Along with the growing number of clinical and epidemiological studies the diversity of methodological approaches increases to an extent that many physicians and their patients experience as confusing. Should we not simplify matters by limiting our view to high-quality studies of evidence class 1 and ignore all the other approaches? In fact, the generation of evidence and a scientific understanding of real-world processes would be less successful if we forewent alternative approaches to randomized studies. While it is certainly true that direct proof of therapeutic efficacy can only be accomplished with high standard randomized trials, non-randomized studies contribute substantially to exploration, to the generation of hypotheses and to the assessment of the transferability of study results to routine health care practice. Moreover, for clinical decision-making and assessment of the expected individual therapeutic benefit physicians and their patients in several regards need additional information that usually cannot be derived from randomized trials such as to the natural course of the disease without the specific treatment, safety and quality of life aspects as well as disease concepts. Should the findings obtained from non-randomized studies not likewise be incorporated into the general assessment of therapeutic benefit within a population as conducted by the German Institute for Quality and Efficiency in Health Care (IQWiG) for different therapeutic approaches? Therapeutic processes could benefit most from research if randomized trials, cohort studies and registers were integrated and interconnected. This way, assessments of transferability will have a solid empirical base.