[Identification of emergencies for asthma and chronic obstructive pulmonary disease in the emergency rooms of a hospital. A reliability study]

Gac Sanit. 1999 Jan-Feb;13(1):30-7. doi: 10.1016/s0213-9111(99)71319-1.
[Article in Spanish]

Abstract

Objective: To analyse the reliability of the obtained results in the identification of cases of asthma and chronic obstructive pulmonary disease (COPD) diagnosed in the medical records of the emergency rooms of the <<Hospital Clínico Universitario>> of Valencia, Spain.

Methods: In the context of a study on the relationship between emergency rooms for asthma and COPD and specific environmental risks, we revised all the emergency rooms medical records of people aged over 14 attended at the <<Hospital Clínico Universitario>> of Valencia, Spain, during 1993 and 1994. The observers were two nurses who received previous training in the identification of cases. One observer revised the year 1993 and the other one 1994. To evaluate the reliability of the results we obtained a 24 days sample for each year and we estimated the inter-observer agreement using the Proportion of Observed Agreement (POA) and the Kappa Index (KI) and the 95 % confidence intervals (95% CI). We also carried out a specific analysis for each study category estimating the specific Kappa Index and the Proportions of Specific Agreement (PSA). To assess the inter-observer agreement we used the first 59 days of 1994, to estimate the same indexes mentioned above.

Results: In the total of emergency rooms, the POA for both observers was 0.99 and the KI was 0.91 (95% CI: 0.88-0.94) for the first observer and 0.79 (0.76-0.82) for the second one. In the specific analysis the PSA for <<asthma>> versus <<not-asthma>> were 0.92 (0.75-1) and 0.48 (0.31-0. 66) respectively. For the analysis of <<COPD>> versus <<not-COPD>> the PSA were 0.78 (0.69-0.87) and 0.75 (0.64-0.86). In the inter-observer agreement study we obtained, for the overall analysis, a POA equal to 0.99 and a KI to 0.87 (0.86-0.88). In the specific study the PSA for <<asthma>> versus <<not-asthma>> were 0.64 (0.53-0. 74) and 0.82 (0.77-0.87) for <<COPD>> versus <<not-COPD>>.

Conclusions: The consistency of the results is good for both observers, but better for the first observer, particularly for asthma. For the COPD category the results are more homogeneous, showing a good concordance for both observers. The results for the inter-observer study also show a good reliability.

Publication types

  • English Abstract
  • Research Support, Non-U.S. Gov't
  • Review

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Asthma / epidemiology*
  • Emergencies / epidemiology*
  • Emergency Service, Hospital / statistics & numerical data*
  • Female
  • Hospital Records
  • Hospitals, University / statistics & numerical data*
  • Humans
  • Lung Diseases, Obstructive / epidemiology*
  • Male
  • Middle Aged
  • Observer Variation
  • Reproducibility of Results
  • Retrospective Studies
  • Spain / epidemiology