Contents lists available atScienceDirect
Health Policy
j o u r n a l h o m e p a g e :w w w . e l s e v i e r . c o m / l o c a t e / h e a l t h p o l
Dental service utilization and costs before and after introduction
of fluoride gel application for preschool children in Taiwan
Shu-Fen Chen, Herng-Ching Lin
∗Taipei Medical University, School of Health Care Administration, 250 Wu-Hsing St., Taipei 110, Taiwan
a r t i c l e i n f o
Keywords:
Fluoride
Fluoride gel application Preventive dental service Utilization
a b s t r a c t
Objective: To examine dental service utilization and costs before and after the introduction of fluoride gel application for preschool children in Taiwan.
Methods: This study used claim data for 101,314 preschool children aged≤5 years. The study sample was divided into two groups: children who received fluoride gel application in dental clinics in 2004 as the study group, with children who visited dental clinics in 2004 but did not receive fluoride gel applications as our control group. A difference-in-difference methodology was employed to compare change in the number of dental-visits and relative dental expenditures before and after fluoride gel application for the study group alongside change over the same time period for the control group.
Results: We found that after fluoride gel application, the total number of dental-visits, caries treatments and pulpitis treatments all increased at slower rates for the study group compared to the control group. Similar patterns were found on dental expenditures. Conclusion: Fluoride gel application is associated with a slower rate of growth in the number of visits for the treatment of dental disease and lower dental care expenditures for preschool children.
© 2008 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Issues of oral health in children revolve principally around dental caries. Early Childhood Caries (ECC) is
defined as dental decay among children≤5 years of age[1],
and it is this group’s most common chronic dental disease [2]. ECC can develop as soon as teeth erupt, but its treatment
is the most prevalent unmet health need in US children[3].
It is estimated that almost 19% of children aged 2–5 years
meet the criteria for ECC in the United States[4]. Decay
in primary dentition can result in severe health conditions
including failure to thrive[5,6], speech impediments and
reduced quality of life [7]. In addition, tooth decay not
only affects overall health; its ramifications include chil-dren’s hours lost from school and parents’ hours lost from
∗ Corresponding author. Tel.: +886 2 2736 1661x3613; fax: +886 2 2378 9788.
E-mail address:henry11111@tmu.edu.tw(H.-C. Lin).
work[8]. Furthermore, while ECC is a predictor of decay
in permanent teeth[9], and has serious implications for
adulthood[7,10], thus early intervention can translate into
significant cost savings for dental care for preschool-aged
children[11–13].
A previous study showed that ECC prevalence in Tai-wan was 5% among 2-year-old children, and ECC prevalence
increased to 89% among 6 year olds[14]. Since ECC
repre-sents a significant financial and societal burden in Taiwan, a new nationwide public dental health policy was initi-ated in July 2004. This new policy provides free preventive fluoride gel applications twice a year for children under 5 years old, with intervals no shorter than 180 days. Preven-tive programs, which are conducted by licensed dentists, include applying fluoride gel and providing information about oral hygiene to parents at dental clinics or hospital dental departments.
During recent decades, topically applied fluoride gels have been widely used in prevention programs in schools in many countries. Furthermore, substantial reduction in 0168-8510/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved.
caries increases through application of fluoride gel has been reported, and the treatment effect increased when the
frequency and intensity of gel application increased[15].
However, the economic implications of fluoride gel appli-cation for preventing early childhood caries have been less clear.
The government-run, single-payer national health insurance scheme in Taiwan, the National Health Insurance (NHI), was initiated in year 1995. Since NHI enrollment is mandatory for all Taiwan citizens, more than 96% of the Taiwanese population of 23 million is now enrolled. Fur-thermore, 94.45% of total dental clinics are NHI-contracted and dental health providers who contract with the NHI can provide fluoride gel application and can claim 500 NT$ compensation from NHI for each fluoride gel application [16]. By using this nationwide population-based data from Taiwan over a 3-year period, we examined dental service utilization and costs before and after the introduction of fluoride gel application for preschool children in Taiwan.
