Management of substance abuse

The health and social effects of nonmedical cannabis use

New WHO publication on cannabis

Chapter 3. Epidemiology of cannabis use, disorders and treatment

3.1 What do we know?

3.1.1 Prevalence of cannabis use Global and regional data

Cannabis is the most commonly used illicit drug globally (UNODC, 2015; WHO, 2010). The UNODC’s World Drug Report 2015 shows the highest prevalence rates among Western Central Africa, North America, and Oceania (See Figure 3.1). The Global Burden of Disease study shows the highest age-standardized rates of use in Australasia and North America but a significant proportion of people using cannabis live in South and East Asia, followed by North America. Historically cannabis use and cultivation has been prevalent in Africa, Central Europe, South Asia and China from prehistoric times.

Figure 3.1. Annual prevalence of cannabis use for population aged 15-64 years

Source: World drug report 2015. United Nations Office on Drugs and Crime.

Today, there is still wide variation in the recorded prevalence of cannabis use within WHO regions. This in part reflects difficulties in collecting comparable data on illicit drug use. Some countries do not conduct surveys of drug use, some conduct surveys annually and others conduct them less frequently. Of those surveys that are conducted, there is variation between countries in assessing frequency of use, and age groups are divided differently or differ in the settings in which the adolescents and young adults are surveyed (e.g. schools vs. homes).

Nevertheless, there are some relatively good data on prevalence in some parts of the world. For instance, in the WHO European Region an estimated 14.6 million young Europeans (aged 15–34 years), or 11.7 % of this age group, used cannabis in the last year, with 8.8 million of these aged 15–24 years (15.2 % of this age group) (EMCDDA, 2015a). Levels of lifetime use differ considerably between countries, ranging from around one third of adults in Denmark, France and the United Kingdom, to 8% or less than 1 in 10 in Bulgaria, Romania and Turkey. Last-year use in this age group varied between 0.4 % and 22.1 %.

Herbal cannabis is becoming more popular in many EU countries – a trend driven by domestic production.

In the USA, the percentage of people aged 12 years or older who were current illicit drug users (8.4%) rose every year between 2002 and 2013 (SAMHSA, 2014). The higher percentage in 2014 than in prior years appears to reflect trends in cannabis use.

Measuring prevalence of cannabis use is difficult and it is even more difficult to measure how much cannabis, and the potency of that cannabis, is typically used by daily or less than daily users (Hall, 2015). There are no standard measures of the quantity used, and the average THC content of cannabis in most countries and regions is unknown. Epidemiological studies have used daily or near-daily cannabis use as an approximate measure of heavy use.

In high-income countries, such as the USA, cannabis use usually begins in the mid-to-late teens. Heaviest use occurs in the early twenties and declines throughout the late twenties into the early thirties. About 10% of people who use cannabis become daily users and another 20-30% use it weekly. Country examples

Most data on the patterns and health consequences of cannabis use has been collected in high-income countries in Europe, North America and Oceania. The earliest systematic study of cannabis use was the Indian Hemp Commission in 1894 and the earliest descriptions of cannabis-induced psychosis were by Chopra (Chopra, Chopra & Chopra, 1942). At the WHO expert meeting on cannabis in Stockholm in April 2015, various experts presented national and regional prevalence data. The following examples are from those countries that have data on prevalence and patterns of cannabis use that were presented at the Stockholm meeting.


The prevalence of use in the last 12 months in Brazil is 2.5% in the adult population and 3.5% in the adolescent population, a rate similar to that in other Latin American countries (UNIAD/INPAD, 2012). Lifetime use was 4.3% among adolescents and 6.8% among adults (UNIAD/INPAD, 2012). The national school survey found that 7.3% of students reported illicit drug use at least once in their lifetime (Horta et al., 2014). The highest rates of cannabis use are in young single males, adults who are unemployed, adults with a higher income, and individuals living in large cities (Jungerman et al., 2009).

Chile and neighbouring countries

Chile has one of the highest rates of cannabis use in Latin America. Until recently Chile’s prevalence was lower than Uruguay's and never exceeded 6 % of the general population in past-year use. Past-year use in the general population has increased to 11.3% (SENDA, 2015). Since 2011, the prevalence of use has increased significantly in Chile, with 30.6% of students in school surveys reporting use in the past year (Castillo-Carniglia 2014, SENDA, 2014). A similar trend has occurred in Uruguay, where cannabis use in the general population grew from 1.4% to 8.3% over 10 years. In contrast, Peru has the lowest prevalence of past year use in the region at around 1% (CICAD/OEA, 2015).

In Chile and other Latin American countries, the herbal form of cannabis is the most usual form, but a third of the Chilean cannabis market is “pressed marijuana”. This is a dried form of cannabis leaves, pressed with varying unidentified components such as glue, honey and tobacco, that comes mainly from Paraguay (SENDA, 2013).


