Objectives. We examined patterns of cervical and breast cancer screening among Asian American women in California and assessed their screening trends over time.
Methods. We pooled weighted data from 5 cycles of the California Health Interview Survey (2001, 2003, 2005, 2007, 2009) to examine breast and cervical cancer screening trends and predictors among 6 Asian nationalities. We calculated descriptive statistics, bivariate associations, multivariate logistic regressions, predictive margins, and 95% confidence intervals.
Results. Multivariate analyses indicated that Papanicolaou test rates did not significantly change over time (77.9% in 2001 vs 81.2% in 2007), but mammography receipt increased among Asian American women overall (75.6% in 2001 vs 81.8% in 2009). Length of time in the United States was associated with increased breast and cervical cancer screening among all nationalities. Sociodemographic and health care access factors had varied effects, with education and insurance coverage significantly predicting screening for certain groups. Overall, we observed striking variation by nationality.
Conclusions. Our results underscore the need for intervention and policy efforts that are targeted to specific Asian nationalities, recent immigrants, and individuals without health care access to increase screening rates among Asian women in California.
Cancer is the leading cause of death among Asian American women, with breast cancer and cervical cancer being 2 of the most commonly diagnosed types of cancer among this population.1–6 Although Asian American women have lower mortality rates for breast cancer compared with women from other racial/ethnic groups, existing research suggests that they have later cancer stage at diagnosis,3 younger age at diagnosis,7 and poorer survival among certain subgroups.8,9 Notably, studies have highlighted significant variation in cancer risk factors and differential cancer burden among Asian nationalities.4–6 Therefore, researchers have increasingly made efforts to disaggregate data on Asian Americans for analyses to identify subgroup differences. In one study conducted in California, McCracken et al.10 found that Vietnamese women had the highest mortality rates for cervical cancer, whereas Filipino women had the highest mortality rates for breast cancer. Another study conducted by Bates et al.11 found that Vietnamese and Korean women had the highest rates of cervical cancer mortality compared with other Asian women and White women.
Despite a clear need for cancer screening among all Asian American women, their screening rates for breast and cervical cancers remain well below national objectives promoted by Healthy People 2020.12,13 Miller et al.2 found that 73.7% of all Asian women reported a mammogram in the past 2 years—nearly 10% lower than the Healthy People 2020 objectives of 81.1%, and lower than all other racial/ethnic groups except for American Indians or Alaska natives.12 In addition, Asian American women consistently have the lowest rates of cervical cancer screening, with 65.6% reporting a Papanicolaou (Pap) test in 2008.2 This rate was almost 10% lower than screening rates for White women and nearly 30% lower than the national recommendation of 93.0%.12 Additional research has noted that these disparities in breast and cervical cancer screening among Asian women have persisted over time for many groups.14–19
Asian American women experience significant challenges and barriers to cancer screening, with notable differences by Asian nationality. Some factors associated with this heterogeneity include disparate levels of access to care,20–27 socioeconomic status,26–28 English proficiency,29,30 immigration status and length of US residency,23–25,27,31–33 screening-related knowledge,24,34,35 and health beliefs among Asian women.22,26,35,36 Studies have also documented that several Asian communities tend to use health care services for treatment rather than for prevention.37–39 Available research suggests that Asian American orientation toward preventive behaviors includes using complementary and alternative medicine rather than health care services for prevention of disease and may be influenced by low knowledge levels about the technology of Pap tests and mammograms.40–42 Additionally, studies have found that some Asian American communities associate cancer with “a death sentence” and describe fatalism as a barrier to cancer screening.43–45
Until release of the California Health Interview Survey (CHIS), population-based data on Asian Americans were limited by small sample sizes that prevented the ability to disaggregate data on Asian American subgroups. One study that used CHIS data found a wide range of screening rates among Asian American women, with certain groups facing greater disadvantage, such as Vietnamese and Southeast Asian women.46 However, this study did not examine changes in cancer screening trends over time. In our study, we pooled data from the 2001 to 2009 CHIS to assess breast and cervical cancer screening rates for 6 different Asian American subgroups and examined the following research questions:
What are the patterns of breast and cervical cancer screening among Asian American women in California, and how have they changed over time?
Which subgroups of Asian American women have lower screening rates?
What differences and similarities exist in factors associated with screening use among Asian Americans?
Findings from this research may provide helpful insights for interventions targeting Asian American women and for future research in this diverse population.
METHODSOur study sample was derived from the CHIS, a random-digit-dialed health survey conducted in English, Spanish, and 4 Asian languages (Cantonese, Mandarin, Korean, and Vietnamese).47,48 The 2001 survey was also conducted in Khmer. We pooled data from 5 cycles of CHIS from 2001, 2003, 2005, 2007, and 2009 surveys to help increase the stability of estimates for Asian American nationalities. The weighting factor used for each CHIS cycle was 0.2. We examined Pap test and mammography use among Asian American women overall and among Chinese, Filipino, Japanese, Korean, South Asian, and Vietnamese nationalities to explore predictors of cancer screening, detect differences between Asian American ethnic groups, and examine screening trends over time. Pap test receipt was not included in the 2009 CHIS; therefore, results for cervical cancer screening include pooled data from 2001, 2003, 2005, and 2007 only.
The overall response rates for the landline and list-assisted adult samples ranged from 37.7% in 2001 to 17.7% in 2009, which are comparable to the Behavioral Risk Factor Surveillance System rates for equivalent years.49,50 We conducted analyses on 7865 Asian American women aged 21 to 64 to assess rates of Pap test receipt and on 4521 Asian American women aged 50 to 74 to assess mammography rates.
Measures Dependent variables.Dependent variables were Pap test receipt in the past 3 years and mammography receipt in the past 2 years. The question used to create the variable for Pap tests was “How long ago did you have your most recent Pap smear test?” Similarly, the item used to create the screening variable for mammography was “How long ago did you have your most recent mammogram?” Individuals were categorized as “yes” if they received screening tests within the defined time frame and “no” if not. Missing and nonresponse data for these variables were dropped from final analyses. Women who reported a hysterectomy were not included in analyses of Pap test use.
