Received from the Division of General Medicine and Primary Care (QNM), University of California Irvine College of Medicine, Irvine, Calif; and the Division of General Medicine and Primary Care, Beth Israel Deaconess Medical Center, (ATRL, RSP), Harvard Medical School, Boston, Mass.
Asian Americans’ Reports of Their Health Care Experiences
Results of a National Survey
Article first published online: 27 FEB 2004
Journal of General Internal Medicine
Volume 19, Issue 2, pages 111–119, February 2004
How to Cite
Ngo-Metzger, Q., Legedza, A. T. R. and Phillips, R. S. (2004), Asian Americans’ Reports of Their Health Care Experiences. Journal of General Internal Medicine, 19: 111–119. doi: 10.1111/j.1525-1497.2004.30143.x
Portions of this work were presented in part at the 13th Annual Meeting of the American Public Health Association, Philadelphia, Pa, which took place November 9–13, 2002.
- Issue published online: 27 FEB 2004
- Article first published online: 27 FEB 2004
- Asian Americans;
- health care;
- ethnic groups;
- quality of care
OBJECTIVE: To examine how Asian race/ethnicity affects patients’ health care experiences and satisfaction with care.
DESIGN: Telephone interview using random-digit dialing, stratified to over-sample adults living in areas with disproportionately large numbers of minorities.
PARTICIPANTS AND SETTING: White (N = 3,205) and Asian-American (N = 521) respondents, weighted to represent all such adults living in the continental U.S. in telephone households.
MEASUREMENTS: Reports of health care experiences and trust in the doctor at the last visit, and overall satisfaction with care and desire to change doctors in the last 2 years.
MAIN RESULTS: Asian Americans were less likely than whites to report that their doctors ever talked to them about lifestyle or mental health issues (P ≤ .01). They were more likely to report that their regular doctors did not understand their background and values (P ≤ .01). When asked about the last visit, they were more likely to report that their doctors did not listen, spend as much time, or involve them in decisions about care as much as they wanted (all P ≤ .0001). In multivariable analyses, Asian Americans were less likely than whites to report that they were very satisfied with care (odds ratio [OR], 0.64, 95% confidence interval [CI], 0.42 to 0.99). However, they were not significantly less likely than whites to trust their doctors (OR, 0.79, 95% CI, 0.52 to 1.20), or to change doctors (OR, 0.93, 95% CI, 0.56 to 1.56).
CONCLUSIONS: In a national survey, Asian Americans were less likely to receive counseling and less likely to report positive interactions with their doctors than white respondents. More research is needed to determine the reasons for these differences.
Asian Americans are one of the fastest growing ethnic groups, with an estimated 12 million living in the United States.1 Despite this rapid growth, they remain one of the most under-studied groups, with relatively few studies of their health status and service use compared to other minority groups.
Asian Americans are culturally and economically diverse, coming from 30 distinct ethnic groups, each with their own traditions, cultures, and languages.2 Statistical reports that aggregate Asian-American subgroups conceal significant disparities by specific cultures and origins.3 The degree of language proficiency and acculturation determines socioeconomic status. For example, third-generation Japanese Americans tend to have higher socioeconomic indices than recent immigrants from Southeast Asia.4 Vietnamese Americans have an average family income that is half that of other Asian Americans, with 30% living below the poverty level.4,5 Among Asian-American subgroups, Vietnamese, Korean, and Chinese Americans are more likely to be uninsured than white Americans.6
Some Asian-American subgroups have significant health problems. Tuberculosis incidence among new immigrants is five-fold that of the total population.7 Lung cancer is the leading cause of death for Asian-American men.8 Vietnamese-American women have 5 times the rate of cervical cancer as white women, yet only half of Vietnamese women aged 18 and older report ever having a Pap smear.8,9
Though substantial research has been conducted regarding patient-centered care, little work has included Asians. Previous research conducted among patients at a few health centers or in a specific health plan suggested that Asian Americans are more likely than whites to report problems with care and are more likely to be dissatisfied with the care they received.10–12 To our knowledge, the Commonwealth Fund 2001 Health Care Quality Survey is the first survey conducted among a nationally representative sample of Asian Americans in the general population. In the spring of 2002, the Commonwealth Fund released preliminary results from this survey.13 The report showed that Asian Americans were less likely than white respondents to report being “very satisfied” with their medical care (45% vs 65%). Whether this difference in satisfaction can be explained by differences in health care experiences or by some other factors (such as patient demographics or health status) is unclear. Determining that differences in satisfaction are related primarily to differences in health care experiences rather than patient characteristics could prompt improvement in specific processes of care provided to Asian Americans. However, how patients report their experiences and rate their care reflects their perceptions and expectations, as well as the actual quality of care. We analyzed these survey data to determine whether differences in satisfaction between Asian Americans and whites are due to differences in problems reported in the clinical encounter or due to other patient-specific characteristics such as patient race/ethnicity, income, or education. We also assessed how respondents’ perceptions of the doctor-patient interaction affect trust and the desire to change physicians.
