In the general population, cigarette smoking leads to loss of weight while quitting smoking often results in gaining it back. This may ameliorate some of the health benefits of quitting in these people. On the contrary, in certain patients with chronic obstructive pulmonary disease (COPD), gaining weight is associated with a more positive outcome. However, it is unclear how cigarette smoking behaviour relates to changes in weight in people with COPD. Weight change may be related to the effects of smoking or to the disease itself (or both).
Smoking may have a different relationship to a change in body-mass index (BMI) in patients with and without COPD. One previous study showed that weight changes were similar in patients with lung disease and smokers with normal lung function, indicating that it is smoking that mainly affects weight. However, this study included people with a narrow age range (70-79) and they had better overall health than typically found in "real world" settings. Importantly, they did not consider whether COPD patients smoked or not. Another study also did not examine smoking behaviour in COPD patients but found that changes in fat-free mass and fat mass were small and comparable between COPD patients, and smoking and non-smoking patients without COPD.
Therefore, in a "real world" setting, we aim to determine the relationship between smoking and BMI change. The knowledge that smoking affects not only the aetiology of COPD but also BMI may be important for patients and physicians when determining treatment options and prognosis.
The main objective of this study is to assess the change in body-mass index (BMI) in chronic obstructive pulmonary disease (COPD) and non-COPD (i.e. controls) cohorts by current, former, and never smoking status measured in a time-varying manner.
One well-conducted study investigated this topic but had poor generalisability and was underpowered to examine the effect of smoking status among COPD patients. Another study also could not examine the effect of smoking status in COPD patients and had only 3-years of follow-up. Neither study examined BMI, which is a measure of body composition that is more accessible in primary care settings. Importantly, none of the previous studies assessed smoking in a time-varying manner. A time-dependent analysis of smoking on BMI will account for changes in smoking behaviour over time, which has not been explored previously.
A cohort of patients aged ?40 years with a recorded Read code (in CPRD) for COPD during the period of valid data collection (from 1 January 2005 to present) and a second cohort of controls without a COPD Read code will be followed from their index date until the date of death, date of transfer out of the practice, the end of data collection, or outcome of interest, whichever comes first. Both cohorts will be grouped into current smoking, former smoking, and never smoking groups over 90-day periods in order to examine the effect of smoking status - measured time-dependently - on clinically relevant change in BMI.
Time-to-first event of a greater than or equal to a clinically relevant 5% decrease or 5% increase in BMI from baseline are the primary outcomes of interest. A 5% increase and a 5% decrease in BMI will be separate outcomes with separate hazard ratios (HR) in each of the study cohorts (i.e. four estimates total). Patients with a change in BMI <5%, and patients without a second BMI measurement will be censored at the earliest of the date of death, date of transfer out of the practice, and the end of data collection.
Frank de Vries - Chief Investigator - Utrecht University
Frank de Vries - Corresponding Applicant - Utrecht University
Frits Franssen - Collaborator - CIRO
Johanna Driessen - Collaborator - Utrecht University
Patrick Souverein - Collaborator - Utrecht University
Peymane Adab - Collaborator - University of Birmingham
Rachel Jordan - Collaborator - University of Birmingham
Spencer Keene - Collaborator - Utrecht University