Fax: (312) 695-0951
Low rates of colorectal, cervical, and breast cancer screening in Asian Americans compared with non-Hispanic whites
Cultural influences or access to care?
Article first published online: 23 MAY 2006
Copyright © 2006 American Cancer Society
Volume 107, Issue 1, pages 184–192, 1 July 2006
How to Cite
Kandula, N. R., Wen, M., Jacobs, E. A. and Lauderdale, D. S. (2006), Low rates of colorectal, cervical, and breast cancer screening in Asian Americans compared with non-Hispanic whites. Cancer, 107: 184–192. doi: 10.1002/cncr.21968
- Issue published online: 16 JUN 2006
- Article first published online: 23 MAY 2006
- Manuscript Accepted: 27 MAR 2006
- Manuscript Revised: 2 FEB 2006
- Manuscript Received: 15 SEP 2005
- American Cancer Society, Atlanta, Georgia. Grant Number: CPHPS-107922
- Asian Americans;
- cancer screening;
- health services accessibility
Asian Americans have lower cancer screening rates compared with non-Hispanic whites (NHWs). Little is known about mechanisms that underlie disparities in cancer screening. The objectives of the current study were 1) to determine the relation between nativity, years in the United States, language, and cancer screening in NHWs and Asian Americans, independent of access to care and 2) to determine whether Asians reported different reasons than NHWs for not obtaining cancer screening.
This population-based study included 36,660 NHWs, 1298 Chinese, 944 Filipinos, 803 Koreans, 857 Vietnamese, and 1036 Other Asians from the 2001 California Health Interview Survey. The main study outcomes were 1) self-reported colorectal, cervical, and breast cancer screening and 2) reasons for not obtaining cancer screening.
After adjusting for access to care, several Asian subgroups still had significantly lower rates of all types of cancer screening compared with NHWs. Adjusting for nativity, years in the United States, and English language attenuated the relation between Asian ethnicity and lower rates of colorectal and breast cancer screening. When they were asked what the most important reason was for not having each screening test, foreign-born Asians were significantly more likely than United States-born NHWs to report that they “didn't have problems/symptoms” (P<.01).
Nativity, years in the United States, and English language may be markers of cultural differences that are mediating cancer screening disparities. Foreign-born Asians may believe that cancer screening is in response to symptoms rather than tests that are used prior to the development of symptoms. Health education messages must consider how to communicate effectively that “cancer screening is valuable, because it finds cancer before it is advanced enough to cause symptoms.” Cancer 2006. © 2006 American Cancer Society.
Cancer disproportionately burdens Asian Americans.1, 2 Studies have documented lower rates of cancer screening in Asian Americans.2–16 However, few studies use population-based samples,11, 17–21 and those that do often aggregate diverse Asian-American subgroups because of sample size limitations.4, 5, 16, 22 To date, there are scarce data about possible determinants of cancer screening from a population-based sample of disaggregated Asian-Americans subgroups. The main objective of this study was to explore whether access to health care or cultural influences explain disparities in cancer screening in a population-based sample of Chinese, Filipinos, Koreans, Vietnamese, other Asians, and non-Hispanic whites. In this study, we used nativity, years in the United States, English language use, and differences in health beliefs about screening as possible markers of cultural influence.
Cancer screening disparities between non-Hispanic whites, Latinos, and African Americans are explained largely by socioeconomic status (SES) and access to health care.2, 4, 9, 23–26 For Asian Americans, there is limited evidence that lower cancer screening rates persist after controlling for SES and access to health care.5, 27 For Asian Americans, nativity and language may play important roles in explaining cancer screening disparities.16 Foreign birth and limited English proficiency are associated with poor health communication,28 language barriers,29 and lower rates of health insurance.30 However, these factors also may be indicators of cultural differences in health-belief models about cancer screening and prevention between foreign-born Asians and non-Hispanic whites. The health-belief model is based on the idea that an individual's beliefs about their risk and susceptibility to disease, combined with the perceived benefits of action, are crucial in their motivation to undertake specific health-related behaviors, such as cancer screening.31 Most Asians in the United States are foreign born: Many come from resource-limited countries where health care systems are less focused on chronic disease prevention and cannot offer cancer screening on a population basis.32–34 In addition to facing structural barriers to cancer screening, Asian immigrants' perceptions of cancer risk and screening benefits for asymptomatic conditions may differ from those of individuals who were born in the United States.