Study Design and Data Collection
The 2001 Health Care Quality Survey, sponsored by The Commonwealth Fund, consisted of 25-minute telephone interviews with a nationally representative sample of adults aged 18 and older living in the continental United States. The survey oversampled adults living in telephone areas with disproportionately large numbers of African Americans, Latinos, and Asian Americans. The study used a stratified minority sample design using random-digit dialing (RDD) methods. Telephone numbers were drawn disproportionately from area code-exchange combinations with higher than average densities of minority households. Results were weighted to correct for this disproportionate sampling. Details of the survey methodology have been published elsewhere.13 In brief, 6,722 adults aged 18 and older were interviewed by telephone between April 30 and November 5, 2001. Up to 20 attempts were made to contact a person at every sampled phone number, and calls were staggered over times of day and days of the week to optimize the chance of making contact. Respondents were interviewed using a questionnaire developed by the Commonwealth Fund and Princeton Survey Research Associates. The questionnaire had 96 items and queried respondents about access to care and their health care experiences in the last 2 years. It was pretested with a small number of respondents from an RDD sample, and was revised and translated into Spanish, Korean, Mandarin, Cantonese, and Vietnamese.
The national random-digit dialing survey yielded interviews with 3,488 non-Hispanic whites and 621 non-Hispanic Asian Americans (for brevity, referred to as whites and Asian Americans throughout the rest of this paper). Seventy-two percent of those contacted by phone for interviews agreed to participate. Counting eligible adults who were not reached by phone, despite 20 call attempts, the overall response rate was 53%. In this study, we excluded respondents of other race/ethnicities, and excluded Native Hawaiians and Pacific Islanders because of their small number in the sample (N = 48). We excluded respondents who did not have a health care experience (at a doctor's office or medical clinic, hospital, or emergency department) in the last 2 years. Our final study sample consisted of 3,205 white and 521 Asian-American respondents, weighted appropriately to represent the 193 million adults aged 18 and older living in the continental United States in telephone households.14
Respondents’ Healthcare Experiences
Respondents were asked about their health care experiences at the last visit. Respondents reported whether the doctor at the last visit: 1) spent enough time with them; 2) treated them with respect; and 3) involved them in decisions about care. Respondents were asked how much they understood of what the doctor said to them, and whether they had questions about their care or treatment that they wanted to discuss but did not. Respondents were also asked whether a doctor had ever discussed with them lifestyle health issues (smoking, diet/nutrition, or exercise) or mental health issues (stress or depression). Respondents reported whether they felt their regular (main) doctors understood their background and values, and whether they have ever been judged unfairly because of the type of insurance they have, their English-language skills, their race/ethnicity, or gender. They were also asked whether a friend or family member has been treated unfairly when seeking medical care because of race or ethnic background.
In response to questions such as “how much time did the doctor spend” and “did the doctor involve you in making decisions,” response categories were: as much as wanted, almost as much, less than wanted, and a lot less than wanted. We considered ideal care to be reflected when respondents answered “as much as I wanted.” In response to questions such as “how much respect did the doctor show,” response categories were: a great deal, a fair amount, not too much, none at all. We considered ideal care to be reflected when respondents answered “a great deal.”
