Gene Testing Does More Harm Than Good, What You Need to Know
A practical and pragmatic analysis with Dr. Tommy Wood.
Dr. Michael Ruscio: Hi everyone. Today’s podcast was with Dr. Tommy Wood, who is back for a second appearance. And what I love about Tommy and his discussion today, which is essentially debunking there being any utility to gene testing. Tommy is a guy who actually goes in and fact checks the research. In this case, not only did he fact check, but he also performed independent analysis on the datasets of some of the effects that a given gene would have on weight, cardiovascular risk, homocysteine as the case for MTHFR.
So a really thoughtful, open gentleman who really does his homework. And what today’s episode really boils down to in terms of the conclusion is that gene testing does more harm than good. It causes worry and fear. And this has a negative impact because when you’re told you have a bad gene or a bad polymorphism of a gene, this has a negative psychological ramification while offering no clinical or treatment benefit. And we’ll elaborate on the rationale behind that in today’s episode.
But a fantastic job by Dr. Tommy Wood bringing some practical and pragmatic analysis to something otherwise confusing and shrouded in misconceptions and that have been propagated by the fact that so many people are doing this without doing the necessary fact-checking work to see if there is anything to the gene testing. We talk about risk, size, effect size, MTHFR, weighed cardiovascular. A fantastic episode.
One of the reasons why I like his approach so much is because he looks at, amongst other things the effect size. What is the effect of a gene on a given marker? One of the parallels I draw is, what I outline in Healthy Gut, Healthy You. So for various gut treatments, what type of impact do these treatments have? And that’s why I am so happy with the response to Healthy Gut, Healthy You in that people really found it to be a practical and reasonable guide to recovering their gut health without making conflations of “Well, Celiac disease affects maybe 2% of the population. And all these bad things happen to celiac patients. So we had to make sure that everyone avoids celiac all the time”. Wait a minute. This occurs in just 1% or 2% of the population. So how do you get from there to everyone? So in any case, if you want that narrative, go to Healthy Gut, Healthy You. If you want to dig in on what I feel to be an amazing job by Dr. Tommy Wood on gene testing, have a listen and I think you will really appreciate this episode.
[Continue reading below]
In This Episode
Episode Intro 00:04:05
A Closer Look at the Fat Gene 00:16:00
Utility of Cardiovascular Genetic Testing 00:29:43
Rebuttal to Gene Testing 00:38:07
Look at the Genetic Studies That You Are Using for Support 00:46:20
Episode Wrap-Up 00:50:16
Download this Episode (right click link and ‘Save As’)
- There is ample data showing that most gene testing is not clinically useful
- Tommy Wood’s paper under review
Hi, everyone. Welcome back to another episode of Dr. Ruscio Radio where today, we’ve got Dr. Tommy Wood here, who hit, in my opinion, a home run with his recent presentation at the Ancestral Health Symposium, essentially picking apart and debunking the case for gene testing. This is probably no shock to the audience because we’ve discussed this ad nauseum on the podcast prior, but if you’re new to this conversation, there doesn’t seem to be a good scientific leg for genetic testing to stand on. Tommy really got in there and made a strong case refuting there’s any benefit from testing your genes.
Now, I’m using strong language there, but I do that intentionally because this is something that I do feel needs to be called out after I’ve seen an overwhelming number of patients who’ve chased down gene testing as if that was going to help them get healthier and, in many cases, it doesn’t. I would argue in the cases where someone purports that it does, it may actually be due to the placebo effect more than anything else. We will get into the details here with Tommy, so Tommy, welcome to the show. Thanks for being here and what a nice job with the presentation.
Tommy Wood: Thanks so much. I’m excited to be back and really excited to talk about this because I think it’s something that’s well-known if you’re a genetic testing or academic, you’re a medical geneticist, but out in the integrative and functional medicine world, we’re still sort of working with old information, assuming that some of this stuff is really useful when it isn’t, so I’m excited to talk about that.
DrMR: Yeah. You make a great point and this is something you mentioned before we started the recording, but I think it’s worth just repeating. When you went to submit your paper, so he’s drawn this up into a paper that is currently in peer review for publication in a peer-reviewed medical journal, which tells you he’s actually willing to subject his ideas to scrutiny, which is really important because if someone’s making a BS argument, my inclination is they’re going to shy away from the peer review process because you’re going to get ripped apart. What did the journal say to you when you submitted this paper or what did the first journal say to you?
TW: Yeah, it was funny. It was actually really useful to me …
DrMR: Validating, yeah.
TW: Yeah, so I put together this talk and then a few people in the audience were like, “Oh, are you going to write this up as a paper?” Because I describe a new approach. I do the statistical analysis. It’s essentially its own study and then so I did. I wrote it up and I submitted it to one journal and the editor who read it was an academic geneticist, so a medical geneticist professionally who studies the effect of genetics and disease risk.
