| Sign In to gain access to subscriptions and/or personal tools. |
A Cost-effectiveness Analysis of Implant Overdentures
1 Clinic for Reconstructive Dentistry and TMJ Disorders, University of Basel, Hebelstr. 3, CH-4056 Basel, Switzerland; and Correspondence: * corresponding author, N.Zitzmann{at}unibas.ch
Placement of dental implants may improve the retention and stability of complete dentures in edentulous patients. Treatment costs, however, substantially increase with implant treatment. We therefore performed a stochastic cost-effectiveness analysis, comparing implant-supported over-denture prostheses (4 implants), implant-retained overdentures (2 implants), and complete dentures, from the patient’s perspective in Switzerland, to assess whether implant treatment in the mandible represents value for money spent. Twenty patients were included in each treatment group and were followed up for three years. Health outcomes were expressed in Quality-adjusted Prosthesis Years, and dental health care costs and time costs were recorded in year 2000 Swiss Francs (CHF 100 = US$61). The cost per Quality-adjusted Prosthesis Year gained for implant treatment was CHF 9100 (2 implants) and CHF 19,800 (4 implants) over 3 years. Over a ten-year period, these threshold ratios were reduced to CHF 3800 (2 implants) and CHF 7100 (4 implants) per Quality-adjusted Prosthesis Year gained.
Key Words: cost-effectiveness • economic evaluation • dental prosthesis • implant-supported denture • complete denture patient satisfaction
Placement of dental implants in the edentulous mandible facilitates improved patient satisfaction and chewing ability (Allen and McMillan, 2002; Bakke et al., 2002), and the decrease in functional impairment may be greater with implant-supported overdenture prostheses (4 implants) than with implant-retained prostheses (2 implants) (Tang et al., 1997). Treatment costs, however, substantially increase with the number of implants placed in the mandible (Zitzmann and Marinello, 2001). This issue is typically addressed in cost-effectiveness analyses evaluating whether more expensive interventions represent value for money spent (Feine et al., 1998; Palmqvist et al., 2004; Takanashi et al., 2004; Attard et al., 2005). Little is known about the long-term cost-effectiveness of complete dentures and implant prostheses in edentulous patients (Heydecke et al., 2005). In most studies, the estimation of costs and/or effects has been restricted to the follow-up period of the clinical trial (Jönsson and Karlsson, 1990; Tang et al., 1997; Palmqvist et al., 2004; Takanashi et al., 2004). However, most of the costs for prosthetic restorations usually accrue during the first year, whereas improved dental outcomes usually last during the functional period of the denture, which may be much longer (Attard and Zarb, 2004). It is therefore decisive to choose a time-horizon that reflects the life-expectancy of the prosthetic restoration, to avoid a bias against the cost-effectiveness of the intervention. Therefore, we conducted a cost-effectiveness analysis comparing the 3 treatment strategies—implant-supported overdentures, implant-retained overdentures, and complete dentures—in edentulous patients by combining clinical trial data with a follow-up of 3 yrs, with mathematical modeling to project long-term costs and effects over a time period of up to ten years.
Patients Sixty edentulous patients who visited the Clinic for Reconstructive Dentistry (University of Basel, Switzerland) during January, 1999, and December, 2001, and who required treatment in the edentulous lower jaw with removable prostheses, with or without implants, were included in the study. The study protocol was approved by the local ethics committee, and patients gave their informed consent as reported previously (Zitzmann et al., 2005). In brief, the investigation was designed so that the patients were informed about the established treatment options and expected costs; they received an individual treatment recommendation from the specialist and then selected the treatment themselves (self-selected trial [Brewin and Bradley, 1989]). Twenty patients decided on an implant-retained overdenture prosthesis with 2 implants and ball abutments, and were consecutively included in the first group; 20 patients opted for 4 interforaminal implants with a bar-retained overdenture (implant-supported overdentures), and these comprised the second group; and 20 patients requested a complete denture and were included in the third group (control). The sample size was guided mainly by the number of patients requiring the most expensive implant treatment during the study period. Patients were offered the option to receive prosthetic treatment in the undergraduate program for a reduced fee but a longer treatment time as compared with the post-graduate clinic.