Independent variables.Independent variables were survey year, Asian nationality, and sociodemographic, acculturation, and health care access measures. Survey year accounted for changes in screening use over time. Asian nationality was defined by the “Asian9” variable in the CHIS data set, which included Chinese, Filipino, Japanese, Korean, South Asian, Vietnamese, Cambodian, other Southeast Asian, and other Asian/multiple race. We did not conduct nationality-specific analyses for Cambodians, other Southeast Asians, and other Asian/multiple race because of small sample sizes. We tried to combine Cambodians and Southeast Asians into a single group, but the sample size remained too small to produce reliable estimates. All tables present a total column for all Asian Americans; these data include Southeast Asians and other Asian/multiple race.
Sociodemographic variables included age (21–29, 30–39, 40–49, 50–64 for Pap testing; and 50–64, 65–74 for mammography), marital status (married or living as married/other), education (high school education or less, any college or technical school, college graduate or higher), and income (< 200% of federal poverty level; ≥ 200% of federal poverty level). Acculturation variables included percentage of life spent in the United States (0%–24.99%; 25%–49.99%; 50%–99.99%; 100%) and English proficiency (yes/no). Finally, health care access variables included health insurance (uninsured, public only, some private health maintenance organization [HMO], some private non-HMO), having a usual source of care (yes/no), and number of doctor visits in the past 12 months (0, 1–2, 3 or more). We based our variable selection on literature on cancer screening both in the general US population and in the Asian American population. Missing and nonresponse data for independent variables were eliminated from final analyses.
Data AnalysisWe calculated descriptive statistics for characteristics and screening rates among Asian American women in our study sample. We examined bivariate associations between screening and independent variables to select relevant predictors of cervical and breast cancer screening among Asian American women. General health status was included in bivariate analyses but subsequently removed from multivariate analyses because of lack of significance with screening. We conducted multiple logistic regressions on selected independent variables. Models stratified by Asian nationality included survey year, sociodemographics, acculturation, and health care access. Models conducted on the aggregate Asian sample included these variables as well as Asian nationality. We calculated predictive margins and 95% confidence intervals. The predictive margins are adjusted percentages directly standardized to the distribution of the covariates for the population that the sample represents.51
All analyses were weighted to address the complex CHIS design and to produce estimates that were representative of the California population. We used SAS 9.2 and SUDAAN 10.0.1 software to conduct all analyses.52,53
RESULTSTable 1 presents descriptive statistics for Asian American women aged 21 to 74 years in the pooled CHIS sample. In the total Asian American sample, most individuals were younger than 50 years (67.8%), were married (70.7%), had some college education or higher (73.4%), had incomes greater than 200% of the federal poverty level (71.6%), had some private HMO insurance (55.4%), had a usual source of care (86.4%), and were English proficient (76.3%). Approximately 1 of 4 women reported that they were US born (22.1%). Beyond these broader trends, Asian American nationalities varied significantly with respect to demographic, acculturation, and health care access characteristics.
TABLE 1—Sample Characteristics of Asian American Women Aged 21–74 Years, by Nationality, Sample Sizes, Weighted Population Estimates (Pop Est), and Weighted Percentages: California Health Interview Survey, 2001–2009
Asian Chinese Filipino Japanese Korean South Asian Vietnamese No. Pop. Est. % No. Pop. Est. % No. Pop. Est. % No. Pop. Est. % No. Pop. Est. % No. Pop. Est. % No. Pop. Est. % Survey year 2001 2662 285 994 18.3 698 87 158 20.0 532 76 495 18.4 260 23 521 19.8 430 26 567 16.2 178 20 378 13.7 366 29 665 17.3 2003 2101 292 882 18.7 661 84 067 19.3 394 75 394 18.1 212 25 425 21.4 286 30 686 18.7 188 26 123 17.6 220 31 648 18.4 2005 2188 323 806 20.7 671 89 374 20.5 387 87 122 20.9 218 23 171 19.5 353 32 893 20.0 189 34 983 23.5 238 35 704 20.8 2007 2384 326 264 20.9 731 89 132 20.4 456 92 702 22.3 279 23 976 20.2 345 32 886 20.0 209 30 182 20.3 216 37 516 21.8 2009 2532 333 968 21.4 547 86 823 19.9 274 84 579 20.3 188 22 653 19.1 551 41 423 25.2 189 37 023 24.9 627 37 221 21.7 Age, y 21–29 1468 308 573 19.7 394 81 