Evidence-based mitigation strategies, specifically targeting preventable patient harm, are necessary to reduce the number of patients exposed to preventable harm across healthcare settings, according to research results published in the BMJ.
Through a systematic review and meta-analysis, researchers sought to examine the prevalence, severity, and nature of preventable patient harm across medical settings. In total, 66 studies including 70 independent samples were included in the review.
The final study total included 33 studies conducted in the United States, 27 conducted in Europe, and 10 conducted elsewhere. The most common design for these studies was retrospective or cross-sectional (71%), followed by prospective (29%). All studies assessed the preventability of patient harm through the use of consensus procedures and trained reviewers.
The total pooled sample included 337,025 patients, 8% of whom experienced harmful incidents and 55% of whom experienced preventable harmful incidents. Overall, 47,148 harmful incidents were identified; 55% of these were preventable.
The pooled prevalence of patient harm was 6% (95% CI, 5%-7%; I²=99%); median prevalence was 5% (interquartile range [IQR] 3%-9%). The pooled prevalence of overall harm, both preventable and nonpreventable, was 12% (95% CI, 9%-14%; I²=99%) with a median of 10% (IQR 7%-15%).
When stratified by medical setting, the highest pooled prevalence estimate of preventable patient harm was found in intensive care and surgical settings (intensive care: 18%; 95% CI, 12%-26%; I²=96%; surgery: 10%; 95% CI, 7%-13%; I²=97%). Conversely, the lowest pooled prevalence was found in obstetrics (2%; 95% CI, 0%-4%; I²=95%). No evidence was identified in psychiatry; researchers stated a need for additional research in this specialty area.
The pooled proportions of mild, moderate, and severe harm were 49%, 36%, and 12%, respectively. Drug management and other therapeutic management incidents “accounted for the highest proportion of preventable patient harm” (25% and 24%, respectively); this was followed by incidents related to surgical procedures, healthcare infections, and diagnosis (23%, 16%, and 16%).
Univariable analyses indicated that preventable patient harm prevalence was higher among studies based in “advanced” specialties, like surgery and intensive care (regression coefficient b=0.08; 95% CI, 0.05-0.11), as well as in studies with “relatively small sample sizes” and studies with populations that included children and older adults (b=0.03; 95% CI, 0.01-0.06 and b=0.03; 95% CI, -0.01 to 0.05, respectively).
Multivariable model analyses found that only the medical care setting (b=0.07; 95% CI, 0.04-0.10) “remained a significant predictor of prevalence of preventable patient harm…suggesting that the prevalence of patient harm is higher in advanced medical specialties compared with studies in general hospitals.”
Study limitations include the variability of the prevalence of preventable patient harm across studies, the retrospective nature of many of the studies included, and the likelihood of evolving preventability rankings due to technological advancements in the healthcare field.
“Our findings provide a useful agenda of priority areas for mitigating preventable patient harm,” the researchers concluded. “Priority areas are the mitigation of major sources of preventable patient harm…and greater focus on advanced medical specialties.”
Panagioti M, Khan K, Keers RN, et al. Prevalence, severity, and nature of preventable patient harm across medical care settings: systematic review and meta-analysis. BMJ. 2019;366:l4185.
This article originally appeared on Medical Bag