Abstract Objectives To evaluate the relationship between age at onset of cannabis use and cannabis abuse in a sample of Portuguese cannabis users, testing the mediating role of the number of friends using cannabis and comparing these relationships between male and female subsets. Method A sample of 529 Portuguese cannabis users comprising 276 males and 244 females aged from 14 to 21 years completed a sociodemographic questionnaire, the Cannabis Abuse Screening Test (CAST), and answered four questions related to cannabis use. Data were analyzed using the PROCESS macro in SPSS version 26.0. Results Age at onset of cannabis use was negatively related to the number of friends using cannabis, while the number of friends using cannabis had a positive relationship with cannabis abuse. As predicted, the number of friends using cannabis seems to have a mediating role in the relationship between cannabis onset age use and cannabis abuse, since the indirect effect was found to be significant. The pattern of the relationships among the variables observed was found to be similar for both male and female subsets of the sample. However, males reported higher cannabis abuse than females. Conclusion The results highlight the mediating role of friends’ use of cannabis in the relationship between age at onset of cannabis use and users’ abuse. These results highlight the importance of early intervention in cannabis use. In particular, the pattern of the relationships observed among the studied variables emphasizes the need to monitor and conduct peer training interventions or interventions to promote of social skills.

Objective: to present some intervening factors in drinking behavior, risky consumption and alcohol-dependence found in a population group of Medellin city in Colombia. Methods: cross-sectional multi-method study of 538 university students. Several expert-validated instruments were applied and through Rash's method, such as a test to identify alcohol consumption-related disorders, an alcohol expectancy questionnaire, expectancies, a questionnaire of beliefs on alcohol consumption and a questionnaire on rites, social and family traditions. Data were collected from March to June 2010. For the correlation analysis among the qualitative variables, Chi-square test of independence, Fisher's exact test and Z test were used to compare proportions. The value used was p £ 0.05. Results: out of the total number of students who were interviewed, 82.3 % had taken alcohol at some time of their lifetime, 66 % had no problems with alcohol consumption, 21.6 % showed harmful consumption and 12.5 were already alcohol-dependent, according to the test of identification of alcohol consumption-related disorders. Conclusions: expectancies, beliefs, advertising, family and social customs have favorable association with the alcohol consumption in university students in Medellin, Colombia.

Objetivo: dar cuenta de algunos elementos intervinientes en la conducta de beber, consumo de riesgo y la dependencia del alcohol, en un grupo poblacional de la ciudad de Medellín, Colombia. Métodos: estudio multimétodo de corte transversal en 538 estudiantes universitarios. Se aplicaron varios instrumentos previamente validados por expertos, entre ellos, el test de identificación de los trastornos debidos al consumo de alcohol, el cuestionario acerca de las expectativas hacia el alcohol y el cuestionario de creencias acerca del consumo de alcohol. Los datos se recopilaron entre marzo y junio de 2010. Para el análisis de correlaciones entre variables cualitativas se utilizó la prueba ji-cuadrada de independencia, la prueba exacta de Fisher y la prueba Z, para comparar las proporciones. En todos las ocasiones se trabajó con p £ 0,05. Resultados: del total de estudiantes entrevistados el 82,3 % había consumido alcohol en algún momento de su vida, el 66,0 %, no tenían problemas con el consumo de esta sustancia, 21,6 % tenían consumo perjudicial y el 12,5 % presentaban ya dependencia, según el test de identificación de los trastornos debidos al consumo de alcohol. Conclusiones: las expectativas, creencias, la publicidad, las costumbres familiares y sociales, tienen una relación favorable para el consumo de alcohol en estudiantes universitarios en Medellín, Colombia.

Objective: to present some intervening factors in drinking behavior, risky consumption and alcohol-dependence found in a population group of Medellin city in Colombia. Methods: cross-sectional multi-method study of 538 university students. Several expert-validated instruments were applied and through Rash's method, such as a test to identify alcohol consumption-related disorders, an alcohol expectancy questionnaire, expectancies, a questionnaire of beliefs on alcohol consumption and a questionnaire on rites, social and family traditions. Data were collected from March to June 2010. For the correlation analysis among the qualitative variables, Chi-square test of independence, Fisher's exact test and Z test were used to compare proportions. The value used was p £ 0.05. Results: out of the total number of students who were interviewed, 82.3 % had taken alcohol at some time of their lifetime, 66 % had no problems with alcohol consumption, 21.6 % showed harmful consumption and 12.5 were already alcohol-dependent, according to the test of identification of alcohol consumption-related disorders. Conclusions: expectancies, beliefs, advertising, family and social customs have favorable association with the alcohol consumption in university students in Medellin, Colombia.

Objetivo: dar cuenta de algunos elementos intervinientes en la conducta de beber, consumo de riesgo y la dependencia del alcohol, en un grupo poblacional de la ciudad de Medellín, Colombia. Métodos: estudio multimétodo de corte transversal en 538 estudiantes universitarios. Se aplicaron varios instrumentos previamente validados por expertos, entre ellos, el test de identificación de los trastornos debidos al consumo de alcohol, el cuestionario acerca de las expectativas hacia el alcohol y el cuestionario de creencias acerca del consumo de alcohol. Los datos se recopilaron entre marzo y junio de 2010. Para el análisis de correlaciones entre variables cualitativas se utilizó la prueba ji-cuadrada de independencia, la prueba exacta de Fisher y la prueba Z, para comparar las proporciones. En todos las ocasiones se trabajó con p £ 0,05. Resultados: del total de estudiantes entrevistados el 82,3 % había consumido alcohol en algún momento de su vida, el 66,0 %, no tenían problemas con el consumo de esta sustancia, 21,6 % tenían consumo perjudicial y el 12,5 % presentaban ya dependencia, según el test de identificación de los trastornos debidos al consumo de alcohol. Conclusiones: las expectativas, creencias, la publicidad, las costumbres familiares y sociales, tienen una relación favorable para el consumo de alcohol en estudiantes universitarios en Medellín, Colombia.