| RMM and RMFE | CCA | MILR and MIPMM | BPA |
---|---|---|---|---|
Number of patients included at 3Â years | 450 | 44 | 450 | 450 |
Total number of non-missing observations included at 3Â yearsa | 1929 EQ-5D, 6861 period costs | 44 total costs, 44 QALY | 450 EQ-5D, 450 period costs | 377 total costs, 44 QALY |
Format of data as input | Longitudinal | Aggregate | Longitudinal | Aggregate |
Statistical model of the missing data | Implicit imputation of missing EQ-5D and period costs | None | Explicit imputation of missing EQ-5D and period costs | Logit model of probability of missingness |
How are total costs and QALY over the desired time horizon predicted at individual level? | Not necessary | Not done | Passively in each imputed dataset | Missing total cost and QALY are parameters to estimate |
How are mean total incremental costs and QALY over the desired time horizon estimated | Weighted sum of EQ5D and period cost coefficients estimated in the statistical model | Bivariate normal regression | Bivariate normal regression for each imputed dataset, synthesised using Rubin’s rules | Bivariate normal regression |
Estimation of standard errors and CEAC | Bootstrap | Parametrically | Parametrically | Parametrically |