Reporting of suspected adverse drug reactions: a systematic review comparing clinical settings and pharmacovigilance systems.

The proportion of suspected adverse drug reactions (ADRs) that are formally reported to official pharmacovigilance databases is unclear, although this is essential for patient safety. This systematic literature review aimed to compare the level of formal reporting of ADRs to pharmacovigilance databases with the level of informal reporting of ADRs identified in clinical settings. Further, it aimed to investigate factors that could influence formal ADR reporting. A systematic review was conducted in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement. A review protocol was registered a priori. Embase and Medline were searched for studies published between 1st January 2015 and 17th June 2025 that were based on original data and provided quantitative comparisons between formal and informal ADR reports for any medication type. Reviews, meta-analyses, randomized controlled trials, case reports, case studies, letters to editors, and nonhuman studies were excluded. Four independent reviewers screened the studies and extracted the data pairwise. For each study, we calculated the proportion of identified suspected ADRs that were not formally reported to pharmacovigilance databases and presented them using median and interquartile range (IQR). Risk of bias was assessed using the Critical Appraisal Skills Programme (CASP) checklist. The certainty of evidence was rated using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach. Of the 12,558 studies identified, 11 cross-sectional studies were included in the review. These studies included a total of 38,962 patients across all age groups (range 0-96 years) and investigated different medications. The nonreporting estimates ranged between 71.7 and 100%, with a median rate of 98.05% (IQR 95.5-99.3). Formal ADR reporting was low regardless of severity or seriousness, suggesting that these factors did not meaningfully influence reporting rates. The studies were heterogeneous and had considerable risk of bias according to CASP. The overall certainty of evidence was rated as very low. A large proportion of ADRs suspected in clinical settings are not formally reported to pharmacovigilance databases. This could constitute a patient safety risk, and future studies should identify key factors that influence formal ADR reporting in order to guide improvements.