Impact of Medicaid Dental Coverage Expansion on Self-reported Tooth Loss in Low-Income Adults

Abstract

Background:

Low-income adults delay dental care due to cost more than any other healthcare service. These delays lead to tooth decay, periodontal disease, and tooth loss. Expanding Medicaid dental coverage has increased dental visits, but the potential impact on previously unmet oral health needs is not well understood.

Methods:

This analysis estimated the association between Medicaid dental expansion and tooth loss. Data on self-reported tooth loss among adults under 138% FPL were obtained from the Behavioral Risk Factor Surveillance System. A differences-in-differences regression was estimated. Additional analyses stratified by age and separated extensive and limited dental benefits.

Results:

Expanding Medicaid dental coverage is associated with increased probability of total tooth loss by 1 percentage-point in the total sample, representing a 20% relative increase from the pre-expansion rate. This increase was concentrated in states offering extensive dental benefits and was largest (2.5 percentage-point greater likelihood) among adults 55–64, for whom both extensive and limited dental benefits were associated with total tooth loss.

Conclusion:

Medicaid expansion with extensive dental benefits was associated with increased total tooth loss among low-income adults. This finding suggests that greater access to dental care addressed previously unmet oral health needs for this population.

Practical Implications:

As public dental coverage continues to expand, dental professionals may find themselves treating a greater number of patients with large previously unmet oral health needs. Further research to understand the long-term effects of Medicaid dental insurance for adults on their oral health is needed.

Low-income adults often delay or forgo dental care due to high cost and lack of adequate dental insurance.¹ Delaying or forgoing dental care can allow periodontal disease and tooth decay to progress, which may eventually lead to increased tooth loss. Adult tooth loss diminishes health and wellbeing, as well as overall quality of life and employment opportunities.^2,3 Low-income adults bear the greatest burden of tooth loss; 60% of low-income adults, age 25–64, have lost a permanent tooth (compared to 40% of higher income adults); one in ten low-income adults have no remaining teeth by the age of 65.^1,4,5,6

To address the unmet dental needs of low-income adults, Medicaid has become a critical source for accessing dental care. In recent years, Medicaid dental services have accounted for an increasing share of Medicaid budgets nationally and for many states.^7,8,9 Yet, the federal government does not require Medicaid plans to cover adult dental services, so Medicaid dental programs vary among states.¹⁰ In 2018, four states did not cover any adult Medicaid dental services and fifteen states only reimbursed providers for emergency dental treatment.⁸ Even among states covering non-emergency dental services, Medicaid plans differ across states in benefits and services covered.⁸ Additionally, state Medicaid eligibility also varies. Most recently, the Affordable Care Act (ACA) expanded Medicaid by raising the income-based eligibility threshold to 138% of the Federal Poverty Level (FPL). However, low-income adults living in states which have not yet opted to expand Medicaid or those that do not cover non-emergency dental services continue to lack access to dental insurance. Differences in dental coverage availability and Medicaid eligibility have resulted in a public dental system rife with inequity, inconsistency, and coverage gaps.

In this study, we examine whether the ACA Medicaid expansions in states offering dental benefits were associated with tooth loss among low-income adults. Tooth loss is a marker of previously unmet dental care needs and a major oral health indicator. New dental coverage through the Medicaid Expansion increased access to dental care, which can address previously unmet needs. For individuals whose teeth cannot be saved and for whom extraction is needed, tooth loss rates might increase following new coverage. Despite the limited research investigating Medicaid dental expansions and tooth loss, there is strong evidence that Medicaid dental expansions were associated with increased dental services. Previous studies show an increase in access to dental care following Medicaid dental coverage expansions.^10,11,12,13 The pre-ACA Medicaid expansion in Oregon was associated with reduced unmet dental needs among adults gaining new dental coverage.^14,15,16 More recently, there is evidence of increased use of dental services in states that expanded Medicaid under the ACA and offered dental benefits.^17,18,19 Furthermore, there is evidence of larger effects of the ACA Medicaid expansion on dental services use with more generous dental benefits. ^10,17 Building off the evidence that Medicaid dental expansions increased dental care utilization, we hypothesize that new access to dental coverage could impact oral health outcomes.

