Table 2. Full-sample cost-effectiveness analysis (mean cost in £ / mean QALY, 95% CI).
| Intervention (205 patients) | Standard care (212 patients) | Incremental cost / QALYs gained | |
|---|---|---|---|
| The cost of care a | 4267 (3697, 4934) | 4057 (3367, 4882) | 210 (-809, 1165) |
| The cost of care—adjusted b | 4203 (4130, 4276) | 4110 (4037, 4182) | 94 (-10, 198) |
| The intervention cost | 208 (192, 227) | 0 | 208 (192, 227) |
| Total cost (care cost + intervention cost) | 4475 (3901, 5141) | 4057 (3367, 4882) | 419 (-597, 1371) |
| Total cost—adjusted (care cost adjusted + intervention cost) | 4412 (4331, 4490) | 4110 (4037, 4182) | 302 (193, 410) |
| QALYs gained | 0.107 (0.097, 0.118) | 0.103 (0.093, 0.112) | 0.003 (-0.012, 0.017) |
| QALYs gained—adjusted c | 0.106 (0.100, 0.115) | 0.107 (0.098, 0.113) | -0.001 (-0.009, 0.007) |
| ICER | £147 087/QALY | ||
| ICER adjusted | The intervention dominated by standard care | ||
Multiple imputation by chained equation (MICE): predictive mean matching (pmm) for utilities and ordered logit (ologit) for Barthel ADL scores; 45 imputations.
aInpatient, day-case and outpatient cost data were collected for both locations, Nottingham and Leicester.
bAdjusted by age, sex, hospital location (Leicester), and baseline utility, permanent care home residence, and Charlson co-morbidity (scores 2–3 and ≥4). A GLM model (family—gamma, link—0.8) was applied.
c Adjusted by age, sex, hospital location (Leicester), and baseline utility. OLS was applied.