SEARCH

SEARCH BY CITATION

Keywords:

  • Primary care;
  • smoking cessation;
  • smoke free legislation

ABSTRACT

Aims  To investigate whether there were changes in the rate of prescribing of smoking cessation medications in the months leading up to, and after, the introduction of smoke-free legislation in England.

Design  Interrupted time–series analysis of prescribing rates using Autoregressive Integrated Moving Average (ARIMA) models.

Setting  A total of 350 general practices in England who contribute data to The Health Improvement Network (THIN) database.

Participants  Patients in THIN aged 16+ identified from their medical records as smokers.

Measurements  Monthly rates of prescribing of nicotine replacement therapy (NRT), bupropion and varenicline were calculated from THIN from 2000 to 2009 for all smokers and for subgroups defined by patient sex, age group, history of chronic disease and quintile of the Townsend Index of Deprivation. ARIMA models were built to assess whether there were changes in prescribing before or after the introduction of smoke-free legislation over and above any long-term and seasonal trends.

Findings  There was a 6.4% (0.7–12.1) increase in prescribing of all smoking cessation medications in the 9 months before the introduction of smoke-free legislation and a 6.4% (1.1–11.7) reduction in the 9-month period afterwards. A 6.2% (1.4–11.0) increase in NRT prescribing and a 13.2% (4.3–22.2) increase in bupropion prescribing occurred in the 6- and 3-month periods, respectively, before smoke-free legislation was introduced, and a 5.5% (2.3–8.7) decline in NRT prescribing and a 13.7% (4.6–22.8) decline in bupropion prescribing in the 9 months post-legislation. The patterns of change in prescribing did not vary with patient demographics.

Conclusions  Numbers of primary care prescriptions for smoking cessation medications increased prior to the introduction of smoke-free legislation but decreased afterwards, suggesting a temporal displacement in prescribing activity rather than a change in the overall volume of prescribing. Effects observed were consistent across all population subgroups, suggesting that the changes in prescribing will neither widen nor reduce smoking-related health inequalities.