Food and diet

'Yo-yo' diets 'not a no-no'

"Yo-yo dieting is NOT bad for you and won’t stop you losing weight in the long run," the Daily Mail informs us.

The news is based on a study of overweight and obese postmenopausal women with a history of repetitive "weight cycling" (also known as yo-yo dieting). It looked at whether they were at a disadvantage compared with those without a history of weight cycling when it came to losing weight. The researchers randomly assigned women to spend a year on a programme of either:

  • reduced diet only
  • exercise only
  • a combination of both exercise and diet
  • a control intervention – that is no change to either diet or exercise

They found that women considered weight cyclers were not at a disadvantage for weight loss.

There are some limitations to this study, however, including the fact that there is no standard definition of weight cycling and that some information was self-reported by the women, which may affect the reliability of the results.

The study does not provide evidence that yo-yo dieting is effective in helping you lose weight. But it does highlight an important point – that yo-yo dieters should not be put off further attempts to lose weight.

Where did the story come from?

The study was carried out by researchers from the US Fred Hutchinson Cancer Research Center, the University of Washington and other US institutions. It was funded by the Canadian National Cancer Institute, the National Institute of Health and the Canadian Institutes of Health Research. The study was published in the peer-reviewed journal Metabolism.

The story was covered by the Daily Mail, and the headline is misleading. The researchers did not find that yo-yo diets are not bad for you. If anything the study found that women with a history of yo-yo dieting tended to be heavier than women who didn’t yo-yo diet.

What the researchers did find was that yo-yo diets were not as harmful as previously thought.

In addition, the headline implies that this study is relevant to the general population when in fact it only applies to postmenopausal overweight and obese women.

What kind of research was this?

This was a randomised controlled trial investigating the effects of dietary or exercise weight loss interventions on body composition and other biological measures. In particular, it aimed to see whether adherence to weight loss interventions and their effects would differ in women with and without a history of weight cycling. 

A randomised controlled trial is a type of study that compares the effects of an intervention with another intervention or a control (such as a placebo). Participants are randomly allocated to which of these they receive. This is the best type of study design to determine the effects of a treatment.

The researchers say that no other studies have looked at this question before.

What did the research involve?

Between 2005 and 2009, the researchers recruited 439 postmenopausal women aged between 50 and 75 years to their study. To be eligible, the women had to be overweight or obese (considered as a body mass index [BMI] of more than 25kg/m2, or more than 23kg/m2 for Asian-American women). Women were excluded if they:

  • did more than 100 minutes of moderate activity a week
  • took medications for diabetes or had a fasting blood glucose of more than 7mmol/L
  • had a history of breast cancer or any other serious medical condition
  • were taking hormone replacement therapy or weight loss medications
  • were a current smoker or drank more than two alcoholic beverages a day

Eligible women were then randomly assigned to one of four groups for a period of one year:

  • reduced calorie diet intake only (n=118)
  • moderate to vigorous intensity exercise only (n=117)
  • reduced calorie diet intake plus moderate to vigorous intensity exercise (n=117)
  • no intervention (n=87)

To assess weight cycling study participants were asked at the beginning of the study: "Since you were 18 years old, how many different times did you lose each of the following amounts of weight on purpose (excluding pregnancy or illness)?" Various weight ranges were provided ranging from 5lb to more than 100lb and possible responses were from one to more than seven for each amount of weight loss. Women who reported losing more than 20lb on three or more occasions were considered by the researchers to be "severe weight cyclers". Women who reported losing between 10lb and 20lb on more than three occasions were considered to be "moderate weight cyclers" and all other women were considered to be "non-cyclers".

The researchers measured changes in body composition known to have an effect on health, such as:

  • weight
  • blood pressure
  • blood sugar
  • insulin levels
  • various other blood chemicals, such as C-reactive protein (a protein associated with inflammation)

They then analysed their results using statistical methods adjusting for BMI.

What were the basic results?

Overall, 24% of women met the criteria for moderate weight cycling and 18% met criteria for severe weight cycling. At the start of the study, these women weighed more and had less favourable metabolic characteristics than women considered to be non-cyclers.

The key finding of this study was that after one year, participants considered to be weight cyclers (moderate and severe), did not significantly differ in most of the body composition measures compared with those considered to be non-cyclers. In addition, adherence to the different study groups did not significantly differ across groups.

How did the researchers interpret the results?

The researchers conclude that a history of weight cycling does not impede successful participation in diet or exercise interventions or alter their benefits on body composition and physiological outcomes.

In discussing the research findings, lead researcher Dr Anne McTiernan says: "A history of unsuccessful weight loss should not dissuade an individual from future attempts to shed pounds or diminish the role of a healthy diet and regular physical activity in successful weight management".


Overall, this study provides evidence to support the use of well structured weight loss interventions for women known to have a history of repetitive weight cycling.

There are some limitations to this study, some of which the researchers note:

  • there is no standard definition of weight cycling, which may make comparisons to other studies difficult
  • weight cycling was determined by self-report, which can make results less reliable
  • there were difficulties in differentiating between unintentional weight loss and intentional weight loss of participants, which may affect the results
  • it is possible that women did not accurately report their exercise and dietary behaviours correctly, which also could affect the accuracy of the results (although the researchers did ask participants to record their activity or dietary intake and wear a pedometer for one week).

In conclusion, these findings may help to encourage overweight and obese postmenopausal women to reach a healthy weight through following structured weight loss interventions. The study does not provide evidence that yo-yo dieting is effective in helping you lose weight.

NHS Attribution