Table 2.
Key cost-effectiveness model parameters
| Parameter | Best estimate (range for PSA) | Distribution for PSA | Source(s) | |
| Proportion of patients under 5 years | 0.40 (0.30–0.50) | Beta | 20 | |
| True underlying diagnosis | ||||
| Malaria cases with bacterial coinfection | 0.06 (0.03–0.174) | Beta | 61 62 | |
| NMFI cases that are bacterial infection | 0.10 (0.013–0.15) | Beta | 20 63 | |
| Diagnosis (intervention arm only) | ||||
| Patients receiving RDT (‘uptake’) | 0.41 (0.08–0.72) | Beta | 8 | |
| RDT sensitivity | 0.948 (0.931–0.961) | Beta | 64 | |
| RDT specificity | 0.952 (0.631–0.967) | Beta | 29 64 | |
| Initial treatment | ||||
| Proportion of patients receiving an ACT, non-ACT antimalarial, antibiotic, other drug, or no drug | See figure 2, online supplementary file S4 | Dirichlet | 28 30 31 36–38 | |
| Treatment effectiveness | ||||
| ACT efficacy (for malaria) | 0.955 (0.82–1.00) | Beta | 39 | |
| Other antimalarial efficacy (for malaria) | 0.78 (0.183–0.97) | Beta | 39 | |
| Antibiotic efficacy (for bacterial infection) | 0.80 (0.72–0.88) | Beta | Assumption (range: ±10%) | |
| Proportion of stated API in drug (drug quality) | 0.92 (0.828–1.011) | Gamma | 40 (range: ±10%) | |
| Proportion of required dose consumed (adherence to treatment) | 0.892 (0.5–1.0) | Beta | 41 (range: assumption) | |
| Reduction in treatment efficacy due to API consumed | Low transmission: | Medium/high transmission: | Assumptions | |
| 80%–85% | 0.15 (0.05–0.25) | 0.10 (0.00–0.20) | Beta | |
| 75%–80% | 0.30 (0.15–0.45) | 0.25 (0.10–0.40) | Beta | |
| 50%–75% | 0.60 (0.45–0.75) | 0.50 (0.35–0.65) | Beta | |
| <50% | 0.95 (0.90–1.00) | 0.95 (0.90–1.00) | Beta | |
| Disease progression and further care | ||||
| Malaria case progresses to severe with no (or not effective) treatment | Low transmission: | Medium/high transmission: | 65 (Medium/high transmission best estimates: assumptions) | |
| <5 years | 0.30 (0.10–0.90) | 0.10 (0.05–0.60) | Beta | |
| 5+ years | 0.18 (0.05–0.50) | 0.02 (0.00–0.15) | Beta | |
| Bacterial case progresses to severe with no (or not effective) treatment | Low HIV: | High HIV: | 65 | |
| <5 years | 0.20 (0.05–0.80) | 0.40 (0.15–0.90) | Beta | |
| 5+ years | 0.20 (0.05–0.70) | 0.30 (0.10–0.90) | Beta | |
| Severe case receives further (inpatient) care | 0.75 (0.19–0.88) | Beta | Assumption (range20) | |
| Final health outcomes | ||||
| CFR of severe malaria receiving inpatient care | 0.10 (0.05–0.15) | Beta | 20 42 | |
| CFR of severe malaria with no further care | Low transmission: | Medium/high transmission: | 65 | |
| <5 years | 0.73 (0.25–0.95) | 0.45 (0.05–0.90) | Beta | |
| 5+ years | 0.70 (0.30–0.95) | 0.60 (0.10–0.90) | Beta | |
| CFR of severe bacterial infection receiving inpatient care | 0.15 (0.10–0.20) | Beta | 20 | |
| CFR of severe bacterial infection with no further care | Low HIV: | High HIV: | ||
| <5 years | 0.40 (0.10–0.90) | 0.50 (0.15–1.00) | Beta | 65 |
| 5+ years | 0.30 (0.10–0.80) | 0.50 (0.10–0.90) | Beta | |
| Implementation costs (2017 US$)* | ||||
| RDT ex-manufacturer price | 0.22 (0.17–0.28) | Gamma | 66 (range: ±25%) | |
| RDT subsidy (% ex-manufacturer price) | 0.50 (0.40–0.60) | Beta | Assumption | |
| ACT ex-manufacturer price | 0.68 (0.51–1.56) | Gamma | 67 | |
| ACT subsidy (% ex-manufacturer price) | 0.80 (0.70–0.90) | Beta | Assumption | |
| Inpatient cost per day† | 4.33 (3.25–17.72) | Gamma | 68 | |
| Supporting intervention cost per febrile patient | 0.43 (0.21–0.64) | Gamma | See online supplementary file S5 | |
*All costs were adjusted to 2017 US dollars using the median of the five year annual average GDP deflator in the six countries participating in the Private Sector Co-payment Mechanism.69
†Inpatient cost is bed-day cost only; excludes cost of treatment.
ACT, artemisinin combination therapy; API, active pharmaceutical ingredient; CFR, case fatality rate; GDP, gross domestic product; NMFI, non-malarial febrile illness; PSA, probabilistic sensitivity analysis; RDT, rapid diagnostic test.