When NOT to Trust a Food Label?

The FDA recently shut down Clifton N.J.-based Butterfly Bakery “for unlawfully distributing mislabeled food products, such as muffins and snack cakes.” The action was based on findings that samples of Butterfly Bakery products labeled as “sugar free” contained sugar. Additionally the fat content of the samples was also significantly higher than what the label claimed. The bakery had been previously warned by FDA about problems with mislabeled products in May of 2011.

Federal law requires that the Nutrition Facts label accurately reflects the nutritional profile of the food product. The bakery has to fix its mislabeling problems before the FDA will lift the injunction. In order to comply the company can either change the label, reformulate the food-items, or both.

You may have never thought about questioning the accuracy of a Nutrition Facts label. The truth is however, that we all depend on the honesty of businesses to accurately label their food products. Why? Because according to Federal regulations, “There is no required oversight of Nutrition Fact information for packaged food BEFORE [emphasis added] distribution for sale.” [i]

What’s more, it’s FDA’s Food Safety division that is tasked with nutrition label oversight, a division that has long been under-funded and must prioritize its resources based public health risk. Mislabeling issues that pose risk include foods that fail to label known allergens (i.e. peanuts, eggs, shellfish, etc.), or certainly if they claim to be sugar-free (and therefor appealing to diabetics). Mislabeling that makes products appear to have less calories, fat, carbs etc. (errors that don’t pose a health risk), are unlikely to get FDA attention.

It should be no surprise then that more often than not it’s a consumer, or consumer watchdog investigation, that brings a potentially mislabeled product to the attention of the FDA.

You’ll notice that Butterfly Bakery had a detailed warning from FDA almost two years ago--time enough one would think to get things right. I saw a similar situation in 2001 with a cookie company is Bellingham WA (Baker’s Breakfast Cookie; Erin Baker's). Several independent laboratory analyses showed that their cookies were mislabeled and actually had 100- to 200-calories more (per cookie) than the labels said. Needless to say the cookies were flying off shelves and the company could barely keep up with demand. The problem with Baker’s labels was twofold: the cookies weighed a half ounce more than they were supposed to (3.5 instead of 3 ounces), and the calories-per-serving (3 ounce cookie) was also understated by around one third. Two years after Baker's was first informed of their mislabeling problems (and after reformulating and relabeling their cookies) their labels were found to still be significantly off in analyses the Seattle Times had done (the two types of cookies analyized still had 40% more calories than the label claimed). Today Erin Baker’s cookie’s labels have the lowest calories by weight (per gram/ounce) of the dozens of cookies I checked on www.calorieking.com. [See Calories of Baking Ingredients for more insight on calories in baked goods.]

Over the past 20 years only a handful of studies have looked at the accuracy of food labeling.  While most labeling is accurate, it's not too hard to find problems.

Most recently (2010) researchers reviewed 29 restaurant entrees/side-dishes and 10 grocery store frozen foods. The study focused on items labeled as low-calorie. “The measured energy [calorie] values of the restaurant foods averaged 18% more than their stated values. So a 500-calorie entrée would actually ring in at 590-calories. However, some individual restaurant items contained up to 200% of stated values and, in addition, free side dishes increased provided energy to an average of 245% of stated values for the entrée they accompanied.” The measured calorie content of the supermarket-purchased meals was greater than stated values, but only by 8%.[i]

Understated item weights—automatically lead to understated calories for single serving baked goods (cookies, muffins and brownies). This may well be the most common mislabeling issue. “Nearly one third (6 of 19) of the sampled packages weighed at least 1 ounce more than was stated on the label” in a small 1995 study of single-serving baked products.[ii]

Since these kinds of items run between 100 and 150-calories per ounce the error introduced when you’re trying to accurately track your calorie intake is frustrating—especially if it’s an item that you are eating regularly.

Federal law allows a food item’s actual weight to exceed what’s stated on the food label by 20% (allowable range: 99% to 120%). Legislators believed that companies would naturally stick closely to the stated weight (as a matter of cost savings).  

