Application of analytic hierarchy process for measuring and comparing the global performance of intensive care units

Seetharaman Hariharan*, Prasanta K. Dey, Deryk R. Chen, Harley S.L. Moseley, Areti Y. Kumar

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


Purpose: To develop a model for the global performance measurement of intensive care units (ICUs) and to apply that model to compare the services for quality improvement. Materials and Methods: Analytic hierarchy process, a multiple-attribute decision-making technique, is used in this study to evolve such a model. The steps consisted of identifying the critical success factors for the best performance of an ICU, identifying subfactors that influence the critical factors, comparing them pairwise, deriving their relative importance and ratings, and calculating the cumulative performance according to the attributes of a given ICU. Every step in the model was derived by group discussions, brainstorming, and consensus among intensivists. Results: The model was applied to 3 ICUs, 1 each in Barbados, Trinidad, and India in tertiary care teaching hospitals of similar setting. The cumulative performance rating of the Barbados ICU was 1.17 when compared with that of Trinidad and Indian ICU, which were 0.82 and 0.75, respectively, showing that the Trinidad and Indian ICUs performed 70% and 64% with respect to Barbados ICU. The model also enabled identifying specific areas where the ICUs did not perform well, which helped to improvise those areas. Conclusions: Analytic hierarchy process is a very useful model to measure the global performance of an ICU. © 2005 Elsevier Inc. All rights reserved.

Original languageEnglish
Pages (from-to)117-124
Number of pages8
JournalJournal of Critical Care
Issue number2
Publication statusPublished - Jun 2005


  • Analytic hierarchy process
  • Global performance
  • Intensive care unit


Dive into the research topics of 'Application of analytic hierarchy process for measuring and comparing the global performance of intensive care units'. Together they form a unique fingerprint.

Cite this