Cannabis is referred to in Kenya as bhang and is prepared from the leaves and stems of the cannabis plant. It is commonly smoked in powder form or consumed as a beverage. It grows easily in the Mt. Kenya area and is readily available (NACADA, 2007). The NACADA rapid assessment of substance abuse in Kenya (2012) found that bhang was more common among urban residents, the unemployed and among men. Use has decreased since 2007 in the general population (6.5% ever use in 2007 reduced to 5.4% ever use in 2012) but has increased among 10-14-year-olds (from 0.3% in 2007 to 1.1% in 2012). Although urban youth and adults are more likely to use bhang, rural use has been on the rise (NACADA, 2012). Within the 15-24-year-old age category, 1.5% of individuals are currently using bhang. As in higher-income countries, cannabis use in Kenya is most common among 18-25-year-olds and falls dramatically in the mid-thirties (NACADA, 2012).


Household surveys conducted in Morocco in 2004-2005 found a past-month prevalence for cannabis use of 4% (Kadri et al., 2010). In 2013, the results of MEDSPAD (the Mediterranean School Survey Project on Alcohol and Other Drugs) showed that, among 15-17-year-old secondary school students, lifetime use of cannabis was 9.5% for boys and 2.1% for girls, and past-month use was 5.8% for boys and 0.6% for girls. Cannabis use increased with age and was consistently higher among males than females. The mean age of onset in the 2013 MEDSPAD sample was 14.9 years (El Omari & Toufiq, 2015).

South Africa

The National Survey of Youth Risk Behaviour indicates that 12.8% of South African students in grades 8-10 (13-15-year-olds) have used cannabis, and 9.2% did so in the past month (Bhana, 2015). A study of young people in grades 8-10 in the Western Cape in South Africa found lifetime use of 23.6%. The South African Stress and Health Study (SASH) surveyed over 4000 adults aged 18 years and up in a household survey and found 8.4% lifetime use. In all studies summarized by Bhana (2015), males were more likely to use than females, and urban dwellers more likely to use than their non-urban counterparts.

3.1.2 Prevalence of cannabis-use disorders

Harmful use of cannabis and dependence are the most common drug-use disorders in epidemiological surveys in Australia, Canada and the USA. Cannabis-use disorder is estimated to affect 1-2% of adults in the past year and 4-8% of adults during their lifetime (Hall & Pacula, 2010; Anthony, 2006). The risk of developing dependence among those who have ever used cannabis was estimated at 9% in the USA in the early 1990s (Anthony, 2006) compared to 32% for nicotine, 23% for heroin, 17% for cocaine, 15% for alcohol and 11% for stimulants (Anthony, Warner & Kessler, 1994).

Approximately 13.1 million people are cannabis-dependent globally (Degenhardt et al., 2013). Global prevalence of cannabis dependence in the general population is below half a percent but there is considerable variation, with higher prevalence in high-income countries where some of the more recent studies showed higher rates of 1-2% (NIH, 2015).

According to the Global Burden of Disease study (Degenhardt et al., 2013); males have higher cannabis dependence prevalence rates (0.23% [0.20-0.27%]) than females (0.14% [0.12-0.16%]), Women exhibit an accelerated progression to cannabis-use disorder after first use, and show more adverse clinical problems than men do (Cooper & Haney, 2014). Prevalence peaks in the 20-24 years age group at between 0.4% and 3.4% among males, and between 0.2% and 1.9% among females in all regions. It thereafter decreases steadily with age. There are some indications that prevalence of cannabis dependence increased worldwide between 2001 and 2010 (Degenhardt et al., 2013).

The USA is one of few countries to have collected epidemiological data on prevalence of cannabis-use disorders in a consistent manner over time. The prevalence of cannabis-use disorders increased in the USA between 1991-1992 and 2001-2002 (Compton et al., 2004) while the prevalence of cannabis use remained stable. The prevalence of cannabis use more than doubled between 2001-2002 and 2012-2013, and there was a large increase in the prevalence of cannabis-use disorders during that time. The past-year prevalence of DSM-IV cannabis use disorder was 1.5% (0.08) in 2001-2002 and 2.9% (SE, 0.13) in 2012-2013 (P < .05). While not all cannabis users experience problems, nearly 3 out of 10 cannabis users manifested a cannabis-use disorder in 2012-2013. Because the risk for cannabis-use disorders did not increase among users, the increase in prevalence of cannabis-use disorders is due to an increase in prevalence of users in the adult population of the USA. With few exceptions, increases in the prevalence of cannabis-use disorders between 2001-2002 and 2012-2013 were also statistically significant (P < .05) across demographic subgroups. (Hasin et al., 2015).

Pooled estimates suggest that the remission rate for cannabis dependence is 17% per annum (Calabria et al., 2010).

3.1.3 Treatment trends

According to WHO data, 16% of countries included in the recent ATLAS survey (Atlas 2015 in press) reported cannabis use as the main reason for people seeking substance abuse treatment. This puts cannabis second only to alcohol as a reason for treatment entry .

