Cost-effectiveness analysis of a school-based dental caries prevention program using fluoridated milk in Bangkok, Thailand

The use of fluorides is recognized as one of the most successful measures for the prevention of disease in the history of public health [1]. Fluoride can be delivered to individuals as a dental preventive measure through a variety of mechanisms. The most commonly used vehicles for community-based fluoridation programs are drinking water and salt fluoridation. Other mechanisms include school water fluoridation, fluoride mouthrinses, topical application of fluoride solution gels and milk fluoridation [2‐5].

In Thailand, the oral health promoting school program is well established. The program consists of oral health education and health promotion activities, regular oral health checkups, pit and fissure sealant programs and fluoride varnish [6]. The program also includes adequate fluoride exposure through supervised tooth brushing using fluoride toothpaste and in some schools, consumption of fluoridated milk. The first milk-fluoridation scheme was launched in Bangkok in 2000 and this has now been extended to eleven provinces, out of 77, involving more than one million children aged 3 to 12 years. Every child under this scheme receives, at no charge, 200 ml of fresh pasteurized milk per day (on school days) containing 0.5 mg F (sodium fluoride). Additionally, on holidays (i.e., 60 days), children receive containers of fluoridated UHT milk. Milk distribution is the responsibility of the local authority (e.g., Bangkok Metropolitan Administration [BMA]). Fluoridated milk is prepared by 20 dairy companies contracted to the local authority (e.g., BMA).

In the last few years, economic evaluations (EE) have become increasingly important in decision-making in health. Despite this, apart from EE of water fluoridation and dental sealants, the use and application of EE in dentistry remains limited [7, 8]. Furthermore, except for studies carried out on fluoridated milk programs in Chile [9, 10], there have been no true economic evaluations of dental caries prevention programs using milk as the vehicle, which reflect Thailand’s conditions.

The purpose of this evaluation was to assess the cost-effectiveness, from a societal viewpoint, of a dental caries prevention program using fluoridated milk in the city of Bangkok, Thailand, for children from 6 to 12 years old, versus non-intervention. The analysis calculated the per-unit and annual costs of the milk-fluoridation program under Thai conditions, and the cost of this program per child.

The form of economic evaluation used in this study was cost-effectiveness analysis (CEA). In CEA, costs of alternative programs are measured as economic costs, and outcomes are measured in units of effectiveness [11]. This study used the “model” format for CEA [12]. This is an aggregate model with no interactions [13]. A decision tree examines populations undergoing different events (arms of the tree) and their associated costs and consequences [12]. Modelling was based, as far as possible, on real data and real values. When this was not possible, we used realistic assumptions based on secondary analysis of data collected as part of the regular practice of existing milk-fluoridation schemes operating in Thailand.

All costs were assessed from a societal perspective, using a 6-year time horizon. A societal viewpoint requires measurement of all costs and benefits to the community, no matter to whom they accrue [14, 15]. The Consolidated Health Economic Evolution Reporting Standards guideline was followed for reporting this economic evaluation [16].

Calculations based on data provided from the Royal Chitralada Project indicate that the undiscounted investment necessary to produce both Pasteurized fluoridated-milk and UHT fluoridated-milk is THB(2011) 386,019.00 (See Table 1).

Production costs. The daily milk consumption in children is 200 ml of pasteurized milk (bag) per day during school days and one carton per day of UHT milk during holidays. Producing 75,000 bags for 200 school days (@ THB 0.00556 per bag) would cost THB 43,168 per year. To produce 75,000 cartons for 60 vacation days (@ THB 0.00678 per carton) would cost THB 16,430 per year. Thus, the total undiscounted annual cost to add fluoride to milk distributed by the BMA would be THB 113,892 per year.

Milk quality control analysis and monitoring. To check for the consistent provision of appropriate concentrations of NaF, milk samples are delivered from schools to the Ministry of Health, at a cost of THB 1600 a year (THB 200 per trip × 8 times a year). There is a cost of THB 30 per sample for the fluoride analysis at the Ministry of Health. Thus, for six samples per dairy, eight times a year, the cost would be THB 8640.

The laboratory technician’s costs to conduct the analyses at the Ministry of Health would be THB 9138 (THB 138.46/h; 66 h per year). The cost of the urinary analysis would be THB 5467.

Promotion, reports, and management. Assuming a wage of THB 40,000 a month for a fulltime equivalent position, the coordinator’s salary would be THB 48,000 a year. The coordinator AA’s salary would be THB 28,800 a year.

The office space rent would be THB 8100 a year (THB 39 per hour, 4 h a week, 52 weeks a year). Additionally, it is estimated that services including electricity, water, internet and phone would cost THB 1620.

A dental nurse would make 3 h visits to each school once a year to answer questions, promote and explain the program in 435 BMA schools. In order to visit all the schools the dental nurse would have to work 1305 h (@THB 173 per hour), equivalent to THB 225,865 a year.

Consent forms (@THB 0.5 per form) were printed in year 1 only. Assuming 75,000 children would need a form filled-in and signed by their parents or guardians, the cost would be THB 37,500.

Salaries for 600 professionals participating in an 8-h teacher meeting would be THB 858,462. The cost of organising the teachers’ meeting was THB 300,000. This meeting took place at commencement and in Year 4 of the scheme.

The report and presentation to the Princess every year had a cost of THB 1667 a year. The cost of the promotion booklet, to be prepared every 2 years, would be THB 33,334.

The once-a-year inspection of the dairies was done by three dental officers who take 16 h per dairy, at a salary of THB 230.77 per hour, per person. The total cost of inspecting the 5 dairies for 16 h a year each was THB 66,462. Thus, the expected undiscounted annual cost of Promotion, Reports and Management would be THB 1,609,810.

