TargetCOPD: A pragmatic randomised controlled trial of targeted case finding for COPD <i>versus</i> routine practice in primary care

Conference: 
SAPC ASM 2015 - Oxford - Evidence and innovation in primary care
Talk Code: 
4B.1

The problem

Many people with clinically significant chronic obstructive pulmonary disease (COPD) remain undiagnosed worldwide. Small uncontrolled studies have examined methods of case finding, but no large trials have adequately assessed the most effective approach in primary care.

The approach

Using a pragmatic cluster randomised controlled trial (RCT) design among general practices in the West Midlands, UK, we investigated the effectiveness of a Targeted approach to case finding compared with routine practice over a 12-month period. With an individual patient RCT nested in the Targeted arm, we also compared effectiveness of Active case finding using a combined approach with postal and opportunistic questionnaires during GP consultations, and opportunistic-only case finding. Ever-smoking patients aged 40-79 years, without a diagnosis of COPD were eligible. Patients in the Targeted arm reporting indicative respiratory symptoms (chronic cough/phlegm, wheeze or dyspnoea) were invited for post-bronchodilator spirometry, conducted to ATS standards by trained research assistants using an Easy One spirometer. COPD was defined as FEV1/FVC<0.7 among patient with symptoms, in line with the current UK guidelines, to ensure comparability across arms.The primary outcome was new cases of COPD (diagnosed through the trial or usual GP processes). A multilevel logistic regression model was used to model the probability of detecting a new case of COPD for each treatment arm, with clustering of patients (by practice and household) accounted for using a multi-level structure.

Findings

54 general practices (with 74,818 eligible patients) were randomised and completed the trial. 1278 (3.9% of patients eligible) were newly diagnosed in the Targeted arm compared with 337 (0.8%) in the routine care arm, adjusted OR 7.45 (95%CI 4.80, 11.55). Among patients in the Targeted arm, 15,393 were further randomised to opportunistic case finding and 15,394 to active case finding. >90% patients consulted their GP during the year. Response rates to the questionnaire were lower in the opportunistic (12.9%) than the active arm (37.7%) but responders reported a similar proportion of respiratory symptoms (54.7% and 56.4% respectively). Overall, the active arm was more effective in identifying new cases (n=822, 5.4%) than the opportunistic arm (n=370, 2.4%) (adjusted OR 2.34 (2.06, 2.66).

Consequences

A Targeted case-finding programme is significantly more effective in identifying new cases of COPD than routine care. An approach combining postal with opportunistic questionnaires, followed by spirometry among those reporting relevant symptoms is more than twice as effective as an opportunistic-only approach. But before implementation it is essential to examine its cost-effectiveness. Trial registration: Current controlled trials ISRCTN14930255

Credits

  • Rachel Jordan, University of Keele, Keele, UK
  • Peymane Adab, University of Keele, Keele, UK
  • Alice Sitch, University of Keele, Keele, UK
  • Alexandra Enocson, University of Keele, Keele, UK
  • Deirdre Blissett, University of Keele, Keele, UK
  • Sue Jowett, University of Keele, Keele, UK
  • Jen Marsh, University of Keele, Keele, UK
  • Richard Riley, Queen Elizabeth Hospital Birmingham, Birmingham, UK
  • Martin Miller, University of Keele, Keele, UK
  • Brendan Cooper
  • Alice Turner, University of Keele, Keele, UK
  • Jon Ayres, University of Keele, Keele, UK
  • KK Cheng, University of Keele, Keele, UK
  • Kate Jolly, University of Keele, Keele, UK
  • Rob Stockley
  • Sheila Greenfield, University of Keele, Keele, UK
  • Stan Siebert, University of Keele, Keele, UK
  • Amanda Daley, University of Keele, Keele, UK
  • David Fitzmaurice, University of Keele, Keele, UK