Evaluation of the viral haemorrhagic fever surveillance system in Uganda (2020-2025): assessing system performance, critical gaps, and opportunities in strengthening surveillance

Abstract

Background: Recurrent outbreaks of viral haemorrhagic fevers (VHFs) continue to threaten public health security in Uganda. Although investments have been made in surveillance and outbreak preparedness, the time required for case detection, notification, and response, along with incomplete multisectoral information sharing, continues to pose significant challenges. This study measured the effectiveness of the VHF surveillance system in some districts of Uganda between 2020 and 2025 and found the key gaps and opportunities for strengthening surveillance based on CDC evaluation attributes and WHO framework 7-1-7. Methods: A convergent parallel design was used, which involved a retrospective review of 640 suspected VHF case records from Mubende, Kyegegwa and Mbarara districts and ten key informant interviews. All indicators of the WHO 7-1-7 timeliness indicators, data completeness, laboratory turnaround time, and positive predictive value (PPV) were used to assess surveillance performance. Quantitative data were analysed in Stata 18, and Qualitative data in a thematic approach to identify gaps in human, animal and environmental health surveillance data flows. Results: There were 640 suspected VHF case records, 75.2% of which were recorded in Mubende. Full 7-1-7 compliance was achieved in 182 (28.4%) records. 62.8% (402) of the records were detected within seven days, 268 (42%) were notified within one day, and 318(49.7%) were initiated within seven days. Data completeness (90% or more of the essential fields completed) was found in 337 (52.7%) records, and 182 records (28.4%) were missing or had inconsistent critical dates. The median laboratory turnaround time was 5 days and 120/312 (38.5%) laboratory turnaround time of 3 days or less. The overall PPV of all suspected cases with definitive laboratory results was 20.6% (48/233). There were no significant differences found between districts on 7-1-7 indicators and data completeness. Qualitative results showed that the community detection was delayed, the One Health coordination was poor, the involvement of the wildlife sector was limited, and the interoperability of information systems on surveillance was poor. Conclusion: The performance of the VHF surveillance system was found to be different across the selected districts with notification being the lowest performing part of the 7-1-7 surveillance pathway. Enhancing community-based detection, notification, surveillance data quality, lab processes and systematic multi-sectoral sharing of information will be critical to enhance early warning and outbreak response.