Reflection

Reviewing the evolution of patient safety over time, we see a discursive shift from harm to “error.” The “error” framing is used to advance the authority of scientific-bureaucratic managerial medicine and to diminish the traditional authority of clinical expertise. Psychologist Sidney Dekker noted that four different voices appear in patient safety discourses: epistemological (what happened?), preventative (how can it be prevented?), boundary-maintaining (were there violations or impermissible activities?), and existential (what is the reason for this suffering?). Discussions in one voice tend to be dissatisfying for the others, but the “error” framing addresses all four: an error occurred, errors can be prevented, violators should be punished, moral wrong leads to suffering. In summary, patient safety’s rise resulted from five factors: a general decrease in risk tolerance, the industrialization of healthcare, reframing harm as “error,” brief input from safety sciences holding out potential for improvement, and medicine’s effort to retain control of healthcare internally. Ironically, these factors also tended to make patient safety activity ineffective since they bounded out insights, skills, and theories from the safety sciences.

Pressure Builds for Reform

Horrific medical accidents widely circulated in the media: Betsy Lehman, Boston Globe health reporter died from a chemotherapy overdose; in Florida, Willie King had the wrong leg amputated. These scandalous stories killed organized medicine’s efforts at tort reform because no one could reasonably support it after such injuries. In the aftermath, the first Annenberg Conference on error in medicine was proposed to help medicine “get on the right side of the issue.” Lucian Leape and James Reason provided keynote addresses, symbolizing a partnership between medicine and cognitive psychology. The Ben Kolb case presented at Annenberg spurred organized medicine to begin serious safety efforts, and the National Patient Safety Foundation was started by the American Medical Association. A second, even larger Annenberg Conference was held with substantial input from nonclinical safety scientists.

The Special Case of Anesthesia

Anesthesia became the only medical specialty to undertake systematic and dramatic improvements in safety over time. Evidence suggests that this process began through the fortuitous engagement of engineers in anesthesia work, supported by respected leaders in the field. The goal was not simply to solve a problem. The aims were too deeply understand the nature of the technology, the work, and the complex interactions that take place in work as carried out. Oddly, healthcare more generally failed to emulate these efforts. This state of affairs may be attributed to the substantive influence of non-clinical safety scientists in anesthesia, and also to differences in widely accepted methodological and investigative research approaches.

Rationalizing Healthcare

Healthcare changed dramatically during the 20th century; industrialization lead to contention for control over medical work and there was a shift from autonomous, professional control to managerial, hierarchical control, which included a strong belief in technical rationality. The rise of scientific-bureaucratic medicine combined ideas from Taylorism (i.e., the “one best way.” separation of planning from activity, and the rise of a managerial class) and a rationalizing turn in medicine (e.g., evidence-based medicine) to produce a hybrid group of clinician-managers, who can retain control of healthcare in health professionals’ hands. The quality improvement effort preceded the safety movement in health care by about a decade.

Beginnings

The To Err is Human report rests upon a series of historical episodes—the early development of safety thinking in healthcare, the evolution of safety thinking in the safety sciences, and accumulating stories of personal suffering and tragedy experiences by patients. Concerns about the safety of healthcare date at least as far back as the Code of Hammurabi. Early efforts at improving safety appeared sporadically in the eighteenth and nineteenth centuries, but these were isolated and not sustained. Malpractice concerns were also entangled with safety, and the first malpractice crisis in the United States came about due to advances, rather than deficiencies, in care. Public and governmental interest in safety more generally developed out of railway accidents in the nineteenth century. Notions of accidents evolved over time, moving from accident proneness to the domino model, to more complex models such as Reason’s Swiss cheese model. Libby Zion's death results in New York state legislation regulating medical trainees’ supervision and duty hours, marking a change from a self-regulating profession to a more typical service industry. Recognition that health professionals involved in adverse events are also, in a sense, victims begins to grow.

Setting the Stage

The Institute of Medicine report, To Err Is Human, which led the news in late 1999 with a charge that “medical errors” were killing 44,000 to 120,000 Americans each year, gave patient safety “celebrity status.” But the emergence of patient safety as a major concern on the public agenda was not a given. In this chapter, the authors raise questions about why it developed at this time even though the problem of medical harm had long been known, why it was understood as “error” as opposed to “risk” or “hazard,” and, finally, who gained control of the reform effort and to what end.

The Future of Patient Safety

Patient safety has been evolving and changing since its conception, and this evolution will inevitably continue as it competes for attention with other social problems and as healthcare struggles with conflicts among increasing workloads, performance and production pressures, technological innovation, and increasingly limited human and economic resources. Patient safety is, at present, a reform movement becalmed, captured by the industry it set out to reform. The authors see four possible futures: congealing into bureaucracy (already in progress), rebranding as something else (similar to quality improvement rebranding itself as patient safety), simply dying out, or fundamental reform. Fundamental reform is unfortunately the least likely possibility. It will require healthcare to relinquish its dominant position in patient safety and to develop substantive, equal partnerships with safety sciences.

Irruption onto the World Stage

The Institute of Medicine (IOM) report To Err Is Human created a media firestorm. Prior to this, interest in patient safety might be described as a cult—a small group of passionate believers—but the IOM report and its aftermath moved patient safety to a prominent place on the broad public agenda, creating pressure for quick action. The federal government announced a safety program at the Agency for Healthcare Research and Quality; President Clinton committed to a “50% reduction in medical errors in five years”; similar events occurred in the United Kingdom. Programmatic efforts in patient safety began to crowd out safety scientists. The litany of fear-provoking incidents of patient harm continued.

Halting Steps

Early studies of medical harm appeared in the 1950s, with rates of injury little different from those reported 50 years later. Philosopher Ivan Illich criticized medicalization of everyday life, using annual deaths from auto accidents as an example. Malpractice concerns became entangled with safety, and the first malpractice crisis in the US came about due to advances, rather than deficiencies, in care. The Harvard Medical Practice Study (HMPS) renewed interest in medical harm as a cause of malpractice suits in a series of four papers in the prestigious New England Journal of Medicine. Although it was not a pre-defined outcome of the study, one paper by Lucian Leape reframed the problem as one of medical “error”; the other three did not use the term. Lucian Leape fortuitously drew on error research in cognitive psychology and other safety science work on “error” stemming from the research triggered by the Three Mile Island nuclear disaster and other accidents. He formulated those concepts in a form digestible by health professionals and published them in the widely read medical journal JAMA—Journal of the American Medical Association. The figure of 100,000 annual deaths was first circulated.

Consolidation and Stagnation

Safety activity is prevalent, but little of value is produced. Several scandals tainted the patient safety movement. Diagnostic “error” became popular but in a highly medicalized way vulnerable to hindsight bias. Safety science moved on from linear, complicated models to emergent, interactive models based on complexity. In 2015, 15 years after the IOM report, four separate reports on the state of patient safety concluded that it had accomplished little and that the approach to safety must be changed dramatically. But the change they recommended was no different from that recommended in the Institute of Medicine report 15 years earlier.