2. Research methods
2.1. Database
This study uses national dental claim data obtained from the National Health Insurance Research Database (NHIRD) of the NHI, published by Taiwan’s National Health Research Institute (NHRI) from January 2003 to December 2005. The NHIRD covers all inpatient and outpatient medical benefit claims. Each claim record has ICD-9-CM codes for one prin-cipal operation procedure, one prinprin-cipal diagnosis, and up to four secondary diagnoses, along with the details of the care provided, patient demographics and provider charac-teristics.
Annual information about the population size of cities/counties in Taiwan was abstracted from the Taiwan-Fukien Demographic Fact Book published by the Taiwan
Ministry of the Interior[17]. The data on the number of
dentists in each city/county was obtained from the Taiwan Department of Health. Since the NHIRD released secondary data protects patient anonymity for research purposes, the study was exempt from full review by the Internal Review Board (IRB) after consulting with the Director of the IRB of our university.
2.2. Study sample
In Taiwan there were 1,526,867 children ≤5 years of
age in 2004. The study group consisted of all preschool
children aged≤5 years old who received fluoride gel
appli-cation from dental clinics or hospital dentistry departments between 1 July and 31 December, 2004. There were a total of 50,657 eligible patients (3.3% of the whole population in this age group). During the same period, there were 172,980
preschool children aged≤5 years who visited dental clinics
or hospital dentistry departments without receiving fluo-ride gel applications. We have randomly selected 50,657 from the 172,980 preschool children (one for every one in the study sample), matched with the study group in
terms of gender, age (0≤ age < 2; 2 ≤ age <3; 3 ≤ age < 4
and 4≤ age ≤ 5) and geographic region (northern, central,
southern and eastern Taiwan) as our control group. Ulti-mately, a total of 101,314 preschool children were included in our study.
2.3. Empirical specification
Our analysis uses a difference-in-difference methodol-ogy or a pre–post design with a control group. We compare the change (or difference) in the variables of interest (uti-lization of dental services and costs) before and after fluoride gel application for the study group to change over the same period for the control group. In a difference-in-difference framework, any trend affecting the study and control groups equally will be removed from consideration and the resulting net difference should be purely the effect of fluoride gel application.
Table A1 and Fig. A1 illustrate the difference-in-difference methodology. Suppose x1 and x2 measure average dental utilization during the periods before and after the fluoride gel application; c1 and c2 measure average dental utilization for the same periods in the control group. x measures the change in average dental utilization as a result of the fluoride gel application and other trend
fac-tors.c measures the change in average dental utilization
as a result of only trend factors. The difference betweenx
andc thus removes the changes in average dental
utiliza-tion that are caused by the trend effects, isolating the pure effect of fluoride gel application. Our analysis is based on the following empirical specifications:
Y = a + ˇ1 fluoride gel application + ˇ2 time period + ˇ3 fluoride gel application × time period
+ (ˇ4 gender + ˇ5 age + ˇ6 region + ˇ7 urbanization + ˇ8 number of dentists)
+ ε and the coefficient (ˇ3) of the interaction term × (fluoride gel application × time period) = (x − c) 2.4. Variables of interest
Utilization and costs of dental services that occurred during the study period were treated as dependent vari-ables. The costs were the aggregate of all itemized charges in NT$ (New Taiwan dollars at the average exchange rate in 2004 (US $1 = NT $33) for services and disposables billed to the Bureau of National Health Insurance. Dental ser-vices utilization and costs were categorized according to three conditions: treatment procedures for dental caries (ICD-9-CM code 521.00), treatment procedure for pulpitis (ICD-9-CM code 522.00) and treatment for other condi-tions. The dummy variable of fluoride gel application is equal to 1 when the children received fluoride gel applica-tion between 1 July and 31 December, 2004. The dummy variable for the time period is equal to 1 when the uti-lization and costs of dental services were accrued within 1 year after the date of fluoride gel application for children in the study group or within 1 year of the first dental-visit since 1 July, 2004 for children in the control group. The difference-in-difference estimate of the impact of fluoride
Table 1
Basic sample characteristics—between 1 July and 31 December, 2004.
