PEDIATRICS Vol. 115 No. 4 April 2005, pp. 906-914 (doi:10.1542/peds.2004-1687)
Estimated Impact of Competing Policy Recommendations for Age of First Dental Visit
* Division of Public-Private Partnerships, National Center for Health Marketing, Centers for Disease Control and Prevention, Atlanta, Georgia
Division of Public Health Services and Research, College of Dentistry, University of Florida, Gainesville, Florida
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective. To compare levels of dental utilization and untreated dental decay among children aged 1 to 3 years that are likely to occur under 2 potential guidance policies: (1) pediatricians refer all toddlers to dentists for screening (consistent with American Academy of Pediatric Dentistry and the American Dental Association recommendations; DENT), and (2) pediatricians receive training in caries risk assessment, screen toddlers, and refer at-risk children to dentists (consistent with American Academy of Pediatrics recommendations; PED).
Methods. Using decision analysis, we estimated the impact of PED and DENT assuming alternately unlimited dental capacity for Medicaid-insured patients and fixed Medicaid dental capacity.
Results With unlimited capacity, if DENT were implemented, then dental utilization is estimated to increase from 27% under the status quo to 65% and untreated decay to decrease from a mean of 0.60 surfaces to 0.52 surfaces per child. If PED were implemented, then dental utilization and untreated decay would decrease from status quo levels to an estimated 11% and 0.47 surfaces, respectively, assuming that diagnostic sensitivity and specificity both equaled 1; they would decrease to 13% and 0.53 surfaces, respectively, if sensitivity equaled 0.76 and specificity equaled 0.95. With fixed capacity, under DENT, untreated decay is estimated to increase to 0.63 surfaces because low-risk private-pay patients would crowd out at-risk Medicaid-insured children, whereas under PED, untreated decay would still be less than under the status quo.
Conclusions. Implementing PED will decrease untreated decay under most plausible scenarios, whereas switching to DENT will increase the burden of disease if Medicaid dental capacity is limited.
Key Words: delivery of dental care • dental health services • dental care for children • dentist's practice patterns • dental economics • health policy • pediatrics • dental caries
Abbreviations: AAP, American Academy of Pediatrics • SQ, status quo • SCHIP, State Children's Health Insurance Program • NHANES III, Third National Health and Nutrition Examination Survey • p_low, proportion of toddlers from families with incomes 
200% of the federal poverty threshold • pSQdvisithigh, proportion of high-income toddlers with past-year dental visit • pSQdvisitlow, proportion of low-income toddlers with past-year dental visit • pDENTdvisithigh, proportion of high-income toddlers with past-year dental visit if AAP lowered recommended age for first dental visit to 1 year • pDENTdvisitlow, proportion of low-income toddlers with past-year dental visit if AAP lowered recommended age for first dental visit to 1 year • ppedvisithigh, proportion of high-income toddlers with past-year visit to pediatrician or primary care provider • ppedvisitlow, proportion of low-income toddlers with past-year visit to pediatrician or primary care provider • pdh, untreated decay prevalence (at least 1 tooth surface has untreated decay) among high-income toddlers • pdl, untreated decay prevalence among low-income toddlers • dh, untreated decay severity (mean number of decayed tooth surfaces) among high-income toddlers • dl, untreated decay severity among low-income toddlers • pch, caries prevalence (at least 1 filled or untreated decayed tooth surface) among high-income toddlers • pcl, caries prevalence among low-income toddlers • ch, caries severity (mean number of filled or untreated decayed tooth surfaces) among high-income toddlers • cl, caries severity among low-income toddlers • PFh, percentage reduction in untreated decay among high-income toddlers attributable to a past-year dental visit • PFl, percentage reduction in untreated decay among low-income toddlers attributable to a past-year dental visit
Over the past several years, much discussion has centered on the age at which a child without identified dental problems should first visit a dentist.1,2 Several dental professional organizations have offered a strong rationale for making that first visit by age 1.3 A visit by age 1 may provide the opportunity to evaluate craniofacial and dental development, assess risk for common dental conditions, and counsel parents and caregivers on primary prevention interventions before disease progresses, poor habits become well established, or irreversible harm occurs.3
Currently, the American Academy of Pediatric Dentistry,3 the American Dental Association,4 and the Bright Futures Project of the National Center for Education in Maternal and Child Health5 recommend that dentists examine toddlers by their first birthday. In addition, an expert panel was convened by the Health Care Financing Administration (now Centers for Medicare and Medicaid Services) and the National Center for Quality Assurance and charged with evaluating and developing performance measures for the assessment of oral health services delivered to Medicaid-insured children and other pediatric populations. The panel suggested revising the current HEDIS measure, Annual Dental Visit, to extend the age range to children age 1 year and younger.6 Currently, the measure includes the proportion of Medicaid-enrolled people, ages 4 through 21 years, who had at least 1 dental visit during the preceding year.