Despite the evidence on dental services use, less is known about how expanding Medicaid dental benefits affects oral health outcomes such as tooth loss. The existing research offers some although inconclusive findings which motivate our research question. One pre-ACA Medicaid expansion study found that new coverage was associated with a reduction in broken teeth and cavities, but not with tooth loss.¹¹ Another study reported that the ACA Medicaid expansion was associated with greater likelihood of edentulism (total tooth loss).²⁰ However, when categorizing states by whether or not they offer dental benefits to Medicaid adults, that study reported an opposite, small and statistically non-significant association between total tooth loss and expansion with dental benefits. Furthermore, that study did not include any other levels of tooth loss or removal (only edentulism). Finally, a recent study reported no association between state Medicaid dental benefits and tooth loss of low-income adults 65 and older.²¹

Methods

Data

Data were obtained from the Behavioral Risk Factors Surveillance System (BRFSS), a nationally representative, cross-sectional survey.²² Each year, BRFSS includes a new cross-sectional sample in each state. During even years, BRFSS administers an oral health module that includes a measure of self-reported tooth loss up to the interview date. The questionnaire specifically asks for the number of teeth removed due to tooth decay or gum disease, excluding wisdom teeth unless those teeth were removed due to decay or disease.²² BRFSS categorizes that measure as zero tooth loss, losing five or fewer teeth, losing six or more teeth (but not all), and losing all teeth. We included data from two years before the Medicaid expansion (2010 and 2012) and three years after the 2014 expansions (2014, 2016, and 2018). BRFSS also obtains data on demographic and socioeconomic characteristics.

Sample

The primary sample included individuals reporting income below 138% of the federal poverty level (FPL). We estimated each respondent’s FPL based on the reported household size, state, and midpoint of the self-reported income category (since income is based on categories in BRFSS). In sensitivity analyses, we included adults reporting household incomes below 100%, 200%, or 400% FPL. Further, as a check of the validity of the research design, we included a sample of adults with income at or above 400% FPL who would not likely be impacted by the Medicaid expansion. Additionally, the primary sample was restricted to adults between age 25 and 64 to minimize effects from the ACA dependent mandate which was associated with dental coverage gains among young adults.^23,24

State Treatment and Control Group Assignment

Following previous studies of Medicaid expansions, we assigned states into treatment and control groups based on their ACA Medicaid expansion status and for expansion states, by their level of dental benefits (Supplemental Table 1). ^8,9,25,26,27 Respondents in states which did not expand Medicaid under the ACA between the years 2014 and 2018 were assigned into the control group. We excluded respondents in Medicaid expansion states that did not cover dental benefits above emergency services for adults. This left respondents in nineteen states that expanded Medicaid between 2014–2018 and whose Medicaid programs covered more than emergency-only dental benefits. Nine Medicaid expansion states offered extensive dental benefits, covering more services with higher or non-existent annual caps.²⁷ The other ten expansion states offered limited dental benefits, defined by lower annual cap on benefits and fewer services covered.

Statistical Analysis

We examined the association between the ACA Medicaid expansions with dental benefits and tooth loss in a regression model that utilized within state variation over time (i.e., removed time-invariant differences between states) and controlled for national time trends in tooth loss shared among states. This model was generally similar to a difference-in-differences design, but included states that expanded Medicaid under the ACA in different years. Of the 19 included expansion states, 15 expanded in 2014, while 4 expanded in later years. We also estimated a classical difference-in-differences model focusing on the states that expanded in 2014 compared to non-expansion states. The regression models adjusted for individual-level covariates conceptually relevant for oral health including age, gender, race/ethnicity, education, marital status, and whether there were children living at home.^28,29 To distinguish between the levels of Medicaid dental benefits, we estimated an additional model that separated expansion states into those offering extensive benefits and those offering limited benefits. For all analyses, standard errors were clustered at the state level.³⁰

The regression models were estimated for each of the four tooth loss categories, one at a time, relative to the other three categories using a linear probability model and sampling weights provided by BRFSS. Specifically, we estimated the association of Medicaid expansion and the likelihood of zero tooth loss, losing five or fewer teeth, losing six or more (but not all) teeth, and losing all teeth, each relative to the other categories combined, in a separate regression for each outcome. In addition to estimating the average effect in the total sample, we estimated effects for subgroups stratified by ten-year age categories from 25–34 to 55–64 since age is associated with greater prevalence of tooth loss and more accumulated unmet dental care needs.

For the difference-in-differences model to generate causal estimates, expansion and non-expansion states are assumed to share unobserved time events. To test this assumption, we compared outcome trends before the expansion between expansion and non-expansion states. More details of the regression model and tests are provided in appendices 1 and 2 online.

