Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jul 8.
Published in final edited form as: Cancer Epidemiol Biomarkers Prev. 2026 Jul 1;35(7):1139–1150. doi: 10.1158/1055-9965.EPI-25-1718

Ethnic enclave residence and breast, cervical, and colorectal cancer incidence among Asian American adults, 2006–2017

Alison J Canchola 1,2, Sandi L Pruitt 3,4, Katherine Lin 1,2, Francis P Boscoe 5, Alice Guan 1, Kevin A Henry 6,7, Robert A Hiatt 1,8, Amy E Hughes 3,4, Tabassum Z Insaf 9,10, Paulo S Pinheiro 11, Aniruddha B Rathod 3,12, Kathryn Shahan 3, Antoinette M Stroup 13, Hong Zhu 14, Scarlett L Gomez 1,2,8, Salma Shariff-Marco 1,2,8
PMCID: PMC13340256  NIHMSID: NIHMS2171820  PMID: 42029124

Abstract

Background

Ethnic enclaves are culturally or ethnically distinct neighborhoods with high concentrations of individuals with shared ethnic origin, immigrants, and/or ethnic-specific businesses. We examined if residence in Asian American enclaves was associated with incidence of three screening-amenable cancers, using a composite enclave index across 5 US states.

Methods

Using cancer registry data, we identified 74,485 breast, 4,134 cervical, and 35,736 colorectal cancer tumors diagnosed in California, Florida, New Jersey, New York, and Texas from January 1, 2006, through December 31, 2017, in Asian American adults. Age-adjusted incidence rates and incidence rate ratios (IRR) were calculated for 2006–2011 and 2012–2017 for census tract (CT)-level enclave and poverty measures, separately and jointly.

Results

Breast and colorectal cancer incidence rates were lower among Asian American adults residing in the least culturally distinct neighborhoods compared to the most distinct (2012–2017: IRR 0.71, 95% confidence interval [CI]: 0.65–0.76 in quintile 1 [Q1; low] compared to Q5 [high] for breast; IRR 0.62, 95% CI: 0.52–0.74 in Q1 vs Q5 for colorectal cancer in males), while no association was found with enclave for cervical cancer. Higher cervical cancer rates were associated with high-poverty areas. For breast cancer, the highest incidence rates were observed in low-poverty enclaves, whereas for colorectal cancer, the highest rates were observed in high-poverty enclaves.

Conclusions

Breast and colorectal cancer incidence rates were associated with residence in Asian American enclaves and CT poverty, though patterns differed by cancer site.

Impact

Cancer prevention outreach may be beneficial within Asian American enclaves.

Introduction:

Ethnic enclaves are culturally or ethnically distinct neighborhoods with high concentrations of individuals with the same ethnic origin, additionally characterized by linguistic isolation, a large share of recent immigrants, and/or ethnic specific businesses and resources.1,2 Asian American individuals often reside in ethnic enclaves, which have been hypothesized to affect health outcomes.3–7 Enclaves may negatively impact health and health behaviors through attributes of disadvantaged neighborhoods (e.g., low access to healthcare, poor street conditions, high traffic density, low safety). Conversely, there are multiple pathways3–7 through which residing in an ethnic enclave could promote health, including: culturally and linguistically relevant information and resources which could facilitate access to healthcare; social support and cultural norms that may support health promoting behaviors (e.g., physical activity such as tai chi or yoga, ethnic foods prepared in traditional ways); reduced exposure to discrimination which could limit the use of unhealthy coping behaviors (e.g., smoking and drinking) and reduce levels of stress; and access to economic opportunities through small businesses or networks. For this study, we chose five states (CA, FL, NJ, NY, TX) with large and/or rapidly growing populations of Asian Americans; as of 2010 these states collectively accounted for 57% of the Asian American population in the US. In an ecologic study, we showed Asian American enclave neighborhoods in these five states to be more metropolitan, have less poverty, fewer uninsured people, and higher primary care accessibility than non-enclave neighborhoods.8

Cancer is the leading cause of death in Asian American individuals.9,10 Asian American enclaves and neighborhoods with high Asian ethnic density have been associated with cancer incidence, mortality, and screening among Asian American adults, with residence in enclaves shown to be both beneficial and detrimental, depending on the cancer site, outcome, or the study.11–13 For example, for Asian American individuals, residing in an ethnic enclave was associated with higher incidence rates of cervical cancer and lower incidence rates of colorectal cancer; with higher risk of breast cancer in high socioeconomic status (SES) enclaves.13 Mixed findings could be due to differing risk factors for different cancer sites, or varying operationalization of enclave measures and geographic scales used, or differences in the sociodemographic composition or social, physical, and built environment characteristics of enclaves across different geographic regions of the United States (US). Moreover, many existing studies about ethnic enclaves are focused on only one state.13 In addition, neighborhood socioeconomic status (nSES), and its correlation and joint effects with enclave status, plays a role in cancer incidence.14,15

To address these limitations, we used the same measure of ethnic enclave across five states applied to three cancer sites with evidence-based screening guidelines from the United States Preventive Services Task Force, to explore whether residence in enclaves is associated with cancer incidence, separately and jointly with neighborhood poverty. Examining enclave associations with incidence rates across multiple cancer sites will inform which may be more or less related to enclave residence, contribute to generation of mechanistic hypotheses, and guide outreach and screening promotion in areas that need it most.

Materials and Methods

Study population

We identified cancer cases using data from five population-based state cancer registries: California (CA), Florida (FL), New Jersey (NJ), New York (NY), and Texas (TX). These states comprised 57.4% of the Asian American population in the US in 2010. Cancer registries provided data on demographics (age at diagnosis, sex, race and ethnicity), tumor characteristics (site, sequence number, behavior, histology), and residential neighborhood at diagnosis (addresses were geocoded, assigned to 2010 census tracts [CT], and characterized by geocoding precision for census tract assignment).

Data were obtained for Asian American adults (age ≥20 at diagnosis) with in situ or invasive breast cancer (females only), invasive cervical cancer (females only), or in situ or invasive colorectal cancer (males and females) from January 1, 2006, to December 31, 2017, with a known cancer sequence number. Cancer sites were identified using International Classification of Diseases for Oncology, 3rd edition (ICD-O-3) site codes C50.0–50.9 (breast), C53.0–C53.9 (cervical), and C18.0–C18.9, C19.9, C20.9, and C26.0 (colorectal), excluding leukemias and lymphomas (histology codes 9050–9055, 9140, and 9590–9992). We did not include in situ cervical cancers as they are not consistently captured in registries.16 We limited to cases diagnosed in 2006 or later and age 20 or older (instead of age 18 or older) to correspond to available population denominator data. For people with multiple tumors during the study period, all eligible tumors were requested. For each person, we excluded tumors of the same site (treating colon and rectum as one site) that were diagnosed within a year. This removed multiple tumors found on the same day (e.g., during a colonoscopy) or within a short time frame that could have represented the same index case. This approach ensured that the cancer count was enumerated consistently with the population data (where each person was counted once per year) and thus the age-adjusted incidence rates were not artificially inflated. After excluding tumors at the same site within one year of an index tumor; with a residential address at diagnosis that was incomplete, not geocodable, geocoded with poor precision, or erroneously geocoded to a county/CT combination that did not exist; or residing in a CT where the Asian American enclave index was missing (Figure 1), the final analytic sample included n=74,485 breast tumors (in n=72,828 females), n=4,134 cervical tumors (n=4,119 females), n=17,158 colorectal tumors in females (n=16,958 females), and n=18,578 colorectal tumors in males (n=18,332 males). In the analysis, 2.2% of breast cancer patients, 0.4% of cervical cancer patients, and 1.3% of colorectal cancer patients had multiple primary cancers of the same site (over one year apart).

Figure 1.

Figure 1.

Open in a new tab

Hierarchical exclusions for analysis for tumors diagnosed in Asian American adults (age >=20 years); 2006–2017; in California, Florida, New Jersey, New York, and Texas. The figure shows hierarchical exclusion numbers separately for female breast, female cervical, female colorectal, and male colorectal cancer tumors diagnosed in Asian American adults 2006–2017 in 5 states. Numbers originally available and after exclusions are also provided at the person-level, and percent eligible for analysis is provided at both the tumor- and person-level.

We identified Asian American individuals as those with Asian American race or ethnic group (Asian Indian, Chinese, Filipino, Hmong, Japanese, Kampuchean [Cambodian], Korean, Laotian, Pakistani, Thai, Vietnamese, or Other Asian or Asian not otherwise specified [NOS]) in any race variable (up to five races recorded in CA, NJ, and NY, up to two races in TX, and one race in FL), regardless of Hispanic ethnicity (i.e., they could also be Hispanic). Race and ethnicity data were extracted from patient medical records and categorized according to registry classification systems.17 In CA, NJ, NY, and TX, race and ethnicity was supplemented with the North American Association of Central Cancer Registries’ Asian Pacific Islander Identification Algorithm (NAPIIA).18 This algorithm classifies Asian American/Native Hawaiian/Pacific Islander (AANHPI) cases identified as Other or NOS into a specific ethnic group (e.g., Chinese) using factors such as race, Hispanic ethnicity, birthplace, and names. In this sample (excluding FL), 16.0% of cases were originally coded as Other Asian or Asian NOS; after applying the algorithm, 10.0% of the overall sample (62.4% of those originally classified as Other/NOS) were reclassified to a specific Asian ethnic group while 6.0% remained coded as Other/NOS.