778 18.7 260 70 151 16.9 100 11 986 10.1 192 37 528 22.8 190 36 553 24.6 173 36 247 21.1 30–39 2829 387 044 24.8 745 104 788 24.0 476 91 427 22.0 199 21 403 18.0 435 37 192 22.6 366 59 747 40.2 391 41 770 24.3 40–49 3034 364 874 23.3 862 105 440 24.2 502 97 926 23.5 319 30 293 25.5 565 43 209 26.3 210 29 820 20.1 401 37 172 21.6 50–64 3216 361 784 23.1 971 103 604 23.7 580 111 053 26.7 343 32 665 27.5 468 33 970 20.7 154 19 002 12.8 527 46 244 26.9 65–74 1320 140 639 9.0 336 40 944 9.4 225 45 735 11.0 196 22 400 18.9 305 12 555 7.6 33 3568 2.4 175 10 320 6.0 Marital status Married or living as married 8182 1 104 869 70.7 2267 309 081 70.8 1366 293 886 70.6 712 83 755 70.5 1427 114 399 69.7 770 120 702 81.2 1131 115 872 67.5 Other 3677 457 284 29.3 1039 127 334 29.2 676 122 344 29.4 445 34 991 29.5 536 49 721 30.3 182 27 958 18.8 535 55 858 32.5 Education completed ≤ High school 3250 416 190 26.6 846 135 514 31.0 339 67 845 16.3 190 22 714 19.1 566 44 104 26.8 98 15 027 10.1 990 95 936 55.9 Any college or technical school 2257 295 552 18.9 585 68 409 15.7 497 102 770 24.7 334 34 464 29.0 278 22 145 13.5 111 16 909 11.4 254 28 660 16.7 College graduate or higher 6360 851 172 54.5 1877 232 631 53.3 1207 245 676 59.0 633 61 569 51.8 1121 98 206 59.7 744 116 753 78.5 423 47 157 27.5 % of time in United States 0–24.99 2612 347 336 22.2 800 114 378 26.2 324 69 578 16.7 77 7946 6.7 528 43 757 26.6 314 52 604 35.4 499 49 007 28.6 25–49.99 3568 455 347 29.2 1026 143 001 32.8 538 117 223 28.2 98 10 842 9.1 734 52 660 32.1 312 46 209 31.1 663 56 745 33.1 50–99.99 3149 414 652 26.5 784 99 584 22.8 636 134 141 32.2 208 21 896 18.4 556 42 220 25.7 234 33 394 22.5 446 46 280 27.0 100 2527 344 524 22.1 697 79 505 18.2 543 95 070 22.9 774 78 063 65.7 144 25 661 15.6 93 16 482 11.1 56 19 526 11.4 English proficiency English proficient 8442 1 191 917 76.3 2269 284 287 65.1 1966 401 537 96.5 1080 110 033 92.7 897 85 280 51.9 911 143 340 96.4 681 79 947 46.5 Limited or no English proficiency 3425 370 998 23.7 1039 152 266 34.9 77 14 754 3.5 77 8713 7.3 1068 79 174 48.1 42 5349 3.6 986 91 806 53.5 Income as a % of federal poverty level (FPL) < 200% FPL 3549 443 789 28.4 942 136 762 31.3 481 93 740 22.5 154 15 461 13.0 640 48 140 29.3 149 22 591 15.2 932 89 110 51.9 ≥ 200% FPL 8318 1 119 125 71.6 2366 299 792 68.7 1562 322 551 77.5 1003 103 286 87.0 1325 116 315 70.7 804 126 098 84.8 735 82 643 48.1 Health insurance Uninsured 1670 222 562 14.2 400 61 126 14.0 179 34 307 8.2 60 6546 5.5 579 52 853 32.1 88 14 964 10.1 271 34 756 20.2 Public only HMO 1561 151 682 9.7 342 40 982 9.4 190 35 053 8.4 152 15 376 12.9 298 14 798 9.0 32 2799 1.9 457 32 841 19.1 Public only non-HMO 511 68 111 4.4 120 22 302 5.1 117 22 685 5.4 30 1828 1.5 72 6021 3.7 21 2615 1.8 111 8781 5.1 Some private HMO 6423 865 037 55.4 1957 247 559 56.7 1211 247 940 59.6 747 79 260 66.7 809 69 280 42.1 605 90 246 60.7 675 77 968 45.4 Some private non-HMO 1701 255 375 16.3 488 64 440 14.8 346 76 306 18.3 168 15 736 13.3 207 21 503 13.1 207 38 065 25.6 153 17 408 10.1 No. of times saw a physician in past 12 mo 0 1892 250 587 16.1 587 80 097 18.4 226 43 134 10.4 157 18 109 15.3 420 41 293 25.1 126 19 866 13.4 276 30 480 18.0 1–2 4629 619 703 39.8 1320 170 352 39.2 882 184 970 44.5 461 46 960 39.6 688 50 272 30.6 411 65 522 44.1 546 57 315 33.8 ≥ 3 5305 687 266 44.1 1391 184 346 42.4 932 187 281 45.1 537 53 472 45.1 855 72 781 44.3 416 63 301 42.6 823 81 791 48.2 Has a usual source of care Yes 10 285 1 349 278 86.4 2918 380 305 87.2 1845 378 393 90.9 1050 108 218 91.1 1546 120 429 73.2 826 127 008 85.4 1415 142 602 83.0 No or usual source is the emergency department 1580 213 181 13.6 388 55 794 12.8 198 37 899 9.1 107 10 529 8.9 419 44 026 26.8 127 21 681 14.6 252 29 151 17.0 General health status Excellent/very good 5469 775 700 49.6 1588 208 745 47.8 1123 221 933 53.3 728 73 110 61.6 664 65 948 40.1 591 97 050 65.3 391 55 842 32.5 Good 3667 480 381 30.7 1045 139 482 32.0 615 128 436 30.9 312 33 326 28.1 709 58 560 35.6 281 42 303 28.5 458 42 155 24.6 Fair/poor 2727 306 436 19.6 674 88 115 20.2 305 65 922 15.8 117 12 311 10.4 591 39 913 24.3 81 9337 6.3 816 73 603 42.9 Rates of Screening by Survey Year, Age, and Asian NationalityTable 2 presents rates of Pap test and mammography receipt by survey year and age among Asian American women. Pap testing rates remained stable over time, but all rates were below the recommended level of screening, ranging from 77.3% in 2001 to 80.8% in 2007. In 2007, Chinese (77.5%) and Korean (78.0%) women had the lowest rates of Pap test receipt, whereas Japanese women had the highest (85.5%). A trend across Asian American nationalities was that women between ages 21 and 29 had the lowest rates of Pap testing compared with other age groups.
