Skip to main content
NCBI home page
JAMA Network logoLink to JAMA Network
. 2025 Sep 5;6(9):e253020. doi: 10.1001/jamahealthforum.2025.3020

Medicaid Primary Care Utilization and Area-Level Social Vulnerability

Jordan Herring 1,✉, Yoon Hong Park 2, Qian Luo 2, Anushree Vichare 2, Clese Erikson 2, Patricia Pittman 2
PMCID: PMC12413652  PMID: 40911326

Key Points

Question

How is the concentration of social vulnerability in an area associated with primary care utilization by Medicaid beneficiaries?

Findings

This cross-sectional study including data for 34 890 932 Medicaid beneficiaries (age <65 years) found that those residing in the most socially vulnerable areas were significantly less likely to have a primary care visit; however, the probability of having a primary care visit at a federally qualified health center for these beneficiaries increased.

Meaning

These findings suggest that health workforce initiatives and outreach efforts in underserviced areas with high concentrations of social vulnerability may be more effective when they use area-level multidimensional indexes for targeted strategies to improve access to primary care.

Abstract

Importance

The concentration of poverty and multidimensional disadvantage has been shown to limit access to health care in these communities. There is a growing interest in using area-level socioeconomic indexes to address the unequal geographic distribution of health care resources. However, the association of area-level socioeconomic indexes with access to primary care—a key area in health policy—has not been determined.

Objective

To investigate the association of Medicaid primary care utilization with the concentration of poverty and multidimensional disadvantage at the zip code level.

Design, Settings, and Participants

This cross-sectional study used the 2019 Transformed-Medicaid Statistical Information System to identify variations in primary care utilization among Medicaid and the Children’s Health Insurance Program beneficiaries (age <65 years) by poverty and multidimensional disadvantage levels of their area of residence. Included beneficiaries were enrolled in Medicaid from January 1 to December 31, 2019, and were not dually eligible for Medicare. The zip code−level Social Vulnerability Index (SVI) was used to assess the likelihood of a beneficiary having an annual primary care visit, while controlling for individual beneficiary demographic and health characteristics. An activity-based approach was adopted to classify clinicians billing Medicaid for primary care and to identify primary care visits at federally qualified health centers (FQHCs). SVI results were compared with results using income-based poverty rates alone. Data analysis was performed from May 1, 2023, through February 28, 2025.

Exposure

Zip code−level deciles of the SVI and poverty rates.

Main Outcomes and Measures

Regression analysis was performed at the beneficiary level, using a binary indicator for having a primary care visit on a set of dummy variables for SVI deciles, controlling for age and sex interactions, disability status, and indicators for having been diagnosed with behavioral health or chronic physical health conditions.

Results

The total population analyzed comprised 34 890 932 Medicaid beneficiaries (<65 years old; 54.2% female and 45.8% male), more than half of whom resided in the top 20% of socially vulnerable zip codes; approximately 33%, in the top 10%; and another 20%, in the ninth decile. Of the total, 68.1% had at least 1 primary care visit in 2019, at either a non-FQHC practice (61.1%) or a FQHC (12.7%). The probability of having a primary care visit was highest for children (age <18 years) but varied substantially by age. Compared to those residing in the first decile of the SVI (least socially vulnerable), beneficiaries in the tenth decile (most socially vulnerable) were 8.9 (95% CI, –9.9 to –7.9) percentage points (pp) less likely to have a primary care visit when not counting FQHC visits, but this increased to 4.7 (95% CI, –5.5 to –3.8) pp less likely when including FQHC visits. Beneficiaries in the tenth decile were 5.9 (95% CI, 4.9 to 6.8) pp more likely to have a FQHC visit than beneficiaries in the first decile. The SVI results identified more beneficiaries with disparities compared to the area-level poverty rate alone.

Conclusions and Relevance

The findings of this cross-sectional study suggest that Medicaid policy should focus on addressing geography-based disparities in access to care using new measures to target resources. The multidimensional SVI is likely a useful tool to identify small geographic areas with barriers to accessing adequate health care. The FQHC findings suggest that substantially increasing investments and support for FQHCs would address geographic inequities in access to health care.

This cross-sectional study assesses variations in primary care utilization associated with the concentration of multidimensional disadvantage at the zip code level among Medicaid beneficiaries.

Introduction

Health care access and utilization in Medicaid vary substantially, both across 1,2,3,4,5,6,7,8,9 and within states,10,11,12,13,14,15,16,17,18,19 which highlights geographic inequities. Targeting areas with unmet health care needs is the cornerstone of health resource planning, but measuring true access is challenging.20,21 Consequently, there is a growing interest in using area-level socioeconomic indexes to target the areas in need of health care resources.22,23,24 Targeting geographic areas, rather than individuals, could be a more effective strategy for incentivizing competition among health plans and practices and for improving the geographic distribution of health care resources.25 If geographic-based socioeconomic indexes identify areas where individuals have worse access to health care, it could then make these tools useful for health care policy and planning. However, how area-level socioeconomic indexes are associated with primary care access in Medicaid—a key area in health policy for improving access to primary care—has not been determined.

