Evaluating the Impact of a Public Health Centre Intervention on Management of Malaria and Health Outcomes of Children in Uganda: Results from the PRIME & PROCESS Studies

"The PRIME intervention had a small positive impact...on proximal outcomes, including health worker behaviour and appropriate treatment of malaria. However, these positive effects did not extend to the community level, where no differences were seen in health indicators between the children from intervention and standard care communities..."

This policy brief describes the impact of the PRIME study, conducted by the ACT Consortium to find out whether a multi-component intervention delivered at public health centres in Tororo district, Uganda could improve health outcomes of children and treatment of malaria, as compared to the current standard of care. The PRIME study findings were supplemented by the PROCESS study, an evaluation of the process, context, and wider impact of the PRIME intervention.

In 2009-2010, the ACT Consortium, which is funded through a grant from the Bill & Melinda Gates Foundation to the London School of Hygiene & Tropical Medicine, carried out formative research, which included a census survey, a survey of health workers, and a qualitative study. Through this research, they identified barriers and aspirations for quality health care in what is a rural area with limited infrastructure and education levels and very high malaria transmission. (Health workers and community members shared ideals of what constituted good care; social challenges were also identified, including low health worker motivation and difficult relationships between health workers and community members due to lack of trust, language barriers, discriminatory behaviours, and requests for informal payments for services.) Prior to the trial, delivery of supplies to health centres in the study area, including the artemisinin-based combination therapy (ACT) drug artemether-lumefantrine or AL (the first-line recommended treatment for malaria in Uganda) was unpredictable.

The findings from this formative research were considered in the context of literature reviews on previous interventions as well as theory of behaviour change and adult learning, and were discussed with stakeholders in Uganda. They then designed an intervention which aimed to attract patients to seek care and to improve the quality of care delivered at public health centres. The hope was that the intervention could be sustainable by the Ministry of Health and district partners in Uganda. In addition to efforts to ensure the supply of mRDTs and AL, the intervention consisted in training in fever case management and use of rapid diagnostic tests for malaria (mRDTs), workshops in health centre management (HCM), and workshops in patient-centred services (PCS). For the PCS modules, organisers aimed to strengthen providers' relationships with patients, colleagues, and the community by reorienting the care-seeking experience towards patients' aspirations for good quality care - communicated in training activities using scenarios and discussion points relevant to health workers' everyday experiences that stressed the importance of asking good questions, employing active listening, reassuring patients, building rapport, creating a positive work environment, etc.). The 18 manuals for delivering the training/workshops are available here.

The workshops and materials were designed from adult learning principles, using interactive processes to support incorporation of new ideas, tools, and principles into everyday practice. Through interactive, weekly 3-hour workshops, the HCM and PCS modules aimed to build a supportive community of practice through a cyclical process of change where individuals' frames of reference are transformed through participation in a community of peers, and their participation in turn transforms the community. Through this process, learners engage with other community members and reflect critically on their practice through a social process of individual and collective learning. In a paper cited below in the Source section, the researchers explain: "We theorized that, as health workers built, demonstrated, and received positive feedback on their clinical, interpersonal, and managerial skills, the social processes emerging from participation in the community of practice would help them to develop their professional identity and sustain positive skills and behaviours." The workshops were structured as a 6-step adult learning cycle drawn from Kolb's experiential learning theory, which includes 4 stages (experience, reflection, conceptualisation, and planning) and from Knowles' theory of adult learning, which asserts that adults must first establish why they should learn something before proceeding to acquiring new knowledge. The PCS module also included weekly self-observation activities (SOA) that aimed to stimulate learners' purposeful critical analysis of their knowledge and experience, enabling them to engage and deal with their emotions and develop appreciation and respect for others. Semi-structured SOAs followed by feedback in groups provided opportunities for both individual learning and change as a community. Editor's note: For more on the process of developing the intervention, see: "Behind the scenes of the PRIME intervention: designing a complex intervention to improve malaria care at public health centres in Uganda", by Deborah D. DiLiberto, Sarah G. Staedke, Florence Nankya, Catherine Maiteki-Sebuguzi, Lilian Taaka, Susan Nayiga, Moses R. Kamya, Ane Haaland, and Clare I. R. Chandler, Global Health Action, Vol. 8, October 23 2015.

