21 Hampers LC, McNulty JE. Professional interpreters and bilingual physicians in a pediatric emergency department: effect on resource utilization. Arch Pediatr Adolesc Med. 2002;156:1108–13.
22 Baker DW, Hayes R, Fortier JP. Interpreter use and satisfaction with interpersonal aspects of care for Spanish-speaking patients. Med Care. 1998;36:1461–70.
23 Burbano O'Leary SC, Federico S, Hampers LC. The truth about language barriers: one residency program's experience. Pediatrics. 2003;111 (5 pt 1):t–73.
25 Barkin S, Balkrishnan R, Manuel J, Hall MA. Effect of language immersion on communication with Latino patients. N C Med J. 2003;64:258–62.
26 Mazor SS, Hampers LC, Chande VT, Krug SE. Teaching Spanish to pediatric emergency physicians: effects on patient satisfaction. Arch Pediatr Adolesc Med. 2002;156:693–95.
27 Weinick RM, Jacobs EA, Stone LC, Ortega AN, Burstin H. Hispanic healthcare disparities: challenging the myth of a monolithic Hispanic population. Med Care. 2004;42:313–20.
28 Flores G, Abreu M, Schwartz I, Hill M. The importance of language and culture in pediatric care: case studies from the Latino community. J Pediatr. 2000;137:842–48.
For the following questions “Spanish-speaking only” refers to patients who are unable to communicate effectively in English. The medical visit will need to be conducted in Spanish.