The American Academy of Pediatrics (AAP) previously recommended first dental visit at age 3 and recently adopted a policy statement that advocates an oral health assessment of infants by a pediatrician or other qualified pediatric health care professional by 6 months of age.7 The policy statement recommends that children who are deemed to be at high risk for dental caries be referred to a dentist no later than 6 months after the eruption of the first tooth or by 12 months of age, whichever comes first. The policy statement concludes, "The ideal deterrence to early childhood caries is the establishment of the dental home when indicated by the unique needs of the child. Although not always feasible because of manpower and participation issues, best practice dictates that whenever feasible, all patients should have a comprehensive dental examination by a dentist in the early toddler years." Presumably, the pediatrician would continue to conduct periodic oral health assessments when feasible if manpower issues preclude early establishment of a dental home.
Toddlers from low-income families have been shown to be at higher risk than higher income children for dental caries.8 They are also more likely to utilize pediatric or primary care services than dental services: a recent study found that among children aged 1 to 4 years, 
85% had at least 1 office-based visit to a primary care provider but only 20% visited a dental care provider.9 Medicaid agencies in at least 2 states now reimburse pediatricians and primary care physicians and extenders for dental screenings and provision of fluoride varnish.10,11 A recent publication that addressed disparities in children's oral health and access to dental care called for greater integration of medicine and dentistry at multiple levels and, among other strategies, advocated advancing the oral health knowledge and skills of primary care practitioners who see children before they are referred to the dentist.12
There seems to be a growing consensus that pediatricians should take an active role in promoting children's oral health. At present, however, it is unclear how frequently pediatricians screen for common oral conditions such as dental caries, which methods they use to conduct screenings, how they document findings, and how accurate such screenings are.10 For example, although 90% of pediatricians in a recent survey responded that they had an important role in both identifying dental caries and providing counsel on caries prevention, 50% of respondents reported no previous training in dental health during medical school or residency and only 9% of respondents correctly answered 4 questions on scientific dental knowledge.13 There are, however, low-cost, evidence-based models for caries risk assessment,14 and 2 recent studies have shown that after a few hours of training, pediatricians can diagnose overt caries in children with probabilities ranging from 76%10 to 100%15 and identify children without caries with probabilities ranging from 87%15 to 95%.10
There may not be sufficient dental capacity for all children aged 1 to 3 years to be seen by a dentist. Parents of children who participate in the Head Start Program, which requires each child to have an oral examination, identified access to oral health services as their main health concern.16 Limited access has been attributed to a low number of dentists accepting Medicaid payment for services, long wait times for appointments, and extensive travel time in rural areas.16 Fifty percent of surveyed pediatricians reported difficulty in successfully referring noninsured children, and 38% reported difficulty in successfully referring Medicaid-insured children.13 Access may be even more limited for very young children. Studies suggest a shortage of pediatric dentists16; many general dentists, who make up 80% of the dentist workforce, report that they are reluctant to see very young children.14 In light of these problems, the debate continues as to effect and efficacy of dental referrals at age 1.
In this study, we compare estimated utilization of dental services (defined as the proportion of children with a dental visit within the preceding 12 months) and levels of untreated dental caries among 1- to 3-year-olds under competing guidance policies: (1) DENT, in which, consistent with dental professional organizations' recommendations, pediatricians refer all children to dentists by age 1, and (2) PED, a more integrated approach, consistent with AAP guidelines, in which pediatricians receive training in risk assessment and refer to a dentist at-risk children who are younger than 3 years, and dentists, in turn, agree to accept these referrals. The expected outcomes of these proposed policies are compared with each other and with the status quo (SQ), in which pediatricians and other primary care providers screen and refer, and dentists examine and treat very young children to varying degrees under conflicting and uncoordinated guidance for children aged <3 years. These policies are analyzed alternately under assumptions of unlimited and limited (or fixed) dental care capacity for publicly insured children (eg, children who are covered by Medicaid or the State Children's Health Insurance Program [SCHIP]).
