Thursday, November 11, 2010

Apparently nobody's good at keeping food records.

[My heartfelt thanks to all those brave men and women who gave their lives for our freedom]

Food diaries are the mainstay of any behavioural weight management program worth its salt, but are people actually good at keeping them?

We use food diaries at my offices and there's no doubt some folks are better than others, but given the incredible access we have to food, the lack of caloric information in most restaurants, the reluctance of most folks to weigh and measure what they're eating, and the nibbles and bites we may randomly have, accurate record keeping's tough.

A study done nearly a decade ago helps to illustrate how tough.

Simply designed, the study compared the calories recorded in 7 day food diaries with the calories inferred by doubly labeled water testing. The extra wrinkle with this study? Half of the subjects were registered dietitians while half were plain old folk, with the assumption being that the dietitians would be record keeping ringers, especially given the fact that the study was presented to them as a challenge to record their dietary intake with enough accuracy so as not to deviate from the estimate obtained via doubly labeled water.

The results?

First the non-dietitians. Despite actually keeping food records they missed 18% of the calories they consumed. Over 40lbs a year worth.

Now for the dietitians who were trying their best to be perfectly accurate. They missed 10.5% of the calories they consumed. Due to a small sample size, this wasn't deemed statistically significant, but over a year were that discrepancy maintained, they'd have missed 22lbs worth of 'em.

At my office we sometimes ask patients to do what we call, "science experiments" where for a one or two week period they become as anal about record keeping as they possibly can. Weighing and measuring everything. Recording food within minutes of eating it and ideally, not eating out even once. Almost invariably patients who felt they were fairly accurate with their records before the experiment find anywhere from 200-500 calories they didn't even realize were there.

Of course just because something's challenging to do, doesn't mean a person shouldn't do it and if you're trying to lose weight, there's no doubt keeping a food diary has been proven to help.

Remember, food diarizing isn't for judgment. It isn't there to tell you if you've been good or bad, or how much you're allowed, or how much room is left for dinner. What a food diary's for is information, because the more information you have before you make any decision in life, the better that decision's going to be, and knowing how many calories you're having will certainly help you in deciding which ones are worth it and which ones aren't.

Champagne CM, Bray GA, Kurtz AA, Monteiro JB, Tucker E, Volaufova J, & Delany JP (2002). Energy intake and energy expenditure: a controlled study comparing dietitians and non-dietitians. Journal of the American Dietetic Association, 102 (10), 1428-32 PMID: 12396160

Bookmark and Share


  1. Anonymous8:39 am

    You mentioned for before, and they've recently made some upgrades to it. I REALLY find this helps as a food record - I set up an anonymous account, and whenever I eat something, I just have to press a couple buttons on my non-smart phone, type what I ate, and voila! - it's logged online for me.

    It makes it really easy to keep real-time accounts of what you eat. Also, there is something called CrowdCal, which is a crowd-sourced calorie tracker. Any time ANYTHING is submitted into the food database, CrowdCal adds it to the database. Then, if I am to add an apple, but am not sure how many calories it has, I can ask for suggestions, and will see that 40 people have said it is 50 calories, 1 person has said it is 1 calorie, and 1000 people have said it is 75 calories... so I'd go with 75. Not pure science, but it certainly is helpful for people who are in a hurry but care to keep track.

    It's definitely worth looking into, in my opinion, readers! Free, simple, and effective as you make it.

  2. Interesting. This may answer some questions I've been having about my own weight management. I count calories and keep a food diary (I use the LoseIt app on my iPhone), and it seems that my daily food budget I've had to stay within (both for slow, steady loss and more recently for maintenance) was kind of low -- lower than what it ought to be.

    Now I'm thinking that, even though I'm keeping a food diary, I've actually been eating more calories than I thought I was, and that on a calorie budget of 1550, I may have actually eating more like 1700 or even 1850. I do measure and weigh my food when I prepare it at home, and I do record what I eat as soon as I eat it... but I do eat out sometimes, and sometime get food at bakeries/ food stands/ other places where I have to do some guesstimating on calories. And even the calorie counter isn't always that helpful. (70 calories for a chocolate chip cookie. Gee, that's really helpful, What *size* cookie, please?)

    I do have one question, though. Why do you say that a food diary "isn't there to tell you... how much you're allowed, or how much room is left for dinner"? That's exactly what I use it for: to plan my day's eating, to figure out what foods and quantities will and will not keep me within my daily calorie budget. If I have 500 calories left in the evening, I eat differently than if I have 300 calories left. Why do you advise against that?

  3. Because it's too judgmental.

    The goal's the smallest number of calories you need each day to be happy, not to come in under a glass ceiling of calories.

    Some days simply are worth more calories than others.

  4. "Because it's too judgmental."

    Hm. I don't experience it as judgmental. I set my own calorie budget, based on what works for me to lose weight at the rate I want to, and to maintain the weight that I want to. Nobody's judging me -- I'm making these decisions for myself. I don't experience it as judgmental at all. I experience it as empowering.

