• M Zwarenstein and W Bryant.
• Health Systems Division, Centre for Epidemiological Research in Southern Africa, Medical Research Council, Fransie van Zyl Drive, PO Box 19070, Tygerberg, South Africa, 7505. merrick.zwarenstein@mrc.ac.za
• Cochrane Db Syst Rev. 2000 Jan 1(2):CD000072.

BackgroundLack of nurse-doctor collaboration contributes to problems in quality and efficiency of patient care.ObjectivesTo assess the effects of interventions designed to improve nurse-doctor collaboration.Search StrategyWe searched the Cochrane Effective Practice and Organisation of Care Group specialised register and database of studies awaiting assessment, the Cochrane Database of Systematic Reviews, the Cochrane Controlled Trials Register, the Database of Abstracts of Reviews of Effectiveness, MEDLINE, and reference lists of articles up to the end of October 1999.Selection CriteriaRandomised trials, controlled before-and-after studies and interrupted time series of interventions to improve collaboration between nursing and medical professionals sharing patient care in primary or hospital care settings.Data Collection And AnalysisOne reviewer assessed the eligibility of potentially relevant studies, extracted data and assessed the quality of included studies; a second reviewer undertook duplicate assessments on the eligibility of some articles and data abstraction on all included studies.Main ResultsTwo trials involving 1945 people were included. One six month trial involving 1102 admissions evaluated daily, structured, team ward rounds, in which nurses, doctors and other professionals made care decisions jointly. There was shortened average length of hospital stay (LOS) from 6.06 to 5.46 days, and reduced hospital charges from US$ 8090 to 6681. There were no differences in mortality rates or the type of care to which patients were discharged. Another three month trial involving 843 admissions compared two female wards and evaluated a four times per week round. There were no significant differences between the intervention and control wards in total average length of stay for all patients (11.7 days in intervention ward versus 11.6 in the control ward). Excluding patients who died in hospital revealed shortened length of stay in the intervention ward (intervention ward 10.5 days, control ward 11.9). Mortality rates were not significantly different.Reviewer's ConclusionsIncreasing collaboration improved outcomes of importance to patients and to health care managers. These gains were moderate and affected health care processes rather than outcomes. Further research is needed to confirm these findings. The logistic challenge presented by the complexity of the interventions and the need for large sample sizes due to the likely modest impact and rarity of outcome events may best be met by multi-centre studies. Before launching such studies qualitative research is needed to identify barriers to collaboration. Interventions other than nurse-doctor ward rounds and team meetings should also be tested.

10 keywords...

hide…