Neighborhood variables

The Asian American enclave index was derived using the first component score from a principal component analysis of four CT-level variables pooled across all five states: proportion Asian American, proportion non-US born Asian American, proportion with limited English proficiency who speak Asian or Pacific Islander languages, and proportion linguistically isolated households where Asian or Pacific Islander languages are spoken (proportions defined using the total population or households of the CT, with age >=5 for language proficiency). The index was calculated using data and population numbers from ACS 2008–2012 for 2010 CTs. 1.3% of census tracts across the 5 states were missing an enclave index value, primarily due to suppression of small population counts per the US Census Bureau standards. Details of the development of the Asian American enclave index and a description of the social and built environment characteristics within these enclaves have been described elsewhere.8,19 The Asian American enclave index was categorized into quintiles, where the first quintile (Q1) represented census tracts with low Asian American enclave index score (i.e., least culturally or ethnically distinct neighborhoods) and the fifth quintile (Q5) represented high Asian American enclave index score (i.e., most culturally or ethnically distinct neighborhoods). We also dichotomized the index into ‘enclave’ and ‘non-enclave’ status. An enclave was defined as CTs in the fourth (Q4) and fifth (Q5) quintile with an Asian American population greater than 250 individuals; with eligible Q4 enclaves additionally required to be adjacent to an eligible Q5 enclave. An enclave trajectory measure was defined based on the enclave status of 2010 CTs compared to their enclave status in 2000, and was categorized as: never enclave (non-enclave in both 2000 and 2010), former enclave (enclave in 2000 but not in 2010), persistent enclave (enclave in both 2000 and 2010), and emergent enclave (enclave in 2010 but not in 2000). To determine the trajectory when CT boundaries changed, the boundaries for 2000 and 2010 were spatially intersected and the 2010 CT was assigned the category that described greater than 50% of its area.

We defined high-poverty CTs as those with >=20% of households with income less than the Federal Poverty Level using five-year ACS estimates from 2008–2012 and 2010 CTs. We used the Federal Poverty Level to ensure consistency across the five states.

Population data

Annual population estimates of the number of non-Hispanic AANHPI individuals in each CT by sex and five-year age groups to >=85 were obtained from the National Cancer Institute via Woods & Poole,20 based on Census 2000 and 2010 data, using inter-censal estimates for 2006–2009 and post-censal estimates for 2010–2017. At the time of the current analysis, annual population-level denominator data by CT, sex, and age group were not available for Asian American separately from Native Hawaiian/Pacific Islander, nor for disaggregated Asian American ethnic groups.

Statistical analysis

We calculated annual age-adjusted incidence rates (AAIR) per 100,000, standardized to the 2000 US standard million population, and 95% confidence intervals (CI) using the Tiwari modification.21,22 We calculated AAIRs using CT-, sex-, and age group- (5-year age groups from 20–24 to >=85) specific cancer counts and population data for Asian American individuals, separately for two time-periods (2006–2011 and 2012–2017), for each cancer site, and for males and females for colorectal cancer. We used these two time periods (rather than looking at one 12-year period) to examine trends over time and to be more comparable with other published studies.

AAIRs were presented across Asian American enclave index quintiles, enclave status, poverty (<20% or >=20% below the Federal Poverty Level), jointly by enclave status and poverty, and enclave trajectory. We conducted tests of linear trend across Asian American enclave index quintile AAIRs using weighted linear regression where the dependent variable was the AAIR, the independent variable was an ordinal variable set to the median value of the enclave index within each quintile (Q1: −1.35663, Q2: −0.63602, Q3: −0.11118, Q4: 0.52978, Q5: 1.38291), and the weight was the inverse of the variance of the AAIR.23 We calculated incidence rate ratios (IRRs) by dividing the AAIR for each category by the AAIR for the reference category and computed 95% CIs.24

As a sensitivity analysis, we calculated AAIRs and IRRs for invasive breast cancer (excluding in situ) because almost a quarter of the cases were diagnosed at the in situ stage, compared with only 4% of colorectal cancer cases. As another sensitivity analysis, we calculated state-stratified AAIRs and IRRs across enclave index quintiles, using the same enclave index and quintile cut-points as the primary analysis (i.e., pooled across the five states; not state-specific). As a secondary analysis of breast and colorectal cancer incidence rates, we calculated AAIRs and IRRs across enclave quintiles and enclave status stratified by age at diagnosis (<50 vs. ≥50 years) to determine whether associations differed by screening-eligible versus typically unscreened age groups; and compared IRRs and CIs for non-enclaves between age <50 and ≥50 years.

Analyses were performed in SAS 9.4 (RRID: SCR_008567) and statistical significance was p<0.05.

Data availability

The ethnic enclave measures for recent years are available to the public on the UCSF Health Atlas website (see https://healthatlas.ucsf.edu/). The registry data are not publicly available due to data confidentiality concerns and can be requested via state cancer registries.

Results

Table 1 shows the distribution of tumors among Asian American adults included in the analysis for each cancer site, sex, and year group. The majority of tumors were diagnosed in CA (59–69%), with 16–22% in NY, and <5% in FL. For breast cancer, 23–24% of tumors were in situ and 4% were diagnosed at distant stage; for cervical, 12–14% were diagnosed at distant stage; and for CRC, 4% were in situ and 17–19% were diagnosed at distant stage. For all sites, 75% or more of tumors were in people who lived in Asian American enclaves at the time of diagnosis. More than a quarter (26–27%) of cervical cancer tumors were in people who lived in high-poverty neighborhoods, compared to 15–16% of breast, and 21–22% of CRC. Supplementary Table S1 shows the distribution of these factors by Asian American enclave index quintiles. Across cancer sites and time-periods, 71–78% of tumors diagnosed in CA were in people who resided in the highest quintile of Asian American enclave index, 36–70% in NJ, NY, and TX, and 5–8% in FL (Supplementary Table S1).

Table 1.

Distribution of tumors in Asian American adults (age >=20) by cancer site, sex, and year of diagnosis; 2006–2017; in California, Florida, New Jersey, New York, and Texas

