Pediatric dental care utilization and parent/caregiver-rated oral health among Medicaid enrollees in Alabama

Abstract

Background:

Reducing tooth decay and improving access to dental care remains a public health challenge. Understanding low utilization of care remains a critical gap. This study estimated parent/caregiver-reported prevalence and identified factors associated with children’s dental care utilization, including the association with children’s oral health.

Methods:

A cross-sectional analysis of Medicaid-enrolled children in Alabama using data from the 2017 statewide Consumer Assessment of Healthcare and Providers and Systems (CAHPS^®) Health Plan Survey. Associations were measured using adjusted prevalence ratios (aPR) and 95% confidence intervals (95% CI) from logit regression and generalized linear model postestimation of least squares means.

Results:

The 6-month prevalence of children receiving dental care was 70.4%. Children aged 0-3 years (aPR, 0.72; 95% CI, 0.53 to 0.91) had lower prevalence of care compared to other age groups. The prevalence of low-rated oral health was 9.2%. Low-rated oral health was associated with not receiving dental care (aPR, 1.50; 95% CI, 1.12 to 1.87) and parental education of 8^th grade or less (aPR 2.59; 95% CI 1.20 to 3.98). Falsification tests determined dental care utilization was not associated with ratings for overall health (aPR, 1.18; 95% CI, 0.83 to 1.52) or emotional health (aPR, 1.06; 95% CI, 0.79 to 1.33).

Conclusions:

We observed children not receiving dental care had low-rated oral health; however, as a cross-sectional study, we were unable to assess the temporality of this relationship.

Practical Implications:

Dental care providers should continue to recognize their role in educating parents and providing anticipatory guidance on children’s oral health.

Introduction

Dental caries remain a common childhood health concern associated with a multitude of adverse effects on the quality of life, including impaired physical growth, and diminished academic performance.^1–5 Furthermore, racial/ethnic and socioeconomic disparities exist in access to care, contributing to the burden of disease.^6,7 Although the prevalence of tooth decay has decreased in the last 10 years, vulnerable populations of children still face challenges accessing dental care and consequently have poor oral health outcomes.⁸ Data from the National Health and Nutrition Examination Survey (NHANES) indicate 36% of children under 8 years of age have caries in their primary teeth, with higher prevalence in children from low income families.⁸ This has implications into adolescence and adulthood, as tooth decay in primary teeth is a known predictor of caries in permanent teeth.⁹ Thus, reducing the prevalence of tooth decay for children remains a public health challenge. Compounding the problem is the limited information regarding how parents’ or caregivers’ beliefs and behaviors affect utilization of dental care. For example, children have higher rates of caries if their parents fear the dentist or do not regularly utilize care.^10,11

Essential dental services for low-income children are mandated under the Early and Periodic Screening, Diagnosis, and Treatment (EPDST) Program, which has improved access to care and led to better prevention, early identification, and management among Medicaid enrollees.^12,13 Other efforts, including community fluoridation, school-based dental clinics, and expanded use of fluoride varnishes and sealants have also led to improvements in oral health.^14–17 However, despite having dental coverage through Medicaid’s EPDST program, a high proportion of children in low-income families have tooth decay and untreated caries and report notable gaps in access.^18,19 Gaining a better understanding of why children do not receive care even when they have coverage, and whether parental/caregiver attitudes play a role therein, remains a critical gap in the ability to improve care delivery and reduce tooth decay.

In this paper, we analyzed data from the Consumer Assessment of Healthcare Providers and Systems (CAHPS^®) annual Health Plan Survey from Alabama for 2017. The scope of the survey, along with the novelty of questions asked within this population provide a unique opportunity to gain insights into dental care utilization and oral health among a vulnerable population. As such, this is the first study to utilize data collected from CAHPS to assess dental care or oral health representative of an entire state’s pediatric Medicaid population. Our objectives for this analysis were to 1) estimate the prevalence of children with dental care utilization; 2) identify factors associated with receipt of dental care; and 3) identify how parent/caregiver-rating of child’s oral health is associated with receipt of dental care.

