Automatic processing techniques applied to the analysis and prediction of drug use

Executive Summary: Background and justification. The consumption of alcohol and other drugs continues to constitute one of the main public health problems due to its consequences for the consumer and for their environment, including health problems, intrafamily violence and loss of productivity in Ecuador. However, prevention programs and/or effective public action policies continue to be a challenge according to the data provided by the different National Observatories on Drugs, the National Commission for Development and Life Without Drugs (DEVIDA), and the Inter-American Commission for the Control of Drug Abuse (CICAD).

Aim. The objective of this project is to apply emerging artificial intelligence algorithms to predict problematic and controlled drug use based on sociodemographic, psychosocial, and health information; and the creation of an "interuniversity observatory for the study of alcohol and other drugs" that guarantees access to updated information, the periodic dissemination of the results, and that serves as support for the design of prevention programs and public policies on drugs, without prejudice to the powers that other institutions from other territories may exercise on this matter. Methodology. Variables of alcohol, tobacco and other drug consumption will be evaluated, as well as sociodemographic variables; psychosocial: stress, social support and tendency to risk; and health: psychological inflexibility, depressive and anxious symptoms, life satisfaction and adverse childhood experiences. The collection of information will be carried out through web-based questionnaires previously selected based on three criteria: brevity, psychometric properties, and open for research purposes. The web administration of the questionnaires minimizes the risk of data loss and allows the automatic calculation of scores immediately, with the consequent optimization of time and delivery of immediate feedback based on the identified needs. The generation of predictive models for the dynamic identification of problematic and controlled consumption of alcohol and other drugs will be carried out using emerging artificial intelligence algorithms and parameterizable prediction systems based on sociodemographic, psychosocial and health data mining algorithms. Expected results. The project will involve the creation of an inter-university observatory for the study of alcohol and other drug use; the detection of levels of alcohol or other drug use, the identification of risk groups according to their sociodemographic, psychosocial and health characteristics; and the generation of predictive models of problematic and controlled consumption of alcohol and other drugs through artificial intelligence. Impact and diffusion. In the first place, the predictive models of drug use, regardless of its legal status, will allow the design of prevention programs and more effective public policies, acting on the psychosocial determinants of health, associated with the drug problem as a social problem and not only like a medical problem. Secondly, the creation of an inter-university observatory for the study of alcohol and other drugs will allow monitoring the impact of prevention programs and current public policies, of interest for compliance with the Andean Standard for the Control of narcotic drugs and psychotropic substances, and the Andean Strategy for Comprehensive and Sustainable Alternative Development, which includes actions such as the support project for reducing the demand for illicit drugs in the Andean community (PREDEM) or the anti-drug program in the Andean community (PRADICAN). Finally, the results will be presented through contributions to national and international conferences and publications through articles in high-impact peer review journals, reinforcing the research activity of the participating institutions.

General Objective: The objective of this project is to apply emerging artificial intelligence algorithms to predict problematic and controlled drug use based on sociodemographic, psychosocial and health information; and the creation of an "observatory for the study of alcohol and other drugs" that guarantees access to updated information and the periodic dissemination of the results, and that serves as support for the design of prevention programs and public policies on drugs, without prejudice of the powers that other institutions from other territories may exercise on this matter.

Specific Objectives The creation of a digital observatory for the study of the consumption of alcohol and other drugs will allow:

  • Collection of information on the consumption of alcohol and other drugs, and sociodemographic, psychosocial and health variables associated with consumption, using an online platform designed specifically for this purpose and gamification systems.
  • Detection of levels of consumption of alcohol or other drugs according to sociodemographic criteria and risk of alcohol consumption according to WHO criteria.
  • Generation of predictive models for the dynamic identification of problematic and controlled consumption of alcohol and other drugs through artificial intelligence.

Participating Institutions:

UTPL, UTN, UPS, U. SALAMANCA, U.LOYOLA ANDALUCÍA.

Participants:

Project manager Pablo Vicente Torres Carrión.

  • Pablo Vicente Torres Carrion
  • Silvia Libertad Vaca Gallegos
  • Priscila Marisela Valdiviezo Diaz
  • Victor Manuel Lopez Guerra
  • Byron Fernando Bustamente Granda
  • Jorge Edmundo Gordon Rogel
  • Gino Eduardo Grondona Opazo
  • Maria Jose Boada Suraty
  • Luis Fernando Garces Velasquez
  • Pablo Ruisoto Palomera
  • Carla Lopez Nunez

Awarded budget: $60174

Project status: Signing of agreements.