From: The statistical approach in trial-based economic evaluations matters: get your statistics together!
Analysis | Societal costs difference (€) (95% CI) | QALY difference (95% CI) | ICER (€/ QALY gained) |
---|---|---|---|
1.- Two t-tests (1 for cost; 1 for effects) complete-case analysis | −142 (− 1298 to 1014) | −0.002 (− 0.057 to 0.54) | 90,989 |
2.- Two bootstrapped t-tests; complete-case analysis | −142 (− 1326 to 986) | − 0.002 (− 0.057 to 0.54) | 90,989 |
3.- Two regressions with correction for confounders; complete-case analysis | −23 (− 1180 to 1134) | −0.009 (− 0.041 to 0.03) | 2536 |
4.- Two bootstrapped regressions with correction for confounders; complete-case analysis | −23 (− 1170 to 1096) | −0.009 (− 0.041 to 0.03) | 2536 |
5.- Two t-tests; mean imputation | − 105 (− 971 to 759) | 0.006 (− 0.037 to 0.049) | − 17,307 |
6.- Two bootstrapped t-tests; mean imputation | − 105 (− 1032 to 739) | 0.006 (− 0.037 to 0.049) | −17,307 |
7.- Two regressions with correction for confounders; mean imputation | −85 (− 954 to 784) | 0.009 (− 0.020 to 0.038) | − 9208 |
8.- Two bootstrapped regressions with correction for confounders; mean imputation | −85 (− 966 to 770) | 0.009 (− 0.020 to 0.038) | − 9208 |
9.- Two t-tests; MI | 438 (− 1033 to 1910) | 0.006 (− 0.041 to 0.053) | 74,862 |
10.- Two bootstrapped t-tests; MI | 438 (− 940 to 1704) | 0.006 (− 0.041 to 0.053) | 74,862 |
11.- Two regressions with correction for confounders; MI | 486 (− 995 to 1968) | 0.010 (− 0.022 to 0.042) | 47,305 |
12.- Two bootstrapped regressions with correction for confounders; MI | 486 (− 904 to 1747) | 0.010 (− 0.022 to 0.042) | 47,305 |
13.- SUR without correction for confounders; MI | 438 (− 936 to 1712) | 0.006 (− 0.042 to 0.053) | 74,862 |
14.- SUR with correction for confounders; MI | 462 (− 921 to 1728) | 0.010 (− 0.022 to 0.0423) | 46,592 |