Literature on concentrated poverty indicates that where an individual resides has a distinct influence on health outcomes, beyond individual factors.26,27,28 Measurement of poverty has now become multidimensional, demonstrating that social, economic, and environmental conditions offer a more comprehensive understanding of barriers to economic and social participation than do poverty rates based on income alone.29,30,31 A substantial body of research on the social determinants of health demonstrates that multidimensional disadvantage is closely associated with reduced access to health care and worse health outcomes.32,33,34

This study was motivated to extend the theoretical foundation of concentrated poverty26 to concentrated multidimensional disadvantage in the context of health care access for the most disadvantaged populations in the US. We aimed to use the Social Vulnerability Index (SVI) to investigate how primary care use among Medicaid beneficiaries varies by the concentration of multidimensional disadvantage at the zip code level.35 The SVI is an area-level measure of multidimensional disadvantage. Developed by the US Centers for Disease Control and Prevention,35,36 it has been widely used to study disparities in health care access, outcomes, and utilization.37,38,39,40,41,42,43,44,45,46,47,48,49,50 We focused on the probability of having a primary care visit, which is different from true access to health care,51,52 but may serve as a proxy for potential barriers.20 We also sought to assess the extent to which federally qualified health centers (FQHCs) ameliorate geography-based disparities. Established as part of an antipoverty program to improve health care access, the initial rollout of FQHCs was associated with declines in mortality rates.53 At FQHCs, more than 50% of patients have Medicaid coverage and more than 80% of patients whose poverty status is known are at 200% or more below the federal poverty line.54 In theory, primary care use at FQHCs should increase with higher levels of the SVI. In addition, to gain further insight into how concentrated multidimensional disadvantage may differ from a more parsimonious measure of concentrated poverty, we aimed to include a comparison of the SVI results with results of using only income-based poverty rates.

Methods

This cross-sectional study was reviewed and deemed exempt by the institutional review board of George Washington University. Informed consent was not required because we used secondary data. We followed Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Data Collection

Using the 2019 Transformed-Medicaid Statistical Information System (T-MSIS) Analytical Files (TAF), we constructed an analytical sample of Medicaid−Children’s Health Insurance Program beneficiaries (age <65 years) who were continuously enrolled in coverage from January 1 to December 31, 2019, and were not dually eligible for Medicare. We excluded those who were also enrolled in Medicare (dually enrolled) and residents of 14 states and the District of Columbia due to data quality concerns (details are available in eAppendix 1 in Supplement 1).

We used a combination of the National Plan and Provider Enumeration System and an activity-based classification developed by O’Reilly-Jacob et al55 to identify primary care clinicians, based on practice setting, clinical concentration, and service types for clinicians meeting an activity threshold. For clinicians not meeting the activity threshold, we defaulted to reported specialties in the National Plan and Provider Enumeration System. We classified any claim that included a primary care clinician as a primary care visit at the beneficiary level. We identified FQHC organizational National Provider Identifiers based on prior work56 and included all evaluation and management57 services from FQHCs as primary care because FQHC billing practices may differ from those of other clinic types and across states (details are available in eAppendix 2 in Supplement 1).

The SVI is an equally weighted, rank-based composite index based on variables that capture socioeconomic status, household characteristics, race and ethnicity minority status, housing type, and transportation type.36 The original SVI is only available at the census tract and county levels; however, the TAF includes only zip code−level data. To address the geographic mismatch, we used the 2020 data from Franchi et al35 who constructed a zip code−level SVI based on the original SVI methods.

Primary Outcomes

The primary outcomes were the probability of having a primary care visit either at a non-FQHC clinic (using the individual clinician classification approach) or at a FQHC and having a primary care visit at all within the calendar year. Having at least 1 primary care visit is likely a key quality indicator for many care strategies and is likely recommended for many of the beneficiaries for preventive services. The absence of having at least 1 primary care visit at all during the year may reflect barriers with accessibility to the health care system overall.20

Statistical Analysis

First, we calculated the distribution of Medicaid beneficiaries across SVI deciles using beneficiary’s residential zip codes. Then, we regressed each outcome at the individual beneficiary level on dummy variables for SVI deciles, controlling for individual demographic and health characteristics. The omitted reference decile was the first SVI decile (the least deprived areas), thus estimating the statistical difference in utilization at each SVI decile relative to the least deprived decile.

Demographic characteristics, which were self-reported in the American Community Survey, included an indicator for having a disability,58 and 5-year age group by sex interactions to account for variation in utilization due to both age and sex. Health characteristics included a set of indicators for having been diagnosed with behavioral health conditions (eg, depression, anxiety, bipolar disorder) and physical health chronic conditions (eg, asthma, diabetes, hypertension). eTable 2 in Supplement 1 reports the conditions used and associated diagnosis codes. Robust standard errors were clustered at the zip code level. We conducted sensitivity checks comparing the main SVI results to results from analyses that did not control for demographic and health characteristics, and to results for only US states with adequate data quality per our quality metric, which was based on the ability to observe individual clinicians on claims among the analytical sample.

To compare the main SVI results to income-based poverty rates, we repeated distribution calculation and the regression estimates but focused on a set of dummy variables corresponding to deciles based on poverty instead of SVI deciles. Data analyses were performed from Data analysis was performed from May 1, 2023, through February 28, 2025, using Stata, version 18 (Stata Corp).

Results

The total study sample analyzed included 34 890 932 Medicaid beneficiaries (age <65 years; 54.2% female and 45.8% male). Descriptive statistics are reported in the Table. We found that 68.1% of Medicaid beneficiaries had at least 1 primary care visit at either a non-FQHC practice (61.1%) or at a FQHC (12.7%). The probability of having a primary care visit was highest for children (age <18 years) but varied substantially by age. Beneficiaries with disabilities and female beneficiaries had higher probabilities of having a primary care visit. Beneficiaries who were diagnosed with chronic conditions had much higher utilization rates overall.

Table. Descriptive Statistics of the Analytical Sample of 34 890 932 Medicaid or Children Health Insurance Plan Beneficiariesa.