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Overview
We used decision analysis to estimate the effects of the proposed guidance policies on (1) dental utilization, the proportion of toddlers, aged 1 to 3 years, with a past-year dental visit (Fig 1), and (2) the average number of tooth surfaces with untreated decay among toddlers aged 1 to 3 years (Fig 2). We compared the values of these outcomes expected under each policy with those observed in the SQ, under 2 different scenarios. First, we assumed unlimited dental capacity; that is, at current public insurance reimbursement levels, all caregivers who want to take a child to the dentist can obtain a visit. We next assumed fixed dental capacity, the more likely scenario because some publicly insured children are unable to obtain dental visits under the current system. Fixed capacity results in the crowding out of some low-income children because increased utilization by higher compensation private patients is offset by lower utilization among lower compensation publicly insured children.
|
|
Market Model
This crowd out is predicted by the dual market model, which has been used in previous studies to measure the impact of Medicaid/SCHIP fee increases on dentist and physician participation.17–19 Dental practices are assumed to serve 2 markets: the private and/or self-insured market and the Medicaid/SCHIP market. In the private market, dentists face a downward-sloping demand curve, and, therefore, marginal revenue decreases as a practice takes on additional patients20; in the Medicaid/SCHIP market, marginal revenue is constant at the government-determined reimbursement rate. The standard economic assumptions are that a practice will take on an additional patient as long as the additional revenue generated from that patient (marginal revenue) exceeds the additional cost (marginal cost) of doing so and that practices prefer to take on patients who generate the highest marginal revenue. We assume, as anecdotal evidence suggests, that marginal revenue in the private-pay market exceeds that in the Medicaid market over some relevant range. As a result, the dental capacity available to publicly insured children is unlimited only if the cost of seeing an additional patient (marginal cost) never increases and this cost is less than or equal to the Medicaid/SCHIP reimbursement rate.
Data
We used data from the US Census Bureau's March 2001 Current Population Survey (for children <6 years of age) to estimate the proportion of 1- to 3-year-olds who are covered by Medicaid or SCHIP using the proportion of children who live in families with incomes 200% of the federal poverty line, the typical threshold for Medicaid/SCHIP eligibility.21 This value is labeled p_low in the decision tree. (Note that the names of decision tree variables appear in italics.) These children are hereafter referred to as Medicaid children, Medicaid toddlers, low-income children, or low-income toddlers; all other 1- to 3-year-olds are collectively referred to as high-income children or high-income toddlers.
We used interview and clinical examination data from the Third National Health and Nutrition Examination Survey (NHANES III; conducted between 1988 and 1994) to estimate utilization and disease parameters. Additional information on NHANES III sampling methods and diagnostic criteria has been published previously.22 No data were available for 1-year-olds regarding utilization of dental services, mean level of caries (filled or untreated decay), or mean level of untreated decay. Thus, we assumed that these values equaled those values for 2- and 3-year-olds. Utilization parameters include (1) the proportion of children who were aged 2 or 3 years (proxy for 1- to 3-year-olds) and had a past-year dental visit as reported by a responsible household adult (decision tree variables labeled pSQdvisithigh and pSQdvisitlow for high-income and Medicaid toddlers, respectively), (2) the proportion of children who were aged 4 or 5 years and had a past-year dental visit as reported by a responsible household adult (variables labeled pDENTdvisithigh and pDENTdvisitlow for high-income and Medicaid toddlers, respectively), and (3) the proportion of children who were aged 2 or 3 years and had a past-year visit to a pediatrician or other primary care provider as reported by a responsible household adult (variable labeled ppedvisithigh and ppedvisitlow for high-income and Medicaid toddlers, respectively). Estimates of the income and utilization parameters are found in Table 1.
|
Disease parameters among 2- and 3-year-olds include prevalence of untreated decay (decision tree variables labeled pdh and pdl for high-income and Medicaid toddlers, respectively), severity of untreated decay (variables labeled dh and dl for high-income and Medicaid toddlers, respectively), prevalence of caries experience (ie, treated or untreated decay; variables labeled pch and pcl for high-income and Medicaid toddlers, respectively), and severity of caries experience (for high-income and Medicaid toddlers, variables labeled ch and cl, respectively). For this same age group, we also calculated the proportion with a past-year dental visit by caries prevalence and, among children with a past-year dental visit and caries experience, the proportion of caries experience that was untreated decay. To estimate the impact of increased dental utilization on untreated decay, we calculated the percentage reduction in untreated decay among 4- and 5-year-olds attributable to a past-year dental visit (decision tree variables labeled PFh and PFl for high-income and Medicaid children, respectively). Estimates of the disease parameters are found in Table 2.
|
Because NHANES III was a complex survey sample, we used SUDAAN statistical software to calculate all point estimates and standard errors and to conduct statistical tests.23 Finally, in our subsequent sensitivity analysis, we used data from a published study to estimate feasible bounds on the probabilities of a pediatrician's or primary care provider's correctly identifying untreated decay (variable labeled sensitivity) and correctly identifying healthy teeth (variable labeled specificity).10
Decision Trees
In analyzing our decision trees, we made the following assumptions.
- Baseline levels of sensitivity and specificity both are 100%.
