Using enrollment records to evaluate self-reports of monthly coverage in the redesigned current population survey health insurance module.

OBJECTIVE: To evaluate the veracity of self-reports of month-level health insurance coverage in the Current Population Survey Annual Social and Economic Supplement (CPS). DATA SOURCES AND STUDY SETTING: The CHIME (Comparing Health Insurance Measurement Error) study used health insurance enrollment records from a large regional Midwest insurer as sample for primary data collection in spring 2015. STUDY DESIGN: A sample of individuals enrolled in a range of public and private coverage types (including Medicaid and marketplace) was administered the CPS health insurance module, which included questions about month-level coverage, by type, over a 17-18-month time span. Survey data was then matched to enrollment records covering that same time frame, and concordance between the records and self-reports was assessed. DATA COLLECTION/EXTRACTION METHODS: Sample was drawn by the insurer's informatics specialists and Census Bureau interviewers conducted the survey. Following data collection, updated enrollment records were matched to the survey data to produce a person-level file of coverage by type at the month-level. PRINCIPAL FINDINGS: For 91% of the overall sample, coverage status and type were reported accurately for at least 75% of observed months. Results varied somewhat by stability of coverage. Among those who were continuously covered throughout the 17-18 month observation period (which comprised 64% of the overall sample), that level of reporting accuracy was observed for 94% of the sample; for those who had censored spells (34% of the overall sample), the figure was 87%; and among those with gaps and/or changes according to the records (2% of the overall sample), for 82% of the group at least 75% of months were reported accurately. CONCLUSIONS: Findings suggest that reporting accuracy of month-level coverage in the CPS is high and that the survey could become a valuable new data source for studying the dynamics of coverage, including the Medicaid unwinding.

Insurance-Based Discrimination Reports and Access to Care Among Nonelderly US Adults, 2011-2019.

Objectives. To report insurance-based discrimination rates for nonelderly adults with private, public, or no insurance between 2011 and 2019, a period marked by passage and implementation of the Affordable Care Act (ACA) and threats to it. Methods. We used 2011-2019 data from the biennial Minnesota Health Access Survey. Each year, about 4000 adults aged 18 to 64 years report experiences with insurance-based discrimination. Using logistic regressions, we examined associations between insurance-based discrimination and (1) sociodemographic factors and (2) indicators of access. Results. Insurance-based discrimination was stable over time and consistently related to insurance type: approximately 4% for adults with private insurance compared with adults with public insurance (21%) and no insurance (27%). Insurance-based discrimination persistently interfered with confidence in getting needed care and forgoing care. Conclusions. Policy changes from 2011 to 2019 affected access to health insurance, but high rates of insurance-based discrimination among adults with public insurance or no insurance were impervious to such changes. Public Health Implications. Stable rates of insurance-based discrimination during a time of increased access to health insurance via the ACA suggest deeper structural roots of health care inequities. We recommend several policy and system solutions. (Am J Public Health. 2023;113(2):213-223. https://doi.org/10.2105/AJPH.2022.307126).

Factors associated with accurate reporting of public and private health insurance type.

OBJECTIVE: To examine factors associated with accurate reporting of private and public health insurance coverage. DATA SOURCES: Minnesota health plan enrollment records provided the sample for the Comparing Health Insurance Measurement Error (CHIME) study, a survey conducted in 2015 that randomly assigned enrollees to treatments that included health insurance questions from the American Community Survey (ACS) or the redesigned Current Population Survey Annual Social and Economic Supplement (CPS). STUDY DESIGN: Reverse record check study that compared CHIME study survey responses to enrollment records of coverage type (direct purchase on and off the Marketplace, Medicaid, or MinnesotaCare), service use, subsidy receipt, and duration of coverage from a major insurer. DATA COLLECTION METHODS: Using matched enrollment and CHIME survey data and logistic regression, we examined correlates of accurate insurance type reporting, including characteristics of the insurance coverage, the covered individual, respondent, and family. PRINCIPAL FINDINGS: Reporting accuracy across treatment and coverage type is high (77%-84%). As with past research, accurate reporting of public insurance is higher for people with characteristics consistent with eligibility for public insurance for both survey treatments. For the ACS treatment, reports of direct purchase insurance are more accurate for enrollees who receive a premium subsidy. CONCLUSIONS: Given the complexity of health insurance measurement and frequently changing policy environment, differences in reporting accuracy across treatments or coverage types are not surprising. Several results have important implications for data editing and modeling routines. First, adding premium and subsidy questions in federal surveys should prove useful given the finding that subsidy receipt is associated with reporting accuracy. Second, across both survey treatments, people whose opportunity structures (race, ethnicity, and income) match public program eligibility are accurate reporters of this coverage. This evidence supports using these commonly collected demographic variables in simulation, imputation, and editing routines.

