Development and Assessment of a Social Media-Based Construct of Firearm Ownership: Computational Derivation and Benchmark Comparison.

BACKGROUND: Gun violence research is characterized by a dearth of data available for measuring key constructs. Social media data may offer a potential opportunity to significantly reduce that gap, but developing methods for deriving firearms-related constructs from social media data and understanding the measurement properties of such constructs are critical precursors to their broader use. OBJECTIVE: This study aimed to develop a machine learning model of individual-level firearm ownership from social media data and assess the criterion validity of a state-level construct of ownership. METHODS: We used survey responses to questions on firearm ownership linked with Twitter data to construct different machine learning models of firearm ownership. We externally validated these models using a set of firearm-related tweets hand-curated from the Twitter Streaming application programming interface and created state-level ownership estimates using a sample of users collected from the Twitter Decahose application programming interface. We assessed the criterion validity of state-level estimates by comparing their geographic variance to benchmark measures from the RAND State-Level Firearm Ownership Database. RESULTS: We found that the logistic regression classifier for gun ownership performs the best with an accuracy of 0.7 and an F1-score of 0.69. We also found a strong positive correlation between Twitter-based estimates of gun ownership and benchmark ownership estimates. For states meeting a threshold requirement of a minimum of 100 labeled Twitter users, the Pearson and Spearman correlation coefficients are 0.63 (P<.001) and 0.64 (P<.001), respectively. CONCLUSIONS: Our success in developing a machine learning model of firearm ownership at the individual level with limited training data as well as a state-level construct that achieves a high level of criterion validity underscores the potential of social media data for advancing gun violence research. The ownership construct is an important precursor for understanding the representativeness of and variability in outcomes that have been the focus of social media analyses in gun violence research to date, such as attitudes, opinions, policy stances, sentiments, and perspectives on gun violence and gun policy. The high criterion validity we achieved for state-level gun ownership suggests that social media data may be a useful complement to traditional sources of information on gun ownership such as survey and administrative data, especially for identifying early signals of changes in geographic patterns of gun ownership, given the immediacy of the availability of social media data, their continuous generation, and their responsiveness. These results also lend support to the possibility that other computationally derived, social media-based constructs may be derivable, which could lend additional insight into firearm behaviors that are currently not well understood. More work is needed to develop other firearms-related constructs and to assess their measurement properties.

Association Between Receipt of Definitive Treatment for Localized Prostate Cancer and Adverse Health Outcomes: A Claims-Based Approach.

OBJECTIVES: This study aimed to examine adverse health outcomes associated with receipt of definitive treatments (prostatectomy, intensity-modulated radiation therapy [IMRT] and brachytherapy). METHODS: We identified men aged 65 years and older who received a new diagnosis of localized prostate cancer from 4 state cancer registries (CA, FL, NJ, and TX) during the years 2006 to 2013. We merged the registry records for this cohort with Medicare enrollment and claims. We constructed indicators of treatment-related adverse outcomes using diagnosis codes reported on the claims. Stage 1 models the choice of definitive treatment versus active surveillance. Stage 2 examines the probability of experiencing a treatment-related adverse health outcome among men who chose definitive treatment. RESULTS: Notably, 81.4% of our cohort of 61 187 men received definitive treatment whereas 18.6% were monitored with active surveillance. The 5-year prostate cancer death rate was 0.28% to 1.75% irrespective of treatment received. Men monitored with active surveillance experienced minimal adverse health outcomes (0.16%-0.75%). The risks of urinary incontinence associated with prostatectomy were 31 and 39.5 percentage points higher than brachytherapy and IMRT, respectively. For erectile dysfunction, the risks were nearly 23 and 27.5 percentage points higher, respectively, than brachytherapy and IMRT. Prostatectomy was associated with lower risk of urinary dysfunction and bowel dysfunction than either brachytherapy or IMRT. Compared with brachytherapy, IMRT was associated with a lower risk of erectile dysfunction (32%), urinary incontinence (84%), and urinary dysfunction (30%). CONCLUSIONS: This evidence should be of value to patient-physician decision making regarding the choice of definitive treatments versus active surveillance for men with localized disease.

Assessing Social Media Data as a Resource for Firearm Research: Analysis of Tweets Pertaining to Firearm Deaths.

