Table 2.

Cost-effectiveness of 10TT under different MNAR assumptions for missing cost and effectiveness quality-of-life data

Scenario description	Incremental cost (£) [95% CI]	Incremental QALYs [95% CI]	INMBa (£) [95% CI]	Probability cost-effectivea (%)
MAR	− 35 [− 504 to 434]	− 0.004 [− 0.074 to 0.066]	− 49 [− 1632 to 1534]	48
Same MNAR parametersb in the two arms
− 10% QoL in both arms	− 35 [− 504 to 434]	− 0.011 [− 0.078 to 0.057]	− 181 [− 1714 to 1352]	41
+ 10% cost in both arms	− 25 [− 512 to 462]	− 0.004 [− 0.074 to 0.066]	− 59 [− 1650 to 1532]	47
− 10% QoL and + 10% cost	− 25 [− 512 to 462]	− 0.011 [− 0.078 to 0.057]	− 191 [− 1733 to 1350]	40
Different MNAR parametersb in the two arms
− 10% QoL in intervention arm	− 35 [− 504 to 434]	− 0.071 [− 0.139 to − 0.002]	− 1378 [− 2932 to 176]	4
− 10% QoL in control arm	− 35 [− 504 to 434]	0.056 [− 0.014 to 0.125]	1148 [− 415 to 2711]	93
+ 10% cost in intervention arm	20 [− 459 to 499]	− 0.004 [− 0.074 to 0.066]	− 104 [− 1691 to 1483]	45
+ 10% cost in control arm	− 80 [− 558 to 398]	− 0.004 [− 0.074 to 0.066]	− 4 [− 1591 to 1583]	50

All results are based on imputed data and comparing the 10TT arm to the control arm (n = 537)

CI confidence interval, INMB incremental net monetary benefit, MAR missing at random, MNAR missing not at random, QALY quality-adjusted life year, QoL quality of life, 10TT Ten Top Tips

^aAt a cost-effectiveness threshold of £20,000/QALY

^bHow missing cost and QoL data are assumed to differ from MAR-imputed values