They basically said, “So what? We know that you can’t use direct to consumer genetic testing to predict disease risk and here’s loads of really good publications talking about how useless they are.”
Some of them were great. I read them and I reference them in newer versions of the talk and even in a revised version of the paper, so there’s this big discrepancy between what is “known” in the academic literature and what is being done out in the real world. That part of what I’d like to do is help sort of link those two worlds together so that people that are trying to help their patients, I do believe that people are trying to help, really know what it is that they’re doing.
DrMR: Yeah. I mean, it’s such a validating remark and it’s also validating for me. I think we’re very similar in how we come at this, which is we’re open-minded and we’re hearing all these people proclaim how important gene testing can be, yet every time we go back and we check the actual scientific basis for these claims, we end up scratching our heads, like what am I not getting here because every time there seems to be a paper trying to assess does this gene help, we don’t end up finding any benefit from the gene testing.
I’ll share a couple of pieces of background just really briefly here because we have covered these on the podcast in the past. There was an Asian study in those who were at risk for stroke who had MTHFR positive. They were given folic acid, which is supposed to be terrible to do, and there was about a 30% reduction in stroke in this group. There was another review paper that concluded that MTHFR testing has no clinical utility. There was Christopher Gardner’s DIETFITS Trial, which tried to use genetic markers of polymorphisms to predict who might do better on a low carb diet, nothing found there. There was the ApoE 4 diet trial that tried to look at an ApoE 4 or gene-guided diet compared to a reference diet for impact on cardiovascular risk testing measures and there was no benefit found there either.
On the one hand, I try to be open. On the other hand, I try not to be someone who just follows trends vacuously, despite there being any good evidence or, in this case, the evidence against that hypothesis that gene testing can really help you make decisions about your diet or the vitamins that you use or what have you. I think you and I, Tommy, come back to the same point of where is all of the excitement in the field coming from if this is what the science says? You really dug into some of that and, essentially, it parallels something that I’ve done in GI.
I know I use this example a lot, so apologies to the audience, but we’ve discussed fiber and/or probiotics for weight loss and yes, there is a documentable, statistically significant amount of weight loss gain from fiber and/or from probiotics. The amplitude of the benefit is so small that any consumer would probably go, “Er, what the heck? I got ripped off. This wasn’t helpful. I lost 0.5 pounds,” or, “I’m obese and I lost 1.2 pounds,” which is really not anything that seems to be clinically meaningful. That kind of seems to be the same conclusion that you came to when you started to pick apart some of this from an analytical perspective. Am I representing that correctly?
TW: Yeah, absolutely. I think there’s definitely something. There’s a difference between what is statistically significant and what is physiologically or clinically relevant. Yes, there are now hundreds of genes or SNPs that we might be interested in that in a population on average you have this SNP and you can have one or two copies of each SNP.
If you have that, then you have a statistically, significantly increased risk of something, say, obesity is one that’s looked at a lot. Type two diabetes is one that’s also looked at a lot and risk of Alzheimer’s disease, obviously something, cardiovascular disease, things that people are definitely concerned about.
When you really try and understand what is the likelihood that having this SNP is going to be associated with a different outcome from what I would’ve had if I didn’t have that SNP, what is the likelihood of zero effect, the vast majority in almost any of the SNPs I looked at for individual SNPs, even if you have two copies of the “bad” SNP, more than 90% of people will have an outcome that is completely the same as what would’ve happened if they had the good gene.
As soon as you start thinking about it like that, you’re thinking great, so on average, I have an increased risk, but the vast majority of people aren’t going to see that. That is what is not being communicated to people because you’re trying to sell a test. You’re trying to sell some recommendations. You’re trying to be useful, but in reality, most people aren’t going to see the outcome associated with whatever gene they might be looking at.
DrMR: Now, there are two things I’d like to achieve. One is to help people better understand that and then two, try to understand where that comes from. Maybe we start with where it comes from. That’s probably the easier one to tackle. It’s more conceptual than it is statistical. My thinking, I’m curious if you would agree with this, is there is a business of laboratory medicine and labs, of course, are looking for labs that not only help people but also that there is consumer demand for because if you have a great lab test that no one cares about, it doesn’t matter how great your lab test is. If no one buys it, then all the money you’ve put into the testing and everything else is in vain.
My thinking here is there is maybe something interesting that was mathematically significant, but not clinically meaningful, yet once one or two labs kind of poke that out into the marketplace, there’s going to be this flood of interest. At the same time, and this is not a dig on clinicians, I don’t think clinicians have the time. You and I have days set aside every week where we do nothing but fact check. That is not most clinicians, so I think clinicians are taking the lab’s word for it, but unfortunately, when you go down and you check the details, it looks like these claims are misleading, but that’s being done in a community where everyone is doing it and it’s kind of logical. Well, if everyone’s doing it, it must be true. It turns out that there can definitely be some violations of that.