Characteristic % % With a primary care visit
Any Non-FQHC FQHC
Total (N = 34 890 932) 100.0 68.1 61.1 12.7
Sex
Female 54.2 69.4 62.3 13.5
Male 45.8 66.4 59.6 11.8
Disability
No disability 89.5 67.5 60.6 12.5
Has disability 10.5 72.5 65.6 14.6
Age group, y
1-9 29.7 79.1 72.4 12.0
10-19 29.0 69.1 62.0 11.9
20-29 12.1 52.3 46.0 12.0
30-39 11.5 56.4 50.0 13.1
40-49 7.7 61.8 54.6 14.4
50-59 7.1 69.0 60.5 16.5
60-64 2.9 71.5 62.7 16.4
Diagnoses
Behavioral health
Anxiety 7.9 87.7 79.6 20.9
Bipolar 1.5 84.7 76.3 23.7
Depression 4.5 87.1 78.3 23.0
Schizophrenia/other psychotic disorder 1.0 77.9 68.0 22.9
Substance use disorder 3.6 79.9 71.9 22.4
Physical health
Asthma 4.4 93.0 87.0 15.9
Chronic obstructive pulmonary disease 1.0 93.3 86.6 20.9
Diabetes 3.1 92.0 82.4 21.9
Kidney disease 0.4 91.3 86.1 18.4
Heart failure 0.2 92.4 87.4 19.6
Hypertension 4.4 93.4 83.7 21.9
Ischemic heart disease 0.2 92.7 86.9 20.8
Open in a new tab

Abbreviation: FQHC, federally qualified health center.

a

Data were obtained from the 2019 Transformed-Medicaid Statistical Information System Analytical Files for 36 states. The analytical sample was composed of beneficiaries from 40 states who were not dually eligible for Medicare, younger than 65 years, and enrolled in Medicaid from January 1 to December 31, 2019. Disability status was determined using Medicaid eligibility codes following MACSTATS, 2015.58 Non-FQHC primary care visits were identified through individual clinicians using an activity-based approach for non-FQHC settings. FQHC primary care visits were identified as all evaluation and management visits at FQHCs.

Figure 1 reports the distribution of Medicaid beneficiaries across SVI deciles. We found that more than half of analytical sample of Medicaid beneficiaries resided in the top 20% of socially vulnerable zip codes; approximately 33%, in the top 10%; and another 20%, in the ninth decile.

Figure 1. Distribution of Medicaid Beneficiaries, by Social Vulnerability Index (SVI) Deciles.

Figure 1.

Open in a new tab

Data obtained from the 2019 Transformed-Medicaid Statistical Information System Analytical Files for 36 states, and 2020 SVI from Franchi et al.35 Beneficiaries are sorted in the SVI deciles based on zip codes. The distribution of beneficiaries across SVI deciles is reported.

Figure 2A and B report the main results from the regressions controlling for individual demographic and health characteristics with 95% CIs (the full regression results are reported in eTable 3 in Supplement 1). Panel A reports the results by any primary care visit, including FQHC visits, and for only primary care visits in non-FQHC settings. Panel B reports the results for primary care visits at FQHCs only. The association between having a primary care visit and the SVI was negative overall but increased for FQHC primary care visits. Compared to beneficiaries living in the first decile of the SVI (the least socially vulnerable), beneficiaries living in the tenth decile of the SVI (the most socially vulnerable) were 8.9 (95% CI, –9.9 to –7.9) percentage points less likely to have a primary care visit when not counting FQHC visits, but this increased to 4.7 (95% CI, –5.5 to –3.8) percentage points less likely when including FQHC visits. Beneficiaries in the tenth decile were 5.9 (95% CI, 4.9 to 6.8) percentage points more likely to have a primary care visit at an FQHC than were the beneficiaries in the first decile.

Figure 2. Probability of Having a Primary Care Visit, by Social Vulnerability Index (SVI) Deciles.

Figure 2.

Open in a new tab

At the beneficiary level, a binary variable indicating having a primary care visit was regressed on a set of dummy variables for SVI deciles (reference category is the first decile), and the coefficients are reported with 95% CIs. The regressions include controls for 5-year age group-by-sex interactions, an indicator for having a disability, and indicators for having been diagnosed with behavioral health and physical health chronic conditions. The full regression results are reported in eTable 3 in Supplement 1. Data obtained from the 2019 Transformed-Medicaid Statistical Information System Analytical Files for 36 states, and 2020 SVI, from Franchi et al.35 FQHC indicates federally qualified health centers.

Sensitivity checks comparing the main results to those without controlling for demographic and health variables and restricted to only states with adequate (low concern) data quality are reported in eFigures 2 and 3 in Supplement 1, respectively. In both cases, the probability of having a primary care visit decreased as the SVI increased. Excluding the demographic and health controls showed a smaller disparity compared to using the controls, likely associated with minor selection effects of socially vulnerable areas having a higher prevalence of chronic conditions, and thus, higher utilization rates in general.59,60 By adjusting for health diagnoses and accounting for these minor selection effects, a slightly larger utilization disparity emerged.

In Figure 3 we report the distribution of beneficiaries across deciles (panel A) and the coefficients on poverty deciles for the probability of having an annual primary care visit (panel B; full regression results reported in eFigure 4 in Supplement 1). The distribution of beneficiaries across poverty deciles increased similarly to the SVI, but the level of concentration in the top deciles was lower than the SVI. The pattern across deciles based on the SVI and poverty both decreased, although the SVI estimate was slightly higher. Beneficiaries living in the tenth decile based on poverty rates were 3.9 (95% CI, –5.0 to –2.8) percentage points less likely to have a primary care visit than those in the first decile.

Figure 3. Social Vulnerability Index (SVI) Deciles Compared to Poverty Deciles.

Figure 3.

Open in a new tab

A and B, Distribution of Medicaid beneficiaries by SVI vs poverty rate deciles. C and D, Probability of having a primary care visit by SVI and poverty rate deciles. Whiskers indicate with 95% CIs and the blue dashed lines denote the point estimate from the tenth decile using the SVI to compare to estimate of the tenth decile using poverty rates. Data were obtained from the 2019 Transformed-Medicaid Statistical Information System Analytical Files for 36 states, and 2020 SVI, from Franchi et al.35

Discussion

We found that Medicaid beneficiaries were disproportionately concentrated in the most socially vulnerable areas per the SVI, and those living in the most socially vulnerable areas were considerably less likely to have a primary care visit within the calendar year. Benchmarked to an overall average of 68.1% of beneficiaries having a primary care visit, those in the most socially vulnerable areas were approximately 6.9% ([4.7 ÷ 68.1] ×100) less likely to have a primary care visit, even when accounting for FQHC visits, than the average beneficiary.