Cost of the Program (See Table 2). Assuming that the annual cost of the fluoridated milk scheme remained constant over time, the value for 2011 was applied retrospectively to all previous years and deflated at 3.25% per year. The total discounted cost of the program offered to 75,000 children for 6 years was THB 5,345,048. That is, the cost was THB 71.26 per child for a six-year program, or THB 11.88 per child per year. This cost included the fluoridation of the milk, tests, transportation, promotion and administration.

The largest cost component was promotion, reports, and management: 75.38% of the total annual fluoridated-milk discounted cost. Production costs and milk quality control analysis and monitoring costs accounted for 141.33% and 3.13%, respectively. Initial investment accounted for 7.16% of the total discounted cost.

The estimated discounted cost of dental treatment over the 6 years for the intervention and control groups is shown in Table 3. Costs were about 33.5% higher in the control group (THB 40,549,920 or THB 90.1 per child per year) compared with the intervention group (THB 26,927,693 or THB 59.8 per child per year).

When the costs of operating the program and the costs of dental treatment in the test community were compared with the costs in the non-fluoridated community, it was found that a public investment of THB 5,345,048 (or THB 11.88 per child per year) over 6 years resulted in a saving of THB 8,177,179 in societal costs attributable to the preventive program over the 6 years. This investment would result in a reduction of 40,500 DMFS avoided within a community with a childhood population of 75.000 [DMFS avoided: 75,000 x (− 0.54)] (See Table 4). Thus, investing in the program would not only result in a reduction in disease, but in a net financial saving to the community.

Sensitivity analysis would result in net savings ranging from THB 18,597,122 to THB 7,920,711 after 6 years for the intervention group, when compared to the non-fluoridated group (See Table 5). This range of variability was caused by uncertainties of the effectiveness of the milk-fluoridation scheme. As expected, the most favourable result was gained by using the best-case scenario of the effectiveness assumption, that is, 58% caries reduction. Conversely, the least favourable result was found using the worst-case scenario of effectiveness (i.e., 31%).

Our primary EE estimated that if a dental caries prevention program using fluoridated-milk from the BMA were available for 75,000 children aged from 6 to 12 years, the net saving from a societal perspective in dental treatment would total THB 8,177,179 over 6 years. These societal savings would be achieved at a yearly cost to a government-sponsored agency of THB 71.26 per child for a 6 year program, or THB 11.88 per child per year. That is, spending THB 11.88 per year per child would save THB 18.17 per child per year, for a 6 year program. From the family perspective, the net present value of savings, due to savings in production losses, transportation, and other uses of time, would be THB 2.55 {([40,549,920–26,927,693]/75,000)/71.26} for each THB spent on the milk-fluoridation program.

Assumptions and uncertainties were explored in the sensitivity analysis to test the robustness of the results. The milk-fluoridation alternative remained dominant even when more stringent and pessimistic assumptions were evaluated. The sensitivity analysis yielded confidence intervals (THB 18,597,122 to THB 7,920,711) resulting from the ranges of variation of two key input parameters. Thus, even with a lower confidence limit in dental caries reduction, the milk-fluoridation program would still be a cost-effective health intervention compared to current oral health care.

Compared with other milk-fluoridation programs [9, 10], the present study appears to be less cost-effective, but still represents an efficient use of resources. This somewhat lower range was highly influenced by the fixed administrative costs included in the total costs. In particular, the cost of promotion and teacher meetings accounted for 79.81% of the promotion, reports and management costs category, over the total period of the scheme. Additionally, an environment of low caries prevalence (i.e., DMFS 1.1 to 1.6) [24] limited the cost-effectiveness of the fluorides program [25]. An EE in areas where communities are at higher risk of dental caries would provide a more cost-effective result.

The results of this research are presented in incremental costs and incremental benefits (i.e., THB 8,177,179 of cost-savings and 40,500 DMFS avoided). However, there are many ways in which they do not account for all possible costs and benefits of a program. Thus, this analysis can be considered conservative. The use of a number of assumptions might have underestimated the savings of the program [26]. For example, only the costs of the initial treatment were considered. Additional costs due to potentially costlier treatments (e.g., cost of space retainers after early extraction of deciduous teeth, or pulpal therapies) were not considered. On the other hand, the study assumed that the decay component of the DMFS index was restored in the year of the increment. There is no guarantee that the necessary treatment would be undertaken.

Pain, infection and tooth loss are the most common consequences of oral disease. The contribution of a healthy dentition to quality of life was not quantified, but is probably highly valued by most people [25]. Moreover, there are many non-health benefits of a preventive approach which were not considered. This analysis ignores issues around social justice and a decrease of inequalities in health. These issues are complex and difficult to assess within the constraints of a research framework, but are not outside the realm of EE [26]. Their inclusion would provide a closer picture of the advantages in any preventive program.

The current study represents only the first step in determining whether a specific oral health care program is an efficient use of scarce community resources. Although appropriate data sources were available for most assumptions used, follow-up studies are required to avoid assumptions. Thus, future studies need to be done under more generalizable conditions and it would also be appropriate to prospectively collect information regarding actual treatment costs along with those of the intervention. Such an economic evaluation would decrease the need to undertake sensitivity analyses [11].

The purpose of this evaluation was to assess the cost-effectiveness, from a societal viewpoint, of a dental caries prevention program using fluoridated milk in the city of Bangkok, Thailand, for children from 6 to 12 years old, versus non-intervention. While the analysis has inherent limitations due to its dependence on a range of assumptions, the results provide important information for further economic evaluation, and may help local program managers to determine the average cost per child and per DMFS avoided of providing fluoridate milk, under the conditions prevailing in Thailand. Since the analysis indicated consistent cost-savings and health gains, in terms of dental caries averted, a reasonable conclusion is that the economic and human benefits to society are significant.