Effects of Early Dental Office Visits on Dental Caries Experience

Abstract

Objectives. We determined the association between timing of a first dentist office visit before age 5 years and dental disease in kindergarten.

Methods. We used North Carolina Medicaid claims (1999–2006) linked to state oral health surveillance data to compare caries experience for kindergarten students (2005–2006) who had a visit before age 60 months (n = 11 394) to derive overall exposure effects from a zero-inflated negative binomial regression model. We repeated the analysis separately for children who had preventive and tertiary visits.

Results. Children who had a visit at age 37 to 48 and 49 to 60 months had significantly less disease than children with a visit by age 24 months (incidence rate ratio [IRR] = 0.88; 95% confidence interval [CI] = 0.81, 0.95; IRR = 0.75; 95% CI = 0.69, 0.82, respectively). Disease status did not differ between children who had a tertiary visit by age 24 months and other children.

Conclusions. Medicaid-enrolled children in our study followed an urgent care type of utilization, and access to dental care was limited. Children at high risk for dental disease should be given priority for a preventive dental visit before age 3 years.

Early childhood caries, or tooth decay in children younger than 6 years, is the most common chronic disease among children. Its prevalence increased to 28% among 2- to 4-year-old children between 1988 to 1994 and 1999 to 2004, and its impact on children is becoming better known.^1,2 Tooth decay can impair young children’s overall health, speech, growth, and school performance; it can also negatively affect families’ quality of life.^2–6

Because of concern about the impact of dental disease on overall health and the stagnant rate of dental care use over the past decade, Healthy People 2020 identified the annual use of dental care for every person aged 2 years and older as 1 of 24 leading health indicators.⁷ Dental use was selected from as many as 1200 objectives as a high-priority health issue.

Early childhood tooth decay is preventable and largely reversible in its early stages through self-care, use of professional services, and exposure to community interventions such as water fluoridation.^8,9 During dental visits, children can receive an assessment for disease risk, early detection and treatment services, preventive care such as fluoride therapy, and anticipatory guidance.^8,10–12 To ensure exposure to prevention early in life, professional organizations recommend that children have a dental home by 12 months of age.^10,12–14 North Carolina Medicaid, along with Medicaid in most other states, also recommends a first dental visit by age 12 months; however, North Carolina Medicaid does not require a visit until age 3 years because of a limited supply of dentists.

Despite evidence that professional preventive dental care leads to good oral health outcomes, no strong evidence indicates the most effective age for the first visit. A previous study found no difference in the severity of disease at age 6 to 7 years between children whose first dental visit occurred before age 2 years and those who were aged 2 to 5 years at their first visit.¹⁵ Another study found that early dentist visits that included prevention were associated with fewer treatments from the time the children were aged zero to 5 years.¹⁶ Three studies found no relationship between early visits and dental costs or treatment use,^17–19 but one found that children with existing disease who received preventive services earlier had fewer subsequent treatments and expenditures.¹⁹

Previous studies on the age of the first dental visit have primarily focused on treatment and cost outcomes, with conflicting results. It is unclear from studies on treatment outcomes whether children with early visits have differences in disease status. Unlike medical claims, dental claims do not include diagnosis codes, so the extent of disease cannot be determined from claims files alone. We combined claims files with public health surveillance files of oral health status to estimate the effect of the timing of the first dental visit on dental disease history and untreated disease in kindergarten students in North Carolina.

METHODS

The North Carolina Medicaid files contained enrollment information and reimbursement claims from October 1999 to December 2006 for all children born on or after January 1, 1998. These claims files provided detailed information on all reimbursement requests that were submitted by dental and medical providers to Medicaid during this period. The enrollment files provided information about enrollment durations for each child enrolled in Medicaid.

