Skip to main content
Fig. 1 | BMC Cancer

Fig. 1

From: A body shape index (ABSI) is associated inversely with post-menopausal progesterone-receptor-negative breast cancer risk in a large European cohort

Fig. 1

Breast cancer risk according to allometric anthropometric index quintiles

ABSI – a body shape index; BMI – body mass index; CI – confidence interval; ER+/- – oestrogen receptor status; HI – hip index; HR – hazard ratio; PR+/- – progesterone receptor status. HR (95% CI) (compared to the lowest quintile) were derived from Cox proportional hazards models, including quintile categories for ASBI, HI, and BMI as exposures (lowest quintile reference, cut-offs: 68.843, 71.525, 73.919, 76.884 for ABSI; 62.763, 64.240, 65.477, 66.933 for HI; 21.63, 23.54, 25.58, 28.55 kg/m2 for BMI). Estimates are plotted at the quintile medians. Models were stratified by age at the anthropometric assessment, country, and categories of menopausal status and, for post-menopausal women, age at menopause (pre-menopausal, peri-menopausal, menopause at < 46 years, menopause at 46 to < 52 years, menopause at ≥ 52 years, menopause at unknown age), and adjusted for height (continuous), smoking status and intensity (never smoker, former quit ≥ 15 years, former quit < 15 years, current ≤ 10 cigarettes/day, current > 10 cigarettes/day), alcohol consumption (none, < 4 g/day, 4 to < 16 g/day, ≥ 16 g/day), physical activity (inactive, moderately inactive, moderately active, active), education (primary/none, technical, secondary, university/longer), hormone replacement therapy use (never, former, current, missing), oral contraceptives use (never, former, current), age at the first period (continuous), parity with age at the first live birth (nulliparous, one at < 25 years, one at ≥ 25 years, two at < 25 years, two at ≥ 25 years, ≥ 3 at < 25 years, ≥ 3 at ≥ 25 years, missing), breastfeeding with duration (never, < 6 months, ≥ 6 months, missing), and energy intake (log-transformed continuous); p-non-lin. – p-value for non-linearity, obtained from likelihood ratio tests comparing the fully adjusted models, including ABSI, HI, and BMI on a linear untransformed scale, with models including restricted cubic splines, individually for each anthropometric index, with knots at the corresponding quintile cut-offs. HR estimates are shown in Supplementary Table S3

Back to article page