Exploring the cost-effectiveness of child dental caries prevention programmes. Are we comparing apples and oranges?

Abstract

Data sources The following seven databases were searched: PubMed, EMBASE, DARE, NHSEED, HTA, Cost-Effectiveness Analysis Registry and Paediatric Economic Database Evaluation (PEDE).

Study selection The review included trial and model-based economic evaluation studies and the participants included children aged from 0 to 12 years old who were healthy except for having dental caries. Studies of mixed populations of parents and children were included where the data for children were presented separately.

The interventions included were:

• Community-based oral-health education/training programs related to healthy oral habits.

• Screening of children's teeth.

• Supervised toothbrushing technique through the provision of toothbrushes, an appropriate amount of fluoride toothpaste, and topical fluoride.

• Advice on dietary control, such as limitation of sugar or carbohydrates consumption, and enhanced fortified nutrition with an appropriate amount of calcium intake.

• The comparators were situations where the populations were the same as the test group, but were receiving no intervention, or a dissimilar one .The interventions were oral-health promotion programs (OHPPs) implemented by oral-health professionals in the contexts of home visits, telephone calls, healthcare centres and primary schools.

The outcome measures were reductions in the Decayed, Missing, Filled Teeth (DMFT) index for permanent teeth or (dmft) index for deciduous teeth among children and OHPP cost, incremental cost (difference between mean costs of intervention and mean costs of the comparator), and cost-effectiveness analysis (CEA).

Data extraction and synthesis The title, abstract and full text of each study were screened. During the first phase screening of titles and abstracts, irrelevant records were removed. The exclusion criteria were: participant with health-related diseases or aged older than 12 years; interventions other than OHPP (such as implant dentistry or other invasive-dentistry programs); other economic-evaluation outcomes such as cost-benefit, cost-utility or cost-minimisation; authors' opinion (unoriginal records); reviews; and study language other than English. The second-phase screening assessed full texts of the articles using the same eligibility criteria. The risk of bias was assessed using the Drummond 10-item Checklist.

Meta-analysis: The costs were converted to 2015 USA dollars. Data analysis was performed through dichotomous outcomes such as the number of children in the intervention and in the control group, the DMFT index in children, and the OHPP cost. Odds ratios (ORs), effect sizes with 95% confidence interval (CIs) and study weights were estimated from random effects analysis. Forest plots were constructed for each outcome, and chi-square tests used to assess homogeneity, where a p-value of less than 0.1 indicated statistically significant heterogeneity. An I2 test was used to quantify inconsistencies between studies as the percentage of variation across studies. Data synthesis was carried out using narrative demonstration, with a summary of the characteristics of each included study. For quantitative synthesis, a summary of the combined estimation related to the OHPP effect was measured. Three types of subgroup analysis were performed: by the age of the children (age under or equal to 6 years, and age 6-12 years), by publication year (studies published in the last five years, and earlier published studies) and by the country of the study. Egger's regression test and a funnel plot were used to assess and demonstrate publication bias. Publication bias was considered present if the p-value of the Egger test was more than 0.05.

Results 19 full texts were included into qualitative synthesis and eight articles used in quantitative synthesis.

Qualitative synthesis results: With regards to the country of origin, 32% of the studies were conducted in the United Kingdom (n = 6), 26% in Australia (n = 5), 16% in the United States (n = 3). There was also one study from each of the following countries Finland, Ireland, Japan, Nigeria and Singapore. Fifty-two percent (n = 10) were model-based economic evaluation studies and 47% (n = 9) were trial-based economic evaluations. The population of 14 studies were younger than six years of age, while in four studies the children were over the age of six years. In one paper the age of the children was not clearly stated. Just under a half of the papers (47%) were published in the last five years.

The majority of the studies had a low risk of bias (n = 12, 63%) and seven (37%) had a moderate risk of bias. Various outcome measures were used in the included studies: DMFT, average number of dental visits, number of prevented caries teeth, average number of cavity-free months, probability of less cost, caries prevalence, number of specific OHPP visits, quality-adjusted life year, cost-effectiveness ratio, and percentages of not having debris.

Quantitative synthesis results: The overall pooled impact of OHPPs showed that children with tooth decay had 81% lower odds of participating in OHPP (95% CI 61-90%, I2: 98.5%, p = 0) with considerable heterogeneity among studies. OHPPs were successful in reducing financial costs in 97 out of 100 OHPPs (95% CI 89-99%, I2: 99%, p = 0) with considerable heterogeneity among studies. The studies with participants under the age of six years old weighted 71% with an OR of 0.14 (95% CI, 0.05-0.39, I2: 99%). These children had the highest benefit of OHPPs to lower DMFT/S. The studies reporting children aged six years and over weighted 29% with an OR of 0.29 (95% CI, 0.08-1.01, I2: 99%), and these children had no benefit from OHPPs in lowering DMFT/S.

Studies with under-six-year-old participants had an OR of 0.07 (95% CI, 0.02-0.32) revealing no cost-effectiveness effect to reduce OHPP incremental cost, whereas studies reporting children aged six years and older had an OR of 0.0 (95% CI, 0.00-48,704.6). The authors concluded that OHPPs involving the later (older) children were cost-effective in reducing the OHPPs' incremental cost.

Conclusions A comprehensive analysis of the OHPPs confirmed that DMFT could be reduced, hence, lowering the financial burden of dental-care treatment. More effort is needed to manage the allocation of scarce resources, taking into account the economic impact of dental caries on healthcare systems. More studies on caries-prevention programmes among young children in high-, middle- and low-income countries are needed, in order to assess the clinical and financial effectiveness.