BACKGROUND: Historic constraints on research dollars and reliable information have limited firearm research. At the same time, interest in the power and potential of social media analytics, particularly in health contexts, has surged. OBJECTIVE: The aim of this study is to contribute toward the goal of establishing a foundation for how social media data may best be used, alone or in conjunction with other data resources, to improve the information base for firearm research. METHODS: We examined the value of social media data for estimating a firearm outcome for which robust benchmark data exist-specifically, firearm mortality, which is captured in the National Vital Statistics System (NVSS). We hand curated tweet data from the Twitter application programming interface spanning January 1, 2017, to December 31, 2018. We developed machine learning classifiers to identify tweets that pertain to firearm deaths and develop estimates of the volume of Twitter firearm discussion by month. We compared within-state variation over time in the volume of tweets pertaining to firearm deaths with within-state trends in NVSS-based estimates of firearm fatalities using Pearson linear correlations. RESULTS: The correlation between the monthly number of firearm fatalities measured by the NVSS and the monthly volume of tweets pertaining to firearm deaths was weak (median 0.081) and highly dispersed across states (range -0.31 to 0.535). The median correlation between month-to-month changes in firearm fatalities in the NVSS and firearm deaths discussed in tweets was moderate (median 0.30) and exhibited less dispersion among states (range -0.06 to 0.69). CONCLUSIONS: Our findings suggest that Twitter data may hold value for tracking dynamics in firearm-related outcomes, particularly for relatively populous cities that are identifiable through location mentions in tweet content. The data are likely to be particularly valuable for understanding firearm outcomes not currently measured, not measured well, or not measurable through other available means. This research provides an important building block for future work that continues to develop the usefulness of social media data for firearm research.

The Influence of Practice Structure on Urologists' Treatment of Men With Low-Risk Prostate Cancer.

BACKGROUND: Vertical and horizontal integration among health care providers has transformed the practice arrangements under which many physicians work. OBJECTIVE: To examine the influence of type of practice structure, and by implication the financial incentives associated with each structure, on treatment received among men newly diagnosed with low-risk prostate cancer. RESEARCH DESIGN: We compiled a unique database from cancer registry records from 4 large states, Medicare enrollment and claims for the years 2005-2014 and SK & A physician surveys corroborated by extensive internet searches. We estimated a multinomial logit model to examine the influence of urologist practice structure on type of initial treatment received. RESULTS: The probability of being monitored with active surveillance was 7.4% and 4.2% points higher for men treated by health system and nonhealth system employed urologists ( P <0.01), respectively, in comparison to men treated by single specialty urology practices. Among multispecialty practices, the rate of active surveillance use was 3% points higher compared with single specialty urology practices( P <0.01). Use of intensity modulated radiation therapy among urologists with ownership in intensity modulated radiation therapy was 17.4% points higher compared with urologists working in small single specialty practices. CONCLUSIONS: Physician practice structure attributes are significantly associated with type of treatment received but few studies control for such factors. Our findings-coupled with the observation that urologist practice structure shifted substantially over this time period due to mergers of small urology groups-provide one explanation for the limited uptake of active surveillance among men with low-risk disease in the US.

Documenting Horizontal Integration Among Urologists Who Treat Prostate Cancer.

Relatively little is known about the extent and effects of horizontal mergers among physician specialists. We developed and implemented a methodology to document changes in physician practice structure resulting from horizontal integration among urology groups. We merged cancer registry records from four large states with Medicare Part B claims to identify all urologists who treated men with prostate cancer. We added information from SK & A surveys and extensive internet searches to assign a practice structure to each urologist-year (2005-2014). Horizontal integration among small urology groups led to a sharp increase in the proportion of urologists who belong to large urology practices with ownership in intensity modulated radiation therapy and/or anatomical pathology services. By 2014, more than half of New Jersey urologists and about 43% of urologists in Florida and Texas were members of such large practices, whereas small percentages (7%-16%) were employed by a health system. In contrast, more than 27% of California urologists were employed but only 17.5% had ownership in intensity modulated radiation therapy and/or pathology services. Importantly, we found our indicators of market share of urologists associated with each practice structure type were highly concordant with indicators of market share based on number of prostate cancer episodes treated by each practice structure type.

Effect of early-stage Alzheimer's disease on household financial outcomes.

