Inequalities in oral health: estimating the longitudinal economic burden of dental caries by deprivation status in six countries | BMC Public Health
A simplified conceptual framework was developed to demonstrate the progression of dental caries from a healthy tooth to an unsalvageable carious tooth that requires extraction. The stages in the progression include the development of an initial white spot lesion caused by demineralisation followed by established more severe stages of dental caries. If left to progress, dental caries can progress to involve the dental pulp (root canal system) and the tooth may eventually become unsalvageable (Fig. 1). The framework also includes interventions for primary prevention at various stages to prevent the development of or limit the progression of caries (secondary prevention). For healthy teeth or those with carious white spot lesions, maintaining good oral hygiene by brushing with fluoridated toothpaste and applying topical fluorides (e.g., fluoride varnish or SDF) or consuming fluoridated products (e.g. fluoridated water or fluoridated salt) are effective preventive measures. Dental fissure sealants effectively prevent caries in healthy teeth or halt the progression of initial carious lesions by sealing the deep grooves and fissures on chewing surfaces, which are prone to decay due to microbial plaque and food accumulation. After established dental caries sets in and forms a cavity, the dental caries process cannot be reversed, and management moves to a “restorative/reparative cycle”. Initially, the tooth can be restored using fillings. If the dental caries progresses to compromise the dental pulp, the tooth will require root canal treatment and/or a crown. The most extensive stages of dental caries may result in an unsalvageable tooth, necessitating extraction followed by replacement, frequently with a dental implant, if affordable, feasible and clinically appropriate.
Based on this conceptual framework, we developed a simplified approach to the dental caries clinical care pathway, based on available data, to enable an estimation of the direct costs of managing dental caries (Fig. 2) [24,25,26]. We adhered to the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) reporting guidelines to ensure comprehensive and transparent reporting of our analysis [27].
Data inputs
Population data
Data from the World Population Prospects (WPP, 27th edition) was used to determine the population size and age structure for each country, broken down into 5-year age groups [28]. The latest population size estimates for the 10–14 year age group were used as the baseline, and projections were made for the population size as the cohort progressed into the 60–64 year age group, based on general population mortality. The death rate assumptions were derived from the WHO data on probability of dying at a specific age [29]. The formula applied was as follows:
$$Prob\_1yr=1- exp(ln\frac(1 – Prob\_5yr)5)$$
where:
Prob_5yr: The probability of dying between 5-year age cohorts.
Prob_1yr: The probability of dying at a specific age.
exp: exponential.
ln: natural logarithm.
Disaggregation by deprivation quintiles
The English Index of Multiple Deprivation IMD 2019 was used to disaggregate the population in each 5-year age cohort into deprivation quintiles. The quintiles were classified as least deprived, second least deprived, middle deprived, second most deprived and most deprived [30]. Following the relative ranking system of the IMD 2019, we assigned 20% of the population in each age cohort into each deprivation group.
Dental caries experience data
We used the commonly used population-based measure of DMFT to quantify the current and past caries experience. DMFT refers to the number of decayed, missing and filled permanent teeth and is calculated as the sum of all decayed, missing and filled teeth among individuals in a specific age group divided by the total population in that age group [31, 32].
DMFT scores for 12-year-olds were sourced from national oral health surveys from Brazil, France, Germany, Indonesia, Italy, and the UK and used as the baseline DMFT to quantify dental caries experience (see Table 1) [33,34,35,36,37,38]. The UK data is extrapolated from a national survey of England, Wales and Northern Ireland. The total caries experience within the age group was calculated by multiplying the average DMFT scores by the population size in that specific age group. We used separate Welsh and German data among 12-year-olds that showed a similar 0.4 difference in DMFT scores between the least and most deprived groups when stratified by socioeconomic status [37, 39]. This difference of 0.4 between deprivation groups was extrapolated to all the studied countries due to the lack of other country specific data.
The average DMFT data reported in each country’s national oral health survey was assumed to apply to the middle-deprived quintile. We derived conversion factors from the Welsh data and adjusted them according to the country’s Gini coefficient relative to the UK, which were used to adjust the DMFT scores for the other quintiles (see Table 2) [40].
The calculation used was:
$$DMFT\_1_i=DMFT\_5_i\times (\fracDMFT\_5UKDMFT\_1UK)\times (\fracGINI_iGINI\_UK)$$
where:
DMFT_1i is the DMFT score for the most deprived quintile in country i;
DMFT_5ii is the DMFT score for the least deprived quintile in country i;
DMFT_1UK is the DMFT score for the most deprived quintile in the UK;
DMFT_5UK is the DMFT score for the least deprived quintile in the UK;
GINIi is the Gini coefficient for country i;
GINI_UK is the Gini coefficient for the UK.
Dental caries progression
Data from a large systematic review and meta-analysis reported an unadjusted annual increase in DMFT scores of + 0.18 with a lower progression rate of + 0.07 after adjusting for preventive interventions. The preventive interventions varied across the studies and included school-based preventive education programs, the use of fluoride-containing lozenges, mouthwash, or toothpaste, the application of topical fluorides, or the use of sugar substitutes like xylitol or poly alcohol [41]. Based on these data, an annual increment in DMFT score of + 0.18 was applied to the middle deprivation cohort and that of + 0.07 to the least deprived cohort. We assumed that the least deprived cohort with best access to interventions would have the lowest annual progression rate. Progression rates were then assumed to evolve across income brackets linearly to arrive at the rate of annual progression of DMFT:
-
Least deprived quintile: + 0.07
-
Second least deprived quintile: + 0.125
-
Middle deprived quintile: + 0.18
-
Second most deprived quintile: + 0.235
-
Most deprived quintile: + 0.29
We assumed that the progression rate in dental caries remains the same across countries and an individual’s lifetime, irrespective of the baseline caries experience.