In another small study researchers analyzed 20 national brand foods, 12 regional brands and 8 locally produced products. The national brands averaged just 2.18% more calories than stated in their labels. Regionally distributed foods averaged 25.2% more energy than labeled and the locally produced items averaged 85.4% more energy than stated.[iii]

Based on this limited data it seems likely that you can generally trust the nutrition information on national brands—their actual content tends to be very close to what the labels say. This makes sense in that big operations, with quality controls (that carefully watch expenses) produce these items.

On the flip side, inaccuracy is higher in regionally produced and labeled brands and worse still in locally labeled foods likely because their production operations (like restaurants) are more human-labor and less mechanized.

So what’s a consumer interested in accurate calorie counts to do?

1. Be skeptical—if the calorie count on a food item seems too good to be true it very well may be! Compare it to other similar foods (www.calorieking.com).
   
   
2. If you buy single serving baked goods weigh them (using a digital kitchen scale) to see if the actual weight matches the label. If it’s over the labeled weight than it’s more than one serving. Next check to see if the calories per serving are plausible: divide the stated calorie amount by the serving size (usually this is in grams). Muffins and cake should be about 3.5 calories per gram (so about 100 calories per ounce). Cookies or brownies should be about 4.9 calories per gram (so about 140 calories per ounce). [1 ounce = 28.35grams]
   
   
3. In restaurants make sure you account for all extra sides, beverages, bread etc. in addition to the calories of the entrée.
   
   
4. Practice makes you a better estimator when you need to be. The more you weigh, measure and track your intake the better you will be at accurately estimating portion sizes. Most of the error in tracking calorie intake comes from underestimating the portion size, or forgetting things (sides, snacks or beverages, etc.).

Your questions or comments are welcome!
-Dorene

Of Further Interest: Some Media reports of mislabeling

2013 – How Accurate are chain restaurant calorie counts? [added 27APR13]

2013 – Potato chips mislabeled Gluten Free…

2013 – Mislabeled fish a national problem

2011 – With Faulty Food Labeling, Who's Minding the Store?

2010 – Nutrition Label Accuracy

2008 – Nutrition Bars (Energy Bars, Fiber Bars, Protein Bars, Meal Replacement Bars, Snack Bars and Whole Food Bars).

2005 – Consumer.lab.com Reports on Nutrition Bars: Highlights Major Differences, Inaccuracies, and Urges Consumers to "Know Your Bar".

2003 – Seattle Times – Cookies’ true calories may dismay diligent dieters.

2003 – Heart Associatioin Praised for 'Laura's Lean Beef' Crackdown

2002 – Woman Seeks $50 Million in Damages Caused by Mislabeled Fat Content.

2002 – Can you judge a bar by its wrapper? J Am Diet Assoc. 2002;102(2):180.

2002 – Low-Carb Bars Aren’t What You think They Are

2001 – Sixty percent of nutrition bars fail to meet claims in Consumerlab.com tests—“Low Carb” Bars Often Loaded with Carbohydrates; excess sodium and Saturated Fat also found.

2001 – I Scream, You Scream… Our Position: It’s Too Bad That The Ice Cream Was Too Good To Be True. Orlando Sentinel; June 20, 2001.

1996 – The Case of the Missing Calories. Vegetarian Times. 1996 Oct. pg. 32-34.

1996 – FDA found food labels mostly accurate

1992 Consumer Reports – Lowfat frozen desserts: better for you than ice cream? 1992;57(8):483-487.

1992 CBS This Morning – Special segment: Study of soft-serve frozen dessert. Diaz A. CBS September 19, 1992.

1989 New York Times – Soft-serve desserts: how low in calories? Burros M. NYTs August 9, 1989:C-1.

---

[i] Urbain LE, et al. The Accuracy of Stated Energy Contents of Reduced-Energy, Commercially Prepared Foods. J Am Diet Assoc. 2010;110:116-123.

[ii] Young LR, et al. Food labels consistently underestimate the actual weights of single-serving baked products. J Am Diet Assoc. 1995;95(10):1150-51.

[iii] Allison DB, et al. Counting Calories—Caveat Emptor. JAMA 1993;(270):1454-56.