Study group (n = 50,657) Control group (n = 50,657) Total (n = 101,314) p-Value
n % n % n % Gender 1.000 Female 24,935 49.22 24,935 49.22 49,870 49.22 Male 25,722 50.78 25,722 50.78 51,444 50.78 Age 1.000 Age < 2 1,879 3.71 1,879 3.71 3,758 3.71 2≤ age <3 8,783 17.34 8,783 17.34 17,566 17.34 3≤ age < 4 17,504 34.55 17,504 34.55 35,008 34.55 4≤ age ≤ 5 22,491 44.40 22,491 44.40 44,982 44.40 Urbanization level <0.001 1 (highest) 16,707 32.98 13,794 27.23 30,501 30.11 2 20,112 39.70 18,596 36.71 38,708 38.21 3 7,370 14.55 8,992 17.75 16,362 16.15 4 5,243 10.35 6,423 12.68 11,666 11.51 5 101 0.20 334 0.66 435 0.43 6 432 0.85 1,074 2.12 1,506 1.49 7 (lowest) 692 1.37 1,444 2.85 2,136 2.10 Geographic region 1.000 Northern 23,109 45.62 23,109 45.62 46,218 45.62 Central 19,236 37.97 19,236 37.97 38,472 37.97 Southern 7,099 14.01 7,099 14.01 14,198 14.01 Eastern 1,213 2.39 1,213 2.39 2,426 2.39
Note: Study group: received fluoride gel application; control group: did not receive fluoride gel application.
gel application is captured by the interaction term fluo-ride gel application× time period. It measures differences in dental service utilization and costs before and after flu-oride gel application for the study group as compared to changes in the control group.
Data on patient characteristics, including gender, age, geographic region, place of residence by urbanization level, and the number of dentists according to city/county of residence per 10,000 people, were collected. Preschool chil-dren were originally grouped into the five age categories, 0≤ age <1; 1 ≤ age < 2; 2 ≤ age <3; 3 ≤ age <4; 4 ≤ age ≤ 5. However, there were few children between 0- and 1-years old, and between 1- and 2-years old, thus we combined these two groups into one group, 0≤ age <2.
Geographic regions consisted of northern, central, southern and eastern Taiwan. Urbanization levels for cities/counties were stratified into seven classifications (from 1 indicating the most urbanized to 7 indicating the least urbanized). All 316 cities/counties in Taiwan were then stratified into the seven levels based upon a com-posite score obtained by calculating population density (people/km2), population ratio of people with university
education or above (%), population ratio of people over 65 years (%), population ratio of agricultural workers (%) and the number of physicians per 100,000 people[18]. Since tap water in Taiwan is not fluoridated, urbanization and geographic location were not correlated with fluoride exposure. We took urbanization and geographic location into consideration in this study for the adjustment of pos-sible difference in medical access between areas.
2.5. Statistical analysis
The SAS 8.2 (SAS Institute, Cary, NC) statistical pack-age was used to perform analysis of the data. Descriptive
statistical analyses including frequency, percentage, mean, and standard deviation were performed for all identified variables. Chi-square tests were employed to examine the distribution of demographic characteristics between the study and control groups. Paired t-tests were also con-ducted to compare the difference in dental service costs and utilization before and after fluoride gel application in the study group, and to compare the difference in den-tal service costs and utilization before and after the first dental-visit for the control group, between 1 July and 31 December, 2004. Then a multivariable regression employ-ing difference-in-difference was performed to assess the independent association between fluoride gel application and dental service costs and utilization after adjusting for gender, age, geographic region, residential urbanization level and the number of dentists per 10,000 in the patient’s city/county of residence. A two-sided p-value of less than or equal to 0.05 was considered to be statistically significant.
3. Results
Table 1 shows basic demographic characteristics for 101,314 dental patients in our analytical sample who were ≤5 years old, as a whole and separately, according to fluo-ride gel application status. Of the total sample, about 51% were males. Approximately 44% of the total sample was in the 4–5-year-old age group. The distribution according to level of urbanization indicated the majority was located in communities situated at urbanization levels 1 and 2.
The2tests show that there were significant relationships
between fluoride gel application and urbanization level (p < 0.001).
Prior to fluoride gel application, the average number of dental-visits for caries per person/year in the study group was higher than in the control group (1.08 vs. 0.41,
respectively), nevertheless, after fluoride gel application, the mean number of visits for dental caries in the study group was lower than that of the control group (1.87 vs. 2.62, respectively) (Table 2). Furthermore, growth in the number of dental-visits for caries was smaller for the study group than for the control group (0.79 vs. 2.21, respectively). Similar patterns were observed with dental-visits for pulpi-tis treatment and total dental-visits in general. In the study group, dental-visits for pulpitis increased only by 0.22, from 0.20 to 0.42 per person/year. Meanwhile, dental-visits for treating pulpitis in the control group increased by 0.65, from 0.12 to 0.77 per person/year. Finally, the increase in the number of overall dental-visits for the study and control groups was 1.77 and 3.69, respectively.
