Udskriv Udskriv
Switch language
Region Hovedstaden - en del af Københavns Universitetshospital

Evidence-based clinical practice

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

  1. Should antidepressants be used for major depressive disorder?

    Publikation: Bidrag til tidsskriftReviewForskningpeer review

  2. Dr. Uhre et al. Reply

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Vis graf over relationer
Evidence-based medicine combines the patient's preferences with clinical experience and the best research evidence. Randomized clinical trials are considered the most valid research design for evaluating health-care interventions. However, empirical research shows that intervention effects may be significantly overestimated (in one study on average by about 50 per cent) in randomized clinical trials with inadequate methods for generation of allocation sequence, allocation concealment, and double blinding. Further, intervention effects may be significantly overestimated if trial results are not analysed by the intention-to-treat method. Unfortunately more than 90 per cent of the randomized clinical trials have inadequate randomization, lack double blinding, or do not use intention-to-treat analysis. These irregularities may explain the considerable variation of clinical practice-within regions, single countries, and single clinics. Accordingly, there is an urgent need to improve this situation. Guidelines for Good Clinical (Research) Practice, conduct of more trials as multicentre trials, The Consort Statement, and The Cochrane Collaboration may all help in the application of the best research evidence in clinical practice. By investments in education, applied research, and The Cochrane Collaboration, evidence-based medicine may form a stronger basis for clinical practice.
Bidragets oversatte titelEvidence-based clinical practice
TidsskriftZhongguo Yixue Kexueyuan Xuebao
Udgave nummer6
Sider (fra-til)541-7
Antal sider7
StatusUdgivet - dec. 2002


  • Evidence-Based Medicine, Humans, Meta-Analysis as Topic, Randomized Controlled Trials as Topic, Research Design

ID: 38927483