Ethiopia's health system records a huge amount of data. Currently, DHIS2 (the major aggregate level data source) has more than a billion records, with eCHIS, EMR and other systems collecting additional vast demographic, clinical and other data. Yet at the subnational level, where care is delivered and most day-to-day decisions are made, much of that data mainly focus on collecting and reporting. Closing the distance between what we collect and what we act on is a remaining task in the health system. The Capacity Building and Mentorship Partnership (CBMP), a collaboration between the ministry of health and six research universities is working a scalable example of how we can triangulate multiple sources of data and translate them into action at point of service delivery.
In this partnership, universities bring the scientific expertise (data triangulation, drawing various data sources) and supporting the quality-improvement actions. The health system owns the problem, sits at the table, and puts the evidence to use. The pathway we follow is deliberate. raw data becomes information; information becomes evidence when several sources are read together; and evidence becomes knowledge for action only when it is shaped into a targeted, monitored action plan.
For instance, Addis Ababa University CBMP is closely working with Kirkos Sub-City triangulated EPI registers, DHIS2, PHEM surveillance and household surveys, revealed a widening MCV1 to MCV2 immunity gap (a major driver behind rising measles cases). A prioritization matrix helped them focus on the “change ideas” and the quality-improvement (strengthened use of tickler boxes, reminder calls and defaulter tracing, is currently being reviewed every week by the sub-city’s EPI unit.
Similarly, at the University of Gondar, the same six-step cycle focused on reducing neonatal mortality at the university’s specialized Hospital. Triangulating the hospital’s data revealed preventable losses, and the hospital is acting on them: renovating the NICU for infection prevention, strengthening referral linkages, and fixing challenges with vital equipment such as incubators, all being led by the hospital’s leadership. Other universities are also applying this data to action cycle focusing on various problems such as increasing CBHI enrolment, and enhancing long-term family planning.
Through these works, we have learned triangulated evidence provides more comprehensive view of the problem, because a single source alone may not inform us the whole story. With data available from various sources, the pressing task is to systematically translate them into action
Scaling this across the health system lies on three priorities; (1) run the data-to-action cycle as a routine, not a one-off; (2) co-identify the problem and co-own the evidence, so researchers and decision-makers discuss it in the same room, from the first question to the action; and (3) the PHCUs and hospitals should invest in the enablers (leadership and financing) that keep it sustained.