OBJECTIVES: The study aims to analyse the cost-effectiveness of the culture-based treatment (CBT) versus empirical treatment (ET) to guide antibiotic use in hospitalized community-acquired pneumonia (CAP).

METHODS: A model was developed from the individual patient data of hospitalized adults with CAP admitted to an academic hospital in Indonesia from 2014 to 2016 with ICD-10 J.18x. The second antibiotic was assessed after microbiological culture results in terms of the impact on hospital costs and life-years (LY). We conducted subgroup analyses for the implementation of CBT and ET in elderly (aged >60), moderate-severe CAP (PSI class III-V), and ICU patients. The model was designed with a lifetime horizon by adjusting the patients’ ages to the average remaining LY expectancy for the Indonesian population with a 3% discount for both cost and LY. We applied sensitivity analyses on 1,000 simulation cohorts to examine the economic acceptability of CBT in practice. The willingness-to-pay (WTP) was defined as 1x or 3x the Indonesian Gross Domestic Product (GDP) per capita (US$ 3,570.28).

RESULTS: CBT would effectively increase the patients’ LY and save US$ 4,313.60/LY. The ET group’s hospitalization cost had the greatest influence on the ICER value. The subgroup analyses show that CBT could be cost-effectively implemented in Indonesia in patients aged >60 (ICER= US$ 6,011.22/LY), patients with moderate-severe CAP (ICER= US$ 4,094.20/LY), and patients in the ICU (ICER= US$ 4,147.01) with acceptability rates of 93.6%, 68.2%, and 77.1% for 1xWTP and 86.7%, 62.5%, 72.4% for 3xWTP, respectively.

CONCLUSIONS: Both sputum and blood cultures provide advantages in terms of cost saving and LY gained for patients with hospitalized CAP. Moreover, CBT is considered cost-effective in patients aged >60, PSI class III or above and in the ICU.