Modelling the cost-effectiveness of a rapid diagnostic test (IgMFA) for uncomplicated typhoid fever in Cambodia

Abstract

Typhoid fever is a common cause of fever in Cambodian children but diagnosis and treatment are usually presumptive owing to the lack of quick and accurate tests at an initial consultation. This study aimed to evaluate the cost-effectiveness of using a rapid diagnostic test (RDT) for typhoid fever diagnosis, an immunoglobulin M lateral flow assay (IgMFA), in a remote health centre setting in Cambodia from a healthcare provider perspective. A cost-effectiveness analysis (CEA) with decision analytic modelling was conducted. We constructed a decision tree model comparing the IgMFA versus clinical diagnosis in a hypothetical cohort with 1000 children in each arm. The costs included direct medical costs only. The eligibility was children (≤14 years old) with fever. Time horizon was day seven from the initial consultation. The number of treatment success in typhoid fever cases was the primary health outcome. The number of correctly diagnosed typhoid fever cases (true-positives) was the intermediate health outcome. We obtained the incremental cost effectiveness ratio (ICER), expressed as the difference in costs divided by the difference in the number of treatment success between the two arms. Sensitivity analyses were conducted. The IgMFA detected 5.87 more true-positives than the clinical diagnosis (38.45 versus 32.59) per 1000 children and there were 3.61 more treatment successes (46.78 versus 43.17). The incremental cost of the IgMFA was estimated at $5700; therefore, the ICER to have one additional treatment success was estimated to be $1579. The key drivers for the ICER were the relative sensitivity of IgMFA versus clinical diagnosis, the cost of IgMFA, and the prevalence of typhoid fever or multi-drug resistant strains. The IgMFA was more costly but more effective than the clinical diagnosis in the base-case analysis. An IgMFA could be more cost-effective than the base-case if the sensitivity of IgMFA was higher or cost lower. Decision makers may use a willingness-to-pay threshold that considers the additional cost of hospitalisation for treatment failures.

Fig 1. Decision tree model for diagnosis and treatment for typhoid fever.

* MDR strains exist equally among true-positives and false-negatives. Since probability of treatment success in azithromycin is the same between MDR strains and non-MDR strains, *1 and *3 are not branched to P9 and (1-P9).

Fig 2. Tornado diagram of ICER change by varying cost parameters.

*Cost of amoxicillin is a total cost of 1500mg/day, 5 days course, including 10% supply cost. **Cost of azithromycin is a total cost of 250mg/day, 5 days course, including 10% supply cost. ICER, Incremental Cost-effectiveness Ratio.

Fig 3. ICER distributions in two-way sensitivity analyses.

Fig 4. Cost-effectiveness plane in incremental number of treatment success and incremental cost.

Fig 5. Cost-effectiveness acceptability curve.