IMPACT OF PROACTIVE PHARMACIST COUNSELING GUIDED BY AN AI-BASED DECISION SUPPORT SYSTEM ON HOSPITALIZATIONS AND EMERGENCY DEPARTMENT VISITS
Author(s)
Lior Apter, PhD1, Talia Wagner, Bsc Pharm1, Adva Tsuk On, MD2, Shahar Dekel, BA3, Tal Raveh, BSc, MBA, MHA1, Shlomo Vinker, MD1.
1Leumit Health Services, Tel Aviv, Israel, 2FeelBetter, Herzalia, Israel, 3FeelBetter, Herzelia, Israel.
1Leumit Health Services, Tel Aviv, Israel, 2FeelBetter, Herzalia, Israel, 3FeelBetter, Herzelia, Israel.
OBJECTIVES: Preventable hospitalizations and emergency department (ED) visits driven by medication safety issues present a major economic burden. This study evaluated the impact of an artificial intelligence (AI)-based decision support system (FeelBetter [FB]) combined with proactive pharmacist counseling on healthcare utilization compared to a matched standard care control group.
METHODS: Leumit-Health-Services implemented the FB system to proactively identify patients at high risk for pending hospitalization due to medication anomalies, enabling pharmacists to prioritize high-risk patients. We retrospectively analyzed data from 1,644 patients who received pharmacist counseling between September 2024 and August 2025. Hospitalization and ED events were calculated for the 3 months pre- and post-intervention. A control group (n=1,660) was matched by age, medications, comorbidities, and baseline health services utilization. Cost of bed-days was based on the Israeli Ministry of Health medical services tariff (~915 € per day).
RESULTS: Study population had a median age of 72 years and a median of 18 medications per patient. The intervention group showed a 40.0% decrease in hospitalizations (0.23 to 0.14 events/patient) versus 11.7% in controls (0.23 to 0.209), making the intervention 3.29-fold more effective than standard care. ED visits dropped 52.33% (0.29 to 0.14 visits/patient) versus 16.49% in controls (0.194 to 0.162), a 3.69-fold higher effectiveness. The association scaled with baseline risk, peaking at risk scores ≥95 percentile with a 54.78% reduction in hospitalizations and 64.5% in ED visits. Total inpatient days decreased 24.6% versus 16.9% in the control group, translating to ~900 saved bed-days in the intervention group. At an estimated cost of 915 € per bed-day, the total 3 months healthcare savings for the intervention group reached ~823,500 €.
CONCLUSIONS: Proactive, AI-guided pharmacist intervention was associated with a significantly reduces hospitalizations, ED visits, and inpatient days, optimizing care and generating substantial cost savings. To maximize efficiency, resource allocation should target the highest-risk patient populations.
METHODS: Leumit-Health-Services implemented the FB system to proactively identify patients at high risk for pending hospitalization due to medication anomalies, enabling pharmacists to prioritize high-risk patients. We retrospectively analyzed data from 1,644 patients who received pharmacist counseling between September 2024 and August 2025. Hospitalization and ED events were calculated for the 3 months pre- and post-intervention. A control group (n=1,660) was matched by age, medications, comorbidities, and baseline health services utilization. Cost of bed-days was based on the Israeli Ministry of Health medical services tariff (~915 € per day).
RESULTS: Study population had a median age of 72 years and a median of 18 medications per patient. The intervention group showed a 40.0% decrease in hospitalizations (0.23 to 0.14 events/patient) versus 11.7% in controls (0.23 to 0.209), making the intervention 3.29-fold more effective than standard care. ED visits dropped 52.33% (0.29 to 0.14 visits/patient) versus 16.49% in controls (0.194 to 0.162), a 3.69-fold higher effectiveness. The association scaled with baseline risk, peaking at risk scores ≥95 percentile with a 54.78% reduction in hospitalizations and 64.5% in ED visits. Total inpatient days decreased 24.6% versus 16.9% in the control group, translating to ~900 saved bed-days in the intervention group. At an estimated cost of 915 € per bed-day, the total 3 months healthcare savings for the intervention group reached ~823,500 €.
CONCLUSIONS: Proactive, AI-guided pharmacist intervention was associated with a significantly reduces hospitalizations, ED visits, and inpatient days, optimizing care and generating substantial cost savings. To maximize efficiency, resource allocation should target the highest-risk patient populations.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
CO195
Topic
Clinical Outcomes, Health Service Delivery & Process of Care
Topic Subcategory
Clinician Reported Outcomes