Direct costs of managing in-ward dengue patients in Sri Lanka: A prospective study

Introduction The cost in managing hospitalised dengue patients varies across countries depending on access to healthcare, management guidelines, and state sponsored subsidies. For health budget planning, locally relevant, accurate costing data from prospective studies, is essential. Objective To characterise the direct costs of managing hospitalised patients with suspected dengue infection in Sri Lanka. Methods Colombo Dengue Study is a prospective single centre cohort study in Sri Lanka recruiting suspected hospitalised dengue fever patients in the first three days of fever and following them up until discharge. The diagnosis of dengue is retrospectively confirmed and the cohort therefore has a group of non-dengue fever patients with a phenotypically similar illness, managed as dengue while in hospital. The direct costs of hospital admission (base and investigation costs, excluding medication) were calculated for all recruited patients and compared between dengue and non-dengue categories as well as across subgroups (demographic, clinical or temporal) within each of these categories. We also explored if excluding dengue upfront, would lead to an overall cost saving in several hypothetical scenarios. Results From October 2017 to February 2020, 431 adult dengue patients and 256 non-dengue fever patients were recruited. The hospitalisation costs were USD 18.02 (SD: 4.42) and USD 17.55 (SD: 4.09) per patient per day for dengue and non-dengue patients respectively (p>0.05). Laboratory investigations (haematological, biochemical and imaging) accounted for more than 50% of the total cost. The costs were largely homogenous in all subgroups within or across dengue and non-dengue categories. Excluding dengue upfront by subsidised viral genomic testing may yield overall cost savings for non-dengue patients. Conclusion As non-dengue patients incur a similar cost per day as the dengue patients, confirming dengue diagnosis using subsidised tests for patients presenting in the first three days of fever may be cost-efficient.