One would expect that the priority of the Canadian Task Force on Preventive Health Care on breast-cancer screening would be to recommend practices that would maximally reduce mortality and morbidity from disease. In the case of breast cancer, this suggests focusing on effective guidelines for those who are at increased probability of developing disease or increased risk of late diagnosis and the consequent higher likelihood of premature death or morbidity associated with treating advanced disease.
But previous guidelines in 2001, 2011 and 2018 have not addressed those at high risk of death or morbidity. They have also ignored women who are at increased risk of incidence of breast cancer due to elevated breast density and those in certain sociodemographic groups for whom outcomes tend to be less favourable than average. Instead, recommendations have mainly focused on women considered to be at “average risk” (although, inexplicably, those with dense breasts appear to have been included in this group without any special considerations).
Selection of evidence
While the randomized controlled trial (RCT) is the gold standard of evidence of efficacy of an intervention, such trials do not usually assess effectiveness. RCTs are evaluated by “intention to treat,” not according to whether an individual in the trial actually received the intervention. So, the lifesaving potential of an intervention that was inherently efficacious would be underestimated if an appreciable number of people in the intervention arm of the trial refused the intervention or if one who had been randomized to the control (non-intervention) group sought and received the procedure outside the trial. This effect could be quite substantial.
In addition, in the case of breast-cancer screening, the available data from RCTs are old, reflecting very different methods of cancer detection and treatment than those used today. While more up-to-date evidence coming from other (non-randomized) studies existed, the task force excluded these in previous evidence reviews. It also did not make use of data from sophisticated computer modelling of the natural history of breast cancer, its detection and treatment. These models are helpful in interpolating gaps between existing empirical data.
Harms vs. benefits
The Task Force emphasized the importance of balancing benefits and harms while considering individuals’ values in making a decision to participate in screening. On the surface, this is a reasonable and laudable position. It is emphasized that a “shared decision” process be used. This term is in itself somewhat misleading in that ultimately the decision regarding screening is for a woman herself to make, based in part on information received from a reliable source, who may be her primary-care provider (PCP). In practice, this has not been clearly understood by some PCPs, who often attempt to dissuade women from screening or outright refuse to provide women with a requisition for screening. This behaviour has been aggravated and facilitated by messaging from the task force that tends to overemphasize the harms and to play down the benefits of screening without ever having presented a meaningful analysis of the balance between benefits and harms.