- Because it is already recommended that 4- and 5-year-olds visit the dentist, implementing DENT increases demand for dental services among 1- to 3-year-olds of each income group to levels comparable to the current utilization rates of 4- and 5-year-olds in that group.
- Because 1- to 3-year-olds generally do not have a dental home and pediatricians are the primary source of health care advice for their parents, under PED, pediatricians become the de facto gatekeeper for dental visits among these children. We further assume that all those with untreated decay are diagnosed and successfully referred to a dentist, and only those who receive a dental referral will seek dental care. Thus, among toddlers with a pediatric visit, the proportion who visit the dentist will equal the proportion with caries, and among toddlers without a pediatric visit, the probability of visiting a dentist and mean level of disease equal those values under the SQ.
- Most dentists prefer to see older children14,24; therefore, increases in demand for dental services among higher income 1- to 3-year-olds will not crowd out utilization among older low-income children.
Figure 1 shows the decision tree used to analyze the level of dental utilization among 1- to 3-year-old children that is both observed in the SQ and expected under DENT and PED, assuming unlimited dental capacity. The decision tree in Fig 2 shows the amount of untreated decay in the SQ and the amount of untreated decay expected if DENT or PED were implemented, assuming unlimited capacity. Plugging the parameter estimates found in Tables 1 and 2 into the decision models, we calculated the utilization rates and untreated decay levels both observed in the SQ and expected under the competing policy recommendations. For example, assuming unlimited capacity, the utilization expected under DENT is found by summing the products of the probabilities along each branch coming from the DENT trunk in Fig 1: [p_low x pDENTdvisitlow x 1)] + [p_low x (1 – pDENTdvisitlow) x 0] + [(1 – p_low) x pDENTdvisithigh x 1] + [(1 – p_low) x (1 – pDENTdvisithigh) x 0] = [0.4 x 0.5391 x 1] + [0.4 x (1 – 0.5391) x 0] + [(1 – 0.4) x 0.7206 x 1] + [(1 – 0.4) x (1 – 0.7206) x 0] = 0.65. Likewise, assuming unlimited capacity, the amount of decay expected under DENT is found by summing the products of all probabilities and the end-node decay value along each branch coming from the DENT trunk in Fig 2: [p_low x pDENTdvisitlow x (1 – PFl) x dl)] + [p_low x (1 – pDENTdvisitlow) x dl] + [(1 – p_low) x pDENTdvisithigh x (1 – PFh) x dh] + [(1 – p_low) x (1 – pDENTdvisithigh) x dh] = [0.4 x 0.5391 x (1 – 0.1536) x 1.28] + [0.4 x (1 – 0.5391) x 1.28] + [(1 – 0.4) x 0.7206 x (1 – 0.6162) x 0.14] + [(1 – 0.4) x (1 – 0.7206) x 0.14] = 0.52.
The same decision trees were used to analyze outcomes under the assumption of limited capacity. However, the probability of a low-income child's visiting a dentist under DENT was changed to 0 (from pDENTdvisitlow), because all low-income children are crowded out by higher paying, high-income children. (If Medicaid capacity were fixed, then any increase in utilization among private-pay patients would have to be offset by a decrease in Medicaid patients. Under DENT, 72.06% of high-income 1- to 3-year-olds are expected to seek dental appointments, ie, the percentage of high-income 4- and 5-year-olds who are already referred to and seek dental appointments under the SQ.) In addition, the amount of untreated decay in the mouths of low-income children who do not visit the dentist equals cl under the assumption of fixed capacity (instead of dl). (Because no low-income children visit the dentist under this scenario, the amount of both untreated decay and treated decay that is observed under the SQ will be untreated under DENT.) So, for example, the expected utilization under DENT with the assumption of fixed capacity is as follows: [p_low x 0 x 1)] + [p_low x 1 x 0] + [(1 – p_low) x pDENTdvisithigh x 1] + [(1 – p_low) x (1 – pDENTdvisithigh) x 0] = [0.4 x 0 x 1] + [0.4 x 1 x 0] + [(1 – 0.4) x 0.7206 x 1] + [(1 – 0.4) x (1 – 0.7206) x 0] = 0.43. Likewise, the amount of decay expected under DENT with the assumption of fixed capacity is as follows: [p_low x 0 x (1 – PFl) x cl)] + [p_low x 1 x cl] + [(1 – p_low) x pDENTdvisithigh x (1 – PFh) x dh] + [(1 – p_low) x (1 – pDENTdvisithigh) x dh] = [0.4 x 0 x (1 – 0.1536) x 1.47] + [0.4 x 1 x 1.47] + [(1 – 0.4) x 0.7206 x (1 – 0.6162) x 0.14] + [(1 – 0.4) x (1 – 0.7206) x 0.14] = 0.63.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Status Quo
At present,
40% of children who are aged 6 years or younger live in families with incomes 200% of the federal poverty level.21 Under the SQ, past-year dental visits among high-income 1- to 3-year-olds were estimated to be 48% higher than among their Medicaid-insured counterparts, whereas past-year pediatric/primary care visits among the high-income group were only 8% higher (Table 1). The mean number of tooth surfaces with untreated decay and mean number of surfaces with caries experience (filled or untreated decay) among Medicaid toddlers in the SQ (1.28 and 1.47, respectively) are 8 times higher than those values among high-income toddlers (0.14 and 0.19; Table 2). In fact, 86% of all untreated decay ([0.4 x 1.28]/[0.4 x 1.28 + 0.6 x 0.14]) was found in low-income children.