Methods

Data source

We analyzed data collected using the child-specific CAHPS^® Health Plan Survey for Medicaid version 5.0 instrument within the state of Alabama during calendar year 2017. The Agency for Healthcare Research and Quality (AHRQ) uses CAHPS^® survey data to advance scientific understanding of patient experiences with health care.²⁰ Among the multitude of CAHPS^® surveys is the Health Plan Survey, which is the standard tool for measuring patient experiences with health plans, including commercial and Medicaid plans. Annual administration of the survey is required for National Committee for Quality Assurance (NCQA) accreditation of a health plan.²¹ In 2016, the Alabama Medicaid Agency expanded the scope of data collection for the CAHPS^® Health Plan Survey to generate state- and regionally-representative samples of Medicaid enrollees by each age-based category of aid for internal quality improvement efforts. The CAHPS Health Plan Survey must be administered without alteration; however, approved supplemental questions may be included. In 2017, supplemental questions related to use of dental services and parent/caregiver-rated oral health were included. The survey was administered using a mixed-methodology approach—postal mailed questionnaires and telephone-based follow-up calls by an NCQA-certified survey vendor to ensure rigorous processes are in place for designing, testing, collecting, analyzing, and reporting survey measures according to specific methodologies outlined by the Healthcare Effectiveness Data and Information Set (HEDIS).²² The institutional review board and Alabama Medicaid Agency approved of the survey and supplemental questions; this analysis was based on de-identified data.

Sampling methodology

To ensure consistency across health plans and states, NCQA-certified vendors must follow pre-specified sampling methodologies. In order to generate representative samples for important health care delivery regions within the state, by age and selected categories of aid, a stratified random sample design was used. The sample frame included all households with current Medicaid-enrolled children. For households with multiple enrollees, one enrollee was randomly selected during each 6-month period according to NCQA/HEDIS methodology. Strata were formed based on children aged 0-12 years and 13-17 years in 6 regions, for a total of 12 strata. Samples were randomly drawn each quarter of calendar year 2017 to achieve at least 400 responses for each stratum. Over the course of the year, the parents or caregivers of 27,318 Medicaid-enrolled children were contacted for participation. Per NCQA/HEDIS methodology, parent/caregiver respondents could refuse or answer “don’t know” to any questions. If parent/caregivers terminated the survey early, a determination was made on whether the questionnaire responses were complete, partially completed, or incomplete based on AHRQ documentation.²³ Specifically, completed questionnaires were required to have at least half of the 20 key survey items and at least one reportable item, partially completed questionnaires must have a least one reportable item but less than half of the key items, while incomplete questionnaires lacked at least one reportable item. The full survey documentation and questionnaire is available for download (https://www.ahrq.gov/cahps/surveys-guidance/hp/index.html). In total, 5,032 questionnaires were deemed complete or partially completed for an overall response rate of 18.3%.

Outcome measures

The primary outcome measures of interest were derived from responses to the approved CAHPS^® supplemental questions on dental care and oral health. The supplemental questions were structured similarly to standard CAHPS^® survey questions that asked about health service utilization and health rating over a 6-month recall window. Regarding dental service utilization, the question was asked, “In the last 6 months, did your child get care from a dentist’s office or dental clinic?” Parent/caregivers were asked to rate oral health based on the question, “How would you describe the overall condition of your child’s teeth and gums?” Response options included “excellent,” “very good,” “good,” “fair,” and “poor.” We dichotomized the outcome with “fair” and “poor” ratings being the primary outcome of interest relative to any other rating (i.e., “excellent,” “very good,” or “good”). Hereafter, we refer to the category represented by “fair” or “poor” ratings as low-rated oral health compared to the reference group of high-rated oral health, which is the other three categories.

