Examining neighbourhood-level disparities in Black, Latina/o, Asian, and White physical health, mental health, chronic conditions, and social disadvantage in California

ABSTRACT

Racial/ethnic minority individuals in the U.S. experience numerous health disparities versus Whites, often due to differences in social determinants. Yet, limited large-scale research has examined these differences at the neighbourhood level. We merged 2021 PLACES Project and 2020 American Community Survey data across 3,211 census tracts (neighbourhoods) defined as majority (>50%) Black, Latina/o, Asian or White. T-tests and hierarchical linear regressions were used to examine differences and associations between neighbourhoods on key health (general health, mental health, obesity, diabetes, cancer, coronary heart disease, chronic obstructive pulmonary disease, stroke), and social outcomes (income, unemployment, age, population density). Results indicated that minority neighbourhoods in California exhibited stark health and social disparities versus White neighbourhoods, displaying worse outcomes on nearly every social and health variable/condition examined; particularly for Black and Latina/o neighbourhoods. Moreover, regression findings revealed that, after considering income, unemployment, and population density, (1) fair/poor mental health and higher percentages of Black, Latina/o and Asian residents in neighbourhoods independently associated with greater neighbourhood fair/poor physical health, and (2) fair/poor mental health significantly associated with greater prevalence of obesity and COPD. This study thus underscores the need to address the profound health and social disparities experienced by minority neighbourhoods for more equitable neighbourhoods.

KEYWORDS:

• Health disparities
• neighbourhood
• social determinants
• mental health

1. Introduction

Racial minorities in the United States are disproportionately affected by poor health and disease relative to their non-Hispanic White counterparts, with these disparities well-documented for decades (Brennan et al., 1990), despite major changes in disease states (e.g. smallpox, HIV/AIDS, COVID-19), medical treatments, and efforts to improve the nation’s health equity over time (National Center for Health Statistics, 2015). This gap is especially stark between non-Hispanic White (i.e. U.S. racial majority) individuals versus Black and Latina/o (i.e. U.S. racial minority) individuals, with health disparities between these groups persisting or even widening over the past two decades (Izenberg et al., 2018; Mahajan et al., 2021; Zimmerman & Anderson, 2019). These disparities are found among individuals in a wide range of health risks and conditions, including infant mortality, cancer, diabetes, obesity, coronary heart disease, chronic obstructive pulmonary disease, and smoking, to name a few (National Center for Health Statistics, 2015).

Although Asian Americans have shown success in certain socioeconomic measures, such as higher aggregate income and education levels, their health status remains uncertain. While some studies suggest that Asian Americans are at higher risk for heart disease, diabetes, and certain conditions such as hepatitis B and liver disease (Office of Minority Health, 2019), others suggest that they may be healthier than other racial groups in some health outcomes, including non-Hispanic Whites (Adia et al., 2020; Office of Minority Health, 2019). However, limited research has been conducted to compare the physical and mental health status and determinants of poor health between different racial groups – including understudied Asian Americans – at the neighbourhood level; leaving significant gaps in our understanding of these disparities.

Moreover, as medicine and public health continue to evolve, the recognition of non-medical social factors as essential contributors to health disparities gains further prominence. Commonly referred to as social determinants of health (SDOH), these factors have been acknowledged to play a fundamental role in perpetuating health disparities by shaping people's health status and health-related behaviours (Frieden, 2010; Kivimäki et al., 2020; Kolak et al., 2020). The World Health Organization has defined SDOH as the conditions in which people are born, grow, work, live, and age, and the wider set of forces and systems that shape their conditions of daily life (World Health Organization, n.d.). Despite this intricate conceptual depiction of SDOH, SDOH are often assessed in studies using socioeconomic measures/indices such as income and employment.