The North Carolina Surveillance of Dental Caries (NCSoDC) contains estimates on dental disease (dental caries experience) for almost all kindergarten students in the state for every school year except 1 since 1996. This information is collected through clinical dental screenings conducted by trained oral health professionals working for the state or local health departments. Surveillance data demonstrated good reliability in a previous study comparing screening results to an experienced, reference dentist who performed a standard dental examination.²⁰

We used surveillance information from the 2005 to 2006 school year, with data for 95 135 kindergarten children from 98 of the state’s 100 counties, or 82% of the state’s public school kindergarten enrollment. We linked the NCSoDC file for this academic year to the Medicaid claims files with Link King version 6.4.9 linking software.²¹ We conducted a study to test the accuracy of the data linkage and found that the North Carolina Medicaid data linked to the NCSoDC with a sensitivity of 0.892 (95% confidence interval [CI] = 0.838, 0.950) and a specificity of 0.882 (95% CI = 0.807, 0.965).²²

We identified children in the Medicaid database who were born between October 1, 1998, and January 1, 2001, who were enrolled before their first birthday, were still enrolled after they turned age 1 year, and did not receive preventive dental services in a medical office (n = 118 564). All children who were in kindergarten in the 2005 to 2006 school year, were recorded in the NCSoDC data set (n = 92 127), and had been enrolled in Medicaid were eligible for the study. The merging of the Medicaid and NCSoDC data sets resulted in a sample of 36 890 children who appeared in both data sets and were thus eligible for our study (match rate = 33%). In light of the percentage of North Carolina children enrolled in Medicaid, we expected the match rate to be between 25% and 35%; previous work showed that the 2 data sets matched with a high level of accuracy.²² We excluded children who did not have a dental visit (24 549) and children who had a missing disease outcome in the surveillance data (n = 947). The final sample therefore comprised 11 394 children, or about 30% of the sample of children who were enrolled in Medicaid and in kindergarten in 2005 to 2006.

Variables

We defined the dependent variable, dental disease status, as a count of decayed, missing (molars only), and filled primary teeth (dmft index) children had during their assessment in kindergarten. Decayed teeth (d) have untreated cavitated lesions, filled teeth (f) have lesions that have been repaired with restorative treatment, and missing teeth (m) have been treated by extraction because of dental caries. The dmft index provides a history of the child’s lifetime caries experience in the primary dentition with the exception of any disease or treatment of anterior teeth missing at the time of the examination. The use of the dmft index from surveillance data allowed us to examine whether teeth were treated for disease, regardless of whether the treatment was included in the Medicaid claims data. Mean age of children at the time of screening was 5.7 years (SD = 0.41 years), which did not differ by age of first visit. Among children who had experienced any disease (dmft > 0), we used a binary dependent variable indicating whether the child had any decayed teeth that were untreated for cavitated lesions to determine the effect of the timing of first visit on untreated disease.

The main explanatory variable of interest was a categorical variable indicating age (in months) at the first visit (< 24, 24–36, 37–48, or 49–60 months). We chose the cutoffs according to current guidelines for the age of the first dental visit. Guidelines state that visits should occur by 1 year, but we had an insufficient sample size to examine visits before age 12 months, so we chose to examine visits before 24 months.

Clinical guidelines recommend early dental visits to prevent and control disease at an early age. Disease prevention can be primary (preventing disease before it starts), secondary (catching disease early and preventing further complications), or tertiary (controlling existing disease and restoring functioning). Dental guidelines for an early preventive dental visit combine components of primary prevention (prophylaxis, fluoride, and counseling to prevent disease initiation) and secondary prevention (examination, fluoride, and improved oral hygiene to prevent disease progression and restoration of carious lesions in their early stages).¹²

Our data incorporated visits with and without prevention and with varying amounts of treatment, which made it difficult to identify the effects of different types of early dental visits. Therefore, we conducted 2 subanalyses on the effect of 2 types of early dental visits on disease status: (1) for children who received a preventive visit in line with clinical guidelines that had elements of both primary and secondary prevention in the form of comprehensive evaluation, fluoride, and small amounts of treatment and (2) for children who received a tertiary visit with 2 or more restorative treatments, so that we could examine the effect of early visits in children at highest risk for future disease. Details of each definition are shown in Table 1.²³

