Validação Concorrente e Confiabilidade da Versão Brasileira do ASSIST-WHO (Smoking and Alcohol Substance Involvement Screening Test) for Teenagers. / Concurrent Validation and Reliability of the Brazilian Version of WHO-ASSIST (Alcohol Smoking and Substance Involvement Screening Test- World Health Organization) for Teenagers
AUTOR(ES)
Vanessa de Jesus Patrick Carminatti
DATA DE PUBLICAÇÃO
2010
RESUMO
Rationale: Given the large number teenagers using alcohol and other drugs, screening tools are necessary to assess drug use and the presence of problems associated aiming an appropriate planning for prevention, intervention or treatment strategies. Objective: The objective of this study was to assess, in a sample of Brazilian teenager, the concurrent validity and to evaluate the internal consistency of the Brazilian Portuguese version of the WHO ASSIST (Alcohol Smoking and Substance Involvement Screening Test) instrument, already used in adults. Method: A hundred eighty three teenagers between the ages of 11 to 19 years were interviewed. Ninety were not users (or occasional users) and 93 met criteria for drug use or addiction. Besides the ASSIST, two other instruments, CIDI (Composite International Diagnostic Interview) and TEEN-ASI (Addiction Severity Index - for teenagers), were used as gold standards. We calculated the sensitivity, specificity, negative and positive predictive values for screening of each drug risk use/dependence. Internal consistency was assessed by Cronbachs alpha and the agreement among the instruments was analyzed by kappa index. Results: The instrument reliability was considered good (Cronbachs alpha: 0.81 for tobacco, 0.78 for alcohol, 0.82 for marijuana and 0.83 for cocaine). The agreement between ASSIST - CIDI and ASSIST - TEEN-ASI was evaluated by kappa indexes which ranged from satisfactory (64% for alcohol) to very good (93% for marijuana). Sensitivity values ranged from 61% for alcohol to 100 % for tobacco, while specificity ranged from 76% to 99% for other drugs. Conclusions: ASSIST performance as screening tool for use of risk/dependence on drugs by teenagers presented reasonable psychometric values. It can be administered for early detection of tobacco, marijuana and cocaine/crack use in this specific population. However, sensitivity for alcohol use was relatively low, suggesting that further adjustments are required on the ASSIST cut-off points for alcohol.
ASSUNTO(S)
adolescente alcoolismo tabaco drogas ilícitas estudos de validação psiquiatria adolescent alcoholism tobacco street drugs validation studies
