Public-private health engagements: Lessons from Kakamega County, Kenya

We’ve previously described the Strengthening Mixed Health Systems (SMHS) project supported by Merck for Mothers, which is working to generate actionable evidence that policymakers, health providers, trusted brokers, and others supporting mixed health systems can use to design and implement public-private engagements to improve maternal and newborn health and achieve UHC.

To generate this evidence, we’ve landscaped the global evidence and guidance around public-private sector engagement for health, proposed an ecosystem of factors influencing public-private engagement, and explored that ecosystem through six secondary case studies as well as primary engagements in Kenya and India.

The SMHS team is now ready to share what we’ve learned from our engagement in Kenya about how we can better understand and support public-private engagements to optimize mixed health systems through sector integration.

Our engagement took place in Kakamega County in western Kenya, where the maternal mortality rate is the 5th highest out of all of Kenya’s 47 counties. Public and private health system stakeholders in Kakamega share a concern about stubbornly high maternal mortality rates and the quality of maternal care in the county — providing a clear entry point for public-private engagement.

We had the opportunity to test out an approach for supporting public-private engagement in Kakamega while conducting robust primary research alongside the engagement. This research was designed to better understand both whether public-private engagement could be associated with outcomes — including improved quality of the engagement between sectors and improved health-related outcomes — and the factors associated with helping or hindering public-private engagement’s success.

So, what did we learn about public-private engagement in Kakamega County? Our research revealed several important lessons that we hope will be of practical use to those working within or supporting mixed health systems:

1. Support for public-private engagements can demonstrate potential to improve outcomes, even within a short timeframe.

During a global pandemic and over a relatively short timeframe, our research in Kakamega County revealed evidence of progress on many actions that were co-designed and implemented by the involved public and private sector actors. While not conclusive evidence of impact, this provides strong signals that even relatively short-term investments have the potential to improve engagement between the public and private sectors and ultimately, lead to improvements in health. The positive outcomes that we observed — both expected and unexpected — included:

Setting up a stakeholder forum for public, private, and other non-state actors to jointly identify and address challenges.
Significant progress on forming a private sector association to better enable private actors to engage with their public counterparts.
Broader health system outcomes such as increased and improved referrals between the two sectors and greater sharing of commodities and supplies from the public to the private sector.
Engagement-related outcomes in Kakamega County and beyond, including the appointment of a private sector liaison in the county government who is already connecting the private sector association to donors and partners to support their work.
Further, it appears that this work catalyzed plans to create similar private sector associations across all of Kenya’s counties — potentially setting up the possibility for greater public-private engagement across the country as a whole.

We did observe more progress on actions that were focused on improving engagement between the sectors, relative to actions that focused on improving health-related outcomes. It’s true that health-related outcomes are likely longer-term and require more time to “move the needle.”

However, it is also likely that engagement-related outcomes are foundational to improving health systems and outcomes — meaning that prioritizing improvements in how the sectors work together now creates greater potential for health improvements in the future. These changes — related to both the health system itself and to engagement between the sectors — provide signs of the ripple effects that even small and time-limited support for public-private engagements can have when the support takes a participatory and adaptive approach.

2. Bureaucracy can be a challenge, but high-level political support and better sector organization help.

The issue of political bureaucracy emerged as a hindering factor that sometimes stymied both high-level and day-to-day operations. Despite this, several other environmental factors were found to have helped the engagement, including high-level political support within the county — including from the governor himself, which made greater engagement possible as demonstrated through the following quote from a public sector actor:

The governor’s administration was open to the whole idea of engaging the private sector. It was an enabler. We didn’t have a policy direction of ‘these are competitors, let’s not engage.’ The environment was ‘they are stakeholders, let’s engage….’ We were encouraged.”

On the side of the private sector, the creation of the private sector association in Kakamega County helped private providers to organize and speak with a unified voice, enabling them to more easily and effectively engage with the public sector — an improvement attributable to the project itself.

3. Resource constraints and a lack of pre-existing policies are hindering, but these can be improved through better and more formalized engagement.

Resource constraints — both in terms of financial and human resources — across both sectors were one of the most cited hindering factors, as a lack of resources may have constrained some of the engagement activities between partners. The lack of pre-existing policies to govern public-private engagement in the county also came up as a structural hindrance: as one stakeholder noted, “I think the weak pre-existing structures for engagement slowed down the process, because it was like building a new house.”

While it may have felt like building a new house from the ground up, we also found that the formalization of the engagement through the formation of a public-private stakeholder forum — including its participants, terms of reference, and associated committees and technical working groups — helped to guide the engagement by providing a structure within which sector partners could interact. We even heard that the stakeholder forum and the engagement it enabled helped to promote joint resourcing, especially for human resources and commodities.

4. Engagement factors are important, too — and can be outcomes themselves.

Overall, we found that nearly all of the engagement factors in the public-private engagement ecosystem are important and can often be improved through better engagement between sectors.

