Abstract
Objectives
When planning primary oral health care services the cost implications of adopting new intervention practices are important, especially in resource-strapped countries. Although on a trajectory to be phased-out, amalgam remains the standard of care in many countries.
Methods
Adopting a government perspective, this study compared the costs of performing amalgam and ART/high-viscosity glass-ionomer cement (HVGIC) restorations and the consequences of failed restorations over 3 years in suburban Brasilia, Brazil. Cost data were collected prospectively; cost estimates were developed for the study sample and a projection of 1000 single- and 1000 multiple-surface restorations per group. Probabilistic sensitivity analysis was conducted in TreeAge Pro.
Results
Results were mixed. For single-surface restorations, ART/HVGIC will cost US$51 per failure prevented, while for multiple-surface restorations, ART/HVGIC was cost-effective with a savings of US$11 compared to amalgam. Probabilistic sensitivity analysis (Monte Carlo simulation) predicted amalgam would be cost-effective 49.2% of the time compared to HVGIC at 50.6% of the time at a willingness to pay threshold of US$237 per failure prevented. Personnel accounted for more than half the cost burden for both methods; instruments and supplies accounted for about one third. The per restoration cost to replace amalgam with HVGIC ranges from US$1 to a savings of US$0.84.
Conclusion
Replacing amalgam with a high-viscosity glass-ionomer as part of the ART method comes at a minimal increase in cost for governments. Increasing the number of restorations seems to diminish the cost burden.
Clinical significance
ART/HVGIC could be considered a viable alternative to amalgam in primary teeth.
1
Introduction
Following the Minamata Treaty on Mercury recommending a phase down of the use of amalgam, international oral healthcare organizations such as the Federation Dentaire International (FDI) and the World Health Organization (WHO), have called for the development of alternative, biomimetic, restorative materials [ , ]. One such material is a restorative glass-ionomer. Its high-viscosity type, in conjunction with the atraumatic restorative treatment (ART) approach, has been shown to produce restorations whose survival is not statistically significantly different from that of comparable amalgam restorations [ ] and that of composite resin restorations in primary teeth [ ]. A recently published study confirmed this conclusion [ ].
There have been many studies comparing the effectiveness of amalgam and other alternative restoration methods, however, scant research has been conducted to estimate the potential additional costs associated with making the change from amalgam to other restoration materials. A search of the literature for studies that have investigated this topic in the primary dentition did not yield any peer review articles.
While the international phase down of the use of amalgam entered into force 16 August 2017 [ ], it is expected and perhaps accepted, that the rate of change over will vary from country to country – with resource (technical and financial) strapped countries being the last to implement the recommended change. Within countries, the communities that would be the last to have these changes implemented are poor and usually rural communities. In implementing programs geared toward facilitating a transition from amalgam to other materials, such as high-viscosity glass-ionomer cement (HVGIC), dental program developers, will be most interested not only in comparative effectiveness, but also in other factors such as: needs for additional technical and administrative resources and the potential cost of additional equipment needs.
This study presents an analysis of the costs associated with switching from amalgam to high-viscosity glass-ionomer ART restorations for the primary teeth in a randomized clinical trial that compares the effectiveness of amalgam and high-viscosity glass-ionomer ART restorations amongst school age children in Brasilia, Brazil. This cost-effectiveness analysis sought to: (i) determine the mean time spent, for each intervention protocol, in carrying out single- or multiple-surface restorations; (ii) assess the costs and net costs associated with the placement of restorations for each intervention protocol, amalgam and ART/high-viscosity glass-ionomer (ART/HVGIC); (iii) conduct an incremental cost-effectiveness analysis comparing amalgam to high-viscosity glass-ionomer (HVGIC) as a restoration material; and (iv) determine how changes in cost inputs for each method affect the cost of the restorations and the cost-difference of switching from amalgam to high-viscosity glass-ionomer. The study time horizon was 3 yr.
