Researchers have used artificial intelligence to predict the recovery of people with Generalized Anxiety Disorder (GAD) after nine years. Analyzing data from 126 patients, they found that factors such as depressive symptoms, discrimination, and frequent medical visits were linked to a lower chance of recovery. Older age, higher education, support from friends, and positive emotions increased the chances of improvement. These findings could help doctors create more personalized treatments for people with GAD.
Generalized Anxiety Disorder (GAD) is a chronic condition that affects millions of people worldwide, causing excessive and persistent worry about many aspects of life.
Identifying factors that can predict the recovery trajectory of these people over time could be essential to developing more personalized treatment approaches.
This study used machine learning techniques to analyze whether it would be possible to predict which patients with Generalized Anxiety Disorder would recover after a period of nine years.
The study included 126 participants diagnosed with Generalized Anxiety Disorder. To predict how the condition would progress over time, researchers collected a variety of information about the participants at baseline, including psychological aspects (such as depressive symptoms), social factors (such as peer support), biological data (such as waist-to-hip ratio), sociodemographic characteristics (such as education level and age), and health information (such as frequency of visits to medical and mental health professionals).
They then applied two machine learning models to analyze this data and predict whether or not the participants would recover from Generalized Anxiety Disorder after nine years.
The models used were gradient boosted decision trees and elastic nets, a type of statistical regression that combines different methods to increase the accuracy of predictions.
The results showed that, over the nine years of follow-up, 95 of the 126 participants (75.4%) showed recovery from Generalized Anxiety Disorder. The elastic net model performed strongly in predicting outcomes, achieving a score of .81 on the metric called "area under the ROC curve" (AUC), which assesses the model's ability to distinguish between individuals who recovered and those who did not.
In addition, this model showed a balanced accuracy of 72%, with a sensitivity of 70% (ability to correctly identify those who did not recover) and a specificity of 76% (ability to correctly identify those who recovered).
The factors that most indicated a higher probability of non-recovery were: greater presence of depressive symptoms, frequent experience of discrimination in daily life, greater number of consultations with mental health professionals and greater number of medical consultations in general.
On the other hand, the factors that increased the chance of recovery included: having at least some level of higher education, older age, greater support from friends, higher waist-to-hip ratio and greater presence of positive emotions.
The findings of this study demonstrate that it is possible to predict, with an acceptable level of accuracy, whether an individual with Generalized Anxiety Disorder has a greater or lesser chance of recovery after nine years.
This represents an advance in research on the outcomes of Generalized Anxiety
Disorder and may contribute to the development of more targeted preventive strategies. With this information, mental health professionals can create personalized approaches for each patient, focusing on the factors that increase the chances of recovery.
This study also represents an important first step towards the practical use of machine learning models in predicting the progression of Generalized Anxiety Disorder, which could significantly improve care for people living with this condition.
READ MORE:
Development of a machine learning-based multivariable prediction model for the naturalistic course of generalized anxiety disorder” by Candice Basterfield and Michelle G. Newman, 25 January 2025, Journal of Anxiety Disorders.DOI: 10.1016/j.janxdis.2025.102978
Abstract:
Generalized Anxiety Disorder (GAD) is a chronic condition. Enabling the prediction of individual trajectories would facilitate tailored management approaches for these individuals. This study used machine learning techniques to predict the recovery of GAD at a nine-year follow-up. The study involved 126 participants with GAD. Various baseline predictors from psychological, social, biological, sociodemographic and health variables were used. Two machine learning models, gradient boosted trees, and elastic nets were compared to predict the clinical course in participants with GAD. At nine-year follow-up, 95 participants (75.40 %) recovered. Elastic nets achieved a cross-validated area-under-the-receiving-operator-characteristic-curve (AUC) of .81 and a balanced accuracy of 72 % (sensitivity of .70 and specificity of .76). The elastic net algorithm revealed that the following factors were highly predictive of nonrecovery at follow-up: higher depressed affect, experiencing daily discrimination, more mental health professional visits, and more medical professional visits. The following variables predicted recovery: having some college education or higher, older age, more friend support, higher waist-to-hip ratio, and higher positive affect. There was acceptable performance in predicting recovery or nonrecovery at a nine-year follow-up. This study advances research on GAD outcomes by understanding predictors associated with recovery or nonrecovery. Findings can potentially inform more targeted preventive interventions, ultimately improving care for individuals with GAD. This work is a critical first step toward developing reliable and feasible machine learning-based predictions for applications to GAD.
Comentarios