    "The goal's the smallest number of calories you need each day to be happy, not to come in under a glass ceiling of calories."

    I see what you're saying. The problem with that for me, though (and I suspect for a lot of other people), is that it'd be too easy to delude myself, and to rationalize decisions that will mess up my health. One of the whole points of conscious weight management is that my "natural" instincts about food are pretty broken. I deluded myself for years into thinking I was happy and healthy at 5'3" and 200 pounds, and that I couldn't possibly be happy eating any less than I was. (And the other side of that is the specter of undereating and anorexia. If the goal is the smallest number of calories I need each day to be happy, then I could easily see myself deluding myself into thinking I'd be happier if I ate just a little less and lost just a little more weight... and then lost just a little more... and then just a little more...)

    Of course my own judgment about what weight I'm healthy and happy with is a major part of these decisions. But I really need some external, evidence-based metrics to help me make those decisions. If I rely on my own instincts, I'm right back where I started, at 200 pounds.

    "Some days simply are worth more calories than others."

    Definitely true. And some days I do go over my calorie budget, and I'm totally okay with that. In fact, a central part of my weight management plan is to have one day a month when I don't count calories at all. But on a day- to- day basis, having a budget that I stay within most of the time has been essential to losing the weight, and to keeping it from sneaking back up again.

  5. Anonymous9:29 pm

    Great subject! I still keep a daily food journal, although I've been at goal weight for a year. I figure 10 minutes a day writing in my food journal helps me stay at goal weight - that is a bargain by any measure. I use an old-fashioned method - pencil and paper. I have a 3-month food journal. As much as anything, writing down my food keeps me *aware* of what and how much I eat. Otherwise I can easily develop food amnesia. This is just as important an aspect to me as the calorie count. As to the exactness of the calorie count, so long as I was losing weight, I figured my journaling was 'accurate enough'. The same concept applies to maintenance - so long as I am in goal range, my calorie estimates are working well enough. ~Carol~

  6. "so long as I was losing weight, I figured my journaling was 'accurate enough'. The same concept applies to maintenance - so long as I am in goal range, my calorie estimates are working well enough."

    That's a good way of thinking about it, Anonymous. The important thing is the bottom line. It could be that I'm journaling 1550 calories a day when I'm actually consuming 1800... but as long as I'm doing that consistently, and as long as the 1800 I'm consuming is the right amount for me to lose/ maintain my weight, it doesn't really matter. It'd be interesting to know for purely academic reasons -- and of course, if I were doing scientific research I'd want to know the accurate answer -- but as a practical matter, the bottom line is what counts.

  7. I was never very good with food journaling until I got a Fitbit ( Easy to enter food, and it gives a rough calculation of calorie deficits, which has been very helpful for me. So far, 18 lbs lost in ~7 weeks. Because the fitbit itself calculates activity (pedometer which includes intensity), it's been more motivational than trying to do all the calculations myself -- I want to have my counts be as accurate as possible.

    And weighing food has been key, because I want the calorie counts to be as accurate as possible. I already had a food scale, so just getting it out more often has helped a lot, too. So far, I think I've managed to do pretty well at logging what I've eaten without getting too judgmental -- I use the total at the end of the day to determine whether I should have a dessert with dinner or a little extra pasta or whatever, but mainly, I've just been trying to be as accurate as possible in recording what I'm eating.

    The fitbit site also has a lot of other metrics you can use or not, like blood pressure, pulse rate, blood sugar, etc. I'm very motivated in general to know more about what my personal statistics are, so it's been great to be able to have graphs showing what I've been eating and doing. Now if they would just open the APIs so I could suck in my data and do some more robust statistical analyses on things like nutrient consumption...

  8. Michelle Meaders12:48 pm

    "Simply designed, the study compared the calories recorded in 7 day food diaries with the calories inferred by doubly labeled water testing."

    What is "doubly labeled water testing"?

  9. You really believe that when RDs were challenged to record intake as accurately as possible they still couldn't do it?

    Come on, that doesn't pass a sniff test.

    And many weight-aware consumers are excruciatingly careful recordkeepers as well. Especially when they know they're being watched.

    Doesn't it seem far likelier that their measurements ARE accurate - but that the actual nutritional content of the food eaten is different from its stated value in official sources?

    Food log accuracy will never solve that part of the problem.

    For example:

    1) Not all oranges are created equal. Not all chicken is created equal. And official nutritional values are averages, often from many, many decades ago.

    (Not convinced? Check out the nutrition differences between turn-of-the-century chickens, pigs and cows and today's industrial chicken, pig or cow. Same goes for many fruits.)

    2) Regulators permit +/- 20% accuracy in nutrition labels compared to the actual food value contained in that package, as I recall.

    And so on.

    To be clear: Food records are valuable tools. And no argument, lots of folks are eating far more than they realize (or log accurately).

    But I think it's important to avoid the suggestion that the recordkeeper is the only causal factor. That type of "monocausal" thinking is flawed and more importantly, very discouraging to folks who KNOW they were scrupulous in their journaling.