This community innovation has been accepted at the 2026 DHIS2 Annual Conference as a physical poster.
Middleware-Based Interoperability for Surveillance
Inpatient morbidity and mortality data provide essential confirmation of disease burden but are often underutilised in routine epidemiological surveillance due to separation between statistical reporting systems and surveillance platforms. In Sri Lanka, the electronic Indoor Morbidity and Mortality Return (eIMMR), maintained by the Medical Statistics Unit, captures ICD-10–coded discharge diagnoses and outcomes from hospitals nationwide. While comprehensive for statistical reporting, this data was not routinely accessible to epidemiological focal points for timely surveillance interpretation or decision-making. This intervention focused on enabling operational interoperability between the national morbidity reporting system and the DHIS2-based epidemiological surveillance platform through a dedicated middleware layer. The middleware was designed to extract daily discharge diagnosis and outcome data from eIMMR, apply mapping, harmonisation, and validation rules, and transmit structured records to DHIS2 using the DHIS2 Web API. The integration was implemented without altering existing hospital or statistical workflows, ensuring continuity of routine reporting while enabling automated data exchange. Daily synchronisation of ICD-10–coded inpatient data allowed epidemiological teams to access near real-time hospital morbidity and mortality trends within the surveillance environment. This supported validation of notifiable disease trends, identification of discrepancies between hospital burden and surveillance notifications, and improved interpretation of disease severity and outcomes. By making discharge diagnosis data routinely available within DHIS2, the integration strengthened evidence-based planning of surveillance activities and prioritisation of epidemiological follow-up. The implementation demonstrates a practical and sustainable middleware-based interoperability model for linking national morbidity statistics systems with DHIS2. This experience provides a replicable approach for countries seeking to operationalise routine use of inpatient discharge data to strengthen epidemiological surveillance and decision-making.
Primary Author: Pramil Liyanage
Keywords:
Interoperability, DHIS2, middleware, inpatient morbidity and mortalitydata, ICD-10, eIMMR, Epidemiology, Medical Statistics, hospital discharge diagnosis, disease surveillance, health information systems, data integration, public health informatics, decision-making, API, Sri Lanka
2 Likes