Mutter 2015.

Methods	Retrospective longitudinal cohort
Participants	Collected data on women at the Mayo Clinic between 1 January 1960‐31 December 1993 BC developed ipsilateral to the RRM in 25 participants (13 after BRRM; 12 after therapeutic mastectomy and CRRM). The study utilized a study‐specific questionnaire (sent from 1995‐1997), and follow‐up surveys at 10 and 20 years after RRM. All participants who underwent RRM were followed up yearly through the Mayo Clinic Cancer Registry for subsequent BC events and outcomes.
Interventions	BRRM CRRM
Outcomes	Incidence Disease‐free survival
Notes	Subjects also included in Hartmann 1999 and McDonnell 2001
Risk of bias
Bias	Authors' judgement	Support for judgement
Free of selection bias?	Unclear risk	This paper doesn't report on the entire cohort. Table 1 reports on the characteristics of the 25 women who developed BC after RRM, and shows that there were baseline differences in family history status, type of mastectomy undertaken, and breast reconstruction choice, although it is not clear if these are statistically significant.
Free of performance bias?	Unclear risk	Because of the 34‐year timeframe (1960‐1993) it is unclear how RRM techniques, and adjuvant therapies, changed over that time period, and how this may have affected the clinical outcomes.
Free of detection bias?	Low risk	A survey was sent to all women or their next‐of‐kin, and data were also collected via the Mayo Clinic Cancer Registry. Nurses called participants to follow up on surveys. All medical records were also reviewed.
Free of attrition bias?	Low risk	No attrition was reported (there were only 25 participants in this study).