Table 5 Findings on the association of muscle-strengthening exercise with recurrence and mortality among breast cancer survivors

Recurrence

Ansa et al., 2015 [34]

The proportion of those participating in muscle strengthening in the past month was examined among those with breast cancer recurrence and those free of breast cancer recurrence. Results are reported below:

• Among those with breast cancer recurrence, 69.7% (n = 23) reported muscle strengthening in the past month vs. 30.3% (n = 10) who did not

• Among those free of breast cancer recurrence, 67.6% (n = 94) reported muscle strengthening in the past month vs. 32.4% (n = 45) who did not

• p = 0.82

*Variables adjusted for: None

Forbes et al., 2015 [38]

The proportion of those who met strength exercise guidelines was compared between those with and without recurrence. Results are reported below:

• 9% (n = 1) with recurrence met guidelines

• 24 (n = 57) of those who were recurrence free met guidelines

• p = 0.430

*Variables adjusted for: None

Troeschel et al., 2023 [37]

The association of different levels strength training with breast cancer recurrence was examined by computing Cox proportional hazard ratios. Results are reported below:

• Referent: None

• 1–2 times/wk: HR = 0.81 (95% CI: 0.50, 1.29)

• > 2 times/wk: HR = 1.26 (95% CI: 0.84, 1.90)

*Variables adjusted for: Age, tumor stage, ER status, chemotherapy, Herceptin, radiotherapy, comorbidities, smoking status, income, and education level, diet, alcohol, aerobic exercise, sedentary behavior, body weight

Breast cancer-specific mortality

Fortner et al. 2023 [35]

The association of different levels of strength training (measured in MET-hrs/wk) with breast-cancer specific mortality was examined by computing Cox proportional hazard ratios. Results are reported below:

• Referent: 0 MET-hrs/wk

• > 0–1 MET-hrs/wk: HR = 0.59 (95% CI: 0.33, 1.05) (Model 1); HR = 0.60 (95% CI: 0.33, 1.08) (Model 2)

• ≥ 1 MET-hrs/wk: HR = 0.51 (95% CI: 0.25, 1.02) (Model 1); HR = 0.49 (95% CI: 0.23, 1.06) (Model 2)

*Variables adjusted for: (Model 1) estrogen receptor (ER)/progesterone receptor (PR) status, treatment with tamoxifen, aromatase inhibitor, Herceptin, and/or chemotherapy, stage at diagnosis (I-III), pre-diagnosis hormone therapy use, pre-diagnosis BMI, BMI change from pre-diagnosis to current, alcohol consumption, smoking status, aspirin use, and neighborhood SES.; (Model 2) same variables as Model 1 + pre-diagnosis exercise

**Results for postmenopausal women only, and stratified by BMI are reported in Supplementary Tables 5 and 6 respectively of Fortner et al. 2023 [35]

Tarasenko et al., 2018 [36]

The association of meeting muscle-strengthening exercise guidelines and breast cancer-specific mortality was examined by computing Cox proportional hazard ratios. Results are reported below:

• Referent: <2 times/wk

• ≥ 2 times/wk: HR = 0.94 (95% CI: 0.59, 1.49)

*Variables adjusted for: Age (continuous), race/ethnicity, education level, marital status, insurance status, self-rated health, activity limitations, smoking status, BMI categories, number of comorbid conditions, and age at first cancer diagnosis

Troeschel et al., 2023 [37]

The association of different levels strength training with breast-cancer specific mortality was examined by computing Cox proportional hazard ratios. Results are reported below:

• Referent: None

• 1–2 times/wk: HR = 0.26 (95% CI: 0.08, 0.86)

• > 2 times/wk: HR = 0.84 (95% CI: 0.40, 1.76)

*Variables adjusted for: Age, tumor stage, ER status, chemotherapy, Herceptin, radiotherapy, comorbidities, smoking status, income, and education level, diet, alcohol, aerobic activity, sedentary behavior, body weight

Cardiovascular disease specific mortality

Tarasenko et al., 2018 [36]

The association of meeting muscle-strengthening exercise guidelines and cardiovascular disease specific mortality was examined by computing Cox proportional hazard ratios. Results are reported below:

• Referent: <2 times/wk

• ≥ 2 times/wk: HR = 0.69 (95% CI: 0.36, 1.33)

*Variables adjusted for: Age (continuous), race/ethnicity, education level, marital status, insurance status, self-rated health, activity limitations, smoking status, BMI categories, number of comorbid conditions, and age at first cancer diagnosis

All-cause mortality

Fortner et al. 2023 [35]

The association of different levels of strength training (measured in MET-hrs/week) with all-cause mortality was examined by computing Cox proportional hazard ratios. Results are reported below:

• Referent: 0 MET-hrs/wk

• > 0–1 MET-hrs/wk: HR = 0.89 (95% CI: 0.64, 1.25) (Model 1); HR = 0.90 (95% CI: 0.64, 1.27) (Model 2)

• ≥ 1 MET-hrs/wk: HR = 0.61 (95% CI: 0.39, 0.97) (Model 1); HR = 0.65 (95% CI: 0.40, 1.06) (Model 2)

*Variables adjusted for: (Model 1) ER/PR status, treatment with tamoxifen, aromatase inhibitor, Herceptin, and/or chemotherapy, stage at diagnosis (I-III), pre-diagnosis hormone therapy use, pre-diagnosis BMI, BMI change from pre-diagnosis to current, alcohol consumption, smoking status, aspirin use, and neighborhood socio-economic status.; (Model 2) same variables as Model 1 + pre-diagnosis physical activity

**Results for postmenopausal women only are reported in Supplementary Table 5 of Fortner et al. 2023 [35]

Tarasenko et al., 2018 [36]

The association of meeting muscle-strengthening exercise guidelines and all-cause mortality was examined by computing Cox proportional hazard ratios. Results are reported below:

• Referent: <2 times/wk

• ≥2 times/wk: HR = 0.91 (0.66, 1.26)

*Variables adjusted for: Age (continuous), race/ethnicity, education level, marital status, insurance status, self-rated health, activity limitations, smoking status, BMI categories, number of comorbid conditions, and age at first cancer diagnosis

Troeschel et al., 2023 [37]

The association of different levels strength training with all-cause mortality was examined by computing cox proportional hazard ratios. Results are reported below:

• Referent: None

• 1–2 times/wk: HR = 0.63 (95% CI: 0.41, 0.98)

• > 2 times/wk: HR = 0.99 (95% CI: 0.69, 1.42)

*Variables adjusted for: Age, tumor stage, ER status, chemotherapy, Herceptin, radiotherapy, comorbidities, smoking status, income, and education level, diet, alcohol, aerobic activity, sedentary behavior, body weight