Repository logo

Coffee consumption increases risk of advanced breast cancer among Singapore Chinese women




Zhu, Lei, author
Bachand, Annette, advisor
Butler, Lesley, advisor
Nelson-Ceschin, Tracy, committee member

Journal Title

Journal ISSN

Volume Title


Background: Experimental data support both a protective and risk enhancing effect of coffee compounds on breast cancer development. A few retrospective studies conducted before 1988 also reported inconsistent findings. However, more recent prospective epidemiological data do not support a relationship between coffee intake and breast cancer risk among western populations, and well-conducted studies among Asian population are lacking. Coffee is one of the most commonly consumed beverages, but more among western populations than most Asian populations. In Singapore with relatively recent economic development, aspects of western diet and lifestyle, including coffee consumption, have become more prevalent. Even though the breast cancer incidence rate kept increasing in Singapore during last three decades, it is still much lower than western countries. Since many contributing factors (such as diet and lifestyle) to the lower risk profile among Asian women compared with US women, conducting prospective studies on evaluating the association between coffee intake and breast cancer risk among this Asian population help us have opportunity to observe the breast cancer-coffee association if there is one, among this lower breast cancer profile population. Objective: To evaluate the effect of coffee consumption and overall caffeine intake on breast cancer incidence among women in the Singapore Health Study. To achieve this objective, we have the following analysis strategies: 1) Evaluate the effect of coffee consumption and overall caffeine intake on breast cancer incidence; 2) Restrict the analysis to the subgroup group: late stage breast cancer cases (metastatic or ≥5cm or present in lymph nodes vs. localized disease), since we only observed the statistically significant findings among the advanced disease cases; 3) Conduct subgroup analysis by estrogen/progesterone receptor (ER/PR) status; 4) Conduct stratified analyses by potential effect modifier: below and above median BMI. Methods: At baseline, between 1993 and 1998, information on dietary history from 35,303 Singaporean Chinese women aged 45 to 74 years was collected by trained interviewers using a structured questionnaire. Using Cox regression models, we calculated hazard ratios (HR) and 95% confidence intervals (CIs) adjusted for potential confounders. Stratified analyses by menopausal status, stage of disease, and body mass index (BMI) were conducted to evaluate potential effect modification. Results: After a mean follow-up of 10 years, 629 women developed breast cancer. Average intake among coffee drinkers (80% of the study population) was 1.5 cups/day (interquartile range=0.8, 2.3), compared with average 3.2 cups/day coffee consumption in United States. Restricting the analysis to coffee, overall, women consuming ≥2 cups of coffee per day had a statistically non-significant increase in risk of breast cancer compared with coffee nondrinkers or monthly drinkers (HR=1.23; 95% CI: 0.98, 1.54). There was no association with either coffee or caffeine intake and localized breast cancer risk. However, the positive overall association strengthened and became statistically significant when the analysis was restricted to advanced breast cancer (n=272) (HR=1.90; 95% CI: 1.30, 2.77). Considering total caffeine consumed, results were similar (HR = 2.23; 95% CI: 1.52, 3.26) comparing the highest (median amount (IQR) = 244.96 mg (60.79)) to the lowest quartile (median amount (IQR) = 18.59 mg (32.77)) of caffeine intake. The associations between overall coffee and caffeine intake and advanced breast cancer did not differ by menopausal status. We conducted stratified analyses by BMI and found that the adverse effect of coffee intake on the development of advanced breast cancer was present only among heavier women (BMI ≥median, 23 kg/m2) (HR=2.35; 95% CI: 1.51, 3.66; P for interaction=0.02). Conclusion: We provide the first prospective evidence among Chinese women suggesting an adverse effect of coffee consumption at relatively low levels of intake on advanced stage breast cancer. Our data suggest that a higher BMI ≥23 kg/m2, may increase the adverse effect of coffee on advanced breast cancer development.


2013 Spring.
Includes bibliographical references.

Rights Access


advanced disease
breast cancer
body mass index


Associated Publications