TW: Yeah, and to your point, I completely agree. I think that this is partly what’s happening. I had a conversation just a couple of days ago with a clinician who I really respect, who is using a new SNP testing service. They told me that this is definitely much better than what’s out there elsewhere. Everything is referenced.
They give a bit of range in terms of the amount of risk an SNP might come with, but they also admitted that that, so I started digging into some of the references and they didn’t seem to quite match up with what was being recommended for each take this supplement because you have this SNP and it’s associated with a risk of Alzheimer’s disease.
They admitted, which obviously, there’s nothing you can do about this and most people can’t do anything about it, that you’re seeing a patient tomorrow. This has shown up. It looks potentially useful and you’re going to act on that information because the patient is saying, “Well, you tested my genes, doc. What’s good? What’s bad? What should I be doing?” Actually, to take the time to do a really significant analysis of the evidence behind what’s being said, what’s being recommended, I mean, it’s going to take hours, days of time and most people just can’t do that.
DrMR: Right. Exactly. Let’s go into what is wrong here, the second point that I wanted to touch on. One of the examples you used in your presentation, I’m assuming also in your paper, is weight. You gave a great analogy, which was if you went to the bathroom before or after you weighed yourself, the genetic variability had less of an impact than if you took a poop before or after you weighed yourself, so it really puts it in context. The clinician’s probably trying to do the best that they can, but now I’m doing all these dietary changes. Maybe I’m popping certain supplements because of this potential risk for weight because of my genes, yet the amount the impact that has is minuscule, so rather, we should just be focusing on the tried and true interventions for weight. Tell us more about pre/post stool and how that factors into the weight analysis.
A Closer Look at the Fat Gene
TW: Yeah, so one of the first genes that I looked at was the FTO gene, which is the fat master, the obesity-associated protein. It’s the fat gene some people may have heard of and there’s multiple different SNPs or polymorphisms you can have in this gene. There’s one, in particular, that’s kind of looked at as the fat gene. It’s the single SNP that’s most associated with an increase in body weight. When you on average, for each copy that you have, you can have up to two copies, your BMI goes up by about 0.3 on average, so maybe two to four pounds in those who have two copies. You think that’s a reasonable amount of weight and you think well, I am destined to be four pounds fatter because I have two copies of this gene.
When you actually look at how much of the variability that having this gene predicts and what I mean by variability is that there’s a whole range of potential weights that any one person could be or that people with a certain genotype can be. When it’s described in a paper, it’s usually described as the mean or just the regular average, with some kind of measure of variability, like the standard deviation. If you go plus or minus one standard deviation outside the mean, you encompass about 68% of people or the middle 2/3 roughly.
When you think about the full possible range, it’s much larger than that, so if you have two copies of the fat gene, on average, in one of the groups of people that I looked at, on average, the median BMI was 25, so basically borderline overweight. The full range of weights could basically be all the way down to 16 or 17 or all the way up to 35, where you’re approaching morbidly obese. It’s a huge potential range and so when you then try and analyze what’s the overall effect of each copy of this gene, you can have one or two versus zero, then you see that it predicts, your genotype for this FTO SNP, predicts about 0.4% of your actual BMI, which as I said, it’s the difference of pooping or not before you weigh yourself is more of a difference than that. I think having a glass of water before you measure yourself or not, that’s the kind of error that we’re looking at. In reality, all these other things that are happening around you are much more important in terms of the weight that you’re going to end up seeing on the scale.
DrMR: Hi everyone. I am very excited to announce we have two new versions of our Elemental Heal available. Elemental Heal in case you haven’t heard of it, is our great-tasting meal replacement, hypoallergenic gut-healing formula. We now have two new versions: One is low carb and the other is a fully elemental whey free and dairy free formula, and they both taste great. The really nice thing about these is they are the only of their kind available on the market. Including the fact that they don’t require a doctor’s note in order to use them.
So whether using Elemental Heal as a morning shake meal replacement, or as a mini gut reset for one, two, three or four days, or even all the way through to the application of exclusive or semi-exclusive for two to three weeks, we now have a formula that should fit your needs no matter how you’re using it.
If you go to drruscio.com and into the store, we’re offering 15% off any one of our Elemental Heal formulas. This discount is limited to one per customer and if you use the code, “TryElementalHeal” at checkout, you’ll get that 15% off. Please give them a try and let me know what you think. I believe you will find them to be not only a great tasting but also really friendly on the gut and will help give you a bump and a boost in how you’re feeling.
