Pharmaceutical Management Responses, Medication Safety and Patient Outcomes during Medicine Shortages and Stock-outs in Health Facilities: A Systematic Review

Background When a medicine is missing, facilities substitute, conserve or seek other supplies, changing what reaches the bedside and the patient. This evidence is divided between pharmacy, nursing and outcome studies and between income settings. This review synthesized studies relating responses to medication-safety consequences and patient outcomes. Methods Following PRISMA 2020, Scopus, PubMed/MEDLINE and OpenAlex were searched on 7 October 2026, and citations of included studies were searched. Eligible studies were English-language journal articles from 2020 with empirical data on at least two of response, consequence and outcome. Two authors independently screened 1,476 records and assessed 170 full texts, plus 23 from citation searching. Appraisal used MMAT 2018, synthesis followed SWiM with effect-direction coding, and confidence was rated with GRADE-CERQual. Results We included 142 studies (143 reports) from 69 countries; 110 (77.5%) were from high-income countries and 123 (86.6%) from hospitals. Therapeutic substitution was the most frequent response (85 studies; high confidence). In measured hospital comparisons, mortality, length of stay and readmission mostly showed no clear difference (moderate confidence), whereas worse disease control was reported mainly when long-term treatment was interrupted or switched (moderate confidence). The share of respondents associating shortages with medication errors ranged from 1.1% to 77.7%, and four studies measured errors (moderate confidence). In low- and middle-income settings, staff and patients reported referral to buy medicines elsewhere (moderate confidence). Procurement, inventory and governance responses were least often connected to any consequence or outcome (high confidence). No study reported occupational exposure or outcomes of nutrition commodity stock-outs. Conclusions Evidence is densest for clinical decisions in high-income hospitals and thinnest for supply and governance responses and for primary care in low- and middle-income countries. Substitutes without a close therapeutic equivalent and interrupted long-term treatment warrant monitoring, and studies with concurrent controls and measured errors and outcomes are needed.