Exploring medication-related safety incidents and avoidable harm in English prisons: a combined national retrospective record review study

Abstract Introduction Medication-related safety incidents in prisons occur frequently and present unique challenges for healthcare delivery. Despite a growing awareness of this phenomenon, there is limited understanding of the nature and system-level factors contributing to medication incidents and avoidable harm in prison healthcare settings to guide improvement efforts. Aim To characterise the nature of medication-related incidents and subsequently map their contributory factors to the Systems Engineering Initiative for Patient Safety (SEIPS) framework.[1] Method Multi-source retrospective descriptive analysis of medication safety events from English prisons using two datasets: National Reporting and Learning System (NRLS) safety incidents submitted by prisons (2018–2019, n = 1123) and cases of avoidable medication incidents identified by doctors and nurses as part of a national cross-sectional systematic medical record review study[2] including both substandard care (n = 32) and avoidable harm (n = 35) (2020–2022). All incidents were reviewed and coded to identify incident type(s), outcomes and harm severity. Additionally, incident free text narratives containing descriptions of locations (e.g. ‘meds hatch,’ ‘prison wings’), care processes (e.g. ‘medication administration,’ ‘prescription expiry’), staff interactions, and/or environmental constraints (e.g. ‘patient placement in distant wings,’ ‘rigid regime schedules’) were coded for systematic identification and mapping of contributory factors to the SEIPS framework[1] across five components: people, tasks, tools/technology, environment, and organisation. Results Most incidents resulted in no harm across NRLS (1015/1123, 90.4%) and substandard care datasets (32/32, 100%), while the avoidable harm dataset showed 65.7% (23/35) resulted in moderate harm and 14.3% (5/35) caused severe harm or death. Incident types varied by dataset: administration/supply incidents predominated in NRLS (681/1123, 60.6%), while prescribing incidents were most frequent in both avoidable harm (15/35, 42.9%) and substandard care incidents (12/32, 37.5%). Omitted medicine/ingredient was the leading incident subtype across all datasets (NRLS: 21.3%; avoidable harm: 68.6%; substandard care: 31.3%). Within 45.2% of reported NRLS incidents there were identified contributory factors, whilst this figure was 87.5% for substandard care cases and 100% of avoidable harm incidents, with 97.1% containing multiple factors. Organisational factors (including communication failures, staffing issues, and policy gaps) were most prominent in both NRLS data (48.8%) and substandard care cases (33.9%), while task/process factors (such as medication reconciliation failures and handover problems) were commonly involved in avoidable harm incidents (35.1%). Conclusions Medication-related safety incidents in prisons arise from complex multi-factorial systemic issues that may vary by incident severity and detection method. The SEIPS framework analysis reveals that system-level contributory factors, particularly organisational and task/process factors, drive medication safety failures in prison healthcare, requiring comprehensive system-wide interventions rather than individual-focused approaches. A strength of this study is the triangulation of two national datasets providing more comprehensive coverage of medication incidents across harm severity levels. However, a limitation is the known underreporting and brevity of recording observed with voluntary incident reporting systems, which may not capture all safety events with enough detail to fully characterise their antecedents.