Results

Table 1 reports summary statistics for tooth loss categories and control variables by year and Medicaid expansion status. Figure 1 depicts the year-by-year rates of tooth loss for the expansion and non-expansion states.

Table 1:

Sample Rates (%) of Outcome and Control Variables by Year Separately for Expansion and Non-Expansion States

	2010		2012		2014		2016		2018
	Expansion	Non-Expansion	Expansion	Non-Expansion	Expansion	Non-Expansion	Expansion	Non-Expansion	Expansion	Non-Expansion
Zero Tooth Loss	38.9	42.1	41.3	42.3	41.3	42.4	41.4	43.8	45.0	45.0
1–5 Teeth Lost	38.8	38.9	36.4	37.4	37.6	37.2	37.2	35.2	34.3	35.2
>=6Teeth Lost, but not all	15.9	14.0	16.1	14.6	14.5	14.9	14.9	14.2	14.5	13.2
Total Tooth Loss	6.4	4.9	6.2	5.7	6.6	5.6	6.5	6.8	6.2	6.7
Age 25–29	12.3	15.5	14.4	14.2	14.4	15.8	13.4	14.4	12.9	14.8
Age 30–34	16.1	17.0	15.7	15.7	15.6	15.0	15.5	15.6	14.8	16.7
Age 35–39	16.1	14.3	13.4	13.5	12.4	12.4	14.2	13.2	13.6	12.2
Age 40–44	15.6	14.0	13.7	13.9	12.7	13.6	13.2	13.6	12.7	12.4
Age 45–49	12.1	11.3	11.0	11.5	11.6	10.6	10.4	10.5	10.5	9.1
Age 50–54	12.8	11.9	13.1	12.8	13.3	13.3	11.5	11.6	11.4	11.9
Age 55–59	7.9	8.9	10.7	10.0	10.3	10.3	11.4	10.6	11.8	12.1
Age 60–64	7.2	7.2	8.0	8.4	9.7	9.0	10.5	10.4	12.3	10.7
Non-Hispanic White	45.0	43.8	39.5	43.2	37.4	43.4	37.2	42.1	37.6	41.9
Non-Hispanic Black	19.9	13.0	22.0	15.6	23.0	14.4	20.7	14.5	22.6	15.3
Non-Hispanic Other	5.1	7.5	5.4	7.6	4.9	8.6	5.0	8.2	4.7	8.2
Hispanic	30.0	35.6	33.1	33.5	34.7	33.6	37.1	35.3	35.1	34.6
Male	45.2	46.0	43.6	44.0	42.4	44.0	43.2	44.1	43.2	43.2
Married	55.1	52.9	44.9	42.8	44.2	42.0	46.5	40.9	42.6	40.7
Childless	36.7	36.7	38.3	39.0	39.6	42.0	39.0	42.8	44.1	44.2
No High School Degree	25.0	27.4	33.6	34.4	33.0	33.8	34.5	33.7	31.8	31.1
High School Degree	38.2	36.8	34.2	31.9	33.7	31.4	33.5	33.1	34.5	32.8
Some College, no Degree	25.1	23.8	24.7	25.9	25.8	25.8	24.1	24.8	25.6	26.8
College Degree	11.8	11.9	7.5	7.8	7.5	9.0	8.0	8.4	8.1	9.3
N	13313	17112	9519	20337	10147	16925	11131	14704	9685	15094

Table 1 reports the weighted rates (%) for all outcome variables (binary measures of tooth loss) and (binary) control variables. Expansion group include all states which expanded Medicaid under the Affordable Care Act between 2014–2018 and cover non-emergency Dental benefits under Medicaid. Non-Expansion group includes all states which did not expand Medicaid under the Affordable Care Act during the study period.

Figure 1: Weighted Mean Tooth Loss Rates, by Expansion and Non-Expansion States.

Figure 1 shows the year-by-year, weighted mean tooth loss rates for states which expanded Medicaid dental coverage (and states which did not expand Medicaid. Expansion states include states which expanded in 2014, as well as states who expanded after 2014 but before 2018.

