Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jul 23.
Published in final edited form as: J Health Care Poor Underserved. 2024;35(1):209–224.

The Relationship Between Dental Provider Density and Receipt of Dental Care Among Medicaid-enrolled Adults

Heather L Taylor 1, Nir Menachemi 2, Ann Holmes 3, Bisakha Sen 4, Titus Schleyer 5, Justin Blackburn 6
PMCID: PMC11264445  NIHMSID: NIHMS2006326  PMID: 38661867

Abstract

Objective.

We sought to measure the association of dental provider density and receipt of dental care among Medicaid-enrolled adults.

Methods.

We used four years of Indiana Medicaid claims and enrollment data (2015 to 2018) and the Area Health Resources File to examine the relationship between any dental visit (ADV) or any preventive dental visit (PDV) and three county-level measures of dental provider density (the total number of Medicaid-participating dentists, a binary indicator of a federally qualified health center (FQHC) with a Medicaid-participating dentist, and the overall county dentist-to-population ratio).

Results.

The likelihood of ADV or PDV increased with greater density of Medicaid-participating dentists as well as dentists accepting Medicaid working at an FQHC within the county. The overall dentist-to-population ratio was not associated with dental care use among the adult Medicaid population.

Conclusion.

Dentist participation in Medicaid program may be a modifiable barrier to Medicaid-enrolled adults’ receipt of dental care.

Keywords: Medicaid, dental health services, oral health care utilization, low-income adults

Low-income adults are less likely to use dental care and more likely to have poor oral health than adults at higher income levels.1,2 Although there exists no federal mandate that requires states to provide Medicaid dental coverage for adults, 34 states have opted to offer dental benefits to low-income adults to increase access to dental care for this population.3 States offering comprehensive or extensive Medicaid dental coverage for adults observe greater dental care utilization, lower rates of untreated caries (dental decay), and fewer non-traumatic emergency department dental visits than other states.2,4,5 However, barriers to dental care remain, as even with coverage, Medicaid-enrolled adults visit the dentist on an annual basis half as often as privately-insured adults.6–8 Thus, dental coverage alone may not practically increase use of dental care or improve oral health, and access to care may continue to be a barrier for this population.

Access to dental care among low-income adult populations may be associated with provider supply factors including density and willingness to accept Medicaid coverage. Over 62 million people live in dental professional shortage areas,9 which means using dental care may be challenging for those with Medicaid given the overall strained dental provider supply.10,11 Disparities in access to dental care have persisted even among states that expanded eligibility for Medicaid under the Affordable Care Act and offered dental coverage to a larger adult population, likely due to heterogeneity in the density of dental providers within a geographic area.12–14 While the relationship between dental provider density and dental care use has been broadly studied,15–17 less is known about whether and to what extent Medicaid participation of dental providers affects enrolled adults. While dentists may self-report they are Medicaid providers, significant variation exists in how many patients they actively treat.18 Furthermore, prior research aggregated at the national level12–14 lacks the ability to control for individual- and community-level factors that may confound receipt of dental care and the density of dental providers. Given that low-income adults have the greatest risk of poor oral health outcomes,1,19 dental provider density and participation in Medicaid may be modifiable barriers for low-income adults to receive dental care and improve oral health; participation could be improved through a variety of policy changes such as increasing reimbursement rates and improving program processes.11,20–22

In this study, we measured the association of dental provider density with receipt of overall dental care and preventive dental care among Medicaid-enrolled adults with dental benefits. Specifically, we examine three measures of dental provider density to provide a comprehensive assessment of the extent to which dental provider density and participation in Medicaid are associated with dental care use among low-income adults. Our findings have direct implications for policies that could reduce oral health disparities among low-income adults.

Methods

Data and population.

