Table 6. Sensitivity analysis: cost-effectiveness of the intervention from an NHS/PSS perspective, excluding non-primary-care-/non-intervention-related costs, on imputed data.

	Usual care mean	Intervention mean	Difference (95% CI)
Costs and QALYs
Total NHS/PSS costs	£488	£623	£135 (£70 to £199)
QALYs	0.540	0.541	0.001 (−0.023 to 0.026)
Cost-effectiveness statistics
ICER: £114 624
Probability that intervention cost-effective at CE threshold of £20 000: 0.33
Probability that intervention cost-effective at CE threshold of £30 000: 0.40
NMB at threshold of £20 000 (95% confidence interval): −£111 (−£132 to −91)

CI, confidence interval; PSS, personal social services; QALYs, quality-adjusted life years; ICER, incremental cost-effectiveness ratio; CE, cost-effectiveness; NMB, net monetary benefit.

We report confidence intervals for the point estimate of net benefit, but not for the ICER. Confidence intervals for the ICER can be both difficult to interpret³⁴ and statistically intractable.³³ Instead, we place an emphasis throughout our analysis on net benefit. We present cost-effectiveness acceptability curves and confidence intervals around net benefit to represent uncertainty in our cost-effectiveness results.