Observationella kohortstudier. Översikt och några exempel från området läkemedelssäkerhet
Sammanfattning
Clinical studies are often categorized as either descriptive, instrumental for generating hypotheses, or analytic, aiming to test such hypotheses. When working with treatment-safety there are no study designs that are always ”better” or ”lesser” suited than others. The choice of study design should be based on the research question at hand.
The aim of any clinical study, regardless of design, is that they have a high internal validity (void of systematic errors leading to over- or underestimations of the true association between exposure and outcome, that there are no confounders of the true association, and that the precision is good) and external validity (the association is true also for the groups of patients in whom the results are meant to be applied).
Threats to internal and external validity differ by study design. Every study – regardless of design –must always be criticized based on its individual strengths and weaknesses. To do this, both subject matter and methodological knowledge are needed.