These findings provide further evidence that geographic areas with a high concentration of multidimensional disadvantage potentially have worse access to health care. Previous work has found that “neighborhood effects” account for up to 15% of the geographic variation in life expectancy in the US, and places with better health outcomes tend to have higher-quality hospitals, more primary care and specialist clinicians per capita, and more frequent health care utilization,61 thus highlighting the importance of addressing geography-based disparities in health care access and utilization.

Many potential mechanisms may provide further insight into our study results. Neighborhood effects that diminish the accessibility of health care could be associated with a lack of investment and diminished economic growth that discourage high-quality health care clinics and hospitals from serving within these areas,26,28 compounding transportation barriers62,63 given that individuals must seek care elsewhere. Peer effects could also affect health care access, use, and outcomes.64,65,66 Moreover, areas where most individuals have equally poor health care access are likely to limit individuals’ abilities to exchange information on accessing high-quality health care outside of their community. Poverty stigma also operates at structural and social levels, more so than the individual level, which may discourage individuals living in high poverty areas from seeking health care.67,68 Racial segregation also contributes to higher rates of concentrated poverty,69,70 and structural racism has been associated with geographic disparities in access to health care.11,71,72,73

FQHCs play a critical role in serving individuals experiencing severe socioeconomic challenges,74 and in line with expectations, the probability of having a primary care visit at an FQHC increased considerably with area-level social vulnerability. Without FQHCs, the most socially vulnerable areas would be 8.9 percentage points less likely to have a primary care visit, but this increased to 4.7 percentage points when we included FQHC visits, reducing the disparity by more than half. However, given that only 12.7% of the overall analytical sample had a visit at an FQHC (compared to 61.1% at a non-FQHC practice), this increase was not sufficient to fully offset the geographic disparity overall and is likely due to the limited availability of FQHCs.

We found similar results in the probability of having a primary care visit using deciles of poverty rates instead of SVI deciles, although the SVI indicated a slightly larger disparity than using income-based poverty rates. Where the SVI differs more from simple poverty rates is on the distribution of Medicaid beneficiaries. For example, approximately 33% of Medicaid beneficiaries live in the top 10% of zip codes based on the SVI, but just over 15% of beneficiaries live in the top 10% of zip codes based on poverty rates. The SVI, and its multidimensionality, identifies a different set of zip codes than the poverty rate alone, capturing substantially more beneficiaries who have lower utilization rates, and potentially worse access to health care. Thus, a multidimensional disadvantage measure may increase the accuracy of identifying communities that face obstacles to accessing health care (compared to poverty rates alone).

These study results have large implications for health policy and resource targeting. The FQHC findings suggest that substantially increasing investment and support for FQHCs would directly address geographic inequities in access to health care. Most FQHCs screen for social risk factors and have many systems in place to track and address social determinants of health75 and employ additional staff, such as community health workers, to better meet Medicaid patients’ needs.76,77 However, although FQHCs have expanded considerably during the past several years, they face workforce challenges—70% of FQHCs report severe staffing shortages and recruitment shortfalls78—indicating more support is needed.

Because the SVI identifies communities with potentially lower access to primary care, resource planners could target the most socially vulnerable areas directly. The SVI could be further integrated into risk adjustment and value-based payment models to increase payments to health care professionals for caring for beneficiaries in socially vulnerable areas to incentivize systematic change. Similarly, the health professional shortage areas as defined by the US Health Resources & Services Administration could be updated to include social vulnerability as a means to proxy for lack of access to health care, and thus, incentivize practitioners and facilities to locate within these areas. For example, the SVI was recently included as a scoring factor for maternity care target areas.22

However, the use of multidimensional indexes in health policy has been based on using preconstructed indexes. Although we found a decline in primary care utilization as the SVI increased, other similar indexes may produce different results. For example, the Area Deprivation Index has been used to provide bonus payments to clinicians caring for beneficiaries in the most deprived areas in the Accountable Care Organization’s Realizing Equity, Access, and Community Health model43; however, flaws in the Area Deprivation Index construction have diminished the reliability of this particular index in improving health equity.79 New indexes may need to be constructed that are specific to certain policy objectives (eg, improved access, better outcomes, or addressing workforce shortages) as well as certain key drivers of disparities.43 Future work may also direct attention to social vulnerability within rural and/or urban stratifications.

Limitations

Differing state administrative procedures and billing practices may limit the accuracy of quantifying national Medicaid utilization data obtained from the using the TAF. The activity-based approach may have misclassified clinicians, particularly those with smaller Medicaid patient panels or more ambiguous practice styles. Accuracy of identifying and classifying FQHC visits in the T-MSIS data are also limited because there is no comprehensive directory of FQHC billing organizations or methods to identify FQHC claims comprehensively in T-MSIS.

In addition, we captured only the variation within the population fully enrolled for the entire year. Conditioning on full enrollment may also have limited the overall evaluation of utilization given that many Medicaid beneficiaries do utilize health care despite being enrolled for only part of the year.

Moreover, the analytical sample was composed of only 36 states. This sampling bias reduces the generalizability to the national level (eg, we excluded California, which has more FQHCs than other states).

Although we used actual health care utilization as a proxy for access,20 utilization does not exactly translate into access. Utilization is 1 important component of accessible health care in certain communities. Additionally, we focused only on the extensive margin of primary care utilization: having any primary care visit at all. This is a narrow outcome and more detailed outcomes that identify types of visits, quality of visits, and health outcomes would provide further insights. We also focused on only 1 calendar year, and further research including multiple years would provide important information on the ability to access health care. For example, relatively healthy individuals may forgo annual wellness checks, but multiple years without primary care visits would yield further insights into access disparities.