The first objective of this study was to determine the relation between nativity, language, and cancer screening by using a population-based sample of non-Hispanic whites, Chinese, Filipinos, Koreans, Vietnamese, and other Asian Americans. We wanted to explore the hypothesis that being foreign-born and speaking a language other than English at home, which may be barometers of health beliefs and knowledge, are associated with reduced rates of cancer screening even after adjusting for SES and access to care. The second study objective was to examine the reasons endorsed by Asians and non-Hispanic whites, both foreign-born and United States-born, for not having cancer screening tests. We hypothesized that foreign birth may be a marker of an individual's health-belief model, attitudes, and knowledge about cancer screening and that foreign-born Asians would be more likely than United States-born non-Hispanic whites to report that they did not obtain cancer screening because, “they had no problems/symptoms.”
MATERIALS AND METHODS
We used cross-sectional data from the 2001 California Health Interview Survey (CHIS), which is a population-based, random digit dial (RDD) telephone survey of civilian households that was designed to study health-related behaviors, health status and conditions, health insurance, and access to health care among California's major racial and ethnic groups. One adult per household was selected randomly to be interviewed; and, after obtaining verbal consent, 55,428 adults were interviewed. Data were collected between November 2000 and October 2001.
The CHIS provides a unique opportunity to study Asians because of the size and diversity of California's Asian population.35 Interviews were conducted among Chinese, Filipinos, Japanese, South Asians, Vietnamese, Koreans, and Cambodians. Respondents were interviewed in English, Spanish, Mandarin, Cantonese, Vietnamese, Korean, or Khmer. The inclusion of these Asian languages makes the Asian samples in the CHIS more representative than any other large health survey in the United States.36
The CHIS 2001 adult overall response rate is 37.7% (the product of a 59.2% of screener completion rate and a 63.7% adult interview completion rate).37 The final data are weighted and account for study design while also adjusting for nonresponse and the absence of nontelephone households. The final CHIS sample is consistent with California population totals from the 2000 Census.37
The sample for this analysis was restricted to adults age 18 years and older who were identified as non-Hispanic white, Chinese, Filipino, Vietnamese, Japanese, Korean, South Asian, or Cambodian. Because of sample size limitations and sampling methods, the CHIS 2001 RDD public use data file allows the Chinese, Filipinos, Koreans, and Vietnamese to be disaggregated, whereas the Japanese, South Asian, or Cambodian respondents are aggregated as “other Asian.” This yielded a sample with 36,660 non-Hispanic whites, 1298 Chinese, 944 Filipinos, 803 Koreans, 857 Vietnamese, and 1036 other Asians.
There were 3 main dependent variables of interest in the multivariate analysis: colorectal cancer screening, breast cancer screening, and cervical cancer screening. We defined adherence to screening based on the cancer-specific recommendations for screening by age and gender endorsed by the American Cancer Society at the time that CHIS data collection began.38
For colorectal cancer screening, we included men and women older than 50 years. Screening was either a fecal occult blood test (FOBT) in the previous year or flexible sigmoidoscopy, colonoscopy, or proctoscopy in the previous 5 years. The CHIS questions on colon cancer screening do not distinguish between sigmoidoscopy, colonoscopy, or proctoscopy: “Have you ever had a sigmoidoscopy, colonoscopy, or a proctoscopy to look for signs of cancer or other problems in your colon?” For cervical cancer screening, we studied women age 18 years and older who had not undergone hysterectomy. Women were considered screened if they reported having a Papanicolaou (Pap) smear in the previous 3 years. For breast cancer screening, we studied women age 40 years and older and considered women screened if they reported having a mammogram in the previous 2 years.
