PURPOSE

The aim of this study was to estimate the prevalence of mortality among patients due to adverse drug reactions that lead to hospitalisation (fatal ADR), to explore the heterogeneity in its estimation through subgroup analysis of study characteristics, and to identify system-organ classes involved and causative drugs for fatal ADR.

METHODS

We identified prospective ADR-related studies via screening of the PubMed and Google Scholar databases with appropriate key terms. We estimated the prevalence of fatal ADR using a double arcsine method and explored heterogeneity using the following study characteristics: age groups, wards, study region, ADR definitions, ADR identification methods, study duration and sample size. We examined patterns of fatal ADR and causative drugs.

RESULTS

Among 312 full-text articles assessed, 49 studies satisfied the selection criteria and were included in the analysis. The mean prevalence of fatal ADR was 0.20% (95% CI: 0.13-0.27%; I = 93%). The age groups and study wards were the important heterogeneity modifiers. The mean fatal ADR prevalence varied from 0.01% in paediatric patients to 0.44% in the elderly. Subgroup analysis showed a higher prevalence of fatal ADR in intensive care units, emergency departments, multispecialty wards and whole hospitals. Computer-based monitoring systems in combination with other methods detected higher mortality. Intracranial haemorrhage, renal failure and gastrointestinal bleeding accounted for more than 50% of fatal ADRcases. Warfarin, aspirin, renin-angiotensin system (RAS) inhibitors and digoxin accounted for 60% of fatal ADR.

CONCLUSIONS

ADR is an important cause of mortality. Strategies targeting the safer use of warfarin, aspirin, RAS inhibitors and digoxin could reduce the large number of fatal ADRcases.