Most COVID-19 vaccine adverse events detected within 48 hours in Ethiopia: implications for cohort event monitoring efficiency in resource-limited settings.

Publication date: Jul 24, 2026

The World Health Organization recommends cohort event monitoring with a 30-day follow-up period for vaccine safety surveillance in low- and middle-income countries (LMICs). However, in resource-limited settings, operational and logistical constraints necessitate evaluation of the efficiency and feasibility of this follow-up duration. We conducted an analysis of a prospective active surveillance study of 10,948 adults who received the Pfizer COVID-19 vaccine in Ethiopia (August 2021-June 2022), using immediate observation and telephone follow-up on days 2, 4, 7, 14, 21, and 30. Among 10,948 participants, 4,236 (38. 7%; 95% CI: 37. 8-39. 6%) reported at least one adverse event following immunisation (AEFI), with 6,174 total events recorded. Of all events, 88. 2% were detected within 48 h (1. 8% immediate, 86. 4% at day 2), while 11. 8% were identified during subsequent follow-up between days 4 and 30. Seven reactogenic symptoms comprised 93. 8% of AEFIs. No serious AEFIs or adverse events of special interest were identified. Over nine months, 138 healthcare professionals captured predominantly mild, expected reactions through extended surveillance. Most COVID-19 vaccine adverse events in Ethiopia occurred within 48 h of vaccination, empirically suggesting that extended full-cohort CEM could yield diminishing informational returns in low- and middle-income country settings. A hybrid surveillance approach combining intensive short-term active surveillance during the first 48 h, enhanced passive reporting, and sentinel hospital networks with adequate diagnostic capacity could offer a more sustainable and operationally realistic framework. These findings are offered as an evidence-based reference for national regulatory authorities planning or conducting emergency vaccine safety surveillance in comparable contexts, supporting a transition from product-centred extended monitoring towards patient-centred, resource-adapted pharmacovigilance. Not applicable.

Semantics

Original Article