FDA cGMP project - overview

**NOTE: all the information presented under the "FDA cGMP compliance open source project" is not intended to be legal advice, nor is it in any way guaranteed that my interpretation of the statute and system for compliance is going to satisfy any individual FDA inspector or compliance officer. These pages are a synthesis of my own understanding of the regulations and how to achieve compliance.

The landscape of FDA compliance may seem vast, and, indeed, it is. I will tell you, however, that it is not as vast as the landscape of human health and disease: it is much easier to understand than medicine and healing are, and certainly easier to learn. If we are to approach this landscape, which includes forms, research papers, logs and checklists, documentation and reference samples, specifications and lab journals, scales and thermometers, and more, I contend that the best way to begin is to do what herbalists do best: start by finding broad patterns, understand intent and intention, and look at the landscape from high above.

Being an herbalist means you can really improve a friend's condition without fully describing the fine details of their pathophysiology: you can observe the state of their tissues, address imbalances using macro-level energetic descriptions, dovetail them with safe herbs, and achieve excellent results. As you gain experience and deepen your study, you add detail and complexity. It can be a similar process - thought much simpler - to understand what goes into FDA compliance. Bringing that understanding into action can be more difficult - but less so if you know the big picture all along. Essentially, complying with regulations isn't a proscribed set of steps - you have to have a, b, and c done in exactly this way. Unfortunately (or fortunately, depending on how you look at it), FDA expects manufacturers to think for themselves within established parameters. So we can't just "copy" something and expect it to work for us (though having examples is really useful). There isn't a "secret formula" to achieve compliance. We've got to understand the patterns and intentions behind the regulations.

If there is a central point to FDA compliance - something akin to the herbalist's understanding that a living being will attempt to nurture and increase life if given the chance - it would be that one must present proof. You can do almost anything if you can present proof. Ask yourself: how do I really know that this is the case? Where's the proof? It isn't that lab tests are what you need to ensure your dandelion tincture is dandelion. You just need proof: an herbalist looking at that tincture, savoring it, shining light on the dropper may be just as good (in fact, like any good sommelier will tell you, the human palate is way better than a machine). If, that is, you can prove the herbalist is competent. Part of why labs are used so often is that it's actually cheaper than to train a competent herbalist. But for those of us who have followed plants from seed to harvest, who have intimate knowledge of how the veins branch in the leaves, how he flowers change when they dry, how the flavor changes with age - identifying plants is a skill woven into our very fabric. We just have to prove it. Do you have a journal where you record your observations of the plants you harvest and use? Having one will improve your skill, but also serve as an important piece of the proof.

When you make and sell an herbal product, there are four key areas that require attention to proof.

First, what goes in to your product? We need to know about everything - from herbs, to bottles, to your chopping surfaces, to labels. It all must be controlled from start to finish: not used until it's proven to be the correct component, secured, sanitized if appropriate, accounted for, and tracked. If you harvested four pounds of Echinacea root, how do you prove it was four pounds? You weighed it? Really? How do you know your scale is accurate? When did you last check? How do you know it's Echinacea? Because you're an herbalist who grew it from seed? Really? Who says? What seeds? 
You get the idea. It may seem daunting, but once you've got the basic pieces together (we can share examples), you don't have to do it again: you calibrate the scale daily with a standardized weight at a standard temperature. You studied herbs for over 10 years - here's the conferences you attended, the field work you did, the teachers you studied with and when, and furthermore, you conduct blind taste tests on yourself a few times a year and you're consistently accurate. These are all examples, but you get the sense of the work. Lay a solid foundation, and you can get to making products confident that what you're using is what you think you're using - and you can prove it to FDA.

Next, how is the product made? If you're going to build a cabinet, you won't just go get some wood from the lumberyard, make some cuts, and put it together. Hopefully, you will make a plan first. Ideally, the plan will cover everything you need to make that cabinet: materials, cuts, tolerances, specialized processes. Making herbal products requires a plan, too: to ensure consistency of execution and also to prove that product consistency exists, and that no errors were made. How do you know this gallon jar of Echinacea tincture, made from 2 of your 4 pounds, is the same as the next gallon jar made with the other two pounds? There are some specific requirements here, mostly related to the possibility of human error. For example, one part of the plan should be to gather all your ingredients first, then mix them. This may seem obvious, but let me give you an example: when I was first starting with tincture making, I'd want, say, a pint of 50% alcohol. So I'd take a pint-sized measuring cup, measure out 8oz of grain alcohol (let's assume, just for simplicity, that grain is 100% alcohol), then pour water into the alcohol until I got to 16oz, and there you have it. Do you see a problem with this? If you do, you are correct. We'll need to cover the checkpoints necessary to prove that the product is made consistently and accurately. 

Then, what is the product - and what is it not? We have to describe and define what our tinctures are so that we can check to make sure, once they're pressed and ready to go, that they turned out right. We have to prove that the Echinacea tincture actually is Echinacea tincture. If you've ever tasted Echinacea tincture, you might think I'm silly. But try watering it down 10%. Come back a few days later and taste it again. Is it Echinacea tincture? How do you know? You need some kind of reference standard. We'll cover ways to make one, and in our experience with FDA, this way works - without expensive lab tests. Part of the solution relies on covering your bases in the two previous steps. 
But you will also need to define what your product is not: not contaminated with bacteria, not loaded with heavy metals, not laced with pesticides. These are harder questions to answer, and may require a lab. One of the big questions I have is how to create a way to reduce these costs for small-scale herbalists. We know some things that don't work (for example, believe it or not, FDA doesn't buy the argument that 60% alcohol basically kills all bacteria in a tincture. At least not in blanket fashion. Has anyone made such an argument successfully? I'd love to know). 

