Chronic obstructive pulmonary disease (COPD) is a lung disease which tends to gradually worsen over time, common amongst people who have smoked, and is responsible for around 5% of deaths in the United Kingdom. There are several drug treatments available to treat COPD, and this project will use the first decision to change or stop treatment as a way of comparing treatment effectiveness and safety. Our first step will be to describe how frequently the different available treatments are prescribed, following a diagnosis of COPD. We will then determine how long people remain on their COPD treatment as initially prescribed and compare this between the different treatments. A decision to change or stop treatment will largely reflect a combination of perceived treatment tolerability and effectiveness, and our comparisons will therefore serve as a way of ranking treatments according to a combination of their risks and benefits, (though we will not be able to distinguish between risks and benefits). This could help patients and prescribers to decide which treatment is likely to be the best one to start with when treating COPD.
This study aims to describe prescribing patterns and compare the time to first treatment change amongst people receiving medication for COPD (using time to treatment change as a proxy for treatment effectiveness and tolerability). We will firstly determine all patients with a valid diagnosis of COPD who start treatment for COPD at the time of, or after this diagnosis. We will then describe the different treatments people are receiving (e.g. long acting beta-agonist (LABA), long acting muscarinic antagonist (LAMA), inhaled corticosteroid (ICS), ICS/LABA combination). Finally, we will use Cox regression to determine any differences in time to treatment change, between these groups, using the group prescribed a LAMA as the baseline group as based on feasibility work we anticipate this will be the largest group. Treatment change will be defined as either the discontinuation of the initial treatment or the addition of a further COPD treatment. Variables considered for their potential confounding nature will be; age, sex, calendar year, lung function (FEV1, FEV1/FVC ratio), recent COPD exacerbations, body mass index, alcohol consumption, history of cardiovascular disease, type 2 diabetes.
Health Outcomes to be Measured:
- COPD medication discontinuation