Prior to fluoride gel application, the average cost of treating caries per person/year in the study group was higher than that in the control group (NT$ 1512 vs. NT$ 520, respectively), but this pattern was reversed after flu-oride gel application (NT$ 2366 and NT$ 3466 in the study and control groups, respectively), and the growth in costs associated with dental caries was smaller for the study group than for the control group (NT$ 854 compared to NT$ 2946, respectively). In addition, after fluoride gel applica-tion, patients in the study group spent NT$ 306 more on average for the treatment of pulpitis, whereas patients in the control group spent NT$ 838 more on average for pul-pitis procedures. The impact of fluoride gel application on costs associated with dental-visits in general is even more significant. Prior to fluoride gel application, the average cost of dental-visits for all conditions per person/year was higher for the study group (NT$ 1945) than for the control group (NT$ 668), but this reversed after fluoride gel appli-cation (NT$ 3310 vs. NT$ 4730 for the study and control groups, respectively). The growth in costs associated with dental-visits for all conditions was likewise smaller for the study than for the control group (NT$ 1365 compared to NT$ 4062).
In Tables 3 and 4, regression analysis confirms the impact of fluoride gel application on dental service
uti-lization and cost. As shown inTable 3, after adjusting for
patient demographic characteristics, the coefficients of the interaction terms for the treatment of dental caries, pulpitis and total dental service utilization are statistically signifi-cant and negative; fluoride application was associated with the reductions of 1.42, 0.43, and 1.92 in treatments for dental caries, pulpitis and total dental utilization, respec-tively. In addition, age, urbanization level, and geographic region were significantly related to total dental service uti-lization. More total dental service utilization was observed as children got older. Generally speaking, the higher the urbanization level of the city/county, the more total uti-lization of dental services was found. The highest total dental service utilization by geographic region was in the South. Although no statistically significant difference was observed by gender for dental services in general, dental treatment for caries was significantly lower among males than among females, but significantly higher rates of dental treatment for pulpitis was found among males.
As shown inTable 4, regression analysis confirms the
impact of fluoride gel application on dental costs was
similar to its impact on dental service utilization. Com- Table
2 A v er ag e dental utilization and cos ts for select e d cat eg ories of service b e for e and af te r fluoride g e l application or firs t dental-visit. Variable S tudy gr oup (n = 50,696) Contr ol gr oup (n = 50,696) Befor e amean (S.D.) Af te r bmean (S.D.) Dif f cmean (S.D). P air e d t-t es ts p -v alue Befor e amean (S.D.) Af te r bmean (S.D.) Dif f cmean (S.D.) P air e d t-t es ts p -v alue Dental utilization (numb e r o f visits) Dental caries 1.08 (1 .7 0) 1.87 (2.25) 0.79 (2.5 7) <0.0 0 1 0.4 1 (1 .1 5 ) 2.62 (3.5 7) 2.2 1 (2.3 5) <0.0 0 1 Pulpitis 0.20 (0.6 6) 0.42 (1 .05) 0.22 (1 .1 3) <0.0 0 1 0.1 2 (0.4 7) 0.7 7 (1 .6 1 ) 0.65 (1 .4 4) <0.0 0 1 Other dental tr eatments 0.23 (0.59) 1.0 0 (1 .40) 0.7 7 (1 .3 7 ) <0.0 0 1 0.08 (0.34) 0.9 1 (1 .7 2 ) 0.83 (1 .30) <0.0 0 1 T o tal 1.5 1 (2.23) 3.28 (3.5 1) 1.7 7 (3.58) <0.0 0 1 0.6 1 (1 .45) 4.30 (3.5 7) 3.69 (3.45) <0.0 0 1 Cos t Dental caries 1 5 1 2 (26 81) 236 6 (32 1 0 ) 8 54 (40 03) <0.0 0 1 520 (1 52 7) 346 6 (36 80) 29 46 (3652) <0.0 0 1 Pulpitis 34 7 (1 382) 652 (1 808) 30 6 (2 1 6 8 ) <0.0 0 1 1 2 3 (6 4 7) 96 1 (23 5 1 ) 838 (2230) <0.0 0 1 Other dental tr eatments 8 6 (328) 292 (60 1) 20 6 (6 4 4) <0.0 0 1 25 (1 65) 303 (585) 2 7 8 (6 4 4) <0.0 0 1 T o tal 1 9 4 5 (3426) 3 3 1 0 (43 55) 1 365 (5 1 62) <0.0 0 1 6 6 8 (1 9 2 7 ) 4 7 3 0 (4542) 40 62 (5225) <0.0 0 1 Note : S.D. = standar d d e viation. a Means b e for e fluoride g e l application or within 365 da y s prior to the firs t dental-visit. b Means af te r fluoride g e l application or within 365 da y s af te r the firs t dental-visit. c Means af te r− be fo re .
