Cost of dengue and other febrile illnesses to households in rural Cambodia: a prospective community-based case-control study

Background: The average annual reported dengue incidence in Cambodia is 3.3/1,000 among children < 15 years of age (2002-2007). To estimate the economic burden of dengue, accurate cost-of-illness data are essential. We conducted a prospective, community-based, matched case-control study to assess the cost and impact of an episode of dengue fever and other febrile illness on households in rural Cambodia. Methods: In 2006, active fever surveillance was conducted among a cohort of 6,694 children aged <= 15 years in 16 villages in Kampong Cham province, Cambodia. Subsequently, a case-control study was performed by individually assigning one non-dengue febrile control from the cohort to each laboratory-confirmed dengue case. Parents of cases and controls were interviewed using a standardized questionnaire to determine household-level, illness-related expenditures for medical and non-medical costs, and estimated income loss (see Additional file 1). The household socio-economic status was determined and its possible association with health seeking behaviour and the ability to pay for the costs of a febrile illness. Results: Between September and November 2006, a total of 60 household heads were interviewed: 30 with dengue-positive and 30 with dengue-negative febrile children. Mean total dengue-related costs did not differ from those of other febrile illnesses (31.5 vs. 27.2 US$, p = 0.44). Hospitalization almost tripled the costs of dengue (from 14.3 to 40.1 US$) and doubled the costs of other febrile illnesses (from 17.0 to 36.2 US$). To finance the cost of a febrile illness, 67% of households incurred an average debt of 23.5 US$ and higher debt was associated with hospitalization compared to outpatient treatment (US$ 23.1 vs. US$ 4.5, p < 0.001). These costs compared to an average one-week expenditure on food of US$ 9.5 per household (range 2.5-21.3). In multivariate analysis, higher socio-economic status (odds ratio [OR] 4.4; 95% confidence interval [CI] 1.4-13.2), duration of fever (OR 2.1; 95% CI 1.3-3.5), and age (OR 0.8; 95% CI 0.7-0.9) were independently associated with hospitalization. Conclusion: In Cambodia, dengue and other febrile illnesses pose a financial burden to households. A possible reason for a lower rate of hospitalization among children from poor households could be the burden of higher illness-related costs and debts.