The overall annual progression rate in DMFT was disaggregated across decayed, filled and missing teeth, based on the progression of decayed, filled and missing teeth values reported in The Dunedin Multidisciplinary Health and Development Study [42]. We assumed that the distribution of the values reported in the Dunedin study was representative of the middle-deprived quintile. We then adjusted the distribution of individual DMFT values to other quintiles relative to the middle quintile, employing assumptions related to the likelihood of receiving a filling versus extraction across deprivation groups. For instance, individuals in the most deprived group are more likely to receive an extraction rather than preventive management or restorative treatments such as fillings, bridges and implants, even in countries with publicly funded dental care, such as the UK [43, 44]. The assumptions used in the distribution of DMFT progression between age cohorts across deprivation quintiles are detailed in Supplementary Table 1 in Supplementary file 1.
Direct costs of managing dental caries
An estimation of the direct costs of managing dental caries was derived through the triangulation of information gathered from country experts and cost data sourced online. The costs associated with dental caries management per tooth are detailed in Supplementary Table 2 in Supplementary file 1. Given the variation in the provision of subsidised care across countries and the lack of information regarding healthcare costs in the public sector, private treatment costs in each country were used as a proxy to estimate the direct costs of dental caries.
An increase in the ‘Fillings’ component of the DMFT by 1 implies that the patient received a new filling. We assumed the need for re-restoration of a filling every 10 years, based on a conservative estimate of the median survival rate of composite fillings [45, 46]. Of those who received a filling, 9.3% were assumed to have had a root canal treatment, based on data reported in a systematic review [46]. Among patients receiving a root canal treatment, a proportion were assumed to have also received a crown. Both root canal and crown interventions were weighted such that they were more common among the least deprived, owing to the cost of the procedures. The assumptions applied to the provision of root canals and crowns across deprivation groups are detailed in Supplementary Table 3 in Supplementary file 1.
An increase in the ‘Missing’ component of DMFT by a value of 1 implies that the patient underwent a single tooth extraction. Following extraction, we assumed that less deprived groups were more likely to receive single tooth implants. Besides dental implants, alternative interventions such as dental bridges or single-tooth removable partial dentures can be used to replace a missing tooth. We included an alternative replacement in the analysis in a treatment-agnostic approach to account for various treatment options that were lower cost to a dental implant, and were more likely to occur in the more deprived groups. To provide a credible range for the estimates and to acknowledge the likelihood that not all patients will receive a replacement for a missing tooth, 60% and 30% of the most and second most deprived groups, respectively, do not receive any replacement for a missing tooth in this analysis. The assumptions applied to the provision of replacements across deprivation groups are detailed in Supplementary Table 4 in Supplementary file 1.
Direct costs for dental caries in each age group were calculated as the cost of treatment per tooth multiplied by the number of teeth requiring the treatment multiplied by the percentage of each deprivation group assumed to have the treatment. Based on the UK’s National Institute for Health and Care Excellence (NICE) recommendation, a 3.5% discount rate for future costs was applied to the calculation [47].
Scenario analysis
Two scenario analyses were performed to assess the decrease in per capita costs between 12–65 years of age based on the following interventions:
-
Scenario 1 – Application of universal interventions with a decrease in caries progression rates by 30% across each deprivation quintile. The 30% decrease in dental caries progression rate was considered conservative, given that the majority of dental caries is preventable via a range of effective public health interventions, such as community water fluoridation, salt fluoridation, reduced sugar consumption (via the implementation of, for example, SSB taxes or enhanced food labelling) and twice daily brushing with fluoridated toothpaste and fluoride varnish (frequency dependent on risk of caries) [48].
-
Scenario 2 – A ‘levelling-up’, or proportionate universalism approach, with the scale of prevention and management interventions proportional to the degree of need across deprivation quintiles. In this scenario, the dental caries progression rate of the least deprived quintile was applied across all quintiles.
Model validation
Consistency checks, face validity assessments and a sensitivity analysis were conducted as part of the model’s internal validation. Consistency checks involved amending data and model parameters such as baseline DMFT data, direct costs and probabilities for tooth health transition states were amended to ensure that the model behaves logically, resulting in the expected changes in outcomes. Face validity was ensured by engaging experts in health economics, epidemiology, and oral health experts to review the model to ensure that the assumptions, data inputs, and overall structure were reasonable and reflected current knowledge. We conducted two sensitivity analyses. Sensitivity analysis A involved adjusting the discount rate from 3.5% to 5%, reflecting the upper limit of discount rates applied in a systematic review of modelling techniques for the economic evaluation of dental caries [49]. Sensitivity analysis B involved adjusting treatment costs to provide a range to account for a variation in treatment costs. We estimate overall direct costs if treatment costs were 10% lower and 10% higher versus the treatment costs applied in the primary analysis.
link