Introduction A cost-utility analysis of unilateral cochlear implantation (CI) in severe to profoundly deaf adults has been undertaken from the primary perspective of the direct medical costs to the Norwegian health care system (all providers) as well as an assessment of those direct medical costs borne by patients (in the form of co-payments).

Methods Cost-effectiveness was measured as the incremental cost per quality adjusted life year (QALY) gained of CI compared to no intervention and was estimated using a simple patient care pathway decision model (with associated costs and outcomes).

The measure of outcome chosen was the likely improvement in the quality of life (in terms of gain in health utility for adult recipients over the useful life of the implant (assumed to be the implantee’s remaining lifetime). A mean value of 0.2 QALYs per year based on a single study included in a systematic review on cochlear implants in severe to profoundly deaf adults from the Norwegian Knowledge Centre for the Health Services’ (Report no 25-2006 ) was used to estimate the “quality weighted health state” gain from CI in the base case analysis. Medical services utilized relating to hospital outpatient and inpatient care were priced using official national tariffs and DRG based reimbursement rates respectively (at 2005/06 levels). Future costs and outcomes were discounted at a rate of 4% per annum.

Results Values for cost per QALY have been calculated for an average adult aged 58 at time of implantation (based on published Norwegian studies) and results calculated for the lifetime of the implant. In the base case, CI yielded an estimated incremental discounted lifetime cost of kr 537,100 per case, compared to no intervention, a total gain in discounted QALYs of 3.12 and a cost per QALY gained of kr 172,000.

To explore uncertainty in the assumptions used we explored how sensitive the results were to changes in the values of a number of parameter estimates. Univariate sensitivity analysis (varying one parameter at a time) performed on the gain in health utility, resource use, unit costs, discount rate and duration of device use indicated that estimates were particularly sensitive to changes in the gain in health utility and duration of devise use but reasonably robust to the most plausible values for other model parameters. Probabilistic sensitivity analysis (in which model parameters were varied simultaneously) across assumed distributions demonstrated that under a mean utility gain of 0.2, there was a 97% probability that CI in adults would be cost-effective if one was willing to pay kr 400,000. Applying a lower value of 0.15 per year for the mean utility indicated a corresponding 92% probability that CI in adults would be cost effective.

Conclusions The results from this cost-utility analysis demonstrate that:

• The estimates of cost per QALY gained for CI in adults in Norway are consistent with other published estimates reported in the adult CI cost-effectiveness literature.
• Cost-effectiveness of CI in adults compares favorably relative to other commonly accepted health care interventions.
• Analyses undertaken on the uncertainty in parameter estimates, including a more conservative assumption on the gain in health utility, indicated that CI in adults would potentially remain cost-effective across a range of possible acceptability thresholds.