Highlighting the interplay of these factors, a substantial body of research has illuminated the pronounced effects of neighbourhood context on health outcomes. The mounting evidence underscores the pivotal role of residents’ living environments in either fostering or undermining individual/patient-level and population-level health and well-being (Franzini et al., 2005; Frieden, 2010; Gustafsson et al., 2014). Consistently, researchers have identified a compelling link between neighbourhood socioeconomic disadvantage and adverse health consequences for residents. Notably, associations have been observed between low-income neighbourhoods, which is characterised in this study by lower median household income; an important contributor to poor health and mental health status due to its association with a lack of healthy food options (Story et al., 2008), decent housing (Angel & Bittschi, 2019), and limited access to recreational facilities, green areas (Kondo et al., 2018; Wood et al., 2017), and health services (Bailey et al., 2017). Moreover, the substandard built environment in socioeconomically disadvantaged neighbourhoods can significant impact the health and mental health well-being of residents by affecting their personal evaluation of their quality of life (Diener et al., 2018) and increasing the risk of crime and violence residents (Rivara et al., 2019).

Aside from socioeconomic status, other social factors have also been shown to influence residents’ health. Aging is one such factor associated with poor health as aging consists of a gradual process of declining physical and mental capacity (World Health Organization, 2021) that is closely linked with increased prevalence of numerous chronic health conditions. Living in population-dense neighbourhoods may also lead to a lower quality of life (Cramer et al., 2004) and higher incidence and mortality from a range of cancers, coronary heart disease, chronic obstructive pulmonary disease, and asthma (Carnegie et al., 2022) due to increased exposure to unhealthy risk factors such as poor air quality and stressful physical environments (Carnegie et al., 2022; Su et al., 2015). In addition, unemployment has been shown in some studies to associate with poor health outcomes (Norström et al., 2019), although the evidence for this relationship remains equivocal in extant data (Böckerman & Ilmakunnas, 2009; Schmitz, 2011), suggesting a key area for empirical investigation.

In considering these social disparities, it's crucial to acknowledge the structural and systemic factors that underlie them. Among these factors, systemic racism has significantly shaped neighbourhood environments and perpetuating health inequalities, with long-lasting consequences for unequal allocation of resources, practices of housing segregation, limited economic opportunities and access to quality healthcare in communities of colour (Bailey et al., 2017; Gee & Ford, 2011; Williams & Collins, 2016).Therefore, this underscores the pivotal role of broader structural determinants that contribute to these disparities, both in health outcomes and mental health conditions.

It is important to note that despite mental health and physical health being historically viewed and treated in Western contexts and health systems as distinct (Foerschner, 2010; Rose, 1985), growing research suggests they may be interconnected with poor mental health having a negative impact on personal health and well-being (Scott et al., 2016). Further, studies have shown that individuals with mental disorders may be at increased risk for physical health problems such as cardiovascular disease (Prince et al., 2007), obesity (Abdalla et al., 2022), diabetes (Robinson et al., 2018), and some unexplained somatic symptoms (Prince et al., 2007). However, studies examining the potential influence of mental health status on physical health within neighbourhoods are scarce; warranting empirical investigation in this study.

Overall, although various non-medical social factors have received significant research attention in the fields of epidemiology, public health, sociology, and medicine, substantial gaps exist in our knowledge of (1) the nature and scope of neighbourhood-level racial/ethnic health disparities between Black, Latina/o, and Asian versus White residents in highly racially diverse contexts, focusing on California as it possesses the nation’s most racially diverse population, and (2) the effects of key social factors on these neighbourhood-level racial/ethnic health disparities. To address this research gap, this study therefore sought to examine potential neighbourhood-level differences between majority Black, majority Latina/o, majority Asian, and majority non-Hispanic White neighbourhoods across California’s five largest counties on key (1) health outcomes including physical health and mental health status, obesity, diabetes, coronary heart disease (CHD), and chronic obstructive pulmonary disease (COPD), and (2) health-relevant sociodemographic factors including income, unemployment, median age, and population density. We focused our study on Black, Latina/o and Asian populations as they are the three largest racial minority groups in the U.S. – thus providing a strong comparison for non-Hispanic White neighbourhoods in our analyses – while demonstrating key potential demographic differences from non-Hispanic White populations such as elevated rates of poverty (Shrider et al., 2021). To obtain sizable numbers of racial minority neighbourhoods for our large-scale study, we obtained, aggregated, and analyzed geospatial, demographic, and health data to compare social and health outcomes across 3,211 majority Black, Latina/o, Asian, and White neighbourhoods.