Data for this pooled cross-sectional study were derived from Medicaid administrative data, including enrollment, provider, and claims data from the Indiana Family and Social Services Administration Office of Medicaid Policy and Planning. To meet our inclusion criteria, each adult was required to be non-elderly (between the ages of 19 and 64) and continuously enrolled in Indiana’s Healthy Indiana Plan (HIP) Plus program or the HIP State Plan Plus program for at least 12 months, with no more than a 30-day gap in coverage. Indiana expanded Medicaid eligibility to non-disabled adults aged 19 to 64 with incomes up to 138% of the federal poverty level (FPL) using a Section 1115 waiver beginning February 1, 2015, which also added dental and vision benefits. The dental benefits include routine dental care, radiographs, and some minor and major restorative services free of charge to enrolled adults. To receive these benefits, enrollees are required to make fixed monthly payments ($1 to $20 depending on income) into a health savings account.23 We excluded from our study any adults who did not maintain their HIP Plus coverage, including those who lost benefits due to non-payment of the required monthly premimum.23

The Health Resources and Service Administration’s Area Health Resources File (AHRF)24 was used to supplement Medicaid claims and enrollment data with economic, population, and provider data attributed to each enrollee by the county of residency.

The total number of unique individuals enrolled in HIP Plus or State Plan Plus insurance programs during the study time period (February 1, 2015 to December 31, 2018) was 610,605. Of these enrollees, 242,022 (39.6%) had less than 12 months of continuous enrollment and were excluded. Continuous enrollment was measured using each individual’s enrollment start and end dates recorded within Indiana’s Medicaid enrollment files.

Dependent variables.

We considered two binary dependent variables: 1) any dental visits (ADV), and 2) the subset of preventive dental visits (PDV) occurring in a 12-month continuous enrollment period. Specifically, ADV was defined as the presence of a dental claim with Common Dental Procedure (CDT) codes D0000 to D9999. We defined a PDV using CDT codes D0120, D0150, or D1000 to D1999 as well as the absence of CDT codes D2000-D9999 on the same claim.25 Our measure of ADVs and PDVs was based on when the claim was submitted and paid, and thus represents one encounter or a complete episode of care. Claims that had both preventive and non-preventive procedures on the same service date were excluded from our definition of a PDV (0.98% of all claims).

Provider density variables (independent variables).

We included three measures of provider density, two measured directly from dental claims and the other from AHRF. Using dental claims, we calculated the number of licensed, Medicaid-participating dentists who submitted at least one dental claim within each year of our study period and the annual county-level number of HIP Plus and State Plan Plus enrollees. We used Medicaid provider data to determine the county where each Medicaid-participating dentist’s practice address was located. After analyzing the distribution of the number of Medicaid-participating dentists per 1,000 enrollees, we categorized this variable into five groups by rounding to the nearest whole integer (0-1, 2, 3, 4, and 5 or more dentists). We also used dental claims to create a binary indicator of whether the county had an FQHC with a Medicaid-participating dentist (yes, no). We used Medicaid provider data to identify the address of the FQHC where the dentist practiced and subsequently the county where the FQHC was located.

We used AHRF to identify the number of dentists with an active license within each county. According to AHRF, business addresses are the primary tool used to identify which counties have licensed, practicing dentists.24 The count of actively licensed dentists was divided by the total county population and expressed as the number of dentists per 5,000 population. After examining the distribution of the number of dentists with active licenses per 5,000 population, we categorized this variable into four groups by rounding to the nearest whole integer (0-1, 2, 3, and 4 or more).

Additional independent variables.

Other included variables were age (i.e., 19-24, 25-34, 35-44 years old, 45-54, and 55-64 years old), sex (male, female), race/ethnicity (i.e., White, Black, Hispanic, other), marital status (i.e., married, never married, divorced, widowed/unknown), calendar year of enrollment, and the total number of months continuously enrolled (i.e., 12 to 19 months, 20 to 28 months, 29 to 40 months, 41 to 47 months). Using enrollee medical claims, we also determined whether enrollees had an annual well visit (yes, no) with a physician during each 12-month enrollment period. Consistent with prior research, we defined an annual well visit as the presence of a medical claim with Current Procedural Terminology Codes 99385, 99386, 99387, 99395, 99396, and 99397.26 Finally, Rural-Urban Commuting Area (RUCA) codes (measures of population density, urbanization, and daily commuting) were used to characterize the rurality of the county where the enrollee resided.27 We aggregated RUCA codes into urban, large rural, small rural, and isolated as outlined in previously published methodology.28

Statistical analysis.