TABLE 2—Weighted Percentages of Papanicolaou Test and Mammography Use Reported by Asian American Women, by Nationality, Age, and Survey Year: California Health Interview Survey (CHIS), 2001–2009
Asian Chinese Filipino Japanese Korean South Asian Vietnamese No. % (95% CI) No. % (95% CI) No. % (95% CI) No. % (95% CI) No. % (95% CI) No. % (95% CI) No. % (95% CI) Papanicolaou Test Survey year 2001 2254 77.3 (74.9, 79.6) 604 72.5 (67.6, 76.9) 442 87.4 (82.6, 91.0) 202 82.1 (72.9, 88.7) 357 72.9 (67.7, 77.6) 163 75.2 (66.4, 82.3) 314 69.3 (62.1, 75.7) 2003 1818 79.6 (76.8, 82.1) 560 73.5 (68.4, 78.1) 351 91.8 (86.3, 95.2) 162 78.3 (67.7, 86.2) 249 70.9 (61.4, 78.9) 180 80.7 (71.3, 87.5) 190 72.2 (62.8, 80.0) 2005 1870 78.3 (75.0, 81.2) 586 76.6 (71.4, 81.1) 322 85.6 (78.6, 90.5) 168 85.4 (78.3, 90.4) 294 70.0 (62.0, 76.9) 176 80.7 (73.2, 86.5) 206 75.7 (66.8, 82.8) 2007 1923 80.8 (77.6, 83.6) 594 77.5 (71.4, 82.6) 351 82.2 (75.9, 87.1) 205 85.5 (76.2, 91.6) 266 78.0 (69.3, 84.8) 192 79.7 (68.9, 87.4) 190 84.6 (77.2, 90.0) Age, y 21–29 1251 63.6 (59.3, 67.6) 359 53.8 (46.7, 60.8) 232 73.5 (63.8, 81.4) 90 63.9 (48.4, 77.0) 153 64.4 (50.1, 76.5) 165 65.5 (55.1, 74.7) 116 65.6 (53.1, 76.2) 30–39 2331 83.2 (81.2, 85.1) 634 79.4 (75.3, 82.9) 426 88.7 (83.8, 92.3) 176 85.4 (76.4, 91.4) 347 76.9 (71.1, 81.9) 296 84.8 (78.1, 89.8) 277 81.5 (74.7, 86.8) 40–49 2210 85.3 (83.3, 87.1) 659 84.0 (80.2, 87.1) 412 91.7 (87.9, 94.4) 249 91.0 (86.0, 94.3) 391 77.7 (72.4, 82.3) 151 85.3 (79.0, 90.0) 228 76.6 (68.2, 83.4) 50–64 2073 82.0 (79.8, 84.0) 692 80.9 (76.8, 84.5) 396 90.0 (86.2, 92.8) 222 81.6 (73.6, 87.6) 275 69.0 (61.0, 75.9) 99 81.7 (67.7, 90.5) 279 78.1 (71.1, 83.8) Total 7865 79.0 (77.6, 80.4) 2344 75.1 (72.5, 77.5) 1466 86.6 (83.8, 89.0) 737 82.7 (78.4, 86.4) 1166 73.0 (69.0, 76.7) 711 79.4 (75.1, 83.2) 900 76.1 (72.1, 79.7) Mammography Survey year 2001 778 76.0 (72.1, 79.4) 206 71.9 (63.6, 79.0) 156 81.1 (71.8, 87.9) 88 82.0 (69.4, 90.2) 134 57.6 (46.1, 68.3) 30 91.6 (73.1, 97.7) 125 75.9 (65.2, 84.1) 2003 736 79.1 (75.3, 82.4) 256 72.7 (63.8, 80.0) 138 86.6 (79.3, 91.6) 101 88.8 (78.0, 94.7) 92 68.7 (54.2, 80.4) 29 79.0 (59.4, 90.6) 84 80.0 (64.7, 89.7) 2005 799 80.2 (76.4, 83.6) 253 82.4 (76.3, 87.1) 152 79.9 (70.3, 87.0) 100 80.0 (68.4, 88.1) 125 69.1 (53.3, 81.4) 36 84.0 (59.3, 95.0) 99 82.4 (70.1, 90.4) 2007 1010 81.2 (77.7, 84.3) 340 83.3 (79.1, 86.8) 196 83.4 (74.9, 89.4) 144 86.4 (77.4, 92.2) 147 64.4 (51.1, 75.7) 44 82.8 (56.3, 94.8) 89 76.9 (63.0, 86.7) 2009 1198 82.6 (77.2, 86.9) 251 81.2 (71.4, 88.3) 157 82.6 (70.6, 90.4) 106 93.3 (83.3, 97.5) 274 64.7 (52.6, 75.1) 47 69.7 (54.3, 81.6) 302 93.6 (89.9, 96.0) Age, y 50–64 3206 80.3 (78.0, 82.4) 971 78.1 (74.2, 81.5) 575 84.1 (78.6, 88.4) 343 83.4 (77.6, 88.0) 467 65.7 (58.3, 72.4) 153 79.0 (70.4, 85.7) 526 83.0 (77.2, 87.6) 65–74 1315 79.5 (76.4, 82.3) 335 79.2 (72.7, 84.4) 224 79.1 (72.0, 84.7) 196 90.1 (83.0, 94.4) 305 63.8 (53.8, 72.8) 33 81.9 (50.3, 95.3) 173 78.8 (67.0, 87.1) Total 4521 80.1 (78.3, 81.7) 1306 78.4 (75.2, 81.3) 799 82.6 (78.3, 86.2) 539 86.1 (81.9, 89.5) 772 65.2 (59.3, 70.7) 186 79.5 (71.3, 85.8) 699 82.2 (77.2, 86.3)Mammography rates for Asian American women in aggregate rose from 76.0% in 2001 to 82.6% in 2009. In disaggregated analyses, mammography rates rose over time for most Asian American ethnic groups, with the greatest increases among Vietnamese, Chinese, and Japanese women. However, rates of mammography may have declined among South Asian women between 2001 and 2009. Relative to the Healthy People 2020 recommendations, all Asian nationalities met the objectives of 81.1% except Korean (64.7%) and South Asian (69.7%) women in 2009.
Multivariate ResultsPredictive margins calculated from multivariate results for Pap test and mammography receipt, respectively, are presented in Tables 3 and 4. Pap test rates remained stable over time, with no significant changes by survey year for nearly all Asian nationalities with the exception of South Asian women who increased screening rates from 72.5% in 2001 to 83.4% in 2007 (Table 3). Results among Asian women in aggregate, which are presented in the first regression column, indicated that Filipino women were more likely to receive Pap tests compared with Chinese women (83.5% vs 77.6%; P < .001). Asian American women who were aged 30 to 39 years and 40 to 49 years were significantly more likely to have Pap tests compared with those aged 21 to 29 years for all Asian American ethnic groups except Korean women. Unmarried women from all Asian American ethnic groups, except Korean women, were significantly less likely to receive Pap tests compared with their married counterparts. This difference was particularly striking for South Asian women (83.8% vs 56.2%; P < .001). Asian American women in aggregate, Filipino women, and Korean women with less than a college degree were less likely to receive Pap tests compared with college graduates. Asian American women who spent more time in the United States were more likely to report Pap test receipt. This finding was true for Asian American women aggregated and all Asian American ethnic groups, with US-born Asian women reporting the highest screening rates. Women from all Asian American ethnic groups with fewer physician visits in the past 12 months also had lower Pap test rates compared with those with 3 or more visits. Uninsured Asian women in aggregate and uninsured South Asian and Chinese women were less likely to report receiving a Pap test compared with their insured counterparts, but this finding was not consistent for the other Asian nationalities.
