medwireNews: US researchers have found that embedding clinical prediction rules for pneumonia and pharyngitis into electronic health records results in significantly reduced antibiotic prescriptions and streptococcal test ordering in primary care.
Furthermore, the system was well received by physicians, achieving a higher adoption rate in practice than observed in previous studies.
“We believe our results indicate that providers may have perceived the tool as being helpful with the clinical diagnosis of pharyngitis and pneumonia, enhancing clinical work flow and improving patient care,” say the authors, Thomas McGinn (Hofstra North Shore-LIJ School of Medicine, Manhasset, New York) and colleagues.
The study, reported in JAMA Internal Medicine, involved two large primary care practices including a mix of 168 providers (including attending physicians, residents, fellows, and nurse practitioners) who were randomly assigned to provide usual care (control group) or the intervention.
The intervention group were given training in the use of clinical prediction rules and provided with evidence supporting the Walsh rule for streptococcal pharyngitis and the Heckerling rule for pneumonia.
Between November 2010 and October 2011, these clinical prediction rules were integrated into electronic health record access systems; triggered by the input of keywords such as “sore throat” or “possible pneumonia,” users were given the opportunity to create a risk score for patients with upper respiratory tract symptoms, and receive subsequent management recommendations.
Over the course of 40,003 patient visits, including 987 patients with pharyngitis or pneumonia, providers in the intervention group were 26% less likely to order antibiotics than were control providers. They were also 25% less likely to order rapid streptococcal tests in patients with pharyngitis.
However, there was no change in the rate of chest radiographs ordered for pneumonia or the number of throat swabs taken for pharyngitis.
Importantly, the authors noted high uptake rates of the tool, with 57.5% of providers in the intervention group using the system for relevant patients. This contrasts with typical rates of 10% to 20% reported in earlier studies.
McGinn and team attribute this success to their development process for the tool, which included physician input, usability testing, and user training.
“High adoption rates of [clinical decision support] interventions that lead to changes in outcomes need to be further researched,” they write.
“Our study showed one such approach to successful [clinical decision support] intervention in a commercial [electronic health record].”
medwireNews (www.medwirenews.com) is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2013