Everyone talks about clinical decision support—but what about ‘medication decision support’? Clinical decision support (CDS) gets a lot of attention, but what about “medication decision support?” Increasingly, health system leaders are putting this issue front and center, especially as it relates to medication alerts that are integrated into EHR and clinician workflows. It’s important that medication alerts improve safety and quality while also making clinicians’ jobs easier. If not properly executed, these alerts can become a burden rather than a decision support tool. In an interview, Anna Dover, PharmD, and Bob Katter can speak to best practices around medical decision support, including medication alerts: Fit the medication decision support into the user’s workflow without interrupting them—and make the alerts actionable. Move alerts to the time when they are relevant and to the best person capable of understanding and making a decision about the risk/medication. Use decision support that incorporates patient characteristics. We are going to see that alerts and decision support that rely only on medication lists and nothing more specific from the patient will start to fade away. That traditional approach, although very sensitive, is not very specific. In other words, the alert was good at telling you about a potential risk, but not good about telling you whether this specific patient was at higher or lower risk. We know have the technical capability to leverage lab results, patient age, gender, diagnosis, and other factors to fire an alert when the risk is high—or automatically suppress an alert when the risk is low. Leverage analytics for better medication decision support. By looking at the data being generated through prescribing, we can make predictions about how a clinician or set of clinicians will respond to alert-tuning or modification. It’s possible to simulate and make educated guesses about whether an alert will have the intended effect and improve care, or whether it will cause additional burden to the prescriber. Medication decision support can also go a step further: This data can be leveraged to make specific predictions about a patient, such as quantifying their risk of opioid addiction—or even better, their risk of facing a whole host of adverse drug events. That determination may be based upon their genetics, co-morbidities, and current physical state. The future of medication decision support will focus on a combination of curated, evidence-based rules/knowledge and AI-driven pattern recognition. When we have better alerts, we can use analytics to prove that they work better; they can be installed side-by-side with traditional alerts to validate that an advanced system is safe and effective—ultimately enabling the removal of traditional alerts. Monitor the impact of deployments, get feedback, and respond, in order to optimize. It’s important to closely manage and maintain your knowledge system to reflect evidence and pivot as needed. First Databank (FDB) is the largest drug database in the United States. Bob and Anna are in discussions with health systems, hospitals, and community and retail pharmacists on a daily basis and have in depth knowledge on pain points and opportunities related to the pharmacy industry across the healthcare landscape. Anna from a Pharm D’s perspective and Bob from a business leadership perspective. Guests: Anna Dover, PharmD, Director, Product Management, (FDB) LinkedIn profile: https://www.linkedin.com/in/anna-dover-02137117/ Bob Katter, President of First Databank (FDB) Bio on FDB website: https://www.fdbhealth.com/company/leadership/bob-katter LinkedIn profile: https://www.linkedin.com/in/bobkatter/ This episode was sponsored by First Databank See omnystudio.com/listener for privacy information. Learn more about your ad choices. Visit megaphone.fm/adchoices