Significant limitations and rapid declines in financial capacity are a hallmark of patients with early-stage Alzheimer's disease (AD). We use linked Health and Retirement Study and Medicare claims data spanning 1992-2014 to examine the effect of early-stage AD, from the start of first symptoms to diagnosis, on household financial outcomes. We estimate household fixed-effects models and examine continuous measures of liquid assets and net wealth, as well as dichotomous indicators for a large change in either outcome. We find robust evidence that early-stage AD places households at significant risk for large adverse changes in liquid assets. Further, we find some, but more limited, evidence that early-stage AD reduces net wealth. Our findings are consequential because financial vulnerability during the disease's early-stage impacts the ability of afflicted individuals and their families to pay for care in the disease's later stage. Additionally, the findings speak to the value that earlier diagnosis may provide by helping avert adverse financial outcomes that occur before the disease is currently diagnosable with available tools. These results also point to a potentially important role for financial institutions in helping reduce exposure of vulnerable elderly to poor outcomes.

Health Care Experiences of Latino Children in Emerging and Traditional Destinations.

BACKGROUND: Latino populations have disproportionately moved to areas in the Southeast and Midwest (emerging destinations), outside traditional Latino enclaves, in recent years. OBJECTIVE: To examine whether health care experiences differ between traditional and emerging destinations for Latino children. RESEARCH DESIGN: We defined traditional and emerging destination counties based on levels and changes in the Latino population between 2000 and 2010. Measures were linked to the restricted National Survey of Children's Health along with county-level data from the Area Resource File and Census of State and Local Governments. To compare outcomes among Latino children across destinations, linear probability models adjusted for individual-level characteristics, area-level characteristics, and state fixed effects. MEASURES: Outcomes were access to a usual source of care, unmet health care needs, preventive health visit in prior year, and family-centered care. RESULTS: Compared with traditional destination counterparts, Latino children in emerging destinations tended to be younger, healthier, and more likely to be in families speaking English at home. Latino children in emerging destinations were significantly less likely to have a usual source of care adjusting for individual-level and county-level variables, but other dimensions of access were similar between destinations. CONCLUSIONS: Differences in usual source of care may reflect lower supply of culturally competent providers or limited information about where and how to seek care. For realized access to care, protective factors in emerging destinations, such as higher average incomes in the area and lower community uninsurance rates, might counteract any negative effects of emerging destinations.

Harnessing Private-Sector Innovation to Improve Health Insurance Exchanges.

Overhauling the individual health insurance market-including through the creation of health insurance exchanges-was a key component of the Patient Protection and Affordable Care Act's multidimensional approach to addressing the long-standing problem of the uninsured in the United States. Despite succeeding in enrolling millions of Americans, the exchanges still face several challenges, including poor consumer experience, high operational and development costs, and incomplete market penetration. In light of these challenges, analysts considered a different model for the exchanges-privately facilitated exchanges-which could address these challenges and deepen the Affordable Care Act's impact. In this model, the government retains control over sovereign exchange functions but allows the private sector to assume responsibility for more-peripheral exchange functions, such as developing and sustaining exchange websites. Although private-sector entities have already undertaken exchange-related functions on a limited basis, privately facilitated exchanges could conceivably relieve the government of its responsibility for front-end website operations and consumer decision-support functions entirely. A shift to privately facilitated exchanges could improve the consumer experience, increase enrollment, and lower costs for state and federal governments. A move to such a model requires, nonetheless, managing its risks, such as reduced consumer protection, increased consumer confusion, and the possible lack of a viable revenue base for privately facilitated exchanges, especially in less populous states. On net, the benefits are large enough and the risks sufficiently manageable to seriously consider such a shift. This paper provides background information and more detail on the analysts' assessment.

Plasma 24-metabolite Panel Predicts Preclinical Transition to Clinical Stages of Alzheimer's Disease.

We recently documented plasma lipid dysregulation in preclinical late-onset Alzheimer's disease (LOAD). A 10 plasma lipid panel, predicted phenoconversion and provided 90% sensitivity and 85% specificity in differentiating an at-risk group from those that would remain cognitively intact. Despite these encouraging results, low positive predictive values limit the clinical usefulness of this panel as a screening tool in subjects aged 70-80 years or younger. In this report, we re-examine our metabolomic data, analyzing baseline plasma specimens from our group of phenoconverters (n = 28) and a matched set of cognitively normal subjects (n = 73), and discover and internally validate a panel of 24 plasma metabolites. The new panel provides a classifier with receiver operating characteristic area under the curve for the discovery and internal validation cohort of 1.0 and 0.995 (95% confidence intervals of 1.0-1.0, and 0.981-1.0), respectively. Twenty-two of the 24 metabolites were significantly dysregulated lipids. While positive and negative predictive values were improved compared to our 10-lipid panel, low positive predictive values provide a reality check on the utility of such biomarkers in this age group (or younger). Through inclusion of additional significantly dysregulated analyte species, our new biomarker panel provides greater accuracy in our cohort but remains limited by predictive power. Unfortunately, the novel metabolite panel alone may not provide improvement in counseling and management of at-risk individuals but may further improve selection of subjects for LOAD secondary prevention trials. We expect that external validation will remain challenging due to our stringent study design, especially compared with more diverse subject cohorts. We do anticipate, however, external validation of reduced plasma lipid species as a predictor of phenoconversion to either prodromal or manifest LOAD.