Estimated Effects of Implementing the Recommended Policies
Unlimited Capacity
A summary of utilization rates and untreated decay observed in the SQ and expected under the competing policy recommendations is found in Table 3. If dental capacity were unlimited, then implementing DENT would increase utilization rates from 27% (in the SQ) to 65% and decrease mean untreated decay per child from 0.60 surfaces to 0.52 surfaces. Implementing PED would decrease dental utilization to 11%; however, because more toddlers with untreated decay then would visit the dentist, the mean number of untreated decayed surfaces would decrease to 0.47 surfaces. Thus, selecting PED would result in 0.05 fewer decayed surfaces per toddler on average than would DENT.
|
Fixed Capacity
Implementing DENT increases demand for utilization among high-income toddlers to an estimated 72.06%, thus crowding out all Medicaid toddlers. As a result, Medicaid toddlers then would have a mean of 1.47 tooth surfaces with untreated decay because there would be no dental care access. Overall untreated decay in this age group is expected to increase to 0.63 surfaces. Because implementing PED does not increase overall dental utilization, no crowding out of Medicaid patients would occur. Thus, even with constrained capacity, the level of untreated decay is expected to decrease to 0.47 surfaces under PED.
Sensitivity Analysis
Unlimited Capacity
Implementing PED becomes less attractive as diagnostic sensitivity and specificity decrease from our assumed value of 1. For example, if diagnostic sensitivity equaled 76% and diagnostic specificity equaled 95%, the values observed in a recent study of the accuracy of pediatrician screenings,10 then utilization and untreated decay under PED would equal 13% and 0.53 surfaces, respectively. Assuming that specificity is only 95%, PED is expected to reduce untreated decayed surfaces relative to the SQ if sensitivity is at least 44%; PED is expected to result in fewer decayed surfaces than DENT if sensitivity is at least 81%. If specificity is <100%, then untreated decay is unaffected under unlimited capacity because mistakenly classifying children as being at risk for dental decay when they in fact are not at risk does not increase their chances of developing decay.
Fixed Dental Capacity
The assumption that dental capacity is fixed decreases the attractiveness of DENT because increases in utilization among higher income children, who are at lower risk for caries, will absorb current Medicaid dental capacity. For no crowding out of Medicaid children to occur (ie, to achieve the unlimited capacity outcome), we would have to assume that we could almost double the dental capacity available to 1- to 3-year-olds without increasing costs per additional patient seen. (To see this, recall that implementing DENT would increase utilization among high-income children to an estimated 72.06% [pDENTdvisithigh]. This means that, overall, 52% of 1- to 3-year-olds would desire a dental visit [0.6 x 0.7206 + 0.4 x 0.2088] under DENT, whereas only 27% of 1- to 3-year-olds visit the dentist under the SQ [0.3098 x 0.6 + 0.2088 x 0.4].)
Inaccuracies in pediatricians' determination of the risk of caries in toddlers also have an affect on the expected outcomes under PED when capacity is limited. The effect of changes in sensitivity under limited capacity are the same as the effect under unlimited capacity because as sensitivity falls below 100%, fewer children are referred to a dentist, leading to less strain on capacity. Decreases in specificity, however, mean more unnecessary referrals. The capacity constraint becomes binding at levels of specificity below 100% if sensitivity is assumed to be 100%, and at levels of specificity below 94% for sensitivity of 76%. Assuming sensitivity were only 76%, at levels of specificity higher than 39%, the expected decay under PED is less than the SQ, and even if specificity were 0%, expected decay is still lower under PED than under DENT.
