ARE DEMOGRAPHIC CHARACTERISTICS AND SOCIAL NETWORK ASSOCIATED WITH OPIOID USE?
Annotation
Objective: This study aims to 1) examine the predictors of opioid misuse 2) build a predictive model for opioid misuse using artificial neural network and compare its performance to logistic regression model.
Methods: Youth Risk Behavior Surveillance System (YRBSS) 2017 data were used for this study. All the participants who were eligible were randomly assigned into 2 groups: training sample and testing sample. Two models were built using training sample: artificial neural network and logistic regression. Receiver operating characteristic (ROC) were calculated and compared for these two models for their discrimination capability and a curve using predicted probability versus observed probability was plotted to demonstrate the calibration measure for these two models.
Results:
About 13.3% of 7274 high school students had opioid misuse, about 14.1% among the female and 12.5% among the male.
According to the logistic regression, Q2 (What is your sex?), Q6 (How tall are you without your shoes on), Q31(How old were you when you first tried cigarette smoking, even one or two puffs?), Q41 (How old were you when you had your first drink of alcohol other than a few sips?), Q47 (How old were you when you tried marijuana for the first time), Q60 (How old were you when you had sexual intercourse for the first time?), Q61 (During your life, with how many people have you had sexual intercourse?), Q64 (The last time you had sexual intercourse, did you or your partner use a condom?) were significantly associated with the Opioid Misuse in the high school students.
According to this neural network, the top 5 most important predictors were Q47 (How old were you when you tried marijuana for the first time), Q4 (Are you Hispanic or Latino?), Q62 (During the past 3 months, with how many people did you have sexual intercourse?), Q1 (How old are you?), and Q3 (In what grade are you).
For training sample, the ROC was 0.79 for the Logistic regression and 0.80 for the artificial neural network. In testing sample, the ROC was 0.76 for the Logistic regression and 0.79 for the artificial neural network.
Conclusions: In this study, we identified several important predictors for opioid misuse e.g., marijuana use, race, sex experience. This correlation between opioid misuse and other risk behaviors was evident. A program to prevent opioid misuse should take into consideration of other risk behaviors.
Keywords
References:
[1] Opioids and Adolescents. https://www.hhs.gov/ash/oah/adolescent-development/substance-use/drugs/opioids/index.html
[2] Monitoring the Future Survey: High School and Youth Trends. https://www.drugabuse.gov/publications/drugfacts/monitoring-future-survey-high-school-youth-trends
[3] Peng, C. J., Lee, K. L., Ingersoll, G. M. An Introduction to Logistic Regression Analysis and
Reporting. The Journal of Educational Research, 96(1), 3-14.
[4] Tabachnick, B., and Fidell, L. Using Multivariate Statistics (4th Ed.). Needham Heights,
MA: Allyn & Bacon, 2001.
[5] StatSoft, Electronic Statistics Textbook, http://www.statsoft.com/textbook/stathome.html.
[6] Stokes, M., Davis, C. S. Categorical Data Analysis Using the SAS System, SAS Institute
Inc., 1995.