Forty-two studies (n=611,801) were included in the review (30 case-control studies, seven cohort studies, one RCT and four cancer registry studies). There were a total of 12,238 cases of ovarian cancer.
Ever-use risk of hormone therapies: There was a significantly increased risk of ovarian cancer in women who had ever used oestrogen/progestin therapy (31 data sets; OR/RR 1.11, 95% CI: 1.02 to 1.21; no statistical heterogeneity between studies, I2=15%), or oestrogen therapy (48 data sets; OR/RR 1.28, 95% CI: 1.18 to 1.40; moderate statistical heterogeneity between studies, I2=48%), but not in women with unspecified menopausal hormone therapy (72 data sets; OR/RR 1.02, 95% CI: 0.98 to 1.07; statistical heterogeneity between studies, I2=63).
Annual risk of hormone therapies: There were significant annual risk increases of ovarian cancer in women who used oestrogen/progestin therapy (22 data sets; OR/RR 1.04, 95% CI: 1.02 to 1.06; no statistical heterogeneity between studies, I2=0%), oestrogen therapy (18 data sets; OR/RR 1.07, 95% CI: 1.06 to 1.08; some statistical heterogeneity between studies, I2=21%), or unspecified menopausal hormone therapy (60 data sets; OR/RR 1.04, 95% CI: 1.03 to 1.04; moderate statistical heterogeneity between studies, I2=43%).
Subgroup analyses: There were no differential impacts of any therapy on histological subtypes. Risks were increased in studies conducted in Europe relative to studies conducted in North America for oestrogen therapy and oestrogen/progestin therapy.
The funnel plots demonstrated publication bias for studies of oestrogen/progestin therapy only.