Practice Pattern Variation in Test Ordering for Low-Acuity Pediatric Emergency Department Patients

Healthcare systems increasingly seek ways to reduce unnecessary diagnostic testing to improve efficiency and lower costs. Prior research has shown that test ordering can vary based on provider type and training. However, no prior work had resolved how much variation exists among licensed independent providers (LIPs) in pediatric emergency departments (PEDs) when treating low-acuity patients. This gap motivated a closer look at how provider background influences diagnostic testing decisions. Existing knowledge includes general trends in diagnostic overuse, but this paper's contribution is to quantify variation within specific provider groups. The study builds on prior work by focusing on low-acuity cases, where unnecessary testing is more likely. It also addresses a gap in understanding how training backgrounds affect clinical decisions. The findings could help inform strategies to standardize care across different provider types.

The study aimed to assess how much diagnostic testing varies among licensed independent providers (LIPs) with different training backgrounds in a pediatric emergency department (PED). The focus was on low-acuity patients, where unnecessary testing is a known concern. The researchers sought to determine if provider type—such as physician versus nurse practitioner—impacted test ordering. They also examined if training, such as pediatric emergency medicine (PEM) fellowship, influenced diagnostic decisions. The motivation stemmed from rising healthcare costs and the need to optimize diagnostic efficiency. The study aimed to identify patterns that could inform future interventions. It sought to measure variability in test ordering across provider groups. The goal was to provide data that could guide efforts to standardize care practices.

The researchers conducted a retrospective analysis of electronic health records from two urban, academic pediatric emergency departments. They included patients 21 years or younger who were triaged as low-acuity between 2012 and 2013. The study focused on encounters involving licensed independent providers (LIPs), including physicians and nurse practitioners. They examined the frequency of specific tests, such as complete blood counts and chest radiographs. The team compared test ordering across provider groups: physicians versus nurse practitioners, and physicians with and without pediatric emergency medicine (PEM) training. They also considered experience level, comparing those with at least five years post-residency to newer providers. Bivariable and multivariable logistic regression analyses were used to assess associations. Sensitivity analyses were performed to confirm findings in a subset of encounters.

The study found significant variation in diagnostic test ordering among providers. Testing rates per patient encounter ranged from 0% to 40% across individual providers. Bivariable analysis showed higher odds of testing when the provider was a physician (OR = 1.2) or had PEM training (OR = 1.3). Multivariable analysis confirmed that PEM-trained providers had higher adjusted odds of ordering tests (AdjOR = 1.2). In a subset of encounters at the PED-based urgent care, PEM physicians had even higher odds (AdjOR = 1.5). Nurse practitioners also had higher odds of ordering tests compared to physicians (AdjOR = 1.2). These findings suggest provider background influences diagnostic decisions. The variability was most pronounced between PEM and non-PEM providers. The results highlight the need for further research into standardizing practice.

The authors concluded that licensed independent providers (LIPs) show substantial variation in diagnostic test ordering for low-acuity patients. They found that provider type and training background significantly influence testing decisions. PEM-trained physicians were more likely to order tests compared to non-PEM physicians. Nurse practitioners also had higher odds of ordering tests than physicians. The study suggests that training and experience impact diagnostic practices. These findings may inform future efforts to standardize care. The authors propose that interventions could help reduce unnecessary testing. They suggest further research to explore ways to align practice across provider groups.