Health insurance literacy: How best to measure and does it matter to health care access and affordability?

BACKGROUND: Health insurance is complex, cost are continuously rising, and people are assuming more of these costs. Health insurance literacy (HIL) is related to healthcare access, yet there is no agreement about how best to measure HIL. OBJECTIVES: Contrast two HIL measures. First, evaluating their association with demographic characteristics, insurance type, and health status. Second, comparing how these distinct measures relate to access, forgone care, and financial burden of health care. METHODS: Data are from a 2017 telephone survey focused on health insurance coverage and access. Participants were randomly assigned either the 4-item likelihood of proactive use scale or a 4-item measure of confidence in use of insurance. Logistic regressions assess correlates of each HIL measure and their association with a range of access measures. RESULTS: For both measures, 25% of insured adults report high HIL. Few demographic and health status measures are associated with high HIL and they are different for each measure. For both measures, high HIL translates into reports of having a usual source of care and confidence in getting care when needed. The HIL measures behave in opposite ways for forgone care due to costs and problems paying medical bills. Adults scoring high on the likelihood measure are more likely to forgo care and report financial burden. By contrast, adults scoring high on the confidence measure are less likely to forgo care and report burdensome medical bills. CONCLUSIONS: The two measures capture different concepts and raise the question of whether reporting a likely behavior or being confident of that behavior are predictive when it is time to use health insurance. Because HIL is measured at the same time as the outcomes, we reason that the likelihood measure is capturing peoples' past experience using insurance and may result in more proactive use of insurance in the future.

State-Level Immigrant Prenatal Health Care Policy and Inequities in Health Insurance Among Children in Mixed-Status Families.

Children in immigrant families are twice as likely to be uninsured as their counterparts, and states may influence these inequities by facilitating or restricting immigrant families' access to coverage. Our objective was to measure differences in insurance by mother's documentation status among a nationally representative sample of US-born children in immigrant families and to examine the role of state-level immigrant health care policy-namely, state-level immigrant access to prenatal coverage. Compared with US-born children in immigrant families with citizen mothers, children with undocumented immigrant mothers had a 17.0 percentage point (P < .001) higher uninsurance rate (8.8 percentage points higher in adjusted models, P < .05). However, in states with nonrestrictive prenatal coverage for immigrants, there were no differences in children's insurance by mother's documentation status, while large inequities were observed within states with restrictive policies. Our findings demonstrate the potential for state-level immigrant health care policy to mitigate or exacerbate inequities in children's insurance.

Assessing the accuracy of survey reports of health insurance coverage using enrollment data.

OBJECTIVE: To measure the accuracy of survey-reported data on features and type of health insurance coverage. DATA SOURCE: Enrollment records from a private insurer were used as sample for primary survey data collection in spring of 2015 using the Current Population Survey health insurance module. STUDY DESIGN: A reverse record check study where households with individuals enrolled in a range of public and private health insurance plans (including the marketplace) were administered a telephone survey that included questions about general source of coverage (eg, employer), program name (eg, Medicaid), portal, premium, and subsidies. DATA COLLECTION/EXTRACTION METHODS: Survey data were matched back to enrollment records, which indicated coverage status at the time of the survey. Concordance between the records and survey data was assessed. PRINCIPAL FINDINGS: Correct reporting of general source of coverage ranged from 77.8 percent to 98.3 percent across coverage type, premium ranged from 91.6 percent to 96.4 percent, and subsidy ranged from 83.0 percent to 91.0 percent. Using a conceptual algorithm to categorize coverage type resulted in sensitivity of 98.3 percent for employer-sponsored enrollees, and 70.6 percent-77.6 percent for the other coverage types, while specificity ranged from 93.9 percent to 99.4 percent across coverage types. CONCLUSIONS: Survey reports of features of coverage suggest they are viable items to include in an algorithm to categorize coverage type. Findings have implications beyond the CPS, particularly for marketplace enrollees.

