Article Text

Download PDFPDF

P25 Inequalities in adolescent smoking in the UK millennium cohort study: estimating the relative contributions of verbal ability and self-regulation
  1. H Creese,
  2. R Viner,
  3. S Hope,
  4. A Pearce,
  5. A Ross
  1. Population, Policy and Practice, University College London, London, UK


Smoking is a leading cause of morbidity and preventable death in the UK. Adolescence is a time when smoking behaviours are often initiated, therefore it is important to understand potential risks and protective factors. Higher cognitive development in adolescence is related to a lower likelihood of smoking. Cognitive development is characterised by the growth of abilities and skills in multiple domains, including verbal ability. Similarly, higher ‘non-cognitive’ ability has been associated with lower levels of smoking. One element of ‘non-cognitive’ ability is self-regulation which refers to an individual’s control of thoughts, emotions and behaviour in order to achieve a goal. Socioeconomic disadvantage is associated with higher rates of initiation and progression to daily smoking. In addition, verbal ability and self-regulation are generally lower in socioeconomically disadvantaged children. We investigated whether the relationship between early socioeconomic circumstances and adolescent smoking was partially driven by cognitive development and/or self-regulation. Using an Oaxaca-Blinder decomposition model, we estimated the relative contributions of verbal ability and/or self-regulation at 11 years to the association between early life socioeconomic disadvantage captured by household poverty status at 9 months and smoking in adolescence (at 14 years) in the UK Millennium Cohort Study (N=6,737). Verbal ability was assessed using a validated, age-appropriate test, the British Ability Scales Second Edition (BAS 2). Five parent-rated items from the Strengths and Difficulties Questionnaire were used to capture two components of self-regulation – task attentiveness and emotional regulation. Children who had lived in poverty at 9 months were almost twice as likely to smoke at age 14. Twelve per cent of the difference in smoking between poverty groups was attributable to self-regulation, with none explained by verbal ability. The remaining difference was largely due to socio-demographic factors. Interventions to improve self-regulation targeted at children from families experiencing socioeconomic disadvantage at the end of primary school may hold promise to reduce inequalities in smoking. However, the detrimental impact of early life poverty on smoking in adolescence is chiefly driven by socio-demographic factors rather than later, individual characteristics such as self-regulation.

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.