The PRIME study, conducted from May 2011 to April 2013, was designed to evaluate the impact of the intervention delivered at public health centres using a cluster-randomised design in Tororo district. Twenty lower-level health centres from 7 sub-counties were randomly assigned to the intervention (10 centres) or to standard care (10 centres). The 10 health centres that were assigned to the intervention received the intervention package. To evaluate the impact, researchers conducted 3 cross-sectional community surveys, followed a cohort of children under 5, and conducted patient exit interviews and monthly surveillance at the health centres. The PROCESS study was carried out alongside PRIME to help understand if and how the intervention worked. It included: an evaluation of the implementation of the intervention activities from the perspective of implementers, health workers, community members, and key stakeholders; a context evaluation to capture information on factors that may have affected the implementation of the intervention or outcomes and; an impact evaluation to assess the wider impact of the intervention beyond outcomes of the PRIME study.

The researchers developed a logic model to set out the intended pathway of change from the PRIME intervention inputs through to the community level outcomes. This illustrated how they expected change to occur - and the conditions required to support change - at the health centres and in the communities. They anticipated that the PRIME intervention would influence treatment seeking behaviour amongst community members, leading them to be more likely to seek care at the health centre, receive better care for febrile illnesses when they attended the health centre, and have better treatment outcomes leading to reduced anaemia and parasitaemia.

The evaluation found that, overall, use of RDTs was much higher in patients attending intervention health centres (52% intervention vs 7% standard care). However, rapid diagnostic testing appeared to have little impact on the use of AL, with approximately half of all patients receiving a prescription for AL in both groups (50% intervention vs 53% standard care). There were no improvements in health outcomes of community children: The primary outcome of PRIME was the prevalence of anaemia (haemoglobin 11.0 g/dL), and there was no difference in prevalence of anaemia or parasitaemia between the intervention and standard care groups. Results of the patient exit interviews suggest that management of febrile illnesses was better in the intervention health centres. However, the difference was small and not significant.

Most of the PRIME intervention was carried out as intended and learning objectives were met, although not all health workers received the training. Of 52 health workers, most attended some training, but only 8 attended all sessions, and 8 attended none; 2 of 10 in-charges did not attend the HCM training. The community was aware of the intervention at the health centres, and was happy with ACTs & RDTs. Small improvements were seen in fever case management, which were attributed to RDTs as well as training and supervision. The researchers also saw positive improvements in patient-centered services in some intervention health centres. Small improvements were also seen in the way health workers interacted with patients. A larger change may have been possible if health workers were less overworked and if change was being taken on more widely in the health system hierarchy and politics. The intervention appeared to improve malaria case management, communication between health workers and patients, and patient satisfaction with care.

However, the intervention was limited compared with demands on health centres, which had further needs of improvement. It was found that HCM training failed to address the changing dynamics at health centres. The frequent supervision visits made by the study to collect data from registers seemed to have a greater impact on information management than the training workshop on this topic. Patients also sought care at numerous non-governmental sources; given that a variety of care options are available (public, private, non-governmental organisation, or NGO), community members appeared to position themselves for opportunities to attain the most convenient and least expensive care.Consequently, the intervention made little difference to treatment seeking overall. In short, improvements that were seen were insufficient to prompt systematic changes in treatment seeking behaviour; broader health centre changes and additional malaria prevention measures will be required in this high malaria transmission setting.

Reflecting on these findings, the researchers surmise that the pathway of change broke down at 2 points: at the point of changing treatment-seeking practices, and at the point of improving fever case management. In theory, introducing RDTs in health centres will reduce "over-prescription" of AL, improving targeting of antimalarial treatment and fever case management, thus resulting in better treatment outcomes. However, this theory is not applicable in high transmission settings such as Tororo, where weak health systems, poverty, and malaria create a cycle of poor health care. Thus, to improve quality of health care within the public sector, infrastructure and wider systems and political issues must be addressed.

Click here for the 2-page brief in PDF format (English).
Click here for the 2-page brief in PDF format (French).
Click here for the 2-page brief in PDF format (Portuguese).