Income eligibility thresholds, premium contributions, and children's coverage outcomes: a study of CHIP expansions.

OBJECTIVE: To understand the effects of Children's Health Insurance Program (CHIP) income eligibility thresholds and premium contribution requirements on health insurance coverage outcomes among children. DATA SOURCES: 2002-2009 Annual Social and Economic Supplements of the Current Population Survey linked to data from multiple secondary data sources. STUDY DESIGN: We use a selection correction model to simultaneously estimate program eligibility and coverage outcomes conditional upon eligibility. We simulate the effects of three premium schedules representing a range of generosity levels and the effects of income eligibility thresholds ranging from 200 to 400 percent of the federal poverty line. PRINCIPAL FINDINGS: Premium contribution requirements decrease enrollment in public coverage and increase enrollment in private coverage, with larger effects for greater contribution levels. Our simulation results suggest minimal changes in coverage outcomes from eligibility expansions to higher income families under premium schedules that require more than a modest contribution (medium or high schedules). CONCLUSIONS: Our simulation results are useful counterpoints to previous research that has estimated the average effect of program expansions as they were implemented without disentangling the effects of premiums or other program features. The sensitivity to premiums observed suggests that although contribution requirements may be effective in reducing crowd-out, they also have the potential, depending on the level of contribution required, to nullify the effects of CHIP expansions entirely. The persistence of uninsurance among children under the range of simulated scenarios points to the importance of Affordable Care Act provisions designed to make the process of obtaining coverage transparent and navigable.

Opportunities and challenges in supply-side simulation: physician-based models.

OBJECTIVE: To provide a conceptual framework and to assess the availability of empirical data for supply-side microsimulation modeling in the context of health care. DATA SOURCES: Multiple secondary data sources, including the American Community Survey, Health Tracking Physician Survey, and SK&A physician database. STUDY DESIGN: We apply our conceptual framework to one entity in the health care market-physicians-and identify, assess, and compare data available for physician-based simulation models. PRINCIPAL FINDINGS: Our conceptual framework describes three broad types of data required for supply-side microsimulation modeling. Our assessment of available data for modeling physician behavior suggests broad comparability across various sources on several dimensions and highlights the need for significant integration of data across multiple sources to provide a platform adequate for modeling. A growing literature provides potential estimates for use as behavioral parameters that could serve as the models' engines. Sources of data for simulation modeling that account for the complex organizational and financial relationships among physicians and other supply-side entities are limited. CONCLUSIONS: A key challenge for supply-side microsimulation modeling is optimally combining available data to harness their collective power. Several possibilities also exist for novel data collection. These have the potential to serve as catalysts for the next generation of supply-side-focused simulation models to inform health policy.

Monitoring Cancer Outcomes Across the Continuum: Data Synthesis and Analysis for the District of Columbia.

This article synthesizes and analyzes available data regarding cancer-related outcomes among District of Columbia residents, highlighting key findings and data gaps across the continuum of cancer prevention, treatment, and outcomes and noting variability across subgroups of District residents. Data sources used in this report include the Behavioral Risk Factor Surveillance System, the National Cancer Database, the Centers for Disease Control and Prevention and National Cancer Institute Wide-Ranging Online Data for Epidemiologic Research database, and the American Community Survey. The findings reveal disparities in cancer-related outcomes between black and white District residents across the continuum. First, black District residents are more likely than white residents to be uninsured. Lack of insurance is associated with lower rates of routine cancer screening among asymptomatic patients and may delay care for patients experiencing early symptoms of cancer. In addition, the rate of smoking is significantly higher, and the rate of exercise is significantly lower, among black District residents than among white residents, placing black residents at higher risk of the most common cancers. Limitations in general access to health care, in primary and secondary prevention, and in access to cancer-related treatment all likely contribute to dramatically higher cancer incidence and mortality among black residents of the District than among white residents.

Health care experiences of Hispanics in new and traditional U.S. destinations.

