My latest article appears as a guest post on LIVESTRONG.com. The article presents 10 tips that may help prevent hip fracture and gives scientific studies in support of each. Click here to read the article.
When I review scientific studies on this website, I always end my review with a “limitations” section. The section is a list of factors or issues that may limit the reproducibility of the study (i.e., self-reporting for food intake), limit who the study is relevant to (e.g., men, women), and so on. When dealing with studies of chemical compounds or drugs, dosing is a very important variable and must be taken into account. I haven’t reviewed a lot of studies where this was an issue, but there were two instances I wanted to point out as they slipped under the radar.
The first pertains to my previous post, “No, a Glass of Wine Does not Equal 1 Hour of Exercise.” This study was overinterpreted by the media, but in my discussion, I left dosing off my limitations list. One of my readers noted that the dose of resveratrol (the compound of interest found in wine) administered to the rats was 4g/kg of body weight. If we take the average rat to weight 300 g (.3kg), then rats were getting about 1.2g of resveratrol. For humans, if they drink a medium sized glass of wine, they are consuming about 0.175 liters of wine. If we are generous and say that the amount of resveratrol in a liter of wine is 7 mg (0.007 g), then one glass of wine gives a human a dose of 1.2x10e-6 (or 0.0000012 g). You can see here that by drinking a glass of wine you are getting nowhere near the amount of resveratrol the rats were getting. In comparison to a human drinking a glass of wine, the rats were overdosed. This is not a fair comparison.
The other instance arose in the article I wrote for Top.me entitled, “5 Supplements That Science Recommends for Fitness.” In the article I noted that science backed the use of caffeine to improve performance, which it does. But the original version of the post didn’t have any mention of dosing. Of course the articles I cited in the article give the dosings used (3-4mg/kg body weight), but most readers likely wouldn’t look for that. This equates to around 320 mg for a 180lb person. Most caffeine supplements on the market are pills of 300 mg or less and suggest taking only 1 tablet, so this is in line with the studies. But caffeine is a stimulant. What if you took more than this amount? Well, it turns out this is exactly what a number of teenagers have been doing and a number of them have died. It appears they were using caffeine as a supplement, but taking it in powdered form. In this form, measurement was up to them and made in teaspoons – and 1 teaspoon is about 5 grams. This is over 15 times the amount of caffeine adults were given in the studies I cited. You can see how this might be a recipe for disaster.
The limitations of studies are extremely important. Virtually every study has one or more limitations that make the findings impossible to directly apply to human health and fitness without additional studies to support their claims. Dosing is a common limitation in studies of chemical compounds and supplements, so keep this in mind the next time you read a headline that seems too good to be true…it probably is.
I was asked to write an article for top.me on fitness supplements. There are thousands of fitness supplements out there, but the number with scientific support for their effectiveness is far less. I narrowed my list down to 5 supplements that have scientific studies (in humans) supporting their use. You can check out the article here:
5 Supplements That Science Recommends for Fitness
Today I came across a summary article reviewing a scientific study on intermittent fasting (skipping one or more meals to promote fat loss). The original article can be read here. The summary article was interesting for two reasons. First, the writer of the article was using the study to “debunk” intermittent fasting and second, the writer brought into play the concept of Bayesian Inference as a major reason why our personal bias can get out of hand.
This is a perfect opportunity for my readers to test their ability to interpret scientific studies and the way those studies are portrayed in the media. Take a few minutes and read the original summary article here. Once you’ve read it, continue reading here…
Now that you’ve read the summary article, what did you think? More specifically:
Quiz Question #1: Do you feel the article satisfies the scientific requirements for “debunking” a theory (i.e., intermittent fasting)? Why or why not?
Quiz Question #2: Do we have to stop using Bayesian Inference in order to achieve unbiased results?
Here is my take on the answers to the above questions:
Answer to #1: This is the easier of the two questions, if you’ve been reading my critiques to scientific studies and have read my Science 101 page, one of the major points is that you cannot take a single study and say “this is how it is.” Science rests on multiple-studies verifying an existing hypothesis or theory. It is the reproducibility of a phenomenon that makes us more comfortable in saying the phenomenon exists. So, contrary to what the summary article says, one study on intermittent fasting cannot rebuke the entire phenomenon. If you’ve been reading my articles you also have a good idea of why one study cannot be used. This is because all studies have limitations. The summary article in question did not highlight any of the study’s limitations, but there were several: the study period was only a month, the fasted group got a protein shake right after their training, the fasting was only removing around 250 calories (what about individuals who normally eat more than 250 calories during the fast period?), and the study only examined women (women are known to be less responsive to intermittent fasting).
