Oh, was it was it Teddy Roosevelt? Uh, which president was who really wanted to use it and he was being recommended against using it and he was like, I'm not listening to you. Yeah.

I need this. I need, I must have it. Um, as far as the, you know, how many people are consuming, what we would call non-nutritive sweeteners. So these are effectively just calorie free sweetening agents. The consumption is estimated to have increased by about 200% in children and adolescents and 54% in adults between 1999 and 2000. So that's over 20 years ago. I, I would say it'd be very unlikely that anybody not only listening to this podcast, but make globally, uh, in developed societies have not been exposed to a non-nutritive sweetener. It'd be highly unlikely, but it's possible, of course. Uh, okay. There are eight non sugar sweeteners currently approved for use in foods in the U S and Europe. Two of them are quote natural. So it's like stevia and derivatives of stevia and also monk fruit extract. It's unclear if these were actually included in the study, by the way, because they specifically go out of their way to say artificially sweetened beverages. And so if you were being, uh, you know, very pedantic, you'd say, well, you can't, can't include those, but unclear if they actually did or didn't. And there are six synthetic ones. So this includes ACE K. So that's stuff like sweet one or soon. Aspartame again, that's equal or nutrissweet. There's neotame, adbantame, saccharin, which is sweet and low and sucralose, which is splenda. Again, those would just be considered the artificial sweeteners, but it's unclear in the study if they just limited it to those six or if they also included stevia, monk fruit extract, et cetera. As far as the lay of the land with respect to artificial sweetener or as I'll refer to them, non-nutritive sweetener intake and health, there's been a lot of kind of, uh, talk about this recently. And may of last year in 2023, the World Health Organization, well, they released a new guideline on these non-sugar sweeteners, which recommends against using them for body weight control or to reduce the risk of non-communicable diseases, things like heart disease, type two diabetes, and so on. There is existing data, pretty good data in my estimation, for those who replaced foods with added sugars in their diet with non-sugar sweetened food, well, it reduces their energy intake by about 136 calories per day, reduces their sugar intake by about 38 grams per day. These are generally beneficial. People lose weight, reduce their sugar intake, disease, processes, and prove. Now, that's in the short term though, when they're doing these sort of controlled experimental studies. The problem is most long-term data sets, you know, sort of analyzing non-nutritive sweetener intake and health, well, they don't measure the replacement of foods with added sugars with non-nutritive sweeteners. They just try to estimate the consumption of non-nutritive sweeteners in and of itself. So you don't really know the behavior, like what's going on there. Non-nutritive sweeteners are also found in lots of ultra-process foods that are generally high in calories, high in added sugars, you can have them together, high in added sodium and high in added fats. So that can be a confounding factor if somebody's consuming a lot of non-nutritive sweeteners or artificial sweeteners, but is also consuming a lot of ultra-process foods, you would predict that would be generally poor dietary pattern, for example. So the long-term data on artificial sweeteners from a health perspective is confounded markedly by the way it's being measured, and particularly the behaviors, food-related behaviors. I can see where the WHO was coming from. They're like, look, man, the existing data doesn't show a big health benefit here. And I would agree with them. It doesn't. But that's the limitation to the data in my estimation, not because artificial sweeteners, or I prefer to refer them as non-nutritive sweeteners are particularly dangerous. So if someone's replacing foods with added sugars with non-nutritive sweeteners or foods with non-nutritive sweeteners, in general, I'm in favor of that unless those foods still contain added sugars on top of that, and again, are ultra-processed in general. We also talked about, so we talked about that in the research review from May of last year. And then more recently, we talked about it in reference to the SWISH trial. Now, this was a study that lasted for a year, and is basically comparing the intake of water only, or non-nutritive sweeteners artificially, sweat and beverages, and the effect on weight loss. So there was a 12-week active period of weight loss, and then they would see how the two different groups maintained the weight loss for the next 40 weeks, so to the completion of a year. So the weight loss with the non-nutritive sweeteners was 7.5 kilos, and then it was 6.1 kilos with the water only. Now, this was barely statistically significant, but one in favor of non-nutritive sweeteners, and again, they increased their weight more, but also their waist circumference more than just the water only group. And again, that was statistically significant. If you thought that there was going to be some untoward outcome from consuming a lot of non-nutritive sweetened beverages, well, it should have shown up in this point. And it didn't really matter if these folks regularly consumed non-nutritive sweetened beverages, or artificial sweeteners, or not. It just seemed to persist regardless. Also, the waist circumference was reduced more in the folks consuming non-nutritive sweeteners compared to water alone, and that again was statistically significant. So before we talk about this actual study, the lay of the land here, based on the quote, barbell medicine interpretation of artificial sweeteners. Artificial sweeteners aren't essential to health, nor are they uniquely health promoting, but they also don't appear to be uniquely harmful either. Any risk, I think, that is attributed to non-nutritive sweeteners is probably due to the co-occurrence in ultra-processed foods and a generally poor dietary pattern, not the non-nutritive sweetener itself. Still, overall, at a population level, I think it's a reasonable recommendation to say that diets shouldn't be that sweet. If the diet is very sweet, it likely contains a lot of food and beverages with added sugars, or a lot of artificial sweeteners from ultra-processed foods. And so I think if you just want to say, the diet shouldn't be that sweet, don't consume a lot of extra sweeteners, I think that's fine. But to say that artificially sweetened beverages, or artificially sweetened foods, or just non-nutritive sweetener consumption in general is deleterious or harmful to someone's health, well, that's a reach, in my opinion. Austin, you agree with that?