Over the past 15 years, striking new settlement patterns have emerged that have brought about unprecedented geographic dispersion in the population of approximately 45 million Hispanics in the United States. In this study, the authors compare the health care experiences of working age U.S.-born Mexican Americans and Mexican immigrants living in new and traditional Hispanic destinations. They use a geocoded version of the Medical Expenditure Panel Survey Household Component linked to contextual data from secondary sources. They characterize destinations as new or traditional using information on the percentage of the population that was Hispanic in 1990 and the growth in percent Hispanic between 1990 and 2000. The authors find that, compared with living in destinations with a well-established Hispanic presence, U.S.-born Mexican Americans living in new destinations have less favorable health care outcomes, including a greater probability of having an unmet need for or delay in receiving medical care and reduced satisfaction with care.

Take-up of public insurance and crowd-out of private insurance under recent CHIP expansions to higher income children.

OBJECTIVE: To analyze the effects of states' expansions of Children's Health Insurance Program (CHIP) eligibility to children in higher income families on health insurance coverage outcomes. DATA SOURCES: 2002-2009 Current Population Survey linked to multiple secondary data sources. STUDY DESIGN: Instrumental variables estimation of linear probability models. Outcomes are whether the child had any public insurance, any private insurance, or no insurance coverage during the year. PRINCIPAL FINDINGS: Among children in families with incomes between two and four times the federal poverty line (FPL), four enrolled in CHIP for every 100 who became eligible. Roughly half of the newly eligible children who took up public insurance were previously uninsured. The upper bound "crowd-out" rate was estimated to be 46 percent. CONCLUSIONS: The CHIP expansions to children in higher income families were associated with limited uptake of public coverage. Our results additionally suggest that there was crowd-out of private insurance coverage.

Behavioral Health in the District of Columbia: Assessing Need and Evaluating the Public System of Care.

This article shares findings from a study of the public behavioral health care system in the District of Columbia, including the prevalence of mental health disorders and substance use, the organization and financing of public behavioral health services, utilization of public behavioral health services, and priorities for improvement. The authors' analyses found that prevalence of mental health conditions resembles patterns nationally, among both adults and youth. Substance use disorders are more prevalent among adults and comparatively lower for the youth population, compared to national patterns. Potentially 60 percent of adults and 72 percent of adolescents enrolled in Medicaid managed care may have unmet need for depression services. Based on claims data, 45 percent of children and 41 percent of adults enrolled in Mental Health Rehabilitation Services programs have gaps in care that exceed six months during a 12-month period. Participants in focus groups and stakeholder interviews highlighted such challenges as gaps in care and difficulties in coordination of care for particular populations and services. High-level priorities include reducing unmet need for public mental health care, tracking and coordinating care, improving the availability and accessibility of substance use treatment services, and upgrading the data infrastructure.

Understanding disparities in health care access--and reducing them--through a focus on public health.

Attempts to explain disparities in access to health care faced by racial and ethnic minorities and other underserved populations often focus on individual-level factors such as demographics, personal health beliefs, and health insurance status. This article proposes an examination of these disparities-and an effort to redress them-through the lens of public health. Public health agencies can link people to needed services such as immunizations, testing, and treatment; ensure the availability of health care; ensure the competency of the public health and personal health care workforce; and evaluate the effectiveness, accessibility, and quality of personal and population-based services. Approaching disparities through a public health framework can provide the foundation for developing more robust evidence to inform additional policies for improving access and reducing disparities.

Spillover effects of community uninsurance on working-age adults and seniors: an instrumental variables analysis.

BACKGROUND: : Previous research suggests, but does not definitively establish, that a high level of uninsurance in a community may negatively affect access to and quality of health care for insured persons. OBJECTIVE: : To assess the effect of the level of uninsurance in a community on access to and satisfaction with care-an important dimension of quality-among insured persons. RESEARCH DESIGN: : The 1996 to 2006 Medical Expenditure Panel Survey Household Component data linked to data from the Current Population Survey, Area Resource File, and the InterStudy Competitive Edge. Analyses include 86,928 insured adult respondents living in approximately 200 large metropolitan areas. MAIN OUTCOME MEASURES: : Measures of whether an individual had a usual source of care, had any delay/difficulty obtaining needed care, used office-based services, used prescription drug services, and used any medical services, and measures of satisfaction with care. RESULTS: : Among privately insured adults, a higher community uninsurance rate resulted in a lower probability of having a usual source of care, having an office-based visit, having any medical expenditures, and reporting being satisfied with the quality of care provided by the usual source of care. A higher community uninsurance rate also led to a higher probability of reporting difficulty obtaining needed care. Among Medicare enrollees, a higher community uninsurance rate resulted in lower reported satisfaction with care and higher probability of experiencing difficulty or delay in getting needed care. CONCLUSIONS: : Our results suggest substantial spillover effects of the community uninsurance rate on access to and satisfaction with health care among insured working-age adults and seniors. Consequently, new efforts to address the problem of the uninsured may bring significant benefits to persons who already have insurance.