Answer to #2: This is a tough question as it requires to reader to understand Bayesian Inference. The writer of the summary article frames it as a process that is loaded with bias and often leads us astray. In truth, it is a statistical approach. Yes, it has its limitations, but it is used in numerous disciplines inducing science, medicine, and law. I think the risk lies not in the approach of Bayesian Inference, but how it is applied. The author of the summary article sates, “My inclination to believe the efficacy of this [intermittent fasting] method grows stronger as my understanding of the science behind it is reinforced by my own experiences.” In other words, if my experiences keep reinforcing the science, then I’m likely to give the science far more credit that it deserves (i.e., personal bias). But it doesn’t have to be this way. And if you think like a scientist it won’t be. Instead you will keep accumulating an understanding of what multiple studies show and then weigh these findings in the context of your own experience. This is a better approach to Bayesian Inference and this is how new hypotheses are generated and theories are refined.
As scientists, clinicians, therapists, etc., we cannot completely remove personal bias. To do so would be to remove us from the situation completely and that cannot be done (the studies won’t perform themselves!). Regardless, science begins with a question. In this question, the questioner invariably has an opinion as to the what the answer will be (a prediction). They just need to be open to the possibility that their prediction might be wrong. As long as we are open and make an attempt to consider all of the available information, then we are doing our due diligence.
And now, my very first book review. My review is of the book “Shred It!” which I read to assist in the training of one of my clients (it was a book they were referring to in our discussions). The book advocates a whole food, plant-based diet for optimum health and fitness and claims to be a step-by-step guide to burning fat and building muscle using this approach.
Unfortunately, right in the Introduction it makes claims that are hard to substantiate, such as 1) the diet being the best diet to prevent and even reverse disease. 2) the diet and lifestyle that yields the most energy, 3) the best diet to aid in recovery after exercise, 4) that fiber is only found in plants and of paramount importance, 5) that cholesterol should be avoided. It seems these claims stem from the author being a disciple of Dr. Campbell and the science given in the Forks Over Knives documentary (which I review here). The science just isn’t supportive though: Studies haven’t properly compared plant-based diets to other diets like Paleo. Arctic peoples have and continue to eat all meat (no-fiber) diets. The cholesterol data has been picked apart by Gary Taubes in Good Calories, Bad Calories. Because I place a premium on how science is presented, all of these points have a major influence on my review.
Regarding the rest of the book, the Getting Started chapter is good. It touches on motivation, goals, and being specific. The Basic Nutrition and Fat Burning chapters are also good. There is then a brief chapter on gaining muscle which focuses on making sure you ingest the right mount of calories. The estimations are certainly reasonable, but they don’t address balance of macronutrients. Indeed, the book espouses a 70/15/15 percent split of carbs, protein, and fat. For everyone. But scientific support for this is not given. The author says these numbers are within healthy ranges, as close to ideal for energy production, muscle recovery/growth, joint care, digestion/assimilation ease, and disease prevention as you can get. But the proof is not given. The book eventually shifts away from scientific claims and focuses more along the lines of how the diet should be chosen because “it works.” I am all for discussing things that one observes to be true, but there is a lot of this. More than 30 pages are devoted to convincing the reader of the diet’s effectiveness with different success stories. This would have been a great opportunity to delve into variations of this diet as clearly the different case studies presented were a mix of men and women of different ages and body types. We get tidbits of how they had to individually alter their diets and training, but not enough to put together a comprehensive picture. The book then concludes with meals, exercises, recipes, and yet another case study section, this time featuring athletes.
In the end this book gets off on the wrong foot scientifically, and although it has some good sections, they are few and far between. It reads more like a sales pitch than a guide, which may have been the intent given how difficult it is for vegans to obtain mainstream acceptance (a reality that baffles me). But for those that are not biased against a plant-based lifestyle, you aren’t getting a whole lot of information that can’t be found elsewhere. There were a lot of missed opportunities here, so Shred It! is not a book I would recommend.
