Tanzania PHIT Partnership: The Connect Project
Thursday, November 12, 2015
Thanks to decades of a national focus on rural health care, Tanzania has one of the highest densities of health facilities among African countries, alongside services proven to reduce maternal and child mortality. Yet, Tanzania is experiencing a “bottleneck” of quality health care at the community level, according to Dr. Helen Semu, the deputy director of public health education at the Ministry of Health and Social Welfare. Utilization of the before-mentioned resources is very low, with Tanzania struggling to reach its targets for improvement. The mortality rate for infants less than five years of age remains about 84 per 1,000 live births for the wealthiest 20 percent of the population; for the lowest 20 percent, the rate is about 103 per 1,000 live births. Neonatal mortality comprises 30 percent of child deaths.
Fifteen years ago, major concern over these areas helped lead to the establishment of Millennium Development Goals 4 and 5, two of the goals set by United Nations member states with a shared goal of substantially improving maternal and child health by 2015. With that said, as government leaders began to institute strategies to improve health outcomes for mothers and children, they also began to realize that their health systems were much weaker than originally assumed. Tanzania specifically has observed stagnant maternal and child mortality rates due to gaps between levels in its health infrastructure. A growing body of research has revealed that a solution to this systemic issue may lie in interventions targeted at primary care delivery, specifically the effectiveness of community health workers (CHWs); also, experiments across Africa and Asia have shown that competent CHWs increase use of existing health services through household visits and treat patients with more cultural sensitivity than facility-based workers. However, Tanzania lacked a context-specific, evidence-based trial to test if a professionally trained and salaried cadre of community health workers could have an impact on reductions in maternal and child mortality rates.
The response to this need came in the form of the Connect Project (Connect), which arose with the support of the African Health Initiative (AHI), a multi-million dollar initiative launched in 2007 at the Doris Duke Charitable Foundation with a focus on strengthening health systems in sub-Saharan Africa. AHI supported five Population Health Implementation and Training (PHIT) partnerships to design, implement and evaluate large-scale models of care that incorporate both research about implementation of health care and workforce training into primary health care delivery.
“Was there any value in bolstering the capacity of the community health workers?” was the question at the heart of the project, explained Dr. James F. Phillips, the Connect Project’s principal investigator at Columbia University’s Mailman School of Public Health.
At the start of the project, the Tanzanian government had already enacted national policies focused on increasing community access to primary health care by supporting a broad array of mostly volunteer community health workers and establishing more village health dispensaries. This set of policies was called the Primary Health Services Development Programme (Mpango wa Maendeleo wa Afya ya Msingi, MMAM). The Connect Project’s role was to provide scientific data proving whether or not a cadre of professionalized trained CHWs could bolster MMAM and accelerate the reduction in mortality rates.
Key to the Connect Project’s effort was the cooperation of primary stakeholders in Tanzanian health. The project’s research was co-led by a team from Columbia University’s Mailman School of Public Health and the Tanzania-based Ifakara Health Institute (IHI), a health research organization possessing one of the world’s largest demographic surveillance systems (DSS). Access to the DSS-observed population gave the team the ability to conduct a randomized control trial testing the new cadre of CHWs and comparing districts that did not have access to the CHWs. Dr. Ahmed Hingora, the principal investigator at IHI, carried a legacy of leadership at the Ministry of Health as a previous architect of MMAM and was central to translating the knowledge gained throughout the project period to key policymakers. Dr. Helen Semu at the Ministry of Health and Social Welfare brought an even deeper national policy perspective to the research. The Tanzania Training Center for International Health (TTCIH), with its intellectual leadership and credibility in the country, offered essential training for the community health workers. Under Connect, these diverse parties united to select, hire and train a new cadre of CHWs to learn how it might relieve the “bottleneck” of quality service delivery at the primary health care level, thereby bolstering Tanzania’s entire health system.
“In the past, these workers have always been volunteers and trained in non-specific activities—an extension of vertical programs,” said Dr. Phillips. “There’s been an absence of professionalization. They start off well, and then the capability of managerial programs atrophies.”
The Connect Project faced a number of challenges while testing the new cadre of community health workers. These challenges often resulted, perhaps not unexpectedly, from crucial gaps in the larger health system. For example, fractures in the medical supply chain resulted in limited supply of essential medicines (e.g., iodine or malaria diagnostic kits); also, the addition of new CHWs created the need for more supervisors in an already understaffed system. Fortunately, the six-year design of the project allowed for on-the-ground learning and adaptations, including the development of feedback mechanisms so that supplies were added later on in the project timeline, and more staff were hired and trained to oversee the program.
