Figure10.

a) The funnel plot for publication bias of the risk estimate of developing breast cancer in patients with proliferative disease without atypia versus non-proliferative disease. No significant publication bias was seen according to the funnel plot. b) The Begg plot for the publication bias of risk estimate of developing breast cancer in patients with proliferative disease without atypia versus non-proliferative disease. The Egger’s test showed a low publication bias (P=0.049). c) The “trim and fill” plot for the publication bias of risk estimate of developing breast cancer in patients with proliferative disease without atypia versus non-proliferative disease. According to the trim test result, publication bias was rejected and none of the studies were added to the present study. d) The funnel for publication bias of the risk estimate of developing breast cancer in patients with the proliferative disease with atypia versus non-proliferative disease. No significant publication bias was seen according to the funnel plot. e) The Begg plot for the publication bias of risk estimate of developing breast cancer in patients with the proliferative disease with atypia versus non-proliferative disease. No significant publication bias was seen according to the Begg plot.