Satisfaction with Care, Trust in Doctor, and Changing Doctor
Our main outcomes included: 1) satisfaction with care received in the last 2 years; 2) how much trust the patient had in the doctor seen at the last visit; and 3) whether patients changed doctors or wanted to change doctors in the last 2 years because they were dissatisfied. Satisfaction was determined by a question that asked: “Overall, how satisfied or dissatisfied are you with the quality of health care you have received during the last 2 years?” Response categories were: very satisfied, somewhat satisfied, somewhat dissatisfied, and very dissatisfied. We dichotomized satisfaction into “very satisfied” versus “somewhat satisfied, somewhat dissatisfied, or very dissatisfied.” Trust was determined by a question that asked: “How much confidence and trust did you have in the doctor treating you?” Response categories were: a great deal, a fair amount, not too much, not at all. We dichotomized trust into “a great deal” versus “a fair amount, not too much, or not at all.” We chose to dichotomize satisfaction and trust in this manner because we wanted a priori to examine variations in health care experiences among those with the best ratings of care compared to all others, as has been done in previous literature.12,15–17 Changing doctor was determined by answering “yes” to either of 2 questions: 1) “In the last 2 years, have you changed doctors because you were dissatisfied with the doctor you were seeing?” or 2) “Have you been dissatisfied with a doctor but could not change to a new doctor?
In addition to race/ethnicity (non-Hispanic white or Asian), we obtained other demographic information including respondents’: 1) place of birth (U.S. vs foreign-born); 2) length of time living in the United States; 3) Asian ethnic subgroup; and 4) primary language at home. We classified respondents’ age as < 65 years or ≥ 65 years. We classified marital status as being married or not. Total household income was classified as ≥ $50,000, < $50,000, or undesignated. Education level was classified as non–high school graduate, high school graduate, technical school, or college/postgraduate schooling. Respondents’ self-reported health status was classified as “excellent, very good, or good” versus “fair or poor.”
We also obtained information on access to medical care, including whether respondents had: 1) insurance in last 12 months (insured continuously, insured now but not in the last 12 months, or uninsured); and 2) choice in place of care (“great deal of choice or some choice” vs “very little choice or no choice”). We also obtained information on whether the respondent had a regular medical provider, and whether the respondent had racial/ethnic concordance with his regular provider.
We conducted bivariable analyses to analyze any differences between white and Asian Americans in: (1) demographic characteristics; (2) self-reported health status; and (3) access characteristics (having continuous insurance in last 12 months and having a regular medical provider). We also examined the health care experiences received at the last visit. We used the Cochran-Mantel-Haenszel test for all categorical variables. We classified the independent variables into 4 categories of predictors: 1) race/ethnicity; 2) other demographic variables (age, gender, place of birth, length of time in US, education, income) and health status; 3) access variables (having continuous insurance in last 12 months and having a regular medical provider); and 4) health care experiences received at the last visit. We then used logistic regression to examine which factors were associated with being “very satisfied” with medical care. We conducted multivariable analyses, adding independent variables to the models in groups, according to the 4 categories as described above. First, we examined the effect of race/ethnicity alone on satisfaction with medical care in the last 2 years. Second, we added other demographic and health status variables to the model. Third, we added access variables (having insurance and regular provider) to the model. Fourth, we added patients’ health care experiences. In our final model, we retained race/ethnicity as the main variable of interest, other variables that were determined a priori to be important to retain as possible confounders (age, gender, income, health status, access variables), and all variables that were significantly associated with satisfaction (P ≤ .05).
In a similar manner, we conducted multivariable modeling for the other 2 outcomes: 1) having trust in doctor (“a great deal” vs “a fair amount, not too much, or not at all”); and 2) having changed or wanted to change doctor because of dissatisfaction in the last 2 years. We based our modeling technique on the conceptual framework for quality assessment developed by Dr Donabedian and modified by others.16,18–20 All bivariable and multivariable analyses were conducted using SAS-callable SUDAAN version 8.0.2 (Research Triangle Institute, Research Triangle Park, NC) software with the STRWR (stratified with replacement) design option and weights computed by Princeton Survey Research Associates to account for the complex sampling design.21,22 We present weighted results as appropriate to the sampling design.
