Where does the nutrition information come from?

[teresa2711]

I'm curious.......what is the source of all the nutrition information in the databse?

I use the Australian database, but the questions is relevant for all of them.

I assume that when the food is a branded product, the information is taken from the manufacturer. But what about fresh, generic food? Can we be confident it is accurate? I have my doubts, as often foods will be listed more than once, with different nutritional info.

It woudl be great if one of the Fitbit team could respond to this question, or maybe point me in the direction of an article that explains it all.

Thanks!

[Ukase]

Teresa, 

  You ask a great question. 

"Fresh Generic food" - I'm not sure what you mean by this term.

Let's take fresh fruit for example. A banana has about 35 calories per ounce. Where do I get this? I get it from a nutritional database. The real question is  - Where do "They" get the information from? I suspect that the information stems from a combination of places. The database I use most often is http://nutritiondata.self.com 

When I look at the specifics of a certain food, at the bottom, I see this: 

Source: Nutrient data for this listing was provided by USDA SR-21

Further clicking and Googling about reveals a nifty FAQ. 

FAQ

If I read between the lines, just skimming over this vast amount of material, it looks as though some fellow named Atwater ran a bunch of tests and reached various conclusions about the amount of heat required to burn various foods until they were transformed from food into heat. That heat was measured in terms of KiloCalories - which we call calories, in a generic sense. 

So - the truth is - if you want to get OCD and calculate every morsel you ingest, and compare it with every energetic expenditure - Good Luck! Both the Fitbit, and the information in all the databases are proximally close ballpark guesses. A calorie is not a calorie. All bananas are not created equal. Some get more water than others as they grow and ripen. As such, their composition is largely the same - but will have variance - and that variance may not be by weight, but by fruit. 

The same with beef, chicken, other meats - the animals the meat comes from eat different things, consume different amounts of liquids, exercise a different amount. As a result, their composition of muscle, bone, sinewy fibers, etc - will be different. One pound of chicken breast is not the same as another pound of chicken breast. It's close - but not the same. How could it be? 

I'm not suggesting the variation is off by a significant amount, mind you. But, if you're OCD, you're going to be disappointed. The math just doesn't work out. 

For best results - use one of the formulas for calculating your BMR, or use what Fitbit calculates. Use the databases information, assume they're as close to accurate as anyone is going to get in the near future. And, for weight loss, eat less than your BMR. For weight gain, eat more. Write it all down, check the scale after a week. If you've not lost, and that was your goal, eat even less, or move around that much more. Check back after a week. If you've still not lost, continue reducing your intake until you do lose. 

Then, you've established what your basic BMR is and how much you can consume and still lose weight. 

Hope that info is helpful. 

[teresa2711]

Hi Ukase,
Thanks for your comprehensive response!
As a medical scientist, I fully understand the science behind how the calorific value of food is calculated, I am also pretty familiar with that causal relationship between BMR, calorie intake, energy expenditure and weight loss/gain.
My question was prompted by an interest in the reliability of the nutrition information. I hope I'm not OCD ☺️, but I do think it's important to know where information comes from, so I can make my own informed judgement about how reliable it is. At MFP, for example, they clearly indicate whether information is provided by a member or if it has been verified. I was curious if at Fitbit there is any validation at all, or whether it's a free for all and anyone can add whatever they like. I don't expect Fitbit to personally experimentally prove the caloric value of every food, I'd just like I know where the information comes from ☺️
And by fresh generic food, I mean for example fresh fruit, veggies or staples such as rice, milk, etc.

[Ukase]

Ah! Well, then...I have no answer for you. 

 I know if you see any thing in the database that's "John's Brand" - I put it in there. Painstakingly weighed out each ingredient and looked up the caloric value by weight and did the math. 

Other than that - I have no idea. I don't think that database is "user-generated", or people wouldn't be complaining about the lack of foods in it. So, we'll have to wait for a FitBit big shot to let us know. 

[Joni63]

My fitness pal "verified" tag is far from.  The verification is obtained from site users, not MFP.  And there are trolls everywhere!  Enter Sara Lee 100% whole wheat bread and click that you ate one slice of the "verified" version, then check your potassium intake on the nutrition page lolol 7000 mg per slice...........

So, my answer to your question is, MFP is a great data base and I use it, but I also check the entries, especially when my nutrition at the end of the day just looks wrong.  Sometimes, the verified version didn't bother adding in the sodium.  I am more trustworthy of an entry that says corrected after it than the little green check mark though. 

On fresh food, I figure it's all close enough though, but you could always use a data base you trust, enter the data placing the items in your meals or recipes section, then use entries you feel better about using!

[ZeljkoManojlovi]

If you have a chance watch documentary "Fed up"... 😉

All info that you get is written by government, sponsored from food industry..