We also compared data from the CHIS about potential reasons why cancer-screening rates are lower in Asian Americans. Respondents who had not been screened according to guidelines were asked to identify the one most important reason they did not undergo sigmoidoscopy, FOBT, mammography, or Pap smear. Respondents were given a list of reasons that was predetermined by those who designed the CHIS survey; some of the reasons for not obtaining screening included: “I had no problems/symptoms,” “it was too expensive,” “don't have insurance,” “did not know the test was needed,” “don't have a doctor,” “I thought the test was painful or embarrassing,” and “the doctor did not tell me I needed the test.” Respondents could also give an answer that was not on the list (coded “other”). For the purposes of this study, some of these response categories were combined because of low frequency and related content.
Our main independent variable was self-reported race/ethnicity. Individuals were classified as non-Hispanic white, Chinese, Filipino, Korean, Vietnamese, or other Asian. Other demographic variables included age, which was used as continuous variable, and marital status. We also adjusted for gender when colorectal cancer screening, which is not gender-specific, was the outcome of interest.39
SES was measured by education and poverty income ratio (PIR). Education was categorized as “less than high school,” “high school graduate,” “some college,” and “college graduate.” Whether education was in the United States or abroad is not available in the data set. PIR is household income adjusted for household size divided by the Federal Poverty Level (FPL) for that size household. The PIR was categorized as “0% to 99% of FPL,” “100% to 199% FPL,” “200% to 299% FPL” and “≥300% FPL.” Education and PIR had linear relations with screening, and both of these variables were entered as a single ordinal variable in the regression models.
Access to care was defined as having health insurance and a usual source of care. Insurance status was categorized as currently being insured (any type) or being uninsured. Respondents were categorized as having access to a usual source of care if they reported having a regular provider of care that was not an emergency room or an urgent care clinic.
In addition to race/ethnicity, SES, and access to care, we also examined the associations of nativity and language use at home with cancer screening. United States-born individuals were those born in the United States, Puerto Rico, or other United States territories. All others were classified as foreign-born. Among the foreign-born, we categorized years in the United States as, “living in the United States for >15 years,” “living in the United States for 5 to 15 years,” and “living in the United States for <5 years.” In multivariate models, years in the United States was used as an ordinal variable, with United States-born individuals as the referent group, because the association between the log odds of cancer screening and the 4 categories was approximately linear. English language proficiency (speaking English “very well,” “well,” etc.) was not associated as strongly with cancer screening as language use at home. Because of their high correlation, only 1 of the language-use variables could be entered into the models. Language use at home has been associated with differences in other health behaviors.40–42 Individuals were categorized as “speaks only English at home,” “speaks English and another language at home,” “does not speak English at home.”
We did not adjust for self-reported health because of concerns that the question does not elicit comparable information from non-Hispanic whites and Asians. Asian Americans have lower self-reported health than non-Hispanic whites, despite having similar or lower rates of chronic disease or health limitations.43 Instead, we accounted for differences in illness burden by including self-report of a serious chronic disease (asthma, arthritis, diabetes, high blood pressure, and heart disease). The regression models also included current smoking status.
United States-born non-Hispanic whites were used as the referent group for all comparisons. First, we used bivariate analysis to compare the sociodemographic characteristics and screening rates for non-Hispanic whites, Chinese, Filipinos, Koreans, Vietnamese, and other Asians. Next, we used logistic regression analysis for the binary outcome of having been screened for colorectal, breast, and cervical cancer. We used separate logistic regression models for each of the 3 screening tests, in which the dependent variable was receipt of each of the screening tests within the recommended time frame. Only individuals for whom the screening test was recommended by gender and age criteria were included in each model. For each outcome, 4 hierarchical models were constructed (only Models 3 and 4 are shown). In Model 1, we adjusted for demographic factors: age, race/ethnicity, and gender. In Model 2, we added education, PIR, smoking status, and chronic illness burden. In Model 3, we added insurance status and usual source of care to determine whether the lower cancer screening rates among the Asian-American groups persisted after adjusting for access to care. In Model 4, we added nativity, years in the United States, and English language use at home to investigate the direct effects of these covariates on cancer screening and to determine whether adjustment for these variables changed the association between race/ethnicity and cancer screening.