During aging and in elderly populations it is very common for bone loss to place individuals at risk for one or more fractures. Mild loss of bone is refereed to as osteopenia, whereas severe loss of bone is known as osteoporosis. Bone fracture risk is higher for those with osteoporosis and a great deal of research is being conducted to diagnose, prevent, and treat fractures in the elderly. Nutrition can pay a role in age-related bone loss and a common recommendation is to make sure you are getting plenty of calcium since calcium is the major mineral building block of bone. This recommendation has resulted in many individuals taking calcium supplements. However supplements are often taken with no consideration to the quality and composition of regular foods being consumed. Perhaps these can have a protective effect if chosen wisely.
In a recent article by Byberg and others (2014) in the Journal of Bone and Mineral Research, fruit and vegetable intake was analyzed in men and women to determine if there were statistical associations between intake levels and incidence of hip fracture. The rationale was that daily intake of at least 5 servings of fruits and vegetables is associated with reduced risk of disease (e.g., diabetes, cardiovascular disease). Further, prior studies have indicated fruit and vegetable intake was associated with increased bone mineral density. Some studies also exist linking fruits and vegetables to reduced forearm fractures, but the specific foods and doses could not be linked to the reduction in fracture risk. Therefore, in the Byberg et. Al 2014 study, the authors sought to clarify the relationship between fruit and vegetable intake and hip fracture risk.
The Takehome: The study indicated that, in both men and women, low intakes of fruit and vegetables (less than 1 serving) were associated with increases of hip fracture compared to 3-5 servings. Over 5 servings of fruits and vegetables did not appear associated with any increased reduction in hip fracture. So, there appears to be a limit to the power fruits and vegetables; you can’t reduce your fracture risk to zero by increasing your intake beyond 5 servings a day. As with all questionnaire-based studies, this study is limited by the ability of the participants to accurately recall the details of their past eating habits. The study also doesn’t address the mechanism by which fruits and vegetables might protect against hip fractures, though the authors speculate it may be tied to the roles these foods might play in reducing oxidative stress and inflammation, and increasing intestinal calcium absorption. Nevertheless, the findings are of note given the study’s large sample size and detailed data set. How many servings of fruits and vegetables are you eating each day?
- Swedish men (40,644) and women (34,947) free of cancer and cardiovascular disease (age 45-83 years old) were studied.
- Fruit and vegetable consumption was recorded based on surveys given to participants asking them how often, on average, they had consumed each food during the previous year.
- Responses were converted to average daily intakes based on age and sex-specific portion sizes.
- There were 14 vegetable entries tracked (carrot, beetroot, broccoli, cabbage, cauliflower, lettuce, onion, garlic, peas, pea soup, pepper, spinach, tomato, and “other” vegetables) and 5 fruit entries tracked (apple, banana, berry, orange/citrus, and “other” fruits)
- The outcome measure was a participant’s first incidence of a hip fracture.
- Information on prevalent diseases, body composition, size, medication, smoking and various other variables were incorporated in the statistical design as covariates.
- Depending on the specific analysis used, 5 or between 3 and 5 servings of fruits and vegetables per day was used as the reference point.
- Cox’s proportional hazards regression models were used for assessing the association between fruits and vegetables and hip fracture.
- The was an inverse association between fruit and vegetable intake and hip fracture such that less than or equal to 1 serving a day was associated with an almost 50% increased rate of fracture, as compared to 3-5 servings per day.
- Compared with 5 servings a day, lower intakes of fruits and vegetables were associated with higher rates of hip fracture.
- Intakes above 5 servings per day did not appear associated with any increased benefit to hip fracture incidence.
- There was no interaction between fruit and vegetable intake and possible co-variants like potassium, magnesium, calcium, or body mass index (BMI). So, at the very least, these factors were not underlying the observed fruit & vegetable association.
- Fruit and vegetable consumption was self-reported by the participants (not monitored by the investigators).
- Only certain covariates (education, physical activity, smoking) were considered and other factors not considered may be contributing to the observed relationship between fruit and vegetable intake and reduced hip fracture incidence.
- This study does not point to causation, only association between fruit & vegetable intake and reduced hip fracture incidence.
One of the people I train suggested that I watch the movie Forks Over Knives as it is the basis for his, and many other individuals’ switch to a vegan lifestyle. For those that might be unaware, vegans consume foods that are plant-based and in no way derived from animals. I watched the movie last night and it was a very good example of the problems we face in objectively and fairly presenting science to the public. The upside is that this experience can be used as a great learning tool (by virtue of this posting) and it has now made me realize that movie and book reviews will be a great addition to this site. So without further ado, my review of Forks Over Knives.