Compared to white respondents, Asian-American respondents were significantly more likely to be foreign-born (P ≤ .0001, Table 1). However, 90% of Asian Americans spoke English as the primary language at home. Asian Americans were younger than white respondents (P ≤ .0001) and were more likely to have an annual household income ≥$50,000 (P ≤ .02). Asian Americans were also less likely to report their health as fair or poor (P ≤ .05), and less likely to have a regular doctor (P ≤ .02).
|Characteristics||White N = 3,205||Asian American N = 521|
|Do not know/refused||2.49|
|18 to 29||16.75||22.19|
|30 to 39||18.58||28.10|
|40 to 49||21.74||21.10|
|50 to 64||22.19||16.92|
|65 and older||19.69||9.29|
|Undesignated or missing age||1.04||2.40|
|Born in the United States*||95.98||17.05|
|Speak English as primary language at home*||99.99||90.08|
|Completed high school||88.71||90.61|
|Self-reported health fair or poor‡||15.20||11.73|
|Have insurance last 12 months||83.38||84.64|
|Have regular medical provider†||84.07||72.43|
When we examined respondents’ health care experiences, we found that Asian Americans were less likely to report that their regular doctor is of similar race/ethnicity as them (P ≤ .0001), and less likely to report having a great deal of choice in where they go for medical care (P ≤ .05), as compared to white respondents (Table 2). They were less likely than white respondents to report that a doctor has ever talked to them about lifestyle issues (smoking/nutrition/exercise, P ≤ .01) or mental health issues (P ≤ .01). They were less likely to report that their regular doctor understands their background and values (P ≤ .01). They were more likely to report that the doctor at the last visit did not spend as much time with them (P ≤ .0001), listen to them (P ≤ .0001), or involve them in decisions about care as much as they wanted (P ≤ .0001).
|Experiences with regular (main) provider|
|Doctor-patient race discordant||29.48||69.75†|
|Doctor ever spoke to patient about lifestyle issues (smoking/nutrition/exercise)||||70.20||58.61‡|
|Doctor ever spoke to patient about mental health issues||||21.32||11.78‡|
|Doctor understood patient's background and values (strongly agree vs somewhat agree, somewhat disagree, and strongly disagree)||61.77||49.62‡|
|Experiences at the last visit|
|Doctor listened to everything patient had to say (vs most, some, or only a little)||69.07||47.13†|
|Doctor spent as much time as patient wanted¶||72.54||50.45†|
|Doctor involved patient in decisions about care as much as patient wanted¶||79.69||59.40†|
|Doctor treated patient with a great deal of respect (vs a fair amount, not too much, or none at all)||76.98||59.36†|
|Had questions about care patient wanted to ask but did not||||10.19||14.51|
|Patient had a great deal of choice in place of care (vs some, very little, or no choice)||85.16||78.13§|
|Doctor or medical staff judged patient unfairly because of patient's insurance status, language, race, or gender||||8.74||11.20|
|Over past 2 years, family or friend has been treated unfairly when seeking medical care because of race or ethnicity||||2.62||7.81§|
When asked about how satisfied they were with their health care overall in the last 2 years, Asian Americans and whites differed significantly in their patterns of response (P ≤ .0001, Table 3). Asian Americans were less likely than white respondents to say “very satisfied.” Asian Americans were also less likely than whites to report having “a great deal” of trust in the doctor at the last visit. Asian Americans were not significantly different from white respondents in having changed doctors or wanting to change doctors in the last 2 years because of dissatisfaction (Table 3).