Cancer site, Sex
Breast, Female Cervical, Female Colorectal, Female Colorectal, Male
2006–2011 2012–2017 2006–2011 2012–2017 2006–2011 2012–2017 2006–2011 2012–2017
N Col Pct N Col Pct N Col Pct N Col Pct N Col Pct N Col Pct N Col Pct N Col Pct
All 31521 100 42964 100 1916 100 2218 100 8154 100 9004 100 8439 100 10139 100
State 20528 65.1 26427 61.5 1177 61.4 1301 58.7 5605 68.7 5785 64.2 5704 67.6 6323 62.4
 California
 Florida 1107 3.5 1905 4.4 85 4.4 98 4.4 280 3.4 359 4 260 3.1 387 3.8
 New Jersey 2425 7.7 3395 7.9 121 6.3 115 5.2 436 5.3 533 5.9 489 5.8 648 6.4
 New York 5187 16.5 7480 17.4 390 20.4 482 21.7 1312 16.1 1551 17.2 1410 16.7 1928 19
 Texas 2274 7.2 3757 8.7 143 7.5 222 10 521 6.4 776 8.6 576 6.8 853 8.4
Age at diagnosis 2399 7.6 2764 6.4 339 17.7 402 18.1 313 3.8 353 3.9 268 3.2 347 3.4
 20–39
 40–49 8013 25.4 9992 23.3 508 26.5 544 24.5 850 10.4 927 10.3 818 9.7 944 9.3
 50–59 9054 28.7 11475 26.7 457 23.9 512 23.1 1741 21.4 1936 21.5 1952 23.1 2391 23.6
 60–69 6873 21.8 11014 25.6 302 15.8 407 18.3 1812 22.2 2263 25.1 2202 26.1 2969 29.3
 70–79 3700 11.7 5597 13 191 10 218 9.8 1821 22.3 1763 19.6 1983 23.5 2114 20.9
 80+ 1482 4.7 2122 4.9 119 6.2 135 6.1 1617 19.8 1762 19.6 1216 14.4 1374 13.6
Summary stage 7386 23.4 10122 23.6 0 0 0 0 339 4.2 352 3.9 368 4.4 428 4.2
 In situ
 Localized 14878 47.2 21006 48.9 842 43.9 947 42.7 2890 35.4 3056 33.9 3129 37.1 3514 34.7
 Regional 7423 23.5 9399 21.9 723 37.7 803 36.2 2942 36.1 3199 35.5 2822 33.4 3509 34.6
 Remote 1129 3.6 1550 3.6 238 12.4 309 13.9 1421 17.4 1702 18.9 1516 18 1902 18.8
 Unknown 705 2.2 887 2.1 113 5.9 159 7.2 562 6.9 695 7.7 604 7.2 786 7.8
Asian American enclave index,a census tract 505 1.6 728 1.7 39 2 47 2.1 131 1.6 203 2.3 119 1.4 151 1.5
 Quintile (Q) 1 (low/least culturally distinct)
 Q2 1338 4.2 1833 4.3 99 5.2 112 5 317 3.9 413 4.6 299 3.5 360 3.6
 Q3 2964 9.4 4294 10 167 8.7 212 9.6 659 8.1 785 8.7 692 8.2 825 8.1
 Q4 6001 19 8593 20 358 18.7 434 19.6 1319 16.2 1615 17.9 1364 16.2 1738 17.1
 Q5 (high/most culturally distinct) 20713 65.7 27516 64 1253 65.4 1413 63.7 5728 70.2 5988 66.5 5965 70.7 7065 69.7
Asian American enclave status,b census tract 7401 23.5 10553 24.6 461 24.1 563 25.4 1728 21.2 2191 24.3 1782 21.1 2155 21.3
 Non-enclave
 Enclave 24120 76.5 32411 75.4 1455 75.9 1655 74.6 6426 78.8 6813 75.7 6657 78.9 7984 78.7
Percent poverty,c census tract 4876 15.5 6512 15.2 508 26.5 578 26.1 1733 21.3 1958 21.7 1856 22 2226 22
 High poverty (>=20%)
 Low poverty (<20%) 26645 84.5 36452 84.8 1408 73.5 1640 73.9 6421 78.7 7046 78.3 6583 78 7913 78
Enclave by poverty, census tract 1338 4.2 1722 4 125 6.5 154 6.9 403 4.9 469 5.2 433 5.1 499 4.9
 High poverty, Non-enclave
 High poverty, Enclave 3538 11.2 4790 11.1 383 20 424 19.1 1330 16.3 1489 16.5 1423 16.9 1727 17
 Low poverty, Non-enclave 6063 19.2 8831 20.6 336 17.5 409 18.4 1325 16.2 1722 19.1 1349 16 1656 16.3
 Low poverty, Enclave 20582 65.3 27621 64.3 1072 55.9 1231 55.5 5096 62.5 5324 59.1 5234 62 6257 61.7
Enclave trajectory,d 2010 census tracts 6196 19.7 8964 20.9 392 20.5 490 22.1 1453 17.8 1865 20.7 1487 17.6 1818 17.9
 Never enclave
 Former enclave 1204 3.8 1584 3.7 69 3.6 73 3.3 274 3.4 326 3.6 295 3.5 337 3.3
 Persistent enclave 21757 69 28781 67 1329 69.4 1455 65.6 5888 72.2 6150 68.3 6092 72.2 7176 70.8
 Emergent enclave 2363 7.5 3630 8.4 126 6.6 200 9 538 6.6 663 7.4 565 6.7 808 8
 Enclave index unknown in 2000 <11 <11 0 0 0 0 <11 0 0 0 0 0 0
Open in a new tab
a

Asian American enclave index developed at the census tract level using principal component analysis of 4 census tract level variables. Quintiles based on census tracts pooled across the 5 states. Quintile cut-points of the enclave index: Q1: −1.35663 to <−0.96913, Q2: −0.96913 to <−0.38415, Q3: −0.38415 to <0.19711, Q4: 0.19711 to <0.89552, Q5: 0.89552 to 3.80902

b

Asian American enclave status defined as census tracts within the two highest quintiles of the enclave index (Q4 or Q5) with an Asian American population >250; Q4 enclaves were additionally required to be adjacent to an eligible Q5 enclave census tract

c

Neighborhood percent poverty defined as percent of households in the census tract below the Federal Poverty Level

d

Asian American enclave trajectory defined based on enclave status of 2010 census tracts compared to enclave status in 2000, and categorized as: never enclave (non-enclave in both 2000 and 2010), former enclave (enclave in 2000 but not in 2010), persistent enclave (enclave in both 2000 and 2010), and emergent enclave (enclave in 2010 but not in 2000)

Col Pct, column percent; Q, quintile

Supplementary Table S2 shows the distribution of tumors by Asian American ethnic group (among the four states with available data). In all, 30–32% of CRC tumors and 23–27% of breast and cervical tumors were diagnosed in Chinese American individuals. Within Asian American ethnic groups, the percent of cancer cases residing in Asian American enclaves varied from 86–90% of Chinese American cases to 63–72% of Asian Indian/Pakistani American cases (Supplementary Table S2).

Breast cancer

Breast cancer incidence rates were lower among Asian American adults residing in neighborhoods which were less culturally distinct compared to those residing in neighborhoods which were more culturally distinct, with a gradient observed across quintiles (AAIR p-trend=0.01 in 2006–2011 and p-trend=0.06 in 2012–2017, although the latter did not reach statistical significance; Table 2). Incidence rates were lower in non-enclaves compared to enclaves in both time-periods (for example, IRR 0.88, 95% CI 0.86–0.90 in 2012–2017). Incidence rates were also lower in neighborhoods with high poverty compared to low poverty in both time-periods (for example, IRR 0.84, 95% CI 0.81–0.86 for high-poverty compared to low-poverty neighborhoods in 2012–2017). Looking at the joint association of Asian American enclave status and poverty, the associations of enclave (lower rates in non-enclaves) and poverty (lower rates in high-poverty) can be seen in both time-periods (Figure 2 showing AAIRs per 100,000), although the enclave association differed by poverty level and time-period. In 2012–2017, breast cancer incidence rates were lowest in high-poverty non-enclaves (IRR 0.70, 95% CI 0.67–0.74) and highest in low-poverty Asian American enclaves (reference group). Former enclaves had incidence rates which were similar to persistent enclaves; whereas for emergent enclaves the rate was somewhat lower (IRR=0.95, 95% CI 0.92–0.98 in 2012–2017), although not as low as for neighborhoods that were never enclaves (IRR=0.85, 95% CI 0.83–0.87 in 2012–2017). Limiting to invasive breast cancer (excluding in situ), patterns were similar but results slightly attenuated (Supplementary Table S3).

Table 2.

Age-adjusted incidence ratesa and incidence rate ratios with 95% confidence intervals for Asian American adults (age >=20) by cancer site, sex, year, Asian American enclave measures, and neighborhood poverty; 2006–2017; in California, Florida, New Jersey, New York, and Texas

