Nearly a quarter of hospital stays involve adverse events from healthcare errors, and nearly one in 10 cause serious harm, according to a study replicating the landmark 1991 Harvard Medical Practice Study (HMPS).
In a random sample of 2,809 admissions at 11 Massachusetts hospitals, 23.6% had at least one adverse event, 32.3% of which required substantial intervention or prolonged recovery, David W. Bates, MD, of Brigham and Women’s Hospital in Boston, and colleagues reported in the New England Journal of Medicine.
Fully 22.7% of the adverse events were judged to be preventable, with a preventable event happening in 6.8% of all admissions and a serious, life-threatening, or fatal preventable event in 1.0%.
These “disturbing” new findings “suggest that the safety movement has, at best, stalled,” said Donald M. Berwick, MD, MPP, of the Institute for Healthcare Improvement in Boston, in an accompanying editorial.
After the publication of the 2000 Institute of Medicine report “To Err Is Human: Building a Safer Health System,” built in large part on the 1991 HMPS data, improving patient safety was a priority in U.S. healthcare for a while, Berwick noted, but the decades that followed brought inklings of progress without a firm answer on whether the national healthcare system is safer since the report rang the alarm.
The study by Bates and co-authors doesn’t provide the definitive answer either, Berwick argued, as the methods differed sufficiently from the original to make direct comparison “tempting but … not warranted.”
The new study, like the one in 1991, randomly sampled admissions but added a trigger tool to help flag suspicious records and looked at certain types of harm that were not examined in the original, such as diagnostic errors and failure to treat decompensating patients. And, of course, the harder one looks for harm, the more one will find, Berwick noted.
In the 1991 HMPS, the rate of adverse events due to medical management seen in the 51 New York State hospitals evaluated was 3.7%, with 27.6% deemed due to negligence.
Also, “judging ‘preventability’ is not only difficult but may also be misleading,” Berwick wrote. “The more valuable approach is to regard all injuries as potentially preventable.” Nor did either iteration of the HMPS pay attention to “near misses.”
Bates and co-authors also acknowledged that many aspects of healthcare have changed over the past 34 years (i.e., in the time since the New York hospital records were first sampled for HMPS in 2018), including the shift to electronic health records and much care moving from the inpatient to ambulatory settings.
Pushing patient safety back to the top of the numerous urgent priorities, like supply-chain shortages and preparedness issues, facing healthcare systems today is a “sacred obligation” for all who sign up to “first do no harm,” Berwick noted. “Without renewed board and executive leadership and accountability for safety and without concerted, persistent investment in and monitoring of change, a summary study 34 years from now may again look all too familiar, with millions upon millions of patients, families, and health care staff paying the price for inaction.”
The updated HMPS was a retrospective look at 11 hospitals in Massachusetts with the same malpractice insurance carrier (a sponsor of the study) and selected to represent the range of large and small hospitals across three healthcare system. The random sample of admissions for adult patients occurring in 2018 excluded hospice, rehabilitation, psychiatric care, addiction treatment, and observation-only stays that didn’t cross two midnights. A group of nine trained nurses reviewed the admissions records to identify possible adverse events, defined as “unintended physical injury resulting from or contributed to by medical care that requires additional monitoring, treatment, or hospitalization, or that results in death” and including both errors of omission and commission. After review of confirmed events, a random sample of 10% were reviewed by a second physician.
Adverse event type prevalence tracked with those previously reported, with drug events being most common (39.0% of all events), followed closely by surgical or procedural events (30.4%). Events associated with nursing care, including falls and pressure ulcers, accounted for 15.0% of the total, and healthcare-associated infections represented 11.9%. The surgical events were most likely to be life-threatening, while the infections were most likely to be fatal.
Admissions involved at least one event that caused unnecessary harm but with rapid recovery — defined as significant harm — occurred in 18.6% of all admissions, while the rate of serious adverse events requiring substantial intervention or prolonged recovery was 7.5%, and 1.2% were life-threatening.
Of the seven deaths (0.2%) from adverse events, one was deemed to be preventable.
Patient characteristics associated with higher rates of adverse events included older age, male gender, Black or white versus Asian race, non-Hispanic versus Hispanic ethnicity, and private or Medicare insurance versus Medicaid.
Notably, the larger hospitals had higher adverse healthcare error adverse event rates than the smaller hospitals in the study. That variation from center to center “suggests that if hospitals had data that were more reliable and more routinely collected, it is possible that monitoring could be improved, adverse event rates could be reduced, and improvement strategies could be shared through careful study of interventions,” the researchers wrote.
“Other key organizational elements such as safety culture and strong leadership with respect to safety and quality are also needed to advance performance,” the team said. “Our findings are an urgent reminder to all health care professionals of the need for continuing improvement in the safety of the care we deliver.”
The study was supported by a grant from the Controlled Risk Insurance Company and the Risk Management Foundation of the Harvard Medical Institutions.
Bates disclosed relationships with AESOP, CDI Negev, EarlySense, FeelBetter, Guided Clinical Solutions, IBM Watson, MDClone, and Valera Health.
Berwick disclosed no relevant conflicts of interest.
New England Journal of Medicine
Source Reference: Bates DW, et al “The safety of inpatient health care” N Engl J Med 2023; DOI: 10.1056/NEJMsa2206117.
New England Journal of Medicine
Source Reference: Berwick DM “Constancy of purpose for improving patient safety — Missing in action” N Engl J Med 2023; DOI: 10.1056/NEJMe2213567.