2. Materials and methods

2.1. Data acquisition

In this ecological cross-sectional study, data were extracted and aggregated from two national data sources: the American Community Survey (2020) and the PLACES Project (2021)–an extension of the national 500 Cities Project. We used census tracts as the unit of analysis, as they are more socially and economically homogeneous than larger geographic units (e.g. zip codes, cities, counties), and thus served as the ideal unit for examining and determining (1) majority Black, Latina/o, Asian, and White neighbourhoods, and (2) neighbourhood-level effects on our study variables.

The study area for this research consisted of California’s largest five counties – all of which are contiguous within Southern California. Collectively, the study area possessed a combined resident population of 20,936,527 residents in 2021. The five target counties were: Los Angeles County (9,829,544 residents), San Diego County (3,286,069 residents), Orange County (3,167,809 residents), Riverside County (2,458,395 residents), and San Bernardino County (2,194,710 residents).(U.S. Census Bureau, 2020) Respectively, these counties composed the 1st, 5th, 6th, 10th, and 14th largest U.S. counties by population in 2021. In addition to their large overall population sizes, we selected these counties for analysis as they contain the 5 largest Latina/o populations and top 15 largest Black and Asian populations in California (U.S. Census Bureau, 2020).

Across the 3,872 total census tracts within the five counties, we identified census tracts composed of either majority Black (> 50% Black), majority Latina/o (> 50% Latina/o), majority Asian (>50% Asian), or majority White (> 50% non-Hispanic White) residents for analysis, resulting in 3,211 census tracts (282 majority Black, 953 majority Latina/o, 481 majority Asian, 1,495 majority White census tracts) across the five county sample.

2.2. Measures

2.2.1. Sociodemographic variables

To develop our dataset, we downloaded census tract-level sociodemographic data from the American Community Survey (ACS), an ongoing annual survey conducted by the U.S. Census Bureau that provides comprehensive U.S. population and housing information. The sociodemographic variables included mean household income, unemployment rate, population density, and median age. All variables were continuous.

2.2.2. Health outcome variables

We incorporated census tract-level health data from the 2021 PLACES Project. This initiative, co-funded by the Centres for Disease Control and Prevention and the Robert Wood Johnson Foundation, offers model-based estimates for unhealthy behaviours, health outcomes, and clinical preventive services use for the 500 largest U.S. cities. The health outcome variables encompassed the percentage of adult residents (≥18 years) within each census tract who reported (1) fair/poor physical health (‘In general, would you say your health is excellent, very good, good, fair, or poor?’), (2) fair/poor mental health (‘In general, would you say your mental health is excellent, very good, good, fair, or poor?’), (3) obesity, (4) diabetes, (5) cancer, (6) coronary heart disease (CHD), (7) chronic obstructive pulmonary disease (COPD), and (8) stroke. The fair/poor health items were ordinal while the health condition variables were continuous.

2.3. Statistical analyses

By employing the census tract FIPS code for seamless integration, we merged the data downloaded from ACS and the PLACES Project. Subsequently, statistical analyses were conducted using SPSS v.27. Independent t-tests were employed to assess mean differences across majority Black, Latina/o, Asian, and White census tracts for both study sociodemographic variables and health conditions.

We conducted a linear regression analysis to explore the associations between physical health status and our target health conditions (dependent variables) and sociodemographic variables and mental health status (independent variables), offering insights into potential key relationships that may underlie physical health disparities in study census tracts. With a primary focus on examining how health disparities were influenced by the minority racial group composition in each census tract, we included the percentages of Black, Latina/o, and Asian in each census tract as predictor variables in Step 1. In Step 2, we introduced several control variables known to be associated with health outcomes, including mean household income, population density, median age, and unemployment rate. Finally, in Step 3, fair/poor mental health was included to explore whether mental health in neighbourhoods influenced physical health in these neighbourhoods – an avenue that could potentially underpin neighbourhood health improvement. By systematically controlling for these factors at each step, we aimed to isolate the unique contribution of each predictor variable to physical health status, fostering a more nuanced understanding of the intricate interplay between sociodemographic factors and health outcomes. Additionally, regression analyses were conducted with the prevalence of health conditions (obesity, diabetes, cancer, CHD, COPD, and stroke) as dependent variables and the same independent variables used in our fair/poor health model.

