According to systematic reviews, the main benefit of mammography screening is to reduce breast cancer mortality[2,3,6,7]. This benefit increases with age.

For every 1000 women aged 40 to 49 who perform mammograms every two years over ten years, one avoids dying from breast cancer. For every 1000 women aged 50 to 69 who perform mammograms every two years over 20 years, 4 avoid dying from breast cancer. [26 ,27]. Another way to express these probabilities: for every 10000 women aged 39 to 49 who perform mammograms every two years over ten years, four avoid dying from breast cancer; for every 10000 women aged 50 to 59 who perform mammograms every two years over ten years, five to eight avoid dying from breast cancer; for every 10000 women aged 60 to 69 who perform mammograms every two years over ten years, 12 to 21 avoid dying from breast cancer[33].

Detecting breast cancer at earlier stages allows for more conservative surgeries [30].

Receiving true negative results from mammographic screening brings a sense of reassurance and satisfaction to women[15].

False negatives, that can give a false sense of reassurance [34].

False positives, that can result in additional testing until the abnormal finding is refuted, leading to unnecessary exams and anxiety in some women [25 ,35].

Overdiagnosis: Breast cancer screening leads to the detection of indolent cancers that would not have caused harm if left undetected [36 ,37].

For every 1000 women aged 40 to 49 who perform mammograms every two years over ten years, 239 will have a false-positive result and seven an overdiagnosis. For every 1000 women aged 50 to 69 who perform mammograms every two years over 20 years, 412 will have a false-positive result and 19 an overdiagnosis.[26 ,27] Estimates on overdiagnosis vary depending on the type of study, measures and methods used, and there is no consensus about the appropriate approach. Data based on trials with long term follow-up found a 22% overdiagnosis rate for invasive cancer for the combined age groups[33].

Women had limited awareness about the harms of screening. This information triggered different emotions such as surprise, concern and mistrust.

Some women consider it appropriate to inform women about the harms of screening, but others fear it could deter women from participating in screening.

Women value better the possibility of an early diagnosis over the risk of false positives and overdiagnosis.

Overdiagnosis might discourage younger women instead of older women from participating, but others found the contrary. According to our interviews:

Women and doctors preferred a decision aid used in a clinical encounter, containing information on the benefits and harms of screening.

The content of our interviews yielded similar findings to those in the systematic review.

Women prefer to talk about the harms of screening with their doctors. This allows them to ask questions and clarify concepts that might be misinterpreted. That is why we developed a decision aid to be used during the clinical encounter.

Physicians’ recommendations for breast cancer screening were not always consistent with institutional and national guidelines. Furthermore, many of them express that false positive and overdiagnosis are complex terms to explain. Hence, we included more detailed information in the tool to serve as an evidence-based summary in lay terms to support clinicians in having these conversations with their patients.

Many women preferred a digital format, so we decided to develop a digital tool. However, we included the option to print a paper-based summary. Information was divided into sections: introduction to breast cancer, risk factors, definition and purpose of screening, benefits and harms, graphics with probabilities of benefits and harms, values and preferences, a glossary (including false positives and overdiagnosis definitions) and references.

We decided to present information about the probabilities in absolute risk, using frequencies with a constant denominator, as the evidence shows that they are easier to understand and interpret [38].

We chose to depict numerical information using graphics that would contextualise the magnitude of the numbers, i.e. using a theatre seating plan representing 1000 women (see Figure 2).

To help patients match their values, we included patients' narratives regarding motivations and feelings about the decision to undergo breast cancer screening that emerged in the interviews and previous studies [15,39]. They can classify these narratives in three different columns if they agree, disagree, or nor agree nor disagree with each one of them.