GLUTEN PRODUCTS MADE BY THE KELLOGG FOOD COMPANY
Report of the Council on Pharmacy and Chemistry
For over two years the Council has had under consideration certain products offered for the use of diabetics by the Kellogg Food Company of Battle Creek, Mich. These are:
Pure Gluten Biscuit.
Pure Gluten Meal.
40 per cent. Gluten Biscuit.
40 per cent. Gluten Flour.
40 per cent. Gluten Meal.
20 per cent. Gluten Meal.
The Council found these products ineligible for New and Nonofficial Remedies because the statements of composition (particularly of starch content) were insufficient and because the exploitation of the products to the laity was objectionable. June 21, 1915, the company promised to place a statement of the starch content on the package of each gluten product, to place on the gluten flour sacks a caution that diabetics use the flour only on the advice of their physicians, and to revise its advertising in accordance with the suggestions of the Council. Nothing further having been heard from the company, in April, 1916, specimens of the product were obtained, through a layman, direct from the Kellogg Food Company. These specimens, together with the advertising matter received at the same time, and also a letter of advice from the company to another layman, were sent to the Council’s referee, whose report follows. As will be seen, the referee finds that the amounts of carbohydrates contained in Pure Gluten Flour, 40 per cent. Gluten Flour and Pure Gluten Meal are greater than the amounts claimed in the company’s published analyses; that in the two first mentioned the amounts of protein are less than the amounts claimed; that exaggerated claims are made on all the labels and in the advertising literature, and that the company prescribes directly to the patient.
The following report was sent to the Kellogg Food Company for consideration. In reply the firm stated that a revision of its advertising was under consideration but would make no statement as to how soon this revision would be carried into effect. As the consideration had already consumed two years, the Council decided to give the profession the facts and authorized publication of the report. At the same time the Kellogg Food Company was advised that its products would be considered further whenever any submitted evidence warranted this.
W. A. Puckner, Secretary.
Referee’s Report
I submit herewith my report on certain foods offered by the Kellogg Food Company for the use of diabetics. I shall discuss these products from the standpoint of the claims made on the label, from the standpoint of the company toward nonmedical treatment as revealed in a letter to a layman, and lastly, on the basis of the claims made for the foods in the company’s literature.
CLAIMS MADE ON THE LABEL
PURE GLUTEN BISCUIT
| Referee | Company | |
| Water | 8.30 | 5–10 |
| Ash | 2.04 | 1–2 |
| Protein (N × 5.7) | 73.87 | 75–80 |
| Fiber | 0.12 | 2.4–3 |
| Carbohydrates | 14.84 | 0–5 |
| Fat | 0.81 | 0.25–0.70 |
| Starch | 4.02 | 0–5 |
The sample analyzed does not contain the amount of protein claimed for it. It also contains more starch than one might suspect from the company’s analysis. A more conservative claim would be “starch less than 5 per cent.” The company makes the error of using the terms “starch” and “carbohydrates” as synonymous. If the maximum figures of the company’s analysis are used, the carbohydrates would amount to 5 per cent., whereas I find 14.84 per cent. The claim on the label “Guaranteed to contain less than 5 per cent. of carbohydrates” is incorrect. The next claim, “Each ounce of this gluten contains 23 grams of protein and represents 95 calories” is approximately correct, as my analysis shows 20.9 grams of protein and 103 calories.
The following remarks under “Vegetable Proteins” are in my judgment exaggerated:
“Leading authorities are now agreed that meat, fish, eggs and other animal proteins are greatly inferior to vegetable proteins in diabetes, often increasing the sugar output and the dangerous acidosis which leads to diabetic coma.... After many years of experimentation, we have succeeded in perfecting a process whereby the carbohydrates are excluded.”
In this connection, von Noorden, whom the company constantly quotes, says:
“In the slighter forms (of diabetes), the influence of meat albumins is not great and it is difficult to demonstrate the reaction of the patient to different forms of albumin. It may be necessary to add more albumin than the patient can actually take before glycosuria indication is reacted.... Once a medium amount of albumin is exceeded, say 70 to 80 grams, the glycosuria increases, no matter what the type of albumin is.”
My analysis also shows that the carbohydrates are not excluded from this food as claimed above.