If, under PED, we relax the assumption that pediatricians become the de facto gatekeeper and instead assume that children who visit the dentist under the SQ would also do so under PED, then fixed capacity would become an issue and some crowding out would result. The worst-case scenario for crowd out is if all at-risk children who are recommended for a dental visit try to schedule an appointment and children who are deemed not to be at risk continue to demand appointments in the same proportions as under the SQ. In this case, assuming sensitivity and specificity of 1, utilization drops to 28.8% and the expected decay is 0.565 surfaces; assuming sensitivity and specificity of 0.76 and 0.95, respectively, utilization is 28.4% and expected decay is 0.557 surfaces. In both cases, the number of decayed surfaces expected per child is still fewer than under SQ or DENT.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Approximately 11% of 1- to 3-year-old children have experienced dental caries (Table 2). Among these toddlers, the condition can be fairly severe: a mean of 5.33 tooth surfaces had fillings or untreated decay (Table 2). A total of 84% of the disease is concentrated in low-income toddlers, who at present are less likely to utilize dental services. Implementing DENT may decrease or increase the burden of disease, depending on the dental capacity available to Medicaid patients. Implementing PED, however, is expected to decrease the burden of untreated disease, regardless of assumptions about dental capacity.
We used a well-established model to analyze the possible consequences of changing the AAP recommended age of first dental visit; however, multiple factors affect demand for and supply of dental services. Thus, the issue of increasing utilization of dental services among low-income children has more complexities than modeled. Although our results do not provide margins of error for our untreated decay or utilization estimates, we do provide plausible point estimates of outcomes under a best-case (unlimited Medicaid dental capacity) and worst-case (fixed Medicaid dental capacity) scenario and sensitivity analysis regarding other assumptions, which can be used as a launching point to discuss the various tradeoffs associated with implementing 1 of the 2 widely advocated policies, DENT or PED.
Our analysis of utilization levels and untreated decay under PED is limited by 2 factors. First, our estimates of sensitivity and specificity come from a study of pediatricians' abilities to identify caries and not precavitated lesions, the more difficult but important task for preventing decay.10 Thus, we include extensive sensitivity analysis of the effects that these values have on expected outcomes. In addition, pediatricians in the study did not refer a substantial proportion of at-risk children for a dental visit.10 If pediatricians did not refer all at-risk children for dental visits, then the expected decay under PED would be higher than our model predicts. However, the study's authors state that "primary care providers may not have referred because they did not have a dental provider to whom they could refer."10 If capacity is constrained, then the crowd out resulting from DENT could only exacerbate this situation.
Several of our assumptions bias our analysis toward finding DENT more attractive than the SQ or PED. First, if utilization and disease levels among 1-year-olds were less than those of 2- and 3-year olds, then our estimates of current utilization and untreated decay would be inflated. Therefore, DENT would result in an even greater increase in utilization among children without caries, and if dental Medicaid capacity were limited, then that would in turn result in even greater crowding out of at-risk children. The assumption that increased dental demand by 1- to 3-year-olds had no impact on capacity available to older children could also underestimate the impact on untreated decay. In reality, because they generate higher compensation for dentists, younger low-risk children could crowd out older at-risk children who have more disease.
We also assumed that higher prices in the private-pay market resulting from increased demand did not reduce utilization among the private-pay group. For example, we assumed that DENT would increase utilization among high-income children >2-fold to 72.06%. If capacity did not expand by this same amount or if higher prices deterred some visits, then untreated decay among the high-income group would be understated. Finally, we assumed that the lower effectiveness of a past-year dental visit in preventing decay observed among Medicaid toddlers was independent of capacity. If the lower prevented fraction were attributable to existing crowding out (eg, Medicaid children have difficulty in scheduling follow-up appointments for treatment), then reducing crowding out by adopting PED would result in an even greater decrease in untreated decay. Our finding that, among Medicaid children with a past-year dental visit, 76% of caries experience was untreated decay (compared with 39% among their higher income counterparts) suggests that crowd out may indeed be present in the SQ.
