Initiation, patterns and risk factors

Figure 21a

Age of initiation to being drunk and cannabis use among 17- to 18-year-old boys in France in 2001

item Cannabis | item Being drunk

Sources: ESCAPAD (2001), OFDT. National representative sample.


In general, the likelihood of young people aged 12–18 years getting drunk or being offered cannabis, or any other illegal drug, as well as their willingness to try drugs, increases sharply with age. This is illustrated here by data from the French ESCAPAD survey (Beck, 2001). Figure 21b shows that, among boys aged 13 and 14, the proportion who had lifetime experience of being drunk was 15.9 % and the proportion who had experimented with cannabis was 13.8 %. Among 17- to 18-year-olds it had increased to 64.5 % and 55.7 % respectively.

In a recent EU young population survey, ‘curiosity’ was given as the main reason for trying drugs (EORG, 2002). Of those who experiment with drugs, the majority do not continue to use them on a regular basis. In a small but significant minority use escalates to intensive levels. This is illustrated in Figure 21b, which shows the distribution of cannabis use among the general population of 18-year-olds in France. General population surveys show that lifetime experience of illicit drug use is significantly higher than recent or current use (94). Comparable information on patterns of use among regular drug users is less developed than in the field of alcohol research. This limits understanding about the patterns of drug use and, consequently, the development of effective responses. Definitions of ‘problem cannabis use’ are being explored in some Member States, and it has been suggested that people who have used cannabis on 20 or more occasions during the past month are most at risk of developing a problematic pattern of use (Beck, 2001; Dutch national report). By this definition, one out of every five people in the Netherlands who have used cannabis during the past month can be classified as ‘at risk’. According to Figure 21b, in France 13.3 % of 18-year-old men, compared with only 3.6 % of 18-year-old women, fall into the ‘at-risk’ category.

Figure 21b

Level of cannabis use at the age of 17–18 in France in 2001

item Never used | item Initiator | item Occasional | item Repeated | item Regular | item Intensive


Intensive = at least 20 times in past 30 days.

Sources: ESCAPAD (2001), OFDT. National representative sample.


One major concern about experimental use of cannabis is related to the ‘gateway effect’ (95). However, the association between cannabis use and other illegal substances is complex and not reducible to a simple causal model. An alternative ‘common factor’ model demonstrates how correlations between the use of cannabis and hard drugs are due to a range of common risk factors, including vulnerability and access to drugs and propensity to use drugs. Findings from cohort studies show that illicit drug use is rarely the first sign of trouble for adolescents. Alcohol use, antisocial behaviour, truancy and crime often occur at a younger age than cannabis use (Figure 50 OL). Adolescents rarely use illicit substances without concomitant exposure to other illicit users and believe that the potential benefits of use outweigh the potential costs (Engineer et al., 2003). Evidence for the ‘gateway effect’ may be explained by cannabis bringing users into contact with an illicit market, increasing access to other illegal drugs and providing a platform of acceptability for using other illicit drugs (Grant and Dawson, 1997; Petraitis et al., 1998; Adalbjarnardottir and Rafnsson, 2002; Brook et al., 2002; Morral et al., 2002; Parker and Eggington, 2002; Pudney, 2002; Shillington and Clapp, 2002).

In the Netherlands, a youth survey conducted in 1999 found that the majority of young cannabis users purchased cannabis from friends (46 %) and from coffee shops (37 %) (De Zwart et al., 2000).

Identifying a range of risk factors that influence both the initiation and escalation of drug use in an extremely heterogeneous adolescent population is an approach that has begun to gain currency. These factors span a continuum from individual to community to macroenvironmental factors and are probably different for recreational and problem drug use.

Risk factors

Targeted surveys have shown that particular groups of young people have much higher levels of drug use than those found in the general national population. These are often young people who have been excluded from school or truanted, committed a crime, been homeless or run away from home, and those whose siblings are drug users (Lloyd, 1998; Swadi, 1999; Goulden and Sondhi, 2001; Hammersley et al., 2003). The United Kingdom youth lifestyles survey 1998/1999 found that prevalence of drug use was significantly higher among these vulnerable groups (Figure 51 OL). The size of these vulnerable groups at national level suggests that current school-based surveys are underestimating drug prevalence by failing to identify the populations of high-risk adolescents not found in the school environment. Comparable EU data on ‘vulnerable groups’ of young people at present are scarce. Young people who go out at night to particular dance music settings constitute another vulnerable group. The links between specific youth cultures and drugs are well documented, most recently in relation to the diffusion of ecstasy (MDMA) use (Griffiths et al., 1997; Springer et al., 1999). In techno dance settings, lifetime prevalence of ecstasy use ranges from 12.5 % (Athens) to 85 % (London), compared with a lifetime prevalence of 1 % (Greece) and 8 % (United Kingdom) among the general young adult populations (EMCDDA, 2002a).


In recent years increased attention has been given to social, economic and cultural determinants including physical environment (Spooner et al. 2001; Lupton et al., 2002). Drug problems are often concentrated in particular geographical areas and housing locations. For example, the Irish national report cites that children in focus groups recounted routine encounters with drug users and made casual reference to the presence of drug paraphernalia on stairs and balconies. Parents living there expressed extreme anxiety about their children’s high level of exposure to drugs (O’Higgins, 1999).

(94) See Figure 1 OL: Patterns of cannabis use among the general population – lifetime experience versus current use (last 30 days), National (drug use) prevalence survey 2001 (the Netherlands).

(95)  This is based on the hypothesis that use of cannabis per se increases risk of initiating hard drugs.