Table 4. Model results under base case and scenario analyses.
| Scenario | AiCure | DOT | AiCure vs. DOT | ||||
|---|---|---|---|---|---|---|---|
| Cost (2017 USD) | QALYs | Cost (2017 USD) | QALYs | Cost (2017 USD) | QALYs | ICER (2017 USD) | |
| Base case | $2,668 | 1.05 | $4,894 | 1.03 | -$2,226 | 0.02 | AI Dominant |
| AI 5% worsea | $2,860 | 1.03 | $4,894 | 1.03 | -$2,034 | 0.00 | AI Dominant |
| AI 10% worsea | $3,011 | 1.02 | $4,894 | 1.03 | -$1,882 | -0.01 | $244,651 (DOT) |
| AI 15% worsea | $3,164 | 1.01 | $4,894 | 1.03 | -$1,730 | -0.02 | $89,889 (DOT) |
| AI WC/DOT BC | $5,278 | 1.08 | $1,607 | 1.08 | $3,672 | 0.01 | $433,646 (AI) |
| AI BC/DOT WC | $2,883 | 1.01 | $10,485 | 0.98 | -$7,603 | 0.03 | AI Dominant |
a x% worse implies an x percentage-point decrease in all monthly completion probabilities up to 15 months, with a floor of zero. The worst/best scenario involves model one-way sensitivity parameters tested at their 95% confidence interval lower or upper bounds. All costs and QALYs are reported per-patient. AI, AiCure; DOT, directly observed therapy; ICER, incremental cost-effectiveness ratio, QALY, quality-adjusted life year; BC, best case; WC, worst case.