We assessed the distribution of included enrollee characteristics using means and frequencies. We used chi-square and z-test statistics to measure associations between enrollees’ use of dental care and dental provider density measures along with all other individual- and county-level variables. Because our dataset contained repeated observations by enrollees and multiple enrollees are nested within counties, we used multi-level linear probability regression models to assess the effect of enrollee- and county-level variables on dental care use while accounting for the interdependency of observations. We tested multiple model specifications to check the robustness of our findings, and the amount of total variation explained at the county-level was consistently small (~0.05%) across each specification. Therefore, a two-level hierarchical model accounting for repeated observations was selected as the most parsimonious and statistically efficient modelling strategy.

All estimates were reported with 95% confidence intervals and standard errors were clustered at the enrollee-level. We used SAS 9.2 software29 for data management and Stata SE version 1730 software for all our statistical analyses. This study received an exemption from review by the Indiana University Institutional Review Board.

Results

A total of 358,685 enrollees contributed 618,722 enrollee-year observations during the study period. Healthy Indiana Plan Plus and State Plan Plus enrollees were primarily White (73.5%), female (65.0%), and continuously enrolled for an average length of 29 months. Approximately 40.1% (n=248,115) of enrollees had ADV and 28.8% (n=177,975) had a PDV during at least one enrollment period. Table 1 presents descriptive statistics and bivariate associations between enrollee characteristics and ADV, or PDV, during a 12-month continuous enrollment period.

Table 1 –

Associations between individual- and county-level variables and any dental visit and any preventive dental visit among Medicaid-enrolled adultsa

Total enrollees N=618,772 ADVb n=248,115 (40.1%) p-value PDVc n=177,975 (28.8%) p-value
Individual-level Factors
Sex <0.001 <0.001
    Female 402,001 (65.0) 173,021 (69.7) 126,572 (71.1)
    Male 216,771 (35.0) 75,094 (30.3) 51,403 (28.9)
Race/Ethnicity <0.001 <0.001
    White 455,058 (73.5) 179,885 (72.5) 128,704 (72.3)
    Black 82,154 (13.3) 35,191 (14.2) 24,816 (13.9)
    Hispanic 33,253 (5.4) 14,108 (5.7) 10,827 (6.1)
    Other 48,307 (7.8) 18,931 (7.6) 13,628 (7.7)
Marital Status <0.001 0.044
    Married 153,502 (24.8) 61,197 (24.7) 44,727 (25.1)
    Never married 297,922 (48.2) 119,217 (48.0) 85,842 (48.2)
    Divorced 133,964 (21.6) 55,989 (22.6) 38,939 (21.9)
    Widowed/Unknown 33,384 (5.4) 11,712 (4.7) 8,467 (4.8)
Age categories <0.001 <0.001
    19-24 76,653 (12.9) 33,613 (13.5) 25,707 (14.4)
    25-34 162,374 (26.2) 69,620 (28.1) 48,834 (27.4)
    35-44 144,619 (23.4) 59,972 (24.2) 42,479 (23.9)
    45-54 134,770 (21.8) 51,746 (20.8) 36,382 (20.5)
    55-64 97,356 (15.7) 33,164 (13.4) 24,573 (13.8)
Distance to nearest DDSd <0.001 <0.001
    <0.5 mile 143,419 (23.2) 58,076 (23.4) 41,032 (23.1)
    0.5 to 1.0 miles 172,529 (27.9) 70,141 (28.3) 50,013 (28.1)
    1.0 to 2.0 miles 128,092 (20.7) 51,877 (20.9) 37,796 (21.2)
    2.0 to 5.0 miles 87,645 (14.1) 35,007 (14.1) 25,605 (14.4)
    >5.0 miles 87,087 (14.1) 33,014 (13.3) 23,529 (13.2)
Year Indicator <0.001 <0.001
    2015 140,539 (22.7) 59,415 (24.0) 43,129 (24.2)
    2016 186,381 (30.1) 75,989 (30.6) 54,133 (30.4)
    2017 200,405 (32.4) 77,755 (31.3) 55,739 (31.3)
    2018 91,447 (14.8) 34,956 (14.1) 24,974 (14.1)
Months of Enrollment <0.001 <0.001
    12 to 19 months 165,110 (26.7) 63,477 (25.6) 44,101 (24.8)
    20 to 28 months 148,601 (24.0) 58,352 (23.5) 41,436 (23.3)
    29 to 40 months 156,439 (25.3) 63,326 (25.5) 45,659 (25.6)
    41 to 47 months 148,622 (24.0) 62,960 (25.4) 46,779 (26.3)
Annual well visit
    Yes 160,365 (25.9) 78,433 (31.6) <0.001 61,226 (34.4) <0.001
    No 458,407 (74.1) 169,682 (68.4) 116,749 (65.6)
County-level variables
RUCA designatione <0.001 <0.001
    Urban 483,513 (78.2) 195,684 (78.9) 140,953 (79.2)
    Large Rural 86,168 (13.9) 33,157 (13.4) 23,408 (13.2)
    Small Rural 31,563 (5.1) 12,531 (5.1) 8,947 (5.0)
    Isolated 17,416 (2.8) 6,695 (2.7) 4,667 (2.6)
Open in a new tab

