People with traumatic brain injuries (TBIs) commonly report memory impairments which are persistent, debilitating, and reduce quality of life. As part of the Rehabilitation of Memory in Brain Injury trial, a cost-effectiveness analysis was undertaken to examine the comparative costs and effects of a group memory rehabilitation program for people with TBI.
Individual-level cost and outcome data were collected. Patients were randomized to usual care (n=157) or usual care plus memory rehabilitation (n=171). The primary outcome for the economic analysis was the EuroQol-5D quality of life score at 12 months. A UK NHS costing perspective was used. Missing data was addressed by multiple imputation. One-way sensitivity analyses examined the impact of varying different parameters, and the impact of available cases, on base case findings whilst non-parametric bootstrapping examined joint uncertainty.
At 12 months, the intervention was GBP 26.89 (USD 35.76) (SE 249.15) cheaper than usual care; but this difference was statistically non-significant (p=0.914). At 12 months, a QALY loss of −0.007 was observed in the intervention group confidence interval (95% CI: −0.025–0.012) and a QALY gain seen in the usual care group 0.004 (95% CI: -0.017–0.025). This difference was not statistically significant (p=0.442). The base case analysis gave an ICER of GBP 2,445 (USD 3,252) reflecting that the intervention was less effective and less costly compared to usual care. Sensitivity analyses illustrated considerable uncertainty. When joint uncertainty was examined, the probability of the intervention being cost-effective at a willingness-to-pay threshold of GBP 20,000 per QALY gain was 29 percent and 24 percent at GBP 30,000.
Our cost-utility analysis indicates that memory rehabilitation was cheaper but less effective than usual care but these findings must be interpreted in the light of small statistically non–significant differences and considerable uncertainty was evident. The ReMemBrIn intervention is unlikely to be considered cost-effective for people with TBI.