Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Dec 18, 2022
Date Accepted: Jun 28, 2023
Analysis of the Adverse Effects and Nonmedical Use of Methylphenidate Before and After the Outbreak of COVID-19 Using Twitter, Facebook, and Instagram: Machine Learning Analysis in Cross-Sectional Study
ABSTRACT
Background:
Attention-deficit/hyperactivity disorder (ADHD) has been increasingly diagnosed in adult including child and adolescents. Methylphenidate is the first-line treatment for ADHD and have several adverse effects, including sleep problems, loss of appetite, anxiety and psychiatric disorders. Social networking services (SNSs) and machine learning can be effective methods for analyzing the status of methylphenidate use, which is difficult to identify using clinical trials.
Objective:
This study analyzed the adverse effects and nonmedical use of methylphenidate and evaluated the change in frequency of nonmedical use before and after the outbreak of Covid-19. The study also evaluated the performance of machine-learning models.
Methods:
In this cross-sectional study, SNS data on methylphenidate from Twitter, Facebook, and Instagram was from January 2019 to December 2020. Among the collected data, first-hand experience data were used and annotated according to their medical use, adverse effects, and nonmedical use. The frequency of adverse effects, nonmedical use, and drug use comparison before and after pandemic was analyzed. Additionally, machine-learning models were built using the data, and their performance was evaluated.
Results:
This study collected 146,352 data points and detected that 4.3% (6,340/146,352) were first-hand experience data. Psychiatric problems (521/1,683, 31.0%) had the highest frequency among the adverse effects. Nonmedical use for study or work had the highest frequency (741/2,016, 36.8%). While the frequency of nonmedical use before and after the pandemic was similar (odds ratio, 1.02; 95% confidence interval, 0.91–1.15), a comparison of each nonmedical use showed significantly different trends. Of the machine-learning models, random forest had the highest performance of 0.75.
Conclusions:
The machine learning models using SNS data applicable to analyze the adverse effects and nonmedical use of methyphenidate. The highest performance was showed in random forest methods. The study results will contribute to the analysis of the status of methylphenidate use and the application of machine learning using SNS data for automatic monitoring tasks. Clinical Trial: none
Citation
The author of this paper has made a PDF available, but requires the user to login, or create an account.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.