Table 6. Cost Effectiveness Analyses.

Scenario/Analyses	Difference, adjusted forbaseline andparticipant/clustercovariates* Mean(95% CI)	ICER, Cost (£)per QALY	Probability Cost-effective at WTP‡per QALY gained:
			£20,000 per QALY	£30,000 per QALY
Base Case:
Total NHS and personal social services	£270.72(–202.98, 886.04)
EQ-5D: QALY (12-month)	0.019(–0.019, 0.06)	£14,248	0.58	0.65
Sensitivity Analyses:
(1) Base Case CEA, withmultiple imputation ofmissing data
Total NHS and personal social services	£292.08(–216.88, 801.04)
EQ-5D: QALY (12-month)	0.017(–0.020, 0.054)	£17.490	N/A	N/A
(2) CEA Using SF6D QALY data
SF6D: QALY (12-month)	0.0168(0.000, 0.032)	£16,114	0.57	0.72
(3) CEA when excludingone high cost participant
Total NHS and personal social services	£63.34(–295.98, 422.67)
EQ-5D: QALY (12-month)	0.019(–0.018, 0.06)	£3,334	0.76	0.79
(4) CEA using higher costestimate for CollaborativeCare, at meancost of £338.80
Total NHS and personal social services	£337.02(–136.67, 952.34)	£17,738	0.54	0.62
(5) CEA using a broaderperspective, including patientcosts and informalcare costs	–£312.83(–2,339.93, 2,035.27)	Collaborative Care is dominant**	N/A	N/A

*Adjusting for baseline measures, and pre-specified covariates for age (individual level), and (at the cluster level) deprivation (IMD), site, and practice size.

**Dominance: lower expected costs, with greater expected QALY gain.

^‡WTP = willingness to pay; based in the assessment of incremental net benefit statistic, and WTP thresholds commonly applied in the UK NHS (NICE, 2013).