TABLE 3—Predictive Margins (PMs) From Multivariate Analyses for Asian American Women Aged 21–65 Years, by Nationality Reporting a Papanicolaou Test in the Past 3 Years: California Health Interview Survey, 2001–2009
Asian Chinese Filipino Japanese Korean South Asian Vietnamese PM (95% CI) PM (95% CI) PM (95% CI) PM (95% CI) PM (95% CI) PM (95% CI) PM (95% CI) Survey year 2001 (Ref) 77.9 (75.7, 80.1) 73.8 (69.7, 77.8) 86.5 (81.9, 91.0) 82.0 (74.1, 89.9) 75.7 (70.5, 81.0) 72.5 (65.0, 80.1) 72.7 (66.5, 78.8) 2003 79.5 (77.0, 81.9) 73.6 (69.3, 77.8) 91.6 (87.4, 95.8) 81.4 (74.3, 88.5) 72.2 (64.8, 79.6) 79.8 (72.7, 87.0) 74.0 (66.6, 81.4) 2005 77.7 (74.9, 80.5) 75.5 (71.5, 79.4) 86.6 (81.4, 91.7) 81.7 (75.2, 88.2) 68.3 (61.6, 75.1) 79.3 (72.3, 86.3) 75.3 (67.7, 83.0) 2007 81.2 (78.4, 83.9) 78.2 (72.7, 83.7) 82.2 (77.2, 87.3) 85.7 (79.5, 92.0) 76.4 (66.7, 86.1) 83.4* (76.8, 90.0) 81.5 (74.4, 88.6) Age, y 21–29 (Ref) 68.3 (64.3, 72.2) 61.2 (54.6, 67.8) 76.7 (68.7, 84.8) 75.8 (65.4, 86.2) 64.6 (49.0, 80.1) 73.1 (65.4, 80.9) 58.6 (46.5, 70.7) 30–39 81.2*** (79.3, 83.2) 76.6*** (72.8, 80.4) 88.4** (84.3, 92.6) 88.4* (82.2, 94.6) 72.7 (66.6, 78.9) 82.4* (76.8, 88.0) 76.6* (70.1, 83.2) 40–49 84.9*** (83.1, 86.7) 83.7*** (80.5, 87.0) 91.2*** (87.8, 94.6) 88.9** (84.8, 92.9) 80.3 (75.8, 84.8) 84.4* (78.7, 90.1) 77.5** (70.5, 84.4) 50–65 81.4*** (79.2, 83.6) 79.6*** (75.8, 83.5) 88.9** (85.3, 92.5) 75.5 (68.6, 82.3) 69.6 (62.6, 76.7) 79.0 (65.6, 92.4) 82.9*** (77.7, 88.1) Marital status Married or living as married (Ref) 83.3 (81.9, 84.6) 79.7 (77.0, 82.4) 90.2 (87.6, 92.8) 85.9 (81.6, 90.1) 73.9 (69.8, 78.0) 83.8 (80.2, 87.3) 81.4 (77.2, 85.5) Other 69.5*** (66.7, 72.2) 66.0*** (61.2, 70.8) 79.5*** (74.6, 84.4) 76.8** (70.8, 82.7) 70.5 (62.8, 78.2) 56.2*** (44.2, 68.3) 63.0*** (54.5, 71.5) Education completed ≤ High school 77.4 (74.6, 80.3) 74.4 (69.7, 79.1) 83.8 (77.0, 90.6) 82.1 (75.1, 89.2) 69.2* (61.5, 76.9) 80.5 (69.1, 92.0) 76.8 (72.0, 81.7) Any college or technical school 77.1* (74.2, 80.0) 73.4 (67.9, 78.9) 83.2* (78.4, 88.0) 84.6 (79.0, 90.2) 56.3*** (44.4, 68.2) 81.1 (71.3, 90.9) 73.3 (62.8, 83.9) College graduate or higher (Ref) 80.7 (78.8, 82.6) 76.4 (73.0, 79.7) 88.7 (85.8, 91.6) 81.8 (76.5, 87.2) 79.5 (74.0, 85.1) 79.0 (74.5, 83.5) 74.9 (66.2, 83.6) % of time in United States 0–24.99 (Ref) 68.0 (64.9, 71.0) 64.7 (59.7, 69.8) 71.8 (63.9, 79.7) 53.6 (33.8, 73.4) 64.4 (57.4, 71.4) 69.8 (62.9, 76.8) 65.3 (58.3, 72.3) 25–49.99 77.5*** (74.7, 80.4) 72.7* (68.5, 76.8) 85.8** (80.3, 91.2) 82.5** (70.7, 94.3) 68.4 (57.9, 78.9) 84.7** (78.3, 91.1) 74.6* (67.1, 82.0) 50–99.99 83.8*** (81.4, 86.2) 84.1*** (79.1, 89.0) 90.2*** (86.5, 93.9) 83.9** (77.0, 90.8) 82.7*** (76.9, 88.6) 82.3* (74.4, 90.1) 78.8* (71.0, 86.6) 100 87.8*** (85.6, 90.1) 84.9***(80.2, 89.6) 92.1*** (89.1, 95.0) 86.5*** (81.8, 91.2) 89.3** (81.2, 97.5) 93.1*** (88.7, 97.5) 94.6** (88.2, 100.0) English proficiency English proficient (Ref) 79.8 (78.1, 81.6) 74.7 (71.4, 78.0) 86.7 (84.0, 89.3) 82.3 (78.1, 86.5) 73.3 (67.5, 79.0) 79.9 (75.8, 84.0) 81.6 (75.5, 87.8) Limited/no English proficiency 77.1 (74.0, 80.2) 76.2 (71.5, 80.8) 85.2 (73.1, 97.3) 86.6 (77.1, 96.0) 72.9 (68.0, 77.8) 65.4 (44.8, 85.9) 71.6 (65.3, 77.9) Income as a % of federal poverty level (FPL) < 200% FPL 78.2 (75.6, 80.7) 72.4 (67.9, 76.9) 86.5 (81.7, 91.3) 77.2 (66.1, 88.4) 73.3 (65.4, 81.2) 82.8 (74.4, 91.1) 75.5 (69.6, 81.4) ≥ 200% FPL (Ref) 79.5 (77.8, 81.3) 76.6 (73.6, 79.6) 86.6 (83.6, 89.6) 83.9 (79.7, 88.0) 72.9 (68.1, 77.7) 78.6 (74.2, 83.1) 76.4 (69.7, 83.2) Health insurance Uninsured 75.2** (72.2, 78.3) 69.2* (63.5, 75.0) 88.7 (83.4, 94.1) 78.0 (64.8, 91.1) 69.5 (62.8, 76.2) 63.2** (48.3, 78.2) 72.8 (63.9, 81.7) Public only 79.2 (75.1, 83.3) 74.6 (65.4, 83.8) 92.2 (85.2, 99.1) 66.4 (46.9, 85.9) 79.9 (70.2, 89.6) 72.7 (37.5, 100.0) 73.8 (66.8, 80.7) Some private HMO (Ref) 80.7 (78.6, 82.8) 77.2 (73.7, 