Clinical decision support systems based on artificial intelligence and pharmacogenomics in community pharmacy: systematic review and proposal of the milfc-mp framework for precision medicine

Abstract

The convergence of artificial intelligence (AI), pharmacogenomics (PGx) and community pharmacy is one of the most promising frontiers of contemporary precision medicine, yet no translational framework has been proposed for community pharmacies within the public health systems of emerging countries. This article aims to conduct an integrative systematic review of AI-based clinical decision support systems (CDSS) with PGx integration in clinical pharmacy, to map gaps in the literature, and to propose the MILFC-MP framework as a translational implementation model for community pharmacies in the Brazilian Unified Health System (SUS). The review followed PRISMA 2020 guidelines, searching PubMed/MEDLINE, Embase, Scopus, Web of Science and SciELO for publications from 2014 to 2026, with analytical structuring based on the Consolidated Framework for Implementation Research and Proctor et al.'s implementation outcomes. Seventeen studies were included, involving 9,847 patients across eleven countries, using rule-based algorithms, supervised machine learning, deep learning and large language models. Seven critical gaps were mapped, including the near-absence of models for community pharmacy in low- and middle-income countries, the lack of telepharmacy integration and the absence of cost-effectiveness analysis for the SUS. In response, the MILFC-MP framework is proposed, operating across seven integrated layers. It is concluded that MILFC-MP is the first integrative framework of AI, pharmacogenomics and Lean specifically designed for community pharmacies in public health systems, awaiting prospective multicentric validation.