Green, Alexander R. MD, MPH; Betancourt, Joseph R. MD, MPH; Park, Elyse R. PhD; Greer, Joseph A. PhD; Donahue, Elizabeth J.; Weissman, Joel S. PhD
Cultural competence is the ability to provide high-quality, effective health care to patients from diverse sociocultural backgrounds.1 Concern about cultural competence in health care has increased in recent years as providers and policy makers strive to eliminate the striking racial, ethnic, linguistic, and socioeconomic disparities that are now well documented.2 Unexplored or misunderstood sociocultural differences between patients and providers can lead to patient dissatisfaction, poor adherence to treatment plans, and poor health outcomes.3,4 Thus, cultural competency among physicians is considered an important step toward the elimination of racial and ethnic disparities in health care and improving the quality of medical care for all patients.2,5,6
Graduate medical education plays a crucial role in preparing physicians to function effectively in a changing and challenging clinical environment. To ensure the availability of high-quality primary health care for all Americans, the Health Resources and Services Administration (HRSA) of the U.S. Department of Health and Human Services has supported training a generalist workforce since 1972 through its Training in Primary Care Medicine and Dentistry (Title VII TPCMD) grant programs. These training grants were awarded first for residency training in family medicine, then later in internal medicine and pediatrics. In 2005, out of a total Title VII, Section 747 budget of $88.8 million that supported grants in six different primary care training categories, HRSA provided $17.2 to 109 family medicine, internal medicine, and pediatric residency programs around the country. In 2006, the total federal budget for the Title VII TPCMD grant program was cut to $41.8 million, which was insufficient to cover the costs of any new residency training grant awards that year.
The HRSA Bureau of Health Professions has managed the Title VII, Section 747 grant programs, one goal of which is to “enable culturally competent health care through improved racial and ethnic diversity and cultural competence in the health professions workforce.” Primary care residency programs apply for Title VII funding and are awarded support based on factors such as recruiting racially and ethnically diverse groups of residents, having track records of graduates practicing primary care—particularly in underserved communities—and implementing innovative teaching methods, including cultural competence training. It is not known, however, whether residents who train in HRSA Title VII-funded programs in fact become more culturally competent physicians.
In 2004, we conducted a national survey of resident physicians in their final year of training, assessing their self-reported preparedness to provide health care to culturally diverse patient populations and the amount of formal training received in cross-cultural care during their residencies.7 This survey showed that many senior residents (19%–25%) felt unprepared to provide specific components of cross-cultural care such as caring for new immigrants or patients with different health beliefs or religious beliefs that affect treatment. Despite 96% of residents responding that it is important to address cultural issues when providing care, between one third and one half reported receiving little or no instruction in specific aspects of culturally competent care during residency. Because of HRSA's emphasis on cultural competence in its primary care training grants, we reanalyzed our data to determine whether residents in Title VII-funded primary care programs reported greater cross-cultural preparedness and skills than residents in other primary care programs without HRSA funding. We hypothesized that residents in HRSA Title VII-funded programs might have better experiences in residency related to training in cross-cultural care than those in non-HRSA-funded programs, and that these might explain any differences in cross-cultural preparedness and skills.
Details of the sample selection process and survey design have been reported elsewhere.7 Briefly, we selected a stratified random sample of residents in three primary care specialties (family practice, internal medicine, and pediatrics) who were scheduled to complete their training in June 2004. We limited our study population to residents in their last year of training so that they might assess the entirety of their graduate medical education experiences, and because they are in the best position to evaluate their preparedness for practice at the completion of their residency. We studied residents at academic health centers (AHCs), defined as medical schools and their closely affiliated or owned clinical facilities.
A draft of the survey instrument was developed based on literature review, focus groups with residents in each specialty,8 and comments from expert colleagues. The instrument underwent cognitive testing by the Center for Survey Research at the University of Massachusetts–Boston and was subsequently revised. The mailed survey was administered by the Center for Survey Research in the winter and spring of 2003–2004. Response-enhancement techniques included multiple mailings, telephone reminders, and monetary incentives (random prize drawings of $1,000 each, and a $20 incentive). The protocol was approved by both the Massachusetts General Hospital and University of Massachusetts–Boston institutional review boards.
