News & Commentary

20 August 2015

Learning together to improve county health systems in Kenya

By Edwine Barasa*


This blog post is part of a series on people-centred research methods for health systems development published in conjunction with a Twitter chat on the same topic. Please see below for links to other blog posts in this series. 

It has been argued that the potential for health policy and systems research (HPSR) to contribute to health system improvement is dependent on, among other things, it becoming more people-centered in how it is conceived, conducted, and utilized. People centered HPSR recognizes that researchers are themselves actors within health systems and that their decisions about research themes, questions, methods, presentation and interpretation of findings are critical influencers of the dynamics and functioning of health systems. Therefore to be people-centered HPSR researchers should see themselves as part of the system, rather than as external observers, and actively engage with various actors within the system. HPSR knowledge should be co-produced through an interactive process among researchers and other health system actors, based on trust, continuous dialogue, and a shared commitment to change. By being people-centered, HPSR produces knowledge that has a shared sense of ownership among actors in the health system, is strategic, in the sense that it is relevant, timely and useful to health systems decision making, and therefore enhances the uptake and translation of this knowledge into practice. 

Our research in Kenya is exploring and examining how the governance and accountability mechanisms of country health systems have interacted with and are influenced by decentralization of government structures. We have adopted a “learning sites” approach in our research. The Kenyan leaning site is one of three RESYST linked learning sites, the others being in Mitchells Plain in Cape Town and Sedibeng in Johannesburg, both in South Africa. Working in two counties in coastal Kenya, Kilifi and Mombasa, our approach entails embedding ourselves in the contexts in which we carry out our research (county health departments, hospitals, PHC facilities), and collaboratively working with health system actors (managers, practitioners, policy makers) to “learn” the system; identify problems, formulate research questions and explore them and propose solutions. The three learning sites (one in Kenya and two in South Africa) are also learning from each other across sites.

Decentralization in Kenya

To understand the context of our research, a little background about what is happening in the Kenyan health system is useful. After the passing of a new constitution in 2010, and the general election in March 2013, Kenya transitioned from a centralized government into a devolved system of government. Under this new governance arrangement, the country is organized into two levels, the national government, and 47 quasi-autonomous county governments. These two levels of government are distinct but interdependent and should conduct business on the basis of consultation and cooperation. The authority for decision-making, budgeting, and management of several key functions were transferred to county governments. The health sector is the most devolved sector in this new arrangement. Under devolution counties have the responsibility for all health service delivery functions while the national government has the responsibility for health policy formulation, regulation, as well as the management of specialized national referral services.

How does the learning site work?

One of our researchers, Benjamin Tsofa, focuses on examining the effects of devolution on county health sector governance and accountability, my work focuses on examining priority setting and resource allocation practices in county hospitals, while Sassy Molyneux and Mary Nyikuri’s focus on examining accountability mechanisms in primary healthcare facilities in the face of changes occasioned by devolution. 

We have employed a number of strategies:

  1. Preparatory phase: Extensive engagement and deliberations between our research team and health managers/policy makers in the development our research protocol. These health system actors are therefore co-investigators in our work.

  2. Qualitative case study methods: A phenomenon is examined and analyzed in detail and depth using research tools that are most appropriate to the nature of the inquiry. The case study approach is suitable for examining complex social phenomena that are often highly contextual.

  3. Prolonged engagement with study sites: For example, I spent a total of 7 months in 2 case study hospitals. This prolonged engagement allows us to integrate ourselves with the systems that we study, develop a deeper appreciation of the system and build trust among the actors in the system that allows us to ask hard questions, and get to the core of the problems faced by the systems. Indeed, often, health managers identify us as one of their own, share their opinions openly and expect us to help is some of the day to day operations in their areas of work.

  4. Regular reflective practice sessions: These are regular formalized meetings where we (the researchers) invite health managers and jointly reflect on our (researcher and health manager/policy maker) work. Health managers often share the challenges faced by learning site health system, relevant research questions and potential problems, while we share our research plans, processes and findings. The reflective practice sessions offer a space for the co-production of a shared understanding of the health system, the challenges faced, relevant research questions, interpretations of research findings and potential solutions to health system challenges.

This is why we call it the “learning sites” approach: a process where both researchers and health managers/policy makers co-learn and co-produce an understanding of phenomena we seek to study, how we should study, how to interpret the outcomes of our study and how to use these outcomes to improve the health system.

Learning sites are a people-centred approach

We see our learning sites approach as inherently people-centered. By “learning together” and co-producing health system knowledge, our approach has resulted in greater ownership and acceptance by health managers and policy makers of the outcomes of our research. Whereas it is commonplace to find researchers struggling to gain audience with policy makers and to get research translated into practice, in our case, health managers continue to request for our research outputs, and our advice in solving health system problems in their contexts. Further, our prolonged engagement, collaborative approach and reflective practice have allowed us to gain trust among health system managers and policy makers at both the county and national levels in Kenya. Health systems managers see us as allies, recognize the utility of our work and continue to consultant us regularly. Our researchers have been incorporated in decision making and technical committees at both the county and national level to provide technical advice and research evidence in these exercises. Our experience in Kenya therefore reiterates the observation that for HPSR to achieve its potential in shaping and changing health systems, it must be people-centered.

* Edwine Barasa is a health economist, working as a post-doctoral research scientist at the KEMRI-Wellcome Trust Research Programme (KWTRP) in Nairobi, Kenya. Over the past three years he, together with colleagues Benjamin Tsofa, Sassy Molyneux and Mary Nyikuri have been involved in an exciting research adventure under the umbrella of the Resilient and Responsive Health Systems (RESYST) research collaboration.


Other blog posts in the People-Centred Research Methods series:


Photo: Learning site in Kilifi county, Kenya


Archived comments