Skip to main content
. 2015 Oct 23;17(8):939–950. doi: 10.1007/s10198-015-0734-5

Table 4.

Overview clinical effect estimates of QALY model for the missing data methods

RA CCA (% bias) MI-PMM (% bias) Log MI-PMM (% bias) Two-step MI (% bias)
M M + H M M + H M M + H M M + H M M + H
17 % Missing data
n 237 193 201 154 237 193 237 193 237 193
Mean (QALY) 0.730 0.798 0.722 0.798 0.728 0.792 0.727 0.791 0.728 0.792
SE mean (QALY) 0.015 0.016 0.017 0.016 0.016 0.016 0.016 0.016 0.020 0.016
QALY difference 0.054 0.060 (11) 0.061 (12) 0.061 (12) 0.061 (12)
SE QALY difference 0.018 0.020 (12) 0.020 (10) 0.020 (10) 0.020 (11)
z for CCA and t for MI 2.970 2.950 3.020 3.020 3.000
p value 0.003 0.003 0.003 0.003 0.003
95 % CI lower limit 0.018 0.020 0.021 0.021 0.021
95 % CI upper limit 0.090 0.100 0.100 0.100 0.100
35 % Missing data
n 237 193 163 122 237 193 237 193 237 193
Mean (QALY) 0.730 0.790 0.715 0.790 0.718 0.790 0.717 0.791 0.718 0.790
SE mean (QALY) 0.015 0.018 0.020 0.018 0.017 0.016 0.017 0.016 0.017 0.016
QALY difference 0.054 0.068 (24) 0.069 (27) 0.071 (30) 0.069 (27)
SE QALY difference 0.018 0.023 (27) 0.021 (17) 0.022 (18) 0.022 (20)
z for CCA and t for MI 2.970 2.910 3.230 3.260 3.150
p value 0.003 0.004 0.001 0.001 0.002
95 % CI lower limit 0.018 0.022 0.027 0.028 0.026
95 % CI upper limit 0.090 0.113 0.111 0.113 0.112
50 % missing data
n 237 193 132 91 237 193 237 193 237 193
Mean (QALY) 0.730 0.782 0.717 0.782 0.705 0.785 0.708 0.784 0.706 0.784
SE mean (QALY) 0.015 0.021 0.021 0.021 0.018 0.017 0.018 0.018 0.018 0.018
QALY difference 0.054 0.047 (13) 0.077 (41) 0.074 (36) 0.075 (38)
SE QALY difference 0.018 0.026 (43) 0.024 (29) 0.024 (30) 0.024 (31)
z for CCA and t for MI 2.970 1.820 3.260 3.110 3.140
p value 0.003 0.069 0.001 0.002 0.002
95 % CI lower limit 0.018 −0.004 0.031 0.027 0.028
95 % CI upper limit 0.090 0.098 0.123 0.120 0.122

M refers to the methadone maintenance treatment group, M + H refers to the group that had medical co-prescription of heroin. SE standard error, QALY quality of life years gained, CI confidence interval, RA reference analysis, CCA complete case analysis, PMM multiple imputation with predictive mean matching, Log MI-PMM multiple imputation with predictive mean matching on log-transformed costs, Two-step-MI two-step multiple imputation with predictive mean matching