TABLE 1—

Definitions for Preventive Dental Visits Used for Subanalyses in Sample of Kindergarten Children Enrolled in Medicaid Before Age 5 Years: North Carolina, 2005–2006

Definition	Aim of Visit	Included Preventive Services	CDT Codes	Treatment Exclusions
Subanalysis 1	Primary or secondary prevention	Comprehensive evaluation and fluoride application	D0120, D0150 (comprehensive evaluation or periodic exam); D1201, D1203 (fluoride, with and without prophylaxis)	No more than 2 restorative treatment procedures (CDT codes D2000–D2999; restorations, crowns, and other restorative services) on the day of the visit or for any visit for 3 mo (92 d) following the initial visit, and no other treatment (CDT codes D0140; problem focused evaluation), D0415–D0460 (tests and laboratory examinations), D0470–D0501 (pathology laboratory) on the day of the visit or in an office visit for 3 mo (92 d) following the preventive visit, or in a hospital visit for 6 mo (182 d) following the preventive visit
Subanalysis 2	Tertiary prevention	Any preventive service (comprehensive evaluation, prophylaxis, or fluoride application)	D1000–D1999, (prophylaxis, fluoride, other preventive services), D0120, D0150	More than 2 restorative treatment procedures on the day of the visit or within 3 mo at an office visit or in a hospital visit within 6 mo

Note. CDT = current dental terminology

Our analysis incorporated both child-level and county-level variables chosen for the regression model because previous research found them to be associated with the likelihood of having preventive dental visits and dental disease.^16,24,25 The child-level covariates in the model were gender, race (White, Black, Hispanic, or other), number of well-child visits from age 12 to 24 months, and whether the child was continuously enrolled in Medicaid. The analysis also incorporated characteristics of the child’s county of residence: percentage of the population under age 18 enrolled in Medicaid²⁶; metropolitan status, derived from the rural–urban continuum codes available from the US Department of Agriculture Economic Research Center²⁷; and number of dentists per 10 000 population, derived from the North Carolina Health Professions Data System.²⁸

Analyses

We constructed a zero-inflated negative binomial regression model to test whether early visits were associated with subsequent dental caries experience in children. A likelihood ratio test comparing an intercept-only model to incorporation of all covariates in the zero-inflated portion indicated that the model with covariates in the zero-inflated portion was a better fit (χ² with 13 df = 233.7; P < .001), so we used all of the covariates in both the zero-inflated and the negative binomial parts of the model.

We calculated incidence rate ratios (IRRs) for estimating the multiplicative increase in mean dmft index in kindergarten in one subgroup of the overall sampled population relative to another.^29,30 (Methods for calculating these overall exposure effects and their 95% CIs are described in Appendix A, available as a supplement to the online version of this article at http://www.ajph.org.) We also included just the observations for children with some disease (treated or untreated) in a logistic regression model to estimate whether having any untreated disease differed between children who had an early visit and other children.

We analyzed data with SAS version 9.2 (SAS Institute, Cary, NC) and Stata version 12.0 (StataCorp LP, College Station, TX). For all analyses, we considered P < .05 to be statistically significant.

RESULTS

Among all 35 943 Medicaid-enrolled children in the NCSoDC, the average dmft score was 2.18, and 48% experienced dental disease by the time they were in kindergarten (Table 2). Among those with disease, the average dmft was 4.58, and 54% had at least 1 tooth with untreated disease. By their fifth birthday, 32% of children had a dental visit paid by Medicaid. Children who had never had a dental visit paid by Medicaid had a statistically significantly lower dmft score than children who had a dental visit by age 24 months.