Independent variables

We investigated the association between outcomes and several child demographic and health service factors based on standard CAHPS^® questions. Demographics included sex, age, race/ethnicity, and parent education level. We also examined whether the child was reported to have a primary care provider (PCP).

Statistical Analysis

Our analyses excluded all observations with missing responses (i.e., “don’t know” or refuse) to dependent or independent variables of interest, except for parent/caregiver-rated teeth. For this measure, we considered an analysis comparing “excellent” or “very good” rated teeth to “unable to rate.” In total, 193 observations were excluded due to “don’t know” or refuse responses to variables of interest. Analyses were weighted using the total number of enrollees in each sample strata divided by the number of stratum-specific responses to represent 493,770 enrollees. All analyses were conducted with SAS 9.4 and were weighted based on the survey’s sampling design to produce estimates representative of the population of Alabama’s Medicaid-enrolled children.

We calculated univariate frequencies to describe the prevalence of each characteristic we investigated. We conducted bivariate analyses to obtain the unadjusted prevalence of the outcome for each independent variable. We conducted weighted multivariable logit regressions to assess the relationship between each outcome measure and the association with each independent variable. Due to the high prevalence of the outcomes of interest (in particular, receiving dental care), the “rare event” assumption was violated. Thus, we intentionally avoided reporting odds ratios which would overestimate the true prevalence ratio.^24,25 Instead, we first estimated a weighted logit regression accounting for the survey design, then used generalized linear model postestimation of the least squares means to derive prevalence ratios and adjusted prevalence ratios.²⁶ We report results as adjusted prevalence ratios (aPRs) and 95% confidence intervals (CIs) comparing the association of each independent variable on the relative prevalence of each outcome measure, controlling for all other independent variables. As a robustness check, we compared the estimated prevalence ratios with odds ratios. For low-rated oral health, a relatively rare occurrence, odds ratios were consistent with prevalence ratios.

Falsification Tests

To test whether the ability to rate oral health was dependent on the receipt of dental care, we separately analyzed parent’s/caregiver’s inability to rate children’s oral health as an outcome. We dichotomized responses into “don’t know” relative to all other ratings and estimated the same model specifications as our primary analyses. To test the specificity of the association of receiving dental care on parent/caregiver oral health rating, falsification tests were conducted with outcomes that were unlikely to be directly influenced by receipt of dental care within the prior 6 months. Specially, the questions “In general, how would you rate your child’s overall health?” and “In general, how would you rate your child’s overall mental or emotional health?” Overall child’s health rating and emotional health rating were dichotomized and analyzed using the same model specifications (i.e., low-rated vs high-rated).

Results

Characteristics of Medicaid-enrolled children with responses to the survey are shown in Table 1, along with both weighted and unweighted frequencies, and the prevalence of each characteristic based on the weighted analysis. Approximately 70% of children enrolled were reported by their parents/caregivers to have had a dental care visit within the past 6 months and 9% had teeth or gums rated “fair” or “poor” by their responding parent/caregiver. Forty-one percent were white, 38.1% black/African American and 12.2% Hispanic/Latino. Approximately 90% were reported to have a PCP, and approximately 7% were reported in “fair” or “poor” general overall health.

Table 1.

Characteristics of Medicaid-enrolled children with responses, their use of preventive dental care, and parent/caregiver-rated dental health.