Relative Affordability of Health Insurance Premiums under CHIP Expansion Programs and the ACA.

Affordability is integral to the success of health care reforms aimed at ensuring universal access to health insurance coverage, and affordability determinations have major policy and practical consequences. This article describes factors that influenced the determination of affordability benchmarks and premium-contribution requirements for Children's Health Insurance Program (CHIP) expansions in three states that sought to universalize access to coverage for youth. It also compares subsidy levels developed in these states to the premium subsidy schedule under the Affordable Care Act (ACA) for health insurance plans purchased through an exchange. We find sizeable variability in premium-contribution requirements for children's coverage as a percentage of family income across the three states and in the progressivity and regressivity of the premium-contribution schedules developed. These findings underscore the ambiguity and subjectivity of affordability standards. Further, our analyses suggest that while the ACA increases the affordability of family coverage for families with incomes below 400 percent of the federal poverty level, the evolution of CHIP over the next five to ten years will continue to have significant implications for low-income families.

Employer Self-Insurance Decisions and the Implications of the Patient Protection and Affordable Care Act as Modified by the Health Care and Education Reconciliation Act of 2010 (ACA).

The Patient Protection and Affordable Care Act as amended by the Health Care and Education Reconciliation Act of 2010 (ACA) changes the regulatory environment within which health insurance policies on the small-group market are bought and sold. New regulations include rate bands that limit premium price variation, risk-adjustment policies that will transfer funds from low-actuarial-risk to high-actuarial-risk plans, and requirements that plans include "essential health benefits." While the new regulations will be applied to all non-grandfathered fully insured policies purchased by businesses with 100 or fewer workers, self-insured plans are exempt from these regulations. As a result, some firms may have a stronger incentive to offer self-insured plans after the ACA takes full effect. In this article we identify factors that influence employers' decisions to self-insure and estimate how the ACA will influence self-insurance rates. We also consider the implications of higher self-insurance rates for adverse selection in the non-self-insured small-group market and whether enrollees in self-insured plans receive different benefits than enrollees in fully-insured plans. Results are based on data analysis, literature review, findings from discussions with stakeholders, and microsimulation analysis using the COMPARE model. Overall, we find little evidence that self-insured plans differ systematically from fully insured plans in terms of benefit generosity, price, or claims denial rates. Stakeholders expressed significant concern about adverse selection in the health insurance exchanges due to regulatory exemptions for self-insured plans. However, our microsimulation analysis predicts a sizable increase in self-insurance only if comprehensive stop-loss policies become widely available after the ACA takes full effect and the expected cost of self-insuring with stop-loss is comparable to the cost of being fully insured in a market without rating regulations.

Community demographics and access to health care among U.S. Hispanics.

OBJECTIVE: To explore the influence of the communities in which Hispanics live on their access to health care. DATA: 1996-2002 Medical Expenditure Panel Survey data, linked to secondary data sources and including 14,504 observations from 8,371 Mexican American respondents living in metropolitan areas. STUDY DESIGN: We use multivariate probit regression models, stratified by individuals' insurance status, for analyses of four dependent variables measuring access to health care. We measure community characteristics at the zip code tabulation area level, and key independent variables of interest are the percentage of the population that speaks Spanish and percentage of the population that is immigrant Hispanic. Each of these measures is interacted with individual-level measures of nativity and length of U.S. residency. PRINCIPAL FINDINGS: For Mexican American immigrants, living in an area populated by relatively more Spanish speakers or more Hispanic immigrants is associated with better access to care. The associations are generally stronger for more recent immigrants compared with those who are better established. Among U.S.-born Mexican Americans who are uninsured, living in areas more heavily populated with Spanish-speaking immigrants is negatively associated with access to care. CONCLUSIONS: The results suggest that characteristics of the local population, including language and nativity, play an important role in access to health care among U.S. Hispanics, and point to the need for further study, including analyses of other racial and ethnic groups, using different geographic constructs for describing the local population, and, to the extent possible, more specific exploration of the mechanisms through which these characteristics may influence access to care.