One more thread on diet, summarised very well by Joanna on email:

I have dieted a lot of my adult life and find that now I have to eat far less than the recommended 2000 calories in order to maintain my weight. In fact if I ate 2000 calories a day I’d put on weight rapidly, unless I was in training for a marathon!

I do feel that government guidelines need to change.

A few years back the government’s Scientific Advisory Committee for Nutrition was asked to review the widely-used targets of 2,000 kcal/day for women and 2,500 kcal/day for men. I reported on it at the time for the Observer (here). I’m guessing from the lack of uproar and despair in the desert aisles that nothing came of this, but I’ll try and find out why before we close for the day.

This is by far the most controversial area, judging by the responses so far.

First up, do we massively under-report what we eat, as my colleague Jonathan Haynes suggests?

The dietry figures come from the National Diet and Nutrition Survey, now a rolling annual programme in which people do a wide ranging interview about their eating habits and other lifestyle factors, and keep a seven day food diary, including weighing everything they eat and drink in and outside the home.

The problems of such self-reporting are widely acknowledged, though, says the H&SCIC:

Under-reporting can cause biased low estimates of intake as respondents under-report their actual intake or modify their diet during the recording period....

Analysis of data from the NDNS adults 2000/01 indicated that energy intake could be under- reported by about 25%.

The 25% figure feels a bit stab-in-the-dark (if the reserachers disagree, please get in touch). But it still doesn’t seem to explain everything: if reported calorie intakes are increased by 25%, that takes average consumption to 2,688 kcal/day for men and 2,017 kcal/day for women. The former is notably above the average recommended of 2,500, but the figure for women is only fractionally above the 2,000 official suggestion.

That leads us to the questions: are all calories the same when it comes to weight gain?



Your suggestions raise questions over dairy products, high sugar and fat contents, the quantity and quality of carbohydrates recommended, the hidden calories in alcohol and other drinks, even...

This seems to be where the intelligent debate about obesity and other food/health issues is happening.

One expert several of you suggested is Robert Lustig, the American professor of paediatrics, who has done so much to popularise the idea that not all calories are equal, and in particular to campaign against the huge quantity of sugar in modern diets.

Last year Lustig and colleagues published a paper in the journal PLoS One, in which they undertook a massive global study of diet, which he describes in the above blog in the Huffington Post:

We asked the question, “What in the world’s food supply explains diabetes rates, country-by-country, over the last decade?” We melded databases from the Food and Agriculture Organization (FAOSTAT), which measures food availability, the International Diabetes Federation (IDF), which measures diabetes prevalence, the World Bank World Development Economic Indicators, and the World Health Organization Global Infobase. We assessed total calories; meat (protein); oils (fat); cereals (glucose); pulses, nuts, vegetables, roots, and tubers (fiber); fruit excluding wine (natural sugar); and sugar, sugarcrops, and sweeteners (added sugar). We controlled for poverty, urbanization, aging, and most important, obesity and physical activity.

This study focused on diabetes, and but the science of why sugar calories are so different raises interesting questions for the study of weight gain:

for every extra 150 calories per day, diabetes prevalence rose by only 0.1 percent. But if those 150 calories per day happened to be a can of soda, diabetes prevalence rose 11-fold, by 1.1 percent.

I’d love to hear of more research on this subject.

Here we might have found the answer to today’s question. Might.

The report headlines activity levels based on the annual Health Survey for England (HSE), which asks people to record their physical activity in the four weeks prior to an interview. This is the “primary” source of official information on activity levels.

Even the report admits, though, that it appears to be deeply flawed. It cites another study by the HSE in 2008 which fitted people with accelerometers to record the “duration, intensity and frequency of physical activity for each minute”.

This study found a shocking 6% of men and 4% of women reached the government’s recomended levels. I keep re-reading this sentence in the report and here on the blog, but it seems that really is what it says.

I’ve found that 2008 report, here, so I’ll have a look in more detail. Self-reporting exercise seems intuitively unreliable, but it would be interesting to know how many people were involved in the study. Also, were most people just below the recommended level, or way off? Worryingly, experts tend to think that people don’t only over-report activity, but act differently when they are being monitored. I know I would...

On Twitter @dan_running challenges whether the obestity figures are a meaningful reflection of the health of the nation:

Today’s publication does address this.

Like most reports its headline figures draw on classifying people as obese or overweight based on their body mass index (calculated by dividing a person’s weight in kilos by the square of their height in metres).

However it recognises that:

Although BMI allows for differences in height, it does not distinguish between mass due to body fat and mass due to muscular physique, or for the distribution of fat.”

For this reason it also includes figures for waist measurements: more specifically the number or proportion of people with “raised waist circumference” of more than 102cm for men and 88cm for women. By this measure, one third of men (34%) and nearly half of women (45%) are judged too large.

For all it’s clumsiness, however, the BMI measure appears to be a useful guide to what really matters about all this: whether or not a person is likely to suffer ill health as a result of carrying too much weight.

In Chapter 7, the report summarises these risks:

Obesity is a major public health problem due to its association with serious chronic diseases such as type 2 diabetes, hypertension (high blood pressure), and hyperlipidaemia (high levels of fats in the blood that can lead to narrowing and blockages of blood vessels), which are major risk factors for cardiovascular disease and cardiovascular related mortality. Obesity is also associated with cancer, disability, reduced quality of life, and can lead to premature death.”

For example, a woman who is obese is considered 13 times more likely to develop type 2 diabetes than a woman of “normal” weight. When somebody with more technology training than my two hours comes along I will try and post the report’s arresting table on this.