10 Lessons From Health Care on Quality Improvement

A multidisciplinary team working on improving the quality of obstetric care in Brazil reviews its data and determines what to test next. (Photo by Pedro Delgado)

For more than a century, the American health care sector has wrestled with the question of how best to improve the quality of the care it provides. Countless reports, commissions, government regulations, and accrediting bodies have sought to introduce basic standards of practice, and improve outcomes for patients across the United States. In the last 30 years, in particular, providers, payers, purchasers, and patients themselves have joined in a global "quality-improvement movement," broadly applying a variety of modern management methods (including Lean, Six Sigma, and Continuous Quality Improvement) in pursuit of progress. While results have been decidedly mixed, the field has made some advances while learning a great deal about the best use of these approaches.

Given the broader social sector’s recent surge of interest in performance management and quality-improvement methods, we reflected on the work we’ve done to improve the effectiveness of hospitals, clinics, and public health services. While health care differs from other fields in several important respects, we believe some of the lessons we’ve learned are broadly relevant. Though this list is hardly comprehensive and undoubtedly reflects the bias inherent in our own experience, we hope it will be a helpful provocation to those beginning to take these techniques more seriously.

1. Avoid Systems Myopia 

W. Edwards Deming, a pioneer in the science of quality improvement, described the field as resting on four pillars: the ability to see and appreciate the system in which we operate; the ability to introduce new ideas into that system to continuously generate new knowledge; the ability to understand when meaningful change is happening in the system (the statistics of variation); and the ability to motivate participants in the system to embrace change. With respect to the first pillar—seeing the system—health care has long been guilty of myopia, focusing on improvement in hospitals and specialty areas, while failing to understand the larger societal factors responsible for unequal outcomes and skyrocketing costs (including the undue power of health insurers and hospitals, underinvestment in communities and public health, lack of access to healthy food and transportation, and deep racial bias). Policy changes driven by the Affordable Care Act, which encouraged care providers to improve health in their communities, have begun to address this, though only in pockets. Our more-limited experience in other parts of the social sector suggests that funders and nonprofits frequently neglect similar "macro" factors, missing the pervasive injustices and underlying structures that give rise to their daily challenges.

2. Start With the Customer

The American health care system prioritizes the financial goals and operational needs of care purchasers, providers, and payers, which results in fragmented systems that are both disrespectful of patients’ time and damaging to their health. By asking patients, families, and community members a simple question—“What matters to you?”—a group of care systems from around the globe has started to turn that experience around. The question prompts health care leaders to deeply understand what patients, families, and community members value and dislike, informing their objectives and deepening their focus on customer needs. Likewise, social sector leaders need to regularly ask themselves whether their processes serve beneficiaries or burden them, relentlessly eliminating activities that don’t add value for the end user and, as much as possible, orienting the system away from the self-serving, idiosyncratic concerns of those who hold the purse strings.

3. Track Rate of Learning

While many health care systems invest in the trappings of quality improvement (memorable slogans, technical jargon, and the like), the most successful avidly study the rate at which they trial new innovations and learn what works. This approach views hospital units, office practices, and clinicians as generators of knowledge—as trailblazers asking new questions and testing new solutions—as opposed to waiting for guidance from experts. This mindset is rare but powerful, and it significantly increases the pace of improvement. At Cincinnati Children’s Hospital, for example, doctors and nurses test scores of new ideas on a small scale every day, then review progress together and generate new concepts for experimentation. If organizations in the social sector start examining how often they try new ideas and embrace this pioneering, fast-learning self-image, they will progress much faster. 

4. Emphasize Adaptation, Not Fidelity 

Much of the work of improvement in health care involves taking ideas and innovations already established in the evidence base, and adapting them to different care settings. Seeking strict adherence to implementation protocols is thus often counterproductive. In most cases, after communicating which components of an intervention are clinically sacrosanct, leaders must trust professionals to make it work within their own context, culture, and operating constraints. Words like "fidelity” and “accountability" are canaries in the coal mine, suggesting a philosophy of rigidity and fear that is unlikely to generate the creativity needed to make practices work locally. By contrast, systems like the Scottish National Health Service, broadly recognized for its success at improving patient safety, encourage and celebrate context-appropriate adjustments, sharing interesting modifications across facilities. The social sector would benefit from encouraging the same.

5. Design Measurement Systems for the Front Line First

In health care, measurement systems too often serve the needs of regulators, administrators, academics, and other third parties who use data for research and inspection. Care providers do the arduous work of entering data into forms and spreadsheets, never to see it again unless leaders use it to rank or admonish them. By contrast, organizations that make regular progress realize that no one is better positioned to improve the health system than the people in it. The first design principle of any effective measurement system is this: Put timely, easy-to-interpret data in the hands of those who can make day-to-day change, including doctors, nurses, patients, and families. Managers can aggregate descriptive, system-level data later as needed. Similarly, foundations and other investors in the social sector must ask themselves if their evaluation systems and data requirements empower front-line learning or distract from it. And while initiatives that pursue user feedback are encouraging, those that go a step further and allow customers and community members to access data and join in the improvement process are even better. This approach, sometimes referred to as “co-production,” is gaining momentum in health care; patients take much more active roles in their own treatment and the treatment of their peers, optimizing care delivery in the process.

