Use the ADF’s typicals and estimators to avoid nutrient deficiencies and hyper-supplementation
The USDA nutrient database is a wonderful tool, 99% of the data are good, but it is not sufficient for detailed-oriented canine, feline, and human diet formulators.
I know. Details are important to me, and I try not to use data I have not analyzed. I entered all the data in the Animal Diet Formulator, entering and analyzing 10 or more ingredients per day for more than 20 years. Most of the nutrient data were from the USDA nutrient database https://fdc.nal.usda.gov/, ingredient suppliers, and my clients’ tests at nutrient testing laboratories.
As I entered the data, I compared it on my spreadsheet to all other data for similar foods. For instance, I have 65 different raw chicken parts in my database. Anomalies, such as thiamine in raw stewing chickens, stick out.
Often I asked myself, “Why am I doing this? Some of the data are really bad.” However, deep down I knew that if I entered enough data, I would figure it out. I entered enough data!
I built tools to analyze the data. These tools were crucial, especially when looking at the amount of calcium and phosphorus in ground bone samples. Some of the tools include analyzing and comparing the 65 chicken types by milligrams (mg) of nutrient per gram of protein, mg of nutrient per gram of ash, mg of nutrient per gram of fat, mg of nutrient per 1,000 kcal, and, for bone-in parts, Ca + P / ash.
Using these tools, I was able to build the “typicals” and estimators. I threw out the outliers and then averaged the rest to get a typical amount of the nutrient per gram of fat or per gram of protein.
We recommend using the typicals whenever possible, the data will be better and more complete than using USDA or any other nutrient database, which will help you avoid nutrient deficiencies or unnecessary supplementation.
The “why we can’t rely on USDA nutrient data” video shows that if you choose the wrong ingredient such as blue mussels for manganese, eastern canned oysters for zinc, and stewing chickens for thiamine on the USDA database and all programs linked to it, your recipe may have nutrient deficiencies. While the computer analyses may show a balanced recipe, in reality, it may be deficient. Using the typicals and estimators will avoid these problems.
Choline is a special problem.
The lack of data on the choline content of foods can lead to hyper-supplementation. If you are not using ADF’s typicals or estimators, you must examine the data.
Go to the USDA nutrient database, Chicken, stewing, dark meat, meat only, raw. https://fdc.nal.usda.gov/fdc-app.html#/food-details/172404/nutrients. You will find no listing for choline. Try New Zealand, Danish, or Australian nutrient databases. No choline listed. All programs using these databases will show no choline. Only about 50% of the ingredients in the USDA nutrient database list choline amounts.
Now go to USDA database Chicken, broilers or fryers, dark meat, thigh, meat only, raw and you will find that dark meat chicken has lots of choline. https://fdc.nal.usda.gov/fdc-app.html#/food-details/173627/nutrients
My client tests have confirmed this. There is choline in stewing chicken dark meat.
If there are no choline data for the ingredient you choose, the recipe analysis will show a lack of choline even though there is ample choline, at least for dogs.
This can lead to hyper-supplementation, for both commercial and veterinary formulated diets, of one of my least favorite supplements. Choline supplements are often fragile and may interact with other nutrients in a commercial “premix.” I advise my commercial and veterinary clients to avoid supplementing with choline, especially choline chloride.
Avoid the problem; use the ADF’s typicals and estimators.
If you use chicken, dark meat, no skin, typicals you are assured of getting an estimated amount of choline, based upon the best data available. My clients have tested, and usually, my estimates are 15% low.
There are three major reasons why people who formulate diets should not rely on USDA, Danish, NZ, or AU nutrient databases:
- Nutrients in natural foods are highly variable
- Testing is not exact, and mistakes are made. Users must analyze the data.
- The nutrient databases are not complete.