Health Conditions, Outcomes, and Service Access Among Filipino, Vietnamese, Chinese, Japanese, and Korean Adults in California, 2011–2017

Abstract

Objectives. To determine the impact of data disaggregation on the ability to identify health disparities and needs for future research for Filipino, Vietnamese, Chinese, Japanese, and Korean adults in California.

Methods. Using available data from the 2011–2017 California Health Interview Survey, we conducted bivariate and multivariable analyses to assess disparities in health conditions, outcomes, and service access compared with non-Hispanic Whites for Asians as an overall group and for each individual subgroup.

Results. As an aggregate category, Asians appeared healthier than did non-Hispanic Whites on most indicators. However, every Asian subgroup had at least 1 disparity disguised by aggregation. Filipinos had the most disparities, with higher prevalence of fair or poor health, being obese or overweight, and having high blood pressure, diabetes, or asthma compared with non-Hispanic Whites (P < .05) in multivariable analyses.

Conclusions. Failure to disaggregate health data for individual Asian subgroups disguises disparities and leads to inaccurate conclusions about needs for interventions and research.

Asians are the fastest growing racial group in the United States, with the population growing by 12.8% from 2013 to 2017, a rate 4 times faster than that of the overall US population.¹ Currently, California has the largest population of Asian Americans.¹ Overall, Asians have historically been branded as a “model minority,” despite evidence that Asians are underserved in health care.² This designation has harmful impacts, as Asian Americans are viewed as healthier than are members of other races and are de-emphasized in research focusing on health disparities.³ Subsequently, few large-scale health surveys specifically tailor methods toward increasing accessibility among Asian subgroups (e.g., providing survey questions in Asian languages) or enabling subgroup analyses rather than an aggregate Asian category.⁴ Calls for disaggregation of Asian Americans into distinct subgroups to properly assess health disparities and needs for interventions have increased over the years.^4–6

Previous efforts to disaggregate health data for Asian Americans have shown substantial variation in mortality patterns,^7,8 insurance coverage and health service utilization,^9,10 health condition,^11,12 and health outcomes.^11–13 However, many previous studies focused on a limited set of health dimensions or analyzed data from small samples, which limited the capacity for quantitative disaggregation of health data by health condition and by ethnic group. Although identification of specific differences between Asian subgroups in smaller studies is useful, disaggregation in broader population-based surveys is necessary to better assess health needs for diverse Asian subpopulations. Problematically, many surveys that collect disaggregated data do not contain adequate samples to allow subgroup-specific analyses,^4,6 which would help to identify priorities for research and programming.

Given this, our focus was to use the California Health Interview Survey (CHIS) to examine patterns and potential disparities in self-reported health between Asian subgroups. CHIS has been used in previous efforts to assess health disparities by race in California.^14,15 Specifically, we examined health indicators (including health condition, health outcomes, and health service access) in a single aggregated Asian category compared with these health indicators disaggregated into Filipino, Vietnamese, Chinese, Japanese, and Korean subgroups. Through these analyses, we provided data-driven identification of how aggregation into a single Asian category limits the identification of health disparities and identified future priorities for research, policy, and health programs.

METHODS

We used 7 years of publicly available data from adults included in the 2011 to 2017 iterations of CHIS. The largest state health survey in the country, CHIS is a random-dial telephone survey using a multistage sampling design conducted in English, Spanish, Cantonese, Mandarin, Korean, Tagalog, and Vietnamese and was initiated in 2001 as a biennial survey. CHIS data have been collected annually since 2011. CHIS employs oversampling of Koreans and Vietnamese to ensure sufficient representation for analysis.

Outcomes

To form a comprehensive understanding about potential health disparities among Asian American subgroups, we selected 4 health indicators based on measures of health condition, health outcomes, and health service utilization assessed in CHIS. Health condition indicators assessed in CHIS included being obese or overweight, self-reported health status, having a disability, and experiencing serious distress. Self-reported height and weight were used to calculate body mass index (BMI; defined as weight in kilograms divided by the square of height in meters). Non-Hispanic Whites with BMIs of 25.0 or above were classified as overweight or obese. Per World Health Organization guidelines, Asians with BMIs of 23.0 or above were classified as overweight or obese.¹⁶ Self-reported health status included excellent, very good, good, fair, and poor and was dichotomized into those reporting fair or poor health versus those who did not.