DrMR: Now, I’m sure our audience is interested about MTHFR. This seems to be the one and, admittedly, it’s almost frustrating at this point because patients come in saying almost like they’re a diagnosed type two diabetic that they have MTHFR and well, also doc, I should let you know I’m MTHFR. I don’t mean to be overly jestful about that, but what’s frustrating is people are being indoctrinated into thinking that this is something that has a huge benefit or not even a huge, just a noticeable benefit. It doesn’t seem to, but so many people in the field are echoing this. Where do you come down on MTHFR?
- Impact on homocysteine is at worst 1%. This is less than lab error
- FTO gene testing predicts risk of .4% of BMI, which is pretty insignificant, about the amount of weight from a glass of water or bowel movement
TW: Yeah. I use MTHFR as an example in the talk and in my paper because what’s nice about it is that rather than there just being you can have either zero, one, or two copies of the SNP and when you’re looking at the effects of these SNPs, most people assume that they are linear so that if you have two copies, that’s twice as bad, and I use bad in inverted commas, twice as bad as having one copy, which is a certain amount bad compared to having zero copies.
What’s nice about MTHFR is there are two SNPs that are talked about together. The 1298AA is the good version and the 677CC is the good version, so people may have heard of those. Some combination of those has an effect on your MTHFR protein. What’s nice is that when the certain number of SNPs you have in each, either zero, one, or two copies of the bad version, so the 1298C or the 677T, then you get a different amount of how much your MTHFR protein works or how much function you’ve lost. Then you can look at a whole range of MTHFR protein functions across all the different combinations of genotypes and you can look at how that affects something that you might actually care about.
The thing that I looked at because it’s been readily measured and published on is homocysteine level. I know you’ve talked about homocysteine in the past. Does homocysteine associate with Alzheimer’s disease, cardiovascular disease? Does giving B vitamins to bring down homocysteine levels, does that affect the outcome?
I think there’s a lot that’s still up in the air about that, particularly because some other studies like randomization studies have said that homocysteine level exactly might not make that much of a difference on those outcomes, but may be beneficial for some things like stroke, potentially, so it’s not a perfect thing to look at. We know that homocysteine isn’t a perfect marker of the MTHFR gene function, but it’s what most people are using to say, “I’m MTHFR. My homocysteine’s high,” or, “I have MTHFR and therefore, need to take these supplements. Otherwise, my homocysteine will be high.” You see that all the time.
TW: What I did is something very, very similar is I plotted all the different ranges of MTHFR function. This is potentially important for me because I am heterozygous for both of those mutations, which means that I have lost more than 50% of the function of my MTHFR gene. I’m somebody who I’m told by well-meaning people who like to manipulate nutrition one micronutrient at a time, I’m told by those people that I have bad MTHFR function. My MTHFR doesn’t work properly. Those are the things you hear all the time.
When you look at the full possible range of MTHFR function and how much it changes homocysteine, if you look at just the main ones and then the most common ones, which go down to me, so I would be the worst possible function in this analysis, then your MTHFR function predicts about 1% of your homocysteine.
That is less than the variability in the lab test, so whether your lab processed your sample properly is more important in terms of your homocysteine level that they give you than your MTHFR gene function.
DrMR: This is so important for people to know. I wish more people understood this. This is perfectly reinforced by what we see in the Asian stroke risk reduction study I mentioned a second ago. The theory and this is why I say and I really try to get this across to our audience, that outcome data always trumps mechanism data. You’ll have people tell you about oh, my gosh. You shouldn’t take folic acid because MTHFR is going to thwart your ability to use folic acid and it’s almost poisonous for someone with MTHFR because you have this reduced ability to convert and to metabolize. Okay. I mean, that’s the mechanistic theory, but it’s a safeguard.
I am not as smart as you, Tommy, with the deep dive and the stats analysis you can do. It’s pretty remarkable how deep you can get there. I would think with your Ph.D. training, that’s square within your wheelhouse, which is awesome. One of the things I use as a safeguard against my shortcomings with stats is looking at the outcome data because that is where you will flush out if there is a small effect size or if there’s been monkey business done with some of how the data’s been represented. We look at the outcome data and what do we see? Wow, actually, it turns out that, again, using the same example, the MTHFR-positive subjects had a 30% reduction of stroke.
That would perfectly reinforce what you’re saying, which is okay, at worst, the impact of not to say that homocysteine dictates all of the risk here, but if at worst, the MTHFR gene variants led to a 1% change in your homocysteine level, then it would stand to reason that it’s not really that impactful, so to withhold using folic acid from these people would put us at a negative. I wish more people understood this because I think it’s so freeing that now you don’t have to be checking every supplement label. You’re at the airport. You’re trying to get a bar and you look on the label and you see folic acid, so you go oh, I have MTHFR. I can’t have this. It’s like oh, you’re not celiac. This is not the same sort of risk quantification.
TW: Yeah, and beyond all this, what I really worry about is how people talk about their genetics. It’s like you say, you’ve been given a death sentence when in reality, that’s so far from the truth. We do know that telling somebody that they have a bad gene negatively affects their physiology and their performance. You can measure it in the lab. It’s been done multiple times.
