Intended for healthcare professionals

Research Article

Intensive Care Society's APACHE II study in Britain and Ireland--II: Outcome comparisons of intensive care units after adjustment for case mix by the American APACHE II method.

British Medical Journal 1993; 307 doi: (Published 16 October 1993) Cite this as: British Medical Journal 1993;307:977
  1. K M Rowan,
  2. J H Kerr,
  3. E Major,
  4. K McPherson,
  5. A Short,
  6. M P Vessey
  1. Department of Public Health and Primary Care, University of Oxford, Radcliffe Infirmary.


    OBJECTIVES--To compare outcome between intensive care units in Britain and Ireland both before and after adjustment for case mix with the American APACHE II method and to validate the American APACHE II method in Britain and Ireland. DESIGN--Prospective, cohort study of consecutive admissions to intensive care units. SETTING--26 general intensive care units in Britain and Ireland. SUBJECTS--8796 admissions to the study intensive care units. MAIN OUTCOME MEASURE--Death or survival at discharge from intensive care unit and hospital. RESULTS--At discharge from both intensive care unit and hospital there was a greater than twofold variation in crude mortality between the 26 units. After adjustment for case mix, variations in mortality were still apparent. For four intensive care units the observed numbers of deaths were significantly different from the number predicted by the American APACHE II equation. The overall goodness of fit, or predictive ability, of the APACHE II equation for the British and Irish data was good, being only slightly inferior to that obtained when the equation was tested on the data from which it had been derived. When patients were grouped by various factors such as age and diagnosis, the equation did not adjust across the subgroups in a uniform manner. CONCLUSIONS--The American APACHE II equation did not fit the British and Irish data. Use of the American equation could be of advantage or disadvantage to individual intensive care units, depending on the mix of patients treated.