Disability was a CHIS-constructed variable whereby participants were categorized as disabled if they indicated yes to any of the following: blind, deaf, or severe vision or hearing problem; difficulty learning, remembering, or concentrating; difficulty dressing, bathing, or getting around the house; difficulty going outside the home alone to shop or visit a doctor’s office; difficulty working at a job or business; or at least 1 limitation in 1 or more basic activity such as walking or climbing stairs. Serious distress was assessed using the Kessler 6 scale, which has a range of 6 to 30, and scores of 13 or higher are indicative of serious mental distress in the past 30 days.¹⁷

Health outcomes assessed in CHIS included having any previous diagnosis of heart disease, diabetes, or high blood pressure and any occurrence of asthma. Health service access assessed in CHIS included whether participants had a usual source of care besides the emergency room, had seen a doctor in the past year, delayed getting care in the past year, and delayed getting medications in the past year.

Race

Race was assessed through self-report. Participants who indicated they were Asian were asked with which what specific ethnic group they identified. In the public use files, 5 Asian subgroups had sufficient respondents for CHIS to create unique analytic categories: Filipino, Vietnamese, Chinese, Japanese, and Korean. Participants who indicated their race as Bangladeshi, Burmese, Cambodian, Indian, Indonesian, Laotian, Malaysian, Pakistani, Sri Lankan, Taiwanese, Thai, or other Asian were aggregated into a separate “other Asian” category; these participants were included in the aggregate Asian category but were not specifically examined in subgroup analyses because our goal was to present disaggregated estimates. We included non-Hispanic Whites to provide a referent group for our analyses; 13 297 Asians and 88 296 non-Hispanic Whites were included in our sample. Of the 13 297 Asians included, 2101 identified as Filipino, 1939 as Vietnamese, 4106 as Chinese, 1343 as Japanese, and 1587 as Korean.

Covariates

We included the following demographic variables: age, sex, marital status, employment, education, annual income, health insurance status, nativity, and English proficiency. We coded age into 3 categories: 18 to 44 years, 45 to 64 years, and 65 years or older. We dichotomized marital status into those who were married or living with a partner versus those who were not. We coded educational attainment as high school, some college, and bachelor’s degree or more. We categorized annual income using the 2011–2017 US Department of Health and Human Services’ federal poverty guidelines (FPG) into 138% or less of FPG, 139% to 400% FPG, and greater than 400% FPG. We assessed nativity using percentage of a participant’s life lived in the United States and we categorized it into 0% to 60%, 61% to 99%, and 100%. We dichotomized English proficiency into participants who spoke English only or had strong English proficiency versus those who did not speak English.

Statistical Analyses

We first created descriptive analyses to assess the distribution of demographic characteristics, general health condition, health outcomes, and health service utilization across the Asian subgroups. We conducted the χ² test by subgroup to assess differences in our study sample. We performed multivariable logistic regressions to assess racial differences for all Asians in an aggregate category compared with non-Hispanic Whites. We then performed specific comparisons for Filipino, Vietnamese, Chinese, Japanese, and Korean adults on each outcome compared with non-Hispanic White adults for all health indicators. Regression models adjusted for all covariates as well as the survey year to minimize correlation based on survey year. Adjusted odds ratios (AORs) and 95% confidence intervals (CIs) are presented. We used sample weights specific to each iteration of CHIS to account for the complex sampling design and to obtain correct variance estimations for all of our analyses. We performed all analyses using Stata version 15 (StataCorp LP, College Station, TX).

RESULTS

Weighted sample characteristics are presented in Table 1. Because of CHIS’s design and sample weights, the sociodemographic characteristics of the sample reflect the non-Hispanic White and Asian population in California overall. A majority or plurality of Asians in our sample were female (53.4%), aged 18 to 44 years (57.3%), making a yearly income of more than 400% of the FPG (45.2%), college graduates (57.2%), employed (66.5%), married or living with a partner (58.3%), proficient in English (82.4%), and insured (89.8%) and had spent 0% to 60% of their life in the United States (49.2%).