Yeah, I agree. I mean, we've seen the pendulum swing all over the place, but even so far as like, soda companies deliberately promoting or producing and promoting natural sugar products or cane sugar products with the idea that, of course, if you label it as an artificial sweetener, people just have this aversion suspicion to quote unquote artificial things, and it natural sugar might sound better. But I don't think either of us would advise that people go out of their way to consume added sugar, even if it is quote unquote natural sugar, over the use of a non-nutritive sweetener in the diet on a habitual basis. Yeah,

I agree. So this particular study, now this is a prospective cohort study where the researchers assessed the individual's dietary patterns at baseline using a fruit frequency questionnaire, and it multiple intervals throughout the data collection period. And basically, we're trying to see if there was a connection between the particular variable or set of variables in this case. So that's artificial sweet beverage consumption, sugar sweetened beverage consumption, and or a genetic risk score, and the outcome of interest, in this case, atrial fibrillation risk. They used the UK United Kingdom biobank data. This is a nationwide data set from the United Kingdom consisting of over half a million people aged 37 to 73. In this particular study, they started out by including just over 200,000 people. Others of the data set were missing dietary recall questionnaires. It didn't have genetic polygenic risk score data, for example. And then they cut that down to just over 140,000 people because they eliminated any family members. So they basically didn't want kinship to sort of sway the results for the genetic risk, risk ratios. As far as beverage consumption, again, this was assessed via a 24 hour food frequency questionnaire, basically as a questionnaire that asked you to look back over the previous day and say, hey, how much of these things did you eat? And sugar sweetened beverages were considered fizzy drinks and squash Austin. Have you ever heard of squash as

a dream?