Furthermore, we used a limited set of individual demographic factors (age, sex, disability, and health diagnoses) as controls but were not able to fully disentangle individual effects from area-level effects. For example, after controlling for other individual factors (eg, transportation access, employment, race and ethnicity), geographic variation may decrease. The TAF includes race and ethnicity variables, but there is a substantial amount of missing data across states. Additionally, we are likely underestimating diagnoses prevalence due to limitations in the T-MSIS and claims that may not comprehensively and accurately capture diagnoses.

Conclusions

This cross-sectional study found that Medicaid beneficiaries who reside in more socially vulnerable areas are less likely to have an annual primary care visit. Medicaid policy should focus on addressing geography-based disparities in access to care using innovative new measures, such as the multidimensional SVI.

Supplement 1.

eAppendix 1. T-MSIS cleaning and data quality

eAppendix 2. Activity-based primary care classification approach

eAppendix 3. Identifying FQHC visits

eTable 1. State-level percent of E&M office/outpatient services that have an identifiable individual provider for non-dually eligible Medicaid beneficiaries, under the age of 65, and enrolled the entire year of 2019

eFigure 1. Steps in classifying ambulatory primary care and specialist visits

eTable 2. ICD-10 diagnosis codes used to identify chronic conditions

eTable 3. Full regression results for SVI decile coefficients plotted in Figure 2

eFigure 2. Comparison of main SVI result to results without adjustments for demographic and health characteristics

eFigure 3. Comparison of main SVI result to results for “low concern” data quality states only

eFigure 4. Full regression results for poverty decile coefficients

jamahealthforum-e253020-s001.pdf (913.8KB, pdf)
Supplement 2.

Data Sharing Statement

jamahealthforum-e253020-s002.pdf (14.4KB, pdf)

