Economic Evaluation of the Cochlear Implant

A cost utility study of the cochlear implant has been undertaken from the perspective of `government/service provider', covering its use in profoundly deaf adults, partially deafened adults and children. Costs include the selection of implant recipients, surgery/implantation, and rehabilitation/implant maintenance over the useful life of the implant. Costs for recipients and their families are included only in so far as they are related to the `service provider' perspective. Costs borne purely by patients (travel time, foregone income, etc) are not included in the analysis. The measure of outcome chosen is the likely improvement in the quality of life for recipients over the useful life of the implant. A brief review of the literature indicated that significant recent developments in cochlear implantation include improved reliability of the device, extension of the useful life of an implant to 15 years or more, further evidence of good outcomes for children who have cochlear implants, and indications that while achievement of a sensation of hearing provides benefits, additional and greater degrees of benefit are obtained from other quality of life consequences of the implant procedure. The HRQOL-15D quality of life instrument developed by Sintonen has been used to derive utility values for the outcome measure, because of its sensitivity for the application and superiority of validation in the context of economic evaluation. Three values (low, middle and high) have been calculated for each of the patient groups considered, based on expert opinion. The different values reflect the number of dimensions from the HRQOL-15D instrument included in each scenario, and the extent of the health state improvement within each dimension. For profoundly deafened adults the improvement in health-related quality of life through use of an implant ranges from 11% to 37%. Amelioration of the hearing disability contributes 3-4%, while the functional consequences of the hearing improvement result in a further 8-33% improvement. Corresponding results for partially deafened adults and for children are 2% and 4-5% improvement respectively through obtaining hearing sensation, and 9-28% and 13-32% improvement respectively due to functional consequences. Values for cost per quality adjusted life year have been calculated for each patient group using a specified treatment pathway (with associated costs and probabilities); the three values obtained from the HRQOL-15D instrument; and 10, 15 and 20 years as the useful life of the implant. A 5% discount rate was used for both costs and outcomes. Results using a value of 15 years for the lifetime of the implant give an indication of the typical cost utility of the technology. Costs per quality adjusted life year range from $11,790 to $38,150 for profoundly deaf adults; $14,410 to $41,000 for partially deafened adults; and $5,070 to $11,100 for children. The results for children in part reflect the offsetting savings through enabling education to be undertaken in mainstream schools. These savings (net present value) are estimated as $7,978 per child. The results also include a more comprehensive costing than in an earlier Australian study, with provisions made for regular software upgrades and on-going electrical maintenance. These results are of the same order as other values reported in the literature, and are more optimistic for children than the values derived in the earlier Australian study. They indicate that the cochlear implant compares favourably with other health care technologies where resources are currently being committed. While there is no universally agreed benchmark on the appropriate trade-off between costs and quality adjusted life years, these results would normally be regarded as good `value for money'. Sensitivity testing indicates that the estimates are reasonably robust to changes in the discount rate and in the cost of the device. The estimates for children are sensitive to changes in the rates of long term rehabilitation and the proportion of those who participate in mainstream education. Further work that might usefully be undertaken includes obtaining Australian values for quality of life factors from empirical studies and considering more specific groups of patients and interest groups than has been possible in the present analysis. The extent of the quality of life improvement due to functional consequences vis-a-vis the hearing impairment itself, reinforces the requirement for such research, as does the need to take a broader societal perspective to the economic appraisal.