Association between Medicaid Expansion and Tooth Loss

Table 2 reports estimates of the associations of expanding Medicaid with dental benefits (above emergency-only services) and the probability of each tooth loss category. There was no association with zero tooth loss (versus any loss); the estimate was very small and statistically non-significant. Medicaid expansion was associated with a marginally significant decline in the likelihood of having lost ≤5 teeth by 1.1 percentage-points. This estimate represents a 2.9% decline relative to the pre-expansion rate of reporting five or fewer lost teeth. In contrast, there was no association with the likelihood of having lost six or more but not all teeth; estimate was very small and statistically non-significant. Finally, the expansion was associated with a statistically significant (p<0.05) increase in likelihood of total tooth loss by 1.1 percentage-points, or by 20.6% increase from the pre-expansion rate. Estimates were overall similar using multinomial logit instead of a linear probability model (Supplemental Table 2). Also, estimates were similar when using a classical difference-in-differences regression including only treatment states that expanded in 2014; estimates were slightly larger in magnitude but not meaningfully different when excluding year 2014 since it was the first year (Supplemental Table 2).

Table 2 –

Effect Estimates of Medicaid Dental Expansion on Likelihood of Tooth Loss

Outcome	Effect	Mean
P(Zero Tooth Loss)	−0.0001 (0.008)	0.427
P(1–5 Teeth Lost)	−0.0111* (0.006)	0.385
P(>=6 Teeth Lost, but not all)	0.0007 (0.004)	0.137
P(Total Tooth Loss)	0.0105*** (0.003)	0.051

Table 2 shows the estimates from model 1 of effect of expanding Medicaid dental coverage on each probability of tooth loss category. Robust standard errors are clustered at the state level. Each outcome is measured as a binary 0/1 variable, and so the effect estimates represent a change in the outcome probability on a 0–1 scale. The proportion of the sample with each outcome before the expansion is shown in column "Pre-Expansion Outcome Mean" as a reference. The Pre-Expansion outcome mean was calculated using years 2010 and 2012.

^*p < 0.1

^**p < 0.05

^***p < 0.01.

When considering alternative sample income ranges, we found similar estimates if we included adults below 100% FPL or 200% FPL. In contrast, estimates were smaller if the sample was limited to adults below 400% FPL, and there was no association for adults over 400% FPL (Supplemental Table 3). Furthermore, there were no significant differences in pre-expansion trends in tooth loss among states that expanded in 2014 (combining extensive and limited benefits) and non-expansion states (Supplemental Tables 4 and 5), lending support to the difference-in-differences estimates. Supplemental exhibit 2, and supplemental tables 2–6, provide more details on the alternative specification results and checks of pre-expansion trends.

Associations by Level of Dental Benefits

Table 3 reports estimates of the Medicaid expansion associations with tooth loss separately for extensive and limited dental benefits. We found a similar pattern of results for extensive benefits as the overall expansion, with a larger increase in the probability of total tooth by 1.4 percentage-points (p<0.01) or by 40% relative to pre-expansion rate. Similarly, there was a marginally significant decline in the likelihood of losing ≤5 teeth by 1.3 percentage-points or by 3.2% relative to pre-expansion rate of this outcome. There was, however, a significant differential pre-expansion trend that could bias this estimate, so it should be interpreted with caution. On the other hand, all estimates of Medicaid expansions for states that offer limited dental benefits were smaller than those with extensive benefits and statistically non-significant; the model for losing ≤5 teeth also had a significant pre-trend.

Table 3 –

Effect Estimates of Medicaid Dental Expansion on Likelihood of Tooth Loss by Level of Coverage

	Extensive Benefits		Limited Benefits
		Pre-Expansion Outcome		Pre-Expansion Outcome Mean
Outcome	Effect	Mean	Effect	Mean
P(Zero Tooth Loss)	−0.0028 (0.010)	0.447	0.0018 (0.011)	0.396
P(1–5 Teeth Lost)	−0.0127* (0.007)	0.396&	−0.0081 (0.007)	0.368#
P(>=6 Teeth Lost, but not all)	0.0015 (0.004)	0.121	0.0004 (0.006)	0.162
P(Total Tooth Loss)	0.0140*** (0.003)	0.035	0.0060 (0.004)	0.074

Table 3 shows the estimates from model 2 of effect of expanding extensive Medicaid dental coverage and limited Medicaid dental coverage on each probability of tooth loss category. Extensive and limited benefits were estimated by separate regressions. Robust standard errors are clustered at the state level. Each outcome is measured as a binary 0/1 variable, and so the effect estimates a represent change in the outcome probability on a 0–1 scale. The proportion of the sample with each outcome before the expansion is shown in column "Pre-Expansion Outcome Mean" as a reference. The Pre-Expansion outcome mean was calculated using years 2010 and 2012.