Source: Enrollment, provider, and claims data from the Indiana Family and Social Services Administration Office of Medicaid Policy and Planning

a –

Specific to Healthy Indiana Program (HIP) Plus and State Plan Plus adult Medicaid enrollees in Indiana with at least 12 months continuous enrollment between Feb. 1, 2015 and Dec. 31, 2018. (N= 618,722 person-enrollment years)

b –

Any dental visit

c –

Preventive dental visit

d –

Dentist (general practitioner)

e –

Rural-Urban Commuting Areas as defined by the U.S. Department of Agriculture Economic Research Service

Table 2 presents descriptive statistics and bivariate associations between dental provider density measures and ADV, or PDV, during a 12-month continuous enrollment period. All three measures of dental provider density (number of Medicaid-participating dentists per 1,000 enrollees, number of active dentists per 5,000 people, and whether the county had an FQHC with a Medicaid-participating dentist) were significantly associated with ADV and any PDV (all p-values < .001). Across all counties in Indiana (n=92), the number of dental visits per year ranged from 22 to 13,265 (data not presented in Table 2).

Table 2 –

Associations between variables of provider density and any dental visit and any preventive dental visit among Medicaid-enrolled adultsa

Total enrollees N=618,772 ADVb n=248,115 (40.1%) p-value PDVc n=177,975 (28.8%) p-value
Provider density variables d
Number of Medicaid-participating DDSe per 1,000 enrollees <0.001 <0.001
    0-1 21,776 (3.5) 7,886 (3.2) 5,631 (3.2)
    2 103,314 (16.7) 39,852 (16.1) 29,156 (16.4)
    3 220,959 (35.7) 87,850 (35.4) 63,320 (35.6)
    4 115,479 (18.7) 47,710 (19.2) 33,884 (19.0)
    5 or more 157,244 (25.4) 64,817 (26.1) 45,984 (25.8)
Dentists with an active licensef per 5,000 population <0.001 <0.001
    0-1 38,625 (6.2) 14,800 (6.0) 10,304 (5.8)
    2 156,701 (25.3) 61,593 (24.8) 43,872 (24.6)
    3 277,925 (44.9) 111,739 (45.0) 81,433 (45.8)
    4 or more 145,521 (23.5) 59,983 (24.2) 42,366 (23.8)
Medicaid-participating DDSe at FQHCg <0.001 <0.001
    No 297,178 (48.0) 117,036 (47.2) 83,489 (46.9)
    Yes 321,594 (52.0) 131,079 (52.8) 94,486 (53.1)
Open in a new tab