Health Care Coverage and Access Among Children, Adolescents, and Young Adults, 2010-2016: Implications for Future Health Reforms.

PURPOSE: We examine changes to health insurance coverage and access to health care among children, adolescents, and young adults since the implementation of the Affordable Care Act. METHODS: Using the National Health Interview Survey, bivariate and logistic regression analyses were conducted to compare coverage and access among children, young adolescents, older adolescents, and young adults between 2010 and 2016. RESULTS: We show significant improvements in coverage among children, adolescents, and young adults since 2010. We also find some gains in access during this time, particularly reductions in delayed care due to cost. While we observe few age-group differences in overall trends in coverage and access, our analysis reveals an age-gradient pattern, with incrementally worse coverage and access rates for young adolescents, older adolescents, and young adults. CONCLUSIONS: Prior analyses often group adolescents with younger children, masking important distinctions. Future reforms should consider the increased coverage and access risks of adolescents and young adults, recognizing that approximately 40% are low income, over a third live in the South, where many states have not expanded Medicaid, and over 15% have compromised health.

Rural-Urban Differences in Access to Preventive Health Care Among Publicly Insured Minnesotans.

PURPOSE: Reduced access to care and barriers have been shown in rural populations and in publicly insured populations. Barriers limiting health care access in publicly insured populations living in rural areas are not understood. This study investigates rural-urban differences in system-, provider-, and individual-level barriers and access to preventive care among adults and children enrolled in a public insurance program in Minnesota. METHODS: This was a secondary analysis of a 2008 statewide, cross-sectional survey of publicly insured adults and children (n = 4,388) investigating barriers associated with low utilization of preventive care. Sampling was stratified with oversampling of racial/ethnic minorities. RESULTS: Rural enrollees were more likely to report no past year preventive care compared to urban enrollees. However, this difference was no longer statistically significant after controlling for demographic and socioeconomic factors (OR: 1.37, 95% CI: 1.00-1.88). Provider- and system-level barriers associated with low use of preventive care among rural enrollees included discrimination based on public insurance status (OR: 2.26, 95% CI: 1.34-2.38), cost of care concerns (OR: 1.72, 95% CI: 1.03-2.89) and uncertainty about care being covered by insurance (OR: 1.70, 95% CI: 1.01-2.85). These and additional provider-level barriers were also identified among urban enrollees. CONCLUSIONS: Discrimination, cost of care, and uncertainty about insurance coverage inhibit access in both the rural and urban samples. These barriers are worthy targets of interventions for publicly insured populations regardless of residence. Future studies should investigate additional factors associated with access disparities based on rural-urban residence.

Data Resources for Conducting Health Services and Policy Research.

Rich federal data resources provide essential data inputs for monitoring the health and health care of the US population and are essential for conducting health services policy research. The six household surveys we document in this article cover a broad array of health topics, including health insurance coverage (American Community Survey, Current Population Survey), health conditions and behaviors (National Health Interview Survey, Behavioral Risk Factor Surveillance System), health care utilization and spending (Medical Expenditure Panel Survey), and longitudinal data on public program participation (SIPP). New federal activities are linking federal surveys with administrative data to reduce duplication and response burden. In the private sector, vendors are aggregating data from medical records and claims to enhance our understanding of treatment, quality, and outcomes of medical care. Federal agencies must continue to innovate to meet the continuous challenges of scarce resources, pressures for more granular data, and new multimode data collection methodologies.

Trends in self-efficacy to quit and smoking urges among homeless smokers participating in a smoking cessation RCT.

