My friend Scott, who has spent enough time on insulin to make my head spin, just vented his frustrations regarding the carb count for a meal of soup and crackers. While I've not had to become quite as precise on carb-counting as those of you on insulin, having to completely rewrite my doctor's office's 1000 mg sodium/day diet (following the flyer they gave me would have had me consuming closer to 2000 mg sodium/day) made me much more sensitive to calculating serving size than the average person with diabetes -- or the average person with hypertension, for that matter.

Carb count aside (I can see many of our readers shuddering at a 66-carb-gram bowl of soup with another 60+ carb grams of crackers -- but as Bennet says, YDMV), there are so many variables between what is inside the can and what gets onto your plate, and what of that gets into your mouth, that it's a wonder any of us comes "within delta" of correctly calculating the effects of our food on our blood glucose.

Let's start with serving size. I was diagnosed during the era of "exchange" diets, so to me, a "serving" of vegetables is 5 g carb, and a "serving" of starch is 15 g carb. The oatmeal in my kitchen cabinet (America's Choice Old-Fashioned Oats) states a serving size of 1/2 c (40 g) dry, for 27 g carb. If I take my half-cup measure and do like we learned in 7th Grade Home Ec class -- use a scoop to pour the oatmeal into my half-cup dry measure, and level off with a knife -- when I weigh the result, I get 50 grams. Fifty grams. One quarter more than the stated serving size. (And yes, I did tare out the weight of the measuring cup first.) So if I were calculating insulin based on dry measurement, I'm already off by a quarter. Now, since I use exchanges, I can either recalculate to the package serving (27 g carb divided by 15 g per exchange = 1.8 exchanges) or I can recalculate down to a single starch exchange (15/27 = 5/9, times 40 g per serving, equals 22.2 g per starch exchange). If I were to trust the density (weight per volume) information on the nutrition label, I would be looking to add just over 1/20 cup to my half-cup serving. 1 Tablespoon is 1/16 of a cup, 1 teaspoon is 1/48 of a cup, so I'd be looking for about 4/5 of a Tablespoon for that extra 4 grams carb. Presuming the discrepancies carry down to the teaspoon level, I'm adding in another 5 g carb to a 33.5 g measurement, for 38.5 g carb instead of 30 -- more than half of a starch exchange!

Now let's take something that is more conducive to "just making the whole container worth". I have a can of Progresso "40% less sodium" soup that says its contents weigh 524 g, that the serving size is 249 g, and that the can contains "about 2 servings". 2 x 249 = 498, leaving another 26 g, or 1/10 of a serving. While the carb count on this difference is arguably insignificant (16 g carb per serving = 1.6 extra carb), the sodium content (480 mg per serving) is not, especially when one is counting every milligram. Another can --this one a 14.5-oz can of Shop-Rite Sliced Stewed Tomatoes (No Salt Added) -- states a net weight of 411 g, a serving size of 123 g, 8 g carb per serving, and 3.5 servings per container. 3 servings are 369 g, 0.5 serving is 61.5 g, so 3.5 servings are 430.5 g -- a nominal deficit of 29.5 g or approximately 0.22 servings. Here, though, we are talking about straight vegetables, and 8 g carb translates to 1.6 vegetable exchanges -- so in reality I'm looking at about 22.5 g carb for the can, or 4.5 vegetable exchanges. Significant difference, no?

This all assumes that the contents of the container weigh exactly what is printed on the label. Weighing a 14.5-oz can of diced or sliced tomatoes, I will get as little as 25 g less than listed and as much as 50 g more. Again, this is a significant proportion of a 123-g serving. And these types of canned tomatoes tend not to leave any residue in the can. Other packaged foods -- such as yogurt -- are another story. My favorite Chobani nonfat vanilla yogurt comes in a nominal six-ounce/170-g container, with a 170-g serving size. Sealed, the container weighs 179 g. Once I remove the foil lid, that goes down a gram or two (177-178 g) -- depending on how much yogurt sticks to the lid. Use a spoon to eat the yogurt, and the empty container weighs 10-12 g, suggesting I've only eaten about 165-168 g. Scale this up to the one-quart container of the average store-brand nonfat yogurt. Spoon out all the yogurt, and the open container weighs 33 g. Wash and dry the container thoroughly, and you'll find it only weighs 16 g. In none of these cases can one trust that the weight listed on the container is the weight of food you will get by emptying its contents into a bowl or a cooking pot -- much less the double-transfer from can to pot to bowl.

These are just a few of the causes of discrepancy between what we think we are eating and what actually goes into our mouths. I'll consider a few more in my next post -- and I promise there will be a lot less math in it.