Source: Enrollment, provider, and claims data from the Indiana Family and Social Services Administration Office of Medicaid Policy and Planning

a –

Specific to Healthy Indiana Program (HIP) Plus and State Plan Plus adult Medicaid enrollees in Indiana with at least 12 months continuous enrollment between Feb. 1, 2015 and Dec. 31, 2018 (N= 618,722 person-enrollment years)

b –

Any dental visit

c –

Preventive dental visit

d –

Measured at the county-level

e –

Dentist (general practitioner)

f –

Professionally active dentist as defined by the US Health Resources and Service Administration’s Area Health Resource File

g –

Federally Qualified Health Center

Table 3 presents results from our preferred multi-level linear probability regression model. After controlling for all other enrollee- and county-level variables, having at least two or more Medicaid-participating dentists per 1,000 enrollees in a county was associated with a higher likelihood that an enrollee had ADV (1.2 percentage points (pps); p=0.038) or PDV (1.1 pps; p=.044) within a year. Similarly, having a Medicaid-participating dentist in an FQHC within the county was also associated with a higher likelihood an enrollee had ADV (0.9 pps; 95% CI 0.6 to 1.2) and PDV (0.9 pps; 95% CI 0.6 to 1.2). In contrast, the total number of actively licensed dentists in the county was not found to be significantly associated with the likelihood of an enrollee having ADV or PDV.

Table 3 –

Multi-level linear probability modelsa of receipt of any dental visit and any preventive dental visit among Medicaid-enrolled adultsa

ADVb PDVc
Partial effects 95% CI Partial effects 95% CI
Individual-level variables
Sex
    Male ref ref ref ref
    Female 0.059*** (0.056, 0.062) 0.054*** (0.052, 0.057)
Race/ethnicity
    White ref ref ref ref
    Black 0.029*** (0.025, 0.033) 0.016*** (0.012, 0.020)
    Hispanic 0.018*** (0.011, 0.024) 0.031*** (0.025, 0.036)
    Other −0.004 (−0.010, 0.002) −0.001 (−0.005, 0.004)
Marital Status
    Married ref ref ref ref
    Never married −0.012*** (−0.016, −0.009) −0.014*** (−0.017, −0.011)
    Divorced 0.015*** (0.011, 0.019) −0.005** (−0.009, −0.002)
    Widowed/Unknown −0.041*** (−0.048, −0.035) −0.035*** (−0.041, −0.029)
Age
    19-24 years old ref ref ref ref
    25-34 years old −0.003 (−0.008, 0.001) −0.029*** (−0.033, −0.025)
    35-44 years old −0.021*** (−0.025, −0.016) −0.038*** (−0.043, −0.034)
    45-54 years old −0.049*** (−0.054, −0.044) −0.059*** (−0.063, −0.054)
    55-64 years old −0.085*** (−0.090, −0.080) −0.072*** (−0.077, −0.067)
Months of Enrollment
    12 to 19 months ref ref ref ref
    20 to 28 months 0.014*** (0.011, 0.018) 0.016*** (0.013, 0.019)
    29 to 40 months 0.028*** (0.025, 0.032) 0.031*** (0.027, 0.034)
    41 to 47 months 0.042*** (0.038, 0.047) 0.050*** (0.046, 0.053)
Year
    2015 ref ref ref ref
    2016 −0.009*** (−0.012, −0.006) −0.012*** (−0.015, −0.009)
    2017 −0.027*** (−0.030, −0.024) −0.024*** (−0.027, −0.021)
    2018 −0.027*** (−0.031, −0.023) −0.022*** (−0.026, −0.018)
Distance to nearest DDSd
    <0.5 miles ref ref ref ref
    0.5 – 1.0 miles 0.001 (−0.003, 0.005) 0.002 (−0.001, 0.005)
    1.0 – 2.0 miles −0.001 (−0.004, 0.004) 0.005** (0.002, 0.009)
    2.0 – 5.0 miles −0.001 (−0.005, 0.004) 0.007** (0.003, 0.011)
    >5.0 miles −0.011*** (−0.016, −0.006) −0.004 (−0.009, 0.001)
Annual well visit 0.083*** (0.080, 0.086) 0.091*** (0.088, 0.093)
County-level variable
RUCA designatione
    Urban ref ref ref ref
    Large Rural −0.006** (−0.012, −0.002) −0.008** (−0.011, −0.003)
    Small Rural 0.004 (−0.003, 0.010) 0.003 (−0.003, 0.009)
    Isolated −0.002 (−0.011, 0.007) −0.007 (−0.015, 0.001)
Dental density measures f
Number Medicaid-participating DDSd/1,000 enrollees
    0-1 ref ref ref ref
    2 0.012** (0.004, 0.020) 0.011** (0.003, 0.018)
    3 0.017*** (0.009, 0.025) 0.010** (0.003, 0.017)
    4 0.031*** (0.023, 0.039) 0.016*** (0.009, 0.024)
    5 or more 0.026*** (0.018, 0.036) 0.017*** (0.009, 0.025)
Number of dentists with active licenseg per 5,000 population
    0-1 ref ref ref ref
    2 −0.002 (−0.008, 0.004) 0.004 (−0.002, 0.010)
    3 −0.006 (−0.013, 0.001) 0.006 (−0.001, 0.012)
    4 or more −0.004 (−0.012, 0.004) <0.001 (−0.007, 0.008)
Medicaid-participating DDSd at FQHCh
    No ref ref ref ref
    Yes 0.009*** (0.006, 0.012) 0.009*** (0.006, 0.012)
Open in a new tab