40 PER CENT. GLUTEN BISCUIT
| Referee | |
| Water | 8.50 |
| Ash | 1.48 |
| Protein (N × 5.7) | 41.15 |
| Fiber | 0.08 |
| Carbohydrates | 47.81 |
| Fat | 0.98 |
| Starch | 36.98 |
No analysis is supplied by the company, but this may be called properly a “40 per cent. Gluten Biscuit.” The company claims, however, that this is “Best for Diabetics,” which is not true.
Here, as in the case of “40 per cent. Gluten Flour,” the company’s label attributes to “Dr. Wm. Osler in ‘Practice of Medicine,’ ” the following quotation: “Of Gluten Foods, many are very unpalatable, others are frauds. A Good Gluten Flour is made by the Battle Creek Sanatarium Co., Mich.” I have no way of knowing to which gluten flour of the company Dr. Osler had reference. The “Pure Gluten Meal” might be called properly a “good gluten flour,” but this “40 per cent. Gluten Flour” is no better, and no worse, than the average gluten flour on the market. The quotation from Osler gives an entirely false impression.
40 PER CENT. GLUTEN FLOUR
| Referee | Company | |
| Water | 8.62 | 5–10 |
| Ash | 0.89 | 0.5–1 |
| Protein (N × 5.7) | 33.63 | 40–45 |
| Fiber | 0.08 | 1–3 |
| Carbohydrates | 55.35 | 40–45 |
| Fat | 1.43 | 0.2–0.5 |
| Starch | 48.04 | ... |
My analysis shows 6.37 per cent. less protein than the company’s minimum, and 10.35 more carbohydrates than their maximum. In past years I have found the protein in this brand to range from 35.0 to 42.9 per cent. (using the factor 5.7). It is true that the manufacturer does not state what protein factor is used in his reported analysis, but as in four other brands 5.7 is used, it is fair to assume that the same factor applies to this as well. At least such should be the case, as otherwise the manufacturer’s analyses would be meaningless. Even using the factor 6.25 this later sample contains only 36.88 per cent. of protein.
The following statement, in my judgment, as applied to a food containing over 48 per cent. of starch, does not hold water: “This food is of special service in cases of Glycosuria and in the milder forms of Diabetes.” With this brand as with “40 per cent. Gluten Biscuit” the manufacturer again uses the misleading quotation from Osler.
40 PER CENT. GLUTEN MEAL
| Referee | Company | |
| Water | 7.30 | 5–10 |
| Ash | 1.36 | 1–2 |
| Protein (N × 5.7) | 41.55 | 40–45 |
| Fiber | 0.10 | 1–2 |
| Carbohydrates | 48.58 | 40–45 |
| Fat | 1.11 | 0.2–0.5 |
| Starch | 36.59 | 40–45 |
The claimed analysis is justified by my findings. I must take exception, however, to the following statement: “Prepared with great care from a good grade of Spring Wheat, by our special process, which preserves the natural food properties of the product.” The company evidently tries to carry water on both shoulders, on the one hand claiming a reduction in the starch content, while on the other claiming the preservation of all “the natural food properties.”
20 PER CENT. GLUTEN MEAL
| Referee | Company | |
| Water | 7.65 | 5–10 |
| Ash | 1.22 | 1–2 |
| Protein (N × 5.7) | 24.68 | 20–30 |
| Fiber | 0.12 | 1–2 |
| Carbohydrates | 65.41 | 65–70 |
| Fat | 0.92 | 1–2 |
| Starch | 51.24 | 65–70 |
The company’s analysis is confirmed. As the company claims directly that this is “Not A Diabetic Food,” any criticism of its use for that purpose is disarmed. However, again exception must be taken to the statement that “the natural food properties of the product” are preserved.
PURE GLUTEN MEAL
| Referee | Company | |
| Water | 4.60 | 5–10 |
| Ash | 0.96 | 1–2 |
| Protein (N × 5.7) | 76.78 | 75–80 |
| Fiber | 0.08 | 1–3 |
| Carbohydrates | 16.77 | 0–5 |
| Fat | 0.81 | 0.25–0.70 |
| Starch | 6.77 | 0–5 |
The minimum claim as to protein is justified. Again the company confuses carbohydrates and starch, and the food instead of containing from 0 to 5 per cent. of “carbohydrates (starch)” actually contains 16.77 per cent. of carbohydrates, of which 6.77 per cent. is starch. Once more the statement that “the natural food properties” are preserved is untrue as applied to a wheat product deprived of most of its starch.