2
Methods
2.1
Community effectiveness trial
The restoration trial was approved by the Research Ethics Committee of the University of Brasília Medical School (reference number 081/2008) and was registered at the Dutch Trial Registration Centre (reference number 1699). Parents and/or caregivers were informed in writing about the investigation and treatments. Children whose parents or caregivers completed and signed the consent forms were included in the study. Details of the study were published previously [ ] and a summary of it is provided below.
The randomized clinical trial used a parallel study design and was carried out amongst 6- and 7-year-old children from a low socioeconomic area of Brasilia, the capital of Brazil. Inclusion criteria were: good general health and having at least two cavitated dentine carious lesions in vital pain-free primary molars assessed according to the ICDAS II system. The study groups consisted of three protocols: conventional restorative treatment (CRT), where amalgam restorations were applied; ART, high-viscosity glass-ionomer restorations (HVGIC); and, Ultra-Conservative Treatment (UCT) which included the cleaning of medium and large cavities to maintain them plaque free under the daily supervision of one of the dental assistants during school time, and restoring small cavities that are difficult to clean according to ART.
Treatments were performed on the grounds of six primary schools by three trained pedodontists, assisted by trained dental assistants. The CRT protocol was performed in a dental clinic that was available in two schools, using rotary equipment and the cavities were restored with a high-copper non-gamma2 spherical and lathe cut amalgam (Permite Regular set ® ; SDI, Melbourne, Australia). The ART restorations were performed in the two ART schools and the two UCT schools, in rooms where each participating child lay on a mobile bed with artificial light, according to the instructions by Frencken et al. [ ]. The restorative material used was Ketac Molar Easymix ® (3 M ESPE, Seefeld, Germany).
Two trained, calibrated and independent evaluators (dentists) assessed the restorations in the primary teeth according to the ART restoration criteria [ ] on the school premises after 6 months, 1, 2 and 3 years. Secondary caries was defined as an obvious cavitated dentine carious lesion. Battery-illuminated dental mirrors (Kudos ® , Hong Kong, China), CPITN probe Golgran) and compressed air aided the evaluation. Re-restoration of primary molars was carried out for cases where secondary dentine carious lesions were found. During the study period, restored teeth that caused pain and were not considered viable were extracted. For the most part, in cases of mechanical failure if part of the restoration remained, the teeth were considered viable and if the child was not experiencing pain, the restorations were left alone and the teeth exfoliated naturally.
Because the researchers did not find a statistically significant difference in cumulative survival percentages between ART/HGVIC restorations cleaned under the UCT supervised tooth brushing protocol and the ART group (p = 0.16), the two groups were combined into one ART/HGVIC group [ ] in the effectiveness study. This study compared the cumulative survival percentages of restorations in the primary teeth using amalgam (CRT) and a high-viscosity glass-ionomer cement (HVGIC; applied in the ART and UCT groups) in primary molars over a period of three years. The baseline intervention and results over three years are presented in Table 1 .
| All restorations | Single-surface | Multiple-surface | ||||
|---|---|---|---|---|---|---|
| Amalgam | ART/HVGIC | Amalgam | ART/HVGIC | Amalgam | ART/HVGIC | |
| Children (N) | 126 | 154 | ||||
| Schools (N) | 2 | 4 | ||||
| Restorations by type (N) | 364 | 386 | 105 | 116 | 259 | 270 |
| Mechanical failure (N) | 79 | 105 | 4 | 8 | 75 | 97 |
| Secondary dentine carious lesion (N) | 3 | 7 | 1 | 2 | 2 | 5 |
| Extractions | 19 | 15 | 1 | 2 | 18 | 13 |
| Survival at 3 yr % (SE) | 72.6 (2.9) | 66.8 (3.1) | 93.4 (3.3) | 90.1 (3.0) | 64.7 (3.6) | 56.4 (3.9) |
| Probability of failure (95% CI) | 0.2749 | 0.332 | 0.066 | 0.099 | 0.353 | 0.436 |
| (0.217–0.330) | (0.271–0.393) | (0.001–0.131) | (0.04–0.158) | (0.282–0.424) | (0.360–0.512) | |
b ART/HVGIC = atraumatic restorative treatment/high-viscosity glass-ionomer cement.