3. Results

Table 1 displays a summary of sociodemographic characteristics and prevalence of chronic health conditions among majority Black, Latina/o, Asian, and White census tracts. Independent t-tests revealed that the minority census tracts significantly differed from majority White census tracts on all examined sociodemographic variables (p < .01). Specifically, majority Black, Latina/o and Asian census tracts possessed significantly lower mean incomes (t = −35.35, t = −38.99, t = −20.12, p < .01), higher unemployment rates (t = 19.90, t = 24.02, and t = 7.54, p < .01), higher population densities (t = 18.50, t = 21.32, and t = 12.22, p < .01), and lower median ages (t = −19.16, t = −22.80, and t = −4.64, p < .01) compared to majority White census tracts.

Table 1. Characteristics and prevalence of health conditions across majority Black, Latina/o, and non-Hispanic White census tracts in California’s five largest counties.

Variables	Majority Black (n = 282)	Majority Latina/o (n = 953)	Majority Asian (n = 481)	Majority White (n = 1,495)
Sociodemographic Factors
Mean Household Income	$66,016.21 ± 1,142.45	$70,147.25 ± 590.44	$91,269.05 ± 1370.01	$131,565.33 ± 1,460.59
Population Density (per square mile)	15,668.44 ± 444.41	15,502.99 ± 358.12	14,265.61 ± 580.10	6,892.60 ± 166.14
Median Age	32.46 ± 0.36	33.67 ± 0.21	38.49 ± 0.35	40.39 ± 0.20
Percent Unemployment Rate	9.55% ± 0.22	7.78% ± 0.09	6.01% ± 0.12	5.02% ± 0.06
Chronic Health Conditions (%, 95% CI)
Fair/Poor Physical Health	30.02 (29.31–30.72)	28.31 (27.95–28.68)	21.45 (20.89–22.02)	14.70 (14.49–14.90)
Fair/Poor Mental Health	16.70 (16.50–16.92)	15.66 (15.45–15.86)	12.31 (12.10–12.53)	11.94 (11.83–12.05)
Obesity	34.95 (34.65–35.25)	31.60 (31.35–31.85)	23.33 (22.91–23.76)	24.23 (24.05–24.42)
Diabetes	13.12 (12.90–13.34)	12.02 (11.90–12.14)	11.40 (11.4–11.59)	8.43 (8.32–8.53)
Cancer	4.21 (4.08–4.33)	4.03 (3.99–4.08)	4.71 (4.63–4.80)	6.74 (6.64–6.84)
Coronary Heart Disease	5.09 (4.98–5.20)	4.79 (4.73–4.85)	4.60 (4.52–4.69)	4.81 (4.72–4.90)
Chronic Obstructive Pulmonary Disease	5.64 (5.51–5.77)	5.07 (5.00–5.13)	4.31(4.23–4.40)	4.86 (4.78–4.94)
Stroke	3.78 (3.68–3.87)	3.04 (2.99–3.08)	2.78(2.73–2.83)	2.69 (2.64–2.73)

Bold-faced: The estimate for the group differed from the estimate of non-Hispanic Whites (p < .05)

In addition, there were significant disparities in health outcomes between majority Black, Latina/o, and Asian census tracts compared to majority White census tracts. Specifically, majority Black, Latina/o and Asian census tracts had significantly greater prevalence rates of fair/poor health (t = 41.01, t = 63.92, t = 22.08, p < .01), fair/poor mental health (t = 29.45, t = 39.76, and t = 2.97, p < .01), diabetes (t = 34.06, t = 42.14, and t = 26.39, p < .01), and stroke (t = 20.17, t = 11.50, and t = 2.66, p < .05) versus White census tracts.

Obesity and COPD prevalence were significantly higher in majority Black and Latina/o census tracts compared to majority White census tracts (t = 59.94, t = 46.17; t = 9.92, t = 3.84, p < .01) while majority Asian census tracts had significantly lower prevalence rates of obesity (t = −3.79, p < .01), CHD (t = −3.24, p < .05), and COPD (t = - 9.11, p < .01). Additionally, Black census tracts had significantly higher prevalence rates of CHD (t = 3.97, p < .01) versus White census tracts.

