The Relationship between a High-Dairy Diet and Breast Cancer in Women
Recent studies have demonstrated that 1 in 8 women will develop invasive breast cancer at some point in their lives being that breast cancer is the second most common type of cancer among women (American Cancer Society, 2018). Since 2000, the rates of breast cancer have been declining and researchers posit this may be due to better screening processes or possibly due to the decrease in hormone replacement therapy because it was associated with increased risk of breast cancer (Breastcancer.org, 2018). Although breast cancer has been linked to genetic inheritance, lifestyle factors such as diet and hormone levels also play a role in developing the disease (American Cancer Society, 2018). One study indicates that dairy, specifically in the Western world, is pumped full of estrogen and progesterone hormones, which may link a diet high in dairy to breast cancer rates (Ganmaa & Sato, 2005). Estrogenic hormones are more common in fats, so there may be a critical difference between the effects of high-fat dairy products, such as whole-milk compared to the low-fat version referred to as skim milk (Pape-Zambito et al., 2010). Research suggests that there are conflicting results in regards to the association between high-dairy intake and breast cancer (Hunter & Willett, 1994; Boyd et al., 1993), which makes it important to study the methods and possible explanations behind this controversial evidence. In this paper, I will analyze two research studies that provide opposing results in regards to the relationship between breast cancer and a high-dairy diet in order to understand why researchers may have found these varying outcomes.
Kroenke et al. (2013) hypothesized that a high-fat dairy diet, compared to a low-fat dairy diet, is more associated to breast cancer recurrence and mortality rates. To test this hypothesis, a team of researchers conducted a prospective cohort design and studied 1893 women that were a part of larger study (LACE) who already had early stage invasive breast cancer. These women previously went through cancer treatment and exhibited no evidence of recurrence or other types of cancer. The number and type of recurrences were assessed by health status questionnaires obtained annually, and mortality was measured using the Kaiser Permanente Northern California Cancer Registry (KPNC) data sources or using information from the participant’s family. Researchers collected data at two time points that encapsulated the women’s diets, measured by the Fred Hutchinson Cancer Research Center Food Frequency Questionnaire, at baseline and at a 6-year follow-up. This measure asked women how often they ate dairy foods in the past year by specifically asking the participants to describe their daily, weekly and monthly eating habits. The questionnaire displays a medium size portion as an example and then asks the participants to indicate the size of their own dairy servings (small, medium, or large). In order to evaluate fat content, the participants were asked the type of fat levels they usually consumed, such as percentage of fat in their milk, cheese, yogurt and dessert products. They gathered the total number of servings of dairy per day and classified certain types of dairy into high or low-fat groups. In their statistical analyses for high and low-fat dairy intake, the researchers adjusted for possible confounding variables such as age, smoking, menopausal status, race, education, BMI, and reproductive factors. For analyses on breast cancer outcomes, researchers controlled for confounding variables such as disease severity, treatment type, and behavioral factors such as other diet components (i.e. red meat consumption, sugar), exercise, alcohol and smoking behaviors (Kroenke et al., 2013)
The results of this study demonstrate that overall, dairy consumption levels among these women were low, and women were consuming more low-fat dairy than high-fat dairy products. Kroenke et al. (2013) also found that high-fat dairy intake was not related to cancer recurrences, but it was positively associated with mortality. Although this study found a link between high fat dairy intake and mortality from breast cancer, they did not find an overall relationship between dairy and breast cancer. Previous research suggests that breast cancer causes may be linked to calcium and Vitamin D (Lipkin & Newmark, 1999), but these results showed no relation between these vitamins and cancer outcomes (Kroenke et al., 2013) Since a high-fat dairy diet is related to an increase in estrogen, it is important for researchers to further study the impact of estrogen levels on cancer. If these results were to be replicated, it would be valuable to advise women with breast cancer to cut back on high fat dairy products, since it may impact their survival.
One of the strengths of this study was that they adjusted for different factors that could have effected breast cancer severity such as stage, nodal status, tumor size, and lifestyle factors such as SES, exercise, demographics and reproductive history. Since this was a correlational study, we cannot confirm causality, but the study’s adjustment for third factors does strengthen their observed effects (Kroenke et al., 2013). One major limitation in studies regarding diet is the tendency for people to underestimate their food intake. Some of the women responded that they didn’t know what type of milk they were consuming or how much they had daily, so it was difficult to categorize dairy products into high or low-fat groups accurately. Previous research suggests an inverse relationship between dairy products and breast cancer outcomes in samples of premenopausal women (Shannon et al., 2003); however, the current study was mainly postmenopausal, and they did not have a large enough sample of premenopausal women to study the differences in menopausal status in regards to diet. Another limitation of this study was its inability to accurately identify estrogen/progesterone receptor status of the reported products since researchers suggest the effects of estrogen may contribute to higher mortality rates in women with breast cancer (Kroenke et al., 2013).
