In-hospital adverse drug reactions in older adults; prevalence, presentation and associated drugs—a systematic review and meta-analysis

Thumbnail Image
Jennings, Emma L. M.
Murphy, Kevin D.
Gallagher, Paul
O'Mahony, Denis
Journal Title
Journal ISSN
Volume Title
Oxford University Press
Published Version
Research Projects
Organizational Units
Journal Issue
Background: the prevalence of adverse drug reactions (ADRs) in hospitalised older patients, their clinical presentations, causative drugs, severity, preventability and measurable outcomes are unclear, ADRs being an increasing challenge to older patient safety. Methods: we systematically searched PubMed, Embase, EBSCO-CINAHL, the Cochrane Library, ‘rey’ literature and relevant systematic review bibliographies, published from database inception to March 2020. We included any study reporting occurrence of in-hospital ADRs as primary or secondary outcomes in hospitalised older adults (mean age ≥ 65 years). Two authors independently extracted relevant information and appraised studies for bias. Study characteristics, ADR clinical presentations, causative drugs, severity, preventability and clinical outcomes were analysed. Study estimates were pooled using random-effects meta-analytic models. Results: from 2,399 abstracts, we undertook full-text screening in 286, identifying 27 studies (29 papers). Final analysis yielded a pooled ADR prevalence of 16% (95%CI 12–22%, I2 98%,τ2 0.8585), in a population of 20,153 hospitalised patients aged ≥65 years of whom 2,479 patients experienced ≥ one ADR. ADR ascertainment was highly heterogeneous. Almost 48.3% of all ADRs involved five presentations: fluid/electrolyte disturbances (17.3%), gastrointestinal motility/defaecation disorders (13.3%), renal disorders (8.2%), hypotension/blood pressure dysregulation disorders/shock (5.5%) and delirium (4.1%). Four drug classes accounted for 57.8% of causative medications i.e. diuretics (19.8%), anti-bacterials (14.8%), antithrombotic agents (12.2%) and analgesics (10.9%). Pooled analysis of severity was not feasible. Four studies reported the majority of ADRs as preventable (55–95%). Conclusions: on average, 16% of hospitalised older patients experience significant ADRs, varying in severity and mostly preventable, with commonly prescribed drug classes accounting for most ADRs.
Older people , Adverse drug reaction , Polypharmacy , Multi-morbidity , Inpatient , Iatrogenic
Jennings, E. L. M., Murphy, K. D., Gallagher, P. and O’Mahony, D. (2020) 'In-hospital adverse drug reactions in older adults; prevalence, presentation and associated drugs—a systematic review and meta-analysis', Age and Ageing, afaa188, (11 pp). doi: 10.1093/ageing/afaa188
Link to publisher’s version
© The Author(s) 2020. Published by Oxford University Press on behalf of the British Geriatrics Society. All rights reserved. This is the submitted pre-copyedited, author-produced PDF of an article accepted for publication in Age and Ageing. The version of record is available online at