References

  • 1.Bond A, Pajerowski W, Polsky D, Richards MR. Market environment and Medicaid acceptance: what influences the access gap? Health Econ. 2017;26(12):1759-1766. doi: 10.1002/hec.3497 [DOI] [PubMed] [Google Scholar]
  • 2.Cheung PT, Wiler JL, Lowe RA, Ginde AA. National study of barriers to timely primary care and emergency department utilization among Medicaid beneficiaries. Ann Emerg Med. 2012;60(1):4-10.e2. doi: 10.1016/j.annemergmed.2012.01.035 [DOI] [PubMed] [Google Scholar]
  • 3.Cunningham PJ, O’Malley AS. Do reimbursement delays discourage Medicaid participation by physicians? Health Aff (Millwood). 2009;28(1):w17-w28. doi: 10.1377/hlthaff.28.1.w17 [DOI] [PubMed] [Google Scholar]
  • 4.Luo Q, Bodas M, Vichare A, et al. Primary care provider Medicaid participation across the United States, 2016. J Health Care Poor Underserved. 2023;34(2):703-718. doi: 10.1353/hpu.2023.0059 [DOI] [PubMed] [Google Scholar]
  • 5.MACPAC . Physician Acceptance of New Medicaid Patients: Findings from the National Electronic Health Records Survey; 2021. Accessed October 22, 2023. https://www.macpac.gov/wp-content/uploads/2021/06/Physician-Acceptance-of-New-Medicaid-Patients-Findings-from-the-National-Electronic-Health-Records-Survey.pdf
  • 6.Polsky D, Richards M, Basseyn S, et al. Appointment availability after increases in Medicaid payments for primary care. N Engl J Med. 2015;372(6):537-545. doi: 10.1056/NEJMsa1413299 [DOI] [PubMed] [Google Scholar]
  • 7.Shen YC, Zuckerman S. The effect of Medicaid payment generosity on access and use among beneficiaries. Health Serv Res. 2005;40(3):723-744. doi: 10.1111/j.1475-6773.2005.00382.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Zuckerman S, Williams AF, Stockley KE. Trends in Medicaid physician fees, 2003-2008. Health Aff (Millwood). 2009;28(3)(suppl 1):w510-w519. doi: 10.1377/hlthaff.28.3.w510 [DOI] [PubMed] [Google Scholar]
  • 9.Zuckerman S, Skopec L, Aarons J. Medicaid physician fees remained substantially below fees paid by Medicare in 2019. Health Aff (Millwood). 2021;40(2):343-348. doi: 10.1377/hlthaff.2020.00611 [DOI] [PubMed] [Google Scholar]
  • 10.Cooper Z, Stiegman O, Ndumele CD, Staiger B, Skinner J. Geographical variation in health spending across the us among privately insured individuals and enrollees in Medicaid and Medicare. JAMA Netw Open. 2022;5(7):e2222138. doi: 10.1001/jamanetworkopen.2022.22138 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Erikson CE, Dent RB, Park YH, Luo Q. Historic redlining and contemporary behavioral health workforce disparities. JAMA Netw Open. 2022;5(4):e229494. doi: 10.1001/jamanetworkopen.2022.9494 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Greene J, Blustein J, Weitzman BC. Race, segregation, and physicians’ participation in medicaid. Milbank Q. 2006;84(2):239-272. doi: 10.1111/j.1468-0009.2006.00447.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Lindner SR, Hart K, Manibusan B, McCarty D, McConnell KJ. State- and county-level geographic variation in opioid use disorder, Medication treatment, and opioid-related overdose among Medicaid enrollees. JAMA Health Forum. 2023;4(6):e231574. doi: 10.1001/jamahealthforum.2023.1574 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Morenz AM, Liao JM, Au DH, Hayes SA. Area-level socioeconomic disadvantage and health care spending: a systematic review. JAMA Netw Open. 2024;7(2):e2356121. doi: 10.1001/jamanetworkopen.2023.56121 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Powers BW, Figueroa JF, Canterberry M, et al. Association between community-level social risk and spending among Medicare beneficiaries: implications for social risk adjustment and health equity. JAMA Health Forum. 2023;4(3):e230266. doi: 10.1001/jamahealthforum.2023.0266 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Shen YC, Hsia RY. Do patients hospitalised in high-minority hospitals experience more diversion and poorer outcomes? A retrospective multivariate analysis of Medicare patients in California. BMJ Open. 2016;6(3):e010263. doi: 10.1136/bmjopen-2015-010263 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Siddiqi A, Jones MK, Bruce DJ, Erwin PC. Do racial inequities in infant mortality correspond to variations in societal conditions? A study of state-level income inequality in the US, 1992-2007. Soc Sci Med. 2016;164:49-58. doi: 10.1016/j.socscimed.2016.07.013 [DOI] [PubMed] [Google Scholar]
  • 18.Zhang Y, Li J, Yu J, Braun RT, Casalino LP. Social determinants of health and geographic variation in medicare per beneficiary spending. JAMA Netw Open. 2021;4(6):e2113212. doi: 10.1001/jamanetworkopen.2021.13212 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Dasgupta S, Bowen VB, Leidner A, et al. Association between social vulnerability and a county’s risk for becoming a COVID-19 hotspot-United States, June 1-July 25, 2020. MMWR Morb Mortal Wkly Rep. 2020;69(42):1535-1541. doi: 10.15585/mmwr.mm6942a3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Sommers BD, Tipirneni R. Reconsidering and measuring patient access in Medicaid. JAMA Health Forum. 2024;5(4):e241399. doi: 10.1001/jamahealthforum.2024.1399 [DOI] [PubMed] [Google Scholar]
  • 21.Rosenberg M, Kowal P, Rahman MM, Okamoto S, Barber SL, Tangcharoensathien V. Better data on unmet healthcare need can strengthen global monitoring of universal health coverage. BMJ. 2023;382:e075476. doi: 10.1136/bmj-2023-075476 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.US Department of Health & Human Services . Maternity Care Target Areas (MCTAs) Scoring. Accessed June 23, 2024. https://bhw.hrsa.gov/workforce-shortage-areas/shortage-designation/scoring#mcta-scoring
  • 23.Bleser WK, Tchuisseu YP, Shen H, et al. ACO REACH and Advancing Equity Through Value-Based Payment, Part 2. Published online May 18, 2022. doi: 10.1377/forefront.20220517.755520 [DOI]
  • 24.Powell WR, Sheehy AM, Kind AJH. The Area Deprivation Index is the Most Scientifically Validated Social Exposome Tool Available for Policies Advancing Health Equity. Health Affairs Forefront. Published online July 20, 2023. doi: 10.1377/forefront.20230714.676093 [DOI]
  • 25.McWilliams JM, Weinreb G, Ding L, Ndumele CD, Wallace J. Risk adjustment and promoting health equity in population-based payment: concepts and evidence. Health Aff (Millwood). 2023;42(1):105-114. doi: 10.1377/hlthaff.2022.00916 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Jargowsky PA. Poverty and Place: Ghettos, Barrios, and the American City: Ghettos, Barrios, and the American City. Russell Sage Foundation; 1998. [Google Scholar]