Table 3
Multivariable regression difference-in-difference estimates of the impact of fluoride gel application on dental service utilization.
Variable Utilization
Dental caries Pulpitis Other treatments Total
Intercept −0.59*** −0.31*** −0.11*** −1.00***
Fluoride gel application Control group (reference group)
Study group 0.77*** 0.14*** 0.15*** 1.05***
Time period
Before (reference group)
After 2.23*** 0.66*** 0.75*** 3.65***
Interaction term
Fluoride gel application× time period −1.42*** −0.43*** −0.06* −1.92***
Gender
Female (reference group)
Male −0.05*** 0.07*** −0.04*** −0.03
Age (years)
Age < 2 (reference group)
2≤ age < 3 0.64*** 0.13*** 0.07*** 0.84***
3≤ age < 4 1.19*** 0.32*** 0.09*** 1.62***
4≤ age ≤ 5 1.26*** 0.41*** 0.15*** 1.81***
Urbanization level
1 (highest) (reference group)
2 −0.03** 0.02 −0.05*** −0.05*** 3 −0.12*** 0.01 −0.09*** −0.21*** 4 −0.08*** 0.00 −0.15*** −0.20*** 5 −0.07 0.01 −0.14*** −0.21** 6 −0.34*** 0.03* −0.13*** −0.45*** 7 (lowest) −0.25*** −0.05** −0.13*** −0.42*** Geographic region Northern (reference group)
Central −0.09*** 0.04*** 0.17*** 0.13***
Southern 0.04*** 0.08*** 0.09*** 0.20***
Eastern −0.06* 0.06*** 0.04*** 0.05*
Number of dentists/10,000 residents 0.01*** −0.01*** 0.01*** <0.01
*p < 0.05. **p < 0.01. ***p < 0.001.
pared to the control group, dental costs for the study group were higher on average. After fluoride treatment, there was a statistically significant reduction in expen-ditures for the treatment of dental caries, pulpitis, other dental conditions and for dental treatments in general. Fluoride application was associated with the reductions of NT$ 2092, NT$ 532, NT$ 72, and NT$ 2697 per per-son/year, for dental caries, pulpitis, other procedures and total dental-visits, respectively. In addition, age, urban-ization level, geographic region and dentists per 10,000 residents by each city/county were significantly related to dental expenditures. More total dental expenditures were observed as children got older. Generally speaking, the higher the urbanization level of the city/county, the more dental expenditures were found.
4. Discussion
The importance of primary and secondary preventive care, such as vision, hearing, and oral screening, which should be conducted before children enter school, is well
recognized[19,20], furthermore, the earlier the preschool
children use preventive dental care, the fewer dentally
related costs the NHI program will incur[13]. According
to the Guidelines of the American Academy of Pediatric Dentistry, all children should have their first dental-visit before the age of 12 months and should continue to
visit dentists every 6 months subsequently[21].
Preven-tive dental programs have been instituted in schools in many countries in accordance with these guidelines during recent decades. Furthermore, topical application of flu-oride has been widely used, and statistically significant caries-inhibiting effect of fluoride gel application has been
documented by various controlled trial studies[15]. Taking
advantage of the preventive dental care policy introduced for preschool children in Taiwan, we employed a difference-in-difference methodology to assess the impact of fluoride gel application on the number of dental-visits and dental cost in the preschool population in Taiwan.