|Non-Hispanic Whites N = 3,205||Asian Americans N = 521|
|Satisfaction with care received in the last 2 years*|
|Do not know/refused||0.83||2.19|
|Amount of confidence and trust in the doctor at the last visit*|
|A great deal of trust||71.61||55.04|
|A fair amount||23.92||37.81|
|Not too much||2.95||6.59|
|None at all||0.88||0.04|
|Do not know/refused||0.63||0.52|
|Changed doctor or wanted to change doctors in the last 2 years because of dissatisfaction†|
Table 4 shows the unadjusted and adjusted odds ratios for satisfaction, trust, and changing doctors. After adjustment for other demographic, health status, access factors, and health care experiences, Asian Americans were still less likely than white respondents to report that they were very satisfied with care (adjusted odds ratio [AOR], 0.64; 95% confidence interval [CI], 0.42 to 0.99). However, they were not significantly different than white patients in having a great deal of trust in their doctors (AOR, 0.79; 95% CI, 0.52 to 1.20), or in changing or wanting to change doctors (AOR, 0.93; 95% CI, 0.56 to 1.56). Respondents who had fair or poor health, racial/ethnic discordance with their providers, or little choice in their place of care were more likely to be dissatisfied and more likely to change doctors (all P ≤ .05, Table 4). Respondents who perceived that their doctors did not understand their background and values or did not listen to everything they had to say were less likely to be very satisfied with care, less likely to have a great deal of trust, and more likely to change doctor or want to change doctor (all P ≤ .05, Table 4). Respondents who perceived that their doctors (or other medical staff) judged them unfairly because of the type of insurance they had, their language proficiency, or their race or gender, were more likely to have changed doctors or wanted to change doctors (AOR, 3.28; 95% CI, 2.15 to 5.02).
|Model 1 Satisfaction OR (95% CI)||Model 2 Trust OR (95% CI)||Model 3 Changing Doctors OR (95% CI)|
|Asian race/ethnicity (unadjusted)||0.41 (0.31 to 0.55)§||0.48 (0.36 to 0.64)§||1.29 (0.88 to 1.90)|
|Final Model 1*||Final Model 2*||Final Model 3*|
|AOR (95% CI)‡||AOR (95% CI)‡||AOR (95% CI)‡|
|Demographic and Access Characteristics|
|Asian race/ethnicity||0.64 (0.42 to 0.99)§||0.79 (0.52 to 1.20)||0.93 (0.56 to 1.56)|
|Female||1.17 (0.90 to 1.52)||1.33 (1.01 to 1.75)§||1.90 (1.37 to 2.62)§|
|Income > $50,000||1.37 (1.02 to 1.85)§||0.81 (0.59 to 1.12)||1.16 (0.80 to 1.66)|
|Not have a regular provider||1.23 (0.79 to 1.91)||0.86 (0.59 to 1.25)||0.53 (0.33 to 0.87)§|
|Health status fair or poor||0.54 (0.37 to 0.80)§||0.74 (0.51 to 1.07)||1.70 (1.15 to 2.52)§|
|Doctor–patient race discordant†||0.61 (0.43 to 0.86)§||–||1.55 (1.04 to 2.29)§|
|Did not have choice in place of care†||0.50 (0.35 to 0.72)§||–||1.85 (1.28 to 2.69)§|
|Doctor did not spend as much time as patient wanted†||0.38 (0.28 to 0.51)§||–||–|
|Had questions about care patient wanted to ask doctor but did not||0.52 (0.33 to 0.80)§||0.32 (0.21 to 0.49)§||2.59 (1.76 to 3.80)§|
|Doctor spoke to patient about lifestyle issues (smoking/nutrition/exercise)†||–||1.40 (1.05 to 1.87)§||–|
|Doctor spoke to patient about mental health issues†||–||–||1.45 (1.04 to 2.03)§|
|Doctor did not listen to everything patient had to say||0.64 (0.48 to 0.84)§||0.53 (0.39 to 0.71)§||1.64 (1.18 to 2.29)§|
|Doctor did not understand patient's background and values||0.36 (0.28 to 0.47)§||0.32 (0.25 to 0.43)§||1.74 (1.25 to 2.41)§|
|Doctor did not involve patient in decisions about care as much as patient wanted†||0.66 (0.48 to 0.92)§||0.60 (0.43 to 0.84)§||–|
|Doctor did not treat patient with a great deal of respect†||0.45 (0.33 to 0.62)§||0.16 (0.12 to 0.22)§||–|
|Doctor or medical staff judged patient unfairly because of patient's insurance status, language, race, or gender†||0.45 (0.30 to 0.70)§||–||3.28 (2.15 to 5.02)§|
|Over past 2 years, family or friend has been treated unfairly when seeking medical care because of race or ethnicity†||–||0.40 (0.19 to 0.82)§||–|
We found that Asian Americans were less likely than white respondents to receive counseling from their doctors about lifestyle or mental health issues. They were more likely to report that their doctors did not understand their background and values. When asked about the last visit, they were more likely to report that the doctor did not spend as much time with them or involve them in decisions about care as much as they wanted, and did not treat them with a great deal of respect. Asian Americans were less likely to be very satisfied with care, and less likely to have a great deal of trust in the doctor. However, they were not more likely than white respondents to change doctors because of dissatisfaction.