TABLE 1—

Weighted Demographic Characteristics: California Health Interview Survey, 2011–2017

Variable	Non-Hispanic White (n = 88 296), %	All Asian (n = 13 297), %	Filipino (n = 2101), %	Vietnamese (n = 1939), %	Chinese (n = 4106), %	Japanese (n = 1343), %	Korean (n = 1587), %	P
Sex								.02
Male	49.2	46.6	46.3	48.5	44.7	42.1	42.1
Female	50.8	53.4	53.7	51.5	55.3	57.9	57.9
Age, y								< .001
18–44	37.9	57.3	57.6	51.1	58.1	37.7	55.2
45–64	37.1	29.8	28.6	33.9	30.6	37.9	26.9
≥65	25.0	12.8	13.7	15.0	11.3	24.4	17.9
Income, % FPGa								< .001
≤ 138	17.1	21.5	21.2	34.7	19.8	13.4	21.8
139–400	32.0	33.3	37.0	30.8	32.8	30.9	39.0
> 400	50.8	45.2	41.8	34.5	47.4	55.7	39.3
Education								< .001
High school	26.1	24.8	19.1	48.3	25.6	27.4	17.6
Some college	27.3	18.0	26.1	17.0	12.8	15.4	23.1
Bachelor’s degree or more	46.6	57.2	54.8	34.7	61.7	57.2	59.3
Employed								< .001
Yes	60.9	66.5	66.4	61.6	67.4	63.2	59.5
No	39.1	33.5	33.6	38.4	32.6	36.8	40.5
Married or living with partner								< .001
Yes	61.5	58.3	53.0	58.5	59.0	57.1	57.8
No	38.5	41.7	47.0	41.5	41.0	42.9	42.2
Speaks English only, very well, or well								< .001
Yes	99.7	82.4	96.1	57.5	76.2	95.2	65.2
No	0.3	17.7	3.9	42.5	23.8	4.8	34.8
Life spent in US, %								< .001
0–60	5.0	49.2	40.4	56.4	51.1	12.2	54.8
61–99	4.6	20.4	21.4	22.4	16.7	10.8	21.6
100	90.4	30.5	38.3	21.3	32.2	77.0	23.6
Insured								< .001
Yes	92.6	89.8	90.8	89.7	89.3	93.2	85.8
No	7.4	10.2	9.2	10.4	10.7	6.8	14.2

Note. FPG = federal poverty guidelines.

^aAs determined by the Department of Health and Human Services 2011–2017.

Prevalence of Health Burdens

Weighted information about the health of respondents is presented in Table 2. Because of varying ages for our sample, we included weighted information about the health of respondents stratified by age in Table A (available as a supplement to the online version of this article at http://www.ajph.org).

TABLE 2—

Weighted Health Condition, Outcomes, and Service Access for Asian American and Non-Hispanic White Adults: California Health Interview Survey, 2011–2017

Variable	Non-Hispanic White (n = 88 296), %	All Asian (n = 13 297), %	Filipino (n = 2101), %	Vietnamese (n = 1939), %	Chinese (n = 4106), %	Japanese (n = 1343), %	Korean (n = 1587), %	P
Health condition
Fair or poor health	13.9	17.8	15.8	36.4	17.1	15.8	11.8	< .001
Obese or overweight	57.7	61.2	72.5	50.3	54.4	65.0	49.1	< .001
Serious distress	3.2	2.5	3.6	3.0	1.5	1.6	5.0	.05
Disability	31.4	21.0	26.5	30.1	16.3	20.3	20.1	< .001
Health outcomes
High blood pressure	31.0	22.9	32.3	24.2	18.2	32.3	24.3	< .001
Diabetes	7.5	8.4	12.6	6.7	5.5	10.6	8.8	< .001
Asthma	16.1	11.5	17.9	8.6	9.9	16.7	6.8	< .001
Heart disease	8.5	4.4	6.1	4.8	3.6	4.7	3.9	< .001
Health service access
No usual source of care	10.9	17.4	15.4	18.4	16.9	13.8	27.0	< .001
No doctor’s visit in past year	14.9	23.0	16.8	23.1	27.1	18.7	26.4	< .001
Delayed medications	11.7	6.6	8.4	4.8	5.8	7.6	9.0	< .001
Delayed care	15.7	9.2	9.8	5.6	8.1	12.0	16.6	< .001

