Can Automated Alerts in the Electronic Health Record Encourage Referrals for Genetic Counseling and Testing Among Patients at High Risk for Hereditary Cancer Syndromes?

PURPOSE: Up to 10% of cancers may be associated with an inherited mutation that increases cancer risk. National guidelines emphasize referral for genetic counseling and testing for patients whose personal and/or family history increases their risk of having a hereditary cancer syndrome. METHODS: To increase appropriate referrals for cancer genetic counseling and testing, we piloted an automated alert known as a Best Practice Advisory (BPA) in the electronic health record, Epic, to notify oncology providers when a patient had a personal and/or family history that merited referral to cancer genetics. Epic could not gather the complex clinical data needed for the referral decision automatically, necessitating staff completion of a questionnaire. After educating providers, the BPA was implemented with resources to support its use. RESULTS: Initial interaction with the alert was high but rapidly dwindled, resulting in questionnaire completion in 7.2% of more than 32,000 encounters and 14.9% of patients over 9 months. However, cancer genetics referrals increased 95.9% during the pilot (P < .0001), with 18.5% placed through the BPA and the rest from a non-BPA mechanism. Semistructured interviews with key stakeholders revealed not only general acceptance of the BPA concept but also barriers to completion, such as pressure to room patients quickly in the face of competing BPAs and lack of buy-in from some providers. CONCLUSION: These results suggest that provider engagement and BPA fatigue are significant obstacles to acceptance of a new automated alert. Despite interest in a tool for cancer genetics, the demand on clinical time for this complex BPA was poorly tolerated.

Complexities of cocaine users presenting to the emergency department with chest pain: interactions between depression symptoms, alcohol use, and race.

Chest pain is the most frequent cocaine-related medical event for which patients seek treatment in inner-city emergency departments (EDs). Given that depression increases risk for poor substance use and cardiac outcomes, knowledge of correlates of depressive symptoms among these out-of-treatment cocaine users is crucial to developing interventions for these inner-city populations that frequently present to the ED. A total of 219 individuals presenting to an inner-city ED with chest pain and recent cocaine use were divided into 2 groups based on scoring positive (42%) for moderate-to-severe depression on the Patient Health Questionnaire (PHQ-9). The depression symptoms group reported a significantly greater number of standard drinks per drinking occasion (7.1 versus 4.6) and a greater number of heavy drinking days (9 versus 5). A significant 3-way interaction effect found males and non-white females reporting a greater number of heavy drinking days were more likely to be in the depression symptoms group, whereas white females with a greater number of heavy drinking days were more likely to be in the nondepression symptoms group. Depression is clearly a problem among not-in-treatment cocaine users presenting to an inner-city ED; heavy drinking in concert with cocaine use increases the risk for depression, with important interactions by race and gender.