Finally, where does your product go? You need to establish a way to trace specific lots, with a granularity of each bottling run, to where they go: what stores, what customers. If you bottled half your pressed Echinacea tincture in 2oz dropper bottles on one day, and the other half a week later, they should have different lot identifiers. FDA wants you to be able to account for each and every bottle. Records of bottling = sold product + what's still on the shelf (no more, no less). You sold some product? Really? To whom? Prove it.
For many businesses, this will be part of an already-existing inventory control and invoicing system. But you can do it using modern e-commerce tools (like Square) too, as long as you enter the lot of the product for each sale. Part of what FDA inspections involve is a mock recall: they'll pick a product and a lot, and ask where every single one went.

The areas where we need proof are, to the best of my understanding, described above. It's pretty broad, but if you keep those four key areas in your mind, you will stay mindful of the intention behind the current good manufacturing practices (cGMPs). Once we start getting into specifics, don't lose sight of those overarching areas of proof that are needed. 
But before that, there is an important area which, as has been suggested, has well-defined guidelines (mostly) and just requires following rules: labeling and language. As David said, "Labels are your brand's face to the world". And, increasingly, so is your website and social media presence. Compliance on this front can be easily achieved, because the rules are fairly clear, especially if you refrain from saying your product does anything at all. Since that's pretty ridiculous, there is a law that says we can make "structure/function" claims about our herbal dietary supplements. What this means is less clear - but there are good resources available to help. Next week I'd like to start here: what does a label need? what is a supplement facts panel and how must it be formatted? what kind of language can go on a label / website / facebook page? If folks have experience and resources, please share them next week and I will add them to the page for everyone to reference. I will start out with some detailed examples (vetted as compliant and FDA-tested), links to the relevant pieces of the statute, and a voice walk-through for both to help navigate the examples. 

Thanks for reading, feedback always welcome, stay tuned.


Jonah Ruh Roberts said...

This is awesome, thank you. It seems that if you are buying from or growing herbs on a certified organic farm, with documentation of soil tests and everything the farm has to prove to gain organic certification, this should satisfy the question of heavy metals.
And if we can show scientific proof that bacteria is killed at 60% alcohol then that should be acceptable to them as well.
It seems like these 2 areas are something we could try to work with the FDA to accept as solutions to avoid expensive lab testing?
I understand interpretation of the regulations is subjective to the inspector right? Is the FDA open at all to input from us and codifying acceptable efforts into their rule book?

Guido Masé said...

Unfortunately, though what you say makes perfect sense, FDA doesn't see it that way. There has been a lot of effort to try to exempt tinctures from microbiological testing based on the alcohol percentage - and these efforts have failed. There are bacteria that can (perhaps) survive. Theoretical analyses don't cut it based on what I've seen. And, contamination could be introduced during bottling, etc... and somehow persist in the product. I agree it makes no sense, but there you go.
The same is unfortunately the case for heavy metals. Presses, bottling lines, just dust could introduce heavy metals into the product - even if the raw material is clear. And just because the soil is clear doesn't mean the herb, once harvested, is clear: what if it sat in a burlap bag next to some tractor exhaust by mistake for a while?

Finally, FDA will probably never codify acceptable responses. There's been some clarity on language for labeling, but that's about it. FDA says we need specifications, but will never tell us what they are - because there are many potential satisfactory possibilities. Interpretation is done mostly by the district offices - the inspector is really just a data collector and messenger, and makes no interpretations without checking in with superiors.

Denise Joy said...

thank you I am a very small company with the FDA on my back and feel like I am drowning, so all of this is helpful.

Summer Singletary said...

Hi Guido, great post! I was at the workshop and roundtable discussion on FDA + GMP compliance at IHS. I'm so grateful and happy that you're providing this information to the public. I have a fairly strong understanding of DSHEA compliance on social media. I would be happy to share information, and give concrete examples if that is helpful.

Bonnie Pooley said...

This is a very helpful post. After reading, I am immediately struck by the irony that GMO foods are not required to be labelled, and yet herbalists are being required to jump through these many hoops to sell basically healthful products. Oh, my, what Big Money can buy!

Guido Masé said...

Summer - thanks for the offer. Would love to hear your input after I post some info on Wednesday. Now that we live in a time when FDA can cite you for "liking" a facebook post with non-compliant language (for example, "Wow! This tincture was great. It really helped me when I had a cold last week.")

Kelly Atterberry said...

Great blog and I look forward to the future post. Thank you Guido!

Sheila Devitt said...

This may be a digression.

I am also wondering about the regulations for other types of herbal products that are not categorized as Dietary Supplements. Ear oils, eye washes, pessaries & suppositories; topical lotions, creams & salves... It is my understanding that these range across categories, including Cosmetics and Drugs. Herbal Teas are Food.

The vast majority of my cGMP training revolves around 21CFR111. What are the relevant CFRs for Food & Cosmetics? As for the herbal "drugs", it seems that these must remain in the clinic, with the one-on-one consultation. I don't see a viable retail manufacturing option. I'd be interested in more info/ other opinions.

Thanks ~