Reducing Practice Variation in Pediatrics

June 23, 2015

With variations in practice existing even between practitioners at the same institution, research can provide the keys to standardizing care and improving outcomes.

Practice variation occurs when care providers diagnose or treat patients differently, even in the setting of similar disease presentation and risk factors. Variations in care have been noted at the geographic level (countries, states, counties), the hospital level (practices differ by institution) and the provider level (among care providers at the same institution/practice). My research has documented marked variations between major U.S. children’s hospitals in medications used to treat bronchopulmonary dysplasia (BPD) and patent ductus arteriosus (PDA).

Although a small degree of variation is to be expected given differences in patient populations or local risk factors, unwarranted practice variation has the potential to cause patient harm. A study by Rysavy et al. recently showed that variation between hospitals in active treatment of extremely preterm infants appeared to explain a large portion of differences in survival. Locally, changes in testing and treatment strategy between cross-covering providers have the potential to cause families stress and emotional harm.

A Lack of Evidence-Based Best Practices

In many instances, a lack of solid evidence for the best way to diagnose or treat a disease is the root cause of variation. Given limited evidence, a provider must often rely on “the way they did it where I was trained” or “what has always worked for me” as a basis for decision-making. In pediatrics, many commonly used treatments were never studied in children and we often are forced to extrapolate findings from adult studies when treating children. Clearly, better evidence is needed to ensure we provide treatments that help, avoid treatments that harm, and save our patients’ money by avoiding unnecessary treatment.

Comparative effectiveness research will play a prominent role in reducing unwanted practice variation. This branch of research aims to determine the real-world impact of treatment on patient outcomes. Such research provides actionable intelligence that will influence physician decision-making and inform quality improvement (QI) initiatives. Fortunately, prioritization and funding for comparative effectiveness research continues to increase. In addition to National Institutes of Health (NIH) funding, Congress recently created the Patient-Centered Outcomes Research Institute (PCORI) to prioritize research that will provide evidence-based information to answer questions clinicians face in daily practice.

Establishing and Following Practice Guidelines

Despite the promise of new research, we clinicians currently work in a setting where many commonly prescribed tests and treatments have not been fully studied. Adoption of practice guidelines, at both the national and local level, is a way to reduce practice variation by making practice more uniform. Guidelines should be based on the best available, current evidence and the strength of evidence within the guidelines should be stated. The “Grades of Recommendation, Assessment, Development and Evaluation” (GRADE) guidelines provide a helpful template. Grading the evidence within guidelines allows clinician groups to identify those areas of weaker evidence to be amended as new evidence emerges. As a teaching institution, stating the quality of evidence within guidelines teaches trainees “what is known” versus “our local group consensus in the face of poor evidence,” to better inform their understanding of best practices as they begin their careers as attending physicians.

Although it may be impossible for large practice groups to come to a unanimous consensus, an open forum for discussion allows all clinicians to jointly evaluate the evidence. As stakeholders in the guidelines, they are more likely to adhere to them, according to Timmermans and Mauk. Following guideline implementation, it is also important to track both compliance and important patient outcomes. Shewhart Control Charts, as commonly used in quality improvement initiatives, are an ideal tracking tool. Finally, guidelines should be continuously updated as new evidence emerges.

In conclusion, unwarranted practice variation has the potential to cause both patient and family harm. Ultimately, a greater emphasis on patient-centered comparative effectiveness research and evidence-based consensus guidelines provides us with an opportunity to reduce harmful variation and improve patient outcomes.


Join the conversation. How have you seen practice variation affect patients and their families? In which areas is this research needed most?




  1. Rysavy MA, Lei L, Bell EF, Das A, Hintz SR, Stoll BJ, Vohr BR, Carlo WA, Shankaran S, Walsh MC, Tyson JE, Cotton CM, Smith PB, Murray JC, Colaizy TT, Brumbaugh JE, Higgins RD for the Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network.  Between-hospital variation in treatment and outcomes in extremely preterm infants. New England Journal of Medicine. 2015;372:1801-1811.
  2. Grades of Recommendation, Assessment, Development and Evaluation. (accessed 1 June 2016).
  3. Timmermans S, Mauck A. The promises and pitfalls of evidence-based medicine. Health Affairs. 2005 Jan-Feb;24(1):18-28.


Jonathan Slaughter, MD, MPH, is an Assistant Professor of Pediatrics and principal investigator in the Center for Perinatal Research at Nationwide Children’s Hospital, and a neonatologist within the Division of Neonatology.  He conducts patient-centered neonatal research and has completed advanced training in healthcare quality improvement.  Dr. Slaughter’s research has demonstrated marked variation between hospitals in commonly used but understudied neonatal medications. His ultimate research goal is to both improve neonatal care and reduce practice variation via research that will determine whether these medications are helpful in improving important patient outcomes.