The cumulative survival percentage for all amalgam restorations in the effectiveness study was 72.6% compared to 66.8% for all ART/HGVIC restorations after 3 years; this difference was not statistically significant [ ]. After three years, the single-surface restorations under both treatment protocols had higher survival percentages than the multiple-surface restorations. Ten secondary dentine carious lesions developed under both treatment protocols; in each case more than half occurred in multiple-surface restorations. The total number of extractions under both protocols was 34; 19 for amalgam and 15 for ART/HVGIC.
2.2
Cost-effectiveness study design
The study adopts a government program perspective and seeks to provide information on the inputs, their effects and the costs that go into the implementation of each dental caries prevention and care intervention protocol. This information can enhance each entity’s capacity to maximize the use of available resources in planning and implementing the intervention protocol that is most feasible to them.
The study outcome was evaluated in two ways: counting the total number of teeth with failed restorations in each group, and counting only the number of restored teeth that developed a cavitated carious lesion or were extracted over the 3yr period. The adverse effects costs include the cost of re-restoring teeth where secondary dentine carious lesions developed, as well as the costs of extractions. Restored teeth that exfoliated naturally and those that had mechanical failures but were judged viable were not included in the second analysis.
Outcomes and costs, including the costs of adverse events, i.e., re-restorations and extractions estimated in the study, were discounted at a rate of 3% [ ]. Cost data were recorded in Brazilian reais (R), adjusted for inflation using the World Bank GDP inflation deflator and then converted into 2012 US dollars [ ].
The intent of this study is to inform dental practitioners and oral health program planners and policy makers and enhance their ability to evaluate the broader consequences and opportunities of replacing amalgam with the alternative of high-viscosity glass-ionomer cement for the care of dental caries in the primary and permanent dentition of young children as they go through an important period in their oral health development. To that end, the analytic study was expanded beyond the study population used in the effectiveness study trial, to include analysis of a projection sample of 1000. This will provide information for making the change on a larger scale than for the study group.
2.3
Data collection: personnel time
Personnel time is a fundamental element for estimating the costs of interventions. Data were collected in two ways to estimate personnel time during the intervention. In the first instance the assistant recorded the actual start and end time for each restoration performed in every intervention session. In the second, a systematic method was used to sample several sessions to obtain estimates for the restoration times as well as other activities not captured in the recording of the actual time method.
Actual time method: The assistant recorded the start and end time for placing each restoration. The start time began when the pedodontist’s instruments were lifted to start the restoration and ended the moment they finished and put down the instruments.
Systematic sampling method: The systematic sampling method used in this study is referred to as activity sampling [ ] time as it serves to capture, as best possible, all activities that occurred during the entire session, in addition to restoration placement.
Of the 66 4-h sessions that were covered in the actual time data recording, only 35 were included in activity sampling data collection. The method was implemented by sampling the 4-h treatment sessions in 15 m intervals using a countdown timer. In an effort to capture activities at varying times during the sessions, the timer was set at a different time after the session began (usually 8:00 am) each day. The number of minutes after 8:00 am was determined by the last number in the identification number of the first child treated that day. The timer sounded at each 15 m interval and the assistant recorded the code that best described the pedodontist’s activity at that moment. For example, if the pedodontist was performing a restoration in the amalgam group during the second interval, the assistant would write the code for that activity in the space for that time interval.
Clinical activities included exam and diagnosis, placing a restoration, performing an extraction, etc. Complementary activities included preparing the clinical area, instrument preparation, awaiting a patient or talk with patient to calm their nerves. Patient absence, equipment failure, dentist absence and other reasons such as coffee break were recorded as other, non-clinical activities. Activity sampling time provided the capacity to better understand the wider range of activities that occured during the session for each of the intervention methods.