Our findings are consistent with previous research. In a study done in 1997, Taira et al. found that Asian Americans at a general medical group at one academic health center rated overall satisfaction and every dimension of primary care lower than white patients.11 In the Medical Outcomes Study, Meredith and Siu reported that Asian Americans across different practice sites in 3 U.S. cities reported lower satisfaction with care and perceived less sharing in the doctor-patient relationship, compared to other ethnic groups.10 Murray-Garcia et al. found that Asian Americans in a large Health Maintenance Organization (HMO) consistently rated their physicians lower than white patients in every dimension of performance.12 They also found, as we did, that Asian Americans reported receiving significantly less counseling than white patients on health habits such as smoking, nutrition, or exercise.12
The reasons why Asian Americans receive less counseling compared to white patients are unclear. In March 2002, the Institute of Medicine (IOM) released the report: Unequal Treatment: Confronting Racial and Ethnic Disparities in Healthcare.5 The report suggested that racial/ethnic disparities in care exist across a wide range of disease areas, clinical services, and clinical settings. The IOM Committee posed the question: “How could well-meaning and highly educated health professionals, working… with diverse populations of patients, create a pattern of care that appears to be discriminatory?”5 The Committee suggested that physicians, like everyone else, might use “stereotyping” as a cognitive short cut. Stereotyping is defined as the process by which people use social categories (e.g., race, gender) in acquiring, processing, and recalling information about others.5 The Committee stated that stereotyping may be “activated in situations characterized by time-pressure, resource constraints, and high cognitive demands,” such as the clinical encounter.5 In our study, “stereotyping” may be one explanation for the decreased counseling regarding lifestyle and mental health issues to Asian Americans by providers. Asian Americans have been previously characterized as the “Model Minority,” with few health problems or negative health habits.5,23 Health care providers may not be aware of high smoking rates and lung cancer mortality among subgroups of Asian Americans.3 They may not know about the high prevalence of cardiovascular disease among Japanese and Chinese Americans;23 or the high rates of depression and post-traumatic stress syndrome among Southeast Asians.24,25 Providers may not be aware of the bimodal distribution of socioeconomic and health status among many Asian-American subgroups. In order to eliminate racial/ethnic disparities in care, clinicians must be better educated regarding cross-cultural care.5
In our study, we found that Asian Americans report significantly different care experiences than white respondents. They were more likely to report that the doctor at the last visit did not spend as much time with them as they would like and did not treat them with a great deal of respect. Murray-Garcia et al. had similar findings in their study of HMO patients.12 They found that Asian Americans valued physicians spending sufficient time and showing appropriate courtesy and respect significantly more than white patients valued these two aspects of care. In our previous research among Chinese- and Vietnamese-American patients, we found that patients preferred doctors who did not behave in a rushed or hurried manner.26 Patients perceived those doctors who “take their time” to be more thorough and complete in their evaluations and diagnoses than other doctors. “Taking time” was also seen as a sign of respect for the patient, allowing them to have enough time to ask questions. Physicians under heavy time constraints may need to be aware that patients who perceive that their doctors do not listen to them or understand them are more likely to change doctors. Health care systems must strive to allow sufficient time for doctors and patients to build communication and establish trust. This is especially important when providers and patients have different cultural or ethnic backgrounds.
Asian Americans were less likely to report that their doctors involved them in decision-making as much as they would like. Cultural differences in communication style may explain this finding. For example, Asians often nod and smile to show respect for the doctor, whom they regard as the “authority.”27 Doctors may misunderstand the gestures as a sign of agreement, when in fact the patient is simply showing respect. The patient may not agree with the treatment plan, but may not feel that it is appropriate to openly disagree with “authority.”28 Because the Asian style of communication is more indirect, the doctor may need to specifically ask the patient if there are any problems with the treatment plan. When asked in this manner, the patient may be more likely to acknowledge problems and suggest an alternative plan.