Cancer site, Sex Year Measure Populationb Tumorsc AAIR LCI UCI IRR LCI UCI p-trendd
Breast, Female Asian American enclave index, e census tract
2006–2011  Quintile (Q) 1 (low/least culturally distinct) 430,178 505 83.8 76.5 91.4 0.72 0.66 0.79 0.0133
 Q2 1,032,577 1,338 94.6 89.5 99.9 0.81 0.77 0.86
 Q3 2,142,821 2,964 104.7 100.9 108.6 0.90 0.87 0.94
 Q4 4,126,284 6,001 112.6 109.7 115.5 0.97 0.94 1.00
 Q5 (high/most culturally distinct) 13,043,126 20,713 116.1 114.5 117.7 1.00
2012–2017  Quintile (Q) 1 (low/least culturally distinct) 564,943 728 87.6 81.3 94.2 0.71 0.65 0.76 0.0613
 Q2 1,306,543 1,833 95.7 91.2 100.2 0.77 0.73 0.81
 Q3 2,640,551 4,294 115.5 112.0 119.0 0.93 0.90 0.96
 Q4 5,028,955 8,593 124.2 121.6 126.9 1.00 0.98 1.03
 Q5 (high/most culturally distinct) 15,455,247 27,516 124.2 122.7 125.7 1.00
Asian American enclave status,f census tract
2006–2011  Non-enclave 5,476,951 7,401 101.2 98.9 103.6 0.87 0.85 0.89
 Enclave 15,298,034 24,120 116.4 115.0 117.9 1.00
2012–2017  Non-enclave 6,795,327 10,553 109.6 107.4 111.7 0.88 0.86 0.90
 Enclave 18,200,912 32,411 125.1 123.7 126.5 1.00
Percent poverty,g census tract
2006–2011  High poverty (>=20%) 4,145,466 4,876 92.9 90.3 95.6 0.79 0.77 0.82
 Low poverty (<20%) 16,629,520 26,645 117.1 115.6 118.5 1.00
2012–2017  High poverty (>=20%) 4,768,661 6,512 104.4 101.8 107.0 0.84 0.81 0.86
 Low poverty (<20%) 20,227,578 36,452 124.7 123.4 126.0 1.00
Enclave by Poverty, census tract
2006–2011 High poverty, Non-enclave 1,203,986 1,338 88.6 83.9 93.5 0.73 0.69 0.77
High poverty, Enclave 2,941,479 3,538 94.8 91.7 98.0 0.78 0.75 0.81
Low poverty, Non-enclave 4,272,965 6,063 104.5 101.8 107.3 0.86 0.84 0.89
Low poverty, Enclave 12,356,555 20,582 121.4 119.7 123.1 1.00
2012–2017 High poverty, Non-enclave 1,440,623 1,722 90.1 85.9 94.5 0.70 0.67 0.74
High poverty, Enclave 3,328,038 4,790 110.7 107.5 113.9 0.86 0.84 0.89
Low poverty, Non-enclave 5,354,704 8,831 114.5 112.0 116.9 0.89 0.87 0.91
Low poverty, Enclave 14,872,874 27,621 128.4 126.9 129.9 1.00
Enclave trajectoryh, 2010 census tracts
2006–2011 Never enclave 4,672,609 6,196 99.1 96.6 101.7 0.85 0.82 0.87
Former enclave 815,707 1,204 113.2 106.8 119.8 0.97 0.91 1.03
Persistent enclave 13,577,936 21,757 117.0 115.4 118.6 1.00
Emergent enclave 1,720,099 2,363 112.1 107.5 116.8 0.96 0.92 1.00
2012–2017 Never enclave 5,917,553 8,970 106.8 104.5 109.0 0.85 0.83 0.87
Former enclave 890,002 1,584 128.3 121.9 134.8 1.02 0.97 1.07
Persistent enclave 15,862,645 28,781 125.8 124.3 127.3 1.00
Emergent enclave 2,338,267 3,630 119.3 115.4 123.3 0.95 0.92 0.98
Cervical, Female Asian American enclave index,e census tract
2006–2011  Quintile (Q) 1 (low/least culturally distinct) 430,178 39 6.9 4.9 9.3 0.98 0.69 1.38 0.2250
 Q2 1,032,577 99 6.9 5.6 8.4 0.98 0.79 1.22
 Q3 2,142,821 167 5.7 4.8 6.6 0.81 0.68 0.96
 Q4 4,126,284 358 6.7 6.0 7.4 0.95 0.84 1.07
 Q5 (high/most culturally distinct) 13,043,126 1,253 7.1 6.7 7.5 1.00
2012–2017  Quintile (Q) 1 (low/least culturally distinct) 564,943 47 6.2 4.5 8.1 0.96 0.70 1.30 0.0916
 Q2 1,306,543 112 6.0 4.9 7.1 0.93 0.76 1.13
 Q3 2,640,551 212 5.7 4.9 6.5 0.88 0.76 1.03
 Q4 5,028,955 434 6.3 5.7 6.9 0.98 0.87 1.09
 Q5 (high/most culturally distinct) 15,455,247 1,413 6.4 6.1 6.8 1.00
Asian American enclave status,f census tract
2006–2011  Non-enclave 5,476,951 461 6.2 5.7 6.8 0.89 0.79 0.99
 Enclave 15,298,034 1,455 7.0 6.7 7.4 1.00
2012–2017  Non-enclave 6,795,327 563 5.9 5.4 6.4 0.92 0.83 1.01
 Enclave 18,200,912 1,655 6.5 6.1 6.8 1.00
Percent poverty,g census tract
2006–2011  High poverty (>=20%) 4,145,466 508 9.8 8.9 10.7 1.58 1.42 1.75
 Low poverty (<20%) 16,629,520 1,408 6.2 5.9 6.5 1.00
2012–2017  High poverty (>=20%) 4,768,661 578 9.5 8.7 10.3 1.68 1.52 1.85
 Low poverty (<20%) 20,227,578 1,640 5.7 5.4 5.9 1.00
Enclave by poverty, census tract
2006–2011  High poverty, Non-enclave 1,203,986 125 8.4 6.9 9.9 1.31 1.08 1.59
 High poverty, Enclave 2,941,479 383 10.4 9.3 11.4 1.63 1.44 1.84
 Low poverty, Non-enclave 4,272,965 336 5.7 5.1 6.4 0.90 0.79 1.02
 Low poverty, Enclave 12,356,555 1,072 6.4 6.0 6.7 1.00
2012–2017  High poverty, Non-enclave 1,440,623 154 8.3 7.0 9.7 1.44 1.20 1.71
 High poverty, Enclave 3,328,038 424 10.0 9.1 11.0 1.74 1.55 1.95
 Low poverty, Non-enclave 5,354,704 409 5.4 4.8 5.9 0.93 0.83 1.04
 Low poverty, Enclave 14,872,874 1,231 5.8 5.4 6.1 1.00
Enclave trajectory,h 2010 census tracts
2006–2011  Never enclave 4,672,609 393 6.3 5.6 6.9 0.87 0.77 0.98
 Former enclave 815,707 69 6.4 4.9 8.0 0.89 0.68 1.14
 Persistent enclave 13,577,936 1,329 7.2 6.8 7.6 1.00
 Emergent enclave 1,720,099 126 5.7 4.7 6.7 0.79 0.65 0.96
2012–2017  Never enclave 5,917,553 490 5.9 5.4 6.4 0.91 0.82 1.01
 Former enclave 890,002 73 6.0 4.7 7.5 0.93 0.72 1.18
 Persistent enclave 15,862,645 1,455 6.5 6.1 6.8 1.00
 Emergent enclave 2,338,267 200 6.3 5.5 7.2 0.98 0.83 1.14
Colorectal, Female Asian American enclave index,e census tract
2006–2011  Quintile (Q) 1 (low/least culturally distinct) 430,178 131 25.3 21.0 30.0 0.73 0.60 0.88 0.0088
 Q2 1,032,577 317 25.2 22.4 28.2 0.73 0.64 0.82
 Q3 2,142,821 659 26.6 24.5 28.8 0.77 0.71 0.84
 Q4 4,126,284 1,319 28.7 27.1 30.3 0.83 0.78 0.88
 Q5 (high/most culturally distinct) 13,043,126 5,728 34.6 33.7 35.6 1.00
2012–2017  Quintile (Q) 1 (low/least culturally distinct) 564,943 203 25.3 21.9 29.0 0.92 0.79 1.06 0.0466
 Q2 1,306,543 413 23.1 20.9 25.4 0.84 0.76 0.93
 Q3 2,640,551 785 22.3 20.7 23.9 0.81 0.75 0.88
 Q4 5,028,955 1,615 24.8 23.6 26.0 0.90 0.85 0.95
 Q5 (high/most culturally distinct) 15,455,247 5,988 27.5 26.8 28.2 1.00
Asian American enclave status,f census tract
2006–2011  Non-enclave 5,476,951 1,728 26.9 25.6 28.3 0.80 0.75 0.84
 Enclave 15,298,034 6,426 33.9 33.0 34.7 1.00
2012–2017  Non-enclave 6,795,327 2,191 23.8 22.8 24.8 0.88 0.84 0.93
 Enclave 18,200,912 6,813 27.0 26.4 27.7 1.00
Percent poverty,g census tract
2006–2011  High poverty (>=20%) 4,145,466 1,733 33.2 31.6 34.8 1.04 0.99 1.10
 Low poverty (<20%) 16,629,520 6,421 31.8 31.0 32.6 1.00
2012–2017  High poverty (>=20%) 4,768,661 1,958 29.7 28.3 31.0 1.17 1.11 1.23
 Low poverty (<20%) 20,227,578 7,046 25.3 24.7 25.9 1.00
Enclave by poverty, census tract
2006–2011  High poverty, Non-enclave 1,203,986 403 28.5 25.8 31.4 0.85 0.76 0.94
 High poverty, Enclave 2,941,479 1,330 34.9 33.0 36.8 1.04 0.98 1.10
 Low poverty, Non-enclave 4,272,965 1,325 26.4 24.9 27.9 0.79 0.74 0.84
 Low poverty, Enclave 12,356,555 5,096 33.6 32.6 34.5 1.00
2012–2017  High poverty, Non-enclave 1,440,623 469 24.2 22.0 26.5 0.94 0.85 1.03
 High poverty, Enclave 3,328,038 1,489 31.8 30.2 33.5 1.23 1.16 1.31
 Low poverty, Non-enclave 5,354,704 1,722 23.7 22.6 24.9 0.92 0.87 0.97
 Low poverty, Enclave 14,872,874 5,324 25.9 25.2 26.6 1.00
Enclave trajectory,h 2010 census tracts
2006–2011  Never enclave 4,672,609 1,454 26.7 25.3 28.2 0.78 0.73 0.83
 Former enclave 815,707 274 28.0 24.7 31.5 0.82 0.72 0.93
 Persistent enclave 13,577,936 5,888 34.3 33.4 35.2 1.00
 Emergent enclave 1,720,099 538 29.5 26.9 32.1 0.86 0.78 0.94
2012–2017  Never enclave 5,917,553 1,865 23.4 22.3 24.4 0.85 0.81 0.90
 Former enclave 890,002 326 26.9 24.0 30.0 0.98 0.88 1.10
 Persistent enclave 15,862,645 6,150 27.4 26.7 28.1 1.00
 Emergent enclave 2,338,267 663 24.0 22.2 25.9 0.88 0.81 0.95
Colorectal, Male Asian American enclave index,e census tract
2006–2011  Quintile (Q) 1 (low/least culturally distinct) 348,960 119 30.4 24.7 36.6 0.69 0.56 0.84 0.0040
 Q2 838,867 299 32.6 28.8 36.7 0.74 0.65 0.84
 Q3 1,803,536 692 36.8 33.9 39.8 0.83 0.76 0.91
 Q4 3,546,011 1,364 37.5 35.4 39.7 0.85 0.80 0.90
 Q5 (high/most culturally distinct) 11,712,753 5,965 44.2 43.1 45.4 1.00
2012–2017  Quintile (Q) 1 (low/least culturally distinct) 457,076 151 24.1 20.2 28.2 0.62 0.52 0.74 0.0006
 Q2 1,063,378 360 26.7 23.9 29.7 0.69 0.61 0.77
 Q3 2,220,756 825 30.5 28.4 32.7 0.78 0.73 0.85
 Q4 4,309,858 1,738 32.9 31.3 34.5 0.85 0.80 0.89
 Q5 (high/most culturally distinct) 13,865,572 7,065 38.9 37.9 39.8 1.00
Asian American enclave status,f census tract
2006–2011  Non-enclave 4,589,333 1,782 36.3 34.6 38.2 0.84 0.80 0.89
 Enclave 13,660,794 6,657 43.2 42.1 44.3 1.00
2012–2017  Non-enclave 5,685,090 2,155 30.3 29.0 31.7 0.80 0.76 0.84
 Enclave 16,231,549 7,984 38.0 37.1 38.8 1.00
Percent poverty,g census tract
2006–2011  High poverty (>=20%) 3,831,078 1,856 44.4 42.3 46.4 1.08 1.03 1.14
 Low poverty (<20%) 14,419,049 6,583 40.9 39.8 41.9 1.00
2012–2017  High poverty (>=20%) 4,382,261 2,226 41.5 39.7 43.2 1.19 1.14 1.25
 Low poverty (<20%) 17,534,379 7,913 34.8 34.0 35.6 1.00
Enclave by poverty, census tract
2006–2011  High poverty, Non-enclave 1,090,089 433 39.1 35.3 43.0 0.92 0.83 1.02
 High poverty, Enclave 2,740,990 1,423 46.3 43.9 48.8 1.09 1.03 1.16
 Low poverty, Non-enclave 3,499,244 1,349 35.5 33.5 37.6 0.84 0.78 0.89
 Low poverty, Enclave 10,919,804 5,234 42.5 41.3 43.7 1.00
2012–2017  High poverty, Non-enclave 1,296,786 499 32.6 29.7 35.6 0.89 0.81 0.98
 High poverty, Enclave 3,085,475 1,727 45.1 42.9 47.3 1.24 1.17 1.31
 Low poverty, Non-enclave 4,388,304 1,656 29.7 28.3 31.2 0.82 0.77 0.86
 Low poverty, Enclave 13,146,075 6,257 36.4 35.5 37.4 1.00
Enclave trajectory,h 2010 census tracts
2006–2011  Never enclave 3,892,752 1,490 36.0 34.0 38.0 0.83 0.78 0.88
 Former enclave 718,950 297 38.6 34.2 43.4 0.89 0.78 1.00
 Persistent enclave 12,132,539 6,092 43.6 42.5 44.7 1.00
 Emergent enclave 1,528,255 565 39.5 36.0 43.1 0.91 0.82 1.00
2012–2017  Never enclave 4,937,297 1,821 29.6 28.2 31.0 0.77 0.73 0.81
 Former enclave 772,875 337 35.3 31.6 39.3 0.92 0.82 1.03
 Persistent enclave 14,146,367 7,176 38.4 37.5 39.4 1.00
 Emergent enclave 2,085,182 808 33.8 31.4 36.3 0.88 0.81 0.95
Open in a new tab
a