^*p < 0.1

^**p < 0.05

^***p < 0.01

^&indicates pre-treatment differential trend test statistic p < 0.1

^#indicates pre-treatment differential trend test statistic p < 0.01

Estimates by Age Groups

Table 4 reports the estimates of the associations of Medicaid expansions with tooth loss from models 1 and 2 separately for age subgroups. There was an increase in likelihood of total tooth loss among ages 35–44 and 55–64. For ages 55–64, there was an increase in total tooth loss with both extensive and limited dental benefits, and a decline in likelihood of no tooth loss with extensive benefits. For ages 45–54, there was a marginally significant increase in losing 6 or more but not all teeth with limited benefits.

Table 4 –

Effect Estimates of Medicaid Dental Expansion on Tooth Loss, by age groups

Age Group	Overall Expansion Effect	Extensive Coverage	Limited Coverage
Age 25–34
P(Zero Tooth Loss)	0.0101 (0.018)	0.0156 (0.018)	−0.0039 (0.021)
P(1–5 Teeth Lost)	−0.0203 (0.015)	−0.0215 (0.016)	−0.0095 (0.016)
P(>=6 Teeth Lost, but not all)	0.0059 (0.006)	0.0044 (0.005)	0.0064 (0.012)
P(Total Tooth Loss)	0.0043 (0.004)	0.0015 (0.003)	0.007 (0.006)
Age 35–44
P(Zero Tooth Loss)	−0.0043 (0.017)	−0.0083 (0.023)	−0.0038 (0.022)
P(1–5 Teeth Lost)	−0.0059 (0.016)	−0.0215 (0.016)	0.0149 (0.017)
P(>=6 Teeth Lost, but not all)	0.0020 (0.008)	0.0137 (0.007)	−0.0091 (0.009)
P(Total Tooth Loss)	0.0082 (0.004)	0.0160 (0.005)	−0.002 (0.005)
Age 45–54
P(Zero Tooth Loss)	−0.0079 (0.014)	−0.0177 (0.016)	0.0075 (0.009)
P(1–5 Teeth Lost)	−0.0127 (0.014)	−0.0093 (0.016)	−0.0249 (0.016)
P(>=6 Teeth Lost, but not all)	0.0251* (0.013)	0.0240 (0.017)	0.0314* (0.017)
P(Total Tooth Loss)	−0.0045& (0.007)	0.0029# (0.008)	−0.014 (0.011)
Age 55–64
P(Zero Tooth Loss)	−0.0015 (0.011)	−0.0179** (0.008)	0.0194 (0.012)
P(1–5 Teeth Lost)	0.0006 (0.020)	0.0184 (0.019)	−0.0241 (0.020)
P(>=6 Teeth Lost, but not all)	−0.0239 (0.015)	−0.0217 (0.018)	−0.0288 (0.018)
P(Total Tooth Loss)	0.0248***	0.0212**	0.0334***

Table 3 shows the estimates from models 1 and 2 of effect of expanding Medicaid dental coverage on each probability of tooth loss category, by ten-year age groups. The overall effect, extensive coverage, and limited coverage were estimated by separate regressions. Robust standard errors are clustered at the state level. Sample size ranges from 29,500 to 60,963. Each outcome is measured as a binary 0/1 variable, and so the effect estimates represent a change in the outcome probability on a 0–1 scale.

^*p < 0.1

^**p < 0.05

^***p < 0.01

^&indicates pre-treatment differential trend test statistic p < 0.1

^#indicates pre-treatment differential trend test statistic p < 0.01

Discussion

This study provides further evidence that Medicaid expansions with dental benefits are associated with increased access to care for previously unmet oral health needs among low-income adults. Medicaid dental coverage expansion was associated with greater likelihood of total tooth loss, reflecting greater access to dental extractions and previously untreated dental problems leading to tooth loss. This association was mainly observed for extensive dental benefits, except for adults 55–64, an age group with the largest estimated association between Medicaid dental expansion with either extensive or limited benefits and total tooth loss. Building off previous evidence showing how low-income adults increased dental service visits following Medicaid dental coverage expansions, our findings similarly suggest that the oral health needs of previously uninsured low-income adults were largely unmet prior to expansion. Prior studies found that dental visits increased after the Medicaid expansions with dental benefits, with estimates ranging from 4 to 12 percentage point increase in likelihood of visits.^10,11,14 Prior research investigating dental treatment use also reported that the Medicaid expansion was associated with increasing likelihood of major dental treatments by 4 percentage points and dental caries treatment by 9 percentage points.^11,17 Our finding that the Medicaid expansion was associated with a 1.1-percentage-point increase in total tooth loss is overall consistent with those prior studies and indicates that low-income adults responded to new access to dental coverage by addressing previously unmet dental needs.