INTRODUCTION: In the U.S., approximately 73% of homeless adults smoke cigarettes and they experience difficulty quitting. Homeless smokers report low self-efficacy to quit and that smoking urges are a barrier to quitting. Self-efficacy to quit and smoking urges are dynamic and change throughout smoking cessation treatment. This study examines changes in self-efficacy to quit and smoking urges throughout a smoking cessation intervention among the homeless and identifies predictors of change in these characteristics. METHODS: Homeless smokers (n=430) participating in a smoking cessation randomized controlled trial in the U.S. completed surveys at baseline, and weeks 1, 2, 4, 6, 8, and 26 on demographic and smoking characteristics (i.e., confidence to quit, self-efficacy to refrain from smoking, and smoking urges). A growth curve analysis was conducted by modeling change in the smoking characteristics over time and examining the variability in the change in smoking characteristics by demographic characteristics and treatment group. RESULTS: Among the full sample, self-efficacy to refrain from smoking increased linearly over time, confidence to quit increased until the midpoint of treatment but subsequently decreased, and smoking urges decreased until the midpoint of treatment but subsequently increased. There were race differences in these trajectories. Racial minorities experienced significantly greater increases in self-efficacy to refrain from smoking than Whites and Blacks had higher confidence to quit than Whites. CONCLUSIONS: White participants experienced less increase in self-efficacy to refrain from smoking and lower confidence to quit and therefore may be a good target for efforts to increase self-efficacy to quit as part of homeless-targeted smoking cessation interventions. Sustaining high confidence to quit and low smoking urges throughout treatment could be key to promoting higher cessation rates among the homeless.

Cessation-related weight concern among homeless male and female smokers.

Concern about post-cessation weight gain is a barrier to making attempts to quit smoking; however, its effect on smoking cessation is unclear. In this study we examine cessation-related weight concern among the homeless, which hasn't been studied. Homeless males (n = 320) and females (n = 110) participating in a smoking cessation RCT in the Twin Cities, Minnesota from 2009 to 2011 completed surveys on cessation-related weight concern, smoking status, and components from the Behavioral Model for Vulnerable Populations. Generalized estimating equations were used to examine baseline predictors of cessation-related weight concern at baseline, the end of treatment, and 26-weeks follow-up. Logistic regression models were used to examine the relationship between cessation-related weight concern and smoking status at the end of treatment and follow-up. Females had higher cessation-related weight concern than males. Among males, older age, Black race, higher BMI, depression, and having health insurance were associated with higher cessation-related weight concern. Among females, nicotine dependence, greater cigarette consumption, indicating quitting is more important, older age of smoking initiation, and less support to quit from family were associated with higher cessation-related weight concern. In multivariate analyses, cessation-related weight concern decreased over time among females. Cessation-related weight concern wasn't associated with smoking cessation. Although several types of characteristics predicted cessation-related weight concern among males, only smoking characteristics predicted cessation-related weight concern among females. Given the small proportion of quitters in this study (8% of males and 5% of females), further research on the impact of cessation-related weight concern on smoking cessation among the homeless is warranted.

A Community-Driven Implementation of the Body and Soul Program in Churches in the Twin Cities, Minnesota, 2011-2014.

BACKGROUND: African Americans have high disease and death rates due to cancer and cardiovascular disease. Health promotion efforts to improve diet have the potential to reduce these rates. COMMUNITY CONTEXT: Given their importance in the community and the extent of their reach, churches are effective avenues for health promotion efforts targeting African Americans. The objectives of this project were to promote healthy eating among African American church members, engage African American churches in the implementation of Body and Soul (an evidenced-based program that encourages healthy eating), and implement the program in the community with minimal resources. METHODS: From 2011 through 2014 we conducted a community engagement project to implement the 12-week Body and Soul program, which includes demonstrations of healthy recipes and peer counseling, in 20 churches. Participants (n = 310) completed baseline and follow-up surveys on their eating habits and experience with peer counseling. Church coordinators (n = 11) completed a survey evaluating the program. OUTCOME: Participants' weekly servings of fruit (baseline, 4.3; follow-up, 5.4; P < .001) and vegetables (baseline, 4.5; follow-up, 5.3; P < .001) increased. Church coordinators reported enthusiasm about Body and Soul at their church, and 10 of 11 church coordinators indicated that their pastor encouraged members to attend Body and Soul events. Program success was promoted by engaging the pastor in program activities and by scheduling events soon after church services. Implementation challenges were variation in peer counseling among churches and low turnout at follow-up events. INTERPRETATION: The project was successfully implemented in the 20 churches, and increases in healthy eating were observed. This project demonstrated that Body and Soul can be implemented in communities with little funds or other resources.

Coverage Gains After the Affordable Care Act Among the Uninsured in Minnesota.

