Artificial Intelligence-Enabled Clinical Decision Support for Chronic Disease Management in Primary Care: Implications for Family Nurse Practitioner Practice

Abstract

Artificial intelligence-enabled clinical decision support is increasingly being applied to chronic disease management in primary care. This structured integrative review examined how AI supports risk prediction, complication detection, patient prioritisation, screening, education, and care coordination, with emphasis on implications for family nurse practitioner practice. PubMed was used as the principal information source, with studies limited to English-language, human, free full-text publications from January 2018 to July 2026. Seventy-one records were identified, one duplicate was removed, and 70 unique records were screened. Seventeen publications were retained, including 15 AI-related studies and two contextual digital decision-support studies. The evidence showed that AI can improve case finding for diabetes, predict chronic kidney disease and cardiovascular risk, identify complications in free-text records, support diabetic-retinopathy screening, and assist with patient segmentation and self-management education. However, performance varied across settings, and several models were limited by poor calibration, internal validation, workflow burden, and uncertain transferability to primary care. Direct evidence involving family nurse practitioners was limited, but the findings were highly relevant to screening, monitoring, referral, patient education, and care coordination. AI should therefore be used as a support tool rather than an independent decision-maker. Safe implementation requires clinical oversight, local validation, transparency, privacy protection, equity monitoring, and appropriate professional training.