OD0358 - Decision Support Tool Reduces Low Probability Cardiology Referrals for Chest Pain from Primary Care: a Quality Improvement Initiative

Purpose: To decrease low probability cardiology referrals in patients ages 7-21 presenting with chest pain to three local pediatric practices from 16% to 5% by 1/31/2020.

Background: When a patient presents with chest pain to the primary care pediatrician, the decision to refer can be challenging. Pediatricians need skills in identifying patients who may have underlying cardiac pathology to allay concerns of those who do not. In previous studies, primary pediatricians identified gaps in their education surrounding the issue of subspecialty referral, specifically citing a lack of knowledge of the medical red-flags that would necessitate a referral. Furthermore, a recent multi-center initiative identified medical red-flag criteria for chest pain associated with cardiac disease with 100% sensitivity in children ages 7-21 years. A pilot study was conducted at one primary care practice from 01/2018 to 12/2018, which reduced low probability referral for children presenting with chest pain.
Methodology: A multidisciplinary team including a pediatric cardiologist, primary care clinicians and a population health nurse designed and implemented an initiative consisting of: conducting chest pain provider educational sessions; adding a validated decision support tool (DST) (including screening red-flag criteria for chest pain referral) to providers’ “Personal Forms” in the electronic medical record (EMR); and implementing site specific reminders to increase compliance. Three pediatric primary care practices, comprising 16 pediatricians and 4 nurse practitioners, were included in this study. We tracked progress via control charts.

Discussion: A total of 301 patients presented with chest pain to their primary care pediatrician from 02/01/2018 to 01/31/2020. The percentage of low probability cardiology referrals in pediatric patients ages 7-21 who presented with chest pain decreased from 16% to 4% after our interventions (Figure 1). Concurrently, the utilization of the DST increased from baseline of 15% to 68% (Figure 2). At a median follow up time of 0.9 years (interquartile range 0.3 - 1.6 years), no patient had a life threatening event or sudden cardiac death.

Conclusion: This quality improvement initiative to reduce low probability referrals for children presenting to primary care practices with chest pain was feasible, effective and safe. Clinicians increased their use of the DST and low-probability cardiology referrals decreased. Evidence-based decision support prompts in the EMR and regular reminders were key elements for success.