“There were significant gaps that we faced that we were only able to fill with donor support,” said Mallory Sheff, Connect’s program manager.
Funding from AHI, in partnership with Comic Relief UK, enabled the Connect Project to fill those gaps and ultimately unearth significant findings about the value of a new cadre of community health workers. With detailed registration information from community patients, household surveys of over 3,000 women and regular health facility audits, the project team uncovered a dynamic and multidimensional picture of the positive influence the new cadre of CHWs had on improving services to community members. For example, they discovered that pregnant women in many villages often made the rational choice to bypass their closest primary health care facility because they knew the quality was lower there compared to facilities farther away. This was despite the fact that Tanzanian law mandates that babies are delivered at the nearest facility. This lack of nearby quality care resulted in the complete absence of services for pregnant women in rural villages—services that the project’s newly trained workers could now provide. Another outstanding result of the project was the finding that workers recruited with the designated qualifications and formally trained were well-suited to manage basic but important cases of diarrhea, malaria and pneumonia and a wide set of reproductive child health services. At the national level, the Connect Project team was also able to operationalize new processes of referrals, feedback and communication between village and district authorities so that they could better respond to CHW problems or locally specific needs.
According to Dr. Phillips, the CHW model promoted a “generalized equity effect” in which paid, well-trained workers boosted the provision of “doorstep” services to remote households characterized by relatively high levels of poverty and lack of proper health care. This new cadre of CHWs was unique in that they also provided treatment in the community for diarrhea, malaria and pneumonia. “It’s this package of services and the investment in the model that now has credibility,” he said, referencing results from the first two years of project evaluation. “If these initial, very promising results are sustained, the Connect scale-up could become a national pillar of MMAM.”
“The country now has evidence that if CHWs are well-supported, they can contribute towards reduction of maternal and child deaths as well as address the drivers of the most prevalent diseases and reach out to rural communities,” said Dr. Semu. “The key is to invest in the development of a system that integrates a CHW cadre.”
A possible next step for the Tanzanian health system could be to move from experimental mode to national implementation of a robust CHW model. This requires extremely thoughtful work to link the existing and varied CHW activities led by other institutions and recruit, train and supervise the next cohort of CHWs in other regions. It is also imperative that local government authorities take ownership of the CHW model in their regions and districts. Given Tanzania’s 188 diverse health councils, 26 regions and the required 1,500 workers per region, Dr. Semu has estimated that this part of the process could take between 7-10 years.
“The roll-out plan is step-by-step, including learning the successes and challenges of implementation in different contexts,” said Dr. Semu, referring to the need for continued research. “There will be unique lessons to be learned in each region as the program evolves.”
Looking ahead, the project team agrees that the Ministry of Health and Social Welfare’s leadership must be empowered by donors and government partners to lead future efforts to strengthen Tanzania’s health system. The Ministry is ideally and exclusively suited to implement nationwide health policies and also delegate the relevant administrative and implementation tasks to regional, district and community level players. Thus, the effectiveness of CHW-based policies across Tanzania ultimately hinges on the Ministry of Health’s centralized capacity to drive them forward.
In the words of Dr. Malick Kante, Connect’s co-principal investigator and monitoring and evaluation coordinator, the Connect Project has supplied the Tanzanian government with a convincing amount of essential, research-based evidence that “adding a strong community health workers program improves Tanzanians’ health and improves maternal and child health and likelihood of survival.” The team continues to interpret final impact data to assess improvements in under-5 mortality rates and total fertility rates among other indicators. In doing so, they aim to understand even further how a long-term, national investment in the CHW model could impact health in Tanzania.
 Ramsey et al.: The Tanzania Connect Project: a cluster-randomized trial of the child survival impact of adding paid community health workers to an existing facility-focused health system. BMC Health Services Research 2013 13(Suppl 2):S6. Available at: http://www.biomedcentral.com/1472-6963/13/S2/S6.
 More information on the Connect Project can also be found through Columbia University's Mailman School of Public Health Web site (http://arches.columbia.edu/implementation/projects/) and through Ifakara Health Institute's Web site (http://www.ihi.or.tz/projects/the-connect-project).