In a second study, Knekt et al. (1996), the researchers hypothesized that there is an inverse relationship between dairy intake and breast cancer. This was a prospective study that followed 4697 healthy women (free of cancer) for 25 years. The population of women in the study were from Finland, which is significant because Finland has one of the highest levels of milk intake among anywhere in the world. Once the women were screened for cancer, the researchers conducted a modified dietary history interview that analyzed the women’s diet in the last year. Dietary assessments were repeated every 4-8 months after the initial interview. The model assessed for food eaten each day, week, month and year and had participants estimate their portion size similarly to the previous study, Kroenke et al. (2013). The participants also completed a mailed questionnaire to gather information such as residence, occupation, parity, and smoking to be assessed as covariates. Cancer incidence was obtained through the nationwide Cancer Registry along with death certificates from the Central Statistical office of Finland over the 25-year period of the study.
This study found that the rates of developing breast cancer were highest among women over 50, who had never married or had kids, and had a white-collar job. They also found that milk intake was dependent on location and occupation and was higher in western agricultural areas. There was an inverse relationship between milk consumption (as opposed to other dairy products) and developing breast cancer after controlling for age. Previous research suggests a relationship between high saturated fat and breast cancer outcomes, (Boyd et al., 1993; Hunter et al., 1994); however, the present study found no relationship between total or saturated fat and breast cancer (Knekt et al., 1996). Another finding was that higher levels of calcium are related to lower breast cancer incidence in the study, suggesting that calcium may play a protective role in breast cancer prevention. Researchers indicate calcium’s protectiveness may be related to its ability to protect the binding of fatty acids and bile acids when processing dairy products. However, past studies have not found a significant relationship between calcium intake and breast cancer incidence (Katsouyanni et. al 1988) and after adjusting for calcium in the current study, the relationship between diary and breast cancer still existed.
A strength in this study was their ability to follow healthy women to see who developed breast cancer after initially collecting diet data. This study also had a large sample of 4697 women, which increases the study’s power and researchers followed them for a significant period of time (25 years), which allows them to account for long term effects. Researchers also used a comprehensive survey of food consumption compared to previous studies that used a simpler measure for diet (Knekt et al., 1996). A limitation of this study was its inability to control for some health and behavioral factors such as alcohol consumption; however, few women in Finland are heavy drinkers, so this possible confounding variable most likely did not impact the results of the study (Knekt et al., 1996). Since reproductive factors and hormones may play a role in breast cancer (Kelsey and Whittemore, 1994), these confounding variables may have impacted their results since the researchers only obtained reproductive information regarding the women’s number of births.
Although these two studies indicate different associations between breast cancer and dairy intake, it is important to note that they are studying very different populations of women. The first study described involved a population of women who already had breast cancer while the second study focused on women who develop breast cancer out of a sample starting with healthy women. It is possible that this difference in timing of disease occurrence accounts for some of the difference in results. Also, the Knekt et al. (1996) study was more focused on milk products specifically, while the Kroenke et al. (2013) study was more concerned about high vs. low-fat dairy products. There were also more postmenopausal women in the sample from the Kroenke et al. (2013) study compared to the Knekt et al. (1996), which may have played in a role in interactions between hormone levels and dairy products (Shannon et al. 2003). Menopausal status is especially relevant because the results from the Knekt et al. (1996) study only demonstrated the inverse relationship after controlling for age. Since dairy intake was associated with cancer mortality, it is possible that the fat in dairy cells negatively interacts with cancer cells, but not healthy cells (Kroenke et al., 2013). In the case of the study in Finland, the dairy production process may differ from the US and thus, alter the levels of estrogen in milk. Also, since milk is very common in Finland, it is possible that this population is more tolerant to its effects on hormone levels. Another major difference between the two studies is their amount of data collected on diet. In the Kroenke et. al (2013) study, they gathered assessments of diet at two time points, one at baseline and the other at a 6-month follow-up, while the Knekt et al. (1996) study interviewed people regarding diet every 4-6 months for 25 years. Considering that diet recall is grossly underestimated (Kroenke et. al, 2013), these differences in diet data collection may suggest why the studies are producing different results.
It would be beneficial to have future studies examine the specific nutrients and hormone levels in dairy products consumed to understand why dairy can be both positively and inversely associated with the risk of breast cancer. Research suggests that some cultures place more value on milk-oriented diets in comparison to others (Cramer et al., 1994), perhaps because of cultural differences and views on dairy product’s health benefits. Future studies should examine dairy consumption cross-culturally to assess whether dairy production differences and metabolism differences among certain groups contribute to the relationship between dairy and breast cancer. Additionally, a major problem with diet research, is that it is usually gathered at one time point (Kroenke et. al, 2013) and researchers need to have more frequent questionnaires to accurately gauge the types of nutrient intake and total consumption. Although these studies both used a detailed questionnaire for food consumption, participants may have been inaccurately showcasing their overall diet patterns, or they may have changed their diet habits after the questionnaire was completed, which could have impacted their breast cancer outcomes. Also, future studies could focus more on differentiating between high and low-fat dairy since Kroenke et al. (2013) and Knekt et al. (1996) were unable to accurately measure these differences. Since menopausal status and age were important factors in the relationship between breast cancer and dairy products, future researchers may want to expand on these findings and study how age may impact the relationship between dairy products, estrogen levels, and breast cancer.