Study finds obese people may struggle to reach a healthy weight

"Obese men have just a '1 in 210' chance of attaining a healthy body weight," The Independent reports. This was the findings of a study that used a GP records database to look at body mass index (BMI) measurements of almost 300,000 people recorded over a 10-year period.

Overall, it found that low proportions of people in the obese categories achieved a normal weight in subsequent measures – only 1 in 210 for men and 1 in 124 for women who had BMIs of 30 to 35, and much lower than that for the higher BMI categories.

However, this should not be interpreted to mean that if you are obese, you should give up trying to lose weight. Moving from "very obese" to a "normal weight" category may not be realistic, particularly in the short term, and achieving steady weight loss may be a better goal. 

Encouragingly, much higher proportions of obese people in this study were able to achieve 5% or more weight loss (around 1 in 5 to 1 in 10 people). Even a modest reduction in BMI can bring important health benefits.

Ultimately, without knowing the wider health and lifestyle circumstances of the individuals in this study, it is not possible to identify what aspects of obesity management may be less effective.

Rather than opting for the latest fad diet, try the NHS weight loss guide: our popular free 12-week diet and exercise plan.

Where did the story come from?

The study was carried out by researchers from King’s College London and the London School of Economics and Political Science, and was funded by the UK National Institutes for Health Research.

The study was published in the peer-reviewed American Journal of Public Health, and the article is open-access – meaning it is freely available to read online or download as a PDF.

The media has reported the findings of this research accurately, but it may have been beneficial to discuss some of the wider contextual issues. For example, it may have been helpful to explain what the figures meant, rather than talking of them as "chances" for someone with obesity to lose weight. That is, they were the proportions of people in each category who had attained a normal BMI each year.

The reporting also didn’t make clear that many people included in the study may not have been trying to lose weight, which could change the assessment of how effective weight loss interventions were. 

What kind of research was this?

This was a population-based cohort study, which followed a sample of obese men and women for 10 years to look at what proportion managed to achieve a normal body weight.

Obesity and being overweight are global problems, and finding effective ways to tackle them is an important ongoing issue. Obesity can trigger a wide range of complications, such as type 2 diabetes, heart diseasestroke and non-alcoholic fatty liver disease.

This study assessed how often either a 5% reduction in BMI, or attainment of normal BMI, happens in the general UK adult population.

What did the research involve?

This study obtained medical records from the UK general practice database, Clinical Practice Research Datalink (CPRD). Over the 10-year period 2004 to 2014, over 2 million adults (aged over 20) had their BMI recorded on three or more occasions.

People were grouped according to their BMI:

  • normal weight: 18.5 to 24.9 kg/m2
  • overweight: 25.0 to 29.9
  • simple obesity: 30.0 to 34.9
  • severe obesity: 35.0 to 35.9
  • morbid obesity: 40.0 to 44.9
  • super obesity: 45.0 or greater

From each category, a random sample of 30,000 people was taken, and they obtained their full medical records. After excluding those who had weight loss surgery (also called bariatric surgery), they had a final sample of 278,982 people. They then analysed changes in their BMI over the study period, from the first recorded measurement, looking for those who attained normal weight or achieved a 5% reduction in weight. This 5% weight loss outcome was chosen because it is a realistic target that is often recommended to people who are obese and trying to lose weight.

What were the basic results?

The average age of the people studied was 55 for men and 49 for women. There were larger numbers of obese women than men. For men, there were around 25,000 (19%) with their first BMI measurement in the normal weight category, then around 27,000 (about 21%) in each of the overweight to severe obesity categories, 14,767 (11%) in the morbid obesity category and 6,481 (5%) in the super obesity category. For women, there were 23,640 (16%) in the normal weight category, then 26,000 to 27,000 (around 18%) in each of the overweight to morbid obese categories, and 18,451 (12%) in the super obese group.

When looking at the proportion of people showing no change in their BMI during follow-up, this was greatest in the normal weight category (men 57%, women 59%).