For health condition indicators, Vietnamese participants (36.4%) reported fair or poor health more than twice as often as did non-Hispanic Whites (13.9%) and Asians overall (17.8%). Filipino (72.5%) and Japanese (65.0%) respondents had a higher proportion of individuals who were obese or overweight than did non-Hispanic Whites (57.7%) and Asians overall (61.2%). Filipino (3.6%) and Korean (5.0%) participants reported serious distress more often than did non-Hispanic Whites (3.2%) and Asians overall (2.5%). No subgroup had higher rates of disability than did non-Hispanic Whites, but Vietnamese (30.1%) and Filipino (26.5%) respondents had higher rates of disability than did Asians as an aggregate group (21.0%).

For health outcome indicators, Filipino and Japanese adults (both 32.3%) reported having high blood pressure more often than did non-Hispanic Whites (31.0%) and Asians overall (22.9%). Filipino (12.6%), Japanese (10.6%), and Korean (8.8%) respondents had higher rates of diabetes than did non-Hispanic Whites (7.5%) and Asians overall (8.4%). Filipino (17.9%) and Japanese (16.7%) participants reported having asthma more often than did non-Hispanic Whites (16.1%) and Asians overall (11.5%). No subgroup had higher rates of heart disease than did non-Hispanic Whites, but Filipino (6.1%), Vietnamese (4.8%), and Japanese (4.7%) respondents had greater rates of heart disease than did Asians overall (4.4%).

For health service access indicators, Korean (27.0%) and Vietnamese (18.4%) participants had higher rates of lacking a usual source of care than did both non-Hispanic Whites (10.9%) and Asians overall (17.4%). Chinese (27.1%), Korean (26.4%), and Vietnamese (23.1%) adults reported going without a doctor’s visit in the past year more often than did both non-Hispanic Whites (11.7%) and Asians overall (23.0%). No subgroup delayed medications more often than did non-Hispanic Whites (11.7%), but Korean (9.0%) and Filipino (8.4%) respondents delayed medications more often than did Asians overall (6.6%). Korean participants (16.6%) delayed seeking care more often than did both non-Hispanic Whites (15.7%) and Asians overall (9.2%).

Multivariable Model Results

In Table 3, we present AORs for all health indicators after controlling for all covariates.

TABLE 3—

AORs for Health Condition, Outcomes, and Service Access for 5 Asian Subgroups with Non-Hispanic Whites as a Referent: California Health Interview Survey, 2011–2017

Variable	All Asian, AOR (95% CI)	Filipino, AOR (95% CI)	Vietnamese, AOR (95% CI)	Chinese, AOR (95% CI)	Japanese, AOR (95% CI)	Korean, AOR (95% CI)
Health condition
Fair or poor health	1.2 (1.0, 1.4)	1.4 (1.0, 1.9)	1.6 (1.2, 2.3)	1.0 (0.8, 1.4)	0.9 (0.5, 1.4)	1.2 (0.8, 1.7)
Obese or overweight	1.8 (1.5, 2.1)	2.9 (2.1, 3.8)	0.9 (0.7, 1.2)	1.4 (1.1, 1.8)	1.4 (1.1, 2.0)	1.1 (0.8, 1.5)
Serious distress	0.8 (0.5, 1.2)	1.1 (0.6, 2.2)	0.7 (0.3, 1.7)	0.4 (0.2, 0.8)	0.6 (0.1, 3.7)	1.5 (0.7, 3.3)
Disability	0.7 (0.6, 0.9)	1.1 (0.8, 1.4)	0.9 (0.5, 1.3)	0.5 (0.3, 0.6)	0.6 (0.4, 0.9)	0.6 (0.4, 0.9)
Health outcomes
High blood pressure	1.1 (0.9, 1.4)	2.0 (1.4, 2.8)	0.9 (0.6, 1.3)	0.8 (0.6, 1.1)	1.2 (0.9, 1.7)	1.1 (0.7, 1.5)
Diabetes	1.8 (1.4, 2.4)	2.9 (2.0, 4.3)	0.7 (0.4, 1.2)	1.1 (0.7, 1.7)	1.7 (1.1, 2.5)	1.4 (0.8, 2.5)
Asthma	1.0 (0.8, 1.3)	1.4 (1.0, 2.0)	0.8 (0.5, 1.2)	0.8 (0.6, 1.2)	1.1 (0.8, 1.6)	0.5 (0.2, 1.1)
Heart disease	0.9 (0.6, 1.1)	1.2 (0.8, 1.8)	0.9 (0.5, 1.8)	0.8 (0.4, 1.3)	0.6 (0.4, 0.9)	0.7 (0.2, 2.4)
Health service access
No usual source of care	1.2 (1.0, 1.5)	1.0 (0.8, 1.3)	1.2 (0.8, 1.9)	1.1 (0.8, 1.5)	1.3 (0.8, 2.2)	2.4 (1.6, 3.7)
No doctor’s visit in past year	1.2 (0.9, 1.4)	0.8 (0.6, 1.1)	1.1 (0.7, 1.8)	1.5 (1.1, 1.9)	1.4 (0.9, 2.1)	1.4 (0.9, 2.1)
Delayed medications	0.6 (0.4, 0.8)	0.8 (0.5, 1.1)	0.4 (0.2, 0.8)	0.5 (0.3, 0.7)	0.6 (0.4, 1.1)	0.6 (0.3, 1.1)
Delayed care	0.5 (0.4, 0.6)	0.5 (0.4, 0.7)	0.3 (0.2, 0.5)	0.5 (0.3, 0.6)	0.8 (0.4, 1.3)	1.2 (0.7, 2.0)