	Nunweighted	Nweighted	Prevalence
All	4,839	493,770	100
Male	2,573	264,120	53.5
Female	2,266	229,650	46.5
Age
0-3 years	432	71,656	14.5
4-7 years	776	129,332	26.2
8-11 years	792	128,245	26.2
12-15 years	1,559	103,713	21.0
16-19 years	1,280	60,824	12.3
Race/Ethnicity
White	2,017	202,536	41.0
Black/Afr. Amer	1,871	188,182	38.1
Hispanic/Latino	473	60,347	12.2
Multiple race/ethnicities	359	33,477	6.8
Asian	37	3,505	0.7
American Indian	40	1,186	0.2
Other	41	4,515	0.9
Parent Education
8th grade or less	254	28,413	5.8
Some high school	681	66,756	13.5
High school graduate / GED	1,699	178,480	36.1
Some college or 2-year degree	1,612	169,701	34.4
4-year college graduate	347	32,173	6.5
More than 4-year college degree	246	18,248	3.7
Primary care provider (PCP)
Has a PCP	4,333	443,665	89.9
No PCP	480	47,844	9.7
DK/Ref	26	2,260	0.5
Child’s Health
Excellent	1,463	168,355	34.1
Very good	1,672	172,984	35.0
Good	1,246	114,211	23.1
Fair	383	32,754	6.6
Poor	58	3,956	0.8
Refused/don’t know	17	1,510	0.3
Dental care in last 6 months
Yes	3,283	347,380	70.4
No	1,207	118,217	23.9
Refused/Don’t Know	349	28,172	5.7
Caregiver-rated dental health
Excellent	1,099	124,770	25.3
Very Good	1,587	165,411	33.5
Good	1,339	130,908	26.5
Fair	389	37,341	7.6
Poor	61	5,312	1.1
Refused/Don’t Know	364	30,027	6.1

Table 2 displays characteristics of children who were reported to have received dental care within the past 6 months, including weighted frequencies, prevalence, and unadjusted and adjusted prevalence ratios. Overall, 70.4% of children received dental care. Among all characteristics, children aged 0-3 years had the lowest prevalence of dental care (46.4%), whereas children aged 8-11 years had the highest prevalence (81.3%). Compared to the oldest group of children covered by Medicaid, 16-19 years, children aged 0-3 had a lower prevalence of receiving dental care (aPR, 0.83; 95% CI, 0.75 to 0.91) whereas all other age groups had higher prevalence of receiving dental care; including 4-7 years (aPR, 1.18; 95% CI, 1.01 to 1.34), 8-11 years (aPR, 1.36; 95% CI, 1.05 to 1.67), and 12-15 years (aPR, 1.16; 95% CI, 1.02 to 1.31). Compared to white children, black or African American children had a higher prevalence of receiving dental care (aPR, 1.11; 95% CI, 1.00 to 1.22).

Table 2.

Characteristics of Medicaid-enrolled children with dental care in prior 6 months.

	Weighted Frequency	Prevalence	Prevalence Ratio (95% CI)	Adjusted Prevalence Ratio (95% CI)
Received dental care	347,380	70.4	--	--
Male	184,640	69.9	0.98 (0.94 to 1.02)	0.96 (0.89 to 1.02)
Female	162,740	70.9	Reference	Reference
Age
0-3 years	33,225	46.4	0.83 (0.75 to 0.90)	0.72 (0.53 to 0.91)
4-7 years	97,763	75.6	1.11 (1.06 to 1.15)	1.18 (1.01 to 1.34)
8-11 years	104,209	81.3	1.21 (1.16 to 1.26)	1.36 (1.05 to 1.67)
12-15 years	74,030	71.4	1.10 (1.05 to 1.14)	1.16 (1.02 to 1.31)
16-19 years	38,153	62.7	Reference	Reference
Race/Ethnicity
White	137,566	67.9	Reference	Reference
Black/Afr. Amer	138,219	73.4	1.07 (1.02 to 1.12)	1.11 (1.00 to 1.22)
Hispanic/Latino	41,397	68.6	1.04 (0.94 to 1.11)	1.15 (0.98 to 1.31)
Multi. race/ethn.	24,252	72.4	1.03 (0.94 to 1.11)	1.03 (0.90 to 1.16)
Other	5,946	64.4	0.93 (0.78 to 1.07)	0.88 (0.68 to 1.09)
Parent Education
8th grade or less	16,876	59.4	0.79 (0.56 to 1.02)	0.75 (0.49 to 1.02)
Some HS	46,677	69.9	0.90 (0.71 to 1.09)	0.90 (0.69 to 1.10)
HS graduate / GED	125,070	70.1	0.91 (0.72 to 1.09)	0.90 (0.71 to 1.10)
Some college/2-yr deg.	121,037	71.3	0.90 (0.72 to 1.09)	0.90 (0.70 to 1.10)
4-yr college degree	24,216	75.3	1.04 (0.85 to 1.23)	1.03 (0.82 to 1.24)
>4-yr college degree	13505	74.0	Reference	Reference
Primary care provider (PCP)
Has a PCP	312,655	70.5	Reference	Reference
No PCP	32,629	68.2	0.97 (0.90 to 1.04)	0.92 (0.82 to 1.24)
Caregiver-rated overall health
Excellent	115,373	68.5	Reference	Reference
Very Good	126,755	73.3	1.06 (1.01 to 1.12)	1.06 (0.98 to 1.15)
Good	78,771	69.0	1.00 (0.95 to 1.05)	0.96 (0.86 to 1.05)
Fair	22,791	69.6	1.01 (0.93 to 1.08)	0.94 (0.81 to 1.08)
Poor	2,574	65.1	0.96 (0.77 to 1.14)	0.90 (0.58 to 1.21)
Refused/Don’t Know	1,118	87.1	1.09 (0.83 to 1.36)	1.06 (0.63 to 1.49)