In justice to the company, it should be noted that on the labels of “Pure Gluten Biscuit” and “Pure Gluten Meal” appears the warning: “Every person suffering from diabetes should be under the care of an experienced physician,” and on the label of “40 per cent. Gluten Meal,” “Persons suffering from diabetes should use this food only on the advice of a physician.” On the other hand, the suggestion on the label of “Pure Gluten Meal,” “Write for a copy of Diabetic Foods and How to Use Them” is a more or less direct invitation to self-treatment. Moreover, a letter dated May 9, 1916, apparently dictated for the Kellogg Food Company by one Ruth French, in reply to an inquiry from a layman, gives direct advice with no reference whatever to a physician.
CLAIMS MADE IN A LETTER TO A LAYMAN
In addition to this inconsistent attitude the letter makes certain clear misstatements, as follows:
“40 per cent. Gluten Flour actually contains 40 per cent. of pure Gluten, making it a perfectly safe article of diet in all but the gravest cases of diabetes. From our Gluten Flour excellent bread, gems and puffs are made that perfectly satisfy the craving for bread with no harmful results.” This flour contains 33.63 per cent. of gluten, not 40 per cent.; it is not “a perfectly safe article of diet in all but the gravest cases of diabetes,” for if one reads the literature correctly, starch restriction is more necessary in mild than in severe cases of diabetes. Furthermore, the bread, gems and puffs made from such a flour do not “satisfy the craving for bread with no harmful results.”
In the next paragraph of the letter, undue emphasis is laid on the “objectionable properties” of flesh foods, a statement only in accord with the tenets of extreme vegetarians. I also doubt very much whether the statement is true that “under a diet of our diabetic foods the thirst to which diabetics are so often subject is usually very much relieved.”
In the next paragraph the assertion is made that “The diet indicated ... is in keeping with the ideas of the highest medical authorities.... Meat is entirely excluded from the dietary.” My reading of the literature does not show that the leading authorities take any such position. Later on reference is made to von Noorden’s claim as to the superiority of vegetable over animal proteins, which I have already discussed under “Pure Gluten Biscuit.” (Certain detached sentences of von Noorden might justify such a statement, but a reading of all he says on the subject leads to a very different conclusion.)
CLAIMS MADE IN AN ADVERTISING BOOKLET
The whole booklet is written from the standpoint of an extreme vegetarian, and therefore is often misleading in its conclusions.
Page 5. “The researches of Ogata and others have shown that cane sugar is a less wholesome food than the natural sugars found in fruits and produced in the body by the digestion of starch, that is, fruit sugars and malt sugars.” In opposition to this I quote from von Noorden, their own authority, “Die Zuckerkrankheit und ihre Behandlung,” Berlin, 1910, page 270:
“That levulose, milk sugar and inulin are more useful than the other carbohydrates is a common opinion, but the importance of their use in practice does not correspond with the theory. In light cases the form of carbohydrates makes little difference; in severe cases the advantage from using levulose, milk sugar, etc., is only slightly greater than from using bread and flour.... Only in certain cases does it appear to me that the special form of carbohydrates possesses any particular significance.”
On page 92 of the same work von Noorden tells us that of the carbohydrates dextrose is the worst, with maltose almost as bad (in spite of the fact that Kellogg exploits his “Meltose,” the “new carbohydrate,” as of special value for diabetics). He also says that levulose increases glycosuria only about half as much as dextrose, when used occasionally, but with long use it is as bad as dextrose and starch.
Page 5. The company refers to sugar as “possibly also causing diabetes.” Sugar or any other carbohydrate may under diabetic conditions cause an increase of glucose in the urine, but I do not believe that any food or any diet can cause diabetes.
Page 7. “That the large use of meat and eggs is not only detrimental but positively dangerous in many cases of diabetes is now a well known and recognized fact.” The dietaries of well known authorities on diabetes are not in harmony with this statement.