SUMMARY: The science in this study is disappointing (see Movie Details below). Some claims are dated and no longer valid, key experimental details of other studies are omitted and appear to have been framed just to fit the theme of the movie. But all of this aside, the major message of the movie – that a whole-food, plant based diet is healthiest – simply cannot be supported. Overall the movie spends a large amount of time on personal stories of how a plant-based diet improved the health of many people. But the diets of these individuals were typical western diets full of excessive calories, dairy, and refined carbohydrates. For sure the movie does a good job of showing you that the plant-based diet is better than the traditional western one, but that hard line is not maintained. The movie strays into claiming meat and dairy are bad. Perhaps they are (there is more and more evidence to support that diary is not the best, at least for adults), but the movie can’t separate them out. The movie ends with what sounds like a nice synopsis: “You have two choices. You can eat yourself into poor health and early death or you can eat yourself into good health and a long healthy life…and that road is on a plant-centered dietary pattern.” Unfortunately this claim can’t be substantiated. There are too many diet permutations out there right now and these must be considered. For example, to make this claim you would have to at the very least compare whole food, plant-based diets to whole food, paleo diets. Paleo diets remove dairy, grains, and refined carbohydrates, while including meat. Thus, for these reasons, I this is not a movie I would recommend.
Background: The movie came out in 2011, so it is several years old. The premise is that a whole food, plant-based diet is the answer to preventing many, if not all, of the diet-related diseases and increased mortality rates seen over the past hundred years. Various individuals including scientists, doctors, and those who have tried the diet, were interviewed and give an array of information and personal experiences.
The Health Problem: Forks Over Knives sets the stage well in terms of highlighting the obesity epidemic and noting that western society’s consumption of meat, dairy, and refined carbohydrates has increased dramatically alongside a variety of diseases (such as cancer) and alongside increased mortality rates. But from this point on the movie lumps these three dietary components together (meat, dairy, refined carbohydrates). Scientists have a hard enough time with factors they didn’t consider, but here we have three known factors (some of which may be a problem, some of which might not be) being grouped together.
The Cholesterol Argument: The writers of the movie fall into a common trap of using the old dietary cholesterol (cholesterol you get in your food) literature as support for staying away from meat. The correlative studies of cholesterol and cardiovascular disease began in the 1960s and are hugely problematic. Gary Taubes does a great job of pulling apart all this literature in his books Good Calories, Bad Calories and Why We Get Fat. The science just doesn’t stand up to the test of time.
The Rodent Cancer Reports: The movie highlights a study where high levels of casein (a protein in milk) caused cancer in mice. But the cancer was in combination with Aflatoxin, a known carcinogenic mold toxin. An interaction such as this is hard and dangerous to interpret in terms of the effects of Aflatoxin and casein alone. The movie does note another study (in rats) where only casein levels were altered and the rats became susceptible to cancer and other systemic disruptions. This is a better study for making the case that dairy might be harmful, but again, it was performed in rats and only looked at one milk component. And of course, this only speaks to dairy, not to meat consumption.
Nutrients From Animals Fat Caused Cancer: The movie passingly notes that Dr. Campbell conducted many studies (presumably in animal model) that showed many nutrients in animal fats caused cancer. However, the movie does not indicate which ones, nor the context in which they were tested. For example, were these nutrients only found in animals, and not in plants?
The China Study: Arguably the pillar on which this movie rests is The China Study conducted by Dr. Campbell. This was a huge correlation study and I mean huge. 94,000 correlations between diet and disease were found. Of these around 9,000 were significant. This is quite a lot, but given the vast number of correlations that were considered, it is not surprising to find many significant results. The summary point made was that plant-based diets were correlated with lower disease occurrences and lower mortality rates. But here is where have to go back to the lumping I mentioned above. The study does not control for what could be a drop in meat, dairy, or refined carbohydrates. In fact the study didn’t measure these differences well at all. Gary Taubes was questioned about this study and you can read his response here. On major issue he hits on is how the study don’t control for carbohydrates and, thus, the plant-based individuals were being compared to individuals that had a higher sugar content in their diet. Thus, low sugar intake may very well have been the factor that lead to the better health in the plant-based groups.