The main finding of this study was the significant association between the protective effect of fluoride gel application and the dental utilization and cost. After adjust-ing for gender, age, urbanization level, geographic region and the number of dentists per 10,000 people, we found
Table 4
Multivariable regression difference-in-difference estimates of the impact of fluoride gel application on dental-visit costs.
Variable Costs
Dental caries Pulpitis Others Total
Intercept −944*** −382*** −52*** −1377***
Fluoride gel application Control group (reference group)
Study group 1142*** 253*** 54*** 1448***
Time period
Before (reference group)
After 2848*** 965*** 255*** 4068***
Interaction term
Fluoride gel application× time period −2092*** −532*** −72*** −2697***
Gender
Female (reference group)
Male −28** 33*** −2 3
Age
Age < 2 (reference group)
2≤ age < 3 978*** 228*** 11 1218***
3≤ age < 4 1384*** 435*** 32*** 1850***
4≤ age ≤ 5 1590*** 552*** 92*** 2233***
Urbanization level
1 (highest) (reference group)
2 −101*** −8 −26*** −135*** 3 −272*** −24** −37*** −332*** 4 −254*** −45*** −45*** −345*** 5 −220*** −66 −32** −318*** 6 −515*** −58** −38*** −610*** 7 (lowest) −466*** −140*** −52*** −657*** Geographic region Northern (reference group)
Central 85*** 182*** 57*** 325***
Southern 107*** 95*** 38*** 240***
Eastern −59* 98*** 44*** 82*
Number of dentists/10,000 residents 21*** −8*** 2*** 15***
*p < 0.05. **p < 0.01. ***p < 0.001.
that total dental service utilization and total costs increased more slowly for the study group. A similarly lower rate of increase was found for caries treatment and pulpitis treat-ments in the year following fluoride gel application. These findings are in line with a previous study that found chil-dren receiving fluoridated drinking water were less likely to receive dental treatment than children in a control group, and they also had lower dental treatment costs compared
to the children in the control group[22].
The effects of fluoride application, whether by pro-fessionals or through self-application, on reducing caries
treatment has been well documented[15], with an up to
20% reduction in caries incidence reported [23]. A
pre-vious meta-analysis study of the caries-inhibiting effect of fluoride treatment reported a mean prevention
frac-tion of 22% among highly caries-active children[24]. Our
study found a significantly lower number of dental-visits for caries treatment (1.42 less), for pulpitis treatment (0.43 less) and significantly lower numbers of total dental-visits per preschool child (1.92 less); furthermore, NT$ 2092, NT$ 532, and NT$ 2697 were saved on caries treatment, pulpi-tis treatment and total dental costs, respectively, for each
preschool child per year in the study group. That children in the study group had significantly more dental-visits and higher dental costs before fluoride gel application than the control group did, suggests the study group children were at higher risk of dental disease. However, studies have indi-cated that the earlier children develop dental caries, the
more susceptible they are to carries in the future[9,25,26].
Therefore, given that the study group is likely facing a higher risk for developing caries, it seems significant that the average number and cost of dental-visits after fluoride gel application are nonetheless significantly lower for the study group than for the control group. Thus, we believe that the fluoride gel treatment itself may contribute to the slower rate of growth in the need for dental care among children under 5 years old.
Assessment of children’s dental condition and educa-tion provided about oral hygiene knowledge by dentists when performing fluoride gel application may also account for the decreasing growth in numbers of dental-visits and costs, in addition to fluoride gel application itself. When participating in the fluoride gel application program, par-ents receive counseling from dentists on children’s oral
hygiene and diet. Moreover, participation in the program means a higher chance of detecting dental disease in its early stages, preventing further worsening of the condi-tion. The increasing severity of untreated dental disease secondary to postponing care necessitates more extensive and costly treatments. Through early detection, significant financial savings can be made. For example, a study con-ducted in Iowa found that although only 5% of the dental care related to the Medicaid-participating children treated for ECC in hospitals or ambulatory care settings, up to 25–45% of all funds allocated for dental resources were
consumed by those children[11]. Similar results were also
reported in Washington State. In that study, 19% of pedi-atric dental emergencies were related to ECC; of those,
more than half involved children≤3.5 years of age[12].
