Development of quality indicators for diagnosis and treatment of patients with non-small cell lung cancer:A first step toward implementing a multidisciplinary, evidence-based guideline

Background: While developing and distributing clinical practice guidelines are important in optimising clinical healthcare, insight into actual care is necessary to achieve successful implementation. Developing quality indicators may be the first step to becoming aware of actual care. The Dutch national practice guideline Non-small cell lung cancer: staging and treatment is one of the first clinical, multidisciplinary guidelines for oncology in the Netherlands for which quality indicators were developed systematically. We describe indicator development based on this guideline as a practical experience.Methods: To develop a set of indicators for diagnosis and treatment of patients with non-small cell lung cancer, we systematically achieved consensus on the basis of a national, multidisciplinary, evidence-based guideline and the opinions of professionals and patients. After the researchers extracted the recommendations from the guideline, we carried out a so-called Rand-modified-Delphi procedure. This consisted of three rounds: a national panel. of professionals and representatives of the national patient organization scored all recommendations, the professionals had a consensus meeting, and the final set of indicators was e-mailed for a last check. Subsequently, some clinimetric characteristics of this final set were assessed in a practice test.Results: Thirty-two of 83 recommendations were selected in the first round. After the consensus meeting, 8 recommendations met the final criteria and were incorporated into 15 indicators, which were tested in practice. The most successful indicators for quality improvement are indicators that are measurable, have potential for improvement, have a broad range between practices and are applicable to a large part of the population.Conclusions: For successful implementation of evidence-based guidelines, each new guideline should be developed and tested with a set of indicators based on the guideline. The procedure we describe can serve as an example for other new guidelines. (C) 2006 Elsevier Ireland Ltd. All rights reserved.