Only 14% of men and 15% of women showed decreases in their BMI category without also showing increases. Around 1 in 5 people in the morbidly and super obese groups showed decreases in their BMI, which was the highest rate seen. More than a third of people overall showed weight cycling – both increases and decreases in BMI. This was also highest in the severely obese category, where around half showed weight cycling.

During the total follow-up period, 1,283 men and 2,245 women who were obese attained a normal BMI. Overall, this represented about 1 in 60 men and 1 in 44 women over the entire period. However, to account for people being followed up for different lengths of time, the researchers calculated these numbers for each weight category for one year of follow-up.

The probability of achieving a normal BMI from each starting BMI category over one year was:

  • simple obesity: 1 in 210 for men and 1 in 124 for women
  • severe obesity: 1 in 701 for men and 1 in 430 for women
  • morbid obesity: 1 in 1,290 for men and 1 in 677 for women
  • super obesity: 1 in 362 for men and 1 in 608 for women 

The probability of achieving a 5% reduction in weight over a year was higher:

  • simple obesity: 1 in 12 for men and 1 in 10 for women
  • severe obesity: 1 in 9 for men and 1 in 9 for women
  • morbid obesity: 1 in 8 for men and 1 in 7 for women
  • super obesity: 1 in 5 for men and 1 in 6 for women 

However, this 5% weight loss was often accompanied by weight cycling and gains of over 5% in weight at other times.

How did the researchers interpret the results?

The researchers conclude that, "the probability of attaining normal weight or maintaining weight loss is low". They go on to say that, "obesity treatment frameworks grounded in community-based weight management programs may be ineffective".


This research makes use of a general practice database providing just under 10 years of BMI observations for a large, nationally representative UK sample.

It demonstrates that low proportions of people in the obese categories were able to achieve a normal BMI over a year of follow-up, and the common problem of weight cycling. However, there are points to consider when interpreting these results:

  • The probability of obtaining a normal BMI over a year was very low: only 1 in 210 for men and 1 in 124 for women in the "simple obese" category of 30 to 35kg/m2, and much lower than that for the higher categories. However, these particular figures are only the proportions per year, and we don’t know how many people were attempting to lose weight, or how they were trying to do this. Other analyses showed a better picture – for example, about 1 in 5 obese people managed to reduce their BMI by at least 1 point and not increase it during follow-up.
  • Though the people in this study had 3 or more BMI measurements, we don’t know how long after the first measurement they were taken. Despite the study period being 10 years, the BMI measures may have been over the period of only 1 or 2 years. Reaching a normal BMI may not be a realistic goal in the short term, particularly if a person is in the severe to super obese categories. Achieving steady weight loss may be a better goal and, encouragingly, much higher proportions were able to achieve 5% or more weight loss.
  • The researchers conclude that community-based weight management programmes may be ineffective. However, some care must be taken in concluding that weight loss programmes do not work, because this study only has data on BMI changes. We don’t know anything about the wider health problems or lifestyle circumstances of any of these individuals, or know what care they may have been receiving. As such, we are not able to identify what aspects of obesity management may be ineffective or require a change. We can only say that many obese people did not lose weight.
  • Finally, though this study is a nationally representative sample, there are some exclusions that could influence results. The study excluded people who had received weight loss surgery. These people are likely to have achieved weight loss, and would likely have been in the more severe obesity categories. This may mean that these proportions do not give a true indication of the proportions in the severe obese categories who achieve weight loss. Also, as the researchers acknowledge, weight change may be different in the people who had fewer than two BMI measures over the course of the study; a group who were also excluded.

Nevertheless, this study highlights the growing obesity problem and the need for effective strategies to help people lose weight. If you are obese and trying to lose weight, you should not be discouraged by these results. Eating healthily and exercising have health benefits, even if you do not lose weight, and losing even small amounts of weight and keeping it off in the long term is likely to be beneficial.

NHS Attribution