With new dental coverage, adults enrolled in Medicaid programs that offered extensive dental benefits (covering more services and higher annual spending caps per beneficiary) may face fewer financial barriers to accessing dental services, increasing the likelihood of receiving care for previously untreated tooth decay and periodontal disease. This result suggests that low-income adults face major barriers and delays in getting necessary dental treatments without Medicaid dental coverage, which ultimately lead to tooth loss. Low-income adults in non-expansion states and expansion states which do not offer extensive dental benefits are a particularly vulnerable group for these risks.

That the increase in total tooth loss was most prominent for individuals aged 55–64 further indicates that the increase in total tooth loss following the Medicaid expansions reflects previously unmet dental care needs. Simply by living longer without adequate dental coverage and access, older low-income adults were more likely to have more untreated teeth in need of removal. At older age, low-income adults continue to face barriers in accessing dental care especially if they are not eligible for Medicaid. Traditional Medicare, which covers the majority of adults 65 and older, does not offer dental benefits. Medicare Advantage plans may offer some dental benefits although coverage may not be comprehensive or accessible to all Medicare beneficiaries.^31,32 Individuals below 100% FPL generally continue to be eligible for Medicaid. However, a recent study found no associations between state Medicaid dental benefits and tooth loss for low-income adults 65 years or older using BRFSS data from 2014 through 2018.²¹ One possible explanation is that changes in tooth loss occurred before age 65 as suggested by our study.

Our study makes several contributions to the literature on Medicaid dental benefits and oral health. This study is among the first to examine the associations between recent Medicaid expansions involving different levels of dental benefits under the ACA and tooth loss of non-elderly low-income adults. Moreover, we examine the severity of tooth loss based on the count of teeth removed and not just total tooth loss. By examining different ranges of tooth loss, the study sheds light on associations of different levels of previously unmet oral health needs and new access to dental care. For some, new access is associated with more teeth removed due to high previously unmet needs. For others, access might prevent the need for further tooth loss. Finally, we evaluate differences by age, providing a more comprehensive understanding of the oral health needs of adults eligible for Medicaid. As more states expand Medicaid eligibility and dental benefits, more work is needed to understand the effects of Medicaid dental coverage expansions on oral health equity. Such future research could examine how new access to dental coverage impacts unmet dental needs across demographic and geographic factors. Further, as younger adults gain dental coverage, understanding how long-term access to coverage impacts periodontitis, cavities, and tooth loss prevention remains important.

One limitation of the study was lack of data on changes in oral health over time for the same individuals. Without such data, we could not definitively assess how individuals transition between the different categories of tooth loss over time. Instead, we evaluated changes in the probabilities of each category to infer this flow. The middle categories (5 or less and 6 or more but not all teeth lost) may have opposite changes in transition that could offset each other. However, this limitation had little effect on our main inference on the likelihood of total tooth loss and the likelihood of zero tooth loss.

The analysis was also subject to constraints from the available BRFSS data which only surveys adults on tooth loss during even years. However, we don’t expect the odd years missing from our sample to limit the generalizability of our findings or introduce bias to our estimates. It is unlikely that the unobserved tooth loss outcomes would differ systematically between expanding and non-expanding states in these missing years. An Additional limitation of the BRFSS data is measuring income in categories, which we use to select study sample. To check limitation, we examined different income ranges and observed an expected pattern of changes in estimates. Also worth noting was the apparent presence of secular trends affecting tooth loss patterns in both expansion and non-expansion states (Figure 1). While our empirical strategy accounts for such secular trends, the question may remain as to the reason for these trends. Whether these trends were indicative of improving oral health or perpetuating unmet dental care needs is left for future research. Finally, all BRFSS data including tooth loss were self-reported by participants. Self-reported data including number of lost teeth might include recall bias and error. We do not, however, expect recall bias and errors to differ between respondents in expansion and non-expansion states.

Conclusion

Medicaid dental coverage expansions revealed large previously unmet oral health needs among low-income adults, reflected in an increase in total tooth loss likely due to needed extractions. This result was concentrated in states offering extensive dental benefits and for adults aged 55–64. Consistent with prior studies, these findings indicate that low-income adults have gained access to needed dental care following the ACA Medicaid expansions that offer generous dental benefits.