  • 27.Iceland J, Hernandez E. Understanding trends in concentrated poverty: 1980-2014. Soc Sci Res. 2017;62:75-95. doi: 10.1016/j.ssresearch.2016.09.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Sampson RJ. Great American City: Chicago and the Enduring Neighborhood Effect. University of Chicago Press; 2012. doi: 10.7208/chicago/9780226733883.001.0001 [DOI] [Google Scholar]
  • 29.Sen A. Development as Freedom. Knopf Doubleday Publishing Group; 2011. [Google Scholar]
  • 30.Alkire S, Foster J. Understandings and misunderstandings of multidimensional poverty measurement. J Econ Inequal. 2011;9(2):289-314. doi: 10.1007/s10888-011-9181-4 [DOI] [Google Scholar]
  • 31.Dhongde S, Haveman R. Multi-dimensional poverty index: an application to the United States. SSRN Journal. Published online 2015. doi: 10.2139/ssrn.2588584 [DOI]
  • 32.Adler NE, Glymour MM, Fielding J. Addressing social determinants of health and health inequalities. JAMA. 2016;316(16):1641-1642. doi: 10.1001/jama.2016.14058 [DOI] [PubMed] [Google Scholar]
  • 33.Braveman P, Gottlieb L. The social determinants of health: it’s time to consider the causes of the causes. Public Health Rep. 2014;129(Suppl 2)(suppl 2):19-31. doi: 10.1177/00333549141291S206 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.US Department of Health & Human Services . Healthy People 2030. Social Determinants of Health. Accessed June 23, 2024. https://health.gov/healthypeople/priority-areas/social-determinants-health
  • 35.Franchi L, Barreto Parra PN, Chorniy A, et al. A comparative assessment of measures of area-level socio-economic status. SSRN Journal. Published online 2023. doi: 10.2139/ssrn.4547382 [DOI]
  • 36.US Department of Health & Human Services . CDC/ATSDR Social Vulnerability Index (CDC/ATSDR SVI): Overview. Accessed June 23, 2024. https://www.atsdr.cdc.gov/placeandhealth/svi/index.html
  • 37.Al Rifai M, Jain V, Khan SU, et al. State-level social vulnerability index and healthcare access: the behavioral risk factor surveillance system survey. Am J Prev Med. 2022;63(3):403-409. doi: 10.1016/j.amepre.2022.03.008 [DOI] [PubMed] [Google Scholar]
  • 38.Bauer C, Zhang K, Xiao Q, Lu J, Hong YR, Suk R. County-level social vulnerability and breast, cervical, and colorectal cancer screening rates in the US, 2018. JAMA Netw Open. 2022;5(9):e2233429. doi: 10.1001/jamanetworkopen.2022.33429 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Catalano G, Khan MMM, Chatzipanagiotou OP, Pawlik TM. Pharmacy accessibility and social vulnerability. JAMA Netw Open. 2024;7(8):e2429755. doi: 10.1001/jamanetworkopen.2024.29755 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Flanagan BE, Gregory EW, Hallisey EJ, Heitgerd JL, Lewis B. A social vulnerability index for disaster management. J Homel Secur Emerg Manag. 2011;8(1). doi: 10.2202/1547-7355.1792 [DOI] [Google Scholar]
  • 41.Govier DJ, Cohen-Cline H, Marsi K, Roth SE. Differences in access to virtual and in-person primary care by race/ethnicity and community social vulnerability among adults diagnosed with COVID-19 in a large, multi-state health system. BMC Health Serv Res. 2022;22(1):511. doi: 10.1186/s12913-022-07858-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Herrera-Escobar JP, Uribe-Leitz T, Wang J, et al. The Social Vulnerability Index and long-term outcomes after traumatic injury. Ann Surg. 2022;276(1):22-29. doi: 10.1097/SLA.0000000000005471 [DOI] [PubMed] [Google Scholar]
  • 43.Kaalund K, Thoumi A, Bhavsar NA, Labrador A, Cholera R. Assessment of population-level disadvantage indices to inform equitable health policy. Milbank Q. 2022;100(4):1028-1075. doi: 10.1111/1468-0009.12588 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Landman JM, Steger-May K, Joynt Maddox KE, et al. Estimating the effects of race and social vulnerability on hospital admission and mortality from COVID-19. JAMIA Open. 2021;4(4):ooab111. doi: 10.1093/jamiaopen/ooab111 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Lines LM, Long MC, Zangeneh S, et al. Composite indices of social determinants of health: overview, measurement gaps, and research priorities for health equity. Popul Health Manag. 2023;26(5):332-340. doi: 10.1089/pop.2023.0106 [DOI] [PubMed] [Google Scholar]
  • 46.Liu S, Morin SB, Bourand NM, et al. Social vulnerability and risk of suicide in US adults, 2016-2020. JAMA Netw Open. 2023;6(4):e239995. doi: 10.1001/jamanetworkopen.2023.9995 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Mah JC, Penwarden JL, Pott H, Theou O, Andrew MK. Social vulnerability indices: a scoping review. BMC Public Health. 2023;23(1):1253. doi: 10.1186/s12889-023-16097-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Srivastava T, Schmidt H, Sadecki E, Kornides ML. Disadvantage indices deployed to promote equitable allocation of COVID-19 vaccines in the US: a scoping review of differences and similarities in design. JAMA Health Forum. 2022;3(1):e214501. doi: 10.1001/jamahealthforum.2021.4501 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Sunderraj A, Rivera A, Gaddam M, et al. Associations of Social Vulnerability Index with pathologic myocardial findings at autopsy. Front Cardiovasc Med. 2021;8:805278. doi: 10.3389/fcvm.2021.805278 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Trinidad S, Brokamp C, Mor Huertas A, et al. Use of area-based socioeconomic deprivation indices: a scoping review and qualitative analysis. Health Aff (Millwood). 2022;41(12):1804-1811. doi: 10.1377/hlthaff.2022.00482 [DOI] [PubMed] [Google Scholar]
  • 51.Waters HR. Measuring equity in access to health care. Soc Sci Med. 2000;51(4):599-612. doi: 10.1016/S0277-9536(00)00003-4 [DOI] [PubMed] [Google Scholar]
  • 52.Gulliford M, Figueroa-Munoz J, Morgan M, et al. What does ‘access to health care’ mean? J Health Serv Res Policy. 2002;7(3):186-188. doi: 10.1258/135581902760082517 [DOI] [PubMed] [Google Scholar]
  • 53.Bailey MJ, Goodman-Bacon A. The war on poverty’s experiment in public medicine: community health centers and the mortality of older Americans. Am Econ Rev. 2015;105(3):1067-1104. doi: 10.1257/aer.20120070 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Health Resources & Services Administration . 2023. Health Center Data. Accessed October 10, 2024. https://data.hrsa.gov/tools/data-reporting/program-data/national/table?tableName=Full&year=2023
  • 55.O’Reilly-Jacob M, Chapman J, Subbiah SV, Perloff J. Estimating the primary care workforce for Medicare beneficiaries using an activity-based approach. J Gen Intern Med. 2023;38(13):2898-2905. doi: 10.1007/s11606-023-08206-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Luo Q, Park YH, Dor A, Moghtaderi A. Identifying National Provider Identifiers for Community Health Centers. Fitzhugh Mullan Institute for Health Workforce Equity, George Washington University; 2025. Accessed April 11, 2025. https://publuu.com/flip-book/319048/1858808