The attractiveness of DENT depends on assumptions about capacity constraints. The dental capacity available to Medicaid patients is limited to current levels if 3 conditions are met: (1) demand for dental services exceeds supply of dental services in the Medicaid market currently, (2) the cost of seeing additional patients is increasing (ie, the supply curve is upward-sloping), and (3) supply does not expand. Medicaid parents' and pediatricians' difficulties in obtaining dental appointments for their children in the SQ suggest that demand already exceeds supply. If we assume that the dentist either examines or supervises the examination of each patient, then seeing additional patients requires additional units of dental time. If the number of dentists is fixed, then, at some point, the cost of seeing additional patients must increase. This is because when deciding whether to see an additional patient, the dentist compares the value of his time spent working with the value of that time spent not working. As more hours are worked in a given period, income increases and available leisure time decreases; the value of additional leisure time increases relative to the value of additional income. Because, from the dentist's perspective, the cost of treating an additional patient includes the value of the leisure forgone, dentists view marginal cost (of treating an additional patient) as increasing. Findings from a recent study suggest that indeed both the supply of dentists is fixed and dentists are forgoing additional income to consume additional leisure. The number of practicing dentists has been declining since 1991, and decreases in dental school enrollment suggest that this trend may continue. In addition, the average number of hours worked per dentist is decreasing: the number of part-time dentists increased by 68% from 1982 to 1995, and in 1998, dentists averaged only 33.3 hours per week treating patients.25
Implementing PED would require additional time of pediatricians and other primary care providers. Whereas dental capacity has been declining, physician capacity has been increasing over the past 40 years25 and physicians average 51.6 hours per week treating patients, 55% more time than their dental counterparts. Furthermore, the supply of general pediatricians is expected to increase by 33% from 2000 to 2010 and by 58% from 2000 to 2020, outpacing the growth of the child population.26 As a result, pediatricians then may look to expand into delivery of other services.26 Although dental capacity may be augmented by increased use of dental auxiliaries, eg, dental hygienists and assistants, requirements that auxiliaries must work under the supervision of a dentist suggests that the declining number of dentists still restricts dental capacity.
The scope of this study was to compare the competing guidance policies on the basis of utilization rates and untreated dental decay expected under each policy. Before implementing any policy, all benefits of the strategy should be considered and compared with the costs of achieving these benefits. PED and DENT will yield different additional benefits. For example, we did not determine whether children who visited a dentist by age 1 year were more likely to utilize dental services in the future or to have established better dental habits than those who had their first visit at age 3 years, and we did not quantify the benefits from increased access to primary prevention. In addition, we did not consider the benefit obtained from pediatricians' or other primary care providers' delivering primary prevention. For example, Medicaid programs in at least 2 states now reimburse primary care providers for application of fluoride varnish.10,11
The costs of the policies will differ as well. DENT will require more dental resources than will PED because more toddlers will use dental services, especially those at low risk for dental caries. PED, however, will require more primary care resources in terms of both additional training to identify caries and additional time consumed in examination, assessment, and referral.
In conclusion, both PED and DENT have been recommended as ways to improve the dental health of toddlers. The success of DENT in directly lowering untreated decay will depend on Medicaid dental capacity and parental compliance, especially among parents of at-risk children, and ability to locate dentists who are willing to see Medicaid-eligible toddlers. If Medicaid dental capacity is currently limited, then implementing DENT without an increase in the Medicaid dental supply will result in decreased utilization among Medicaid children and thus an overall increase in untreated decay. The success of PED depends on primary care providers' accuracy in diagnosing at-risk toddlers and their willingness to participate in such programs, dentists' willingness to work with primary care providers to establish a referral base for at-risk toddlers, and compliance among parents of at-risk children who receive a referral. All of these factors should be considered in the formulation of policy regarding the recommended age of a child's first dental visit.
|
FOOTNOTES |
|---|
Accepted Aug 5, 2004.
Address correspondence to Kari Jones, PhD, Division of Public-Private Partnerships, National Center for Health Marketing, Centers for Disease Control and Prevention, 4770 Buford Hwy NE, MS K-39, Atlanta, GA 30341. E-mail: kjones5{at}cdc.gov
No conflict of interest declared.
This work was completed while Dr Jones was at the Terry College of Business, University of Georgia, Athens, Georgia.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- Edelstein B. Policy issues in early childhood caries. Community Dent Oral Epidemiol. 1998;26(suppl) :96 –103[Web of Science][Medline]
- Edelstein BL. Public and clinical policy considerations in maximizing children's oral health. Pediatr Clin North Am. 2000;47 :1177 –1189[CrossRef][Web of Science][Medline]
- Guideline on infant oral health care. In: American Academy of Pediatric Dentistry. 2002–03 American Academy of Pediatric Dentistry Reference Manual. Chicago, IL: AAPD; 2003:47. Available at: www.aapd.org/members/referencemanual/pdfs/02-03/Infant%20Oral%20Health.pdf. Accessed May 12, 2003
- American Dental Association. Manage your oral health. Available at: www.ada.org/public/manage/stages/parents.asp. Accessed November 3, 2003
- Casamassimo P. Bright Futures in Practice: Oral Health. Arlington, VA: National Center for Education in Maternal and Child Health; 1996. Available at: www.brightfutures.org/oralhealth/pdf. Accessed October 8, 2002
- Crall JJ, Szlyk CI, Schneider DA. Pediatric oral health performance measurement: current capabilities and future directions. J Public Health Dent. 1999;59 :136 –141[Web of Science][Medline]
- Hale KJ. American Academy of Pediatrics Section on Pediatric Dentistry. Oral health risk assessment timing and establishment of the dental home.
