“…However, few studies were included in some subgroups leading to wide 95% CIs and limiting the ability to assess/explain heterogeneity. Meta‐regression analyses indicated that tumor invasiveness or stage may partly explain the heterogeneity observed for breast cancer recurrence ( I 2 = 54%), with the studies of invasive and in situ breast cancers showing a higher average positive association (five studies/publications) 55,96,110,112,127 compared with studies with only invasive cancers (50 studies, 41 publications), 40,41,43‐46,48,51,54,56,59,63‐66,68‐70,72,74,79,84,85,88,89,92,94,102,104,106,115,116,120‐125,130,131,136 and the studies of metastatic breast cancers showing inverse associations (eight studies, two publications) 70,121 compared with the positive associations observed for early stage disease cases (27 studies/publications) 41,43,44,46,48,51,54,56,59,63,66,68,69,72,74,79,88,89,94,106,116,120,122‐124,127,130 ( P meta‐regression = .02 and .003, respectively). A similar pattern by tumor stage was observed for all‐cause mortality (30 studies, 28 publications on early stage diseases 41,43,44,46,54,63,66‐68,72,78,79,88,89,91,94,95,98,100,111,116,120,123,125‐127,129,130 ; five studies, three publications on metastatic disease 70,81,105 ; P meta‐regression .14; Table 2).…”