Prevalence is the percentage of children with each characteristic with a parent/caregiver-reported dentist visit within the last 6 months, divided by all children with that characteristic. Prevalence ratios derived from logistic regression estimation accounting for survey design, followed by generalized linear model postestimation of least squares means. Adjusted prevalence ratios control for all other covariates.

Table 3 displays the results of analyses for the outcome of low-rated child’s oral health (low being the combined rated of “fair” or “poor” compared with high rating— “excellent,” “very good,” or “good”). Overall, 8.7% of children had low-rated oral health (the sum of “fair” and “poor” as shown in Table 1), however after excluding children with responses of “don’t know” on oral health rating, the prevalence of low-rated oral health was 9.2%. Among children with no report of dental care in the last 6 months, the prevalence of low-rated oral health was 11.5%. Compared with children that have had at least 1 dental visit, children without any visits had 1.50 times greater prevalence of low-rated oral health (95% CI, 1.12 to 1.87), controlling for all other characteristics. Compared to the reference group of children 16-19 years, children aged 0-3 years had about half the prevalence of low-rated oral health (aPR, 0.50; 95% CI, 0.23 to 0.78). There was no statistical difference between other age-groups. Compared to parent/caregivers with >4 year college degrees, children with parent education of 8^th grade or less had greater prevalence of low-rated oral health (aPR, 2.59; 95% CI, 1.20 to 3.98). There were no statistical differences between white race compared to other racial and ethnic groups, nor any statistical differences between those with and without a PCP.

Table 3.

Characteristics of Medicaid-enrolled children with low parent/caregiver-rated teeth and gum health relative to all other ratings.

