3.1. Adolescents’ Perception of Substance Use
In response to the question Do you feel sufficiently informed about the drugs issue? in the 2016 ESTUDES questionnaire, 66.7% of the adolescents replied that they were sufficiently or perfectly informed and 27% said that they were poorly or only partly informed, while 6.3% did not answer. Teenage girls believed that they were less and/or worse informed (33.7%) than teenage boys (23.9%) (χ2 = 399.8; p < 0.000; the effect size was low, i.e., d = 0.214). This difference increases with age: at 18 years the percentage points difference is 18 in favor of teenage boys (who believe they are well informed), while, at 14 years, this difference is nine percentage points (also in favor of teenage boys). For the sample as a whole, age 14 is when the lack of information is perceived to be greatest (by 31.5% of all adolescents).
With regard to the age at which adolescents begin to consume these substances, 27.2% of adolescents who smoke smoked their first cigarette at the age of 14 and 22.9% did so at the age of 15, while 31.4% of those who drink drank their first alcoholic beverage at 14 and 23.6% did so at 15. We observe that a higher percentage of girls (52.1%) than boys (47.7%) started smoking between the ages of 14 and 15 when the data are disaggregated by sex. Similarly, a higher percentage of girls (56.7%) than boys (53.3%) had their first alcoholic drink between the ages of 14 and 15.
With regard to adolescents who have been drunk, 27.3% got drunk for the first time at the age of 14, while 31.1% did so at the age of 15. Teenage boys and teenage girls did not present significant differences, even in the age at which they began to drink alcoholic beverages practically every week.
shows the percentages of adolescents who stated that consumption implied few or no problems. The points indicate the percentage of adolescents who did not know how to reply to this question and the dashed line indicates the percentage of adolescents who reported that they were not sufficiently informed.
A proportion of adolescents believed that low levels of consumption have no negative consequences on their health. For example, 27.5% believed that consuming 1–5 cigarettes a day implied hardly any problems, while the figures for drinking five or six beers or other alcoholic beverages at weekends, drinking one or two beers or other alcoholic beverages every day, smoking hashish or marijuana occasionally, and smoking electronic cigarettes were 36.2%, 39%, 37.3%, and 44.6%, respectively.
Adolescents are more aware of the dangers involved in consuming less common substances, such as cocaine, ecstasy, heroin, and amphetamines, etc. Nevertheless, the percentages of non-responses due to ignorance about whether consuming such substances can be harmful to one’s health are high and a proportion of adolescents believes that occasionally consuming them has hardly any consequences on health at all. For cocaine, this figure is 17.2%, while those for ecstasy, amphetamines/speed, hallucinogens, heroin, magic mushrooms, anabolics, and methamphetamines are 14.6%, 14.1%, 14.3%, 13.3%, 12.2%, 10.7%, and 10.1%, respectively.
Teenage boys, more than teenage girls, underestimate the consequences of consuming the most prevalent substances (cigarettes, alcohol, hashish/marijuana, and electronic cigarettes), especially when their consumption is habitual. For example, 7.8% of teenage boys, as opposed to 6.8% of teenage girls, believe that smoking a packet of cigarettes every day causes few or no problems. The figures for other substances are as follows: smoking electronic cigarettes, 45.1% (for boys) as opposed to 44.1% (for girls); drinking one or two beers or alcohol beverages every day, 41.3% as opposed to 36.7%; drinking five or six beers or alcohol beverages every day, 9.2% as opposed to 7.2%; smoking hashish/marijuana every day, 9.8% as opposed to 6.0%; taking powdered cocaine powder habitually, 3.4% as opposed to 2.5%; regularly taking ecstasy, 3.5% as opposed to 2.5%; taking amphetamines/speed habitually, 3.6% as opposed to 2.4%; and, taking hallucinogens habitually, 3.4% as opposed to 2.4%. In all cases, the differences between teenage boys and teenage girls are statistically significant (p < 0.001).
3.2. The Consumption of Tobacco and Hashish/Marijuana
The logistic regression model for the consumption (Yes/No) of tobacco in the previous 30 days enables correct estimation in 82.0% of cases (χ2 = 8375.4; p < 0.001), with Nagelkerke R2 estimating an adjustment value of 0.356. The probability that an adolescent had smoked tobacco in the previous 30 days when either their mother or father (or both) are smokers was 1.30 times higher than for adolescents whose parents do not smoke (CI = 1.21–1.39; d = 0.145). For adolescents, most of whose friends smoke, the probability was 94.0 times higher (CI = 79.3–111.1; d = 2.504), while if only some of their friends smoke, the probability was 14.2 times higher (CI = 12.04–16.82). The odds ratio corresponding to the level of information shows that the probability that an adolescent will have smoked tobacco when they believe they are well informed is 1.30 times higher than when they believe that they are not so well informed (CI = 1.21–1.39).
