Evidence evaluation in ranking target areas of interest for a national patient safety campaign

Article type
Authors
Lauvrak V1, Norderhaug I1, Saunes I2, Krogstad U2
1Norwegian Knowledge Centre for the Health Services (NOKC), Department of Evidence-Based Practice, Specialist Health Care Unit, Norway
2NOKC, Department of Quality Measurements and Patient Safety, Secretariat for the Norwegian Patient Safety Campaign, Norway
Abstract
Background: In 2009, the Norwegian Knowledge Centre for the Health Services (NOKC) was commissioned by the Ministry of Health and Care Services to prepare a national patient safety campaign. A broad group of health care experts was established to use a formalized consensus process to rank eleven possible target areas according the following set of criteria: 1) Potential for improvement in Norway; 2) Interventions documented as effective; 3) Access to data and methods to evaluate the effect of interventions; and 4) Support among health care professionals in Norway. Target areas were: safe surgery, postoperative infections, urinary tract infections, CVK-infections, medication errors and harm, stroke, waiting time cancer patients, patient falls, decubitus ulcer, early rehabilitation hip fracture, and rapid response team heart failure.

Objectives: Provide information on the effect of specified interventions.

Methods: For one area, a systematic review of primary studies was prepared. For eight areas, short reports, based on systematic reviews identified in The Cochrane Library, the CRD - databases and by hand searches, were prepared. The most updated (2005-2010) high quality systematic review for each predefined intervention was selected. The level of evidence for predefined patient safety relevant outcomes was graded according to the GRADE approach. Methods and results were communicated written and orally to the expert group who scored each area against all four predefined criteria.

Results: Most interventions were supported by low or very low quality of evidence. No intervention was rated high quality, and only a few were rated moderate quality. For some interventions, relevant scientific evidence was not found. Nevertheless, the consensus process resulted in ranking four areas (safe surgery, postoperative infections, medication errors and harm, and urinary tract infections) as appropriate for the campaign.

Conclusion: Uncertainty of effectiveness did not in itself disqualify an area from being ranked as appropriate for the campaign, but it was transparently acknowledged together with the other three assessment criteria.