Annual age-adjusted incidence rates using census tract level data; calculated per 100,000 and standardized using the U.S. 2000 standard million population

b

Non-Hispanic Asian American/Native Hawaiian/Pacific Islander adult (≥20) population

c

Tumors in Asian American adults (age ≥20)

d

Linear tests of trend across enclave quintile AAIRs using weighted linear regression where the dependent variable was the AAIR, the independent variable was an ordinal variable set to the median value of the enclave measure within each quintile (Q1: −1.35663, Q2: −0.63602, Q3: −0.11118, Q4: 0.52978, Q5: 1.38291) and the weight was the inverse of the variance of the AAIR

e

Asian American enclave index developed at the census tract level using principal component analysis of 4 census tract level variables. Quintiles based on census tracts pooled across the 5 states. Quintile cut-points of the enclave index: Q1: −1.35663 to <−0.96913, Q2: −0.96913 to <−0.38415, Q3: −0.38415 to <0.19711, Q4: 0.19711 to <0.89552, Q5: 0.89552 to 3.80902.

f

Asian American enclave status defined as census tracts within the two highest quintiles of the enclave index (Q4 or Q5) with an Asian American population >250; Q4 enclaves were additionally required to be adjacent to an eligible Q5 enclave census tract

g

Neighborhood percent poverty defined as percent of households in the census tract below the Federal Poverty Level.

h

Asian American enclave trajectory defined based on enclave status of 2010 census tracts compared to enclave status in 2000, and categorized as: never enclave (non-enclave in both 2000 and 2010), former enclave (enclave in 2000 but not in 2010), persistent enclave (enclave in both 2000 and 2010), and emergent enclave (enclave in 2010 but not in 2000). AAIRs excluded census tracts with enclave status unknown in 2000

AAIR, age-adjusted incidence rate; IRR, incidence rate ratio; LCI, lower confidence interval; Q, quintile; UCI, upper confidence interval

Bold indicates statistical significance: IRRs which do not include 1.0 and p-trend<0.05

Figure 2.

Figure 2.

Open in a new tab

Age-adjusted incidence rates (AAIR) per 100,000 and 95% confidence intervals (CI) showing the joint association between Asian American enclave statusa and neighborhood percent povertyb; 2006–2017; in California, Florida, New Jersey, New York, and Texas. The figure shows AAIR and 95% CI for a joint census tract level-variable of Asian American enclave status (non-enclave and enclave) and neighborhood percent poverty (high poverty defined as >=20% below the Federal Poverty Level and low poverty defined as <20% below the Federal Poverty Level), separately for female breast, female cervical, female colorectal, and male colorectal cancer tumors diagnosed in 2006–2011 and 2012–2017 in Asian American adults in 5 states.

a Asian American enclave status defined as census tracts within the two highest quintiles of the enclave indexc (Q4 or Q5) with an Asian American population >250; Q4 enclaves were additionally required to be adjacent to an eligible Q5 enclave census tract

b Neighborhood percent poverty defined as percent of households in the census tract below the Federal Poverty Level

c Asian American enclave index developed at the census tract level using principal component analysis of 4 census tract level variables. Quintiles based on census tracts pooled across the 5 states. Quintile cut-points of the enclave index: Q1: −1.35663 to <−0.96913, Q2: −0.96913 to <−0.38415, Q3: −0.38415 to <0.19711, Q4: 0.19711 to <0.89552, Q5: 0.89552 to 3.80902

Cervical cancer

Cervical cancer incidence rates were not associated with Asian American enclave index quintiles (Table 2). However, incidence rates were lower in non-enclaves in the earlier time period (IRR: 0.89, 95% CI 0.79–0.99 in 2006–2011). Incidence rates were higher with high poverty in both time-periods (for example, IRR 1.68, 95% CI 1.52–1.85 in high-poverty neighborhoods compared to low-poverty in 2012–2017). Among estimates of joint associations, high poverty was associated with higher cervical cancer incidence regardless of enclave status (Figure 2). In the early time-period only, emergent and never enclaves, compared to persistent enclaves, had slightly lower incidence rates.

Colorectal cancer in females

Colorectal cancer incidence rates in females were generally lower in less culturally distinct neighborhoods than in more culturally distinct neighborhoods, with a gradient observed across quintiles in the early time-period (AAIR p-trend=0.01 in 2006–2011; AAIR p-trend=0.05 in 2012–2017, although the latter did not reach statistical significance; Table 2). Incidence rates were lower in non-enclaves compared to enclaves in both time-periods (for example, IRR 0.88, 95% CI 0.84–0.93 in 2012–2017). Incidence rates were higher with high poverty in the later time-period only. For joint associations, incidence rates were lower in non-enclaves compared to enclaves in both poverty levels; however, rates were only higher in high-poverty compared to low-poverty neighborhoods in enclaves in the later time-period (Figure 2). In 2012–2017, the IRRs were generally similar in all groups except higher in high-poverty Asian American enclaves (IRR 1.23, 95% CI 1.16–1.31) compared to low-poverty enclaves. In the early time-period, both former and emergent enclaves had lower incidence rates than persistent enclaves; whereas in the later time period, only emergent enclaves had lower rates.

