More than 1 in 20 patients will experience some kind of harm during their healthcare, of which half of all cases could be prevented, conclude UK researchers in a systematic review that highlighted advanced specialties such as surgery and intensive care as conferring the highest risk.
Maria Panagioti, PhD, senior lecturer, NIHR Greater Manchester Patient Safety Translational Research Centre, University of Manchester, and colleagues looked at data on more than 330,000 patients in 70 samples from around the world.
Preventable Patient Harm ‘A Serious Problem’
The research, published by The BMJ on 17th July, showed that 12% of patients experienced harmful incidents, with 6% having incidents that the researchers considered to be preventable.
Of those, 12% were severe, causing permanent disability or death. Almost half were drug and treatment management-related incidents and almost a quarter related to surgical procedures.
The highest prevalence of preventable harm was seen in surgical and intensive care units, while the lowest was recorded for obstetrics, with no difference between different regions of the globe.
The team writes that the results “affirm that preventable patient harm is a serious problem across medical care settings”.
The researchers say that the “priority areas” in tackling the problem are the “mitigation of major sources of preventable patient harm”, such as drug incidents and a greater focus on advanced medical specialties.
“It is equally imperative to build evidence across specialties such as primary care and psychiatry, vulnerable patient groups, and developing countries,” as well as improvements in the “assessment and reporting standards of preventability”.
Leading Cause of Morbidity and Mortality
Harm experienced by patients during healthcare is a leading cause of morbidity and mortality, equivalent to multiple sclerosis or cervical cancer in developed countries and tuberculosis or malaria in developing countries, say the authors.
With increasing recognition that some harm, such as adverse drug reactions, are not preventable, there has been greater focus on preventable harms.
This, the researchers write, could include errors or omissions by healthcare professionals, health system failures, or both.
Noting that none of the systematic reviews on patients harm so far have focused on preventable harm, they conducted a search of the Medline, Cinahl, Embase, Pubmed, WHOLIS, Google Scholar, SIGLE, and PsycINFO databases.
They looked for prospective, retrospective and cross sectional studies conducted in any medical care setting in any geographical area and published since the turn of the century.
They classified preventable harms by severity, and into those that were drug-related, diagnostic, medical procedure-related, and healthcare-acquired infections.
Review of Evidence
From an initial 7313 citations, the researchers included 66 studies reporting on 70 independent samples, of which 33 were in the USA, 27 in Europe and 10 elsewhere. Fifty samples were retrospective or cross-sectional in design, and 45 originated in general hospitals.
From these studies, a pooled sample 337,025 patients was selected, which included a total of 47,148 harmful incidents, of which 25,977 (55%) were considered preventable.
The pooled prevalence of preventable patient harm was 6%, versus a prevalence for overall patient harm of 12%.
The highest prevalence of preventable patient harm was reported for intensive care, at 18%, and surgery, at 10%, while the lowest prevalence was seen in obstetrics, at 2%.
Almost half (49%) of preventable patient harms were mild in nature, while 36% were moderate, and 12% were severe or led to death.
The most common types of incidents related to drug management, in 25% of cases, followed by other treatment management incidents in 24%, surgical procedures in 23%, healthcare infections in 16%, and diagnosis in 16%.
Univariate analysis suggested that the prevalence of preventable patient harm was higher in studies of advanced specialties such as surgery and intensive care, in those with smaller sample sizes, and in studies involving children and older adults.
On multivariate analysis, only the medical care setting remained a predictor of the prevalence of preventable patient harm, with the risk significantly increased in advanced hospital specialties (p<0.001).
There was no association between preventable patient harm and the World Health Organisation region in which the study was conducted, the study design or the way in which preventable harm was assessed.
‘Serious Concerns’ Raised
Irene Papanicolas, PhD, Department of Health Policy, London School of Economics and Political Science, London, and Dr José Figueroa, Department of Medicine, Harvard Medical School, Brigham and Women’s Hospital, Boston, USA, discussed the study in an accompanying editorial.
They say that it “raises serious concerns about the safety of health systems”.
However, they note that there is “no consensus as to what constitutes preventable harm, and even experienced clinicians vary in the extent to which they agree on whether an error is preventable”.
Papanicolas and Figueroa believe that the definition used in the current study has its limitations, in particular the linking of a patient harm to a specific process or standard of care, which misses the complex interplay that characterises harm.
Moreover, they believe that many harms are “only partially preventable, and therefore distinguishing the preventable from the inevitable becomes difficult”.
They suggest that, to better understand preventable patient harm, it will be necessary to capture all potential risks, not just adverse events, improve the ability to detect harm, and to engage patients and the public in identifying causes of harm.
Nevertheless, Irene Papanicolas and José Figueroa say that the study “serves as a reminder of the extent to which medical harm is prevalent across health systems, and, importantly, draws attention to how much is potentially preventable”.
The study was funded by the UK General Medical Council. The NIHR Greater Manchester Patient Safety Translational Research Centre funded the corresponding author’s time spent in this project. The research team members were independent from the funding agencies.
No other funding or conflicts of interest declared.
BMJ 2019;366:l4185 doi: 10.1136/bmj.l4185
BMJ 2019;366:l4611 doi: 10.1136/bmj.l4611