What accounts for differences in uninsurance rates across communities?

Causal connections between socioeconomic disparities and COVID-19 in the USA

With the increasing use of machine learning models in computational socioeconomics, the development of methods for explaining these models and understanding the causal connections is gradually gaining importance. In this work, we advocate the use of an explanatory framework from cooperative game theory augmented with do calculus, namely causal Shapley values. Using causal Shapley values, we analyze socioeconomic disparities that have a causal link to the spread of COVID-19 in the USA. We study several phases of the disease spread to show how the causal connections change over time. We perform a causal analysis using random effects models and discuss the correspondence between the two methods to verify our results. We show the distinct advantages a non-linear machine learning models have over linear models when performing a multivariate analysis, especially since the machine learning models can map out non-linear correlations in the data. In addition, the causal Shapley values allow for including the causal structure in the variable importance computed for the machine learning model.

Prescription Drug Insurance and Cost-Related Medication Nonadherence Among Lesbian, Gay, and Bisexual Individuals in Canada

Purpose: This study estimates the frequency of uninsurance for prescription drugs and cost-related medication nonadherence (CRNA) among lesbian, gay, and bisexual (LGB) persons in Canada, compared with the heterosexual population. Methods: Logistic regression was used to quantify associations between sexual orientation, insurance status, and CRNA within the national probability-based Canadian Community Health Survey, 2015-2016. This sample included 98,413 individuals aged 15-80 years, including 2803 LGB individuals. Results: From our sample of Canadians, 22.2% of LGB respondents reported being uninsured for prescription drugs, compared with 20.0% of heterosexual persons (unadjusted odds ratio [UOR] 1.00, 95% confidence interval [CI] 0.75-1.33). LGB individuals had more than twice the odds of reporting CRNA compared with heterosexual individuals (UOR 2.48, 95% CI 1.99-3.10). This disparity was most pronounced among bisexual respondents, who had over three times the odds of reporting CRNA in comparison to heterosexual respondents (UOR 3.45, 95% CI 2.65-4.51). The odds ratio (OR) for CRNA comparing bisexual with heterosexual individuals remained statistically significant after adjustment for race/ethnicity, gender/sex, and age (OR 2.67, 95% CI 1.97-3.61) and was further attenuated with adjustment for partnership status, employment status, income, educational attainment, prescription drug insurance status, general health status, and immigration status (OR 2.09, 95% CI 1.51-2.89). Conclusion: LGB Canadians reported more CRNA but comparable prescription drug insurance frequencies to heterosexual persons. Factors pertaining to medication access (e.g., income, partnership status) and health needs appear to be the most important contributors to disparities.

Neighborhood context and hypertension outcomes among Latinos in Chicago

Although a health advantage in hypertension has been documented among Latinos, this advantage appears to be eroding. Of particular concern is the observation that Latinos are less likely to be screened and treated for hypertension and to having it controlled. Scholars have suggested that, above and beyond individual-level factors, neighborhood characteristics may be important predictors of health and health care. We analyzed 2001-2003 data from the Chicago Community Adult Health Study to examine (a) the relationship between the Latino and immigrant composition of neighborhoods in Chicago and several outcomes among Latinos: having hypertension, utilizing hypertension-related health care, and being treated for hypertension; and (b) whether there was a differential effect of neighborhood Latino/immigrant concentration by language of interview and nativity status. We controlled for additional neighborhood characteristics relevant to hypertension and to the availability and accessibility of health care resources. Neighborhoods with higher concentrations of immigrants and Latinos were associated with Latinos having lower odds of hypertension (OR = 0.60, p = 0.03). However, among those with hypertension, our results point to deleterious effects on hypertension care (OR = 0.55, p = 0.06) and treatment (OR = 0.54, p = 0.04) associated with living in neighborhoods with higher concentrations of immigrants and Latinos. We detected no significant interaction effects between immigrant/Latino neighborhood composition and language of interview or being an immigrant in this sample. These results suggest that improving access to care for Latinos with hypertension requires enhanced placement of community clinics and other safety-net health centers in neighborhoods with higher proportions of immigrants and Latinos.

