Community case management and referral of children with fever within the primary health care system in Uganda.

Malaria remains a leading cause of under-five childhood morbidity and mortality in Sub- Saharan Africa and a third of febrile children lack prompt access to effective malaria case management services. Community health workers (CHW) provide an opportunity to deliver malaria diagnosis and treatment in primary health care settings closer to populations at risk. Recognising the potential of CHWs, many malaria endemic countries have national CHW programmes to provide case management services for malaria and other common childhood infections. However, for children that present to CHWs with signs and symptoms they are unable to manage current CHW guidelines indicate referral to the nearest health centre. Despite the scale-up of CHW programmes, there is little understanding of the referral processes and this thesis aimed to examine the processes as part of malaria case management trials with CHWs conducted in Rukungiri District, Uganda. The referral process was explored in four objectives that mirrored a child’s continuum of care beginning at the CHW and ending at the public health centre. The first objective assessed CHWs adherence to the referral guidelines; the second examined the effect of CHWs using malaria rapid diagnostic tests (mRDT) on referrals; and the third examined caregiver’s compliance to referral advice. Finally, the fourth objective described changes in outpatient department (OPD) visits and case-mix at health facilities over a 24- month period, before and after the start of the trials. The referral study was carried out as part of two cluster randomised trials which assessed the impact of CHWs using mRDTs on appropriate treatment of malaria when compared to presumptive diagnosis. Secondary analyses of routine CHW records revealed that two-thirds of children eligible for referral were not referred, including children with severe malaria. Although, the use of mRDTs improved the odds of referral compared to presumptive diagnoses, almost 90% of caregivers failed to comply with CHW referral advice which was associated with factors such as distance from health facilities, seasonality and treatment seeking during the weekend. Compared to the pre-trial period, OPD visits at health facilities declined and case-mix changed following the start of the trial. Overall these findings show the referral processes can be monitored using routinely available data using appropriate indicators. Changes in the number and type of visits seen at health facilities, indicate that health resources could be allocated differently. Most importantly, these data show that the referral process intended to ensure the care of children with referral signs and symptoms did not perform well. Referral is key component of primary health care and the process requires strengthening to ensure a continuum of care in community case management settings. New interventions should be developed involving CHWs, caregivers and health facilities to overcome the referral barriers and consequently improve CHWs adherence to referral guidelines as well as caregivers compliance to referral advice.