To assess their self-perceived preparedness to provide cross-cultural care, we asked residents how prepared they felt to care for a series of types of patients, or pediatric patients' families (1 = “very unprepared,” 2 = “somewhat unprepared,” 3 = “somewhat prepared,” 4 = “well-prepared,” 5 = “very well prepared”). The list comprised the following: patients from cultures different from their own; with health beliefs at odds with Western medicine; with distrust of the U.S. health care system; with limited English proficiency; who are new immigrants; whose religious beliefs might affect treatment; who use alternative or complementary medicines; and who are members of racial/ethnic minorities. We dichotomized the responses, combining “very prepared” and “somewhat prepared” to indicate preparedness.
We also asked residents to assess their skill levels in performing selected tasks or services thought to be useful in treating culturally diverse patients, or pediatric patients' families (1 = not at all skillful to 5 = very skillful). These skills were determining how to address patients from different cultures; taking a social history; assessing patients' understanding of their illness; identifying mistrust; negotiating treatment plans; assessing patients' English proficiency; identifying patients' relevant cultural customs and religious beliefs; understanding decision-making roles; and working with interpreters. We dichotomized the responses (as for preparedness above), with scores of four or five indicating skillfulness.
We asked residents about their experiences during residency that we believe might have influenced their ability to provide culturally competent care. Four questions focused on the usefulness of certain residency experiences: lectures and seminars, case-based discussions, on-the-job training, and diversity of colleagues (1 = “not at all useful,” 2 = “somewhat useful,” 3 = “useful,” 4 = “very useful,” 5 = “did not have during residency”). We dichotomized the responses, combining “useful” and “very useful.” We included “did not have” with the less-useful category (“somewhat/not at all useful”). Eight questions focused on how much of a problem were various aspects of the residency experience for providing cross-cultural care. These included lack of cross-cultural training, absence of good role models, and dismissive attitudes about cross-cultural care, among others (1 = “no problem,” 2 = “small problem,” 3 = “moderate problem,” 4 = “big problem”). We dichotomized these, combining “no problem” and “small problem.”
In addition to preparedness, skills, and training, we asked residents to report how many, if any, role models they had who were good at providing cross-cultural care. We also asked what percentage of the patients they cared for were of minority backgrounds, had limited English language proficiency, and were of cultures different than their own. Other questions assessed residents' characteristics—specifically, gender, U.S. versus international medical graduate (IMG) status, and birth country (United States or other). Residents self-identified their race and ethnicity from a list of categories provided by the investigators, corresponding to designations used by the national Graduate Medical Education Census of the American Medical Association (AMA) / Association of American Medical Colleges (AAMC). We determined HRSA Title VII funding status by comparing the list of residency programs we surveyed against a list of funded residency programs provided by HRSA.
In bivariate analyses, we used the chi-square statistic to test for significant differences in residents' characteristics, residency training experiences related to cross-cultural care, and preparedness and skills to provide cross-cultural care (stratified by specialty) between residents in HRSA Title VII-funded and non-HRSA-funded programs. We used multivariable logistic regression to determine relationships between HRSA Title VII funding status and cross-cultural preparedness and skills, adjusting for resident gender, race/ethnicity, country of birth (United States or other), and specialty. We did not include IMG status as a covariate because the percentages were nearly identical in the HRSA Title VII-funded and non-HRSA-funded groups. In additional logistic regression analyses for these same preparedness and skills variables, we further adjusted for residency training characteristics that were significantly different (P < .10) between HRSA Title VII-funded and non-HRSA-funded programs, to determine whether these reduced the odds ratios (ORs) and statistical significance of the HRSA Title VII-funding variable. All analyses were performed using SPSS software, version 12.0 (SPSS Inc, Chicago, Illinois) and were weighted to account for the complex sampling design, correcting for nonresponse and for the probability of selection within stratum (physician specialty).
