Method of determining the severity of alcohol dependence syndrome by neuropsychological indicators

Abstract

   Objective. To develop a method of determining the severity of alcohol dependence syndrome using the automated program complex NeuroSMG.
   Materials and methods. The study involved 397 practically healthy participants, as well as 251 patients of the narcology department of the Vitebsk regional clinical center of psychiatry and narcology and citizens staying in the treatment and labor preventorium № 4 of Vitebsk. Vitebsk. The participants were tested in the automated program complex of neuropsychological diagnostics NeuroSMG. Statistical processing of the results was carried out in the R environment version 4.4.1.
   Results. The following indices had the greatest predictive ability: time of correct CBT responses, time of No-Go GNG, proportion of yellow ball RAT saves, number of CBT errors, wrong CBT time, sum of RAT gains, SCT response duration, duration of red ball RAT touches, average TMT path length. Based on these, a decision tree was constructed. During validation, the high quality of the obtained decision tree was shown: thus, the accuracy of the model was 91.05 % (with 95 % confidence interval equal to 88.58—93.13 %).
   Conclusion. As a result of this study, a method for determining the severity of alcohol dependence syndrome using the NeuroSMG automated software package was proposed. The obtained data demonstrate high sensitivity of the method: 97.73 % for the comparison group, 86.78 % and 74.62 % for the groups of patients with the initial (first group) and middle (second group) stages of alcohol dependence syndrome, respectively. Specificity rates were also high, amounting to 86.85 % for the comparison group and 98.10 % and 97.10 % for the first and second groups.