The interesting thing also is that we can never be good and get a benefit from that. There’s a couple of exceptions to that rule, but in general, you’re either normal or you’re bad, so there’s an automatic nocebo anytime you talk about your genetics. You’ll see that in even fancy reports. They’ll say here are your bad genes. What does that even mean? You’ve automatically set up this person for a negative effect.
Another example that I have is so I work with various different athlete groups. I was sent a genetic report that was given for free to a Formula One driver that I work with. They had an analysis of SNPs that can affect muscle function.
This one SNP in the exon 3 gene that’s associated with how well you might respond to exercise training, if you believe the research, which isn’t true, but without any references, you give this person here’s a big, red stoplight. This is a bad gene. You have two copies. This is bad. They even say you will not respond well to strength training, so you’ve told an elite athlete, who spends a lot of their time training, who needs to be maximally confident when they’re driving at a brick wall at 200 miles an hour, you have told them that they don’t respond to the training that’s supposed to make them stronger and more resilient to the environment they put themselves in with no evidence to support it.
This is what we’re doing every day when we tell people they have bad genes. They can’t do this. They can’t do that when there’s very, very little evidence to support it. I mean, as you can probably tell, I mean, that’s the thing that really makes me mad.
DrMR: Yeah. I mean, and that’s part of the reason why, even though at baseline, I try to be pretty objective and dispassionate, my passion here is not because I want to convince people of my beliefs. It’s because unfortunately, the majority of people’s opinion has been skewed by incorrect and non-scientifically-backed opinions and this ends up hurting people. I mean, that’s the ultimate travesty. You cause worry and anguish and reduced quality of life while at the same time, you’re not yielding any benefit.
Utility of Cardiovascular Genetic Testing
Cardiovascular gene testing
- Does this change treatment, not currently. You make the same healthy changes irrespective of your genes.
- Getting closer to have a predictive measure here but nothing yet
- Might help motivate people, but advise is the same
Now, let’s look at celiac as an example or maybe even someone with IBS who goes low FODMAP. Those dietary restrictions will decrease one’s quality of life because it interferes with social engagement. It interferes with travel and food choices. On the other hand, it leads to a significant measurable clinical benefit, so okay, there we’re at a net gain. We’re at a net positive. In this case, it’s all negative. The clinical utility is nil, but you do get all the negative side effects of expectations, nocebo, whatever you want to call it.
TW: Yeah, absolutely.
DrMR: Okay. Where else do you want to go? I mean, I’d like to give a few more examples either in cardiovascular or Alzheimer’s. Which one you think is the best one for us to dig into?
TW: Yeah, so I think cardiovascular disease is a good one to pick because what the current state of the academic genetic literature in this field is going towards is rather than saying you have these SNPs, these put you at increased risk. Therefore, you should do these things. What they’re looking at is a combination of clinical risk factors and genetic risk factors. There have been a few papers that have shown that those with a very high genetic risk factor for cardiovascular disease may be the ones who respond most favorably to some kind of pharmaceutical that modifies lipids. Or they’ve looked at what’s your lifestyle risk and your genetic risk for cardiovascular disease and how do they interact.
What they’re hoping is that you’ll get to a point where you’ve measured phenotype. I mean, that’s the thing that we care about and you have some idea of how that affects longterm disease risk. Then you have genetics and then you will intervene earlier in somebody who has a combination of a high-risk lifestyle and high-risk genetics because they’re more likely to be affected. I think that that does potentially have a meaningful endpoint. That’s something we can do, but it’s worth mentioning that at this point, we still don’t know how that looks. I mean, this is like right at the edge of what we’re doing and what people are doing in this field.
I think it’s also worth stepping back and saying okay, does this change what I’m going to do. There was an example that came out. It was in the New England Journal I think 2016. The paper’s called Genetic Risk: Adherence to a Healthy Lifestyle and Coronary Disease. They looked at 50 risk SNPs for coronary disease in 55,000 people, so you think okay, that’s a reasonable group size. That’s something we may want to look at.
Then they looked at genetic risk. They created a polygenic risk score between these 50 SNPs and they made people low risk, medium risk, and high risk. Then they looked at lifestyle factors. Those were whether they smoked or not, whether they were obese or not, whether they have physical activity at least once a week, just once a week, and whether they ate a healthy diet. The healthy diet aspect is standard American Heart Association, whole grains, low fat, low red meat, that kind of stuff. Some of that may be up for debate, which I would be very happy if was the case.
What they said was that those who had three of those, they didn’t smoke, they weren’t obese, and they moved at least once a week, they were the low risk in terms of lifestyle, low risk for cardiovascular disease.
