COPD in England: a comparison of expected, model-based prevalence and observed prevalence from general practice data

J Public Health (Oxf). 2011 Mar;33(1):108-16. doi: 10.1093/pubmed/fdq031. Epub 2010 Jun 3.


Background: Primary care data show that 765 000 people in England have a general practice (GP) diagnosis of chronic obstructive pulmonary disease (COPD). We hypothesized that this underestimates actual prevalence, and compared expected prevalence of COPD for English local authority areas with prevalence of diagnosed COPD.

Methods: Cross-sectional comparison of GP observed and model-based prevalence estimates (using spirometry data without clinical diagnosis) from the Health Survey for England. Local underdiagnosis of COPD was estimated as the ratio of observed to expected cases. We investigated geographical patterns using classical and geographically weighted regression analysis.

Results: Both observed and expected prevalence of COPD varied widely between areas. There was evidence of a 'north-south' divide, with both observed and modelled prevalence higher in the north. The ratio of diagnosed to expected prevalence varied from 0.20 to 0.95, with a mean of 0.52. Underdiagnosis was more pronounced in urban areas, and is particularly severe in London. The inclusion of GP numbers in the analysis yielded a stronger regression relationship, suggesting primary care supply affects diagnosis.

Conclusion: Both observed and modelled COPD prevalence varies considerably across England. Cost-effective case-finding strategies should be evaluated, especially in areas where the ratio of observed to expected cases is low.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • England / epidemiology
  • Epidemiologic Methods
  • Female
  • General Practitioners / statistics & numerical data*
  • Geography
  • Health Status Indicators
  • Humans
  • Male
  • Middle Aged
  • Models, Theoretical
  • Prevalence
  • Pulmonary Disease, Chronic Obstructive / epidemiology*
  • Regression Analysis
  • Risk Factors
  • Spirometry