	Weighted Frequency	Prevalence	Prevalence Ratio (95% CI)	Adjusted Prevalence Ratio (95% CI)
Fair or poor rated teeth	42,653	9.2	--	--
Dental care last 6 months
At least once	28,640	8.3	Reference	Reference
None	13,280	11.5	1.36 (1.05 to 1.66)	1.50 (1.12 to 1.87)
Refused / don’t know	733	23.4	2.85 (0.47 to 5.22)	2.71 (0.30 to 5.12)
Gender
Male	24,979	10.0	1.22 (0.94 to 1.52)	1.14 (0.90 to 1.39)
Female	17,675	8.3	Reference	Reference
Age
0-3 years	2,960	4.5	0.49 (0.22 to 0.75)	0.50 (0.23 to 0.78)
4-7 years	10,501	8.5	0.88 (0.59 to 1.17)	0.95 (0.68 to 1.22)
8-11 years	13,108	10.6	1.15 (0.79 to 1.50)	1.18 (0.86 to 1.50)
12-15 years	10,987	11.5	1.25 (0.89 to 1.60)	1.27 (0.96 to 1.59)
16-19 years	5,097	9.3	Reference	Reference
Race/Ethnicity
White	14,963	7.9	Reference	Reference
Black/Afr. Amer	16,253	9.1	1.17 (0.86 to 1.49)	1.16 (0.89 to 1.43)
Hispanic/Latino	305	14.9	1.95 (1.29 to 2.61)	1.33 (0.88 to 1.78)
Multi. race/ethn.	8,047	9.5	1.22 (0.64 to 1.80)	1.18 (0.68 to 1.69)
Other	3,085	3.4	0.45 (0.01 to 1.10)	0.46 (0.01 to 1.05)
Parent/Caregiver Education
8th grade or less	5,920	23.4	3.25 (1.78 to 4.71)	2.59 (1.20 to 3.98)
Some high school (HS)	8,362	13.4	1.83 (0.90 to 2.76)	1.71 (0.81 to 2.62)
HS graduate / GED	15,269	9.1	1.27 (0.59 to 1.96)	1.24 (0.58 to 1.89)
Some college/2-yr deg.	10,568	6.5	0.92 (0.37 to 1.48)	0.93 (0.39 to 1.47)
4-yr college degree	1,332	4.4	0.67 (0.11 to 1.22)	0.69 (0.13 to 1.24)
>4-yr college degree	1,202	7.0	Reference	Reference
Primary Care Provider (PCP)
Has a PCP	35,904	8.6	Reference	Reference
No PCP	6,453	14.4	1.67 (1.13 to 2.21)	1.35 (0.93 to 1.77)

Prevalence is the percentage of children with each characteristic for whom their parent/caregiver gave a low rating for the health of the child’s teeth and gums, divided by all children with that characteristic. Prevalence ratios derived from logistic regression estimation accounting for survey design, followed by generalized linear model postestimation of least squares means. Adjusted prevalence ratios control for all other covariates.

Table 4 displays the associations of children having no reported dental care visits in the prior 6 months on two dental outcomes of interest: 1) unable/refused to rate oral health versus any rating, 2) low-rated oral health vs all high ratings (same model as shown in Table 3), and two falsification test outcomes, 3) low-rated overall health versus all high ratings, and 4) low-rated emotional health versus all high ratings. While having no dental visits (versus at least one) had a statistically significant association with low-rated oral health, there was no statistically significant association between no dental visits and the other outcomes.

Table 4.

Associations of having no parent/caregiver-reported dental care in prior 6 months with the outcomes of inability to rate dental health, low-rated teeth and gums, low-rated overall health, and low-rated emotional health.

Outcome	Prevalence Ratio of No Dental Care (95% CI)	Adjusted Prevalence Ratio of No Dental Care (95% CI)
Unable to rate dental health (vs giving any rating)	2.54 (1.34 to 3.73)	1.55 (0.51 to 2.59)
Low-rated teeth and gums (vs high rating)	1.36 (1.05 to 1.66)	1.50 (1.12 to 1.87)
Low-rated overall health (vs high rating)	1.09 (0.79 to 1.38)	1.18 (0.83 to 1.52)
Low-rated emotional health (vs high rating)	0.92 (0.72 to 1.12)	1.06 (0.79 to 1.33)

Prevalence ratios derived from logistic regression estimation accounting for survey design, followed by generalized linear model postestimation of least squares means. Adjusted prevalence ratio controls for sex, age, race/ethnicity, parent education, having a primary care provider.