Colorectal cancer in males

Colorectal cancer incidence rates in males were lower in less culturally distinct neighborhoods compared to more culturally distinct neighborhoods, with a gradient observed across quintiles in both time-periods (AAIR p-trend=0.004 in 2006–2011 and p-trend=0.001 in 2011–2017; Table 2). Incidence rates were lower in non-enclaves compared to enclaves in both time-periods (for example, IRR 0.80, 95% CI 0.76–0.84 in 2012–2017). Incidence rates were higher with high poverty in both time-periods. Examining joint associations, incidence rates were lower in non-enclaves compared to enclaves in both poverty levels; however, rates were higher in high-poverty compared to low-poverty neighborhoods only in enclaves especially in 2012–2017 (Figure 2). In 2012–2017, the IRRs were lowest in low-poverty non-enclaves (AAIR 0.82, 95% CI 0.77–0.86) and highest in high-poverty Asian American enclaves (IRR 1.24, 95% CI 1.17–1.31) compared to low-poverty enclaves. In the early time-period, both former and emergent enclaves had similar incidence rates to persistent enclaves; whereas in the later time period, emergent enclaves had lower rates.

State-stratified age-adjusted incidence rates by Asian American enclave index quintiles are presented in Supplementary Table S4. For breast cancer, patterns were similar to the overall five-state results (with lower AAIRs in low enclave) for CA, NY, and TX (later time-period); there was no association with enclave index quintiles in NJ; and results were in the opposite direction (with higher AAIRs in low enclave) in FL, particularly in the early time-period. For cervical cancer, patterns were similar to the overall results (no association with enclave index quintiles) or numbers were too small to assess associations. For colorectal cancer in females, patterns were similar to the overall results (with lower AAIRs in low enclave) only for CA and NY in the early time-period; otherwise, results were not statistically significant and showed an inconsistent pattern for other states and time-periods. For colorectal cancer in males, patterns were similar to the overall results (with lower AAIRs in low enclave) for CA, FL (early time-period), NY (later time-period), and TX (later time-period); there was no association with enclave index quintiles in NJ. Thus, in general, the overall results were driven by CA, NY, and TX.

Examining IRRs stratified by age (<50 vs. >=50 years) for enclave quintiles and enclave status for breast and colorectal cancer, there was no difference by age (overlapping CIs for the two age groups within each category and the estimate for the >=50 age group fell within the CI for the <50 age group for each category; Supplementary Table S5).

Discussion

We compared annual age-adjusted incidence rates of three cancers amenable to screening in Asian American adults by neighborhood Asian American enclave index quintiles, enclave status, and neighborhood poverty, across five states comprising 57.4% of the US Asian American population. While breast and colorectal cancer incidence rates were higher in Asian American enclaves, no association was found for cervical cancer. For breast cancer, both enclave and low-poverty neighborhoods were associated with higher incidence rates, with the highest incidence rates seen in low-poverty enclaves. For colorectal cancer, high-poverty enclave neighborhoods had the highest incidence rates. We found joint associations with residence in enclaves and neighborhood poverty suggesting synergistic impact but this varied across cancer site and time-period.

The relevance of residence in ethnic enclaves and interaction with neighborhood-level SES in cancer incidence among Asian American populations is consistent with existing, albeit limited, research pointing to the importance of these neighborhood-level attributes in cancer risk. For this study, stratifying enclave status by low/high poverty neighborhoods showed a synergistic association for breast cancer in both time-periods and colorectal cancer in 2012–2017 when comparing estimates and confidence intervals for non-enclaves across poverty levels (Supplemental Table S6). As summarized in a review paper, higher incidence of breast and cervical cancer was found in Asian American enclave census tracts in CA, but the inverse was observed in colorectal cancer.13 For breast cancer,25 the highest risk was in high-SES enclaves and for cervical cancer,26 the highest incidence was in low-SES enclaves. A recent study in AANHPI females residing in CA found lower incidence of breast cancer in high enclave and higher incidence in high nSES; with the highest incidence in low-enclave high-nSES neighborhoods.14 Inconsistent findings between studies are likely due to different methods used, both in how the population denominator data were estimated and how the enclave index measure was defined, especially given the that the majority of cases resided in the more culturally distinct areas (i.e., higher enclave index quintiles).

The observed associations between enclave residence and cancer incidence may reflect the influence of—or correlation with—several unmeasured factors, such as characteristics of the residents (individual-level SES, immigration status, insurance status, cancer screening behaviors) or attributes of the neighborhoods themselves. Though we could not directly assess these individual-level factors as they are not available in registry data, prior literature suggests they may play a role. Neighborhood features of enclaves such as ethnic-specific resources, co-ethnic social support and collective efficacy, and/or other environmental factors influence exposures and behaviors. In a previous ecologic analyses of these same five states, neighborhoods classified as enclaves by our measure of Asian American enclave status were characterized by a lower percentage of uninsured individuals and higher primary care accessibility, which could lead to more cancers being diagnosed through cancer screening.8 Greater access to healthcare may contribute to the higher CRC incidence rates observed in more culturally distinct neighborhoods. The same study demonstrated that enclaves were more likely to be metropolitan compared to non-enclaves, and also have lower crime and less poverty. Higher breast cancer incidence within more culturally distinct neighborhoods may be shaped by reproductive patterns (lower parity, later age at first birth) prevalent in metropolitan, higher-SES places, potentially interacting with screening access and acculturation-related lifestyle changes. In addition, non-US born status has been shown to be associated with higher breast cancer risk in Asian American women27,28, which could contribute to higher breast cancer incidence within enclaves. Further research is needed to tease apart and identify the specific factors underlying these differential cancer risks. Additional multi-state studies using the same methods across states could provide more clarity.

We observed higher incidence of breast cancer in low-poverty neighborhoods and higher incidence of cervical cancer in high-poverty neighborhoods. These results are consistent with prior studies which show higher incidence of breast cancer among those with high individual and neighborhood SES, due to different distributions of risk factors by SES29–33 and higher incidence of cervical cancer in those with low SES, due to different uptake of screening for cervical cancer, different rates of follow-up for abnormal testing, as well as differences in HPV infection rates and HPV prevalence by SES34,35. While there is evidence of lower CRC mortality with higher SES in non-Hispanic White patients,36 the association between CRC incidence and SES is less clear.37,38 We observed higher incidence of CRC in high-poverty neighborhoods in both time-periods for males and in the later time-period only for females. The emergence of a SES disparity is likely influenced by the unequal adoption of CRC screening during this time period.39

We found no meaningful differences in incidence rates of breast and colorectal cancer when stratified by age (<50 vs. >=50 years); although numbers were small in the <50 age group. As rates were not higher in the screening-eligible age group, this suggests incidence was not due to healthcare access or screening but other factors.

Similar to other studies,40,41 we saw higher CRC incidence rates in males than females; this is suggested comparing AAIRs and CIs across categories for the different measures by sex. When comparing IRRs, however, patterns are generally similar by sex across the enclave measures.

Limitations

While strengths of this analysis include using population-based cancer registry data from five states representing 57.4% of Asian American individuals in the US and a composite measure of Asian American enclave index based on CT-level data, there are several limitations. Results may be specific to the Asian American population in the states included and may not be generalizable to this population in the rest of the US. We were underpowered to examine AAIRS and IRRs stratified by age, as numbers were small in the <50 age group. We used residential address at time of diagnosis; however, cases could have lived elsewhere during the critical exposure and screening windows for cancer. To measure ethnic enclave and poverty we used ACS data from 2008–2012, which was from the earlier years included in our study and therefore may not have reflected the enclave status of neighborhoods for cases diagnosed in later years. However, statewide comparisons have shown that these measures were relatively stable over time, with 88–92% of census tract not changing their enclave status designation when using ACS data from 2015–2019 (compared to 2008–2012). Cancer cases were Asian American individuals (regardless of Hispanic ethnicity) as the Asian American enclave index was defined predominantly based on neighborhood characteristics specific to Asian American populations. However, the Woods & Poole population estimates were enumerated for non-Hispanic Asian American/Native Hawaiian/Pacific Islander individuals, as data for the non-Hispanic Asian American population were not available at the time of this analysis. This slight mismatch between the numerator and denominator is a limitation, but we believe the extent of under-estimation of rates is small given that, in these five states, Native Hawaiian and Pacific Islander individuals comprise approximately 2.2% of the total AANHPI population. Moreover, in a sensitivity analysis of the enclave index quintile analysis when we recalculated AAIRs to include non-Hispanic Asian American/Native Hawaiian/Pacific Islander adult cases, results were similar. For example, for breast cancer in 2012–2017, IRR 0.73, 95% CI 0.68–0.79 in Q1 compared to Q5; AAIR p-trend=0.08 (compared to IRR 0.71, 95% CI 0.65–0.76 in Q1 compared to Q5; AAIR p-trend=0.06 in the primary analysis). We were unable to show AAIRs disaggregated into specific Asian American ethnic groups, because although we had this data for the cases (excepting Florida), the corresponding disaggregated Asian American population data was not available. These limitations highlight the need for population denominator data that disaggregates Asian American, Native Hawaiian, and Pacific Islander populations overall, and ideally by specific ethnic groups, by CT, sex, and 5-year age intervals—data that are not currently available but necessary for population-based studies of cancer incidence relevant to these diverse populations. Additionally, as higher incidence could reflect greater detection rather than higher risk, having data on insurance status and screening would better inform our understanding of what may be contributing to these cancer incidence rates as well as inform potential opportunities for interventions.

