H J Stoevelaar1, J McDonnell, H Stals, L Smets. 1. Institute for Health Care Policy and Management, Erasmus University Medical Center Rotterdam, The Netherlands, stoevelaar@bmg.eur.nl
Abstract
OBJECTIVE: To refine the appropriate indications for gastro-protective treatment in patients using non-steroidal anti-inflammatory drugs (NSAIDs). METHODS: Using the RAND Appropriateness Method, a multidisciplinary expert panel was asked to rate the appropriateness of gastro-protection for 4608 different cases. Logistic regression was used to construct a decision framework to identify patients for whom gastro-protective medication should be considered. RESULTS: Complete consensus existed on the appropriate use of gastro-protection in patients with a history of (un)complicated peptic ulcer. For other patients. agreement was found for 39% over 1536 cases. Logistic regression demonstrated strong consistency of the panel ratings (Hosmer Lemeshow coefficient 0.92), allowing the development of a comprehensive decision support model. CONCLUSIONS: Using the RAND Appropriateness Method, we were able to develop a clear and internally consistent decision framework for the appropriate use of gastro-protection in patients taking NSAIDs. The validity of this model should be tested in further studies and practice.
OBJECTIVE: To refine the appropriate indications for gastro-protective treatment in patients using non-steroidal anti-inflammatory drugs (NSAIDs). METHODS: Using the RAND Appropriateness Method, a multidisciplinary expert panel was asked to rate the appropriateness of gastro-protection for 4608 different cases. Logistic regression was used to construct a decision framework to identify patients for whom gastro-protective medication should be considered. RESULTS: Complete consensus existed on the appropriate use of gastro-protection in patients with a history of (un)complicated peptic ulcer. For other patients. agreement was found for 39% over 1536 cases. Logistic regression demonstrated strong consistency of the panel ratings (Hosmer Lemeshow coefficient 0.92), allowing the development of a comprehensive decision support model. CONCLUSIONS: Using the RAND Appropriateness Method, we were able to develop a clear and internally consistent decision framework for the appropriate use of gastro-protection in patients taking NSAIDs. The validity of this model should be tested in further studies and practice.
Authors: Carol Tan; Tom Treasure; John Browne; Martin Utley; Christopher W H Davies; Harry Hemingway Journal: J R Soc Med Date: 2007-01 Impact factor: 5.344