Discussion

In this analysis of supplemental questions from the 2017 CAHPS^® Health Plan Survey of children in Alabama’s Medicaid, we report on three main findings according to our objectives. First, we observed the prevalence of children that received some form of dental care within the prior 6 months was 70.4%. Second, we observed child’s age and race were associated with receipt of dental care. Third, not receiving dental care was associated with higher prevalence of low-rated oral health by parent/caregivers. Furthermore, although a cross-sectional study, the association of not receiving dental care and low-rated health was specific to oral health, as falsification tests of other parent/caregiver health ratings were not statistically significant. This study also highlights the potential for innovative use of CAHPS^® Health Plan Surveys as a novel source of data on oral health status and dental utilization that may be used to triangulate estimates on dental utilization and parent/caregiver perceptions about oral health.

Several sources have reported dental care utilization among various population and subpopulations. Most comparable to our population of Medicaid-enrolled children, parent-reported data from the 2018 National Health Interview Survey (NHIS) estimated among children from families with income below 100% of the Federal Poverty Level (FPL), 62.9% aged 2-4 had a dental visit in the past year, 89.2% of children aged 5-11, and 80.5% of children aged 12-17.²⁷ Likewise, analyses from the National Survey of Children’s Health (NSCH) used parent-reported data to estimate 76.4% of low-income children (<17 years) had a dental visit within the past year.²⁸ State specific data from NSCH also reflect 75.5% of low-income children within Alabama had a dental visit in 2017.²⁹ Another parent-reported source of data, NHANES, estimated 67.2% of low-income children aged 2 to 5 years received a dental visit in the past year.³⁰ However, estimates using administrative data or EPDST data (i.e. CMS form 416) show substantially lower utilization than parent-report. National data from the Medical Expenditures Panel Survey (MEPS) from 2010 to 2012 estimated the prevalence of dental care among low-income children to be 32.9%, and rising to 41.7% in 2015.^31,32 Data from Alabama’s 2017 EPDST indicated 48.4% of Medicaid-enrollees received any dental care.³³ The use of more restrictive methodologies in MEPS and EPSDT reports may reduce the influence of social desirability bias that could occur with parent-reporting a lead to overestimated utilization. In particular, MEPS data is important because it forms the baseline estimates used in tracking Healthy People 2020 objectives aimed at increasing dental care utilization.³¹ We estimated any dental care was received within the past 6 months for 46.4% of children aged 3 years and under, 75.6% of children 4-7 years, 81.3% for children 8-11 years, 71.4% of children 12-15 years, and 62.7% for children 16-19 years. Although higher that MEPS or EPDST, our prevalence estimates of care are consistent with other parent-reported data. Notably, to be consistent with CAHPS^® methodology, a 6-month recall window was used to minimize recall bias. While other published estimates and measures vary, our estimated prevalence of dental care via parent/caregiver self-reporting falls within the range of estimates provided from these major national surveys. However, given the difference in time frames (i.e. 1 year versus 6 months), we cannot exclude the possibility that parents overestimate the utilization of dental care due to recall or social desirability bias.³⁴ Furthermore, we are unable to assess whether parent/caregivers reported utilization differs from administrative records.

The American Dental Association and the American Academy of Pediatric Dentistry recommend children establish a dental home within 6 months of their first tooth eruption or by 12 months of age.^35,36 However, many parents do not initiate dental care for their children within this recommended time frame. We observed children between the ages of 0-3 years had the greatest prevalence of not receiving dental care within the past 6 months compared to children 16-19 years. Furthermore, we observed a somewhat “U-shaped” association, such that all age groups other than 0-3 were more likely to have dental care than children 16-19 years. Similar findings in other data sources have been noted previously and may be related to lack of awareness of guidelines to establish a dental home by 1 year of age or underlying parental/caregiver beliefs.^37,38

We observed black/African American children had a higher prevalence of receiving dental care compared to white children. Racial disparities have been consistently observed in access and utilization of dental care, with black/African American and Hispanic/Latino children showing lower levels of access and utilization compared to other races.^39–44 Although racial disparities have narrowed,⁴⁵ it is uncommon to observe black/African American children with greater utilization of care. It is possible that within Alabama, black/African American children utilize more dental care compared to white children. Alternatively, the true association may be masked by the lack of our ability to control for urban-rural status resulting in a specious association favoring black/African American children. Specifically, access to dental care in rural areas is known to be lower than in urban areas,^46,47 and urban-rural status may also be correlated with race/ethnicity. Geographic information was unavailable for the current analysis and we were therefore unable to control for this possible confounding factor. Additionally, factors associated with social desirability bias in parent reporting of oral health status are not well characterized, but our observations could also be explained if overreporting or underreporting of dental care utilization is differential by race.