Page 13. “It has been discovered that the complete suppression of carbohydrates from the dietary is not only unnecessary but is highly detrimental and even dangerous.” “The complete suppression of carbohydrates from the dietary” is the only means the physician has to determine the diabetic’s carbohydrate tolerance. If carbohydrate-poor foods are so “highly detrimental and even dangerous,” why does the company exploit foods like “Pure Gluten Flour” and “Pure Gluten Biscuit,” whose chief claim to excellence is their comparative freedom from carbohydrates?
Page 17. “Cream is an emulsion, and, with the exception of egg yolk, is the only form in which animal fat is found in an emulsified state.” Milk, Nature’s most wonderful emulsion, is apparently overlooked.
Page 19. “... these foods ... will be found of great value ... especially as substitutes for the breads and meats which are the most objectionable features of the ordinary diet, and which should, as far as possible, be interdicted in this class of cases.” This is simply special pleading for the Kellogg vegetarian diet.
Page 19. “Our glutens ... are all thoroughly standardized, so that in their use the physician and the patient know just the amount of starch eaten.” This standardization is largely mythical. For instance, “Pure Gluten Biscuit” claims 0 to 5 per cent. “carbohydrates (starch),” whereas I find 14.84 per cent. carbohydrates with 4.02 per cent. starch. “40 per cent. Gluten Flour” claims 40 per cent. gluten and 40 to 45 per cent. carbohydrates, whereas I find 33.63 and 55.35 per cent., respectively. “Pure Gluten Meal” claims 0 to 5 per cent. “carbohydrates (starch)” whereas I find 16.77 per cent. carbohydrates and 6.77 per cent. starch. I have a record of six analyses each of “40 per cent. Gluten Flour” and “40 per cent. Gluten Biscuit,” which show the hollowness of this claim of “standardization.” The flour showed 33.6, 35.0, 42.9, 36.8, 35.6, and 40.9 per cent. of protein, with from 40.8 to 55.4 per cent. of carbohydrates; the biscuits 32.7, 33.2, 39.5, 43.3, 33.9, and 41.2 per cent. of protein, with from 41.1 to 54.0 per cent. of carbohydrates. In fact, my experience shows that the Kellogg products are more poorly “standardized” than most of the diabetic foods on the market.
Page 20. “May be made to carry a large amount of fat in the form of butter, a most desirable thing in the treatment of diabetes,” while on page 16 the company claims that in an experiment of Minkowski on a dog, butter “passed through the body without change, none being absorbed”; these are certainly contradictory statements. The explanation is that on the one page the company is exploiting its biscuits, and on the other its nut preparations.
Page 20. Again the incorrect claim is made for “40 per cent. Gluten Flour” that “we believe this to be the only standardized gluten flour made.”
Page 21. The claim is made that flesh foods are “objectionable on account of the large amounts of ptomains and toxins which they contain.” I was not aware that fresh meats contained any ptomains whatever. On the same page the claim is again made that by the use of the Kellogg nut foods “diabetics lose their thirst,” a claim which I think is more than doubtful.
Page 22. “Nuts are a whole food, containing all the elements required for the perfect nutrition of the body.” A marked characteristic of nuts is that they are not “a whole food,” as with the exception of a few varieties, such as the chestnut, they are extremely poor in carbohydrates, which fact gives them their value in the diabetic diet.
Page 23. “With the exception of the potato, the beet and the carrot, vegetables contain little sugar or starch.” Corn, beans and peas are all vegetables which are relatively high in carbohydrates, and for this reason are specifically excluded from the diabetic’s dietary.
From the foregoing considerations I would recommend that the company’s analyses of “40 per cent. Gluten Biscuit,” “40 per cent. Gluten Meal,” and “20 per cent. Gluten Meal” be accepted as correct. Before the Council can accept any of these products, the following steps should be taken:
The company on all its labels should correct the impression that “carbohydrates” and “starch” are synonymous terms.
The labels of all the preparations examined should be changed in accordance with the criticisms given above.
In all cases in which analytic data are given, it would be preferable to state only the minimum of protein and the maximum of carbohydrates.
The booklet, “Practical Suggestions About Diet in Diabetes,” should be radically changed along the lines noted above.—(From Reports of Council on Pharmacy and Chemistry, 1916, p. 56.)