Table 2 shows the results of the linear regression model of fair/poor physical health across all study census tracts. When controlling for sociodemographic factors, higher percentages of Black, Latina/o residents, and Asian residents associated with greater prevalence of fair/poor physical health in study census tracts (p < .001). Notably, fair/poor mental health predicted fair/poor physical health (p < .001). Sociodemographically, mean household income (p < .01), population density (p < .01), median age (p < .001), and unemployment rate (p < .01) negatively associated with fair/poor physical health.

Table 2. Linear regression of fair/poor health.

DV = Fair/Poor Health
Variables	b (SE)	P-Value	95% CI	F	R^2 (ΔR^2)
Majority Racial/Ethnic Group
Percent Non-Hispanic Black	5.91*** (0.32)	.000	[5.29, 6.54]	3658.18	0.76
Percent Hispanic/Latino	12.70*** (0.27)	.000	[12.16,13.24]
Percent Non-Hispanic Asian	4.75*** (0.28)	.000	[4.20,5.30]
Demographics / Social Determinants
Mean Household Income	–4.14E-6** (0.00)	.008	[0.00,0.00]	1944.36	0.80 (0.04)
Population Density	2.98E-5*** (0.00)	.000	[0.00,0.00]
Median Age	0.02** (0.01)	.001	[0.01,0.04]
Percent Unemployment Rate	11.56*** (1.94)	.000	[7.75,15.36]
Fair/Poor Mental Health	1.07*** (0.03)	.000	[1.01,1.14]	2366.75	0.84 (0.04)

* p < .05; **p < .01; *** p < .001; DV = Dependent Variable; b = Unstandardized Regression Coefficient; SE = Standard Error; R²= Coefficient of Determination; ΔR²= Change in Coefficient of Determination

Additional regression analyses with obesity, diabetes, cancer, CHD, COPD and stroke as the dependent variables revealed that after accounting for our neighbourhood sociodemographic and percentage of Black, Hispanic/Latino, and Asian residents, fair/poor mental health status significantly associated with obesity, cancer, and COPD. Further, median age and unemployment rate positively associated with every health condition except obesity, while household income and population density negatively associated with health conditions (Please see supplemental file for the tables).

4. Discussion

The current investigation revealed that across over 3,000 neighbourhoods in California’s 5 largest counties, residents living in predominantly Black and Latina/o neighbourhoods reported higher rates of nearly every health condition examined, except for cancer, relative to residents living in predominantly White neighbourhoods. In particular, residents living in majority Black, Latina/o, and Asian neighbourhoods experienced significantly higher prevalence of poor (1) physical health status with Black and Latina/o neighbourhoods enduring approximately 2 times, and Asian 1.5 times the rates of fair/poor health versus White neighbourhoods, and (2) mental health status with 4.76, 3.72 and 0.37 percentage points higher in Black, Hispanic/Latino, and Asian versus White neighbourhoods, respectively. As expected, within our sample, residents in majority Black and Hispanic/Latino neighbourhoods also experienced higher rates of obesity, diabetes, COPD, and stroke – a primary concern as these health conditions commonly predispose affected individuals to developing a number of chronic diseases that carry heavy burdens of morbidity and mortality.(Braveman et al., 2011; National Center for Health Statistics, 2015; Singh et al., 2017) Concerning Asian neighbourhoods, despite exhibiting some favourable health outcomes, Asian residents had significantly higher rates of poor physical health and mental health status, as well as diabetes and stroke versus their White counterparts.

While our findings of geographically linked health disparities in Black and Latina/o neighbourhoods closely align with prior literature, a unique finding from our study was that despite the pervasive ‘model minority’ stereotype, Asian neighbourhoods in California – the state with the largest U.S. Asian population – experienced a number of key social and health disparities. Belying new data from the National 2023 STAATUS Index (Social Tracking of Asian Americans in the U.S.) that indicated most (61%) White Americans view Asians as closer to White people than people of colour in status – whereas 71% of Asians report viewing themselves as closer to people of colour – our results suggest that the widely held perception of Asian success across social, academic, and economic domains (Yi & Museus, 2015) and the absence of significant health disadvantages (Izenberg et al., 2018) might not accurately reflect the social and health reality of many Asian residents in California. Rather, our data indicate that Asians may represent a vulnerable but under recognised racial population that warrant culturally focused research, programmes, and services to improve health and well-being within Asian communities. Further, given that Asian Americans are highly diverse with different subgroups demonstrating vastly different social and health outcomes (e.g. Southeast Asians vs. East Asians), these efforts should seek to disaggregate Asian American data to better identify and then target services toward the most vulnerable Asian communities.