To compare the reasons for not obtaining screening, we created 4 groups: United States-born non-Hispanic whites, foreign-born non-Hispanic whites, United States-born Asians, and foreign-born Asians. The Asians were aggregated because of sample size (the questions were only asked of individuals who did not receive the screening; therefore, the samples are smaller than for the previous analyses). Chi-square tests were used to compare the reasons that each group reported for not obtaining a screening test, with United States-born non-Hispanic whites as the referent group.
All estimates and analyses were weighted using replicate weights, provided by the CHIS, to adjust for nonresponse and the complex survey design. Statistical significance was measured at the 95% confidence interval level. Analyses were conducted using Stata software (version 8.0; Stata Corporation, College Station, TX).
All Asian groups were significantly younger, more likely to be uninsured, and more likely to be immigrants compared with the group of United States-born non-Hispanic whites (Table 1). There also was heterogeneity in characteristics among the Asian groups. In general, Vietnamese were poorer and less educated than the other Asian-American subgroups. Fifty percent of Chinese, 40% of Koreans, and 55% of Vietnamese reported speaking no English at home compared with 19% of Filipinos, 14% of other Asians, and 3% of non-Hispanic whites. Korean Americans were the least likely to report having health insurance and a usual source of care.
|Characteristic||Non-Hispanic White (n = 36,660)*||Chinese (n = 1298)||Filipino (n = 944)||Korean (n = 803)||Vietnamese (n = 857)||Other Asian (n = 1036)|
|Mean age, y||47.8||43.7†||42.6†||41.8†||39.5†||41.8†|
|Does not speak English at home (%)||3.0||49.8†||18.7†||40.4†||54.6†||14.3†|
|Immigrant, living in U.S. >15 years (%)||6.4||39.3†||46.8†||47.7†||42.2†||27.2†|
|Immigrant, living in U.S. >5-15 years (%)||1.8||28.5†||20.8†||24.7†||44.4†||21.1†|
|Immigrant, living in U.S. <5 years (%)||<1.0||12.4†||8.2†||13.8†||8.9†||15.1†|
|Education less than high school (%)||8.4||17.3†||10.0||11.3‡||30.7†||5.5‡|
|Income <100% Federal poverty level (%)||6.5||14.0†||6.8||11.6†||33.0†||10.1†|
|Currently insured (%)||91.1||84.9†||87.0‡||65.1†||79.4†||89.4|
|Has a usual source of care (%)||89.6||85.7†||89.1||64.9†||86.9||85.8‡|
|Current smokers (%)|
|Reports ≥1 chronic illnesses||45.1||28.8†||36.4†||25.9†||26.0†||30.5†|
Chinese, Filipinos, and Koreans all had significantly lower unadjusted rates of colorectal, cervical, and breast cancer screening than non-Hispanic whites (Table 2). Vietnamese had significantly lower rates of colorectal and cervical cancer screening, and other Asians only had significantly lower rates of cervical cancer screening.
|Screening Test||Percent Screened (SE)|
|Non-Hispanic White (n = 36,660)*||Chinese (n = 1298)||Filipino (n = 944)||Korean (n = 803)||Vietnamese (n = 857)||Other Asian (n = 1036)|
|Sigmoidoscopy/FOBT†||61.1 (0.004)||49.2 (0.028)‡||46.3 (0.038)‡||41.3 (0.038)‡||42.2 (0.030)‡||54.3 (0.038)|
|Pap smear§||88.8 (0.004)||67.6 (0.022)‡||81.1 (0.026)∥||63.4 (0.025)‡||62.3 (0.032)‡||70.3 (0.030)‡|
|Mammography¶||77.9 (0.005)||64.9 (0.031)‡||68.6 (0.034)∥||53.1 (0.040)‡||72.7 (0.032)||72.6 (0.033)|
SES, insurance, and access to a usual source of care were associated significantly with each of the 3 cancer screening tests, as expected, with the strongest association observed for having access to a usual source of care (Table 3). After adjusting for access to care (Table 3), Chinese, Filipino, and Vietnamese Americans still had significantly lower rates of colorectal cancer screening compared with non-Hispanic whites. However, after adjusting for access to care, Koreans no longer were less likely to report colorectal cancer screening than non-Hispanic whites (Table 3). The significantly lower rates of mammography among Chinese, Filipino, and Korean women compared with non-Hispanic whites persisted independent of insurance and access to care (Table 3). After adjusting for access to care, other Asian women also were significantly less likely to report mammography in the prior 2 years compared with non-Hispanic whites. Chinese, Filipino, Korean, Vietnamese, and other Asian women all reported significantly lower rates of cervical cancer screening than non-Hispanic white women even after adjusting for access to care (Table 3).