TABLE 2—

Disease Status and Descriptive Characteristics of Kindergarten Children Enrolled in Medicaid Before Age 5 Years: North Carolina, 2005–2006

				Age at First Dentist Office Visit
Variable	Overall Sample (n = 35 943), % or Mean (SD)	No Dental Visits (n = 24 549), % or Mean (SD)	With Dental Visits (n = 11 394), % or Mean (SD)	< 24 mo (n = 623), % or Mean (SD)	24–36 mo (n = 2371), % or Mean (SD)	37–48 mo (n = 4358), % or Mean (SD)	49–60 mo (n = 4042), % or Mean (SD)
Dmft	2.18 (3.19)	1.76** (2.86)	2.42 (3.28)	3.49 (4.03)	3.57 (3.92)	3.11 (3.66)	2.73** (3.33)
Disease history (dmft > 0)	47.66	42.28**	59.23	59.39	62.59	59.06	57.42
Dmft for those with disease	4.58 (3.22)	4.17** (3.05)	5.22 (3.36)	5.88 (3.65)	5.70 (3.52)	5.27 (3.37)	4.75 (3.12)
Any with untreated disease among those with disease	53.50	62.89**	39.06	41.35	33.89**	38.85	42.22
Race
White	44.85	47.89*	38.28	42.05	38.72	36.60	39.16
Black	42.22	39.17	48.52	43.98	48.38	51.58**	46.02
Hispanic	9.27	8.62	11.01	11.56	11.39	9.73	12.07
Other	3.66	4.31	2.19	2.41	1.52	2.00	2.75
Female gender	49.22	49.09	49.16	47.51	50.65	49.50	48.17
Metropolitan statusa	57.22	55.07**	62.17	66.29	64.91	61.86	60.27*
Nonmetro 1	36.34	38.18**	32.12	28.09	30.20	31.78	34.24*
Nonmetro 2	6.44	6.75	5.70	5.62	4.89	6.36	5.49
Dentists/capitab	3.73 (1.77)	3.67** (1.71)	3.89 (1.97)	4.11 (2.04)	4.10 (2.01)	3.83** (1.97)	3.79** (1.91)
Well-child visits, no.			1.25 (1.12)	1.68 (1.13)	1.34** (1.11)	1.24** (1.12)	1.15** (1.09)
Medicaid enrollees aged < 18 y in county of residence	35.81	36.00**	35.58	34.76	34.75	35.96*	35.77*
Children continuously enrolled in Medicaid			49.50	64.04	55.08**	51.06**	42.31**

Note. Dmft = decayed, missing, and filled teeth. Differences between children who had a visit by age 24 months and other age categories derived from pairwise mean comparisons with Tukey’s Honestly Significant Difference test.

^aDerived from rural–urban continuum codes from the US Department of Agriculture Economic Research Center.²⁵

^bDentists/10 000 population, derived from the North Carolina Health Professions Data System.²⁶

*P < .05; **P < .01.

Among the sample of 11 394 children enrolled in North Carolina Medicaid who had a dentist visit by age 60 months, the average dmft score was 2.42. Children who had their first visit at age 49 to 60 months had a statistically significantly lower dmft score than children whose first visit was by 24 months. Children who had a visit by age 24 months had a statistically significantly higher number of well-child visits and were more likely than other children to be continuously enrolled in Medicaid (Table 2). Children who had a visit by age 24 months resided in counties with more dentists per capita and a lower proportion of the population enrolled in Medicaid than did children who had their first visit at age 37 to 60 months.

After adjustment for child- and county-level covariates, children who had a dental visit at age 37 to 48 months or 48 to 60 months had a statistically significantly lower dmft score than children who had a visit before age 24 months (IRR = 0.88; 95% CI = 0.81, 0.95; IRR = 0.75; 95% CI = 0.69, 0.82, respectively; Table 3). We observed no statistically significant difference in the dmft score between children who had a visit before age 24 months and between 24 and 36 months (results of full model in Appendix B, available as a supplement to the online version of this article at http://www.ajph.org).