What you see is that those who are low risk from a lifestyle perspective, but high risk from a genetic perspective, that has the biggest effect on reducing risk, so you get more benefit from having a healthy lifestyle if you have a high genetic risk.
You still get a benefit from those lifestyle factors regardless of your genetic risk. Even if you have a low genetic risk, you will still have a low risk of cardiovascular disease if you have a healthy lifestyle, so you don’t need to know your cardiovascular risk based on your genetics because the advice is still going to be the same from a lifestyle standpoint. Yes, they do interact, but the environment, the things that you can change, that you are already going to be doing if you want to reduce your disease risk, those still matter regardless of your genetics.
DrMR: Yeah, and that’s the other key thing, which is unless the test changes how you treat someone, then, I mean, the test, I think, has merit for research and for observing trends. This is how I’m looking at more of the gut testing. I use it, but with most patients, I give it to them as an optional. We can do a SIBO breath test if you’d like to. We don’t have to. We can do a stool test of whatever sort if you’d like to, but we don’t have to. The reason is so many of the treatments that we use, diets, probiotics, antimicrobials, elemental diets, immunoglobulins, gut repair with nutrients, none of those are really steered by lab testing.
Some people make arguments, but in my opinion, those are really not well-supported arguments and so unless the test is changing the treatment, then the test isn’t really helping you with the clinical process.
Again, with cardiovascular, I guess the thing that we could claim could occur as a benefit is, perhaps, motivational, sure. For some people who need that motivation, maybe.
I think most people, I mean, if you’re motivated enough to listen to this podcast, you’re probably not someone who really and, unfortunately, I think it’s the people who are the most motivated who get the most misled by these tests because they’re already doing all this stuff. Then they’re told, “Well, you’re at risk for this.” Then they go beyond the inflection point of the law of diminishing returns. They start being fanatical about things and that’s a lot of the patients I see, sadly, who have made their lives unnecessarily difficult because of the lab test that unbeknownst to them isn’t really even beneficial for them to be following.
TW: Yeah. This is where there’s a dichotomy in the population. There’s a split in the population of how this might affect them. It is different, so you’re going to have a group of people who are motivated. They want to do whatever they can to improve their longterm health. For them, having a genetic test is maybe going to be partially motivating to finally make that diet change or make that lifestyle change. However, I think they are the people who are also going to be most negatively affected by being told that they have bad genetics because they’re going to worry about that. They’re going to think am I doing the right thing based on my genetics. There’s going to be this constant stress based on-
DrMR: Liability of caring, I guess.
TW: Yeah, and so that’s the downside. Then the other side is that there’s plenty of published research. There’s a metanalysis that came out a couple of years ago that said that for the rest of the population, for the average person, if you tell them about their genetics, it does not change their behavior. Then for both groups, either nothing’s going to happen or you’re basically going to get this nocebo effect. Again, you end up at this point where the net benefit just really isn’t there.
Rebuttal to Gene Testing
- There have not been any substantial rebuttals to prove with a high level of evidence that gene testing has substantial clinical utility
- Be really honest about the level of evidence that supports what you’re doing.
- If then the patient agrees to take that risk and you want to try something, that’s one thing
- But you cannot say that it’s because of this SNP and this is the outcome that we’re going to get because you just don’t know
DrMR: Yup. Well, what about some of the rebuttals? I know that there’s been … I guess let me actually even take a step back. I want to make sure that Tommy and I are clear in expressing something that I know he and I both agree upon. We’ve already touched on it in this conversation, which is by outlining this kind of repudiation of gene testing, we are not saying that those who have been doing gene testing have ill intent, are in it for the money or anything like that. I think everyone is trying to do the best they can. It’s just in an emerging field like functional and alternative medicine, we are going to have some stuff that needs to be updated, like anything else.
This is one of the things that requires some updating, so I think if we can get on the same page where this is not you’re doing it wrong. You should be doing it this way. If you’re not doing it this way, you’re a jerk. No, that’s not it at all. I used to be doing things wrong. Namely, adrenal testing is one example. I’m no longer doing that. I used to do repeat and serial SIBO breath tests. I’m no longer doing that. I think it helps if we can all just embrace the attitude of doing less stuff wrong over time, which in that, you’re automatically acknowledging you’re doing a bunch of stuff wrong, but you can get better if you do less stuff wrong. This is we’re all on the same team here, but this is one thing you can do less wrong and that can really be beneficial to our patients.
With that, what are some of the rebuttals that you’ve seen? Are there any that you feel to be credible rebutting your argument?
TW: Yeah. I think that the problem comes down to the fact that when you talk about it how I did in the talk and in the paper, it’s very it’s maths and statistic heavy and it’s sort of like this bombardment of information that most people don’t have time to sit and digest and understand. Usually when you tell somebody that what they’re doing is wrong and I made it very clear to say that there’s no judgment.