Our study has several limitations. Although the study was designed to generalize to the U.S. population aged 18 and older and to target minority households, respondents who did not have a working telephone number were excluded from this study. Although respondents had the option of responding to the survey in another language (Vietnamese, Korean, or Cantonese or Mandarin Chinese), very few chose to answer in a language other than English (99.6% of respondents overall answered in English; by race, 99% of whites and 91% of Asians answered in English). This finding suggests that limited-English proficient persons may have elected not to participate in the study. Random-digit dialing may not be the optimal method to survey these respondents, and these results cannot be generalized to limited-English proficient Asian Americans. We also had too few numbers of respondents in each group to analyze health care experiences by Asian subgroups. We did not have direct information about respondents’ acculturation levels, but used information such as place of birth and length of time in the US as proxies for acculturation. We also did not have detailed information on other aspects of care, such as respondents’ area of residence (rural vs urban), their insurance plan (fee for service vs managed care, etc.), or on their doctors’ specialties and practice types (solo, group practice, etc.). Our outcome measures, satisfaction and trust, were each measured by a single question; and changing doctor was measured by 2 questions (changed doctor or wanted to change because of dissatisfaction in last 2 years). Using multi-item scales to measure satisfaction or trust may have given more sensitive results. There may be other health care experiences and culturally mediated expectations, not measured in this survey, that may contribute to the differences in care found between Asian-American and white respondents. Finally, this study was a cross-sectional study and can only suggest associations, not causal relationships.
Despite these limitations, our study is unique in several ways. To our knowledge, it is the first study to assess the health care experiences of a large, nationally representative sample of Asian Americans. It is also the first study to assess how Asian-American respondents’ perceptions of the doctor–patient interaction affect overall satisfaction, trust, and desire to change physicians. Our findings suggest that Asian-American patients are similar to white patients in many important ways. Previous research has shown that problem experiences in the doctor-patient encounter may lead to dissatisfaction, loss of trust, and changing physicians.16 In our study, respondents who perceived that their doctors did not understand their background and values or did not listen to them were more likely to be dissatisfied, less likely to trust their doctors, and more likely to change doctors. Patients who had questions they wanted to ask the doctor but did not ask (because of lack of time, lack of trust, or other constraints), were also less likely to be very satisfied, less likely to have a great deal of trust, and more likely to change doctors. These associations exist independently of the respondents’ racial/ethnic background. After adjustment for respondents’ perceptions of their health care experiences, Asian Americans were not more likely to distrust their providers, nor were they more likely to change providers than white Americans.
In our study, Asian Americans were more likely than white respondents to report being “somewhat satisfied” with care, as opposed to report being “very satisfied.” We also found that Asian Americans were more likely to report having “a fair amount” of trust in their doctors, as opposed to having “a great deal” of trust. They were less likely than white patients to rate degree of satisfaction or amount of trust in the most positive categories. They were not more likely than whites to choose the most negative categories, but were more likely to choose responses in the middle of the scale. These findings warrant further study. Previous research has suggested that systematic differences may exist in the way different racial/ethnic groups respond to surveys. Blacks and Latinos have been reported to be more likely than whites to choose both extremes when answering Likert response scales.29–32 Asian Americans have been reported, anecdotally, to be more likely to choose responses in the middle of such scales.12 In their study, Taira et al. reported that the ratings of Asian Americans peaked at the center of distribution more than those of whites, with fewer Asian Americans giving ratings of excellent or very good.11 Further research is needed to determine whether systematic differences exist in the way Asian Americans respond to Likert rating scales, as compared to respondents of other race/ethnicities.
