“Dr. Chang? Oh my, it’s Dr. Chang! And his little son!” I called them “mall moments.” I would be at the local shopping mall with my father, picking up new clothes for the upcoming school year, when suddenly an elderly woman would approach. My father, despite his inability to remember my own birthday, would warmly grasp the woman’s hands, gaze into her eyes, ask about her family, then reminisce about her late husband and his last days in the hospital. After a few minutes, she would say something like, “Well, your father is the best doctor in Bakersfield, and you’ll be lucky to grow up to be just like him.”
And this would be fine, except the same scene would replay at the supermarket, the dry cleaners, and the local Chinese restaurant (the only place my father would eat out until he discovered the exotic pleasures of sushi). I wondered how my father ever got any errands done, with all his patients chatting with him along the way. Looking back on these “moments,” it is clear to me that this was my father’s measure of quality—his patients loved him. Other doctors loved him. The nurses—well, maybe not so much. He was a doctor’s doctor.
Quality measures? After working in his office, I only knew of two: The waiting room must be empty before the doors are closed and locked, and no patient ever gets turned away, for any reason. By seven o’clock in the evening, these measures got pretty old. But simple credos made him one of the most beloved physicians in Kern County, Calif.
Quality, in whatever form it takes, has a cost, however. My father divorced twice. My own “quality” time with him was spent making weekend rounds at the seemingly innumerable nursing homes around Bakersfield, Calif., although this was great olfactory training for my future career as a hospitalist. Many a parent’s day was spent with only my mother present, and I would be lying if I said I didn’t envy the other children with both parents doting over their science projects.
As we in pediatric hospital medicine (PHM) embark on a journey to define and promote quality in our care of children, we are well aware that adhering to our defined standards of quality will have a cost. What has been discussed less, but is perhaps even more elementary, is the cost of simply endeavoring to define and measure quality itself. This has not slowed down the onslaught of newly defined quality measures in PHM. Quality measures from the adult HM world, such as readmission rates, adherence to national guidelines, and communication with primary care providers, have been extracted and repurposed.
Attempts to extrapolate these measures to PHM have been less than successful. Alverson and O’Callaghan recently made a compelling case debunking readmission rates as a valid quality measure in PHM.1 Compliance with Children’s Asthma Care (CAC) measures was not found to decrease asthma-related readmissions or subsequent ED visits in a 2011 study, although a study published in 2012 showed an association between compliance with asthma action plans at discharge and lower readmission rates.2,3 Documentation of primary care follow-up for patients discharged from a free-standing children’s hospital actually increased the readmission rate (if that is believed to be a quality measure).4
Yet quality measures continue to be created, espoused, and studied. Payments to accountable care organizations (ACO), hospitals, and individual providers are being tied to performance on quality measures. Medicare is considering quality measures that can be applied to PHM, which might affect future payments to children’s hospitals. Paciorkowski and colleagues recently described the development of 87 performance indicators specific to PHM that could be used to track quality of care on a division level, 79 of which were provider specific.5 A committee of pediatric hospitalists led by Paul Hain, MD, recently proposed a “dashboard” of metrics pertaining to descriptive, quality, productivity, and other data that could be used to compare PHM groups across the country.6 Many hospitalist groups already have instituted financial incentives tied to provider or group-specific quality measures.7 Pay-for-performance has arrived in adult HM and is now pulling out of the station: next stop, PHM.
The Rest of the Cost Story
Like any labor-intensive process in medicine, defining, measuring, and improving quality has a cost. A 2007 survey of four urban teaching hospitals found that core QI activities required 1%-2% of the total operating revenue.8 The QI activity costs fall into the category of the “cost of good quality,” as defined by Philip Crosby in his book, Quality is Free (see Figure 1).9 Although hospital operations with better process “sigma” will have lower prevention and appraisal costs, these can never be fully eliminated.