Source: Enrollment, provider, and claims data from the Indiana Family and Social Services Administration Office of Medicaid Policy and Planning

Note: Analysis clustered at recipient-level to account for repeated observations within the same individual

ref = Reference category, omitted from model

a –

Specific to Healthy Indiana Program (HIP) Plus and State Plan Plus adult Medicaid enrollees in Indiana with at least 12 months continuous enrollment between Feb. 1, 2015 and Dec. 31, 2018

b –

Any Dental Visit

c –

Preventive Dental Visit

d –

Dentist (general practitioner)

e –

Rural-Urban Commuting Areas as defined by the US Department of Agriculture

f –

Measured at the county-level

g –

Professionally active dentist as defined by the US Health Resources and Service Administration’s Area Health Resource File

h –

Federally Qualified Health Center

***

p<0.001

**

p<0.05

*

p<0.10

Other variables associated with a higher likelihood of having ADV included having an annual well visit (8.3 pps; 95% CI 8.0 to 8.6) and being female (5.9 pps; 95% CI 5.6 to 6.2). Both individuals who are Black (2.9 pps; 95% CI 2.5 to 3.3) and Hispanic (1.8 pps; 95% CI 1.1 to 2.4) had a greater likelihood of having ADV than individuals who are White. Variables found to be associated with a lower likelihood of an enrollee having ADV included further distance to the dentist (greater than 5 miles) (−1.1 percentage points (pps); 95% CI −1.6 to −0.6), age over 34 (−2.1 pps; 95% CI −2.5 to −1.6) and living in a large rural area (−0.6 pps; 95% CI −1.2 to −0.2). Findings were similar when the dependent variable was PDV, although the effect sizes were attenuated. For instance, having an annual well visit was associated with a 9.1 pps higher likelihood (95% CI 8.8 to 9.3) of having a PDV.

Discussion

We investigated characteristics associated with ADV and PDV, with a particular focus on the relationship between dental provider density and receipt of dental care among Medicaid-enrolled adults in Indiana. Outlined below, we draw several conclusions based on these analyses that are specific to Medicaid-provider density, FQHCs with a dentist, annual well-visits, enrollee demographics, and overall low utilization rates within this populations.