Quality of care is measured by a variety of criteria and outcomes. Patient-centered care is one measure of quality, and it involves more than simply “patient satisfaction.” Previous studies suggest that measuring specific aspects of patients’ care experiences may be more useful in quality improvement efforts than measuring overall ratings.33,34 By examining specific processes of care (access, information/education, continuity of care, etc.) and distinct aspects of the patient-provider encounter (knowledge, respect, trust, etc.), we may be able to identify specific areas for improvement. In their study, Taira et al. gave an illustration of how “risk-adjusting” for patients’ Asian ethnicity could significantly improve the quality ratings of providers who have a large percentage of Asian-American patients.11 Before we “risk-adjust” Asian ethnicity on patient ratings, more research must be done. We need to determine whether patient reports about care (i.e., what happened during a visit) may be more reliable and valid than patient ratings of care, especially when comparing quality of care across ethnic groups. This could be evaluated by conducting surveys that include both reports and ratings in the same content areas, and by identifying systematic variations in the responses of patients from diverse racial/ethnic backgrounds. Observing and behavior coding direct patient-doctor interactions is another way to validate patient reports and ratings of care.
As the US population increases in diversity, it is more important than ever for clinicians to be able to provide quality care to all patients, regardless of their backgrounds. Cross-cultural education should be a part of all medical and nursing curricula. However, cross-cultural education can be more detrimental than helpful if it is not done properly.35 Cross-cultural education should focus on the basic tenets of patient-oriented care and on community assessment. For example, clinicians need to learn about individual patients’ personal and social context (new immigrant or second-generation), health habits, and health beliefs and practices. Disease incidence, prevalence, and ethnopharmacology among distinct populations should be part of the curriculum. However, to prevent stereotyping and oversimplification of ethnic groups, intragroup variability must be emphasized. Factors such as patients’ religion, acculturation, or socioeconomic status may lead to much diversity within ethnic groups. Ultimately, quality medical care involves knowledge of individual patients, including knowledge of their social and cultural contexts. In order to eliminate racial/ethnic disparities in care, more research is needed to better understand how culture, language, and socioeconomic factors affect the provider-patient encounter, and how diverse patients experience, perceive, and evaluate their care.
The authors thank Drs Anne Beal, Dora Hughes, Alice Chen, and Michael Massagli for their review of previous versions of the manuscript. The Commonwealth Fund provided support for this research. The views presented are those of the authors and are not necessarily the views of the funding agency.
- 1The Asian Population: 2002. Washington, DC: US Census Bureau; 2000: C2KBR/01–16., .
- 2Cancer registries and data for Asian Americans and Native Hawaiians and Pacific Islanders: what registrars need to know. J Registry Manage. 1999;26: 128–41 ., , , , , .
- 4Asian/Pacific Islander Consortium. Asian and Pacific Islander Americans: A Profile of Socioeconomic Characteristics—1990. San Francisco; 1992.
- 5Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Institute of Medicine Committee on Understanding and Eliminating Racial and Ethnic Disparities in Health Care. Washington, DC: National Academy Press; 2002., , , eds.
- 6The Henry J. Kaiser Family Foundation. Health Insurance Coverage in America – 1999 Data Update. Washington, DC; 2000.
- 7National Center for Health Care Statistics. Healthy People 2000 Review, 1997. Hyattsville, Md: Public Health Service; 1997.
- 8et al, eds. Racial/Ethnic Patterns of Cancer in the United States 1988–92, National Cancer Institute. NIH Pub. No. 96–4104. Bethesda, Md: NIH; 1996., , ,
- 13Diverse Communities Common Concerns: Assessing the Health Quality for Minority Americans. New York: The Commonwealth Fund; 2002., , , , , .
- 14US Bureau of Census for the Bureau of Labor Statistics. Current population survey. Bureau of Labor Statistics. Available at: http://www.bls.gov/cps/home.htm. Accessed September 2, 2003.
- 18The Definition of Quality and Approaches to its Assessment, vol. 1. Ann Arbor, Mich: Health Administration Press; 1980..
- 19Evaluating the quality of care. In: AndersenRM, RiceTH, KominskiGF, eds. Changing the US Healthcare System, 2nd edn. San Francisco: Jossey-Bass; 2001..
- 21SAS. Statistical Software.[computer program], Version 8.1 for Windows. Cary, NC; 1999.
- 22SUDAAN. Software for the Statistical Analysis of Correlated Data. [computer program], Version 7.5.6 for Windows. Research Triangle Park, NC; 2002.
- 27Issues affecting Asian American and Pacific American women. In: Hassey-Dow, ed. Contemporary Issues in Breast Cancer. Boston: Jones and Bartlett; 1996..