Spillover effects of the uninsured: local uninsurance rates and Medicare mortality from eight procedures and conditions

The presence of a large uninsured population may create incentives to providers that affect the care delivered to all individuals in a health care market. Using Current Population Survey data on uninsurance rates and hospital discharge data on Medicare beneficiaries, this study investigates the relationship between the uninsurance rate at the metropolitan statistical area (MSA) level and inpatient quality of care delivered to Medicare beneficiaries, as measured by mortality from eight procedures and conditions. The results do not indicate large or widespread negative effects of the uninsured on Medicare beneficiaries. However, some evidence suggests that the relationship between the local uninsurance rate and Medicare mortality does vary by market size.

Uninsured persons with disability confront substantial barriers to health care services

BACKGROUND

Despite social "safety net" programs, many U.S. residents with disabilities lack insurance coverage and thus risk financial barriers to accessing care. The study objectives were to characterize working-age adults with disabilities who lack health insurance and to examine their self-reported barriers to care.

METHODS

The authors conducted analyses of nationally representative Medical Expenditure Panel Survey data from 2000 through 2006.

RESULTS

During this time period, 14.8% of working-age U.S. residents lacked health insurance, including 11.6% of persons with disabilities. Focusing only on uninsured individuals, persons with disabilities were significantly (p = .001) more likely than those without disabilities to have a usual source of care. However, on 6 other access measures (those that comprised our composite indicator of access barriers), uninsured persons with disabilities reported barriers significantly (p = .001) more often than did individuals without disabilities: 36.0% of uninsured persons with disabilities reported being unable to get necessary medical care, compared with 9.5% of uninsured, nondisabled persons; and 26.9% of uninsured persons with disabilities reported being unable to get necessary medications, compared with 5.3% of uninsured individuals without disabilities. Having a cognitive impairment produced the largest adjusted odds ratio (AOR) of reporting any access barrier (1.64, 95% CI 144-1.87), while having lower body functional limitations or hearing deficits also produced relatively high AORs (1.47, 1.32-1.65 and 1.48, 1.11-1.98, respectively).

CONCLUSIONS

Uninsured individuals with disabilities confront significantly more barriers to accessing care than do nondisabled persons without health insurance. Certain types of disabilities appear especially associated with experiencing access barriers, suggesting areas requiring particular attention.

Associations between minimum wage policy and access to health care: evidence from the Behavioral Risk Factor Surveillance System, 1996-2007

OBJECTIVES

We examined whether minimum wage policy is associated with access to medical care among low-skilled workers in the United States.

METHODS

We used multilevel logistic regression to analyze a data set consisting of individual-level indicators of uninsurance and unmet medical need from the Behavioral Risk Factor Surveillance System and state-level ecological controls from the US Census, Bureau of Labor Statistics, and several other sources in all 50 states and the District of Columbia between 1996 and 2007.

RESULTS

Higher state-level minimum wage rates were associated with significantly reduced odds of reporting unmet medical need after control for the ecological covariates, substate region fixed effects, and individual demographic and health characteristics (odds ratio = 0.853; 95% confidence interval = 0.750, 0.971). Minimum wage rates were not significantly associated with being uninsured.

CONCLUSIONS

Higher minimum wages may be associated with a reduced likelihood of experiencing unmet medical need among low-skilled workers, and do not appear to be associated with uninsurance. These findings appear to refute the suggestion that minimum wage laws have detrimental effects on access to health care, as opponents of the policies have suggested.

Uninsured adults with chronic conditions or disabilities: gaps in public insurance programs

Among nonelderly U.S. adults (ages 25-61), uninsurance rates increased from 13.7 percent in 2000 to 16.0 percent in 2005. Despite the existence of public insurance programs, rates remained high for low-income people reporting serious health conditions (25 percent across years) or disabilities (15 percent). Residents of southern states had even higher rates (32 percent with health conditions, 22 percent with disabilities). Those who did not belong to a federally mandated Medicaid eligibility category were about twice as likely as others to be uninsured overall, and uninsurance among this group increased more rapidly over time. These regional and categorical differences reflect gaps in current policy that pose challenges for incremental health reform.

