"Fat women who refuse to diet 'are more likely to get cancer'," states the Mail Online, using a headline that is both inaccurate and offensive.
The study it reports on looked at the relationship between weight during adulthood, and cancer risk.
The researchers found that duration of time spent overweight or obese, as well as the degree, seems to have a compounding effect on cancer risk. But they did not look at whether the women in the study were asked to lose weight by dieting.
This latest study of more than 70,000 women took multiple measurements over about 12 years and also used women's own estimates of their weight at ages 18, 35 and 50, to calculate how many years they had been overweight or obese. They then calculated the risk of getting an obesity-related cancer, linked to decades of being overweight or obese.
They found that each decade of overweight was linked to a 7% increased risk of obesity-related cancer. Womb cancer (specifically endometrial cancer; a cancer of the lining of the womb) was most strongly linked to obesity. Both duration and degree of overweight increased cancer risk.
The study has limitations, but suggests that keeping to a healthy weight throughout life may help women to avoid some cancers.
If you are concerned about your weight then try the NHS weight loss plan. This is a 12 week plan designed to help people lose weight in a sustainable way through a combination of healthy eating and exercise.
Where did the story come from?
The study was carried out by researchers from nine different universities or research institutions in the US and one in Israel. It was funded by the US National Institutes of Health and the World Cancer Research Fund.
The Mail Online's headline saying "Fat women who refuse to diet 'are more likely to get cancer'," is both unhelpful and offensive; straying into the realm of "body shaming". Women who took part in the study received no advice about dieting; let alone actively refused to diet. (The paper version of the headline in the Daily Mail avoids using any offensive language).
Once you get past the unpleasant and misleading headline, with its implications that overweight women are to blame for cancer by "refusing" to diet, the Mail's report is reasonably accurate. However, it repeatedly states that excess weight "feeds" cancers, which is an overly simplistic way of describing the theory that weight may be linked to cancer through its effect on hormone levels, inflammation and damage to DNA. It does not mention that observational studies such as this one cannot prove that overweight causes cancer.
What kind of research was this?
The research is a cohort study, looking at what happens to a large group of people over time. Cohort studies are good at assessing links between different factors (in this case duration of overweight and risk of certain cancers) but cannot show that one causes the other.
What did the research involve?
Researchers looked at information from a big group of postmenopausal women without cancer, taking part in a long-term cohort study in the US, called the Women's Health Initiative.
They calculated how long they had been either normal weight, overweight or obese, and followed them to see how many of them got one of 10 cancers thought to be linked to weight, over a 12-year period.
Weight was assessed with the widely used body mass index (BMI) measurement, where:
- 18.5 to 24.9 means you're a healthy weight
- 25 to 29.9 means you're overweight
- 30 to 39.9 means you're obese
- 40 or above means you're severely obese
After adjustment to take account of confounding factors that could affect cancer risk, including age, smoking, physical exercise, diet and whether women took hormone replacement therapy (HRT), they calculated risks of cancers per decade of overweight or obesity.
The researchers used measurements of weight and height taken during the study, and asked the women to remember their measurements at ages 18, 35 and 50. Using this information, they calculated how long the women had been normal weight, overweight or obese during their adult lives. The cancers monitored were:
- breast cancer (postmenopausal)
- colon cancer
- endometrial (womb) cancer
- gallbladder cancer
- liver cancer
- ovarian cancer
- pancreatic cancer
- rectal cancer
- renal (kidney) cancer
- thyroid cancer
They also calculated the effect of different degrees of overweight, by looking at the number of units of BMI over the healthy limit of 25 units that women were, for each time period.
This allowed them to compare both time and level of overweight.
What were the basic results?
There were 6,301 cancers among the 73,913 women in the study, over 12 years. About two thirds of women were overweight or obese at some point during adulthood. On average, overweight women were overweight for 31 years of their adult lives.
For every 10 years of being overweight, women had a 7% higher chance of being diagnosed with one of the cancers (hazard ratio (HR) 1.07, 95% confidence interval (CI) 1.06 to 1.09). The risk was increased for colon and postmenopausal breast cancers but highest for endometrial cancer and kidney cancer. No link was seen between time spent overweight and rectal, liver, gallbladder, pancreatic, ovarian or thyroid cancer.
When researchers took into account the degree of overweight, the link became stronger, especially for endometrial cancer. Each additional decade spent with a BMI of 35 (10 units of BMI over normal weight) carried a 37% increase in the risk of endometrial cancer (HR 1.37, 95% CI 1.29 to 1.46).
How did the researchers interpret the results?
The researchers said they cannot show that time spent overweight causes cancer, but that their findings "suggest that reducing overweight duration in adulthood could reduce cancer risk and that obesity prevention is important from the outset". They said this meant that healthcare services should recognise that "excess body weight in women is important to manage, regardless of the age of the patient."
This study adds to evidence that being overweight or obese for long periods of time may increase the risk of certain cancers, just as it increases the risk of cardiovascular disease and diabetes.
The study's size and use of BMI measurements over time mean it is likely to be more reliable than smaller studies, or those that look at BMI only at one time point. The design allows researchers to look at how weight during a lifetime, rather than at one single point in life, may affect cancer risk.
However, there are limitations. It's an observational study, so while researchers took account of known confounding factors such as smoking and exercise, it's always possible some factors were not accounted for. This means the study cannot prove that overweight directly caused cancer. The other main concern is that it relied on women remembering and correctly reporting their weight decades earlier, at ages 18 and 35.
These caveats aside, the study is a serious attempt to quantify the risk that overweight and obesity contribute to cancer risk. Obesity levels have been rising in recent decades and figures from Public Health England show 65% of men and 58% of women in England were overweight or obese in 2014.
The best way to keep to a healthy weight throughout life is to eat a healthy, balanced diet and take plenty of exercise. Some people struggle with their weight more than others, and it can be hard to shift weight once you've put it on.
If you're worried about your weight and want help in reaching a healthy weight, you can talk to your GP for advice or see our information on healthy weight loss.
"Fat women who refuse to diet 'are more likely to get cancer'," states the Mail Online, using a headline that is both inaccurate and offensive. The study it reports on looked at the relationship between weight during adulthood, and cancer risk.
Links to Headlines
Links to Science
Arnold M, Jiang L, Stefanick ML, et al. Duration of Adulthood Overweight, Obesity, and Cancer Risk in the Women's Health Initiative: A Longitudinal Study from the United States. PLOS Medicine. Published online August 16 2016
- NHS Choices links
- Find Weight loss support groups services
- Editor's pick of the blogs
- Evidently Cochrane
- Fact Check Central
- Research the Headlines
- Science-Based Medicine
- Science blog - Cancer Research UK
- Science Media Centre
- Sense about Science
- Understanding Health Research