2.4
Cost data collection
Cost data were collected prospectively. This is a financial analysis, thus all costs were recorded, regardless of whether items were purchased or donated. A Microsoft Excel (2013) data collection instrument was created for the study and given to the principal investigators to fill out. The quantities and costs of instruments and supplies used were recorded by group. Data on the costs of equipment, personnel salaries and the pedodontists’ transportation costs were obtained from the records of the University of Brasilia and apportioned by group based on the number of interventions performed in each group. Capital equipment and instrument and materials costs were obtained from actual expenditures; the costs of the dental equipment were annualized. In the case of capital items that were not acquired, the replacement cost was used. The costs of items acquired outside of Brazil such as the ART instruments, which were donated, were priced in euros and converted to reais .
Anesthesia was used in the CRT and ART/HVGIC interventions and the cost was calculated separately, utilizing information on the costs of the supplies and the number of times it was used for each group.
2.5
Data recording and analysis
The data from recording the individual procedure times (actual time method) for both single- and multiple-surface restorations during each of the treatment sessions were included in the study’s main database and analyzed using SAS to produce mean times for performing for single- and multiple-surface restorations and their standard errors (SEs). In addition, the times of both the single- and multiple-surface restorations were combined to estimate an overall mean time per restoration for each intervention method.
The activity sampling time data (activity sampling time method) were entered into an ACCESS database by one person and verified by a second. The data were transposed into and analyzed in SAS ® 9.3 (SAS Institute, Cary, NC, USA) to obtain mean times per session. The proportion of time consumed by clinical and other, non-clinical activities in each session, for each intervention method was calculated. The time for performing restorations was divided by the number of restorations per session to obtain a proportion. This proportion was used to calculate the time it took to perform each restoration.
The mean number of minutes it took to perform a single- or multiple-restoration for each intervention group was converted to hours. The pedodontists worked in 4-h sessions, approximately 66, representing about 264 h. The proportion of time dedicated to each activity by group, was used to approximate the number of hours for each intervention group; close to 121 h for CRT, 77 h for ART, and 66 h for the UCT group. The difference between total treatment time and the total number of hours per method was apportioned to each procedure to obtain an overall per unit cost.
To calculate the proportion of time it took to complete the baseline intervention, the time the operators took to complete the intervention was estimated to be 4.5% of a year. This proportion was applied to the cost of the annualized equipment for sample costs.
2.6
Cost and incremental cost-effectiveness analysis
Once the mean times were obtained from the actual time data, the value of personnel time per minute was calculated by multiplying the value of salary time per minute by the number of minutes to produce a cost per single- and multiple-surface restoration for the amalgam intervention method and for the combined ART/HVGIC intervention method. The value of personnel time per minute was calculated using salary data. The costs of single- and multiple-surface restorations performed for each treatment group were averaged to calculate one cost for all amalgam and one cost for all ART/HVGIC restorations. Because the activity sampling time data only captured whether an amalgam or ART/HVGIC restoration was done, and not whether they were single- or multiple-surface restorations, the costs per restoration for the activity sampling data are averages for all amalgam and ART/HVGIC restorations.
For the incremental cost-effectiveness analysis, the differences in the costs and outcomes between the amalgam and ART/HVGIC study groups are evaluated in a ratio where the difference in costs is divided by the difference in outcomes [ , ]. As ART/HVGIC is one of the candidate approaches to replace amalgam, the ratio provides information about what additional costs will be incurred using ART/HVGIC to prevent one additional failure.
2.7
Projection sample
The results for the sample are presented alongside a projection of production of 1000 single-surface and 1000 multiple-surface restorations for each method. To create the projection, inputs such as personnel time, instruments and supplies were increased at the same rate as in the original sample. The production of a much larger number of restorations in the projection would be expected to take longer than 4.5% of a year. The estimate was closer to about 25% of a year and was applied to the projection.
2.8
Sensitivity analyses
The impact of increasing the costs of personnel time alone and in combination with the costs of instruments and supplies (5% to 20% change) on the net cost per restoration was explored in sensitivity analyses. To increase the cost, the proportion of the total cost of a restoration represented by a particular input, 57% in the case of personnel, was subtracted from the total cost per restoration and multiplied by the increase proportion and added to the total. Changes in the cost of HVGIC (25% increase and 25% decrease) and their effect on the cost difference between the sample size groups were also evaluated.
