Use of blood eosinophil count to predict inhaled steroid responsiveness in patients with COPD using primary care health records

Date of ISAC Approval: 
01/06/2016
Lay Summary: 
Chronic obstructive pulmonary disease (COPD) is a long-term respiratory condition which affects over one million people in the UK, usually as a result of smoking and causes cough and breathlessness which generally worsen with time. Steroid inhalers are commonly prescribed, but there is uncertainty over how beneficial they are to all patients living with COPD, and steroid inhalers are expensive and have been associated with a range of adverse effects including an increased risk of pneumonia. Some patients with COPD have higher levels of a type of cell called eosinophils in their blood and lungs suggesting greater inflammation in their lungs and thus greater chance of benefit from steroid inhalers. However, this hasn't been investigated yet in general practice using results of blood levels of eosinophils that have been routinely tested for other reasons. In this study, we will use general practice records to evaluate whether patients who have higher levels of blood eosinophils on routinely taken blood tests before starting treatment with inhaled steroids are more likely to have fewer exacerbations and less deterioration on treatment compared to patients with lower levels of blood eosinophils.
Technical Summary: 
This study will use a new user cohort design to assess outcomes for a group of COPD patients who have not previously been prescribed an inhaled corticosteroid (ICS-naive) and who then commence treatment with a new inhaled maintenance treatment (the index date). The descriptive component of the study will examine records of routine blood eosinophil testing in general practice and their values prior to starting a new inhaled maintenance treatment, their relationship to other baseline variables, and their stability over time. The hypothesis-testing component will test whether baseline blood eosinophil values predict disease outcomes, and outcomes under treatment (ICS responsiveness). The primary outcome will be acute exacerbations of COPD. Patients will be divided into two groups based on whether the new inhaled maintenance medication is an ICS or long-acting bronchodilator. Disease outcomes over time will be compared between those patients starting treatment with ICS and those starting treatment with a long-acting bronchodilator, with adjustment for baseline characteristics, and this will be stratified by baseline blood eosinophil values to determine whether this modifies effectiveness of treatment.
Health Outcomes to be Measured: 
Primary outcome: exacerbations of COPD (time-to-first exacerbation, as well as sensitivity analyses on number/year, and 2 exacerbations/year) Secondary related outcomes - Pneumonia, death (ONS), hospitalisation (HES), rate of decline in % predicted FEV1 per year.
Collaborators: 

Dr Helen Ashdown - Chief Investigator - University Of Oxford
Christopher Butler - Collaborator - University Of Oxford
Dr Emily McFadden - Collaborator - University Of Oxford
Dr Helen Ashdown - Corresponding Applicant - University Of Oxford
Ian Pavord - Collaborator - University Of Oxford
Ms Margaret Smith - Collaborator - University Of Oxford
Professor Mike Thomas - Collaborator - University Of Southampton
Dr Mona Bafadhel - Collaborator - University Of Leicester

Linkages: 
HES Admitted;ONS;Patient IMD;Practice IMD (Standard)