Of 1,467 eligible residents in primary care training programs (family medicine, internal medicine, and pediatrics), 870 returned surveys (response rate = 59%). We excluded from the analyses four residents for whom we could not determine HRSA Title VII funding status. We divided the remaining sample of 866 into two groups: residents in HRSA Title VII-funded programs (no. = 403) and residents in non-HRSA-funded programs (no. = 463).
Table 1 illustrates the characteristics of our sample by HRSA Title VII funding status. HRSA Title VII-funded residents were more likely to be Asian/Pacific Islander and born outside the United States, and they were somewhat less likely to be black. The weighted distributions of gender and race/ethnicity are nearly identical to those of all U.S. residents as reported from AMA and AAMC surveys.9 HRSA, through the Title VII, Section 747 Residency Training in Primary Care grant program, funds considerably more family medicine programs than pediatrics and internal medicine programs, which is reflected in Table 1. As described in the original study,7 we found no significant differences between respondents and nonrespondents in terms of gender, Hispanic ethnicity, IMG status, and whether or not they were born in the United States. Compared with nonrespondents, more respondents were white (based on AMA data, 64% versus 59%, P = .02), and in family medicine residencies (P < .01).
HRSA Title VII-funded residents were more likely than non-HRSA-funded residents to report having experiences in residency that were useful or very useful for treating culturally diverse patients, including case-based discussions and on-the-job training in community-based and hospital-based sites (see Table 2). They also reported that lectures and seminars were helpful more than their non-HRSA-funded peers. HRSA Title VII-funded residents were less likely than non-HRSA-funded residents to report experiencing the following problems for delivering cross-cultural care: inadequate cross-cultural training during residency, absence of good role models from cross-cultural care, and dismissive attitudes about cross-cultural care among attending physicians. HRSA Title VII-funded residents reported a higher mean number of role models good at providing cross-cultural care (4.49 versus 3.35, P < .001), but they did not take care of a significantly higher percentage of minority patients, patients with limited English proficiency, or patients who were culturally different from them.
Residents in the three different primary care specialties—family medicine, internal medicine, and pediatrics—differed substantially in their reports of preparedness to provide specific components of cross-cultural care. Because of this difference, the unadjusted analyses in Table 3 (and Table 4, described below) are stratified by specialty. For family medicine residents, those in HRSA Title VII-funded programs reported higher cross-cultural preparedness (well prepared or very well prepared) to care for racial/ethnic minority patients compared with non-HRSA-funded residents (76.3% versus 60.8%, P = .001). This was the only significant difference of the eight measures of preparedness. Internal medicine residents in HRSA Title VII-funded programs were significantly better prepared in five of the eight measures, and they showed similar trends in two other measures. For pediatric residents, those in HRSA Title VII-funded programs reported higher cross-cultural preparedness in caring for patients who use complementary/alternative medicine (31.9% versus 20.7%, P = .05), and a near-significant difference in caring for patients who are new immigrants (44.4% versus 34.6%, P = .06).
Given that the study was not powered to detect differences within specialties but that the bivariate analyses almost all favored HRSA Title VII-funded programs, we combined the data for all specialties in logistic regression analyses adjusting for resident gender, race/ethnicity, birth country, and specialty. In these analyses, residents in HRSA Title VII-funded programs were significantly more likely than those in non-HRSA-funded programs to report being prepared to provide cross-cultural care across all eight measures (OR = 1.54–2.61, all P values <.01).
Table 4 shows bivariate analyses and logistic regression analyses similar to those in Table 3, but for specific self-reported cross-cultural skills. Overall, these differences based on HRSA Title VII funding status were somewhat less prominent than the differences in residents' preparedness, with only two significant differences, all among internal medicine residents. HRSA Title VII-funded internal medicine residents were more likely than non-HRSA-funded residents to report being skilled (four or five on a five-point scale) at determining how a patient wants to be addressed (60.2% versus 47.9%, P = .05) and delivering services effectively through a medical interpreter (73.8% versus 57.5%, P = .01). In combined logistic regression analyses, residents in HRSA Title VII-funded programs were significantly more likely to report being skilled in 6 of the 10 specific cross-cultural skill measures, and they were nearly significantly more likely in two other measures. The ORs were somewhat lower than the cross-cultural preparedness domain, ranging from 1.25 to 1.95.