These studies emphasize that early prevention can decrease health care costs.
4.1. Limitations
Since this study is not a controlled trial, our results must be viewed with caution. First of all, children in the study group had significantly more dental-visits and higher den-tal costs before fluoride gel application than the control group did, which suggests the study group children were at higher risk of dental disease, and therefore the effect of fluoride gel application was probably greater than it would be for an average group of children.
Second, data for preventive dental care paid out-of-pocket or not covered by the NHI program were not collected for this study. However, out-of-pocket preventive dental care, such as fluoride tablets and the application of sealants, is uncommon in Taiwan, so the proportion of preschool children with parents paying out-of-pocket for preventive dental care should be quite small. Finally, although we tried to include every important variable, restrictions inherent to administrative databases mean we were not able to access parents’ characteristics, such as maternal age, educational level, and socioeconomic status. However, since we employed the difference-in-difference method in our study, we believe those variables would not influence our results greatly.
4.2. Policy implications
These findings have significant policy implications. The majority of preschool children who saw dentists went mostly for restorations, especially for the treatment of den-tal caries and pulpitis. Nevertheless, according to the data extracted from the NHIRD, only 23% of preschool children utilized the preventive dental care program from July to
December 2004. This is probably a result of limited and ineffective government promotion of the program result-ing in the lack of parental awareness of this fluoride gel application policy for preschool children. This low uti-lization rate can be overcome through several measures. First of all, preschool kindergartens, news media, and most importantly, pediatricians can raise parental awareness of the significance of preventive dental care for preschool children. Because Taiwan’s National Health Insurance Pro-gram provides well-baby care for children aged 0–7 years old, every eligible child is entitled to a total of nine well-baby/child care visits, making pediatricians the type of physician preschool children are most likely to come across. This puts pediatricians in a position to address the importance of oral health through anticipatory guidance, informing parents about the preventive dental care policy,
and to refer high-risk children to dentists[27]. Second, it
normally requires more time for dentists to treat young children and not all dentists have the skills and equipment to provide care for young children. Thus providing dentists with training in providing care to young children should increase access to preventive dental care, particularly for very young children.
5. Conclusion
In conclusion, our findings underscore the potential impact of preventive dental interventions. We found that early provision of preventive dental services is associated with a slower rate of growth in the number of dental-visits for the treatment of dental disease and lower dental care costs for preschool children for dental caries, pulpi-tis and other procedures. These findings have significant policy implications; we hope that these results will help to mobilize the dental profession in Taiwan to support pre-ventive dental services, especially for preschool children. In the future, more research is needed to examine this effect over a longer period of time in order to evaluate the long-term impact of fluoride application as part of preventive dental service for preschool children.
Conflict of interest
None declared.
Appendix A
SeeTable A1andFig. A1.
Table A1
The difference-in-difference methodology.
Before After Difference (after− before)
Study group (received fluoride gel application)
x1 (dental cost or utilization before fluoride gel application)
x2 (dental cost or utilization after fluoride gel application)
x = x2 − x1 Control group (did not receive
fluoride gel application)
c1 (dental cost or utilization before the first dental-visit)
c2 (dental cost or utilization after the first dental-visit)
c = c2 − c1
Fig. A1. The difference-in-difference methodology. References
[1] Drury TF, Horowitz AM, Ismail AI, Maertens MP, Rozier RG, Sel-witz RH. Diagnosing and reporting early childhood caries for research purposes. A report of a workshop sponsored by the National Institute of Dental and Craniofacial Research, the Health Resources and Services Administration, and the Health Care Financ-ing Administration. Journal of Public Health Dentistry 1999;59: 192–7.
[2] Edelstein BL, Douglass CW. Dispelling the myth that 50 percent of U.S. schoolchildren have never had a cavity. Public Health Reports 1995;110:522–30.
[3] Newacheck PW, Hughes DC, Hung YY, Wong S, Stoddard JJ. The unmet health needs of America’s children. Pediatrics 2000;105: 989–97.
[4] Vargas CM, Crall JJ, Schneider DA. Sociodemographic distribution of pediatric dental caries: NHANES III, 1988–1994. Journal of the Amer-ican Dental Association 1998;129:1229–38.
