Prevalence of drug and alcohol use in urban Afghanistan: epidemiological data from the Afghanistan National Urban Drug Use Study (ANUDUS)

Lancet Glob Health. 2014 Oct;2(10):e592-600. doi: 10.1016/S2214-109X(14)70290-6.


Background: Previous attempts to assess the prevalence of drug use in Afghanistan have focused on subgroups that are not generalisable. In the Afghanistan National Urban Drug Use Study, we assessed risk factors and drug use in Afghanistan through self-report questionnaires that we validated with laboratory test confirmation using analysis of hair, urine, and saliva.

Methods: The study took place between July 13, 2010, to April 25, 2012, in 11 Afghan provinces. 2187 randomly selected households completed a survey, representing 19 025 household members. We completed surveys with the female head of the household about past and current drug use among members of their household. We also obtained hair, urine, and saliva samples from 5236 people in these households and tested them for metabolites of 13 drugs.

Findings: Of 2170 households with biological samples tested, 247 (11·4%) tested positive for any drug. Overall, opioids were the most prevalent drug in the biological samples (5·6%), although prescription drugs (prescription pain pills, sedatives, and tranquilliser) were the most commonly reported in the past 30 days in the questionnaires (7·6%). Of individuals testing positive for at least one substance, opioids accounted for more than 50% of substance use in women and children, but only a third of substances in men, who predominantly tested positive for cannabinoids. After controlling for age with direct standardisation, individual prevalence of substance use (from laboratory tests) was 7·2% (95% CI 6·1-8·3) in men and 3·1% (2·5-3·7) in women-with a national prevalence of 5·1% (4·4-5·8) and a prevalence of 5·0% (4·1-5·8) in Kabul. Concordance between laboratory test results and self-reports was high.

Interpretation: These data suggest the female head of household to be a knowledgeable informant for household substance use. They also might provide insight into new avenues for targeted behavioural interventions and prevention messages.

Publication types

  • Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

  • Adolescent
  • Adult
  • Afghanistan / epidemiology
  • Alcoholism / epidemiology
  • Female
  • Health Surveys
  • Humans
  • Male
  • Prevalence
  • Sex Factors
  • Substance Abuse Detection
  • Substance-Related Disorders / epidemiology*
  • Urban Population
  • Young Adult