The number of persons seeking treatment for cannabis-use disorders and associated conditions have increased since the 1990s in many developed and some developing countries. Cannabis is now the drug of primary concern in a significant proportion of treatment episodes in the UNODC regions of Africa, Oceania, the USA and EU (UNODC, 2015). The number of cannabis users seeking help has increased over the past two decades in Australia, Europe and the USA (EMCDDA, 2015a; Roxburgh et al., 2010; WHO, 2010).

The widespread use of cannabis across the EU and the increase in the use of the drug in recent years is reflected in the high number of cannabis users now seeking treatment in Europe (Figure 3.2). In 2012, 110 000 of those enrolling in specialized drug treatment in the EU reported cannabis as the primary drug for which treatment was being sought. Cannabis is the second most commonly reported primary drug in both inpatient (18% of clients) and outpatient (26% of clients) treatment settings (EMCDDA, 2015b). For instance, in 2011 cannabis was the primary drug problem of 48% of persons entering drug treatment, and of 58% of new treatment entrants in the Netherlands (EMCDDA, 2014). It is uncertain how much increased treatment-seeking may be linked to the use of higher-THC cannabis products in, for instance, the Netherlands and the USA (Hall, 2015).

Figure 3.3. New clients entering treatment by primary drugs, 2006-2013

Source: EMCDDA (2015c). Statistical bulletin. Lisbon: European Monitoring Centre for Drugs and Drug Addiction (, accessed 16 February 2016.)

Emergency departments reported that, from 2004 to 2011, for cannabis alone or in combination with other drugs, cannabis involvement in managed cases increased substantially. Cannabis thus represents 36% of all illicit drug use that is mentioned in the USA and 31% mentioned in an urban emergency department in Switzerland (SAMHSA, 2011; Liakoni et al., 2015). In a consortium of 16 sentinel centres across Europe reporting acute drug toxicity presentations in emergency departments, cannabis ranked third among drugs after heroin and cocaine (Dines et al., 2015b). It has also been reported that cannabis is a small but increasing burden on emergency services in Australia (Kaar et al., 2015). There are indications from the USA and the EU that acute cannabis-induced physical symptoms, anxiety and sometimes psychotic symptoms are among the reasons that illicit drug users present to hospital emergency departments (Dines et al., 2015a; Liakoni et al., 2015; SAMHSA, 2009; Davis et al., 2015).

In some countries treatment uptake is also likely to have been influenced by an increased availability and diversity of treatment options for cannabis users, as well as greater recognition among service providers of the need to address problems related to the consumption of this drug.

There have also been some changes in the age structure of persons seeking treatment by primary drug. Figures 3.3 and 3.4, which are based on data from 26 European countries, show the age structure of clients entering treatment by primary drug in 2006 and 2013.

Figure 3.3. Age structure of clients entering treatment by primary drug, 2006


Figure 3.4. Age structure of clients entering treatment by primary drug, 2013

Source: EMCDDA (2015a), European drug report 2015; and EMCDDA (2015c), Statistical bulletin. Lisbon: European Monitoring Centre for Drugs and Drug Addiction (available at, accessed 16 Februzary 2016).

The adverse health and social consequences of cannabis use reported by cannabis users who seek treatment for dependence appear to be less severe than those reported by persons dependent on alcohol or opioid (Hall & Pacula, 2010; Degenhardt & Hall, 2012). However, rates of recovery from cannabis dependence among those seeking treatment are similar to those treated for alcohol dependence (Florez-Salamanca et al., 2013).

3.1.4 Areas that require more research

·       Global data are required on the frequency of cannabis use (more than once daily, daily, near daily, weekly, etc.) and the prevalence of health and social consequences.

·       Data are also required on the typical doses of THC and other cannabinoids (e.g. cannabidiol or CBD) that users receive through different modes of use (smoked, vaporized, ingested). There are limited data on cannabis potency trends over time and their impact on health (e.g. cognition, psychosis, accidents, motivation, emergency department mentions, cannabis-use disorders).

·       Most epidemiological research on cannabis has focused on smokers in a small number of high-income countries. More research is needed on cannabis use in low- and middle-income countries.

·       Global assessments are needed of the relationship between cannabis use and the use of other drugs.

·       Cannabis and tobacco are often mixed together and there is a need for more data from well-designed studies on the prevalence and health consequences of:

o   smoked cannabis only;

o   different routes of cannabis administration;

o   potential added health risks from the use of cannabis in combination with tobacco;

o   the THC and other cannabis preparations, including pressed marijuana, in different parts of the world (e.g. in Latin American countries, and at different periods of time).

·       Most of the studies on risk and protective factors for cannabis use have been conducted in a limited number of high-income countries. There is some uncertainty as to whether the same risk factors apply in low- and middle-income countries.

·       Global data are lacking on trends in the prevalence of harmful cannabis use and of cannabis dependence (cannabis-use disorders).