Wednesday November 2 2011
Women are advised to drink no more than 2–3 units of alcohol a day
Women who drink within the recommended limits are still putting their health at risk, according to The Daily Telegraph. It said new research shows that less than one small glass of wine a day increases the risk of breast cancer.
This large, well-conducted study assessed women’s drinking habits over many years. It found that even low levels of alcohol were associated with a small increase in breast cancer risk. Women who drank three to six drinks a week (5–9.9 grams of alcohol a day) had a 15% greater risk of breast cancer than women who did not drink alcohol. The risk of cancer increased with the amount of alcohol consumed.
The study has a few limitations. For example, it relied on the women to report their own alcohol consumption over yearlong periods, which introduces the possibility of error. Despite this, the link between breast cancer and alcohol is not a new one, and the findings back up the results of several previous studies.
The increase in risk for women who drank low levels of alcohol was modest. Individuals will need to weigh up the modest increase in risk against the pleasure of occasional drinking and the potential benefits for cardiovascular health. There is currently no definite evidence that stopping drinking altogether will reduce a woman's risk of breast cancer.
Where did the story come from?
The study was carried out by researchers from a number of centres in the US, including Brigham and Women’s Hospital and Harvard Medical School, Boston. It was funded by the National Institutes of Health.
The study was published in the peer-reviewed Journal of the American Medical Association.
The report was covered accurately in the media, and most papers included comments from independent experts, which put the significance of the research into context. Most of them also noted that even women who drink within the current recommended limits may be at modest risk.
What kind of research was this?
This prospective cohort study followed nearly 106,000 women for 28 years to evaluate the association between alcohol consumption and breast cancer. Cohort studies that follow large groups of people for long periods of time are often used to look at the effects of lifestyle on health outcomes, although they cannot prove cause and effect.
The researchers point out that many studies have linked alcohol consumption to breast cancer risk, but that the risk of consuming smaller amounts of alcohol has not been well quantified. Their aim was to look at the association in more detail, including the quantity and frequency of alcohol consumption, and the women’s age at the time.
What did the research involve?
The researchers used data from a large US study, called the Nurses’ Health Study, which began in 1976 and involved 121,700 female US nurses aged 30 to 55 years old. The women completed a questionnaire at the beginning of the study that included questions about risk factors for cancer and cardiovascular disease. They were sent follow-up questionnaires every two years to provide up-to-date information and to record any diseases they had developed.
For the present study, researchers analysed data from 1980 onwards, when alcohol intake was first assessed. After they excluded women who had died or developed cancer since 1976, or who did not respond to questions on alcohol, the researchers had data on 105,986 women. Information on alcohol consumption was taken from a food frequency questionnaire in which women were asked to report their drinking during the previous year. Their reported number of drinks a day and type of alcohol consumed were used to estimate their daily alcohol consumption in grams. These data were updated seven times over the next 26 years, and the cumulative average alcohol intake was assessed for this period.
Data on the participants’ current drinking patterns and those during their earlier adult life were first collected in the 1988 questionnaire, which included 74,854 participants. In 1988 (and at three further follow-up points), participants were asked to provide the number of days they drank alcohol in a typical week and the largest number of alcoholic drinks that would be consumed in one day in a typical month. In the 1988 assessment only, they were asked about the number of alcoholic drinks they had each week at three different age periods: 18–22, 25–30 and 35–40 years old.
The participants were categorised according to how many grams of alcohol they consumed a day: none, 0.1–4.9g, 5–9.9g, 10–19.9g, and more than 20g of alcohol a day.
Throughout follow-up, each questionnaire asked participants whether they had been diagnosed with invasive breast cancer and, if so, when. These reports were confirmed using medical records. In their analysis, the researchers only looked at data on alcohol intake before a diagnosis of breast cancer.
The researchers used standard statistical methods to analyse the data, and the results were adjusted for other established risk factors for breast cancer, including whether they had reached the menopause, family history of breast cancer and use of hormone replacement therapy.
What were the basic results?
From 1980 to 2008 (during 2.4 million person years of follow-up), 7,690 cases of invasive breast cancer were diagnosed. The researchers calculated the association between breast cancer and cumulative alcohol intake (a person’s average lifetime alcohol intake, taken from an average of their intake at each follow-up point). They found that:
- Women who consumed 5.0–9.9g of alcohol a day (equivalent to 3–6 drinks a week) had a 15% greater risk of breast cancer compared to women who did not drink at all (relative risk [RR] 1.15, 95% confidence interval [CI] 1.06 to 1.24). The size of the increase was small. Among women who drank no alcohol at all, breast cancer occurred at a rate of 281 cases per 100,000 person years. Among women who drank 3–6 drinks a week, the rate was 333 per 100,000 person years (an extra 52 cases).
- The size of the risk progressively increased with each of the four intake categories, so that women in the highest consumption category, who consumed on average 30g or more daily (at least two drinks a day) had a 50% greater risk of breast cancer than non-drinkers (RR 1.51, 95% CI 1.35 to 1.70).
- No significant risk was associated with the lowest consumption of 0.1–4.9g of alcohol a day (equivalent to one to three drinks a week).
- When the researchers looked at two broad age categories – earlier life (ages 18–40) and later adult life (40 or over) – alcohol intake in both these separate periods of life was associated with risk.
How did the researchers interpret the results?
The researchers say their results highlight the importance of considering lifetime exposure when evaluating the effect of alcohol on breast cancer risk. They say that alcohol may affect breast cancer risk by increasing blood levels of the hormone oestrogen, which is implicated in many but not all types of breast cancer.
In this large, well-conducted study, women were followed over a long time and their alcohol intake was assessed during different age periods. This gives a comprehensive evaluation of the effects of alcohol throughout a woman’s life. The observed link between breast cancer and alcohol is not new, and alcohol is already an established risk factor for breast cancer. This study provides valuable, in-depth data on the effects of a woman’s average lifetime alcohol intake and the risks associated with different levels of consumption.
One unavoidable limitation of this study is its reliance on women remembering and reporting their alcohol use over the past 12 months. There is a risk that average alcohol intake may have been categorise wrongly, particularly as the cumulative intake was estimated using an average of the intakes reported at each follow-up point. It is possible that women inaccurately reported their intake or that their intake did not remain the same over time. As one independent expert pointed out, it is common for people to underestimate their alcohol use in questionnaires (although the authors say the study took measures to validate the participants’ responses). A second limitation of this type of study is that other factors might be associated with both alcohol intake and risk of breast cancer (called confounders). The authors made careful attempts to adjust their analyses for established risk factors for breast cancer (such as use of hormone therapy, family history of breast cancer, age at first period and menopause). However, it is still possible that other unknown or unmeasured factors could have had an effect.
This study found that even a low level of alcohol consumption, assessed from information throughout a woman’s lifetime, slightly increased the risk of breast cancer, and that the risk increased with the amount of alcohol consumed. Alcohol consumption has already been associated with several cancers, including breast cancer. As the authors point out, women will need to weigh up the modest risks of light alcohol consumption against both the pleasure of occasional drinking and the potential benefits for cardiovascular health.
An accompanying editorial to this study also addresses an important question: as breast cancer risk increases with age, should postmenopausal women consider stopping drinking altogether to reduce their risk? The author of the editorial says that a woman’s individual risk factors for breast cancer should be taken into account, but notes that currently “there are no data to provide assurance that giving up alcohol will reduce breast cancer risk.”
Further research looking at the relationship between risks and benefits is warranted.