A probabilistic sensitivity analysis (PSA) was conducted to create a cost-effectiveness acceptability curve (CEAC) comparing the cost-effectiveness of the two strategies given a willingness to pay (WTP) threshold for a gain in preventing one additional failed restoration. The WTP threshold chosen originated in 1993 World Bank recommendations for a minimum health package where interventions valued at US$150 per DALY or disability adjusted life year, were considered acceptable for low-income countries [ ]. The DALY has been used by nations, international organizations (such as the World Bank and the WHO), as well as academics, to measure health status globally.
To conduct the PSA, a model comparing amalgam and ART/HVGIC was built in TreeAge Pro 2017 [ ] using data derived from the study. A Monte Carlo simulation of 10,000 bootstrap re-samples was run. Lognormal distributions were used for costs and effectiveness; beta distribution were applied to probabilities. The WTP threshold of US$150 in 1993 was adjusted to US$237 in 2012.
2
Methods
2.1
Community effectiveness trial
The restoration trial was approved by the Research Ethics Committee of the University of Brasília Medical School (reference number 081/2008) and was registered at the Dutch Trial Registration Centre (reference number 1699). Parents and/or caregivers were informed in writing about the investigation and treatments. Children whose parents or caregivers completed and signed the consent forms were included in the study. Details of the study were published previously [ ] and a summary of it is provided below.
The randomized clinical trial used a parallel study design and was carried out amongst 6- and 7-year-old children from a low socioeconomic area of Brasilia, the capital of Brazil. Inclusion criteria were: good general health and having at least two cavitated dentine carious lesions in vital pain-free primary molars assessed according to the ICDAS II system. The study groups consisted of three protocols: conventional restorative treatment (CRT), where amalgam restorations were applied; ART, high-viscosity glass-ionomer restorations (HVGIC); and, Ultra-Conservative Treatment (UCT) which included the cleaning of medium and large cavities to maintain them plaque free under the daily supervision of one of the dental assistants during school time, and restoring small cavities that are difficult to clean according to ART.
Treatments were performed on the grounds of six primary schools by three trained pedodontists, assisted by trained dental assistants. The CRT protocol was performed in a dental clinic that was available in two schools, using rotary equipment and the cavities were restored with a high-copper non-gamma2 spherical and lathe cut amalgam (Permite Regular set ® ; SDI, Melbourne, Australia). The ART restorations were performed in the two ART schools and the two UCT schools, in rooms where each participating child lay on a mobile bed with artificial light, according to the instructions by Frencken et al. [ ]. The restorative material used was Ketac Molar Easymix ® (3 M ESPE, Seefeld, Germany).
Two trained, calibrated and independent evaluators (dentists) assessed the restorations in the primary teeth according to the ART restoration criteria [ ] on the school premises after 6 months, 1, 2 and 3 years. Secondary caries was defined as an obvious cavitated dentine carious lesion. Battery-illuminated dental mirrors (Kudos ® , Hong Kong, China), CPITN probe Golgran) and compressed air aided the evaluation. Re-restoration of primary molars was carried out for cases where secondary dentine carious lesions were found. During the study period, restored teeth that caused pain and were not considered viable were extracted. For the most part, in cases of mechanical failure if part of the restoration remained, the teeth were considered viable and if the child was not experiencing pain, the restorations were left alone and the teeth exfoliated naturally.
Because the researchers did not find a statistically significant difference in cumulative survival percentages between ART/HGVIC restorations cleaned under the UCT supervised tooth brushing protocol and the ART group (p = 0.16), the two groups were combined into one ART/HGVIC group [ ] in the effectiveness study. This study compared the cumulative survival percentages of restorations in the primary teeth using amalgam (CRT) and a high-viscosity glass-ionomer cement (HVGIC; applied in the ART and UCT groups) in primary molars over a period of three years. The baseline intervention and results over three years are presented in Table 1 .
Stay updated, free dental videos. Join our Telegram channel
VIDEdental - Online dental courses