Pediatrics. 2003;111
:1113
–1116
[Abstract/Free Full Text] - US Department of Health and Human Services. Oral Health in America: A Report of the Surgeon General. Rockville, MD: Department of Health and Human Services, National Institute of Dental and Craniofacial Research, National Institutes of Health; 2000 [NIH Publication No. 00–4713]
- McCormick MC, Kass B, Elixhauser A, Thompson J, Simpson L. Annual report on access to and utilization of health care for children and youth in the United States—1999.
Pediatrics. 2000;105
:219
–230
[Free Full Text] - Pierce KM, Rozier RG, Vann WF Jr. Accuracy of pediatric primary care providers' screening and referral for early childhood caries. Pediatrics. 2002;109(5) . Available at: www.pediatrics.org/cgi/content/full/109/5/e82
- Access to Baby and Child Dentistry Expanded Program. ABCD"E" program creates partnership between dental and medical providers in Spokane County [news release]. Available at: www.smileabcd.org/news.html. Accessed November 3, 2003
- Mouradian WE, Wehr E, Crall JJ. Disparities in children's oral health and access to dental care.
JAMA. 2000;284
:2625
–2631
[Abstract/Free Full Text] - Lewis CW, Grossman DC, Domoto PK, Deyo RA. The role of the pediatrician in the oral health of children: a national survey. Pediatrics. 2000;106(6) . Available at: www.pediatrics.org/cgi/content/full/106/6/e84
- Kanellis MJ. Caries risk assessment and prevention: strategies for Head Start, Early Head Start, and WIC. J Public Health Dent. 2000;60 :210 –217[Web of Science][Medline]
- Serwint JR, Mungo R, Negrete VF, Duggan AK, Korsch BM. Childrearing practices and nursing caries.
Pediatrics. 1993;92
:233
–237
[Abstract/Free Full Text] - Jones CM, Tinanoff N, Edelstein BL, et al. Creating partnerships for improving oral health of low-income children. J Public Health Dent. 2000;60 :193 –196[Web of Science][Medline]
- Mayer ML, Stearns S, Norton EC, Rozier RG. The effects of Medicaid expansion and reimbursement increases on dentists' participation. Inquiry. 2000;37 :33 –44[Web of Science][Medline]
- Sloan FA, Mitchell JB, Cromwell J. Physician participation in state Medicaid programs. J Hum Res. 1978;8(suppl) :211 –245
- Adams EK. Effect of increased Medicaid fees on physician participation and enrollee service utilization in Tennessee 1985–88. Inquiry. 1994;31 :173 –187[Web of Science][Medline]
- Mankiw NG. Principles of Microeconomics. Orlando, FL: The Dryden Press; 1998
- US Census Bureau. Distribution of people by ratio of income to poverty level by selected characteristics: 2000 [table]. Data from the March, 2001 Current Population Survey. Available at: http://ferret.bls.census.gov/macro/032001/pov/new15_000.htm. Accessed November 3, 2003
- Drury TF, Winn DM, Snowden CB, Kingman A, Kleinman DV, Lewis B. An overview of the oral health component of the 1988–1991 National Health and Nutrition Examination Survey (NHANES III-Phase1). J Dent Res. 1996;75 :620 –630[Medline]
- Shah BV, LaVange LM, Barnwell GB. Killinger JE, Wheless SC. SUDAAN 8.0 Software for Survey Data Analysis. Research Triangle Park, NC: Research Triangle Institute; 1989
- Damiano PC, Kanellis MJ, Willard JC, Momany ET. A Report of the Iowa Title XIX Dental Program. Iowa City, IA: Public Policy Center and College of Dentistry; 1996
- Mertz E, O'Neil E. The growing challenge of providing oral health care services to all Americans. Health Aff (Millwood). 2002;2 :65 –77
- Shipman SA, Lurie JD, Goodman DC. The general pediatrician: projecting future workforce supply and requirements.
Pediatrics. 2004;113(suppl)
:435
–432
[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2005 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Caspary, D. M. Krol, S. Boulter, M. A. Keels, and G. Romano-Clarke Perceptions of Oral Health Training and Attitudes Toward Performing Oral Health Screenings Among Graduating Pediatric Residents Pediatrics, August 1, 2008; 122(2): e465 - e471. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Iida, P. Auinger, R. J. Billings, and M. Weitzman Association Between Infant Breastfeeding and Early Childhood Caries in the United States Pediatrics, October 1, 2007; 120(4): e944 - e952. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. H. Chung, L. M. Kaste, A. Koerber, S. Fadavi, and I. Punwani Dental and Medical Students' Knowledge and Opinions of Infant Oral Health J Dent Educ., May 1, 2006; 70(5): 511 - 517. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
