The dominant framework for medical technology regulation and evaluation has historically been based on therapeutics, whereas early detection relies on screening and diagnostic methods. The evaluation of therapeutic and detection technologies, however, may be intrinsically different. The stages of development for drugs are more standardized, and therapeutic interventions generate direct outcomes that can be observed in patients. In contrast, most patient-level effects of screening and diagnostic tests are mediated by subsequent therapeutic decisions.
Screening and diagnostic tests also generate information that is subject to interpretation. Furthermore, this information is only one of the inputs into the decision-making process. Hence, the evaluation of detection technologies is fundamentally an assessment of the value of information. The development process for devices also tends to be iterative, and thus, assessment at early stages of development may not recognize the full potential of a new medical device. That is, most technologies that ultimately achieve widespread use go through successive stages of development, variation, and appraisal of the actual experience in the market.
With the exception of mammography, new breast cancer detection technologies have been evaluated by diagnostic studies that primarily measure sensitivity (the proportion of people with the disease who test positive) and specificity (the proportion of people without the disease who test negative). Even if the technologies ultimately are intended to be used for screening, they are generally not evaluated through screening studies that measure health outcomes.
Adoption of new detection technologies for screening purposes before assessment of their effects on clinical outcome has been common and quite problematic for technologies used to screen for other diseases because data on detection accuracy are not adequate to assess the potential value of new technologies for screening. The ideal end points for assessment of screening technologies are reductions in disease-specific mortality or morbidity, or both, but the clinical trials needed to measure those end points are quite large, lengthy, and costly.
Surrogate end points for morbidity and mortality are difficult to define because the net effect of new detection technologies could be either positive (more accurate detection, leading to lower breast cancer mortality) or negative (capable of identifying more lesions but not changing disease-specific mortality and thus leading to greater morbidity and higher screening costs).
Sharyl J. Nass, I. Craig Henderson, and Joyce C. Lashof
Committee on Technologies for the Early Detection of Breast Cancer
National Cancer Policy Board INSTITUTE OF MEDICINE and Division of Earth and Life Studies
NATIONAL RESEARCH COUNCIL