The impact of residency experience on preparedness
In additional logistic regression analyses, we sought to determine whether differences in the residency training experience between HRSA Title VII-funded and non-HRSA-funded programs could account for differences in resident preparedness and skills to provide cross-cultural care. The first column of Table 5 shows the unadjusted ORs for all eight preparedness measures (upper section) and for all 10 skills measures (lower section) comparing the two study groups. The second column shows these ORs after adjusting for residents' demographic characteristics and specialty. These are identical to the right column in Table 3 (preparedness) and Table 4 (skills). All of these show HRSA Title VII-funded residents reporting better preparedness than non-HRSA-funded residents, and 6 of 10 show HRSA Title VII-funded residents reporting greater skillfulness. The final column in Table 5 shows these ORs after adjusting for eight characteristics of the residency training experience that were significantly (or nearly significantly) different between the groups, as shown in Table 2.
For three of the cross-cultural preparedness measures, the statistically significant difference between HRSA Title VII-funded and non-HRSA-funded residents became small and nonsignificant after accounting for the residency experience variables. These three measures were preparedness to care for patients whose cultures are different than theirs, whose health beliefs are at odds with Western medicine, and whose religious beliefs affect their treatment. For four preparedness measures, the ORs decreased but still show HRSA Title VII-funded residents to be significantly better prepared than others. For one measure, preparedness to care for patients with limited English proficiency, the OR increased. For 4 of the 10 cross-cultural skills measures, the significant difference between HRSA Title VII-funded and non-HRSA-funded residents disappeared after accounting for the residency experience variables. These four measures were taking a social history, assessing the patients' understanding of the cause of his or her illness, negotiating with the patient about key aspects of the treatment plan, and identifying how well a patient can read or write English. For four other measures of skillfulness, the ORs decreased towards one but were not significantly different from one to start.
This study analyzed data from a national survey of the cross-cultural preparedness of residents in their last year of training and compared residents in HRSA Title VII funded programs with those in non-Title VII-funded programs. Primary care residency programs that apply for HRSA Title VII funding are assessed according to various factors, including whether they are situated in a federally designated underserved community, whether they graduate residents who go on to practice in underserved communities, and their emphasis on cultural competence training. This screening process and the financial support HRSA provides would ideally help create a training environment within Title VII-funded residency programs that would promote cultural competence and prepare residents to provide high-quality care to culturally diverse populations. Our study supports this notion.
Our findings were interesting in that, overall, residents in HRSA Title VII-funded programs were considerably more likely to report preparedness than non-HRSA-funded residents (ORs between 1.5 and 2.6 for all eight measures of preparedness). They also felt more skilled than non-HRSA-funded residents in 6 out of 10 cross-cultural skill measures (ORs between 1.5 and 1.9). For HRSA, whose mission is to “improve the nation's health by assuring equal access to comprehensive, culturally competent, quality health care for all,”10 these are encouraging results. They support the idea that these federally funded residency programs are serving an important role in training primary care physicians who are better prepared to care for a diversifying U.S. population at a time when racial and ethnic disparities in care are widespread.
Primary care specialties differed in the influence of HRSA Title VII funding status on cross-cultural preparedness. In general, residents in internal medicine programs had the greatest differences by HRSA Title VII funding status in the likelihood of feeling prepared to provide various aspects of cross-cultural care. Residents in family medicine programs had fewer and smaller differences in preparedness by funding status, despite having the greatest number of funded programs. This supports previous work suggesting that family medicine residencies may already be doing well at providing effective cross-cultural training experiences for their residents, whereas internal medicine programs may be more variable, with HRSA Title VII funding playing a more important role.11–13 Alternatively, these differences could have more to do with the characteristics of the programs that applied for funds and how funds were allocated.