In addition, perhaps the key finding of this research was the strong relationships revealed between neighbourhood mental health status and a multitude of critical physical health outcomes in our linear regressions. Specifically, we found that even after controlling for our key social determinants and the percentages of Black, Latina/o, and Asian residents in each census tract, for every one-unit increase (on a 5-point scale) in fair/poor mental health (e.g. from ‘poor’ to ‘fair’ or from ‘good’ to ‘very good’) was associated with a (1) 1.07 unit increase in fair/poor physical health (which also used the 5-point scale from ‘poor’ to ‘excellent’); (2) 0.68 unit increase in prevalence of obesity, and (3) 0.25 unit increase in prevalence of COPD. These findings suggest poor mental health status has a robust negative effect on multiple physical health outcomes/conditions at the neighbourhood-level even after controlling for other important physical health neighbourhood risk factors including income, age, unemployment, and greater racial/ethnic minority composition.

Thus, our findings were unique in extending prior research indicating a link between poor mental health and greater rates of disease at the individual level (Abdalla et al., 2022; Prince et al., 2007; Scott et al., 2016) to neighbourhood contexts. Our research reveals a strong potential linkage between poor mental health in neighbourhoods with poor physical health and chronic disease. Thus, our findings highlight the importance of addressing mental health as a potential risk factor for, and avenue to reduce, public health problems such as poor general health, obesity, and COPD affecting California’s diverse resident population. Thus, given the growing prevalence of mental health problems in nationally and in California, and the associated burden of chronic diseases, our study underscores the need for targeted mental health or integrated care interventions that address mental and physical health holistically, particularly for underserved minority communities.

While our findings indicate that residents in majority Black, Hispanic/Latina/o, and Asian neighbourhoods endure a variety of health disparities relative to residents in White neighbourhoods, it is especially notable that residents in Black, Latina/o, and Asian neighbourhoods were, on average, about 8, 6, and 2 years younger than residents in White neighbourhoods. Because age is one of the most profound risk factors for poor physical health and chronic disease (World Health Organization, 2021), the fact that residents in these neighbourhoods are significantly younger than their White peers – who should therefore demonstrate greater relative risk for poor health and disease than racial minorities – suggests a potentially critical role of social factors in driving racial health disparities in our study sample.

Unsurprisingly then, given the consistent relationship shown in the literature between health disparities and disproportionate exposure to social risk factors (Kivimäki et al., 2020), residents in majority Black, Latina/o, and Asian neighbourhoods in California reported numerous socioeconomic disparities, including significantly lower mean household incomes, higher unemployment rates, and greater population densities versus majority White neighbourhoods. Most notably, residents in majority Black, Hispanic/Latino, and Asian neighbourhoods had (1) population densities that were over twice those of White neighbourhoods, (2) higher unemployment rates (9.55%, 7.78%, 6.01% versus 5.02%), and (3) for Black and Hispanic/Latino neighbourhoods, household incomes that were almost half those in White neighbourhoods while residents in majority Asian neighbourhoods had incomes 69% lower than White neighbourhoods. This disparate pattern of lower incomes and higher population densities and unemployment rates in the social and built environment suggests that racial minority neighbourhoods experience severe relative social disadvantage versus majority Whites neighbourhoods in California. These socioeconomic disparities may partially explain the persistent health disparities across numerous health conditions found between the racial minority versus racial majority neighbourhoods in our study.