TABLE 3—

Overall Adjusted Incidence Rate Ratios for Dmft Index Among Kindergarten Children Enrolled in Medicaid Before Age 5 Years: North Carolina, 2005–2006

Variable	All Children With a Dental Visit (n = 11 394), IRR (95% CI)	Children With a Primary/Secondary Preventive Dental Visit (n = 6920), IRR (95% CI)	Children With a Tertiary Preventive Dental Visit (n = 4276), IRR (95% CI)
Age at first preventive visit, mo
< 24 (Ref)	1.00	1.00	1.00
24–36	0.98 (0.90, 1.07)	0.86 (0.73, 1.01)	1.14 (0.97, 1.35)
37–48	0.88* (0.81, 0.95)	0.80* (0.68, 0.94)	1.05 (0.90, 1.24)
49–60	0.75* (0.69, 0.82)	0.80* (0.68, 0.93)	0.98 (0.84, 1.15)
Race
White (Ref)	1.00	1.00	1.00
Black	0.85* (0.82, 0.89)	0.88* (0.82, 0.95)	0.83* (0.78, 0.89)
Hispanic	1.25* (1.18, 1.32)	1.58* (1.43, 1.73)	1.21* (1.10, 1.32)
Other	1.23* (1.10, 1.38)	1.02 (0.80, 1.29)	1.39* (1.20, 1.61)
Gender
Female	0.92* (0.88, 0.95)	0.95 (0.89, 1.02)	0.86* (0.81, 0.91)
Male (Ref)	1.00	1.00	1.00
Well-child visits, no.	0.95* (0.93, 0.97)	0.95* (0.92, 0.98)	0.90* (0.84, 0.96)
Continuously enrolled in Medicaid	0.95* (0.92, 0.99)	1.00 (0.93, 1.07)	0.94* (0.91, 0.96)
Enrolled in Medicaid,a %	1.07 (0.82, 1.38)	0.91 (0.56, 1.47)	1.05 (0.70, 1.58)
Metropolitan statusb
Urban (Ref)	1.00	1.00	1.00
Nonmetro 1	1.05* (1.01, 1.10)	1.05 (0.97, 1.13)	1.05 (0.98, 1.13)
Nonmetro 2	1.03 (0.95, 1.12)	0.84 (0.71, 1.01)	1.09 (0.96, 1.22)
Dentists/capitac	0.99 (0.98, 1.01)	0.99 (0.97, 1.01)	1.00 (0.98, 1.01)

Note. CI = confidence interval; dmft = decayed, missing, and filled teeth; IRR = incidence rate ratio. IRRs estimated overall effects according to a zero-inflated negative binomial model with covariates in excess of zero and negative binomial parts of the model, results from both model parts combined.

^aComparison of children from counties with 30% enrolled in Medicaid (reference group) to children from counties with 36% enrolled in Medicaid, which correspond to the lower and upper quartiles of the distribution of percentage of children enrolled in Medicaid.

^bDerived from rural–urban continuum codes from the US Department of Agriculture Economic Research Center.²⁵

^cDentists per 10 000 population in county of residence, derived from the North Carolina Health Professions Data System.²⁶

*P < .05.

Children who had a primary or secondary preventive visit at age 37 to 48 months or 48 to 60 months had a statistically significantly lower dmft score than children who had a visit before age 24 months (IRR = 0.80; 95% CI = 0.68, 0.94; IRR = 0.80; 95% CI = 0.68, 0.93, respectively; Table 3). Age at first visit was not significantly associated with the rate of caries among children who visited a dental office for tertiary prevention.

Forty-one percent of children who experienced dental caries (n = 6749) had at least 1 untreated decayed tooth (Table 2). Children who received a visit at age 24 to 36 months had lower odds of having any untreated disease than children who had their first visit before age 24 months (odds ratio = 0.71; 95% CI = 0.56, 0.90; Table 4). We found no statistically significant difference in the odds of having any untreated disease for children who had their first primary, secondary, or tertiary visit before age 24 months and other children.