80.7) 85.9 (81.9, 89.9) 83.3 (77.8, 88.7) 76.6 (69.7, 83.5) 82.8 (77.4, 88.1) 77.6 (70.7, 84.5) Some private non-HMO 78.7 (75.8, 81.7) 76.0 (70.7, 81.4) 84.5 (79.6, 89.3) 86.9 (79.5, 94.4) 69.4 (59.1, 79.8) 80.0 (72.6, 87.3) 83.3 (71.8, 94.8) No. of times saw a physician in past 12 mo 0 65.1*** (61.3, 68.9) 60.8*** (54.9, 66.7) 76.6** (67.7, 85.4) 57.7*** (45.9, 69.5) 58.8** (49.8, 67.8) 71.9** (62.2, 81.7) 60.4*** (49.9, 70.9) 1–2 80.1*** (78.0, 82.1) 76.0** (72.1, 79.8) 86.4 (83.1, 89.7) 88.7 (84.1, 93.3) 74.6* (68.8, 80.4) 75.6** (69.7, 81.6) 78.2 (71.9, 84.4) ≥ 3 (Ref) 84.8 (83.0, 86.6) 82.8 (79.4, 86.3) 89.9 (86.4, 93.5) 88.0 (83.0, 93.0) 81.9 (76.3, 87.4) 87.0 (81.7, 92.3) 80.3 (75.3, 85.3) Has a usual source of care Yes (Ref) 79.9 (78.5, 81.3) 76.0 (73.3, 78.7) 87.5 (84.8, 90.2) 83.4 (79.0, 87.8) 72.8 (67.9, 77.7) 81.6 (77.2, 86.1) 76.0 (71.8, 80.2) No or usual source is emergency department 75.7* (71.7, 79.7) 72.0 (64.9, 79.1) 82.0 (73.5, 90.5) 79.2 (68.3, 90.1) 73.6 (64.4, 82.8) 71.0 (59.1, 83.0) 75.1 (65.7, 84.5) Asian nationality Chinese (Ref) 77.6 (75.4, 79.8) Filipino 83.5** (80.6, 86.3) Japanese 75.0 (70.3, 79.6) Korean 76.8 (72.8, 80.7) South Asian 78.6 (74.8, 82.4) Vietnamese 80.0 (76.5, 83.5) Southeast Asian 78.3 (71.8, 84.7) Other/multiple Asian types 74.8 (68.0, 81.7) TABLE 4—Predictive Margins (PMs) From Multivariate Analyses for Asian American Women Aged 50–74 Years, by Nationality Reporting a Mammogram in the Past 2 Years: California Health Interview Survey (CHIS), 2001–2009
Asian Chinese Filipino Japanese Korean South Asian Vietnamese PM (95% CI) PM (95% CI) PM (95% CI) PM (95% CI) PM (95% CI) PM (95% CI) PM (95% CI) Survey year 2001 (Ref) 75.6 (72.0, 79.3) 70.2 (61.9, 78.6) 78.8 (69.8, 87.8) 80.8 (72.4, 89.2) 63.7 (54.2, 73.1) 86.9 (64.9, 100.0) 72.9 (62.3, 83.4) 2003 78.8 (75.1, 82.4) 73.0 (65.2, 80.8) 83.8 (77.1, 90.5) 86.3 (79.2, 93.5) 69.1 (58.4, 79.7) 74.4 (53.4, 95.4) 78.1 (64.7, 91.5) 2005 80.6 (77.1, 84.0) 81.7* (75.6, 87.9) 81.6 (73.8, 89.3) 83.1 (74.7, 91.5) 66.5 (55.3, 77.6) 90.4 (81.8, 98.9) 80.7 (69.4, 92.0) 2007 81.8* (78.5, 85.1) 83.8** (79.9, 87.6) 83.1 (76.2, 90.1) 86.5 (80.5, 92.5) 63.3 (53.4, 73.2) 76.0 (58.8, 93.2) 82.7 (73.0, 92.4) 2009 81.8* (77.5, 86.0) 81.4* (75.0, 87.7) 84.2 (77.8, 90.6) 93.8** (89.6, 97.9) 63.3 (52.5, 74.1) 73.0 (64.7, 81.2) 92.9*** (89.2, 96.7) Age, y 50–64 (Ref) 80.7 (78.3, 83.0) 77.8 (73.1, 82.5) 84.1 (79.7, 88.5) 86.8 (82.6, 91.0) 67.1 (59.6, 74.6) 76.5 (67.2, 85.8) 83.2 (77.9, 88.4) 65–74 78.2 (74.3, 82.1) 79.3 (71.1, 87.5) 79.1 (72.1, 86.0) 84.7 (78.4, 91.0) 60.8 (49.8, 71.8) 90.4 (70.9, 100.0) 76.3 (65.7, 86.9) Marital status Married or living as married (Ref) 82.0 (80.2, 83.8) 78.8 (75.4, 82.3) 86.2 (82.4, 90.0) 86.1 (81.6, 90.5) 67.0 (60.8, 73.1) 87.1 (77.2, 97.0) 80.1 (74.0, 86.2) Other 74.7*** (70.9, 78.5) 76.5 (70.0, 83.0) 75.5** (68.4, 82.6) 86.1 (81.7, 90.6) 60.1 (51.4, 68.8) 56.1* (31.8, 80.4) 85.6 (78.8, 92.4) Education completed ≤ High school 80.2 (76.6, 83.9) 80.6 (76.4, 84.8) 80.1 (71.1, 89.2) 87.2 (81.6, 92.8) 62.5 (54.8, 70.3) 85.7 (71.3, 100.0) 82.4 (77.4, 87.4) Any college or technical school 78.0 (73.7, 82.3) 71.5 (62.9, 80.1) 84.7 (78.0, 91.4) 86.0 (80.8, 91.3) 65.3 (54.7, 76.0) 82.1 (63.8, 100.0) 74.5 (58.5, 90.4) College graduate or higher (Ref) 80.6 (77.9, 83.2) 78.3 (73.4, 83.1) 82.7 (78.6, 86.8) 85.4 (79.8, 91.0) 70.0 (62.7, 77.3) 76.5 (65.9, 87.1) 83.4 (72.4, 94.4) % of time in the United States 0–49.99 (Ref) 65.0 (45.9, 84.2) 0–24.99 (Ref) 70.9 (65.7, 76.1) 72.3 (65.8, 78.9) 65.9 (55.3, 76.6) — 63.7 (53.6, 73.9) 58.9 (34.4, 83.5) 80.1 (72.4, 87.7) 25–49.99 80.6*** (77.5, 83.7) 81.5* (76.8, 86.3) 84.5** (79.0, 90.0) — 63.0 (55.6, 70.5) 95.6*** (89.3, 100.0) 83.3 (76.3, 90.4) 50–99.99 82.6*** (79.7, 85.5) 76.0 (69.1, 82.9) 87.4*** (82.7, 92.1) 83.0* (75.5, 90.4) 69.7 (59.6, 79.9) 83.8 (74.2, 93.4) 83.9 (73.7, 94.1) 100 86.2*** (81.9, 