We observed greater density of Medicaid-participating dentists was associated with increased enrollee dental care use, whereas the overall dentist-to-population ratio was not associated with dental care use among this population. Total number of dentists in a county is a commonly used measure for studying how dental provider density affects utilization of dental care and/or oral health outcomes. For example, studies of Medicaid expansion relied on the number of dentists in a county.12,14 However, our findings within the Indiana Medicaid population suggest that the total number of dentists in a geographic area is an imprecise measure of dental provider density. For vulnerable, low-income groups enrolled in Medicaid, dentists practicing within their county may not all accept and participate in Medicaid or may do so on a limited basis. Ignoring Medicaid participation among dentists masks true access within this population, and consequently policy analyses may be biased for failing to account for these access barriers. Furthermore, as a result of low dentist participation in Medicaid, enrollees may experience challenges scheduling a dental appointment in a timely manner or face greater out-of-pocket costs when seeking dental care.31,32 Given the implications of low dentist participation in Medicaid, policies that incentivize dentists to practice in shortage areas and participate in Medicaid programs should be considered as well as increasing Medicaid reimbursement levels. 21,22 State and federal loan forgiveness programs could be used to provide loan repayment in exchange for requiring the dentist to work in designated health professional shortage or medically underserved areas.33 However, to ensure equity in dental service access, our findings underscore the need for these programs to require provider participation in Medicaid.

Having a Medicaid-participating dentist in an FQHC was also associated with a greater likelihood of dental care use among adult enrollees. Nasseh et al. found that dentists who practice in FQHCs treat a significantly higher proportion of Medicaid patients than dentists who do not practice in FQHCs.34 However, not all FQHCs are configured to provide oral health services directly to patients, as some FQHCs refer patients out to dentists in the local community who agree to treat FQHC patients.35 Co-location of medical and dental services at an FQHC may allow patients to receive same-day services without having the need to take additional time off work.36 Dentists not co-located within an FQHC introduce the possibility of additional barriers to care including transportation, scheduling delays, or the inability to take additional time off work. Thus, our evidence suggests that ensuring the configuration and space for dental providers to practice within FQHCs would increase dental utilization. However, a number of barriers may influence whether FQHCs decide to directly provide oral health services, including difficulty recruiting and retaining oral health professionals, Medicaid reimbursement policies, and state scope-of-practice laws and regulations.31 Since FQHCs rely heavily on reimbursement from Medicaid to cover the costs of providing services to their patient populations, increasing reimbursement rates for oral health services may be important to incentivize, recruit, and retain dentists to practice in FQHCs.35 In addition, expanding scope of practice laws that allow dental therapists and midlevel providers to provide routine and preventive dental care to Medicaid populations could also help alleviate the shortage of dental providers, particularly at FQHCs.37 However, as of 2019, Indiana was one of 39 states that have not legalized the practice of dental therapists or other midlevel dental providers, despite the evidence of potential benefits among underserved populations.38,39

Enrollees who had an annual well-visit were also more likely to have ADV and a PDV. Previous research has demonstrated annual well visits to be associated with adults’ use of key preventive services such as vaccinations, tobacco cessation programs, and cancer screenings.40,41 This preventive care-seeking behavior is likely related to individual characteristics, perhaps health consciousness and health beliefs.42 Adults who regularly seek preventive health care may also be more likely to seek routine dental care. Administrative claims lack data on individuals’ health beliefs or their perceived need for care. However, our results suggest co-location of adult medical and dental services within FQHCs may promote more comprehensive use of preventive care. Additional research may consider concurrence of preventive medical and dental care to further test this hypothesis.

We observed enrollee characteristics such as race, sex, and marital status associated with dental care use. In this study, we found individuals who are Black or Hispanic were more likely than individuals who are White to have ADV and a PDV during an annual enrollment period. Previous studies of dental care utilization disparities have been equivocal: in some, racial/ethnic minorities have lower utilization,43 while others have observed the opposite.44 We were unable to determine whether the differences by race/ethnicity observed in our study are driven by lack of clinical need for dental care or other barriers to the receipt of dental care. A better understanding of racial/ethnic disparities in dental care utilization and how this relates to barriers or facilitators of receiving care is needed. Ultimately, alleviating cost barriers by providing dental coverage for low-income adults may increase dental care utilization. However, the extent to which dental coverage through Medicaid may reduce racial disparities in dental care utilization is not well-studied given most Medicaid programs have not historically provided this type of coverage.