I don’t think that people are doing bad things by doing this and equally, there are millions of potential SNPs for us to look at and that you could analyze in a similar way to how I did.
There may well be ones that are really important where you definitely know that you can do an intervention based on an SNP and get a better outcome. I don’t know about those. I haven’t heard about those. Certainly, nobody’s given me a good example, but I’m fully open to the fact that those things may exist. I’m not saying that this is all wrong, but as soon as you tell somebody that they’re doing something wrong, the automatic response is to double down on what they already do. It’s just a natural human emotion. When you ever tell me that I’m wrong, I’m going to do the same thing. I’m going to go hunting for reasons why you’re wrong rather than look at why I’m wrong. It’s just natural human nature.
What most people say is things like, “Oh, it’s just a tool in my toolbox,” or I heard this after I gave my talk at AHS. I heard, “It’s just a place for me to start.” In my mind, that’s the absolute worst way to use it because you’re starting with the assumption that the SNP you’re looking at is in any way connected to the disease or the phenotype that’s in the patient in front of you when we already have evidence that suggests that it’s very likely that that’s not the case.
Often, people will … It’s definitely, particularly … Then I saw another talk where people were talking about maximizing testosterone in men or improving hormonal function. Then a lot of people will talk about the different detox pathways and are your detox pathways working or not. I mean, what does that even mean, right? Of course, your detox pathways are working or else you’d be dead, but there’s obviously a huge amount of variability, but we don’t know. If you have a very good test for whether you have a specific load of a certain toxin, which is hard to do, and you have a way of looking at how much is being cleared out or metabolized, which is also hard to do, and you have some way of saying this SNP definitely does that, great. You should be using that in your clinical practice, but that evidence doesn’t exist.
I think it’s just people need to be more honest about why it is they’re doing the things they’re doing. If you have some kind of clinical tool, a genetic tool that says based on this SNP, this person’s going to have this inflammatory protein that’s going to be higher and, therefore, we recommend this supplement. I can’t say that you shouldn’t do that because I have no idea whether the patient is going to benefit or not. If they do benefit, I don’t know whether it was the SNP that was responsible for that because there’s no high-level evidence to say that’s the case, so you have to be honest with the patient and say, “You know what? We’re at a point where we’re really stuck and we’re just going to try this.”
That’s the only place where I would now use this, even though I don’t really, but if somebody comes to you with genetics already done and you’ve tried everything and you need something to generate some hypotheses to just try something, I’ve used that occasionally in the last couple of years. The problem is you have to then say to the patient, “Do you know what? I have no idea whether your SNPs have anything to do with any of this, but they gave us an idea to try something else.”
That’s okay, but that’s the level of certainty that we’re at, so it’s very important to just be really honest about the level of evidence that supports what you’re doing. If then the patient agrees to take that risk and you guys want to try that, that’s okay, but you cannot say that it’s because of this SNP and this is the outcome that we’re going to get because you just don’t know.
DrMR: Yup. I would agree with that wholeheartedly. For clinicians, if you want to find a way to really gain trust with your patients, this is the way to do it. Over my tenure as a clinician, I’ve become more cautious and very, very discerning in the claims that I make and the language that I use. The response you get from a patient is much better when you say something to the effect of and this is a narrative I have with patients regarding urinary mycotoxin testing, which is a very new area. There’s not great data here. This is really kind of inferential. I’ve been running this test. I’m looking for correlations. It’s too early for me to even be able to say how confident I am in this test, but we can run it. We can pair that with treatment.
Then this is the most important thing, kind of a second point here, the first point of how you discuss these things. The second point I would pair that with we’ll try the treatment and look to your response to reinforce if this is meaningful because that’s the other thing that gets left out is well, you have this gene test, so forever, you’ve got to do this without any symptomatic change to verify that you’re on the right track.
If you do those two things, I think, as a clinician, it’s very freeing because you’re not setting an unreasonable expectation. The patients tend to have more trust and confidence in you because you’re actually being as truthful and as precise as you can. Then ultimately and finally, you’re trying to tie that to a therapeutic outcome, which is really the end result that we’re trying to use all this for at the end of the day. Then you do that for a while and you should be able to get a clinical sense for if a given test, whether it’s gene or not, has any merit in helping you steer the ship from a treatment perspective.
Look at the Genetic Studies That You are Using for Support
- If you are still going to use genetic testing as a tool for treatment, at the very least go back and look at all the profile of subjects/patients that the research was performed on.
- In some cases the research may not have been done on any ethnic people or perhaps it was only done on sick people
TW: Yeah, absolutely. I think one of the other points that I think is really important to take home from all of this is something that we talked about. The last time that I was on the podcast we talked a bit about blood testing. You always have to go back. If you’re going to do this, you’re going to use genetic SNP testing with your patients, go back and look at the studies that you’re using to support your clinical decisions. I know that that’s hard to do and it’s hard to find the time to do that, but if you’re going to do this, I do feel it’s something that you have to take the time to do.