Note. AOR = adjusted odds ratio; CI = confidence interval. We controlled for sex, income, education, marital or partnered status, employment, age, nativity, insurance status, English proficiency, and California Health Interview Survey year.

Health condition.

For health condition, although Asians overall had higher odds (that were ultimately nonsignificant) of reporting fair or poor health (AOR = 1.2; 95% CI = 1.0, 1.4; P = .13) compared with non-Hispanic Whites, Vietnamese (AOR = 1.6; 95% CI = 1.2, 2.3; P < .05) and Filipino (AOR = 1.4; 95% CI = 1.0, 1.9; P < .05) respondents reported fair or poor health more often than did non-Hispanic Whites. Asians overall reported higher odds of being obese or overweight (AOR = 1.8; 95% CI = 1.4, 2.4; P < .001) compared with non-Hispanic Whites, whereas Filipino participants reported odds of being obese or overweight (AOR = 2.9; 95% CI = 2.1, 3.8; P < .001) nearly 3 times that of the odds for non-Hispanic Whites, and Chinese participants also reported higher odds of being obese or overweight (AOR = 1.4; 95% CI = 1.1, 1.8; P < .05) compared with non-Hispanic Whites.

Neither Asians overall nor any subgroup had significantly higher odds of serious distress compared with non-Hispanic Whites after disaggregating. Asians overall had lower odds of disability (AOR = 0.7; 95% CI = 0.6, 0.9; P < .05) compared with non-Hispanic Whites, but Filipino (AOR = 1.1; 95% CI = 0.8, 1.4; P = .62) and Vietnamese (AOR = 0.9; 95% CI = 0.5, 1.3; P = .49) participants were the only Asian subgroups to have odds of disability that were not significantly lower than those of non-Hispanic Whites.

Health outcomes.

For health outcomes, Asians overall had no significant difference in odds of having high blood pressure (AOR = 1.1; 95% CI = 0.9, 1.4; P = .20) compared with non-Hispanic Whites, but Filipino respondents had 2 times the odds (AOR = 2.0; 95% CI = 1.4, 2.8; P < .001) of having high blood pressure compared with non-Hispanic Whites. Asians overall had nearly twice the odds of having diabetes (AOR = 1.8; 95% CI = 1.4, 2.4; P < .001) compared with non-Hispanic Whites, whereas Filipinos overall had nearly 3 times the odds of having diabetes (AOR = 2.9; 95% CI = 2.0, 4.3; P < .001) as did non-Hispanic Whites.