Nope. So this is like a fruit based syrup that you add water or soda to, kind of like a cordial, if you will. So yeah, so you like you have some of that at home. If you just added it to some water, like a topo chico, you'd have a squash. I'd never seen this term outside of vegetable or like a particular type of racquetball. So I was like, wait, what is squash? So that was sugar sweetened beverages. Artificially sweetened beverages, they classified as low calorie drinks, but the authors don't go in a great detail over this. And this is, we should contrast to some of the other studies like the switch trial we referred to earlier on weight loss, which was no calorie drink. So effectively a diet soda has no calories in it, for example. They also did another analysis on pure fruit or vegetable juice, which will completely ignore because the results there aren't interesting. And they eliminated milk tea or coffee ingestion from exposure despite folks usually, particularly in the UK adding sugar, potentially non-nutritive sweeteners to those beverages. They separated exposure into four separate categories. So zero liters per week, no exposure effectively to either sugar sweetened beverages or artificially sweetened beverages, less than one liter. So not zero, but not a lot. One to two liters per week, and then greater than two liters per week. And the atrial fibrillation diagnosis was obtained by linkage to their medical record. So effectively they were diagnosed in clinic, they looked up these people and said, Oh, got a fib where you didn't. The genetic risk was measured by what's known as a polygenic risk score, which represents the total number of genetic variants that an individual has to assess their heritable risk of developing, in this case, atrial fibrillation. There are a number of potential confounders, like Austin was alluding to, there are many risk factors for developing atrial fibrillation. So increasing age, male sex, specific ethnicities, different socioeconomic status is smoking, alcohol use, physical activity, particularly lack thereof, elevated BMI, poor sleep or obstructive sleep apnea, high triglycerides, and the list goes on and on and on. So they have multiple models here, which are beyond my statistical comprehension. But as a general role, there are three separate models that they included here, model one adjusted just for age and sex, and model three attempted to adjust their findings for all of these variables. I am again, I know it just enough about statistics to be dangerous, but I cannot comment on whether their adjustment was adequate, inadequate, completely off the walls, or whatever. That is the case in a lot of these large observational studies, but it's kind of the devil that we're working with here, the devil that we know. All right, so let's talk about the results. Again, the sample was just under half men. So 45% men, the mean age was 56 for artificial sweetener beverage consumption. 67 and a half percent consumed zero liters. So no artificial sweetened beverages per week. 15.3% of the study group consumed less than one liter and 10 and a half percent consumed one to two liters per week and 6.6% consumed greater than two liters of artificial sweetened beverages per week. They tended to be younger, there were more women, higher BMI, and there was a greater prevalence of type two diabetes in the greater than two liter group. For sugar sweetened beverages, almost 80% of the study population consumed no sugar sweetened beverages per week. Again, they probably have milk, sugar in their coffee, sugar in their tea, but other fizzy drinks or squash, I guess, not big in this group. 8.2% consumed less than one liter, 6.9% consumed one to two liters, and 5.5% consumed greater than two liters of sugar sweetened beverages per week. They tended to be again younger, in this case male, not female, lower socioeconomic status, again, they had higher BMI and more coronary heart disease in the group who consumed greater than two liters. Okay, so with respect to AFib, the follow-up period was just under 10 years and there was just over 9,000 cases of AFib in the study group. How they calculated risk, well, there was no correlation between those consuming less than one liter or one to two liters of artificially sweetened beverages and atrial fibrillation. So what they did is they compared zero liters, so no exposure, to consuming greater than two liters of artificially sweetened beverages per week. And the risk was about 20% greater in those consuming greater than two liters of artificially sweetened beverages per week compared to those consuming zero. For sugar sweetened beverages, the same relationship was found or lack thereof, I should say, meaning that there was no correlation between those consuming less than one liter or one to two liters per week, or rather just at greater than two liters of sugar sweetened beverages per week. So when comparing no exposure to sugar sweetened beverages to greater than two liters per week, the increase in risk was about 10%. Now, here's where things get interesting. We talked about them developing this polygenic risk score. And so those who had an intermediate genetic risk score has nothing to do with artificially sweetened beverages or sugar sweetened beverages. This is just the genetic risk alone. Well, the increase in risk was 74%. And to put that in perspective, again, the risk for consuming greater than two liters of artificially sweetened beverages is 20%. And for those consuming greater than two liters per week of sugar sweetened beverages, the increase in risk for atrial fibrillation was 10%. But intermediate genetic risk alone, independent of consuming artificially sweetened beverages or sugar sweetened beverages was 74%. And those who had a high genetic risk, well, the increase in risk was 216%. Now they went one step further, and they compared those with a low genetic risk of atrial fibrillation and consuming zero liters of artificially sweetened beverages or sugar sweetened beverages per week to those with a high genetic risk and consuming greater than two liters of artificially sweetened beverages. Well, the increase in risk was 251%. And I think the comparison here, the thing we need to point out is that the genetic risk here dwarfs, absolutely dwarfs the effect of artificially sweetened beverages or sugar sweetened beverages on the risk of atrial fibrillation. That's impressive. Did you know that about the genetic risk, like the magnitude of effect there?

So I wouldn't have been able to cite to you specific figures on the genetic risk. I'm certainly aware that it was a major factor. I think that my lack of intimate familiarity with those specific numbers is mainly because at this point in time, there's not much I can do about people's genetic risk for atrial fibrillation. And what I'm recommending for people who either have it or who are trying to reduce their risk for it is generally the same across the genetics, the genetic spectrum, right? Relating to things like managing obesity, diabetes, metabolic syndrome, alcohol, use and sleep apnea would be like overwhelmingly the top things that I'd be looking at for somebody who has atrial fibrillation not due to structural heart disease.

