Statistics from Altmetric.com
Recruitment of patients into clinical trials is challenging and many publicly funded trials in the UK miss their targets.1 Many studies investigating COPD may wish to recruit from patient groups or primary care; however, in contrast to the low rates reported by recent large interventional trials (eg, 7%2), evidence suggests that patients on UK general practice (GP) databases do not fulfil diagnostic criteria for COPD when retested.3 Optimal strategies for recruiting these patients have not been adequately explored.
Three approaches were used during recruitment of subjects to a recent non-commercial trial of oral antibiotics in stable COPD.
Local GP surgeries wrote to patients on their COPD register, and interested patients replied directly to the study team using preaddressed reply slips.
Similarly, local pulmonary rehabilitation (PR) groups wrote to COPD patients on their database.
Study team members gave educational talks to local PR and patient support groups, and at the end described current research plans. Interested patients approached the study team directly.
Centres were reimbursed for sending letters according to standard research tariffs.
Interested patients were contacted and a screening visit arranged which included a full medical history and postbronchodilator spirometry. Patients were considered eligible if COPD was confirmed with FEV1 <80% predicted and FEV1 to FVC ratio <0.7.
The efficiency of identifying eligible patients was assessed by comparing the reply and screening failure rates. External time and financial costs were estimated per eligible patient identified, excluding researcher salaries. Patient characteristics were compared by recruitment source.
Between January 2012 and May 2013, 37 GP surgeries and four PR groups sent letters to 2300 and 469 patients, respectively. Reply rates were similar from both sources (21% and 22%) and 156 (7%) and 37 (8%) of these patients attended screening. The educational talks identified 53 patients, of whom 23 (43%) were screened. Approaches 1 and 2 had screen failure rates of 35% and 19%, while approach 3 had the lowest (13%). Figure 1 summarises these pathways and gives information on the estimated time and financial commitments per eligible patient identified.
Eligible patients recruited from GP surgeries had significantly milder disease than those from PR and patient groups (mean (SD) FEV1 63% predicted (20) vs 53% (18, p=0.003), FEV1 to FVC ratio 0.55 (0.11) vs 0.50 (0.11, p=0.007), exacerbations in the previous year 1.7 (2.0) vs 2.7 (3.4, p=0.029)), as well as lesser smoking history and fewer medication prescriptions. Further information is detailed in the online supplement.
We found that recruiting patients from GP COPD databases was less efficient, more costly, and identified patients with milder disease than through PR and patient groups. However, the pool of patients accessible via GPs was larger and more representative of the wider population. Recruitment via PR and patient groups may therefore target an enriched population useful for smaller studies, and researchers planning future studies of COPD should prioritise resources accordingly.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online supplement
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.