[Study on solvent abuse among junior high school students in Saitama, Japan]

The purposes of this study were to estimate the prevalence of past solvent abuse among junior high school students in Saitama prefecture and to investigate smoking and drinking as "gateway drugs" for solvent abuse.

The survey was conducted from Dec 2002 to Jan 2003, in 3 junior high schools in Saitama, using anonymous self-reporting to preserve privacy. A total of 2,049 (93.2%) students responded to our questionnaire. Two groups were categorized and analyzed according to history of solvent abuse in the past. To investigate "gateway effects" of smoking and drinking, we analyzed their relationship with solvent abuse, order of precedence of usage and confounding factors.

The prevalence of history of solvent abuse was 1.1% of the total. There were notable differences depending on sex and grade. 1.9% of male students had some experience as compared to only 0.3% of female students. An increase was evident according to grade (grade 1-0.6%, grade 2-0.8%, grade 3-2.1%) and there was a strong association between smoking, drinking and solvent abuse. The order of precedence of solvent abuse could not be established because some users had first experienced smoking, drinking and solvent abuse in the same grade and there was also a high recall bias among the subjects. A number of respondents also reported the reverse order, of experiencing solvent abuse before smoking and drinking. The user group was habitual smokers were recommended to smoke by their parents, spent longer periods in solitude, witnessed solvent abuse and had a friend who abused solvents. After adjusting for confounding factors using multiple logistic regression analysis, a relationship between smoking and solvent abuse remained but not the relationship with drinking.

The study conformed that there is solvent abuse among junior high school students, even in the lowest grade. Health education must focus on this fact and the finding that the user group reflects peer pressure and is faced with poor communication and advice from their families. Therefore health education focused on life skills and social support considering family ties must be emphasized. It could not be established which was first experienced, but a strong relationship between smoking and solvent abuse was observed even after multivariable analysis. On the other hand, the results suggested that habitual drinking is not a "gateway drug" to solvent abuse.