In a recently released issue of Pediatrics (10.1542/peds.2017-3187), Dr. Michael Tchou and colleagues share results of a quality improvement (QI) project whose goal was to reduce the number of electrolyte tests ordered per patient by 25% among hospitalized children on their inpatient services. There are several intriguing and unique features of this QI project. First and foremost, it was successful! Second, the project engaged trainees at the front line, and used multiple simultaneous strategies (see Table 2 for a summary) to support improvement and change. Finally, the authors aimed their QI at all electrolyte tests, rather than specifically focusing on those that could be considered “unnecessary,” which had the advantages of both simplicity (for the project) and generalizability (for the results).
Aiming the QI at all electrolyte testing, rather than at testing that could be considered unnecessary for a specific patient, diagnosis, or time, may be less satisfying intellectually and medically than a more focused project. But the outstanding strength of this approach was the ability to both implement strategies and gather results broadly and across multiple patients. Some fine tuning may be interesting and indeed the author team addresses this in the Discussion and under “Next Steps.” One could postulate that a QI project targeting electrolyte testing for inpatients who are NPO and receiving only intravenous fluids would need to begin with some definition of an appropriate minimum frequency of tests needed for patient safety, or that projects targeting patients with bronchiolitis, as compared to dehydration, would need different end points for frequency of electrolyte testing. Thus this QI project is not specifically about “choosing wisely” (see http://www.choosingwisely.org/ ), whose stated aims are to promote conversations between clinicians and patients that help patients choose tests that are supported by evidence, non-duplicative, free of harm and truly necessary. This QI project is more about reducing the burden for patients and clinicians, and about reducing costs. I applaud the authorship team on their SMART (Specific, Measurable, Attainable, Relevant and Timely) aim and their very “smart” and broad approach to a hospital-wide problem.
A final interesting point in thinking about generalizability is the authorship team’s important acknowledgement of their own time and effort, and of their own hospital’s comfort with QI. A meaningful amount of time was required to plan and execute this successful initiative. The cost of the authors’ salaried time is not free, and while provider time at any institution that planned to replicate the work would likely be substantially less due to the ability to use the tools created here, salary cost for the time would not be free. Additionally, the authors’ institution has an honorable history of multiple successful QI initiatives, and this culture of improvement and change is not present at every hospital. However, identification of barriers is the first step to overcoming them, and hopefully there will be a will and a way to overcome such barriers for those who seek to replicate this important work.