Submission ID 130134
| Issue/Objective | Routine immunization remains one of the most effective and equitable public health interventions, yet in many low- and middle-income countries, paper-based systems still delay reporting, weaken follow-up of children who miss doses, and limit real-time accountability. As countries invest in digital public infrastructure to sustain essential services during disruption, electronic immunization registries are increasingly promoted as tools to strengthen continuity of care, data quality, and service responsiveness. However, rigorous evidence of their effect under routine implementation conditions remains limited. In Rwanda, where immunization performance is strong but paper-based tracking constrained timely, individual-level monitoring, the government introduced a DHIS2-based electronic immunization registry, e-Tracker, with national rollout between September 2019 and January 2020. This study assessed whether rollout altered routine immunization trends and what Rwanda's experience offers for equitable and resilient health systems. |
|---|---|
| Methodology/Approach | We conducted a national quasi-experimental interrupted time series evaluation of Rwanda's electronic immunization registry (e-Tracker), a DHIS2-based platform for individual-level immunization data. The analysis was embedded within a broader mixed-methods evaluation of early e-Tracker implementation but focused specifically on its programmatic effect on routine immunization uptake. We used 66 months of aggregated monthly data from January 2018 to December 2024 on DTP-HepB-Hib3 doses administered, together with monthly information on COVID-19 and lockdown periods from the Rwanda Biomedical Centre; population estimates from the Fifth Rwanda Population and Housing Census were used to derive vaccination rates. National rollout occurred between September 2019 and January 2020 following system customization and training that began in May 2019, with implementation support delivered to vaccination focal persons and data managers in 505 health centres and 40 district hospitals across public, faith-based, and non-profit facilities. We first examined temporal patterns descriptively at district, province, and national levels, then fitted segmented interrupted time series regression models to compare pre- and post-rollout trends. After descriptive exploration of trends by district, region, and nationally, we fitted segmented regression models comparing pre- and post-intervention trends. Adjusted models accounted for calendar month to capture seasonality, district-level heterogeneity, lockdown periods, the broader COVID-19 shock, and autocorrelation through a lagged outcome ter to improve model fit and account for temporal dependence. |
| Results | Monthly DTP-HepB-Hib3 doses varied substantially over the study period, with visible fluctuation before rollout of e-Tracker and major disruption during the COVID-19 period. In the unadjusted interrupted time series model, the post-intervention slope shifted in a positive direction, but this change was not statistically significant (β=1.09, p=0.484), and model fit was poor. After adjustment for seasonality, district-level heterogeneity, lockdown periods, the broader pandemic effect, and temporal dependence, model fit improved markedly (adjusted R²=0.923). In this adjusted model, the post-rollout slope remained positive but was still not statistically significant (β=0.71, p=0.238), while the COVID-19 period was associated with a substantial decline in administered doses (β=-44.46, p<0.001); lockdown periods themselves were not significantly associated with doses administered. Overall, the findings indicate that any early programme effect of e-Tracker on immunization uptake was difficult to isolate in the context of concurrent pandemic disruption and transition in reporting practices. However, the positive change in trend direction after rollout is consistent with the registry contributing to stronger monitoring foundations during a period of major health-system stress. |
| Discussion/Conclusion | This study suggests that the value of electronic immunization registries should not be judged only by immediate changes in coverage, particularly when implementation coincides with major health-system disruption. In Rwanda, national rollout of e-Tracker occurred during the COVID-19 period and before full transition to paperless workflows and complete interoperability, making early programme effects difficult to isolate. Yet the experience remains highly relevant for global health: it shows how embedding a registry within national digital health architecture can strengthen the enabling conditions for equitable service delivery, including timeliness of data, accessibility of records, and capacity for routine monitoring and follow-up. In fragmented and uncertain settings, these system effects matter. Rwanda's experience therefore supports a broader shift in digital health policy, from viewing registries as stand-alone technologies to treating them as part of digital public infrastructure for resilient primary care. The implication for countries and partners is clear: investment should combine technology, interoperability, workforce support, and evaluation approaches that measure both health-service performance and system adaptability. |
| Presenters and Affiliations | Francine Uwamahoro University of Rwanda, Center of Excellence in Biomedical Engineering and eHealth Francine Uwamahoro University of Rwanda Andrew Muhire Ministry of Health Piero Mazimpaka Irakiza Ministry of Health_National Health Intelligence Centre Emmanuel Christian Nyabyenda University of Rwanda, Center of Excellence in Data Science Hassan Sibomana Rwanda Biomedical Center Jeanine Condo CIICHIN |