Effects The importance of adjustment for baseline utility in economic evaluations has been emphasized elsewhere; hence the difference in the patient’s dental health state preference before and after posthetic treatment was used to calculate Quality-adjusted Prosthesis Years (Manca et al., 2005). Patients’ preferences recorded at 6 and 36 mos were used to estimate Quality-adjusted Prosthesis Years in years 1 and 3, respectively. For year 2, each patient’s dental health state preference was determined by linear interpolation, and this value was then used for the estimation of the Quality-adjusted Prosthesis Years during the second year. Since the time horizon of the clinical trial was 3 yrs, but dental restorations may have a much longer life-expectancy, we projected Quality-adjusted Prosthesis Years experienced after year three up to 10 yrs by assuming that the patient’s dental health state preference observed at month 36 would be stable over the projected time horizon.
Costs
Cost-effectiveness
Statistical Analysis To assess the uncertainty around the point estimate of the incremental cost-effectiveness ratio, we computed the bivariate distributions of mean total costs and Quality-adjusted Prosthesis Years for each of the 3 treatment strategies, using 5000 bootstrap samples, and summarized them in terms of three-way cost-effectiveness acceptability curves (Briggs et al., 2002). The cost-effectiveness ratio above which an intervention is deemed cost-ineffective corresponds to the decision-maker’s maximum willingness to pay per Quality-adjusted Prosthesis Year gained (Glick et al., 2001).
The mean age of the patients was 71.4 ± 8.8 yrs (range, 44–98 yrs), and the group was comprised of 39 women and 21 men (Zitzmann et al., 2005). Within group 2 (implant-supported overdentures), a higher proportion of patients (n = 10) preferred treatment in the postgraduate clinic as compared with group 1 (n = 3) and group 3 (n = 5, p = 0.045). In all 3 groups, maintenance efforts were greatest during the first year and decreased in the following years (Table 1  ). Follow-up was complete for all patients but one, who selected implant-retained overdentures and who died 34 mos after the prosthetic treatment.
The average pre-treatment dental health state preference was higher in the complete denture group than in the implant-supported overdentures or implant-retained overdentures group (Table 1  ), but did not reach statistical significance (p = 0.112). Patients in the implant-supported overdentures group had the highest average dental health state preference after treatment, but this was not statistically significant (p = 0.472 after 6 mos and p = 0.086 after 3 yrs). The average gain in Quality-adjusted Prosthesis Years adjusted for pre-treatment dental health status was 0.68 for complete dentures, 1.46 for implant-retained overdentures, and 1.57 for implant-supported overdenture patients (p = 0.0478, boxplots in Fig. 1A). The between-group comparisons yielded significant differences between complete dentures and implant-supported overdentures (p = 0.0193), but no significant differences between complete dentures and implant-retained overdentures (p = 0.0678), or between implant-retained and implant-supported overdentures (p = 0.725).
The average total costs in the three groups after 3 yrs were statistically significantly different (p < 0.0001), and amounted to CHF 3675 for complete dentures, CHF 8874 for implant-retained overdentures, and CHF 17,837 for implant-supported overdentures patients (Table 1 and boxplots in Fig. 1B). Initial prosthetic treatment and time costs represented the largest part of the costs in each group, whereas maintenance costs and time costs for follow-up controls and treatments were much lower (Table 1).
The average incremental cost-effectiveness ratios for the 3 treatment options are shown in Table 2
Cost-effectiveness acceptability curves for the base-case analysis and the 10-year time horizon were calculated (Figs. 2A, 2B ). The threshold ratio at which implant-retained overdentures became preferable (i.e., more likely to be cost-effective) to complete dentures is the same as in Table 2, where mean values were used. However, the stochastic three-way comparison shows that implant-supported overdenture treatment becomes the strategy with highest likelihood of being cost-effective at a threshold ratio of CHF 19,800 (base-case) and CHF 7100 (10-year horizon) per Quality-adjusted Prosthesis Year gained (Figs. 2A, 2B).