OBJECTIVES: We determined whether and how Minnesotans who were uninsured in 2013 gained health insurance coverage in 2014, 1 year after the Affordable Care Act (ACA) expanded Medicaid coverage and enrollment. METHODS: Insurance status and enrollment experiences came from the Minnesota Health Insurance Transitions Study (MH-HITS), a follow-up telephone survey of children and adults in Minnesota who had no health insurance in the fall of 2013. RESULTS: ACA had a tempered success in Minnesota. Outreach and enrollment efforts were effective; one half of those previously uninsured gained coverage, although many reported difficulty signing up (nearly 62%). Of the previously uninsured who gained coverage, 44% obtained their coverage through MNsure, Minnesota's insurance marketplace. Most of those who remained uninsured heard of MNsure and went to the Web site. Many still struggled with the enrollment process or reported being deterred by the cost of coverage. CONCLUSIONS: Targeting outreach, simplifying the enrollment process, focusing on affordability, and continuing funding for in-person assistance will be important in the future.

Reports of insurance-based discrimination in health care and its association with access to care.

OBJECTIVES: We examined reports of insurance-based discrimination and its association with insurance type and access to care in the early years of the Patient Protection and Affordable Care Act. METHODS: We used data from the 2013 Minnesota Health Access Survey to identify 4123 Minnesota adults aged 18 to 64 years who reported about their experiences of insurance-based discrimination. We modeled the association between discrimination and insurance type and predicted odds of having reduced access to care among those reporting discrimination, controlling for sociodemographic factors. Data were weighted to represent the state's population. RESULTS: Reports of insurance-based discrimination were higher among uninsured (25%) and publicly insured (21%) adults than among privately insured adults (3%), which held in the regression analysis. Those reporting discrimination had higher odds of lacking a usual source of care, lacking confidence in getting care, forgoing care because of cost, and experiencing provider-level barriers than those who did not. CONCLUSIONS: Further research and policy interventions are needed to address insurance-based discrimination in health care settings.

Measurement Error in Public Health Insurance Reporting in the American Community Survey: Evidence from Record Linkage.

OBJECTIVE: Examine measurement error to public health insurance in the American Community Survey (ACS). DATA SOURCES/STUDY SETTING: The ACS and the Medicaid Statistical Information System (MSIS). STUDY DESIGN: We tabulated the two data sources separately and then merged the data and examined health insurance reports among ACS cases known to be enrolled in Medicaid or expansion Children's Health Insurance Program (CHIP) benefits. DATA COLLECTION/EXTRACTION METHODS: The two data sources were merged using protected identification keys. ACS respondents were considered enrolled if they had full benefit Medicaid or expansion CHIP coverage on the date of interview. PRINCIPAL FINDINGS: On an aggregated basis, the ACS overcounts the MSIS. After merging the data, we estimate a false-negative rate in the 2009 ACS of 21.6 percent. The false-negative rate varies across states, demographic groups, and year. Of known Medicaid and expansion CHIP enrollees, 12.5 percent were coded to some other coverage and 9.1 percent were coded as uninsured. CONCLUSIONS: The false-negative rate in the ACS is on par with other federal surveys. However, unlike other surveys, the ACS overcounts the MSIS on an aggregated basis. Future work is needed to disentangle the causes of the ACS overcount.

Understanding cancer screening service utilization by Somali men in Minnesota.

This study examined factors that influence use of cancer screening by Somali men residing in Minnesota, USA. To better understand why recent immigrants are disproportionately less likely to use screening services, we used the health belief model to explore knowledge, beliefs, and attitudes surrounding cancer screening. We conducted a qualitative study comprised of 20 key informant interviews with Somali community leaders and 8 focus groups with Somali men (n = 44). Somali men commonly believe they are protected from cancer by religious beliefs. This belief, along with a lack of knowledge about screening, increased the likelihood to refrain from screening. Identifying the association between religion and health behaviors may lead to more targeted interventions to address existing disparities in cancer screening in the growing US immigrant population.

The projected effect of the Affordable Care Act on dental care for adult Medicaid enrollees.

Fewer than half of all U.S. states provide dental care for non-elderly adult Medicaid enrollees. Although the Affordable Care Act (ACA) expands Medicaid eligibility for adults, states are not required to offer dental care to adults. We project the effect of the ACA on patient-identified barriers to dental care based on a framework developed using data from a 2008 survey of Minnesota Medicaid enrollees with and without an annual dental visit. The rate of annual visits (55%) was below that of all Minnesotans (79%) with 40% reporting difficulties accessing services. We found no racial/ethnic disparities in annual dental visits among adult Medicaid enrollees. Adult Medicaid recipients with no annual visit reported individual (51%), provider (27%), and system-level (22%) barriers. Hmong, Somali, and American Indian adults were more likely than others to report barriers to receiving dental care. We project that the ACA will not reduce barriers to dental care for adult Medicaid enrollees.

