Is there a risk that the chosen end points, cut-off values or comparators/controls may give a biased description of the results of the technology?
Clinical effectiveness should ideally be directly related to the disease under treatment. This is not always fully possible so other endpoints may have to be used (e.g. surrogate markers for preventing a life-threatening disease). In addition, the technology may have several aims (e.g. those related to treating the disease and preventing secondary morbidity).
The choice of cut-off values for sensitivity and specificity should be done considering the moral value of different results – for example, high specificity is required if false positives have serious consequences.
Other domains:, safety, effectiveness. Expert opinion, Stakeholder hearing
See methodological description in EFF and SAF