A (Not So) Perfectly Designed System: The Paradox of Medical Stewardship and Quality Measurement

The Lack of Overuse Quality Measures

Present-day measures tend to favor assessments of underuse rather than overuse. As an example, >75% of the currently used quality process measures in pediatrics are used to target the underuse of health care services, such as documenting hydration status in gastroenteritis, obtaining a blood culture in patients with community-acquired pneumonia, and performing hearing testing for patients with otitis media with effusion.⁴ The implementation of overuse measures, such as the administration of albuterol and/or steroids in bronchiolitis and appropriate testing for children with pharyngitis, is more challenging than that of underuse measures. Improving the adoption of a recommended test or intervention may be easier to conceptualize, accept, and measure. Deimplementation, or forgoing a test or intervention that evidence suggests should not happen, is more difficult to achieve.⁵ When a patient receives a low-value test, even 1 that ultimately results in downstream harm, we do not typically view it as low-value care. In addition, many tests and interventions seen as the standard of care and subsequently proven to be of low value may be particularly challenging to deimplement.⁶ For example, despite a randomized controlled trial from 2007⁷ revealing no benefit of steroids for moderate-to-severe bronchiolitis, many young children continue to receive this therapy.⁸

Patient Satisfaction Does Not Necessarily Equal Quality

In addition to quality measures, another important variable in the quality measurement equation is the patient satisfaction survey. Surveys, such as the Press Ganey Patient Satisfaction Survey and the Hospital Consumer Assessment of Healthcare Providers and Systems Survey (which is publicly reported), reflect patient perceptions about care and may not be used to measure the true adequacy of clinician practice. Unfortunately, higher patient satisfaction may be tied to practices that run counter to the idea of medical stewardship. In fact, some evidence reveals that higher satisfaction scores are associated with greater inpatient use and higher health care expenditures.⁹ In many cases, these patient satisfaction surveys are tied to reimbursement both at the individual clinician and hospital levels. Often, a physician’s job retention or ability to be promoted and some group contracts are contingent on meeting certain levels of patient satisfaction. These patient satisfaction benchmarks may lead to the unintended consequences of overused treatments and services.¹⁰ Little evidence exists regarding the association between high satisfaction scores and improved outcomes. In fact, the opposite may be true, with higher scores being associated with higher patient mortality in at least 1 study in adults.⁹

Why are patients seemingly satisfied with more care being delivered? Many patients are more reassured when they endure some form of testing for a specific complaint or perceive tests as a meaningful measure of their overall health.¹¹ The presence of health-related information, including online networks of patients and families with similar diseases and symptoms, and direct-to-consumer advertising can overwhelm patients and families and result in expectations that are not necessarily rooted in net benefit. Moreover, when we as clinicians fail to practice evidence-based medicine, as reflected in examples such as unnecessary imaging in minor head trauma and bronchiolitis and antibiotics for viral illnesses, the cycle of unnecessary testing and treatment is perpetuated.

Redesigning the System

Although changing practice and the system is challenging, it is important that quality measurement strategies are used to accurately and adequately capture high-quality health care delivery and do not result in unintended consequences, such as an overuse of resources. Specifically, we believe the following are important key goals for improving the quality measurement system:

1. Measuring patient net benefit: We must move from the current culture that values doing more to a culture that values a net benefit for patients. Campaigns such as Choosing Wisely have assisted in highlighting overuse. In its list of recommendations, there are now >40 in which children are targeted (www.choosingwisely.org). Future national as well as local quality measures should incorporate and prioritize these recommendations. There must be a more balanced assessment of quality such that medical overuse (as well as underuse) is measured and weighted evenly.

2. Overuse as a patient safety measure: Safety issues within hospitals and health care systems are often relegated to errors of omission (best practices we failed to follow). Errors of commission (things we did that are of unproven benefit) are given less weight. Today, hospitals foster a culture in which reporting safety issues is encouraged and taken seriously. Zapata et al¹² suggest including cases of overuse in patient safety and other hospital review committees. Practitioners should be given timely feedback on their performance relative to peers locally and nationally and to evidence-based guidelines. Given that both errors of commission and those of omission can lead to harm, leaders within health care systems should recognize overuse as a safety issue and encourage reporting and monitoring with as much weight as is currently given to other unsafe practices.

3. Patient satisfaction and financial incentive realignment: It is important that evidence-based practices are not abandoned for favorable patient satisfaction scores. Patient satisfaction data can and should be used to improve the quality of care delivery and not as a punitive tool. However, surveys must ask the right questions, for example, “Did my child’s health care provider spend enough time explaining the plan of care, providing education regarding my child’s diagnosis, and answering all of my questions?” and ”Did my child’s health care provider discuss the benefits and risks of any laboratories, imaging studies, and treatments prescribed?” With this, there must be a realignment of financial incentives such that not performing a test and other previously unmeasured behaviors, such as patient education, are rewarded. The Centers for Medicare and Medicaid Services has made efforts toward this goal with its value-based programs, which reward health care providers with incentive payments for quality rather than quantity of care. More time is needed to study the effects of such programs on reducing overuse as well as their influence on patient satisfaction scores. Finally, just as there is targeted patient advertising of tests and medications, so, too, should there be targeted patient education about the consequences of overuse. Clinicians and practices should leverage the reliance and impact of social media, including Facebook and Twitter, to provide accurate information to patients and families. Armed with this knowledge, patients, too, may begin to judge health care on the basis of quality instead of quantity.