Network News

X My Profile
View More Activity

The Problem With Food Diaries

As part of my research for an upcoming Food section column, I've been spending some time with David Cutler, Ed Glaeser and Jesse Shapiro's excellent paper exploring the rise in obesity. More on the main conclusions to come, but this bit on the difficulty of using food diaries — literally, diaries where people write down what they eat so researchers can get an idea of consumption patterns — is pretty entertaining.

Detailed food diaries are available for 1977-78 and 1994-96 from the Continuing Surveys of Food Intake by Individuals, conducted by the U.S. Department of Agriculture. In principle, all food consumption is recorded. In practice, however, consumption is surely understated. People do not record everything they eat, and the act of keeping a diary lowers consumption for some people. Evidence of this underreporting is seen in average caloric intake recorded in these surveys. The average male in 1994-96 reports consuming 2,347 calories — corresponding to roughly 106 lbs in steady state. The average female reports caloric intake of 1,658 calories, consistent with a steady-state weight of 64 lbs.

Now, if only we could self-report our weights, too...

By Ezra Klein  |  August 25, 2009; 2:23 PM ET
Categories:  Food  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: The Coming Budget Battle
Next: Don't Like Michael Steele's Position on Medicare? Wait a Day.


Something's wrong with that quote -- 2000 calories a day is at least pretty close to steady-state for an average adult, and while 1600 is a dieters number, it's not terribly extreme.

Posted by: LizardB | August 25, 2009 3:57 PM | Report abuse

What? All the food packages tell me that 2000 or 2500 calories is the standard daily intake. I'm 200 lb and would be quite healthy at 185 (although BMI suggests 174.) Based on a linear extrapolation of the above, though, that suggests I should be eating 3600 calories a day? I'm eating less than that but I'm sure not losing weight (maybe it's not linear)?

Posted by: _SP_ | August 25, 2009 4:01 PM | Report abuse

Those calculations must be wrong - the only way for those to be steady-state weights for a the given calorie intake is if the person is far more active than . . . anyone.

If you work backward from a BMR formula, A sedentary 30-year-old man of average height who consumes 2347 calories a day will be about 195 pounds; a similar woman who consumes 1658 calories a day will be about 130 pounds.

Posted by: dcamsam | August 25, 2009 4:19 PM | Report abuse

Food diaries can be helpful in losing weight by forcing one to calculate calories in dishes. It also forces one to think about what one eats, knowing it has to be written down. But it is tedious, and backsliding is easy when one quits. Losing is still hard.

For people who don't cook, posting calorie counts in restaurants is an incentive to keep calories, fats and sugars down.

Posted by: Mimikatz | August 25, 2009 4:44 PM | Report abuse

I must admit being chastised for under-reporting my own size. As a result, though, I've implemented a personal regimen in which I nibble only on those things I enjoy the most, engage in appropriate calorie-burning exercise thereafter, allow periodic measurement by skilled personnel, and smile while doing all of the foregoing.

At least that regimen has a good first step... at this time of day, I don't need much encouragement to go for a snack.

Posted by: rmgregory | August 25, 2009 4:55 PM | Report abuse

Huh? How would that work? I'm trying to lose weight right now and I'm way under 1658 calories to lose just a pound a week. I imagine 1658 is what I would eat just to maintain my current weight.

Posted by: tracy2 | August 25, 2009 5:39 PM | Report abuse

Yeah, I don't know where these economists got their steady-state weight numbers, but they are way off. A man can maintain a weight of 185 lbs on just 2347 calories, if they are sedentary. Their exercise adjustment has to be way off.

Posted by: PatrickEarnest | August 25, 2009 6:05 PM | Report abuse

A clarification needs to be made here. The CSFII did NOT use food diaries (I work for the statistical agency that puts out NHANES, the survey which was used in the Harvard paper). Instead they used 24-hour recalls. These two things are very different in the world of nutritional epidemiology. Food diaries are prospective records kept by an individual, as Ezra aptly explains above. 24-hour recalls are done by trained interviewers who retrospectively ask individuals about their food consumption in the previous 24 hours. I'm very surprised and concerned that the authors of the Harvard paper would have classified the CSFII intake measurements as food diaries. It makes me believe that they do not truly understand some of the basic principles of nutritional epidemiology which adds doubt to a lot of their findings. I also agree with the previous comments regarding the confusing statements that were made about steady state weight. These statements make little sense as they stand and I think the authors have an obligation to explain these in layman's terms.

Posted by: amyg8r | August 26, 2009 9:54 AM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company