Article Text

Download PDFPDF

Poster Programme
PS40 Profiles of Polydrug Use at a Local and a National Level: Risk Factors and Associations with Mental Health and Functioning
  1. JL Clark,
  2. S Frissa,
  3. S Hatch,
  4. M Hotopf
  1. Psychological Medicine, Institute of Psychiatry, King’s College London, London, UK


Background Polydrug use is an understudied phenomenon with potential implications for individual functioning and health. Its frequency within the general public and its individual costs thus need to be documented. This study’s aims were to compare the types of polydrug users in the general public at a national and a local level, and to examine how risk factors and impairments differed between them.

Methods Using comparable datasets from a local community in South East London during 2008–2010 (n=1,698) and from England in 2007 (n=7,403), latent class analysis was used to construct profiles of polydrug users. Eight binary indices reporting drug use in the last year for cannabis, amphetamine, cocaine, ecstasy, LSD, tranquilisers, crack, heroin, and two items for hazardous alcohol drinking and current cigarette smoking, were regressed on a latent variable of polydrug use with age and sex covariates. The latent class solution from each dataset was then used in a multinomial logistic regression comparing risk factors from socio-demographic, neighbourhood, social support, and mental health domains.

Results Both datasets produced three class solutions with comparable proportions: in London, 61.6% were in the “Nondrug user” group, 30.5% in the “Moderate user group” (characterised by hazardous drinking, cigarettes and some cannabis use) and 7.9% in the “High Drug User” group (characterised by drug use across all substances); in the national dataset the proportions were 57.4%, 37.9 %, and 4.6% respectively. In a logistic regression comparing both polydrug user groups to the nonuser group, both samples reported higher odds for polydrug use and common mental disorder, suicidal ideation, and functional limitation; higher levels of education, stressful life events and a never-married status were also associated with polydrug use. Differences between the local and national samples for polydrug use were found on factors such as ethnicity, social support, and employment status. Further analysis revealed no difference between the “Moderate” and “High Drug” groups in the local sample for these risk factors and mental health indicators, whereas several dose-response relationships between these groups were found nationally.

Conclusion Approximately a third of the general public both nationally and locally exhibit a pattern of moderate polydrug use associated with mental health and daily functioning impairments. Furthermore, local services and policy makers should note that higher education is a risk factor for polydrug use, and that factors related to employment and social support may be differentially linked to substance use depending on geographical residence.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.