Sensitivity and subgroup analyses
App-based treatment remained cost-effective when assessed with fewer app users, extra developmental and higher maintenance costs (Supplemental table S6). Sensitivity analysis using the mean costs at 4 and 12 months’ follow-up revealed comparable results, demonstrating the robustness of the cost calculation.
Subgroup analysis revealed differences in effects and costs by UI type and recruitment type (Supplemental table S7). App-based treatment for urgency UI resulted in higher IIALYs gained (0.74) compared with care-as-usual (0.60). The costs for UI-specific treatment were also approximately \euro60 higher for urgency UI compared with stress UI mainly due to the cost of incontinence material. Subgroup analysis by recruitment type showed that, for care-as-usual, the group recruited through (social) media had lower costs (\euro131) and a lower treatment effect (IIALY 0.64) than the group recruited by a GP (\euro235, IIALY 0.68). These cost differences were mainly based on lower use of physical therapy (\euro56 versus \euro122) and other treatments (e.g. pessary or tension-free vaginal tape) (\euro2 versus