Evidence-based medicine

Evidence-based medicine (EBM) is not limited to randomized controlled trials and meta-analyses. It involves the search for the best scientific basis in each case to answer clinical questions.

To learn about the accuracy of a diagnostic procedure, one needs well-conducted cross-sectional studies of patients in whom the disease sought is clinically suspected, but not necessarily a controlled trial.

For prognostic questions, one needs methodologically sound follow-up studies of patients who were enrolled in the study at a consistent, early stage of their disease (consider inclusion criteria!). Sometimes the required evidence can be found in basic experimental disciplines such as genetics or immunology. However, experimental approaches should be avoided, especially when asking about therapeutic methods, as they often lead to false-positive conclusions regarding the efficacy of interventions.

Randomized controlled clinical trials (RCTs) and especially systematic reviews of these trials are more likely to inform correctly; false conclusions are less likely. They are the "gold standard" for answering the question of whether therapeutic interventions benefit or harm.

For certain questions, controlled trials are not necessary (such as successful interventions for otherwise fatal conditions) or there is no time for clinical trials. If no results of controlled trials are available, the next best evidence (see Evidence, Level of Evidence below) must be found and considered.

1. Ben-Arye E et al (2003) Complementary medicine and psoriasis: linking the patient's outlook with evidence-based medicine. Dermatology 207: 302-307
2. Bigby M et al (2003) Appraising systematic reviews and meta-analyses. Arch Dermatol 139: 795-798
3. Bigby M (2000) Evidence-based medicine in dermatology. Dermatol Clin 18: 261-276
4. De Korte J et al (2002) The suitability of quality-of-life questionnaires for psoriasis research: a systematic literature review. Arch Dermatol 138: 1221-1227