With better engagement, foundational engagement factors such as trust can transform from challenges to positive outcomes. One defining characteristic of the foundational factors in Kakamega County — willingness to engage, trust, and mutual understanding between sectors — is the difference between our baseline and endline findings. In the baseline, these factors were overwhelmingly framed by public and private respondents as likely challenges, including perceptions that willingness to engage was low and quotes that demonstrated misunderstandings and mistrust between sectors. The responses at the endline, however, were much more positive.For example, partners from all sectors stated that will to engage was strong, as demonstrated through the following quote: “The private sector was very willing to engage as well as public sector. So everyone saw value to engage with the other…it was like a pre-existing hunger, it didn’t require so much energy to tell either party to engage.”

In fact, the engagement itself may have helped to improve the foundational engagement factors. Public sector stakeholders shared that the engagement helped government actors to change their perspective on the value of engaging with the private sector. They mentioned “bad blood” between the sectors in the past and shared that while at first public-private engagement may have been a novel idea, through this engagement they have become more receptive to engaging with private sector actors. While work remains to build trust and mutual understanding between the sectors, we heard that the engagement also helped to improve these factors.

Having a common goal across sectors is critical, but other operational factors are more of a mixed bag. The day-to-day operational engagement factors — engagement rationale, communication, technical and managerial capacities of the sector actors, and accountability — also came up frequently in our endline analysis. Engagement rationale, or the basis and motivation for engagement based on mutually beneficial common goals, was of particular importance and was largely viewed as a positive. Most respondents suggested that working toward similar goals — whether around reducing maternal mortality, improving health services, or tackling COVID-19 — helped the engagement to be more effective. However, our findings on the rest of the operational engagement factors were more mixed, suggesting that we might need more time to see how these factors play out in practice.

5. External shocks can be both crises and opportunities for public-private engagement.

The SMHS project faced many shocks over the course of the engagement in Kakamega County — not only did the engagement occur in the context of the COVID-19 pandemic, but it also faced political transitions and a medical worker strike, any of which could have upended the engagement’s progress.

While COVID-19 was cited as a challenge by many partners, there is also evidence that partners utilized these shocks in some ways to strengthen the engagement across sectors. In the case of COVID-19, our analysis found that the pressure of the pandemic gave diverse actors an incentive to work together and reprioritize, which may have ultimately helped the engagement. The medical worker strike, which started near the end of the engagement, acted as a catalyst for the public sector to increase referrals and strengthen the referral system with new private sector partners.

While shocks like these could have stopped progress, we instead observed great resilience and adaptation by partners to utilize the new public-private engagement to combat what could have been crippling challenges to the system.

So, what does all of this mean for supporters of public-private engagements?

While our research in Kenya focused specifically on outcomes and factors, our approach and learnings from Kakamega County do provide valuable insights into another key learning question: What approaches can engagement brokers or supporters take to facilitate factors that help engagements, mitigate factors that hurt, and ultimately help engagements achieve better health outcomes?

The support of a trusted broker is an important factor in itself. As you learned from our approach blog, a trusted broker played a critical role in the SMHS project by providing technical assistance and support to public and private sector partners throughout the engagement. What we heard made it clear that the continued capacity development support, facilitation, and technical assistance provided by the trusted broker was perceived as critical to this work and can help to ensure that the ecosystem of factors is as enabling as possible for effective engagement.
A trusted broker should be a local partner that knows the context — and should have adequate and flexible resources to succeed in their role. In addition to learning about the importance of the trusted broker role in facilitating public-private engagement, we also learned about what to consider when designing third-party support for an engagement — including: 1) having support from present, flexible and responsive local partners who understand the local context, and 2) the availability and flexibility of resources for the trusted broker to perform their role, including time, technical support, and in some cases, financial resources. These resources can be critical for enabling the trusted broker to help create a strong foundation, especially for new public-private engagements.
There is a need for further research on the role of a trusted broker. These learnings point to the immense value for flexible and intensive third-party facilitation through a trusted broker as support to foster new engagements. The role of a trusted broker is one that does not exist in many public-private engagements, but it is one that is worthy of further piloting and study, given the promise that emerged from the SMHS project.

From designing a trusted broker role, to considering important factors that could help or hinder success, to designing research around relevant system, engagement, and health outcomes, the SMHS project has aimed to start filling in gaps in the global evidence base around public-private engagement. We hope these lessons from Kakamega County, Kenya, are useful to researchers, practitioners, and supporters of these kinds of engagements in health in health. If you are interested in learning more about the Kenya engagement or other components of the SMHS work, you can find our full compilations of resources here.

The Strengthening Mixed Health Systems project is supported by funding from Merck, through Merck for Mothers, the company’s global initiative to help create a world where no woman has to die while giving life. Merck for Mothers is known as MSD for Mothers outside the United States and Canada.