Assessment of physical activity

At recruitment, physical activity was assessed during in-person interviews using a questionnaire that was designed and evaluated for participants in our study based on the existing validated questionnaires [11] and experiences from previous analyses of physical activity questionnaires [12]. At the follow-up, physical activity was assessed during telephone interviews using a questionnaire that was based on the recruitment questionnaire. Physical activity assessment included self-reported participation in walking and cycling for the purposes of commuting/transportation as well as recreational activities, sports, and fitness. Metabolic equivalent task-hours per week (MET-h/week) were calculated by multiplying the average hours per week spent on each activity with an individual intensity score [13]. Leisure-time physical activity is defined as additional activities related to recreational physical activities, sports, and fitness [14]. We have evaluated leisure-time physical activity and not total physical activity in relation to prognosis in this analysis. Total physical activity would also include walking and cycling for commuting/transportation in addition to other domains. In Germany, walking and cycling are usual methods for transportation. They are generally not performed with moderate intensity but with light intensity and so do not produce noticeable increases in breathing and heart rate [15]. For this reason, we focus on leisure-time physical activity.

Prediagnosis physical activity was determined from the information collected at the recruitment interview (median 3.8 months after diagnosis). Women were asked about their physical activity from the age of 50 until diagnosis. They were asked to list up to three leisure-time activities in which they most frequently participated. For each activity, they were asked to provide, from the age of 50 until diagnosis, the number of years, months per year, either days per week or days per month, and the number of hours/minutes per day they participated. Additionally, they were asked how much time (hours/minutes) they spent walking outside the home and cycling as a form of commuting or everyday cycling.

Postdiagnosis physical activity was determined from information collected at the follow-up interview in 2009 (median 5.8 years after diagnosis) and was the physical activity performed from 3 months after diagnosis of breast cancer to the follow-up interview. Patients were asked if they had (re-)started any (other) leisure-time activities and to list up to four of these. For each activity, they were asked when they started (month/year), if they stopped (month/year) or if they were still participating, and the number of days per week and hours or minutes per day that they participated in each activity. They were additionally asked how much time (hours/minutes) they spent walking outside the home and cycling as a form of commuting or everyday cycling.