The form and context of federalism: meanings for health care financing

This article examines the meaning of federalism for health care financing (HCF) and is based on two considerations. First, federal institutions are embedded in their national context and interact with them. The design and performance of HCF policy will be influenced by contexts, the workings of the federal institutions, and the interactions of these institutions with different elements of the context. This article unravels these influences. Second, there is no unique model of federalism, and so we have to specify the particular form to which we refer. The examination of the influence of federalism and its context on HCF policy is facilitated by using a transnational comparative approach, and this article examines four mature federations: the United States, Australia, Canada, and Germany. The relatively poor performance of the U.S. HCF system seems associated with the fact that it operates in a context markedly less benign than those of the other national HCF systems. Heterogeneity of context appears also to have contributed to important differences between the United States and the other countries in the design of HCF policies. An analysis of how federalism works in practice suggests that, while U.S. federalism may be overall less favorable to the development of well-functioning HCF policies, the inferior performance of these policies is to be principally attributed to context.

Use of a diagnosis-based risk adjustment model to estimate costs of indigent care in a community at Medicaid reimbursement rates

OBJECTIVES

This study used a diagnosis-based risk adjustment model to estimate the annual costs of uninsured patients in Austin, Texas, and describe the prevalence and costs of their chronic conditions. The data were supplied by the Indigent Care Collaboration, a partnership of local safety-net hospitals and clinics.

METHODS

This study used the Diagnostic Cost Groups prospective Medicaid All-Encounters model, which uses diagnoses, age and gender to assign relative risk scores to patients. The relative risk scores were multiplied by the per capita Texas Medicaid expenditure to obtain estimated annual costs. Chronic diseases were described in terms of prevalence and total estimated annual cost.

RESULTS

A total of 471,194 encounters were recorded for 163,729 patients meeting the study inclusion criteria between the 1st March 2004 and the 28th February 2005. The mean estimated patient yearly cost was US $1,307, and the total estimated yearly population cost was $228,909,529. The most common chronic conditions included hypertension, diabetes, depression, substance abuse, pregnancy, asthma, chronic obstructive pulmonary disease and congestive heart failure.

CONCLUSIONS

This study demonstrates how the unknown costs associated with caring for indigent uninsured patients in a community can be estimated at Medicaid reimbursement rates using the Diagnostic Cost Group model on aggregated patient encounter data.

Differential impacts of public health insurance expansions at the local level

Dramatic expansions in public health insurance eligibility for U.S. children have only modestly reduced the aggregate number of uninsured at the national level. This paper shows that Medicaid and SCHIP expansions had different impacts on child health insurance coverage patterns based upon local labor market characteristics. Metropolitan areas with high levels of unemployment were most likely to have seen improvements in overall insurance coverage for children between 1990 and 2001. Areas with greater fractions of employment in services, retail or wholesale trade were more likely to have experienced increases in public coverage but not overall coverage rates.

Community-level uninsurance and the unmet medical needs of insured and uninsured adults

OBJECTIVE

To examine the relationship between community-level uninsurance rates and the self-reported unmet medical needs of insured and uninsured adults in the U.S.

DATA SOURCES

2000-2001 Community Tracking Study, which includes data from 60 randomly selected U.S. communities. The sample is representative of the contiguous U.S. states.

STUDY DESIGN

Multilevel logistic regressions were employed to investigate whether the local uninsurance rate was related to having reported unmet medical needs within the last year. The models also included individual and community variables that could be potentially related to both community uninsurance rates and having reported unmet medical needs.

PRINCIPAL FINDINGS

The community uninsurance rate was positively associated with having reported unmet medical needs, but only for insured adults. On average, a five percentage point increment in the local uninsured population is associated with a 10.5 percent increase in the likelihood that an insured adult will report having unmet medical needs during the 12-month period studied.

CONCLUSION

Local health care delivery systems seem to be negatively affected by high uninsurance rates. These effects could have negative consequences for health care access, even for individuals who are themselves insured.

Variations in health insurance coverage for rural and urban nonelderly adult residents of Florida, Indiana, and Kansas

CONTEXT

Evidence exists for differences in health insurance coverage among states, but less is known about variations across different kinds of communities within states.

