Pharmacist-led integration of autonomous AI screening and conversational AI for ocular disease management

The growth of autonomous artificial intelligence (AI) systems for diabetic retinopathy screening and conversational AI tools for glaucoma management has the potential to substantially enhance community pharmacy practice. However, limited evidence exists regarding the real-world use and integration of these technologies. This study assessed the effectiveness of pharmacist-facilitated AI-driven diabetic retinopathy (DR) screening on screening uptake and referral completion and evaluated the effect of a pharmacist-supervised large language model (LLM) chatbot on medication adherence and intraocular pressure (IOP) in patients with primary open-angle glaucoma. A pragmatic 6-month study (July through December 2024) was conducted in four community pharmacies in Al-Karak, Jordan. Phase 1 included adults with diabetes receiving AI-assisted fundus photography (AEYE-DS), whereas Phase 2 randomized glaucoma patients to receive either text-message reminders or pharmacist-supported chatbot interactions using WhatsApp (GPT-4). In Phase 1, 402/518 participants (77.6%) completed screening; 16.9% needed referral, and 81.0% attended ophthalmology visits. The chatbot group in Phase 2 ( n = 94) showed improved medication possession ratio (MPR) adherence (MPR 0.91 ± 0.06 vs. 0.79 ± 0.11; p < 0.001), fewer missed doses (3.2 ± 2.1 vs. 7.0 ± 3.4; p = 0.002), and greater reductions in IOP (2.8 ± 3.6 vs. 0.9 ± 4.1 mmHg; p = 0.010). These results indicate that the integration of AI-based services into community pharmacy care may be beneficial.