That’s really important because the population in which this SNP or polygenic risk score was studied in is incredibly important. I mean, just like a broad scale, we know that on average, the US population is sick. The average US adult is prediabetic, has at least one chronic disease, takes at least one prescription medication and so that is the population in which these SNPs are being studied. When I look at population data on obesity-related genes, type two diabetes-related genes, even those with the best genetics, they are still, on average, prediabetic and overweight. That’s because of the environment that this is being studied.
If you take away that environment, if you’re somebody who’s not doing all the things that the average US American does, so you eat a good quality diet. You sleep. You have some kind of stress modification practice. You have social interaction. You move frequently. All of a sudden, the effects of these SNPs, small as they are, start to disappear, so any risk score for obesity in terms of genetics, becomes almost completely irrelevant if you’re somebody who is active for at least an hour a day. By active, I mean, just like a standing desk is enough. You don’t have to be going cross-training, CrossFit, to get that effect.
Then when you think about ethnicity is also super important, so that FTO gene, the one that I talked about right in the beginning, no effect in African-Americans, mainly in Caucasians and a smaller effect in certain Asian populations.
Or that cardiovascular disease, lifestyle and genetic risk paper that I talked about, only done in white people, so you really need to know who this study was done in and know if it’s relevant to your patient because ethnicity really matters. So does economic status and childhood environment.
You have to make sure that these things line up for the patient in front of you or else something that is already not particularly useful is even less useful. The kind of work that people listening to this are doing with their patients, like the lifestyle factors they’re putting into place, really make most of these genes even less relevant.
DrMR: Yeah, I mean I … Oh, hang on. I just unplugged a wire. Can you still hear me, Tommy?
TW: Yeah, I can hear you.
DrMR: I’m waving my hands in such agreement and enthusiasm over here I’m unplugging wires. I mean, a fantastic note to close on, which is hey, everyone, the stuff that you’re already doing is probably making even the small, nonsignificant impact of the genes now go from nonsignificant to even less than that, so empowering and hopeful.
TW: Yeah, that’s so empowering and so hopeful. That’s why I think this stuff is really good. It’s not being a downer. It’s saying that all that stuff that you’re doing makes these things really not worth worrying about.
DrMR: Yup. Then for the person who’s saying, “But I’m not feeling well and I’m looking for answers,” totally get it. A good clinician will have a deep toolkit of therapies that can be employed, so I also try to always keep that in mind that one of the main reasons for the gene testing is because someone’s looking for answers to a complaint that they have and so there are answers. We’re not saying there’s nothing else you can do. There’s a wealth of tools out there, but just be careful not to use your time and resources to explore this particular avenue.
TW: Yeah, absolutely.
DrMR: Tommy, where can people connect you or hear more from you if they wanted to track you down?
TW: Yeah, so I probably do most of my stuff on Instagram at the moment, @drtommywood, also occasionally on Twitter, @drragnar. I have a website with the same URL www.drragnar.com. That’s where you’ll usually find me. If I give a talk or something like that, I try and maybe give, so particularly on Instagram, give a few summary slides, give an idea of what I talked about, hopefully, make it fairly accessible, so that’s the best place to find me at the moment.
DrMR: Sweet. Well, thanks for doing the hard work on this, just fantastic. I really feel this is almost like a public service announcement at this point because it’s such a common misconception, so thanks for doing all the heavy lifting on that. Please do let me know when your paper comes out of peer review and we’ll make sure to share with our audience and update our show notes here. Just again, awesome job. The presentation, if people haven’t watched it, if you search AHS, as in Amanda, Harry, Sam, Tommy Wood, you should find it come up pretty quickly. What was the title, something like Lies, Damn Lies, wasn’t that the title?
TW: Yeah, Lies, Damn Lies, and Genetics.
TW: Yeah. You should be able to find it if you look for that.
DrMR: Sweet. Well, Tommy, awesome work, my man. Thank you so much for doing it and for discussing it today on the show.
TW: Thanks so much for having me. This was a lot of fun. DrMR: Sweet.
Stroke. 2017 May;48(5):1183-1190. doi: 10.1161/STROKEAHA.116.015324. Epub 2017 Mar 30.
Aust Fam Physician. 2016 Apr;45(4):237-40.
JAMA. 2018 Feb 20;319(7):667-679. doi: 10.1001/jama.2018.0245.
Am J Clin Nutr. 2016 Sep;104(3):827-36. doi: 10.3945/ajcn.116.135012. Epub 2016 Aug 10.
Links & References
- on Instagram, @drtommywood
- on Twitter, @drragnar
I care about answering your questions and sharing my knowledge with you. Leave a comment or connect with me on social media asking any health question you may have and I just might incorporate it into our next listener questions podcast episode just for you!