PURPOSE

This article assesses the role of residential setting (metropolitan county, rural adjacent, and rural nonadjacent) in health insurance coverage for adult residents, under age 65, using data from large-scale surveys collected in 3 diverse states (Florida, Indiana, and Kansas).

METHODS

Descriptive statistics are provided, and logistic regression models are used to examine the relationship between uninsurance status and residential settings while controlling for personal characteristics. Adjusted uninsurance rates by residential settings are presented for each state.

FINDINGS

Residential settings are significantly associated with uninsurance status in 2 of the 3 states we examined. We find that adult Floridians of rural adjacent counties are more likely to be uninsured than those in urban counties, but, for Indiana residents, uninsurance status is comparable between urban and rural adjacent residents. Rural nonadjacent Indiana residents are more likely to be uninsured compared to those in urban counties. The insurance status of adult Kansans does not vary across residential settings.

CONCLUSION

Residential settings are significantly associated with being uninsured, but the significance of this link between residential locations and uninsurance status varies from state to state.

Neighborhood effects on primary care access in Los Angeles

Individual health outcomes differ greatly between neighborhoods, and recent research has begun to examine how neighborhood environment affects individual health. A common hypothesis is that the inequitable distribution of healthcare resources limits access to health care for individuals in disadvantaged neighborhoods, causing poorer long-term health. Yet, research has not examined if neighborhood environment actually affects an individual's ability to access primary care. Data from the Los Angeles Family and Neighborhood Survey suggests there is significant variation between neighborhoods in an individual's ability to access primary care. This neighborhood-level effect is not explained by the composition of individuals living in the neighborhood. Four mechanisms through which neighborhood environment could affect an individual's ability to access primary care are examined: (1) neighborhood information networks, (2) neighborhood health behavior norms, (3) neighborhood social capital and (4) neighborhood healthcare resources. Social capital and healthcare resources significantly predict an individual's primary care access. Since differences in primary care access may explain individual-level health disparities between neighborhoods, policies designed to improve primary care access must account for both individual and neighborhood effects.

Rural-Urban Differences in Employment-Related Health Insurance

CONTEXT

Rural residents are disproportionately represented among the uninsured in the United States.

PURPOSE

We compared nonelderly adult residents in 3 types of nonmetropolitan areas with metropolitan workers to evaluate which characteristics contribute to lack of employment-related insurance. RESEARCH DESIGN AND ANALYSIS: Data were obtained from the Medical Expenditure Panel Survey, pooled across 3 panels (1996--1998) to enhance the rural sample size. Econometric decomposition was used to quantify the contribution of employment structure to differences in the probability of being offered employment-related health insurance.

FINDINGS

The most rural workers are 10.4 percentage points less likely to be offered insurance compared with urban workers; the difference is smaller for residents of other rural areas. In rural counties not adjacent to urban areas, lower wages and smaller employers each account for about one-third of the total difference.

CONCLUSIONS

Health insurance disparities associated with rural residence are related to the structure of employment. Major factors include smaller employers, lower wages, greater prevalence of self-employment, and sociodemographic characteristics.

Reducing the Numbers of the Uninsured

The objective of the study described in this article was to identify a model that best predicts state uninsurance rates and quantifies the contribution of socio-economic factors to enable targeted state programs to reduce uninsurance. Linear regression analysis was carried out using state uninsurance rate as the dependent variable and state-level data on demographic, employment, income, and health care environment data (independent variables). For 2000 data, the model R is 0.77, indicating that 77% of the variation in uninsurance rates is explained by the percentage of immigrant population, the workforce in very small businesses, the Black population, the state's median income, and the Medicare-aged population (model R = 0.77 for 1999 and 0.68 for 1998 data). A 1% increase in immigrant population is associated with 0.18% increase in uninsurance rate. A 1% increase in workforce employed in very small businesses associates with 0.79% increase in uninsurance. The findings indicate substantial potential for reducing uninsurance through targeted state policies. Policy recommendations are made to alleviate the insurance hurdles faced by immigrant and small business employee populations.