Asians overall had no significant difference in odds of having asthma compared with non-Hispanic Whites (AOR = 1.0; 95% CI = 0.8, 1.3; P = .84), but Filipino respondents had higher odds of having asthma (AOR = 1.4; 95% CI = 1.0, 2.0; P < .05) compared with non-Hispanic Whites. Neither Asians overall nor any subgroup had significantly higher odds of heart disease compared with non-Hispanic Whites after disaggregating.

Health service access.

For health service access, Asians overall had odds of having no usual source of care (AOR = 1.2; 95% CI = 1.0, 1.5; P = .09) that were not significantly different from those of non-Hispanic Whites, but Korean participants had more than 2 times the odds of having no usual source of care (AOR = 2.4; 95% CI = 1.6, 3.7; P < .001) compared with non-Hispanic Whites. Although Asians overall did not significantly differ from non-Hispanic Whites on having no doctor’s visit in the past year (AOR = 1.2; 95% CI = 0.9, 1.4; P = .15), Chinese respondents had higher odds of having no doctor’s visit in the past year (AOR = 1.5; 95% CI = 1.1, 1.9; P < .05).

Asians overall had lower odds of having delayed medications (AOR = 0.6; 95% CI = 0.4, 0.7; P < .001) compared with non-Hispanic Whites, although Filipino (AOR = 0.8; 95% CI = 0.5, 1.1; P = .21), Japanese (AOR = 0.6; 95% CI = 0.4, 1.1; P = .08), and Korean (AOR = 0.6; 95% CI = 0.3, 1.1; P = .08) respondents did not significantly differ from non-Hispanic Whites in having delayed medications. Asians overall also had lower odds of having delayed care (AOR = 0.5; 95% CI = 0.4, 0.6; P < .001) compared with non-Hispanic Whites, but Korean (AOR = 1.2; 95% CI = 0.7, 2.0; P = .52) and Japanese (AOR = 0.8; 95% CI = 0.4, 1.3; P = .28) participants did not significantly differ from non-Hispanic Whites in having delayed care.

DISCUSSION

Data disaggregation for Asians in large-scale research efforts has been long identified as a way to combat the pervasive notion of Asian Americans as the model minority assumed to be healthier.³ Using CHIS data, we identified many health disparities through disaggregation that would otherwise be obscured when an aggregate Asian category was used. These findings point to several disparities that, because of disaggregation, can be prioritized in research and interventions that would have otherwise been masked by aggregation. Without disaggregation, these disparities may continue to go unaddressed when distinct Asian subgroups are combined into an aggregate category. Building on previous arguments regarding the need for disaggregation,^4–6 we extend points made in previous articles about disaggregation, illustrating the distortive effect of using a single Asian category through our analyses.

Overall, we identified several disparities in both our descriptive statistics and our models that would have otherwise gone unidentified for each of the subgroups included in our analyses. Relying on aggregate statistics may have encouraged action to address obesity and overweight disparities, higher prevalence of diabetes, and gaps in accessing a usual source of care or having a doctor’s visit in the past year. However, using an aggregate Asian category would not prompt further investigation and action into serious distress faced by Korean adults or the third of Vietnamese respondents in fair or poor health. Therefore, the aggregation of subgroup data into a single Asian category should be reframed as a choice that leads to inaccurate conclusions about health. Doing so ensures that efforts to protect Asian Americans’ right to health are properly informed and targeted.

Importantly, merely including different subgroups as options during data collection must be identified as insufficient. Guidelines for better data collection practices that can address the need for disaggregation already exist.^4,6 Investment in data collection practices, such as oversampling Asian subpopulations and surveying in their languages, that enable disaggregation should not be viewed as good, but unnecessary, features of population-based health research. Instead, these investments are critical steps toward producing accurate evaluations of the health of Asians in the United States. Population-based surveys drive our understanding of what health needs exist and what actions should be taken to correct these challenges. Therefore, aggregation acts as a barrier to identifying interventions that can be used to reduce costs in health service provision (through improved prevention activities) and improve health outcomes among minority populations.