Health reform and the US Virgin Islands: high-need-limited impact.

OBJECTIVES: There are 4.1 million residents living in the US territories, which is more than the combined population of many US states, yet the territories and their citizens are often overlooked from a policy perspective, because most individual territories are relatively small, geographically isolated, and have been treated differently than the states historically. This tendency to fall beneath the radar is clear in the realm of health policy, especially in the area of insurance coverage. This article provides an initial assessment of the potential impact of health reform on the US Virgin Islands (USVI) and, in light of this assessment, considers how the results of a USVI household survey conducted in 2003 and 2009 might be used as a baseline for future monitoring of the impact of national reform. METHODS: A study by the Virgin Island's Bureau of Economic Research, Office of the Governor, and the University of Minnesota, was conducted in 2003 and 2009. The Virgin Islands Health Insurance telephone Surveys were random digit dial landline telephone surveys of households in the USVI. A stratified sample was drawn to produce precise estimates of insurance coverage for the USVI as a whole and for the 3 islands separately. RESULTS: Almost one-third of the residents (28.7%) in the Virgin Islands were uninsured in 2009. This rate is twice the US average (15.4%) and significantly higher than the uninsured rate of 24.1% when a similar survey was last conducted the Virgin Islands in 2003. CONCLUSIONS: The Patient Protection and Affordable Care Act of 2010 provides special funding to the territories through a mix of increased Medicaid caps for each territory and the provision of premium subsidies through newly established health insurance exchanges to low-income populations. However, the Affordable Care Act's Medicaid expansions to newly eligible adults--primarily adults without children--are limited to current eligibility levels in the territories, which is $5,500 in annual income for adult coverage in the USVI. Within these abbreviated parameters, the Medicaid expansion can go so far only toward mitigating uninsurance among the lowest income groups in the territories. With certain low-income childless adults overlooked, the Affordable Care Act does not fully address the high need for affordable health insurance coverage in the territories.

Monitoring health reform efforts: which state-level data to use?

This study compares estimates of health insurance coverage from the American Community Survey (ACS) to those in twelve state-specific surveys. Uninsurance estimates for the nonelderly are consistently higher in the ACS than in state surveys, as are direct purchase insurance estimates. Estimates for employer-sponsored insurance are similar, but public coverage rates are lower in the ACS. The ACS meets some but not all of the states' data needs; its large sample size and inclusion of all U.S. counties in the sample allow for comparison of insurance coverage within and across states. State-specific surveys provide the flexibility to add policy-relevant questions, including questions needed to examine how health insurance translates into access, use, and affordability of health services.

Bias in telephone surveys that do not sample cell phones: uses and limits of poststratification adjustments.

OBJECTIVE: To examine how biased health surveys are when they omit cell phone-only households (CPOH) and to explore whether poststratification can reduce this bias. METHODS: We used data from the 2008 National Health Interview Survey (NHIS), which uses area probability sampling and in-person interviews; as a result people of all phone statuses are included. First, we examined whether people living in CPOH are different from those not living in CPOH with respect to several important health surveillance domains. We compared standard NHIS estimates to a set of "reweighted" estimates that exclude people living in CPHO. The reweighted NHIS cases were fitted through a series of poststratification adjustments to NHIS control totals. In addition to poststratification adjustments for region, race or ethnicity, and age, we examined adjustments for home ownership, age by education, and household structure. RESULTS: Poststratification reduces bias in all health-related estimates for the nonelderly population. However, these adjustments work less well for Hispanics and blacks and even worse for young adults (18 to 30 y). Reduction in bias is greatest for estimates of uninsurance and having no usual source of care, and worse for estimates of drinking, smoking, and forgone or delayed care because of costs. CONCLUSIONS: Applying poststratification adjustments to data that exclude CPOH works well at the total population level for estimates such as health insurance, and less well for access and health behaviors. However, poststratification adjustments do not do enough to reduce bias in health-related estimates at the subpopulation level, particularly for those interested in measuring and monitoring racial, ethnic, and age disparities.