TABLE 4—

Adjusted Odds Ratios for Having Any Untreated Dental Disease Among Kindergarten Children With Any Dental Disease Enrolled in Medicaid Before Age 5 Years: North Carolina, 2005–2006

Variable	All Children With a Dental Visit (n = 6749), OR (95% CI)	Children With a Primary/Secondary Preventive Dental Visit (n = 3524), OR (95% CI)	Children With a Tertiary Preventive Dental Visit (n = 3095), OR (95% CI)
Age at first preventive visit, mo
< 24 (Ref)	1.00	1.00	1.00
24–36	0.71** (0.56, 0.90)	0.88 (0.61, 1.27)	0.75 (0.50, 1.13)
37–48	0.82 (0.66, 1.03)	1.07 (0.75, 1.52)	0.82 (0.56, 1.22)
49–60	0.97 (0.77, 1.22)	1.11 (0.79, 1.58)	0.73 (0.50, 1.08)
Race
White (Ref)	1.00	1.00	1.00
Black	1.22** (1.09, 1.37)	1.13 (0.96, 1.32)	1.25** (1.06, 1.49)
Hispanic	0.68** (0.58, 0.81)	0.58** (0.46, 0.74)	0.80 (0.62, 1.03)
Other	0.94 (0.68, 1.29)	1.35 (0.81, 2.24)	0.78 (0.51, 1.21)
Gender
Female	0.99 (0.89, 1.09)	1.00 (0.87, 1.15)	0.98 (0.85, 1.14)
Male (Ref)	1.00	1.00	1.00
Well-child visits, no.	1.02 (0.97, 1.07)	1.03 (0.96, 1.10)	1.00 (0.93, 1.07)
Continuously enrolled in Medicaid	0.86** (0.78, 0.96)	0.70** (0.60, 0.81)	1.05 (0.90, 1.23)
Enrolled in Medicaid, %	10.57** (5.16, 21.65)	15.58** (5.64, 43.04)	6.72** (2.34, 19.30)
Metropolitan statusa
Urban (Ref)	1.00	1.00	1.00
Nonmetro 1	1.11 (0.99, 1.26)	1.27** (1.07, 1.50)	0.95 (0.79, 1.14)
Nonmetro 2	1.56* (1.24, 1.96)	1.73** (1.21, 2.48)	1.39* (0.86, 0.95)
Dentists/capitab	0.92** (0.89, 0.95)	0.92** (0.88, 0.96)	0.91** (0.86, 0.95)

Note. CI = confidence interval; OR = odds ratio. Results are from a logit model adjusted for covariates.

^aDerived from rural–urban continuum codes from the US Department of Agriculture Economic Research Center.²⁵

^bDentists per 10 000 population in county of residence, derived from the North Carolina Health Professions Data System.²⁶

*P < .05; **P < .01.

DISCUSSION

In this North Carolina Medicaid sample, children who had a dental visit by age 24 months had similar numbers of teeth affected with dental caries at the time they were examined in kindergarten as children who had a visit at age 24 to 36 months, but higher rates of disease than those who had a visit at age 37 to 60 months. However, children who had a visit with 2 or more treatments before 24 months had the same rate of disease as children who had visits with 2 or more treatments at older ages.

The utilization patterns of low-income families⁶ likely explain our findings that children with earlier visits had higher rates of disease than children with visits at later ages. Children who needed care because of known disease or pain were probably more likely to see a dentist early. One of the strongest predictors for future disease severity is having disease at a very young age.³¹ In a sample of children younger than 3 years in Boston, children who had an early dental visit were more likely to have disease.³² We found that children who received 2 or more restorative treatments by age 24 months had the same amount of disease as children who had their first visit at older ages. This finding, along with the higher risk associated with having disease at a young age, suggests that early visits may have been effective in suppressing disease for these high-risk children.

Whether early prevention can suppress disease is an important question for state Medicaid agencies seeking to set policies for when and how often children should see the dentist.³³ Professional organizations recommend a preventive dental visit by age 1 year; however, when the supply of dentists is limited, the American Academy of Pediatrics recommends a first visit at age 3 years for those not at highest risk.⁸ North Carolina has a consistent shortage of dentists available to provide care for young children on Medicaid. Our findings support a policy that gives priority to children at high risk of disease for dental office visits before age 3 years and allows other children to delay their first visit to a dentist or to receive oral health care in alternative settings³⁴ when the supply of dentists in a community is limited. This high-risk policy for dentist visits would be most appropriate for communities that have nondental professionals who can provide screening, risk assessment, and referral services.