We found that, compared with others, residents in HRSA Title VII-funded training programs reported that their experiences with on-the-job training and case-based discussions were more helpful for providing cross-cultural care. They also reported a higher mean number of role models good at cross-cultural care and fewer problems with dismissive attitudes among attending physicians. This should be interpreted in the context of having a similar amount of diversity among their colleagues and their patient population (the average proportion of minority patients residents cared for was very high—more than two thirds for both groups). This suggests that HRSA Title VII-funded programs may provide an environment—a local culture—that places more emphasis than other programs' environments on being responsive to the needs of its culturally diverse patient population. This “hidden curriculum” has been shown in multiple studies to have a strong impact on the attitudes and values of physicians-in-training.14–16 Residents in HRSA Title VII-funded programs were also less likely than others to cite problems with inadequate cross-cultural training, and they reported a somewhat higher incidence of more useful lectures and seminars on cross-cultural care.
On the basis of the findings shown in Tables 3 and 4, it could be argued that differences in self-reported cross-cultural preparedness and skills between the study groups are attributable to HRSA Title VII-funded residency programs attracting residents who are already better prepared to provide culturally competent care. However, Table 5 shows that most of these differences decrease, and many become statistically insignificant, after accounting for various characteristics of the residency training experience related to cross-cultural care, as described in the previous paragraph. In a related study from this same survey, we analyzed which of these characteristics were most strongly predictive of cross-cultural preparedness for residents in general—not related to HRSA Title VII funding status.11 The amount of training in cross-cultural care and the percentage of patients from diverse cultural backgrounds were the two most important predictors of cross-cultural preparedness; however, other factors such as good role models were also important. Because the diversity of the patients cared for was not significantly different between the groups in our study, it is likely that the amount of cross-cultural training (both on the job and, to a lesser extent, in lectures and seminars) and the availability of role models good at cross-cultural care account for much of the greater preparedness and skills among residents in HRSA Title VII-funded programs.
Given the emphasis that HRSA and the Title VII funding mechanism place on culturally competent care, it is not surprising that the residents in these programs feel better prepared and more skilled to provide care to culturally diverse populations. However, it is somewhat surprising that the cultural diversity of the patients the two groups care for is similar, considering HRSA's emphasis on serving the underserved. This may reflect a large number of HRSA Title VII training programs in rural areas that serve lower socioeconomic white communities offsetting the predominantly minority and immigrant patient populations in many large, urban AHC residency clinics. Future studies should sample more rural, nonacademic training programs. Another surprising finding was that residency experience characteristics did nothing to account for HRSA Title VII-funded residents' greater preparedness to care for limited-English-proficiency patients and greater skills in using medical interpreters. It may be that residents in HRSA Title VII-funded programs use interpreters more often and may engage in self-directed learning, which we did not measure.
This study has several limitations. All data are from resident self-report and may not accurately represent measures such as how prepared or skilled residents are to provide care to diverse patient populations or the actual amount of formal training received. Self-reported skills in particular may not correlate well with true skills.17 Also, because this is a cross-sectional study, we cannot make conclusions regarding a causal relationship between HRSA Title VII funding and cross-cultural training, role models, or cross-cultural preparedness for those residents in HRSA Title VII-funded programs. Although we have accounted for several important covariates in regression models, there may still be some selection bias whereby residents who decide to train in HRSA Title VII-funded programs are characteristically different than those who choose to train in other programs. However, this is unlikely to explain a large portion of the differences found. Finally, despite a reasonably high response rate, our study may be subject to some degree of nonresponse bias.
As the health care system strives to improve the quality of care for all patients while eliminating racial and ethnic disparities, training a culturally competent physician workforce has become an imperative. The U.S. government, through HRSA and its Title VII, Section 747 TPCMD grant funding mechanism for primary care residencies, has played an important role in this effort. Our study shows that residents graduating from HRSA Title VII-funded primary care training programs feel better prepared and more skilled to provide culturally competent care across a range of measures. If the U.S. health care system is to improve the cultural competence of its physician workforce and to strive to eliminate racial/ethnic disparities in care, the budget cuts to HRSA's Title VII primary care training programs should be carefully reconsidered. Without these programs, the future may witness worsening inequities in health care, not improvement.