Bisphenol-A exposure and risk of breast and prostate cancer in the Spanish European Prospective Investigation into Cancer and Nutrition study

Table 2 Cox regression and risk of breast and prostate cancer

		Breast cancer					Prostate cancer
		N	HR	SE	p	95% CI	N	HR	SE	p	95% CI
Model A	BPA levels (for 5 ng/ml increase)	2306	1.047	0.037	0.200	0.98–1.12	2328	0.989	0.036	0.749	0.92–1.06
Model B	log₂(BPA)	2306	1.011	0.024	0.655	0.97–1.06	2328	1.035	0.021	0.093	0.99–1.08
Model C	Categorized BPA (values in ng/ml)
	<LOD	705	1	–	–	–	658	1	–	–	–
	Tertile 1 [0.2–1.8)	562	0.820	0.123	0.185	0.61–1.10	534	1.404	0.208	0.022	1.05–1.88
	Tertile 2 [1.8–5.1)	556	0.875	0.132	0.376	0.65–1.18	540	1.365	0.200	0.034	1.02–1.82
	Tertile 3 [5.1–68.9]	483	1.127	0.169	0.425	0.84–1.51	596	1.305	0.193	0.072	0.98–1.74

Model A: linear BPA stratified by center and age group, and adjusted by age, education level, BMI, physical activity, smoking status, alcohol consumption (and menopause, n° of pregnancies, breastfeeding, oral contraceptives, HRT and the interaction between BMI and menopause for women)
Model B: log2BPA stratified by center and age group, and adjusted by age, education level, BMI, physical activity, smoking status, alcohol consumption (and menopause, n° of pregnancies, breastfeeding, oral contraceptives, HRT and the interaction between BMI and menopause for women)
Model C: categorized BPA (<LOD category plus tertiles based on measurable values in ng/dl) stratified by center and age group, and adjusted by age, education level, BMI, physical activity, smoking status, alcohol consumption (and menopause, n° of pregnancies, breastfeeding, oral contraceptives, HRT and the interaction between BMI and menopause for women)
LOD: limit of detection; CI: confidence interval

ISSN: 1476-069X