Our results underlined the strong relationships between the uneven distribution of social risk factors within our study neighbourhoods and disparate racial and ethnic health outcomes among residents (Braveman et al., 2011). Specifically, neighbourhood unemployment rate, population density, and household income were independently associated with greater prevalence of every health condition studied, including poor physical health status in study neighbourhoods. Further, even after controlling for these social risk factors in our primary regression model, higher percentages of Black, Latina/o, or Asian residents in neighbourhoods independently associated with a greater prevalence of neighbourhood fair/poor health. Consequently, we speculate that other critical social determinants not accounted for in our analyses, such as potential exposure to discrimination, racism, and racial segregation experienced by residents in majority Black, Latina/o, and Asian neighbourhoods, which are known to have health-depriving consequences on minority individuals (Ortega & Roby, 2021), coupled with potentially reduced access to health-protective resources (e.g. affordable healthy foods, safe recreational spaces for physical activity, adequate health care) (Douglas et al., 2018; Frieden, 2010; Subica et al., 2016) may have further contributed to the observed health disparities found in our study.

Accordingly, by conducting our analyses at the neighbourhood (versus individual resident) level, our study findings confirm that for residents across California’s five largest counties, living in communities with higher proportions of racial minority residents (whether Black, Hispanic/Latino, or Asian) was associated with greater rates of poor physical health, mental health, and chronic disease exposure as well as greater exposure to social risk factors such as low income and unemployment.

Additionally, it is worth noting that while majority Black, Latina/o, and Asian neighbourhoods experienced sizable social and health disparities – to the degree that Black, Latina/o, and Asian residents in these neighbourhoods are health-poor relative to their White counterparts – residents living in majority Black neighbourhoods experienced the worst health and social disparities in our study, consistent with previous literature in California (Ellis et al., 2018; Izenberg et al., 2018). These disparities include possessing the lowest mean incomes, highest unemployment rates and population densities, and greatest prevalence of every health condition examined except cancer of the four racial/ethnic groups. Consequently, policymakers, practitioners, and researchers should seek to reduce these disparities by developing focused policies and interventions to improve the health and well-being of residents in majority Black California neighbourhoods.

Several limitations to this study should be acknowledged. Firstly, this study's ecological cross-sectional design provides valuable insights into neighbourhood-level patterns but cannot establish individual-level causal relationships. Using aggregated data at the neighbourhood level restricts our ability to infer causality between individual variables and health outcomes. Therefore, caution should be exercised when interpreting findings in terms of causation. Additionally, while using a logic-based threshold of over 50 percent to identify majority racial and ethnic neighbourhoods can increase the validity and reliability of findings, as in prior studies by our team (Subica et al., 2016), it may not accurately capture the true diversity and composition of a census tract, potentially leading to misclassification and underestimation of health disparities. Thus, future research should compare different levels of Black, Latina/o, Asian, and White neighbourhood compositions to more thoroughly understand the effects of race/ethnicity on neighbourhood-level health disparities. Finally, using census tracts as the unit of analysis may not be the most optimal, and examining alternative units of analysis such as blocks, cities, and counties may offer additional information linking neighbourhood-level race/ethnicity to community health and social disparities.

5. Conclusion

Overall, this study’s findings indicate that residents living in majority Black, Latina/o, and Asian neighbourhoods in the largest counties in California endure stark health and social disparities versus those living in majority White neighbourhoods. Specifically, residents living in majority Black, Hispanic/Latina/o, and Asian neighbourhoods were more likely to be socioeconomically disadvantaged and suffer from heightened prevalence of poor physical and mental health status and chronic diseases. Additionally, belying the well-established model minority stereotype that has long affected Asians in the U.S., residents living in Asian communities evidenced disparities in every social risk factor examined, including greater unemployment and poorer socioeconomic status compared to White communities. The reduced health status experienced by many residents in the examined racial minority neighbourhoods – expected to become an increasing part of the nation’s social landscape as the U.S. becomes a majority-minority nation by 2045 (Frey, 2018) – were associated with similar inequities in income, unemployment, and population density. As such, these findings add to the growing body of literature linking neighbourhood social and racial composition to health, reflecting the harmful effects of racial segregation, a fundamental cause of health disparities(Williams & Collins, 2016). Consequently, this study provides valuable empirical evidence to support further research and intervention efforts to promote health equity in Black, Latina/o and Asian communities.

Disclosure Statement

No potential conflict of interest was reported by the author(s).