Supporting tools in psychiatric treatment decision-making: Sertraline outcome investigation with artificial neural network method

• Published in: Psychiatry research 2005-04, Vol.134 (2), p.181-189
• Main Authors: Politi, Ernestina, Franchini, Linda, Spagnolo, Chiara, Smeraldi, Enrico, Bellodi, Laura
• Format: Article
• Language: English
• Subjects: Adult; Aged; Aged, 80 and over; ANN; Biological and medical sciences; Decision Making; Female; Forecasting; Humans; Information theory; Male; Medical sciences; Mental Disorders - therapy; Middle Aged; Models, Theoretical; Neural Networks (Computer); Neuropharmacology; Pharmacology. Drug treatments; Psychoanaleptics: cns stimulant, antidepressant agent, nootropic agent, mood stabilizer..., (alzheimer disease); Psychology. Psychoanalysis. Psychiatry; Psychopharmacology; Sampling Studies; Selective serotonin reuptake inhibitor; Serotonin Uptake Inhibitors - therapeutic use; Sertraline; Sertraline - therapeutic use; Treatment Outcome
• ISSN: 0165-1781; 1872-7123
• DOI: 10.1016/j.psychres.2004.07.011

Controlled trials in clinical psychopharmacology may fail to provide reliable information about the benefit of treatment for the patient when considered in a real-life setting rather than as a part of a well-defined sampling procedure. Previously, we applied the mathematical model of an artificial neural network (ANN) to a pool of clinical information gathered through case descriptions provided by senior psychiatrists in clinical charts of patients receiving their first exposure to sertraline. In the present study, we applied the same mathematical model to a larger sample. The performance of the ANN model in forecasting successful and unsuccessful treatment showed an overall accuracy of classification of 97.12%. This result supports our previous finding about the potential application of this method as a reliable predictor of a given psychiatric patient's outcome during a specific psychopharmacological therapy.