The lack of reliable information about health facilities across sub-Saharan Africa became very clear during the COVID-19 pandemic. Amid a surge in emergency care needs, information was lacking about the location of facilities, bed capacity and oxygen availability, and even where to find medical specialists. This data could have enabled precise assessments of hospital surge capacity and geographic access to critical care. Peter Macharia and Emelda Okiro, whose research focuses on public health and equity of health service access in low resource settings, share the findings of their recent study, co-authored with colleagues.
What are open health facility databases?
A health facility is a service delivery point where healthcare services are provided. The facilities can range from small clinics and doctor’s offices to large teaching and referral hospitals.
A health facility database is a list of all health facilities in a country or geographic area, such as a district. A typical database should assign each health facility a unique code, name, size, type (from primary to tertiary), ownership (public or private), operational status (working or closed), location and subnational unit (county or district). It should also record services (emergency obstetric care, for example), capacity (number of beds, for example), infrastructure (electricity availability, for example), contact information (address and email), and when this information was last updated.
The ideal method of compiling this list is to conduct a census, as Kenya did in 2023. But this takes resources. Some countries have compiled lists from existing incomplete ones. Senegal did this and so did Kenya in 2003 and 2008.
This list should be open to stakeholders, including government agencies, development partners and researchers. Health facility lists must be shared through a governance framework that balances data sharing with protections for data subjects and creators. In some countries, such as Kenya and Malawi, these listings are accessible through web portals without additional permission. In others, such facility lists do not exist or require extra permission.
Why are they useful to have?
Facility listings can serve the needs of individuals and communities. They also serve sub-national, national and continental health objectives.
At the individual level, a facility list offers a choice of alternatives to health seekers. At the community level, the data can guide decisions like where to place community health workers, as seen in Mali and Sierra Leone.
Health lists are useful when distributing commodities such as bed nets and allocating resources based on the health needs of the areas they serve. They help in planning for vaccination campaigns by creating detailed immunisation microplans.
By taking account of the disease burden, social dynamics and environmental factors, health services can be tailored to specific needs.
Detailed maps of healthcare resources enable quicker emergency responses by pinpointing facilities equipped for specific crises. Disease surveillance systems depend on continuously collecting data from healthcare facilities.
At the continental level, lists are crucial for a coordinated health system response during pandemics and outbreaks. They can facilitate cross-border planning, pandemic preparedness and collaboration.
During the COVID-19 pandemic, these lists informed where to put additional resources such as makeshift hospitals or transport programmes for adults over 60 years of age.
The lists are used to identify vulnerable populations at risk of emerging pathogens and populations that can benefit from new health facilities.
They are important when it comes to making emergency obstetric and newborn care accessible.
What goes wrong if you don’t have them?
Many problems arise if we don’t know where health facilities are or what they offer. Healthcare planning becomes inefficient. This can result in duplicate facility lists and the misallocation of resources, which leads to waste and inequities.
We can’t identify populations that lack services. Emergency responses weaken due to uncertainty about where best to move patients with specific conditions.
Resources are wasted when there are duplicate facility lists. For example, between 2010 and 2016, six government departments partnered with development organisations, resulting in ten lists of health facilities in Nigeria.
In Tanzania, over 10 different health facility lists existed in 2009. Maintained by donors and government agencies, the function-specific lists didn’t work together to share information easily and accurately. This prompted the need for a national master facility list.
What needs to happen to build one?
A comprehensive list of health facilities can be compiled through mapping exercises or from existing lists. The health ministry should take responsibility for setting up, developing and updating this list.
Partnerships are crucial for developing facility lists. Stakeholders include donors, implementing and humanitarian partners, technical advisors and research institutions. Many of these have their own project-based lists, which should integrate into a centralised facility list managed by the ministry. The health ministry must foster a transparent environment, encouraging citizens and stakeholders to contribute to enhancing health facility data.
Political and financial commitment from governments is essential. Creating and maintaining a proper list requires significant investment. Expertise and resources are necessary to keep it updated.
A commitment to open data is a necessary step. Open access to these lists makes them more complete, reliable and useful.
Peter Macharia is funded by Fonds voor Wetenschappelijk Onderzoek- Belgium (FWO, number 1201925N) for his Senior Postdoctoral Fellowship.
Emelda Okiro receives funding for her research from the Wellcome Trust through a Wellcome Trust Senior Fellowship (#224272).
By Peter M Macharia, Senior postdoctoral research fellow, Institute of Tropical Medicine Antwerp And
Emelda Okiro, Head of Population Health Unit, KEMRI Wellcome Trust Research Programme