90.4) 86.1** (79.7, 92.5) 83.5* (73.9, 93.0) 89.7*** (86.0, 93.4) 79.6 (55.2, 100.0) 86.5 (60.3, 100.0) — English proficiency English proficient (Ref) 81.0 (78.4, 83.6) 79.7 (74.7, 84.7) 82.7 (78.5, 87.0) 85.2 (81.0, 89.4) 69.4 (59.8, 79.0) 79.9 (72.7, 87.1) 85.5 (75.0, 96.1) Limited/no English proficiency 78.4 (75.0, 81.7) 77.1 (72.3, 82.0) 80.1 (67.3, 92.9) 91.9 (84.9, 98.8) 64.1 (57.3, 70.8) 74.4 (31.5, 100.0) 81.2 (75.9, 86.6) Income as a % of federal poverty level (FPL) < 200 FPL 79.6 (76.8, 82.5) 75.7 (70.0, 81.5) 82.7 (76.6, 88.8) 86.7 (79.7, 93.7) 60.9 (52.5, 69.2) 81.6 (62.9, 100.0) 83.8 (79.2, 88.4) ≥ 200 FPL (Ref) 80.2 (77.7, 82.7) 80.3 (76.4, 84.1) 82.4 (77.7, 87.1) 86.0 (82.1, 89.9) 68.9 (61.1, 76.7) 78.8 (70.6, 86.9) 75.8 (62.9, 88.7) Health insurance Uninsured 69.6*** (63.7, 75.6) 71.3 (60.6, 81.9) 63.5** (43.5, 83.6) 36.6*** (6.9, 66.2) 58.9 (48.1, 69.7) 70.3 (48.9, 91.8) 77.3 (68.4, 86.3) Public only 79.6 (75.2, 83.9) 77.0 (67.7, 86.3) 81.1* (74.2, 88.1) 90.6 (84.6, 96.6) 68.1 (56.4, 79.8) 73.0 (39.2, 100.0) 79.9 (71.7, 88.2) Some private HMO (Ref) 84.5 (82.1, 86.9) 81.7 (76.5, 86.8) 89.3 (85.4, 93.3) 89.2 (84.9, 93.5) 68.7 (57.6, 79.7) 82.9 (74.9, 90.9) 86.0 (76.4, 95.7) Some private non-HMO 75.6** (68.6, 82.7) 75.5 (65.5, 85.5) 73.8** (61.9, 85.7) 75.2* (59.2, 91.3) 71.5 (54.5, 88.5) 80.6 (62.4, 98.8) 90.8 (79.6, 100.0) No. of physician visits in past 12 mo 0 60.8*** (54.4, 67.2) 64.0*** (54.8, 73.2) 74.4** (63.3, 85.4) 44.6*** (26.3, 62.8) 32.2*** (17.9, 46.6) 66.6* (45.6, 87.6) 60.6*** (42.6, 78.5) 1–2 77.9*** (74.6, 81.3) 79.1 (74.1, 84.2) 76.9*** (70.6, 83.3) 89.7 (85.3, 94.1) 62.9** (53.0, 72.7) 73.1* (60.3, 85.8) 82.9 (74.4, 91.4) ≥ 3 (Ref) 86.5 (84.5, 88.5) 83.4 (79.0, 87.7) 88.9 (84.9, 92.8) 92.7 (89.0, 96.5) 78.1 (72.2, 83.9) 93.6 (84.7, 100.0) 86.5 (82.1, 91.0) Has a usual source of care Yes (Ref) 80.8 (78.9, 82.8) 80.6 (77.4, 83.8) 82.9 (78.7, 87.1) 85.5 (81.5, 89.4) 65.5 (59.0, 72.0) 78.4 (70.8, 86.0) 83.6 (78.9, 88.3) No or usual source is emergency department 73.2* (66.8, 79.5) 60.9** (45.2, 76.6) 78.5 (67.2, 89.8) 91.7 (85.1, 98.2) 64.1 (52.6, 75.7) 85.6 (70.5, 100.0) 68.9* (53.8, 84.1) Asian nationality Chinese (Ref) 80.3 (77.3, 83.3) Filipino 79.5 (75.7, 83.4) Japanese 79.2 (73.7, 84.6) Korean 72.1** (67.2, 77.0) South Asian 78.6 (71.6, 85.6) Vietnamese 86.3* (82.7, 90.0) Southeast Asian 78.2 (65.1, 91.3) Other/multiple Asian types 83.3 (73.7, 92.9)Table 4 presents the multivariate results for mammography receipt, which indicate an upward trend over time, with increasing rates of mammography for Asian American women overall and Chinese, Japanese, and Vietnamese women between 2001 and 2009. In 2009, Japanese (93.8%) and Vietnamese (92.9%) women had the highest rates of mammography receipt (83.9%), whereas Korean (63.3%) and South Asian (73.0%) women had the lowest after controlling for other factors. Among Asian American women in aggregate, Korean women were less likely than Chinese women to receive mammograms (72.1% vs 80.3%; P < .01), whereas Vietnamese women were more likely (86.3% vs 80.3%; P < .05). Unmarried Asian women in aggregate, Filipino women, and South Asian women were less likely to receive mammograms compared with married women. Neither education nor income level was a significant predictor of mammography receipt. Women who reported being in the United States for longer reported higher mammography rates, except Korean, South Asian, and Vietnamese women. Compared with women with 3 or more physician visits in the past year, Asian women with fewer visits had significantly lower rates of mammography. Notably, uninsured Asian women overall, Filipino women, and Japanese women were less likely to report receiving a mammogram; additionally, Asian women overall, Chinese women, and Vietnamese women without a usual source of care were also less likely to receive mammograms in the past 2 years.
