Inconsistency in manufacturer's micronutrient measurement methods and inaccurate labels


#1

I’ve been taking this high-end cod liver oil from a company called Green Pastures for a couple of years as cod liver oil is commonly considered one of the most effective general health and well-being supplements a person can take.

Green Pastures uses traditional fermentation methods that were common up until around 1850 to produce its cod liver oil. Green Pastures makes a whole lot of claims about how it’s product is superior. It’s a small company with a large following in the Weston A Price Foundation community. It’s fish oils were also written about by Tim Ferris in the 4 Hour Body, where he claims that the company’s cod liver oil was one of the biggest factors in his increased testosterone levels.

Anyway as I was looking at Green Pastures site I found an interesting dialogue between the guy who runs Green Pastures and a nutritionist who called in to ask about the quantities of vitamin A and vitamin D that were in the oils.

If you go to this page you click play on the audio player for “Discussion on Vitamin A and D” and hear the phone call where he describes the following in greater detail:

The Green Pastures manufacturer goes into a very interesting discussion about the process Green Pastures went through when it was first entering the cod liver oil business. He needed to get his product’s vitamin D levels measured so he went to a certified nutritional analysis lab. The first lab he used claimed there was no vitamin D in his product at all. He knew there’s no way that could be right, so he went to a second lab and got them to test his oils. The second one claimed there was 800 IU of D2 per ml of his cod liver oil. He couldn’t figure out who to believe but he finally resolved the dispute by finding a guy who performed rat bioassays: a process where they feed rats your product and measure the resulting effects - in this case on calcium in the rats. The rat bioassay guy came back and said there was definitely vitamin D and that in his opinion measuring vitamin D in terms of IUs is "all bull and irrelevant. The only way you can measure vitamin D is by what it does, and the only way you do that is with a rat bioassay." The manufacturer of Green Pastures then goes on to say he found yet another accredited lab that used HPLC mass spectrometry to measure vitamin D and they consistently found that his cod liver oil products had a tremendous 20,000 IUs.

If this small, well-respected, organic, high-end manufacturer of fish oil found all these inconsistencies in testing of micronutrient levels of his products, I can only imagine that many big companies focused purely on profit and not on quality of nutrition have run into the same issues but simply don’t care. The Green Pastures manufacturer himself believes that most nutritional measurements that large industrial-scale manufacturers put on their labels are completely arbitrary. He says that if you were to pick any target level of vitamin D IUs you wanted he would be able to find a certified lab that would say thats’s how much his product contained - in other words, the labels of supplements are completely untrustworthy and depending on the technician and the method used you can get dramatically different results - in his case anywhere absolutely zero D in his product to 20,000 IUs of D.

I wanted to at least try to verify some of this so I did some googling as I was listening. HPLC MS is the mass spectrometry method that he says the labs use to measure vitamin D IUs. Searching “hplc ms vitamin D” on Google the sixth search result is a study called Misleading measures in Vitamin D analysis: A novel LC-MS/MS assay to account for epimers and isobars. The first two sentences of the study are as follows:

“Recently, the accuracies of many commercially available immunoassays for Vitamin D have been questioned. Liquid chromatography tandem mass spectrometry (LC- MS/MS) has been shown to facilitate accurate separation and quantification of the major circulating metabolite 25-hydroxyvitamin-D3 (25OHD3) and 25-hydroxyvitamin-D2 (25OHD2) collectively termed as 25OHD. However, among other interferents, this method may be compromised by overlapping peaks and identical masses of epimers and isobars, resulting in inaccuracies in circulating 25OHD measurements”.

I have zero background in chemistry beyond an American high school class so I really can’t do this or other studies justice, but if anyone else can listen to the audio I posted above or look into this study or others, that would be awesome.

This is just a case of one fish oil manufacturer’s experiences with measuring his products vitamin D content, but I think it raises the larger question of how we can usefully make accurate initial calculations as well as adjustments in our DIY recipes.

I’ve seen other people on this forum suggest that the only real way to make a good Soylent recipe is to treat the digestive system as a black box: record what you’ve put in, then measure your blood work some time later, make adjustments and try again. This really lines up with what the Green Pastures guy says about the rat bioassays. But if the nutritional labels of supplements and foods cannot even be trusted how can we expect to alter our recipes by switching between different products which may have used different measuring techniques for their micronutrients. If you start with a product that has 400 IU of D on its label, get blood work done, and find that you need to make an adjustment - let’s say you need to increase D by another 400 IU - and so you switch to a vitamin D supplement from a different company that has 800 IU of D on its label, you don’t actually know if you’ve increased your D, if you are getting the same amount, or if you are getting less. The only way to be sure is to have your body tested.

Can anyone shed light on this issue? I don’t want to believe that it’s that hard to make accurate, useful micronutrient measurements because it means that creating a good DIY recipe is going to be even harder, but I don’t have enough of a background in this area to know what to think.


#2

I agree with this statement, in part because I don’t care for IU conversion, in part because everybody might not get the same value from the same volume of a nutrient, and in part because different testing methods can cause a plethora of testing issues.

Well, I have taken a class where we used a Mass Spectrometer, although I wouldn’t say I’m an expert on it. Its been having issues with giving erroneous results years before I took the class. These results where caused by

  1. contaminated insertion point,
  2. bad sensors
    3)bad pumps
    4)bad program (requires user input to prep and program doesn’t control correctly)
    5)bad connections (power, input, output, etc.)
    just from what they fixed when I was dealing with it. So MS machines are easy to mess up and get bad results with, and that is not including if your sample is contaminated, or not purified enough. Its easy to get erroneous results with sample issues, assuming the machine is working correctly, and the user didn’t use the program incorrectly. That being said, when it works correctly, and the user knows what they are doing and how to use it (I took one look and decided it was more complicated than it should be at my colleges setup), AND the sample is correctly prepared, it should give a fairly accurate analysis. Keep in mind this is an old machine (10 years I was told), and I haven’t seen or used a newer one (one made within a year or two ago) so I don’t know how much better one made recently is, and if any of these issues I experienced are resolved in a new model.

They didn’t say what method they used to measure it? Not knowing what they did, I wouldn’t be able to guess what went wrong. I’d have to say the rat bioassay would be the most accurate one done.

This may be the only way to make your own personal recipe, and while I agree that there needs to be a measurement method-standard, I doubt the companies will get together with each other and their teams, with or without government interference, and make the standard.
The testing itself is relatively easy, its accounting for every error generating step that’s hard. The general rule of thumb is, the more steps, the more chances for an error to be induced, either by human or machine.

That being said, Rosa Labs are using actual chemical nutrients (eg. calcium carbonate) for the most part, ones that we know the body can break down and use. I think that means that this issue with measurements and measuring are only in compounds or liquids and liquid dispersions.