Conclusion

Asian American enclave residence was associated with higher incidence rates of breast and colorectal cancers, with no association for cervical cancer. In breast cancer, the highest incidence rate was in low-poverty enclaves, whereas for colorectal cancer, the highest incidence rate was in high-poverty enclaves. Going forward, understanding incidence rates for other cancers will be important, especially those where Asian American individuals have a higher burden (e.g., lung, liver, and gastric cancers). Further disaggregated research is needed to continue to disentangle both positive and negative health effects of residing in Asian American enclaves for Asian ethnic groups, and culturally-informed interventions are warranted to increase screening and reduce cancer incidence in areas experiencing disproportionate burden.

Supplementary Material

1
2
3
4
5
6

Acknowledgments

Funding: This work was supported by a grant from the National Cancer Institute (R01CA237540) (multiple principal investigators: SSM and SLP). The collection of California cancer incidence data used in this study was supported by the California Department of Public Health pursuant to California Health and Safety Code Section 103885; Centers for Disease Control and Prevention’s (CDC) National Program of Cancer Registries, under cooperative agreement 1NU58DP007156; the National Cancer Institute’s Surveillance, Epidemiology and End Results Program under contract HHSN261201800032I awarded to the University of California, San Francisco, contract HHSN261201800015I awarded to the University of Southern California, and contract HHSN261201800009I awarded to the Public Health Institute. The ideas and opinions expressed herein are those of the author(s) and do not necessarily reflect the opinions of the State of California, Department of Public Health, the National Cancer Institute, and the Centers for Disease Control and Prevention or their Contractors and Subcontractors. The Florida cancer incidence data used in this study were collected by the Florida Cancer Data System (FCDS), the statewide cancer registry funded by the Florida Department of Health (DOH) and the Centers for Disease Control and Prevention’s National Program of Cancer Registries (CDC-NPCR). The views expressed herein are solely those of the author(s) and not necessarily reflect those of the DOH or CDC-NPCR. The New Jersey cancer incidence data used in this study were collected by the New Jersey State Cancer Registry and the New Jersey Department of Health. The collection of New York cancer incidence data used in this study was supported in part by cooperative agreement NU58DP007218 awarded to the New York State Department of Health by the Centers for Disease Control and Prevention and by Contract HHSN261201800005I, task order HHSN26100001, from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services. Texas cancer data have been provided by TCR, the Cancer Epidemiology and Surveillance Branch, Texas Department of State Health Services, 1100 West 49th Street, Austin, TX 78756. Data from TCR is supported by the following: Cooperative Agreement #1NU58DP007140 from the Centers for Disease Control and Prevention (CDC), Contract #75N91021D00011 from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program, and the Cancer Prevention and Research Institute of Texas (CPRIT).

Footnotes

Conflict of interest disclosure statement: Unrelated to this work, SLP reports receiving personal fees from Pfizer and Gilead. The authors have no additional relationships to disclose.