  • 57.US Centers for Medicare & Medicaid Services . Restructured BETOS Classification System RBCS Final Report; 2022. Accessed October 15, 2023. https://data.cms.gov/sites/default/files/2022-10/dad0f7ef-7ebe-4de8-95a4-8cad3895d2d5/RBCS%202022%20Final%20Report_V03.pdf
  • 58.MACPAC . Section 6: Technical Guide to MACStats; 2015. Accessed July 24, 2023. https://www.macpac.gov/wp-content/uploads/2015/01/Technical-Guide-to-MACStats.pdf
  • 59.Dyer Z, Alcusky MJ, Galea S, Ash A. Measuring the enduring imprint of structural racism on American neighborhoods. Health Aff (Millwood). 2023;42(10):1374-1382. doi: 10.1377/hlthaff.2023.00659 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Butler DC, Petterson S, Phillips RL, Bazemore AW. Measures of social deprivation that predict health care access and need within a rational area of primary care service delivery. Health Serv Res. 2013;48(2 Pt 1):539-559. doi: 10.1111/j.1475-6773.2012.01449.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Finkelstein A, Gentzkow M, Williams H. Sources of geographic variation in health care: evidence from patient migration. Q J Econ. 2016;131(4):1681-1726. doi: 10.1093/qje/qjw023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Labban M, Chen CR, Frego N, et al. Disparities in travel-related barriers to accessing health care from the 2017 National Household Travel Survey. JAMA Netw Open. 2023;6(7):e2325291. doi: 10.1001/jamanetworkopen.2023.25291 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Wolfe MK, McDonald NC, Holmes GM. Transportation barriers to health care in the United States: findings from the National Health Interview Survey, 1997-2017. Am J Public Health. 2020;110(6):815-822. doi: 10.2105/AJPH.2020.305579 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64.Christakis NA, Fowler JH. The spread of obesity in a large social network over 32 years. N Engl J Med. 2007;357(4):370-379. doi: 10.1056/NEJMsa066082 [DOI] [PubMed] [Google Scholar]
  • 65.Godlonton S, Thornton R. Peer effects in learning HIV results. J Dev Econ. 2012;97(1):118-129. doi: 10.1016/j.jdeveco.2010.12.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Chatterjee C, Joshi R, Sood N, Boregowda P. Government health insurance and spatial peer effects: New evidence from India. Soc Sci Med. 2018;196:131-141. doi: 10.1016/j.socscimed.2017.11.021 [DOI] [PubMed] [Google Scholar]
  • 67.Inglis G, McHardy F, Sosu E, McAteer J, Biggs H. Health inequality implications from a qualitative study of experiences of poverty stigma in Scotland. Soc Sci Med. 2019;232:43-49. doi: 10.1016/j.socscimed.2019.04.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Allen H, Wright BJ, Harding K, Broffman L. The role of stigma in access to health care for the poor. Milbank Q. 2014;92(2):289-318. doi: 10.1111/1468-0009.12059 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Massey DS, Denton NA. American Apartheid: Segregation and the Making of the Underclass. 10. print. Harvard Univ. Press; 2003. [Google Scholar]
  • 70.Quillian L. Segregation and poverty concentration: the role of three segregations. Am Sociol Rev. 2012;77(3):354-379. doi: 10.1177/0003122412447793 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Hardeman RR, Medina EM, Boyd RW. Stolen breaths. N Engl J Med. 2020;383(3):197-199. doi: 10.1056/NEJMp2021072 [DOI] [PubMed] [Google Scholar]
  • 72.Dent RB, Vichare A, Casimir J. Addressing structural racism in the health workforce. Med Care. 2021;59(suppl 5):S409-S412. doi: 10.1097/MLR.0000000000001604 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 73.Yearby R, Clark B, Figueroa JF. Structural racism in historical and modern US health care policy. Health Aff (Millwood). 2022;41(2):187-194. doi: 10.1377/hlthaff.2021.01466 [DOI] [PubMed] [Google Scholar]
  • 74.Health Resources & Services Administration . National Health Center Program Uniform Data System (UDS) Awardee Data; 2022. Accessed February 21, 2023. https://data.hrsa.gov/tools/data-reporting/program-data/national
  • 75.Cole MB, Nguyen KH, Byhoff E, Murray GF. Screening for social risk at federally qualified health centers: a national study. Am J Prev Med. 2022;62(5):670-678. doi: 10.1016/j.amepre.2021.11.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 76.Greenwood-Ericksen M, DeJonckheere M, Syed F, Choudhury N, Cohen AJ, Tipirneni R. Implementation of health-related social needs screening at Michigan health centers: a qualitative study. Ann Fam Med. 2021;19(4):310-317. doi: 10.1370/afm.2690 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 77.Malcarney MB, Pittman P, Quigley L, Horton K, Seiler N. The changing roles of community health workers. Health Serv Res. 2017;52(Suppl 1)(suppl 1):360-382. doi: 10.1111/1475-6773.12657 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 78.Horstman C, Lewis C, Bryan A, Federman S. Community Health Centers’ Progress and Challenges in Meeting Patients’ Essential Primary Care Needs. The Commonwealth Fund; 2024. Accessed October 15, 2024. https://www.commonwealthfund.org/publications/issue-briefs/2024/aug/community-health-centers-meeting-primary-care-needs-2024-FQHC-survey
  • 79.Petterson S. Deciphering the Neighborhood Atlas Area Deprivation Index: the consequences of not standardizing. Health Aff Sch. 2023;1(5):qxad063. doi: 10.1093/haschl/qxad063 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

eAppendix 1. T-MSIS cleaning and data quality

eAppendix 2. Activity-based primary care classification approach

eAppendix 3. Identifying FQHC visits

eTable 1. State-level percent of E&M office/outpatient services that have an identifiable individual provider for non-dually eligible Medicaid beneficiaries, under the age of 65, and enrolled the entire year of 2019

eFigure 1. Steps in classifying ambulatory primary care and specialist visits

eTable 2. ICD-10 diagnosis codes used to identify chronic conditions

eTable 3. Full regression results for SVI decile coefficients plotted in Figure 2

eFigure 2. Comparison of main SVI result to results without adjustments for demographic and health characteristics

eFigure 3. Comparison of main SVI result to results for “low concern” data quality states only

eFigure 4. Full regression results for poverty decile coefficients

jamahealthforum-e253020-s001.pdf (913.8KB, pdf)
Supplement 2.

Data Sharing Statement

jamahealthforum-e253020-s002.pdf (14.4KB, pdf)

Articles from JAMA Health Forum are provided here courtesy of American Medical Association

RESOURCES