6. Embrace Nuance in Evaluation 

The social sector's fascination with isolating an intervention’s impact through approaches like randomization and randomized control trials was preceded by a similar infatuation in health care. While these methods are, at times, appropriate and powerful in studying new treatments and technologies, health care has dispensed with the idea that any single evaluation approach is a "gold standard." Instead, thoughtful researchers have come to understand that there are many approaches to learning and evaluation, and each is appropriate at different times. For example, when spreading an innovation that already has sound evidence, it is not necessary to reassess its impact; instead we should be studying local implementations and adaptations that will facilitate broader adoption. Funders in the social sector need to school themselves on when to apply different learning approaches, and relieve nonprofits from the burdens of time-consuming and costly evaluations that don’t support urgent change.

7. Live in the Field

A good measurement system doesn't provide conclusions; it provides clues. And no amount of speculation in the central office can replace investigating those clues in person, in the field. The most successful quality-improvement efforts we have observed in health care use data—interesting patterns and troubling trends—as a prompt to go out to visit communities, hospital units, and other care settings. Once there, they can begin to understand what is really happening in all of its texture, harvesting the wisdom of those giving and receiving care and then actively removing impediments to their progress. The most successful systems, like Jonkoping County in Sweden, convene care providers to study failures and successes up close, and even fund them to visit breakthrough work in other countries. Multidisciplinary rounds—where everyone who has a role in giving care, including non-clinicians, visit patients together to coordinate efforts and assess effectiveness—offer another practical form of active reflection at the point of service delivery that social sector organizations could emulate with their clients.

8. Understand the Psychology of Change 

Deming's fourth dimension is about motivating groups to improve and is sometimes referred to as "the psychology of change." Because this can be highly personal, requiring deep investigation into feelings and values, health care has often focused on more-technical aspects of organizational change like process mapping and data analysis. But technical work that fails to connect with the reasons people are called to their professions soon becomes drudgery. By contrast, a few very successful health care initiatives, such as McLeod Regional Health System’s surgical checklist implementation efforts and the Saúde em nossas Mãos effort to reduce infections acquired in public hospitals across Brazil, have actively surfaced participants’ values, thereby tapping huge stores of passion and creativity that they apply to improvement.

9. Approach Payment Incentives With Caution

Hardly a week passes in health care without an insurance company or regulatory agency introducing a new payment incentive program that aims to influence provider behavior. Doctors and nurses are endlessly ranked and rated, and compensated accordingly, but there is very little evidence to suggest that this leads to better outcomes for patients. Possible reasons for the failure of pay-for-performance programs include the fact that they make faulty comparisons between dissimilar organizations, induce groups to misreport their performance, and belittle and discourage care providers, resting on the problematic assumption that financial incentives are what drive their behavior. The social sector should approach payment incentives with caution, and invest more time in cultivating the intrinsic motivations we described in the previous lesson.

10. Address Inequity Proactively

Despite health care's investments in improving quality, major racial disparities in health—and discrimination in the provision of health care services—have persisted. All improvement efforts must proactively address inequity or risk deepening it, exacerbating gaps on the very dimensions they seek to improve. Organizations must pursue interventions that target injustice, and any quality project must solve for the needs of oppressed groups first, studying stratified data on progress in addition to overall changes in outcomes. Colleagues at the Institute for Healthcare Improvement have produced helpful materials on this, and other parts of the social sector are also leading the way. A committee convened by the education team at the Bill and Melinda Gates Foundation, for example, is systematically analyzing how equity can inform every element of quality improvement, including the composition of improvement teams and the framing of aims and norms of operation. (This group, to which co-author Joe McCannon contributes, intends to release its findings later in the fall, and some of the resources developed by its members, such as this tool designed by Shift-Results with the National Equity Project for the Building Equitable Learning Environments (BELE) Network, are available already.)

Other industries with deep experience in improvement science and those applying quality-improvement methods elsewhere in the social sector (such as Carnegie Foundation for the Advancement of Teaching) should add their insights to our own, as their knowledge about the specifics of certain problems is critically important. That said, we believe the lessons we’ve outlined here are durable, and that the way organizations carry out quality-improvement work is highly predictive of success. Understanding the sources of injustice in the systems we want to improve, getting curious about customer needs, supporting front-line testing and learning, using evaluation thoughtfully, and limiting fear and intimidation will all make our attempts at improvement considerably more effective.

Read more stories by Pedro Delgado, Maureen Bisognano & Joe McCannon.