Notably, the overall use of dental care in this population was infrequent; less than half of our study population received an annual dental visit and less than one-third regularly had preventive dental care. These rates of dental utilization are below general population estimates derived from the 2020 National Health Interview Survey which found that 63% of U.S. adults over the age of 18 received a dental exam or cleaning in the past 12 months.45 This raises questions regarding how accessible dental care may be for Medicaid-enrolled adults despite having coverage. Given that national health objectives in Healthy People 2030 recommend increasing the proportion of individuals who use the oral health care system,46,47 state policy-action and targeted interventions specific to the barriers faced by low-income populations seeking dental care are warranted.

Some study limitations should be noted. First, county boundaries may not represent actual community dental service areas for enrollees. Enrollees may live very close to the border of another county and choose to travel to a dentist outside of the county in which they reside. Second, we were unable to account for potentially important enrollee-level variables, such as the need for dental care, oral health beliefs, transportation barriers, or travel costs because these are not available in claims data. Third, Indiana’s HIP program was approved under a Section 1115 waiver unique to Indiana. This may limit the generalizability of this study’s findings to other state Medicaid programs with different benefit packages or provider networks. Fourth, we excluded adults who were not continuously enrolled in Medicaid for at least one year, and thus our findings cannot estimate patterns of dental care use among low-income groups who lose Medicaid coverage within a shorter period. Fifth, our study examined dental care use following Indiana’s Medicaid expansion in 2015 which provided a large number of previously uninsured adults with dental benefits (40,000 new enrollees were added to Indiana’s Medicaid program in first month of expansion).48 This influx of new patients into the dental care system may have had an effect on dental utilization rates as providers adjusted to new demand. Sixth, some dentists may choose to provide pro bono dental services to Medicaid enrollees and thus these services would not be captured in Medicaid claims files. However, this occurrence is likely rare. Finally, this study is cross-sectional, and findings should not be interpreted as causal relationships.

In summary, the density of Medicaid-participating dentists and whether the enrollee’s county has a dentist practicing within an FQHC are relevant factors associated with receipt of dental care. Ultimately, for low-income adults to use their Medicaid dental benefits, dental providers must be available in their community and actively participate in state Medicaid dental programs.18 Policies aimed at increasing dentist participation in Medicaid programs may lead to downstream improvements in oral health outcomes among enrollees, such as greater receipt of dental care.

Acknowledgement

Research reported in this publication was in part supported by the National Library of Medicine of the National Institutes of Health under award number T15LM012502. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the National Library of Medicine.

List of abbreviations used in the article:

ADV

Any dental visit

PDV

Preventive dental visit

HIP

Healthy Indiana Plan

AHRF

Area Health Resource File

CDT

Common Dental Procedure

FPL

Federal Poverty Level

NPI

National Provider Identifier

DDS

General dentist

FFS

Fee for service

RUCA

Rural-urban commuting area

MCO

Managed care organization

FQHC

Federally qualified health center

DHPSA

Dental health professional shortage area

Footnotes

Conflict of Interest: All authors have no potential conflicts of interests to report.

Statement of Institutional Review Board Approval

This study received an exemption from review by the Indiana University Institutional Review Board (#2007943716).

Contributor Information

Heather L. Taylor, Department of Health Policy and Management, Indiana University Fairbanks School of Public Health, 1050 Wishard Blvd, Indianapolis IN 46202.

Nir Menachemi, Indiana University Fairbanks School of Public Health Scientist, Regenstrief Institute, Inc., Indianapolis IN 46202.

Ann Holmes, Department of Health Policy and Management, Indiana University Fairbanks School of Public Health, Indianapolis IN 46202.

Bisakha Sen, Department of Health Care Organization & Policy, School of Public Health, University of Alabama at Birmingham, Birmingham, AL 35294.

Titus Schleyer, Center for Biomedical Informatics, Regenstrief Institute, Indianapolis IN 46202.

Justin Blackburn, Department of Health Policy and Management, Indiana University Fairbanks School of Public Health, Indianapolis IN 46202.

References.

RESOURCES