An AI-Powered Decision Support System in Smoking Cessation Clinics
Purpose of the Study
Patient assessment in smoking cessation clinics requires the comprehensive evaluation of numerous factors, including nicotine dependence, motivation to quit, previous quit attempts, behavioral triggers, and the social environment. This study was conducted to evaluate whether the AI-supported Clinical Decision Support System (CDSS), developed for the smoking cessation field, can analyze information obtained from patient interviews and provide clinicians with meaningful and structured recommendations.
Method
The study was designed as an early-stage Proof of Concept (PoC) research. Two sample smoking cessation consultations, created according to the standard medical history structure, were uploaded to the system, and the resulting outputs were compared with the results of independent thematic analyses conducted by the researchers. This allowed for an assessment of the extent to which the system can recognize behavioral and clinical patterns in patient narratives.
Findings
The study results demonstrated that the AI-supported system was able to accurately identify key behavioral and clinical patterns present in patient consultations to a large extent. In particular, stress, the influence of the social environment, previous cessation attempts, history of relapse, and difficulties experienced during the cessation process were consistently identified by the system.
The system did not merely analyze the current situation; but it has also been able to generate clinically meaningful recommendations such as behavioral coping strategies, the need for psychosocial support, a follow-up plan, and potential pharmacological approaches. The fact that the recommendations are tailored to the specific characteristics of each case demonstrates that the system can operate in a context-sensitive manner.
Importance of the Study
This study indicates that artificial intelligence can be used not to replace clinicians in smoking cessation clinics, but to support decision-making processes. The rapid and systematic evaluation of information obtained from patient consultations can help reduce clinicians’ workload while contributing to the development of more personalized counseling processes.
Conclusion
The findings demonstrate that the AI-powered Clinical Decision Support System can recognize clinical and behavioral patterns derived from smoking cessation consultations and convert them into supportive recommendations for the physician. However, this study is an early-stage proof-of-concept. Further validation studies using larger datasets and real patient-physician consultations are needed to demonstrate the system’s clinical potential.
For details on the study, which was also presented at the National Tobacco Control Congress;
https://tjtc.org/article/view/84