The effect of reducing the direct cost of care on health service utilization and health outcomes in Ghana : a randomized controlled trial

ABSTRACT AIM To improve malaria control in children less than five years of age. OBJECTIVES To assess the impact of reducing the direct cost of health care on utilization of health services and morbidity associated with malaria among children under 5 years old. Primary objective To compare the prevalence of anaemia (Hb<8g/dl) among children six to fifty-nine months of age, from households with and without improved financial access. Secondary objectives included: To compare the health service utilization rate among children from households with or without improved financial access; to compare household all-cause mortality between the trial arms, to examine primary and major secondary outcomes by socio-economic status of households, to compare outcomes among children from self-enrolled and trial-enrolled households; and to document community knowledge and attitudes on user fees. METHODS 2332 households containing 2757 children less than 5 years of age in Dangme West, Ghana were randomised either to be enrolled into a pre-payment scheme operating allowing free access to primary care, including drugs or to a control which paid user fees for their health care as pertained normally. These included 138 households, comprising 165 children, who had enrolled in the pre-payment scheme prior to the closure of the registration window were included in the study as an observational arm. 2 RESULTS 2194 households comprising 2592 children were allocated randomly into control and intervention groups. 138 households, comprising 165 children had themselves enrolled in the pre-payment scheme. At baseline, the two randomised groups were the same, but the group who self-enrolled were significantly less poor and had better health outcome measures. Introducing free primary healthcare significantly altered the healthcare seeking behaviour of households, with those randomly allocated to the intervention arm using formal healthcare more (95% Cl 1.04-1.20; p=O.OOI) and home treatment and chemical sellers less than those in the control group. This did not, however, lead to any measurable difference in any of the health outcomes at the end of the six-month trial period. For the primary outcome of moderate anaemia there was no difference seen OR 1.05 (0.66-1.67). CONCLUSIONS This study suggests that reducing the direct cost of health care has a significant impact on healthcare-seeking behaviour but cannot, on its own, be assumed to have an impact on health outcome