|Race/Ethnicity||OR (95% CI)|
|Chinese||0.74 (0.57-0.96)§||0.69 (0.49-0.98)§||0.27 (0.22-0.35)∥|
|Filipino||0.57 (0.41-0.80)∥||0.59 (0.42-0.85)∥||0.52 (0.35-0.76)∥|
|Korean||0.75 (0.51-1.12)||0.53 (0.34-0.83)∥||0.26 (0.21-0.33)∥|
|Vietnamese||0.68 (0.52-0.88)∥||1.46 (0.93-2.32)||0.29 (0.21-0.40)∥|
|Other Asian||0.76 (0.55-1.04)||0.70 (0.49-1.00)§||0.26 (0.19-0.36)∥|
|Currently insured||2.50 (2.01-3.11)∥||2.65 (2.16-3.26)∥||1.56 (1.29-1.90)∥|
|Has a usual source of care||3.61 (2.84-4.60)∥||3.61 (2.89-4.51)∥||2.37 (1.92-2.91)∥|
|Increasing education||1.18 (1.12-1.23)∥||1.10 (1.03-1.18)∥||1.36 (1.25-1.48)∥|
|Increasing poverty income ratio||1.07 (1.01-1.12)∥||1.15 (1.09-1.21)∥||1.21 (1.12-1.31)∥|
|Reports a chronic disease||1.42 (1.30-1.55)∥||1.37 (1.26-1.50)∥||1.21 (1.02-1.44)§|
|Married||1.31 (1.21-1.43)∥||1.30 (1.19-1.42)∥||2.79 (2.39-3.27)∥|
|Male gender||1.42 (1.30-1.54)∥||—||—|
|Age (years)||1.02 (1.02-1.03)∥||1.01 (1.00-1.01)∥||1.00 (0.99-1.00)|
|Current smoker||0.73 (0.64-0.82)∥||0.66 (0.58-0.75)∥||1.47 (1.21-1.79)∥|
Adjusting for nativity, years in the United States, and language use at home attenuated the association between race/ethnicity and lower rates of colorectal and breast cancer screening for several Asian subgroups. After adjusting for nativity, years in the United States, and language use at home, only Filipinos still were significantly less likely to report colorectal cancer screening than non-Hispanic whites (Table 4). There no longer were any significant differences in self-reported mammography, except that Vietnamese women were now significantly more likely to report mammography than non-Hispanic whites. The lower rates of cervical cancer screening among all Asian-American women were attenuated in part after adjusting for access to care; however, Chinese, Korean, Vietnamese, and other Asians still had significantly lower rates of cervical cancer screening compared with non-Hispanic whites (Table 4).
