Real World Evidence on Impact of Clinical Decision Support System (CDSS) Alerts on General Practitioners Adherence to the New Gastro-Oesophageal Reflux Disease Treatment Guidelines in German Primary Care Settings
Speaker(s)
Aluko P1, deGreyWarter F1, Plehhova K1, Wray J1, Haering M2, Horvath E1, Kovacic B3, Coyle C1
1Reckitt Health, Hull, UK, 2Reckitt Benckiser Deutschland GmbH, Heidelberg, Germany, 3Reckitt Benckiser LLC, Parsippany, NJ, USA
Presentation Documents
OBJECTIVES: The German Society for Gastroenterology, Digestive and Metabolic Diseases (DGVS) issued a guideline recommending that treatment of mild gastro-oesophageal reflux disease (GORD) with proton pump inhibitors (PPI) is limited to 4-8 weeks, reduced in dosage, or switched to alternative therapies to decrease the prevalence of inappropriate prescriptions. Efficacy of the Clinical Decision Support System (CDSS) in encouraging adherence to guidelines has previously been established. This study aims to assess the impact of CDSS alerts on the implementation of new treatment guidelines at the point of prescription in Germany’s primary care settings.
METHODS: Artificial intelligence and machine learning tools were used for identification and delivery of patient specific messages. Alerts were triggered over 12 months based on ICD codes, previous PPI prescriptions and GI symptoms. Alerts included short messages on the latest guideline, and peer-reviewed publications around appropriate PPI use. This study includes 54% of primary care practices in Germany using CDSS. Analysis was conducted using ICD codes for uncomplicated reflux and diagnosed GORD to determine changes in uncoded diagnosis and prescription numbers (pack sizes and dosage).
RESULTS: Contrary to the new guideline recommendations, analysis revealed that <50% of the patients per physician had a documented diagnosis justifying PPI use, with ratio 0.34-0.47 patients per GP. For diagnosed GORD, an average of 47 more GPs, of 333 alerted, prescribed PPIs without an indicated diagnosis (95%Cl: 19.7-74.3; P<0.0017), while 15 more, of 66 alerted, did so for uncomplicated reflux (95%Cl: 9.7-20.10; P<0.001). In addition, further analysis showed consistently large pack sizes being prescribed for >3 months (86% uncomplicated reflux and 100% for diagnosed GORD).
CONCLUSIONS: Findings from this study suggest that CDSS alerts had no significant impact on the adherence to new GORD guidelines and minor changes to prescription patterns. This emphasizes the need for continuous, multidimensional improvement strategies on prescribing practice alignment with guidelines.
Code
HSD72
Topic
Epidemiology & Public Health, Methodological & Statistical Research, Real World Data & Information Systems
Topic Subcategory
Artificial Intelligence, Machine Learning, Predictive Analytics, Disease Classification & Coding, Health & Insurance Records Systems
Disease
Gastrointestinal Disorders