In the present study, we have shown that (i) the time horizon of the economic evaluation of implant treatment may substantially influence cost-effectiveness, that (ii) implant-retained overdentures are the most cost-effective treatment option for a ceiling ratio between CHF 9100 and 19,800 per Quality-adjusted Prosthesis Year over 3 yrs and between CHF 3800 and 7100 per Quality-adjusted Prosthesis Year over 10 yrs, and that (iii) complete dentures and implant-supported overdentures become cost-effective at ceiling ratios below and above those reported for implant-retained overdentures, respectively. Implant treatment leads to greater improvements in dental health outcomes, but also requires substantially higher initial costs. These were almost 3 and 6 times higher for 2 and 4 implants, compared with conventional treatment with complete dentures, as was shown in a previous study (Zitzmann et al., 2005). The proportion of maintenance and time costs in relation to the initial costs also decreased with the number of implants used and represented 46% of the initial costs in complete denture patients, 28% in implant-retained overdentures patients, and only 13% in implant-supported overdentures patients. In contrast to these findings, Takanashi et al.(2004) observed that the total time spend for initial treatment and six-month follow-up was similar for complete dentures and implant-retained overdentures, with the additional time required for implant placement being balanced by extra visits needed for marginal adaptation after the insertion of complete dentures. In a cost comparison including direct and indirect costs for complete dentures and implant-retained overdentures over 1 yr, direct costs were almost 3 times higher for implant-retained overdentures as compared with complete dentures (Takanashi et al., 2004). In a recent Canadian cost-effectiveness analysis by the same research group (Heydecke et al., 2005), implant-retained overdentures vs. complete dentures were reported to cost $14 per Oral Health Impact Profile (OHIP-20) point. In this one-year follow-up study, the annual maintenance costs in the subsequent years were based on expert opinion ($273 for complete dentures and $395 for implant-retained overdentures), which are somewhat higher than those observed in our study in Switzerland. Assuming a life-expectancy of 18 yrs, the estimated discounted total costs and effects were $8852 and 443 OHIP-20 points for implant-retained overdentures, and $5646 and 666 OHIP-20 points for complete dentures (Heydecke et al., 2005). Whereas the costs for implant-retained overdentures are in line with those reported in our study, those for complete dentures are somewhat higher. The effects cannot be directly compared with those from our study, since we used Quality-adjusted Prosthesis Years as an outcome measure as opposed to OHIP-20 points. The advantage of a Quality-adjusted Prosthesis Year is that it provides a preference-based outcome measure commonly used in health economics, with an intuitive interpretation (i.e., one Quality-adjusted Prosthesis Year corresponds to 1 yr in the best possible state). In addition, some further methodological differences merit attention. The follow-up period in our study was 3 yrs, which allowed us to estimate actual maintenance costs and effects. Furthermore, we compared 3 treatment alternatives (i.e., complete dentures, and implant-retained and implant-supported overdenture prostheses), and hence provided an economic analysis including the most expensive option (i.e., implant-supported overdentures). Finally, we used a stochastic three-way analysis to account for uncertainty (Briggs et al., 2002). As shown in this paper, the deterministic and stochastic analyses may differ substantially with respect to the thresholds, where one treatment becomes preferable to the other. However, given that Heydecke et al.(2005) reported similar results, the modeling process validity of both research groups is supported (Sendi et al., 1999). Our study has several limitations. First, the number of patients included in each group may have been too small for us to detect differences in effects between complete dentures and implant-retained overdentures, due to a lack of statistical power. However, a sample size calculation would have entailed a randomized controlled trial with the cost of treatment being allocated randomly to patients, which was not possible due to the substantial differences in costs per treatment alternative. Second, we projected maintenance costs and effects observed in year three over a time horizon of up to 10 yrs. Although costs and effects may vary over the modeled time period, analysis of our data suggests that these variations may not substantially affect total costs and effects. The chosen time horizon is also in line with recent evidence demonstrating a longer life-expectancy of 12 yrs for implant overdentures (Attard and Zarb, 2004). Third, we used Quality-adjusted Prosthesis Years to measure dental health effects. Quality-adjusted Prosthesis Years assume no preference between the sequence of experienced dental health states, which may be addressed by the use of Healthy Year Equivalents (HYE) (Birch et al., 1998). However, since dental health may represent only a part of the patient’s health-related quality of life, the responsiveness of HYE in dentistry may need further investigation. In conclusion, over an assumed time horizon of 10 yrs, implant treatment becomes cost-effective with implant-retained overdentures being the treatment of choice, if the patient is willing to pay at least CHF 3800 for a Quality-adjusted Prosthesis Year gained.
This study did not receive any external funding and was supported by departmental resources. Received for publication September 6, 2005. Revision received April 10, 2006. Accepted for publication April 28, 2006.
Journal of Dental Research, Vol. 85, No. 8,
717-721 (2006)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