|Characteristic||OR (95% CI)|
|Chinese||1.00 (0.75-1.33)||0.85 (0.59-1.22)||0.45 (0.32-0.62)∥|
|Filipino||0.71 (0.50-1.00)¶||0.74 (0.51-1.07)||0.81 (0.53-1.22)|
|Korean||0.99 (0.64-1.53)||0.66 (0.41-1.08)||0.42 (0.31-0.58)∥|
|Vietnamese||1.00 (0.73-1.38)||1.93 (1.17-3.17)∥||0.52 (0.35-0.79)∥|
|Other Asian||0.84 (0.60-1.16)||0.77 (0.53-1.11)||0.38 (0.26-0.55)∥|
|Currently insured||2.41 (1.94-3.00)∥||2.57 (2.10-3.15)∥||1.47 (1.20-1.81)∥|
|Has a usual source of care||3.58 (2.81-4.55)∥||3.62 (2.91-4.51)∥||2.29 (1.85-2.83)∥|
|Speaks only English at home||Referent||Referent||Referent|
|Speaks English and Asian language at home||0.96 (0.82-1.12)||1.20 (0.991.45)||0.79 (0.62-1.00)¶|
|Speaks no English at home||0.75 (0.58-0.97)¶||0.99 (0.72-1.35)||0.77 (0.54-1.11)|
|Years in the U.S. (4 levels)#|
|1||0.90 (0.77-1.05)||0.86 (0.71-1.05)||0.98 (0.73-1.32)|
|2||0.76 (0.38-1.52)||0.50 (0.27-0.92)¶||0.56 (0.35-0.90)¶|
|3||0.47 (0.27-.82)∥||0.38 (0.22-0.65)∥||0.34 (0.21-0.55)∥|
Being born outside the United States and living in the United States for fewer years were associated significantly and independently with lower rates of breast and cervical cancer screening compared with being born in the United States (Table 4). Speaking a language other than English at home was associated with lower rates of cervical cancer screening, and speaking no English at home was associated with lower rates of colorectal cancer screening (Table 4).
For every screening test, foreign-born Asians were more than twice as likely as United States-born non-Hispanic whites to report that the single most important reason they did not get the test was because “they haven't had problems or symptoms,” (P<.01) (Table 5). In contrast, United States-born Asians were not more likely to report, “haven't had problems or symptoms” as a reason. Reasons such as “pain/embarrassment,” and “do not have a doctor/no insurance/too expensive,” were not more likely to be reported by foreign-born Asians.
|Most important reason for not having sigmoidoscopy (%)|
|Have not had problems/symptoms||16.0||20.1||18.6||34.5†|
|Doctor did not tell me I needed it/did not know test was needed||34.2||28.9||34.7||27.5†|
|Do not have a doctor/no insurance/expensive||3.8||6.0‡||<0.1||4.5|
|Had another type of test||1.0||0.7||0.23||<0.1†|
|No specific reason/lazy/other||37.4||36.0||41.3||30.8‡|
|Most important reason for not having FOBT (%)|
|Have not had problems/symptoms||12.5||18.4‡||2.5†||30.4†|
|Doctor did not tell me I needed it/did not know test was needed||38.3||36.2||51.2||30.8†|
|Do not have a doctor/no insurance/expensive||1.6||1.5||<0.1||3.5†|
|Most important reason for not having Pap smear (%)|
|Have not had problems/symptoms||5.8||10.4||8.1||16.9†|
|Doctor did not tell me I needed it/did not know test was needed||11.7||11.6||10.6||19.0†|
|Do not have a doctor/no insurance/expensive||16.3||13.6||0.6†||5.9†|
|No specific reason/lazy/other||59.4||60.1||75.1‡||54.7|
|Most important reason for not having mammogram (%)|
|Have not had problems/symptoms||6.5||13.4‡||6.3||21.1†|
|Doctor did not tell me I needed it/did not know test was needed||16.8||16.3||13.0||14.1|
|Do not have a doctor/no insurance/expensive||11.6||13.5||<0.1||9.6|
|Had another type of test||3.1||3.1||5.6||1.6|
|No specific reason/lazy/other||57.2||47.7†||67.4||48.7†|
When they were asked why they did not have a Pap smear, foreign-born Asian women were significantly more likely than United States-born non-Hispanic white women to report both “haven't had problems/symptoms” (17% vs. 6%) and “doctor did not tell me I needed it/did not know the test was needed” (19% vs. 12%).
