There is currently little published information on the characteristics of people prescribed NRT or on NRT prescribing by GPs. This study provides information describing people prescribed NRT patch and combination regimens involving a patch by GPs in UK primary care.
The demographics of our study population were much as expected and consistent with the demographics of the overall UK smoking population  and the findings of Brose et al. . The equal distribution of men and women in the study population is consistent with the overall prevalence of smoking in the UK, where the difference in smoking prevalence between men and women has decreased considerably over the years. In 2010, the prevalence of smoking was similar among men and women with 20% of men and 19% of women reporting smoking . There were slight differences in the gender distribution across initial NRT, for example a higher proportion of males received initial NRT of patch and nasal spray (62.8%) or patch and gum (53.4%) compared to females.
The difference in the distribution of social deprivation for the study population compared with the overall THIN population might be reflective of the difference between the overall smoking population and the sub-group of people prescribed smoking cessation products. While smoking tends to be more prevalent among those in groups who work in routine and manual jobs , there may be a greater willingness for these people to seek out prescribed smoking cessation products if they qualify for free prescriptions. More affluent smokers may be more willing to access products over the counter, which have to be paid for out of pocket. The Townsend score distribution was similar across all initial NRT types.
NICE recommends three pharmacotherapies – NRT, bupropion and varenicline – as first-line therapy for smoking cessation . The choice of pharmacotherapy within the NICE guidance is recommended to be a tailored choice arising from discussion between the individual seeking help and the advising health professional. There is currently no evidence to suggest that any specific form of single NRT is more effective than another form . However, two recent network meta-analyses (NMAs) based on evidence from randomised controlled trials indicate that varenicline and combination NRT are more effective than single forms of NRT [15, 16].
Our study is an incidence study – i.e. including people who had a first NRT patch prescription and then following them through the study period. However, in the entire NRT (patch + non-patch) population (N = 128,115) identified during the study period prior to study exclusion criteria being applied, only 14.7% (N = 18,838), had a history of bupropion or varenicline prescribed as a smoking cessation treatment, leading to their exclusion. This indicates that GPs appear to usually prescribe NRT as the initial smoking cessation treatment in primary care despite the NICE guidance on smoking cessation pharmacotherapy which recommends equal first-line positioning for NRT, varenicline and bupropion.
This study uses a first NRT patch prescription as the starting place – index medication – for our study. The majority of the study population (64.3%) received NRT patch monotherapy during their first recorded smoking cessation attempt. However, the proportion starting on combination therapy increased across the duration of the study period, from 25.7% of people identified in 2008 to 44.8% of people identified in 2011. This indicates that the use of combination NRT appears to be on the increase amongst GP prescribers.
The high rates of use of NRT patch monotherapy in second (50.2%) and third episodes (45.7%) is surprising as it appears that similar NRT strategies to the first episode are being employed in subsequent attempts, and it is unclear why the strategy of use of NRT patch monotherapy would be used again in this way.
The majority of the included study population had only one recorded smoking cessation attempt. However, 39.3% (N = 50,388) of people from the total patch and non-patch NRT cohort were excluded owing to a history of NRT products and 20.2% (N = 7,868) of the included study population started a subsequent NRT episode after their first episode. Although this is an incidence study, rather than a prevalence study, this rate of exclusion suggests that, if these two groups of people are considered together, a substantial overall proportion, 45.5% (58,256/128,115), of NRT study subjects do proceed to a second episode of NRT. Furthermore, this is likely to be an underestimate of the true rate of repeat NRT use since some of the first-line NRT patch users in the analysis may have already had NRT in over the counter settings, which would not be recorded in the THIN database.
The majority of people who received non-patch NRT as a first-line NRT received either an inhalator only (50.6%) or gum only (16.8%). These deliver ‘short term’ doses of nicotine, rather than the patch which delivers nicotine over a longer period. Furthermore, 13.6% received more than one type of non-patch NRT prescribed in combination. This is unexpected as there is no recommendation in the NICE guidelines for the use of exclusive non-patch NRT combinations .
We observed an average of one year between NRT episodes with a range of 91 to 1693 days. This is a longer interval than expected. NICE guidance for smoking cessation states that “if a smoker's attempt to quit is unsuccessful using NRT, varenicline or bupropion, do not offer a repeat prescription within 6 months unless special circumstances have hampered the person's initial attempt to stop smoking, when it may be reasonable to try again sooner” .
Strengths and limitations of the research methods
The primary weakness of this study is that it was not possible to include treatment bought by people over the counter or NRT vouchers provided by smoking cessation clinics. The study may therefore overestimate the number of first-time NRT patch users identified in the analysis and underestimate the proportion of repeat NRT use in the study population. In addition, as THIN data is based on records of prescribing, therapy prescribed does not necessarily equate to therapy dispensed or take account of patient compliance with treatment.
Reasons for changes in treatment are not captured in THIN. The success of a patient’s smoking cessation attempt is unavailable. It is also worth noting that records of the NRT oral spray had a shorter amount of follow-up time available as oral sprays were not authorised until November 2010 .