DISCUSSIONThe purpose of our study was to examine patterns of cervical and breast cancer screening among Asian American women in California and to assess their screening trends over time. Our findings indicate that rates of Pap test receipt were consistently below Healthy People 2020 objectives for all Asian nationalities and that mammography receipt was below these objectives for certain groups. In examining trends over time, we found no significant changes in Pap test rates, but mammography use increased among Asian American women overall, especially among Chinese and Vietnamese women. Our study confirmed striking variation among Asian American groups, with screening rates varying from 64.7% to 93.6% for mammography receipt in 2009 and from 77.5% to 85.5% for Pap test use in 2007. Notably, Chinese and Korean women had the lowest Pap test rates in 2007, and Korean and South Asian women had the lowest mammography rates in 2009. Therefore, these groups should be the focus of culturally based targeted interventions or programs to promote breast and cervical cancer screening among Asian American women.
Several potential factors could help explain the variation among Asian women that we observed. One of the most consistent findings among all Asian American nationalities was that longer time in the United States was associated with increased use of cancer screening for both Pap test and mammography receipt. Surprisingly, English language proficiency was not associated with screening use. This finding could be associated with California’s relatively high proportion of in-language services offered in Asian languages, particularly in areas with a high population density of Asian Americans. Within Asian ethnic enclaves, such as Koreatown or Chinatown, for example, services are offered by health care providers of the same cultural and linguistic background.54,55 Longer time in the United States may result in changing knowledge and attitudes about the use of preventive health care services. This could help explain why increased length of time in the United States was significantly associated with higher screening rates among nearly all Asian American groups, despite lower English proficiency. More time in the United States may also enable women to have better access to and navigation of the US health care system.
Sociodemographic and health care access factors had varied effects, with education and insurance coverage being significant predictors of screening for certain groups but not all. Contrary to expectations, income level and usual source of care did not appear to play a significant role for most women. For most groups, neither income nor insurance was a significant predictor of mammography or Pap test receipt in multivariate results. This finding is likely a result of a combination of factors. First, California State policies and programs may increase access to screening among Asian women independent of health insurance coverage. The Every Woman Counts program, cofunded by the Centers for Disease Control and Prevention and the state of California, provides in-language services to uninsured and lower-income Asian women and has more generous eligibility criteria in California than in other states.56 The Every Woman Counts program is supplemented by funding for emergency treatment from the federal government and the California tobacco tax, sometimes in collaboration with ethnic community groups. Second, a long-standing network of community-based organizations in California provide primary care services and focus outreach and education efforts on Asian Americans.41,42,57–59 Community-based organizations also work closely with California’s state-based and federal programs as well as ethnic-specific clinics to improve preventive care among Asian individuals in California. Third, several community-based organizations and clinics provide in-language navigation services, which have also been shown to improve preventive care use. Finally, California has a large HMO penetration rate, and HMOs promote the use of preventive care services.60–62 Given this context, California’s state policies and organizations targeting screening in Asians may help explain relative improvements in screening for particular Asian groups (e.g., mammography among Vietnamese women) that we found in our study. Therefore, California may provide a model of health care delivery for increasing cancer screening use among underserved ethnic populations locally and in other states. However, it is important to note that despite the availability of a variety of screening and education programs in California, we still found low and variable rates of cancer screening for breast and cervical cancers, suggesting that underlying cultural attitudes and beliefs may play a stronger role than has been shown to date.
Our study had some limitations that should be noted. First, response rates ranged from 37.7% to 17.7% during the cycles of CHIS included in our analyses and may affect generalizability of findings.49,50 However, these rates are comparable to other random-digit-dial samples during the same period, such as the Behavioral Risk Factor Surveillance System.63,64 Second, the CHIS did not include measures of culturally framed health beliefs and attitudes, which may play an important role in screening behaviors and could explain some of the variation observed between groups. Thus, we were limited in our ability to explain how cultural views may affect screening for breast and cervical cancer among Asian American women. Third, for Asian women who were not offered the survey in their native language, the sample captured was more likely to include English-speaking, acculturated women who also may be more likely to be screened. Fourth, data were self-reported. Last, data were cross-sectional, prohibiting us from making causal inferences.
Despite these limitations, our study had several strengths. Importantly, these findings were based on several years of pooled data that provided larger sample sizes than in other data on Asian Americans. Unlike other population-based surveys, the CHIS was conducted in several Asian languages, which helped capture a larger, more representative sample of Asian Americans in California. CHIS data also included several important measures of sociodemographic, acculturation, and health care access factors, enabling us to examine the role of these factors in screening use among Asian American women. Finally, by measuring differences in survey year with the pooled data, we were able to detect an increase in mammography use among Asian women between 2001 and 2009.
In conclusion, our findings have several implications and directions for future research. Population-based surveys should incorporate measures of health beliefs and attitudes toward health, prevention of disease, health care use, and the capacity to navigate the complex health care system to better address barriers to screening among Asian Americans. Additional research that uses diverse methodological approaches, such as qualitative studies, is needed to tease out the specific barriers to screening for Asian women, including structural barriers such as transportation and out-of-pocket costs, cultural and health beliefs, attitudes toward cancer, and understanding of the basic message that screening use may help prevent the onset of disease. Studies are also needed to assess the cultural sensitivity of health care practitioners and to better understand the role of patient–provider interaction and trust in providers among Asian American women. Perhaps most important, our results underscore the need for targeted interventions to increase rates of mammography and Pap test use among specific Asian nationalities.
AcknowledgmentsThe authors would like to acknowledge the support of the Cancer Prevention Fellowship Program for this research.
Findings from this study have been presented at the American Public Health Association Research Meeting, October 27–31, 2012, San Francisco, CA.
Human Participant ProtectionHuman subjects’ approval for recruitment and data collection for the California Health Interview Survey was obtained from UCLA and the state of California. Therefore, the survey was exempted from review by the National Institutes of Health, Office of Human Subjects Research Protection.
ReferencesRetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.3