Children at higher risk must first be identified before such a policy can be effectively implemented. Existing tools, such as the American Academy of Pediatrics' Bright Futures caries risk assessment tool³⁵ or those developed by other groups, can be used by primary health providers to identify children at higher risk.³⁶ The American Academy of Pediatrics is experimenting with strategies such as quality improvement projects that will meet maintenance of certification requirements to help providers incorporate oral health screening into their practices.³⁷ However, more research is needed to develop better measures for use in identifying young children at highest risk for dental disease, to incorporate these measures into risk assessment tools, and to promote effective dissemination and implementation of risk assessment among appropriate primary care providers and community public health professionals.³⁸

Our findings also indicated that this Medicaid population followed an urgent care type of dental utilization: children went to the dentist when they had pain or known disease. As a result, children who appeared to most need dental care because of dental disease were also the most likely to go to the dentist at an earlier age and receive treatment, which could indicate that high-risk children are receiving appropriate priority. However, in the entire sample of children enrolled in Medicaid, utilization of dental care was low. Only 32% of children had any dental visit paid by Medicaid by the time they were in kindergarten, although 48% had experienced dental caries. In the overall sample, 54% had untreated disease. Moreover, only 7% of the sample had received a preventive visit in accordance with clinical guidelines before age 3.5 years. To meet the Healthy People 2020 oral health objectives,⁷ future research should focus on ways to improve access to dental care for young children enrolled in Medicaid.

Children with disease who had a visit by age 24 months had more untreated disease than children who had a visit by age 24 to 36 months, but did not differ from children with visits at older ages. This finding could suggest that children who received a visit before 24 months in our sample lacked a dental home, as recommended in the guidelines for an early preventive visit.^8,10 Children in areas with limited access—living in counties with fewer dentists per capita and higher concentrations of children enrolled in Medicaid—had higher rates of untreated disease. These results indicate that access to care continued to inhibit utilization of dental care in this Medicaid population.

Limitations

Our study was observational and thus subject to selection bias from unobserved factors, especially unobserved risk of dental caries. Children at higher risk for disease may have been more likely than low-risk children to have an earlier visit. Our data on individuals were limited to claims files and the NCSoDC, so we did not have individual demographic data, such as family income or mother’s dental utilization, which may be associated with a child’s oral health and dental utilization. The study data only contained claims submitted for care reimbursed by Medicaid; children could have received dental care from outside sources that we did not measure.

The measurement of our outcome variable might also have suffered from bias. An examiner is more likely to count a diseased tooth that has a filling than one without. A dentist is also more likely to diagnose a cavity after examination with x-rays than in the type of visual examination done in our study. Moreover, dentists may be aggressive in treatment of young, high-risk children such as those in our study, so children in our sample with dental visits and subsequent treatment may have had overestimates of disease. The count of caries for children who had not received treatment of all of their decay might therefore be biased downward. If children who had an early visit were more likely to use dental care, this would bias our results such that children with an early visit would appear to have more disease.

Because children enrolled in Medicaid have different risk and utilization patterns than other children, our results may not apply to other populations. Moreover, our sample comprised only users of dental care in the Medicaid population. Future studies should examine nonusers of dental care to understand the full policy relevance for the recommendation of the age of the first dental visit. We did not examine frequency of dental visits; future studies could examine whether children with multiple early dental visits have better oral health outcomes.

Conclusions

Children with early dental visits had higher rates of disease than children who had later visits in our sample of North Carolina Medicaid dental users. However, children with an early visit that included more than 2 treatments had no difference in rates of disease in kindergarten than children with visits at older ages. Children in areas with limited access to dental care had higher rates of untreated disease. These results suggest that recommendations for a preventive dental visit before age 3 years could focus on children at high risk for dental disease, especially when the supply of dentists is inadequate. Our findings indicate that access to dental care was still limited for Medicaid-enrolled children in 1999 to 2006. Future research should continue to examine the effectiveness and costs of ways to increase access to oral health care for young children enrolled in Medicaid.