An examination of enrollment of children in public health insurance in New York City through facilitated enrollment

A cohort of families was followed through the enrollment process for Medicaid and Child Health Plus in New York City to determine success in enrollment and the time it takes to enroll. Families were recruited into the study by enrollers in community-based organizations and managed-care organizations. In our sample, three of four families were successful in enrolling. On average, it took 60 days to attain insurance. Most applicants (76%) received some sort of assistance from enrollers, most frequently in determining which documents were needed (74%). In a multivariable analysis, some of the factors associated with success in enrollment included being assisted by a community-based facilitated enroller, knowledge of required documents, and having lost a child's other health insurance.

Voting with their feet: patient exit and intergroup differences in propensity for switching usual source of care

Many analysts advocate patient exit as a strategy for consumers who experience poor-quality care. Exit is believed to have the potential to improve patient welfare by having patients leave (or "exit") poor-performing health care providers, thus signaling their dissatisfaction with the quality of care they have received and thereby admonishing those providers to improve. However, the validity of exit as a signal of consumer dissatisfaction hinges on how closely it reflects dissatisfaction. Intergroup differences in the propensity to exit could also result in unintended consequences. This article examines the association between consumer experience and the decision to change one's usual care providers. It also investigates if there are any intergroup differences in the propensity for changing providers according to insurance status, gender, and race or ethnicity. Data come from household surveys conducted by the Center for Studying Health System Change. Results show significant intergroup differences in propensity for switching usual source of care for voluntary or involuntary reasons related to insurance, rural residency, age, income, race, and ethnicity. Policy implications of the empirical results on exit, voice, and consumerism are discussed.

The Kindness Of Strangers: Community Effects On The Rate Of Employer Coverage

The rate of employer-sponsored insurance (ESI) varies greatly across states. We analyze the factors that account for that variation. We find that the likelihood that a worker is covered by ESI depends on workers' own characteristics and also on those of other workers in the same metropolitan statistical area. Further, in almost all states the percentage of workers covered by ESI is close to the predicted level of coverage, which suggests that state policies that could affect insurance coverage have had little net effect on ESI rates. Hawaii is an exception: Its mandate on employers to offer coverage results in a rate of ESI that is much higher than expected.

A Method for Identifying Places in Rural America at Risk of Not Being Able to Support Adequate Health Services

CONTEXT

Public policymakers and their advisers struggle with the problem of specifying criteria by which health care providers in rural areas are eligible for special consideration in payment policies and for special grant programs. A means of designating places can provide a basis for assistance and can help target public resources for any providers who deliver services in those places.

PURPOSE

This paper provides the details underlying a place-based approach to identifying rural areas that are at risk for not being able to provide requisite health services.

METHODS

A population size criterion is utilized first to eliminate metropolitan areas and other large agglomerations from consideration. Any territory not included in a place of 3500 or more people, including a 25-mile buffer around that place, is a priori considered to be at risk. All places, including buffers, that have populations between 3500 and 100,000 are further analyzed using population compositional data and principal components analysis.

FINDINGS

In 10 states and 24 bordering states selected for developing and testing the method, there were 1907 block groups outside the boundaries of any place with a population of at least 3500. In addition, the analysis suggested that 66 out of 236 places and buffers with populations between 3500 and 100,000 also should be classified as vulnerable.

CONCLUSIONS

The results are discussed in regard to how a place-based approach can advance the study of rural health needs. By focusing on the needs of the people residing in a defined area, as determined from the aggregate characteristics of the population, a model is generated that can be used to predict special circumstances confronting any service provider. The public policy implications of the findings are also considered. Special payment policies could be written on the basis of place instead of provider characteristics, and grant programs providing technical assistance could be targeted to places of greatest need.

Why is there state variation in employer-sponsored insurance?

Using the National Survey of America's Families in 1997 and 1999, we investigate the sources of variation in employer-sponsored health insurance across states. We find that demographics and family characteristics (such as race/ethnicity and citizenship status), individual employment characteristics (such as firm size and labor-force attachment), and local labor market characteristics (such as unionization) consistently explain the relative position of all of the states with either high or low rates of employer coverage. Income plays a smaller role in explaining the state variation but is still an important determinant, especially among states whose average income is far from the national average.