[5] Acs G, Lodolini G, Kaminsky S, Cisneros GJ. Effect of nursing caries on body weight in a pediatric population. Pediatric Dentistry 1992;14:302–5.
[6] Acs G, Shulman R, Ng MW, Chussid S. The effect of dental rehabil-itation on the body weight of children with early childhood caries. Pediatric Dentistry 1999;21:109–13.
[7] Hollister MC, Weintraub JA. The association of oral status with sys-temic health, quality of life, and economic productivity. Journal of Dental Education 1993;57:901–12.
[8] Gift HC, Reisine ST, Larach DC. The social impact of dental problems and visits. American Journal of Public Health 1992;82:1663–8. [9] Greenwell AL, Johnsen D, DiSantis TA, Gerstenmaier J, Limbert N.
Lon-gitudinal evaluation of caries patterns form the primary to the mixed dentition. Pediatric Dentistry 1990;12:278–82.
[10] General Accounting Office. Oral health: dental disease is a chronic problem among low-income populations. Report GAO/HEHS-00-72. Available at:http://www.gao.gov. Accessed August 31, 2000. [11] Kanellis MJ, Damiano PC, Momany ET. Medicaid costs associated with
the hospitalization of young children for restorative dental treat-ment under general anesthesia. Journal of Public Health Dentistry 2000;60:28–32.
[12] Sheller B, Williams BJ, Lombardi SM. Diagnosis and treatment of dental caries-related emergencies in a children’s hospital. Pediatric Dentistry 1997;19:470–5.
[13] Savage MF, Lee JY, Kotch JB, Vann Jr WF. Early preventive den-tal visits: effects on subsequent utilization and costs. Pediatrics 2004;114:e418–23.
[14] Tsai AI, Chen CY, Li LA, Hsiang CL, Hsu KH. Risk indicators for early childhood caries in Taiwan. Community Dentistry and Oral Epidemi-ology 2006;34:437–45.
[15] Marinho VC, Higgins JP, Logan S, Sheiham A. Systematic review of controlled trials on the effectiveness of fluoride gels for the pre-vention of dental caries in children. Journal of Dental Education 2003;67:448–58.
[16] Department of Health. Taiwan public health report 2005. Taipei, Tai-wan: Department of Health, the Executive Yuan; 2006.
[17] Ministry of the Interior. Taiwan-Fukien Demographic Fact Book 2004. Republic of China: Ministry of the Interior; 2004.
[18] Liu CY, Hung YT, Chuang YL, Chen YJ, Weng WS, Liu JS, Liang KY. Incorporating development stratification of Taiwan townships into sampling design of large scale health interview survey. Journal of Health Management 2006;4:1–22.
[19] Navarro-Rubio MD, Jovell AJ, Schor EL. Socioeconomic status and pre-ventive health-care use by children in Spain. American Journal of Preventive Medicine 1995;11:256–62.
[20] US Preventive Services Task Force. Guide to clinical preventive ser-vices (2nd ed.). Baltimore: Williams & Wilkins; 1996.
[21] Hale KJ. American Academy of Pediatrics Section on Pediatric Den-tistry. Oral health risk assessment timing and establishment of the dental home. Pediatrics 2003;111:1113–6.
[22] Centers for Disease Control and Prevention. Water fluoridation and costs of Medicaid treatment for dental decay—Louisiana, 1995–1996. MMWR Morbidity and Mortality Weekly Report 1999;48:753–7. [23] Lawrence HP, Leake JL. Fluoride gels reduce caries incidence by 20
percent. Evidence-Based Dentistry 1999;11(June).
[24] van Rijkom HM, Truin GJ, van’t Hof MA. A meta-analysis of clinical studies on the caries-inhibiting effect of fluoride gel treatment. Caries Research 1998;32:83–92.
[25] Reisine S, Litt M, Tinanoff N. A biopsychosocial model to predict caries in preschool children. Pediatric Dentistry 1994;16:413–8. [26] Raada M, Espelid I. Caries prevalence in primary teeth as a predictor of
early fissure caries in permanent first molars. Community Dentistry and Oral Epidemiology 1992;20:30–4.
[27] dela Cruz GG, Rozier RG, Slade G. Dental Screening and referral of young children by pediatric primary care providers. Pediatrics 2004;114:e642–5.