Despite our attempts at controlling costs, most ongoing QI efforts focused on improving clinical quality alone are doomed to fail with regard to providing bottom-line cost reductions.10 QI efforts that focus on decreasing variability in the use of best practices, such as the National Surgical Quality Improvement Program (NSQIP), have brought improvements in both outcomes and reduced costs of complications.11 Not only do these QI efforts lower the “cost of poor quality,” but they may provide less measurable benefits, such as reduced opportunity costs. Whether these efforts can compensate by reducing the cost of poor quality can be speculative. Some HM authorities, such as Duke University Health CMO Thomas Owens, have made the case, especially to hospital administrators, for espousing a more formulaic return on investment (ROI) calculation for HM QI efforts, taking into account reduced opportunity costs.12
But measured costs tell only part of the story. For every new quality measure that is defined, there are also unmeasured costs to measuring and collecting evidence of quality. Being constantly measured and assessed often leads to a perceived loss of autonomy, and this can lead to burnout; more than 40% of respondents from local hospitalist groups in the most recent SHM Career Satisfaction Survey indicated that optimal autonomy was among the four most important factors for job satisfaction.13 The same survey found that hospitalists were least satisfied with organizational climate, autonomy, and availability of personal time.14
As many a hospitalist can relate, although involvement in QI processes is considered a cornerstone of hospitalist practice, increased time spent in a given QI activity rarely translates to increased compensation. Fourteen percent of hospitalists in a recent SHM Focused Survey reported not even having dedicated time for or being compensated for QI.
Which is not to say, of course, that defining and measuring quality is not a worthy pursuit. On the contrary, QI is a pillar of hospital medicine practice. A recent survey showed that 84% of pediatric hospitalists participated in QI initiatives, and 72% considered the variety of pursuits inherent in a PHM career as a factor influencing career choice.15 But just as we are now focused on choosing wisely in diagnosing and treating our patients, we should also be choosing wisely in diagnosing and treating our systems. What is true for our patients is true for our system of care—simply ordering the test can lead to a cascade of interventions that can be not only costly but also potentially dangerous for the patient.
Physician-defined quality measures in adult HM have now been adopted as yardsticks with which to measure all hospitals—and with which to punish those who do not measure up. In 1984, Dr. Earl Steinberg, then a professor of medicine at Johns Hopkins, published a seminal article in the New England Journal of Medicine describing potential cost savings to the Medicare program from reductions in hospital readmissions.16 This was the match that lit the fuse to what is now the Affordable Care Act Hospital Readmissions Reduction Program. Yet, this quality measure might not even be a quality measure of…quality. A 2013 JAMA study showed that readmission rates for acute myocardial infarction and pneumonia were not correlated with mortality, the time-tested gold standard for quality in medicine.17 That has not stopped Medicare from levying $227 million in fines on 2,225 hospitals across the country beginning Oct. 1, 2013 for excess readmissions in Year 2 of the Hospital Readmissions Reduction Program.18 It seems that we have built it, and they have come, and now they won’t leave.
What is the lesson for PHM? Assessing and improving quality of care remains a necessary cornerstone of PHM, but choosing meaningful quality measures is difficult and can have long-term consequences. The choices we make with regard to the direction of QI will, however, define the future of pediatric healthcare for decades to come. As such, we cannot waste both financial and human resources on defining and assessing quality measures that may sound superficially important but, in the end, are not reflective of the real quality of care provided to our patients.
My father, in his adherence to his own ideal of quality medical care, reaped the unintended consequences of his pursuit of quality medical care. Sometimes, though just sometimes, there are unintended consequences to the unintended consequences. I learned, and was perhaps inspired, just by watching him interact with patients and their families. Somehow I don’t think my own children will learn much by watching me interact with my computer.
Dr. Chang is pediatric editor of The Hospitalist. He is associate clinical professor of medicine and pediatrics at the University of California at San Diego (UCSD) School of Medicine, and a hospitalist at both UCSD Medical Center and Rady Children’s Hospital. Send comments and questions to firstname.lastname@example.org.