A healthcare cognitive decision support system for doctor–patient co‐decisions in primary care using sentiment analysis

Abstract Decision support systems (DSS) in the artificial intelligence era require the integration of cognitive and affective information to support human‐centred decision‐making. This study proposes a healthcare cognitive decision support system (HC‐DSS) based on lexicon sentiment analysis for doctor–patient co‐decisions in primary care. Thirty audio transcriptions from actual consultations involving patients with hypertension (HTA) or dyslipidaemia (DLM) are analysed using the Spanish sentiment analysis‐decision support system (SSA‐DSS) updated with automatic speaker identification. Results show that HTA consultations exhibit greater affective intensity and variability across both positive and negative emotional dimensions, while DLM consultations present comparatively more stable and less reactive affective patterns. Phase‐based analyses further reveal that emotional valence and trust follow structured, condition‐specific trajectories, with peak sensitivity during treatment and decision‐making stages. Based on these exploratory findings, a four‐phase consultation model is proposed as a sentiment‐aware DSS framework to support more emotionally informed and personalised doctor–patient co‐decisions in primary care, as well as the education and empowerment of individuals and society in healthy lifestyle habits.