There has been considerable discussion and many publications on the frequency, cost, and harm of adverse events in our hospitals. The most influential publication on this matter has been the Institute of Medicine report, “To Err is Human” (Kohn LT, Corrigan JM, Donaldson MS, eds.), which was issued by National Academies Press in 1999.
The report was and is eye opening in its description of the frequency, magnitude, and breadth of medical and nursing errors, administrative and trustee laxity, harms to patients, and errors that can result in fatalities and permanent disability. The report set what the authors believed was an achievable target of a 50% reduction in adverse event in 10 years.
The reaction to the report of governments, state and federal, and hospital systems and associations was relatively swift and loud, and the public awareness of the problem was increased substantially.
But has all this activity paid off? Have outcomes improved since 1999? A paper in the 25 November 2010 issue of the New England Journal of Medicine attempts to answer this question (Landrigan CP, et al. Temporal trends in rates of patient harm resulting from medical care, NEJM 363: 2124-2134, 2010).
A brief description of this retrospective study follows.
A stratified, random sample of 10 hospitals in North Carolina was chosen for study. (North Carolina has been a leader in efforts to improve hospital safety, so their results may be a bellwether of wider hospital trends.)
The review included a total of 100 admissions per quarter from January 2002 through December 2007 that were reviewed by both internal and external nurse reviewers. They used the Institute for Healthcare Improvement's Global Trigger Tool for Measuring Adverse Events. Two independent physician reviewers evaluated suspected harms identified on initial review. The authors measured trends in the occurrence of harms and the frequency of preventable harms.
The study found that there were 25 “harms” per 100 hospital admissions. Multivariate analyses by internal reviewers showed no significant changes in harms per 1000 patient days or in the rate of preventable harms. External reviewers found a reduction in preventable harms that was not statistically significant and found no significant change in the overall rate of harms.
Harms Remain Relatively Common
In short, there was no significant improvement over the six-year period. This supports the belief of investigators in Europe and North American that harms remain relatively common.
Why this is so is the key question. The authors point to several potential weaknesses in this type of study, such as: despite the random selection, the hospitals chosen were not representative of North Carolina hospitals or hospitals elsewhere; only 10 hospitals were studied; record reviews are limited to what is found in the record.
However, the care taken in the design and execution of the study to avoid chronological, reviewer, and other kinds of bias seemed impressively scrupulous to me. If we then accept that the study had no significant technical flaws or biases, we must consider other reasons for the lack of improvement over the six-year period despite national attention and the commitment of substantial resources to diminish these problems.
The authors state that they cannot rule out smaller positive changes that might be detectable by larger or longer studies and they point to impressive changes reported from individual health systems in specific areas such as the reduction of catheter-related infections.
But that begs the question of why no improvement was found in this sophisticated study with experienced investigators and an all-star team of advisors.
So I offer a few conjectures from my own unscientific observations of many hospitals over many years.
* Hospitals: The technology of medicine has changed radically over the past decade or two, but hospitals have changed little in structure and function. The organizational structure of most hospitals is corporate with layers of vice presidents and middle managers, endless committee meetings with and without physicians or nurses, and too often, with patients viewed as clients or customers as they would in a department store.Hospitals certainly are in business and must make a profit to survive. But the corporate “top-down” mindset is ill suited for first-rate patient care that requires specialized personal attention from a series of professionals and staff, all synchronized for the patient's benefit and all personally accountable for what happens to the patient. A prime responsibility of hospital management is to put structures, processes, and personnel in place to assure the best possible environment for reducing errors and eliminating preventable errors and adverse outcomes.However, actions by this top-down structure reflect largely the experience and values of non-caregivers. Doctors and nurses are often included in safety committees and discussions, but who is really responsible? Responsibility for harms and errors is not firmly placed and neither hospital administrators (nor anyone else) suffers any consequences because the rate of preventable adverse events has not improved over time.I have never heard of anyone being fired or even having all or part of their substantial bonuses withheld for consistently poor performance in patient safety. Until trustees take safety and efficiency that seriously, it will be difficult to improve the outcomes.
* Physicians: The role of the physician in reducing hospital errors and harms is complex. The relationship of the hospital and physicians varies, but is often one of tolerance rather than collegiality, especially so in most hospitals whose physicians that provide hands-on patient care are self-employed and use the hospital rarely except if they are specialists that require sophisticated equipment and staff, such as operating rooms and ICUs.Thus, most physicians have a purely utilitarian relationship with hospitals and though they may sit on committees, they have little or no responsibility for safety outcomes. Hospitals may have medical directors who are deeply engaged and bear responsibility. But in my experience most medical directors are appointed to blunt criticism from attending physicians, but have little significant authority and bear little or no responsibility for safety outcomes.
* Nurses: The greatest resource for reducing errors is underutilized in my view. Care-giving nurses (not the clipboard-carrying type) spend more time with patients by far than any other professional in a hospital. Yet, “bedside” nurses remain underappreciated and often underpaid in the modern hospital. Nurses are often in the best position to identify a structure or process that increases the risk of adverse events. But few hospitals empower these in-the-trenches nurses to make changes or try new, safer approaches than the “same old, same old” way of doing things.In summary, it is my opinion that more authority and responsibility should be assigned to those in a position to identify and make changes, particularly those who provide care. I also believe that there should be consequences for the lack of improvement in the rate of preventable harms at all levels that bear responsibility, starting at the top with trustees and administrators on down to caregivers. This may mean no bonus, required training attendance, or more serious consequences for repeated failure.Lip service and the lack of improvement are unacceptable at a time when the hospital has become the single most dangerous place for many patients.