We observed an association between not receiving dental care and low parent/caregiver rating of children’s oral health. This association is less likely to be spurious, given the falsification tests showing not receiving dental care did not have an effect on outcomes such as overall or emotional health ratings. Notably, we observed parent education was associated with low-rated oral health; low education has previously been associated with poor dietary habits and risk of caries.^48,49 Taken collectively, this leads to several possible conjectures. Recent dental care visits may be associated with actual or perceived oral health improvement, for example through prophylaxis measures including cleaning, scaling, sealants, or fluoride varnishes, and may undergird parent/caregiver’s confidence in their child’s oral health. Although our data are limited to a single timepoint, prior studies indicate regular dental care can improve oral health.⁵⁰ Care received from a dentist can provide education and anticipatory guidance to alleviate concerns about oral health status. The observed association could be explained if parents valuing dental care are more engaged in practices to promote better oral health including behaviors and adherence to recommended preventive dental care intervals. However, as a cross-sectional study, we cannot assess the temporality of this relationship. For children who have not received dental care, parents/caregivers may be more uncertain about the actual health of their child’s teeth due to incomplete clinical and professional information. However, findings do not support this hypothesis as receipt of dental care was not associated with parents stating they “don’t know” the rating of their child’s oral health. In this current study, parents who reported their child did not receive dental care within the previous 6 months, were more likely to report their child had lower oral health status. This appears to indicate that parents have concerns about the oral health of their children if they had not recently received dental care. Alternatively, parents’ responses may reflect their own beliefs regarding the effect of dental care on oral health status. Therefore, one hypothesis generated is that parents indeed value dental care. Although this study did not assess barriers, future work is warranted to understand potential barriers to care-seeking including the roles of the availability of dental providers who accept Medicaid, unreliable transportation, work scheduling, or lacking other resources in the context of parental value of dental care.

This study must be interpreted with consideration of its limitations. As a cross-sectional study, we were unable to assess the temporal relationship of characteristics with the outcomes. In particular, it is unclear whether children’s oral health promotes the receipt of dental care or vice versa, and our ability to understand the full complexity of this relationship was limited. Self-reported data have the potential to be affected by recall bias, particularly when asking about events over the past 6 months, as well as misclassification or desirability biases. Despite not being formally validated, supplemental questions were approved by the NCQA, and were consistent with the overall CAHPS^® structure which have been psychometrically validated.⁴⁶ We did not have information on the type of dental care received, whether preventive care only, restorative, or problem-focused. Furthermore, we did not have objective measures of clinical oral health among children. The survey did not ask the respondent’s relationship to child, and thus we were unable to control for caregivers other than the primary parent(s) completing the survey. Further, we did not have information on reasons why dental care was not utilized, and what specific barriers may have prevented utilization.

Conclusions

A sizable proportion of Alabama Medicaid-enrolled children may not receive regular dental care despite the program’s coverage of preventive and restorative dental treatments. Our findings indicate that parents of children who did not recently utilize dental care are more likely to report a low-rating of their children’s oral health status. If this reflects parental values of care or concerns about oral health, more attention should be devoted to understand care-seeking behaviors in the context of barriers to care. Further research, especially longitudinal designs, is warranted to better identify barriers as well as more fully elucidate how parent/caregiver perceptions and oral health knowledge promote or inhibit timely receipt of dental care.