The Value of Patient Safety Learning Laboratories

Patient safety learning laboratories (PSLLs) are teams that have come together to improve patient safety, supported by the Agency for Healthcare Research and Quality (AHRQ), and this special issue of Journal of Patient Safety (JPS) includes multiple reports from these teams. The concept of PSLLs was designed to develop, test, and implement innovative solutions for improving patient safety in health care settings. The “secret sauce” is that laboratories bring together clinicians, engineers, human factors experts, data scientists, and other stakeholders to apply a systems approach to patient safety challenges. Systems engineers and health care safety experts have not come together to the extent they should have, and the PSLL concept was intended to enable this. PSLLs have been designed to develop, test, and implement innovative solutions for improving patient safety in health care settings.

Most health care institutions have struggled to support systems engineers over the long term to work on designing care more successfully. There are a few organizations that have been exceptions, for example, Henry Ford under the leadership of Dr Vinod Sahney. Dr Sahney got his PhD from the University of Wisconsin, Madison, in 1970, and for 25 years served as Senior Vice President at Henry Ford Health System, where he worked on improving the diagnostic processes there, as well as patient engagement and care delivery, and developed a “closed loop” system for diagnostic tests and referrals.¹ Johns Hopkins University and the Armstrong Institute for Patient Safety and Quality, under Dr Peter Pronovost’s leadership, also involved systems engineers in many safety efforts. Another would be the University of Wisconsin under Dr Pascale Carayon’s leadership, where she developed the Systems Engineering Initiative for Patient Safety (SEIPS) models, which offer frameworks that enable integration of human factors and ergonomics into health care quality and patient safety. SEIPS 1.0 focused on work systems and their impact on health care processes and outcomes, SEIPS 2.0 expanded the model to include patient and caregiver perspectives,² and SEIPS 3.0 expanded it further to include the patient journey, adding a longitudinal dimension.³ But overall, such successes have been hard to generalize.

The tools of systems engineering can be highly effective at designing, evaluating and improving numerous processes in healthcare, ranging from throughput in the coronary catheterization laboratory, to treatment of sepsis, to prevention of bloodstream infections to give a few examples. AHRQ recognized the potential benefits of this work if it could be spread more broadly across health care, especially focused on safety. The Harvard Medical Practice Study⁴ and the subsequent “To Err Is Human” report of the National Academy of Medicine⁵ had made it clear that far too many people were being harmed by the care they received, and AHRQ devoted considerable attention to how safety might be improved, with one of the major programs being supporting the PSLLs.

This special edition of JPS includes a series of reports from the PSLLs related to the work they have accomplished in improving safety. Examples of the topics covered include medication reconciliation in the emergency department, interruptions during trauma cases, clinical decision support failures in pediatric intensive care, an evaluation of safety concerns for pediatric mental and behavioral health patients and providers in the emergency department, the design of general internal medicine care, the use of a crossdisciplinary approach in obstetrics and neonatology, improving telehealth safety especially regarding readmissions, and improving clinician communication in pediatrics. This work obviously covers a wide array of clinical areas, many of which have been relatively understudied, such as pediatrics, mental health care, and telehealth, for example, and it represents an important set of contributions that will help inform how to improve care broadly. Common threads include bringing multiple disciplines together to improve safety, especially by leveraging a systems lens.

Overall, though, recent data suggest that care remains much less safe than it should be, and in one large inpatient study, about one in four patients suffered an injury related to the care delivered.⁶ This figure was nearly identical to that of a large study conducted by the HHS Office of Inspector General in a hospitalized Medicare population.⁷ In another study addressing care delivered in the outpatient setting, injuries were found in about 1 in 15 per year.⁸ Thus, harm occurs both inside hospitals and outside them, where patients spend most of their time. While some safety areas have improved—such as hospital-acquired infection rates—others have not, such as adverse drug event rates.

At the same time, recent approaches have made it much easier than in the past to track the frequency of harm. These include widespread implementation of electronic health records, so that records are always accessible, legible, and do not get lost. New treatments are continually being introduced and need to be evaluated.

These data make it clear that the work of measuring and improving safety is far from over, and despite all that we have learned, more changes are needed if patients are to receive care that it is as safe as it should be, especially as care changes and evolves, with new therapies and approaches. The PSLLs have made major contributions to our knowledge around improving safety, and safety is better as a result. At the same time, the federal government has been the primary supporter of research on improving the care of all Americans, and while safety has improved in some areas, much more work needs to be done.