HARCP Archives

Browse by article

Browse by volume

Heroin Addiction and Related Clinical Problems: 2024, 26, 16

Using genetic parameters to predict somatization in females with heroin dependence

Kadir Uludag, Hang Su, Haifeng Jiang, Jiang Du, Na Zhong, and Min Zhao

Digital Object Identifier:
https://doi.org/10.62401/2531-4122-2024-16

Summary: Background. Heroin dependence is a severe psychiatric disorder that is often associated with various comorbidities, including somatic symptoms such as pain, anxiety, weakness, tremors, muscle pain, sleep disturbances, and shortness of breath, as well as personality disorders, mood, and psychotic disorders. While heroin use may initially alleviate these symptoms, it can also exacerbate them over time. There is a notable gap in our understanding of the genetic factors, such as catechol-O-methyltransferase (COMT), that can potentially aid in predicting susceptibility to somatisation. Aim. We aimed to utilise specific genetic parameters, such as COMT and stress-related factors, to indicate the somatisation status accurately. Methods. Only female patients with heroin dependence were recruited for the study. We have compared the groups with and without somatisation. Cross-validation was conducted to validate the results obtained from the machine learning (ML) process. Results. Random Forest (RF) demonstrated exceptional performance in predicting somatisation status, achieving an impressive accuracy of 99% (sensitivity:100%, specificity:98%). The mean accuracy of cross-validation for the model was 90%, indicating a reliable performance. Conclusions. The genetic factors investigated in this study did not show any association with the status of heroin use. We achieved a successful prediction of somatisation status solely based on genetic parameters.

Keywords: heroin; heroin addiction; heroin dependence