Yeah. All right. So let's talk about the take home messages here. So my interpretation here is that the exposure to non-nutritive sweeteners or artificial sweetened beverages is correlated to an increased risk of atrial fibrillation, not through a causal mechanism per se, but rather more by complex associations with established risk factors. So in other words, those consuming a lot of sugar sweetened beverages or artificially sweetened beverages are more likely to have established risk factors for atrial fibrillation than those who don't. So things like a higher BMI, more type 2 diabetes, lower socioeconomic status, reduced physical activity, smoking, alcohol use, environmental exposure, and so on. It's really difficult, yes, I alluded to earlier with the statistical modeling to fully adjust for these things, particularly in an observational study. I mean, you would need a bunch of randomized controlled trials to to suss this out, particularly those where you knew the dosing and eliminated all the other variables to come up the causal sort of mechanism

here. Think about the scope here. I mean, this trial that you're discussing, they had what over 200,000 people originally included in the study that got wheeled down to, let's say just under 150,000 people, and then they had like 9,300 cases of AFib. No way that that that's feasible in a randomized fashion. So prospective cohort data is the next best thing that we can have here. But I agree, there are just like AFib is so complex and it remains honestly pretty poorly understood. A lot of aspects about it remain poorly understood. But so even though some of that statistical adjustment can be useful to try to eliminate the impact of certain variables, I just don't think that we're fully there. And so any residual correlation that we see between the use of these sweeteners in AFib, I think is much more likely to be explained by something that we haven't caught or figured out or detected yet than it is due to the ingestion of the actual sweetener itself.

Yeah, or like I said, the correlation of those who consume artificially sweetened beverages have particular lifestyles, behaviors, etc, or environment and or environmental risks that ultimately we've already know about. But like the artificial sweetener itself is maybe not the smoking gun. And my rationale for that thought is that there wasn't a dose-dependent relationship here. So anything less than two liters per week, there was no increased risk of AFib. And what you would have expected to see if there was some like smoking gun, some sort of causal relationship here is slightly increased risk at less than one liter compared to zero liters, some slightly more increased risk from one to two liters compared to that, and then a even greater risk at greater than two liters. And that's not really what we saw here. And again, as I just, you know, let the cat out of the bag earlier, the genetic risk dwarfs this sort of exposure, it's like, wow, okay, so that's clearly a huge, huge risk. It's not modifiable, unfortunately. But I don't think that there's good support for an artificially sweetened beverage mediated mechanism of atrial fibrillation. There's also issues with the data here. So there's no caffeine intake data here, which many people would think is a potential confounder. I'm not sold on that because the existing data on caffeine intake at typical doses, that doesn't really support that it provokes atrial fibrillation or other arrhythmias. But, you know, you would want to know that just to like cross your T's and dot your I's. Also, the intake of artificially sweetened beverages and sugar sweetened beverages in this data set was self reported, which we know that there are issues with that. So the dosing, you know, it's hard to come up with a prescriptive dose limitation. If somebody was like, look, I'm doing everything that you guys have recommended, all of the things I just I'm trying to optimize, bro. What do I do? I would feel very, I wouldn't feel very confident saying, well, look, diet soda, diet drinks, if you limit them to less than two liters, I feel good. I just don't know that that dose is representative there, you know, it's self reported here. Again, it's an observational study. Also, the inclusion of low calorie drinks in the artificially sweetened beverages versus no calories. I don't, I'm thinking I'm like, these like T's, you know, for example, you know, that have like 15 or 20 calories in them from added sugars and then some extra artificial sweeteners to make it sweeten up for people to drink. That's a different, that's a different deal than a no calorie food. And also, the population studied is not generalizable everywhere. This is a relatively affluent sort of group in the UK. And you know, the you would want to have maybe a more diverse sample if you wanted to generalize this to everybody. So overall, I don't think exposure to artificially sweetened beverages independently increased the risk of atrial fibrillation. Though I do think it likely correlates with other risk factors that have already been established and will be and further ones that will be established in the future at a population level. Most people consuming artificially sweetened beverages aren't doing so as part of a health promoting dietary pattern. Most people don't need to help promoting dietary pattern period, right? And direct studies on artificial sweetened, artificially sweetened beverages remain generally supportive of their inclusion when they alter dietary patterns. And so far as they reduce added sugar intake, reduce overall calorie intake, but they these changes may not promote actual long term changes in the dietary pattern, because again, long term dietary changes are relatively difficult period. If non-nusative sweeteners, including those that are artificial, help you better manage body weight, fat, body fat, diabetes or metabolic syndrome, and or avoid or reduce the use of alcohol, that is likely be far more beneficial from an atrial fibrillation risk in general than like going cold turkey and ultimately not making those changes. And again, I go back to my sort of general heuristic that at a population level, diets shouldn't be that sweet. I

agree.