References

  • 1.Logan JR, Alba RD, Zhang W. Immigrant enclaves and ethnic communities in New York and Los Angeles. American Sociological Review. 2002;67(2):299–322. doi: 10.1177/000312240206700207. [DOI] [Google Scholar]
  • 2.Lim S, Yi SS, Lundy De La Cruz N, Trinh-Shevrin C. Defining ethnic enclave and its associations with self-reported health outcomes among Asian American adults in New York City. J Immigr Minor Health. 2017;19(1):138–46. Epub 2015/12/25. doi: 10.1007/s10903-015-0334-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Diez Roux AV, Mair C. Neighborhoods and health. Ann N Y Acad Sci. 2010;1186:125–45. Epub 2010/03/06. doi: 10.1111/j.1749-6632.2009.05333.x. [DOI] [PubMed] [Google Scholar]
  • 4.Yen IH, Michael YL, Perdue L. Neighborhood environment in studies of health of older adults: a systematic review. Am J Prev Med. 2009;37(5):455–63. Epub 2009/10/21. doi: 10.1016/j.amepre.2009.06.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Meijer M, Rohl J, Bloomfield K, Grittner U. Do neighborhoods affect individual mortality? A systematic review and meta-analysis of multilevel studies. Soc Sci Med. 2012;74(8):1204–12. Epub 2012/03/01. doi: 10.1016/j.socscimed.2011.11.034. [DOI] [PubMed] [Google Scholar]
  • 6.Krieger N Theories for social epidemiology in the 21st century: an ecosocial perspective. Int J Epidemiol. 2001;30(4):668–77. Epub 2001/08/21. doi: 10.1093/ije/30.4.668. [DOI] [PubMed] [Google Scholar]
  • 7.Gallo LC, Penedo FJ, Espinosa de los Monteros K, Arguelles W. Resiliency in the face of disadvantage: do Hispanic cultural characteristics protect health outcomes? J Pers. 2009;77(6):1707–46. Epub 2009/10/03. doi: 10.1111/j.1467-6494.2009.00598.x. [DOI] [PubMed] [Google Scholar]
  • 8.Guan A, Pruitt SL, Henry KA, Lin K, Meltzer D, Canchola AJ, et al. Asian American enclaves and healthcare accessibility: An ecologic study across five states. Am J Prev Med. 2023;65(6):1015–25. Epub 2023/07/11. doi: 10.1016/j.amepre.2023.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin. 2016;66(1):7–30. Epub 2016/01/09. doi: 10.3322/caac.21332. [DOI] [PubMed] [Google Scholar]
  • 10.Chen MS, Lee RJ, Madan RA, Ta Park V, Shinagawa SM, Sun T, et al. Charting a path towards Asian American cancer health equity: A way forward. J Natl Cancer Inst. 2022;114(6):792–9. Epub 2022/04/20. doi: 10.1093/jnci/djac055. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Gomez SL, Shariff-Marco S, DeRouen M, Keegan TH, Yen IH, Mujahid M, et al. The impact of neighborhood social and built environment factors across the cancer continuum: Current research, methodological considerations, and future directions. Cancer. 2015;121(14):2314–30. Epub 2015/04/08. doi: 10.1002/cncr.29345. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Fang CY, Tseng M. Ethnic density and cancer: A review of the evidence. Cancer. 2018;124(9):1877–903. Epub 2018/02/08. doi: 10.1002/cncr.31177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Shen H, Vos RO, Liu L. Relationships between living in ethnic enclaves and cancer outcomes: A systematic review of literature. J Racial Ethn Health Disparities. 2025. Epub 2025/04/22. doi: 10.1007/s40615-025-02439-0. [DOI] [PubMed] [Google Scholar]
  • 14.Truong A, McKinley M, Gomez SL, Kim MO, Shariff-Marco S, Cheng I. The role of ethnic enclaves and neighborhood socioeconomic status in invasive breast cancer incidence rates among Asian American, Native Hawaiian, and Pacific Islander females in California. Cancer Causes Control. 2025;36(2):183–9. Epub 2024/10/23. doi: 10.1007/s10552-024-01907-y. [DOI] [PubMed] [Google Scholar]
  • 15.Sangaramoorthy M, Yang J, DeRouen MC, Ho C, Somsouk M, Tana MM, et al. Disparities in hepatocellular carcinoma incidence in California: An update. Cancer Epidemiol Biomarkers Prev. 2020;29(1):79–87. Epub 2019/11/14. doi: 10.1158/1055-9965.EPI-19-0560. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Thornton ML, (ed). Data dictionary for carcinoma in situ of the cervix, CIN, and the Bethesda System. Standards for Cancer Registries Volume II: Data Standards and Data Dictionary, Version 24, 25th ed. Springfield, Ill: North American Association of Central Cancer Registries; July 2023. (Rev. July 2023, Aug. 2023, Oct. 2023, Feb 2024, Mar 2024). [Google Scholar]
  • 17.Thornton ML, (ed). Data dictionary for race (data items #160–164) and Hispanic/Spanish origin (data item #190). Data Standards and Data Dictionary. Version 26, 27th ed. Springfield, Ill: North American Association of Central Cancer Registries; June 2025. (Rev. July 2025, Aug. 2025, Sept. 2025). [Google Scholar]
  • 18.Group NRaEW. NAACCR Asian Pacific Islander Identification Algorithm. Springfield, IL: North American Association of Central Cancer Registries; 2011. [cited Aug 2011]; NAPIIA v1.2.1:[Available from: https://www.naaccr.org/analysis-and-data-improvement-tools/#NHAPIIA. [Google Scholar]
  • 19.Shariff-Marco S, Lin K, Meltzer D, Allen L, Boscoe FP, Canchola AJ, et al. Developing measures of Asian American and Hispanic/Latino ethnic enclaves for five states using U.S. Census and American Community Survey data San Francisco, CA: University of California San Francisco; Oct 2021. Available from: https://cancerregistry.ucsf.edu/resources/research-tools. [Google Scholar]
  • 20.Populations - Total U.S. (2006–2020), Census Tract Estimates by Race/Origin Controlling to Vintage 2020 <2010 Tract Geographies>, National Cancer Institute, DCCPS, Surveillance Research Program, released January 2023. Source: Woods & Poole Economics, Inc. Washington, D. C. Copyright 2022. [Google Scholar]
  • 21.Fay MP, Feuer EJ. Confidence intervals for directly standardized rates: A method based on the gamma distribution. Statistics in Medicine. 1997;16(7):791–801. doi: 10.1002/(sici)1097-0258(19970415)16:7<791::aid-sim500>3.3.co;2-r. [DOI] [PubMed] [Google Scholar]
  • 22.Tiwari RC, Clegg LX, Zou Z. Efficient interval estimation for age-adjusted cancer rates. Stat Methods Med Res. 2006;15(6):547–69. Epub 2007/01/31. doi: 10.1177/0962280206070621. [DOI] [PubMed] [Google Scholar]
  • 23.Boyle P, Parkin DM. Cancer registration: principles and methods. Statistical methods for registries. IARC Sci Publ. 1991(95):126–58. Epub 1991/01/01. [PubMed] [Google Scholar]
  • 24.Fay MP. Approximate confidence intervals for rate ratios from directly standardized rates with sparse data. Communications in Statistics - Theory and Methods. 1999;28(9):2141–60. doi: 10.1080/03610929908832411. [DOI] [Google Scholar]
  • 25.Morey BN, Gee GC, Wang MC, von Ehrenstein OS, Shariff-Marco S, Canchola AJ, et al. Neighborhood contexts and breast cancer among Asian American women. J Immigr Minor Health. 2022;24(2):445–54. Epub 2021/04/14. doi: 10.1007/s10903-021-01196-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Froment MA, Gomez SL, Roux A, DeRouen MC, Kidd EA. Impact of socioeconomic status and ethnic enclave on cervical cancer incidence among Hispanics and Asians in California. Gynecol Oncol. 2014;133(3):409–15. Epub 2014/03/29. doi: 10.1016/j.ygyno.2014.03.559. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Morey BN, Gee GC, von Ehrenstein OS, Shariff-Marco S, Canchola AJ, Yang J, et al. Higher breast cancer risk among immigrant Asian American women than among US-born Asian American women. Prev Chronic Dis. 2019;16:E20. Epub 2019/02/16. doi: 10.5888/pcd16.180221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.John EM, Koo J, Ingles SA, Kurian AW, Hines LM. Changes in breast cancer risk and risk factor profiles among U.S.-born and immigrant Asian American women residing in the San Francisco Bay Area. Cancer Epidemiol Biomarkers Prev. 2023;32(5):666–77. Epub 2023/02/14. doi: 10.1158/1055-9965.EPI-22-1128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Palmer JR, Boggs DA, Wise LA, Adams-Campbell LL, Rosenberg L. Individual and neighborhood socioeconomic status in relation to breast cancer incidence in African-American women. Am J Epidemiol. 2012;176(12):1141–6. Epub 2012/11/23. doi: 10.1093/aje/kws211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Dong JY, Qin LQ. Education level and breast cancer incidence: a meta-analysis of cohort studies. Menopause. 2020;27(1):113–8. Epub 2019/09/04. doi: 10.1097/GME.0000000000001425. [DOI] [PubMed] [Google Scholar]
  • 31.Trewin CB, Hjerkind KV, Johansson ALV, Strand BH, Kiserud CE, Ursin G. Socioeconomic inequalities in stage-specific breast cancer incidence: a nationwide registry study of 1.1 million young women in Norway, 2000–2015. Acta Oncol. 2020;59(11):1284–90. Epub 2020/04/23. doi: 10.1080/0284186X.2020.1753888. [DOI] [PubMed] [Google Scholar]
  • 32.Berger E, Maitre N, Romana Mancini F, Baglietto L, Perduca V, Colineaux H, et al. The impact of lifecourse socio-economic position and individual social mobility on breast cancer risk. BMC Cancer. 2020;20(1):1138. Epub 2020/11/25. doi: 10.1186/s12885-020-07648-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Katuwal S, Tapanainen J, Pukkala E. Multivariate analysis of independent roles of socioeconomic status, occupational physical activity, reproductive factors, and postmenopausal hormonal therapy in risk of breast cancer. Breast Cancer Res Treat. 2022;193(2):495–505. Epub 2022/04/03. doi: 10.1007/s10549-022-06571-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Cham S, Li A, Rauh-Hain JA, Tergas AI, Hershman DL, Wright JD, Melamed A. Association between neighborhood socioeconomic inequality and cervical cancer incidence rates in New York City. JAMA Oncol. 2022;8(1):159–61. Epub 2021/11/25. doi: 10.1001/jamaoncol.2021.5779. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Jensen KE, Hannibal CG, Nielsen A, Jensen A, Nohr B, Munk C, et al. Social inequality and incidence of and survival from cancer of the female genital organs in a population-based study in Denmark, 1994–2003. Eur J Cancer. 2008;44(14):2003–17. Epub 2008/07/30. doi: 10.1016/j.ejca.2008.06.014. [DOI] [PubMed] [Google Scholar]
  • 36.Sturley C, Norman P, Morris M, Downing A. Contrasting socio-economic influences on colorectal cancer incidence and survival in England and Wales. Soc Sci Med. 2023;333:116138. Epub 2023/08/15. doi: 10.1016/j.socscimed.2023.116138. [DOI] [PubMed] [Google Scholar]
  • 37.Steinbrecher A, Fish K, Clarke CA, West DW, Gomez SL, Cheng I. Examining the association between socioeconomic status and invasive colorectal cancer incidence and mortality in California. Cancer Epidemiol Biomarkers Prev. 2012;21(10):1814–22. Epub 2012/08/23. doi: 10.1158/1055-9965.EPI-12-0659. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Doubeni CA, Laiyemo AO, Major JM, Schootman M, Lian M, Park Y, et al. Socioeconomic status and the risk of colorectal cancer: an analysis of more than a half million adults in the National Institutes of Health-AARP Diet and Health Study. Cancer. 2012;118(14):3636–44. Epub 2012/08/18. doi: 10.1002/cncr.26677. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Santiago-Rodriguez EJ, Shariff-Marco S, Gomez SL, Hiatt RA. Disparities in colorectal cancer screening by time in the U.S. and race/ethnicity, 2010–2018. Am J Prev Med. 2023;65(1):74–82. Epub 2023/02/22. doi: 10.1016/j.amepre.2023.01.033. [DOI] [PubMed] [Google Scholar]
  • 40.Abotchie PN, Vernon SW, Du XL. Gender differences in colorectal cancer incidence in the United States, 1975–2006. J Womens Health (Larchmt). 2012;21(4):393–400. Epub 2011/12/14. doi: 10.1089/jwh.2011.2992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Murphy G, Devesa SS, Cross AJ, Inskip PD, McGlynn KA, Cook MB. Sex disparities in colorectal cancer incidence by anatomic subsite, race and age. Int J Cancer. 2011;128(7):1668–75. Epub 2010/05/27. doi: 10.1002/ijc.25481. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2171820-supplement-1.docx (6.3MB, docx)
2
NIHMS2171820-supplement-2.docx (6.3MB, docx)
3
NIHMS2171820-supplement-3.docx (6.3MB, docx)
4
NIHMS2171820-supplement-4.docx (6.3MB, docx)
5
NIHMS2171820-supplement-5.docx (6.3MB, docx)
6
NIHMS2171820-supplement-6.docx (6.3MB, docx)

Data Availability Statement

The ethnic enclave measures for recent years are available to the public on the UCSF Health Atlas website (see https://healthatlas.ucsf.edu/). The registry data are not publicly available due to data confidentiality concerns and can be requested via state cancer registries.

RESOURCES