Previously, reports using 2004–2006 data from the National Health Interview Survey (NHIS) that attempted to show disaggregated data had inadequate sample sizes (with large SEs).¹⁸ A 2016 report using 2010-2014 NHIS data concluded that Asian subgroups were healthier than average,¹⁹ but consistent with our findings, the disaggregated trends showed variation across Asian subgroups. In both reports, limitations from inadequate samples and non-Asian language administration of the NHIS present opportunities for improving Asian American population health insights. Currently, in addition to English and Spanish, CHIS is conducted in several Asian languages, and oversamples for Koreans and Vietnamese. Although investment in practices that enable meaningful disaggregated analyses do have upfront costs, increased prevention and improved targeting can yield savings once efforts are more appropriately tailored.

Although all Asian subgroups certainly were disadvantaged by aggregation of health data, Filipinos in particular had the most disparities identified after disaggregation. Indeed, a key finding of this study is that Filipinos appeared to be in the worst health of all Asian subgroups included in these analyses, which suggests a need to address Filipino-specific needs in future policies and interventions. These findings are consistent with a growing body of research demonstrating a need to address chronic diseases among Filipino Americans.^20–22 Paradoxically, Filipinos are overrepresented in the provision of health^23,24 and long-term care services in the United States^25–27 but, according to the analyses presented here, may face unique drivers of health disparities in their own lives. Further research and interventions are thus needed to specifically address the needs of Filipino Americans.

Importantly, further research is needed to assess the drivers of disparities identified. Social determinants of health were included as covariates,²⁸ although other cultural, behavioral, and structural factors not included in CHIS may also drive disparities. Previous research has suggested that different Asian subgroups have distinct risk profiles in health behaviors,^29,30 which may contribute to the underlying trends observed in this study. Additionally, insurance coverage and annual provider visits may not reflect access to or utilization of care because distinct barriers to care—including unfamiliarity with patient–provider dynamics, perceived discrimination in health care, and a general avoidance of care until absolutely necessary—may limit access and utilization of care, even when patients are insured.^31,32

Qualitative research is necessary, as perspectives into patterns of behavior or other disparities not captured by quantitative research can better inform interpretations of results, targeting for interventions, and intervention content tailored to each Asian subpopulation. Lastly, contextualizing disparities we have identified alongside those faced by other racial groups (e.g., Black and Hispanic Californians) is important for identifying priorities for research and interventions. Future research including the disparities faced by these groups alongside Asian subgroups can aid in the identification of communities or shared needs for interventions that can yield more efficient targeting of resources.

Limitations

There are limitations to this study. First, these findings may not be generalizable outside California, marking a need to study similar indicators in health condition, outcomes, and service access in other areas with high Asian populations. These findings should not be considered representative of all Asians in the United States; New York, Texas, New Jersey, and Hawaii, among other states, also contain substantial Asian populations.

Second, we relied on self-reported metrics of health. Stigmatized health conditions, such as mental health,³³ may be underreported, whereas socially desirable actions, such as seeing a doctor in the past year, may be overreported. Additionally, relying on reported diagnoses of outcomes like high blood pressure and diabetes may substantially underestimate the prevalence of these health outcomes in our sample. For example, previous research suggests that more than half of Asian Americans with diabetes may be undiagnosed.³⁴

Lastly, although we critique aggregation, many Asian groups were included in CHIS but were aggregated into another Asian category because of insufficient sample size. These groups were included in our overall Asian category for analyses, but we could not perform a subgroup-level analysis. Future research focused on members of these groups is warranted and necessary, as disparities among these subpopulations may also be masked by the current practice of aggregating their members into a single Asian category. Additional research is necessary to identify health disparities in these groups as efforts continue to ensure the right to health for every Asian American.

Public Health Implications

Overall, these findings support further data disaggregation in other large-scale research efforts to support interventions tailored specifically to Asian subpopulations in need. Interpreting the aggregated results using a single Asian category would provide little support for the need to dedicate resources to improving the health of any Asians in California. Disaggregation showed that each Asian subgroup faced disparities in health condition, outcomes, and service access that would have been masked. Altogether, disaggregation is a necessary process to ensure that health needs can be properly identified where they exist among diverse sets of Asians in the United States. Stronger and more accurate research and programming might be enabled by greater investment in practices such as oversampling Asian subgroups and providing surveys in Asian languages that can ensure appropriate sample size for analysis by subgroup, as well as a broader commitment toward presenting data by Asian subgroup.