Using a large, population-based sample of diverse Asian-American groups, we observed that nativity, years in the United States, and speaking a language other than English at home were associated significantly with lower rates of cancer screening independent of SES and access to care. After adjusting for these factors, Asian-American subgroups no longer were less likely to be screened for mammography and colorectal cancer (except for Filipinos) than non-Hispanic whites. Prior studies also have found an association between foreign birth, language use at home, and lower rates of cancer screening in Asian Americans.5, 11, 13–15, 18, 21, 27, 29, 39 However, comparing cancer screening rates across these studies is difficult, because few of the other studies utilized population-based samples of Asian Americans in California or elsewhere. The findings in the current study suggest that foreign birth and language are not just markers of SES and access to health care: They also may be markers of health beliefs, knowledge, attitudes, and patient-provider communication regarding cancer screening and prevention.
The data on the reasons for not obtaining screening also support this argument. Our findings suggest that foreign-born Asian Americans perceive that cancer screening tests are a response to a specific symptom of cancer rather than tests that are used prior to the development of symptoms. This pattern is consistent with findings from limited prior study of this issue; 2 small studies, including a survey of Korean women and a qualitative study of Chinese women, found that the women who were interviewed did not obtain cancer screening tests because they “felt fine” and “did not have any problems.”18, 21
Similar to prior studies,5, 13, 16, 27 our results suggest that Asian-American women remain vulnerable to cervical cancer, which is the leading type of cancer affecting Korean, Vietnamese, and Cambodian women.44 Cervical cancer mortality rates are high among Chinese women.45 Although improvements have been made in increasing cervical cancer screening rates in Asian women,4 our results confirm that large cervical cancer screening disparities still exist for Asian Americans, regardless of SES, insurance, access to care, language, or nativity.
There are several limitations to this study. Although a majority of Asian Americans live in California, it is unclear whether these finding are generalizable to Asians throughout the United States. The response rate for the CHIS was 64% after households were identified by screening for eligibility, and there are limited data on the nonresponders. However, comparisons of CHIS with 2000 Census data confirm that the racial/ethnic profile of respondents matches that of the California population.37 These data are based on self-report, and the accuracy of self-reported cancer screening rates may vary by nativity, language proficiency, and race/ethnicity. Although we could disaggregate several Asian groups, these results ma vary among Asian subgroups that were not disaggregated.
We observed that lower rates of colorectal, cervical, and breast cancer screening exist in a population-based sample of diverse Asian-American subgroups. Although insurance and access to care are associated strongly with cancer screening for Asian Americans, these factors explain only in part the lower rates of cancer screening. Nativity, years in the United States, and language use at home also are important determinants of cancer screening for Asian Americans. Although nativity, years in the United States, and language use at home may be markers of structural barriers to cancer screening, they also may be markers of an individual's health-belief model regarding cancer screening and prevention. More research is needed to elucidate the mechanisms involved in the association between nativity, years in the United States, language use at home, and screening. In addition to addressing structural barriers, such as access to care, there remains a need for improved communication methods to overcome linguistic and cultural barriers to cancer screening in Asian Americans. Qualitative methods also may help determine whether there are cultural barriers to cancer screening and whether culturally tailored health communication messages are more effective than nontailored messages. On an individual level, health care providers need to take the time to elicit the cancer-screening beliefs of Asian patients and counsel them about the potential benefits of screening even in the absence of symptoms. Health education messages aimed at Asian Americans must consider how to effectively communicate the message that “cancer screening is valuable, because it finds cancer before it is advanced enough to cause symptoms” in a language that can be understood.
We thank David Baker, MD, MPH and Mita Goel, MD, MPH for comments on this article.
- 1Asian American Cancer Awareness Research and Training. The unequal burden of cancer among Asian Americans, 2003. Asian American Network for Cancer Awareness, Research and Training. Available at URL: http://www.aancart.org/unequal%20.Burden.htm Accessed September 1, 2005.
- 14Exploring access to cancer control services for Asian-American and Pacific Islander communities in Southern California. Ethn Dis. 2004; 14: S14–S19., , , et al.
- 35The Asian population: Census 2000 brief: 2004 vol, 2002. Washington, DC: Bureau of the Census, 2004., .
- 37California Health Interview Survey. CHIS 2001 methodology series: report 4—response rates. Los Angeles: UCLA Center for Health Policy Research, 2002.
- 44Racial/ethnic patterns of cancer in the United States 1988-1992. Bethesda: National Cancer Institute, 1996., , , .