Education and debate

Linking guideline to regular feedback to increase appropriate requests for clinical tests: blood gas analysis in intensive care

Paolo Merlani, intensive care senior registrar, ^a Philippe Garnerin, quality manager, ^b Marc Diby, intensive care nurse, ^a Martine Ferring, intensive care junior registrar, ^a Bara Ricou, intensive care consultant. ^a

^a Division of Surgical Intensive Care, Department of Anaesthesiology, Pharmacology, and Surgical Intensive Care, Geneva University Hospital, 1211 Geneva 4, Switzerland, ^b Hospital Quality of Care Unit and Department of Anaesthesiology, Pharmacology, and Surgical Intensive Care, Geneva University Hospital

Correspondence to: B Ricou bara.ricou{at}hcuge.ch

Problem: Need to decrease the number of requests for arterial blood gas analysis and increase their appropriateness to reduce the amount of blood drawn from patients, the time wasted by nurses, and the related cost.
Design: Assessment of the impact of a multifaceted intervention aimed at changing requests for arterial blood gas analysis in a before and after study.
Background and setting: Twenty bed surgical intensive care unit of a tertiary university affiliated hospital, receiving 1500 patients per year.
Key measures for improvement: Number of tests per patient day, proportion of tests complying with current guideline, and safety indicators (mortality, incident rate, length of stay). Comparison of three 10 month periods corresponding to baseline, pilot (first version of the guideline), and consolidated (second version of the guideline) periods from March 1997 to August 1999.
Strategies for change: Multifaceted intervention combining a new guideline developed by a multidisciplinary group, educational sessions, and monthly feedback about adherence to the guideline and use of blood gas analysis.
Effects of change: Substantial decrease in the number of tests per patient day (from 8.2 to 4.8; P<0.0001), associated with increased adherence to the guideline (from 53% to 80%, P<0.0001). No significant variation of safety indicators.
Lessons learnt: A multifaceted intervention can substantially decrease the number of requests for arterial blood gas analysis and increase their appropriateness without affecting patient safety.