In the model for hashish/marijuana consumption (Table 3
), estimation is correct in 88.6% of cases (χ2 = 6937.4; p
< 0.001; Nagelkerke R2
= 0.370). The probability that an adolescent will have consumed hashish/marijuana in the previous 30 days increases when their mother is a smoker (OR = 1.29; CI = 1.19–1.40: d
= 0.1404) or their father is a smoker (OR = 1.15; CI = 1.05–1.24; d
= 0.077), and especially when some of their friends consume hashish/marijuana (OR = 16.32; CI = 14.46–18.41; d
= 1.539) or most of their friends do (OR = 93.05; CI = 80.55–107.5; d
= 2.499). The probability that those who believe they are sufficiently informed will have consumed tobacco or hashish/marijuana is 1.59 greater (CI = 1.45–1.75; d
= 0.2557), in each case, than for those who believe that they are not so well informed.
We observe that the prevalence relationship is stronger if the majority of the adolescent’s friends smoke in the group who feel better informed (OR = 54.44; CI = 43.11–68.74) than in the group who feel worse informed (OR = 40.10; CI = 27.47–58.55) when consumption is controlled for the variable sufficiently or perfectly informed vs. partly or badly informed. However, there are no significant gender differences in either of these two groups.
3.3. Consumption of Alcohol and Forms of Alcohol Consumption
Estimation is correct in 78.6% of cases (χ2 = 5021.9; p < 0.001; Nagelkerke R2 = 0.23) in the model for alcohol consumption. The probability that an adolescent will have drunk is 37.3 times higher if most of their friends drink than if they do not (CI = 30.49–45.59; d = 1.995). It is also 8.08 times higher when only some of their friends drink (CI = 6.57–9.92; d = 1.152). An adolescent’s consumption of alcohol also increases when their mother is a regular drinker (OR = 1.21; CI = 1.08–1.37; d = 0.1051) and when they feel better informed (OR = 1.28; CI = 1.19–1.37; d = 0.1361).
In the model for getting drunk (Table 4
), estimation is correct in 80.9% of cases (χ2
= 6874.5; p
< 0.01; Nagelkerke R2
= 0.303). The probability that an adolescent will have got drunk is 44.8 times higher when most of their friends have got drunk than when they have not (CI = 39.44–50.89; d
= 2.096). Whether an adolescent gets drunk also depends on whether their mother is a regular drinker (OR = 1.17; CI = 1.03–1.32; d
= 0.086), whether only some of their friends have got drunk (OR = 8.97; CI = 7.98–10.15; d
= 1.209), and whether they feel well informed (OR = 1.23; CI = 1.14–1.31; d
Regarding binge drinking, the model provides correct estimation in 80.5% of cases (χ2 = 2912.0; p < 0.001; Nagelkerke R2 = 0.136). The probability that an adolescent will have binge drunk is 7.36 times higher if most of their friends have binge drunk than if they have not (CI = 6.77–8.00; d = 1.1005). This form of consumption depends on whether their fathers drink habitually (OR = 1.11; CI = 1.01–1.20; d = 0.0575), whether any of their friends have binge drunk (OR = 2.88; CI = 2.64–3.14; d = 0.5832), and whether they perceive themselves to be well informed (OR = 1.23; CI = 1.15–1.31; d = 0.1141).
We observe that adolescents who feel worse informed are slightly more exposed to both alcohol consumption and getting drunk when most of their friends have these behaviors than adolescents who feel better informed (unlike what occurred with the consumption of tobacco and hashish/marijuana) when the consumption of alcohol is controlled for the variable sufficiently or perfectly informed vs. badly or partly informed. The OR for alcohol consumption among adolescents when most of their friends drink is 21.4 (CI = 19.46–23.53) in the group who feel better informed and 22.05 (CI = 19.00–25.58) in the group who feel worse informed. When it comes to drunkenness, the OR are 40.34 (CI = 34.94–45–58) and 62.52 (CI = 47.43–82.42), respectively. There were no significant gender differences in either of the two groups or between them, nor were there any differences in the prevalence ratios between those who feel better or worse informed regarding binge drinking.