Here's another in an occasional series of blog posts on matters of health, fitness and longevity, a subject I explored in detail in my book "The Life of Riley"
The Lancet, one of the world's leading medical journals, in a sub-publication on Public Health, published a report this week claiming there were life shortening effects of eating too few, or too many, carbs, and how plant based fats and proteins were superior to animal fat and protein in extending lifespans if you were unfortunate enough to be eating too few carbs to begin with. You can read the study here.
This study has gained a significant amount of press coverage, essentially saying it's safe to eat a significant proportion of your diet as carbs again (50% is the suggestion), and that we shouldn't adopt a low carb/high fat approach to eating, and should replace meat with more plant foods, if we want to live longer.
I've looked at an awful lot of scientific studies over the last few years which have suggested eating too many carbs is not good for you, leading to insulin resistance and then weight gain, and listened to a lot of folk who have embarked upon a lifestyle which has reduced their carb intake in order to successfully lose weight and in some cases combat the effects of type 2 diabetes, and I have to say this latest study seemed to pull the rug from under that approach.
Except...having read it, I'm a little dubious about the basis for its findings.
The main study (the scientists also added in other studies but didn't give specific details on them in the report linked to above) was of just over 15,000 adults in the US. These folk were asked to fill out a food diary when they enrolled in the study in the late 1980s, and then again in the early to mid-90s (1993-1995). Around 20 years later (by 2013) the authors analysed who had died out of this cohort (about 6,000 out of the original 15,000), and from that produced a risk ratio for various levels of carbohydrate consumption, having split the participants into five groups, with the lowest fifth eating just 37% of their calories as carbs, and with the highest fifth eating 61% of their calories as carbs.
However, this 2013 mortality analysis was based entirely on what these people were claiming to have eaten 20-25 years earlier! No food intake questionnaire was undertaken after 1995. This seems like a significant omission. I certainly don't eat today what I ate 25 years ago, and I think knowledge of healthy eating has advanced markedly in the last two to three decades, irrespective of the percentage of intake you have as carbs or protein or fat. So for all of these study participants, some tracking over a longer time period of changes in dietary habits would seem like a pre-requisite before pronouncing on what is likely to kill you sooner rather than later.
Just to confirm this point, anyone who, between survey one in the late 80s and survey 2 in the early 90s, had already succumbed to heart disease, diabetes or stroke was removed from the study. This was done, the authors said, to eliminate the possibility that these people would change their diet following these illnesses and therefore skew the results. The implication of this of course, is their belief that those who remained in the study would not change their dietary habits. But, as the authors themselves said "....Another limitation of this study is that diet was only assessed at two time intervals, spanning a 6-year period, and dietary patterns could change during 25 years....."
Looking at the data for the participants, it is worth noting that their average age at the start of the study was 54, with a mean BMI of around 27.7. In fact the group with the lowest carb intake at the start of the study had the highest mean BMI at 28, with the highest carb group having the lowest BMI at 27.4. Interestingly, on average, the study participants put on around 1 point in BMI between the first and second questionnaire, to 28.7. In addition, 70% of this initial low carb group were current or former smokers, vs 51% for the highest carb group, and had the lowest levels of physical exercise measured at just 15% being active, vs 20% for the highest carb group. It almost seems like the low carb group started out as the unhealthiest...and perhaps stayed that way?
The description of this low carb group in the report itself bears repeating here "....Participants who consumed a relatively low percentage of total energy from carbohydrates ....... were more likely to be young, male, a self-reported race other than black, college graduates, have high body-mass index, exercise less during leisure time, have high household income, smoke cigarettes, and have diabetes...." This profile of unhealthy, overweight ex or current smokers and potential diabetics, is nothing like that which you would attach to those who have chosen to live a low carb lifestyle today - all of whom know how important physical exercise and maintaining a lean body weight is as part of their approach to life.
The other significant query I have with this study relates again to the initial questionnaire, which was used to produce a data point, recorded in the main table, showing that the "mean" calorific intake recorded for all participants in the late 1980s was around 1,600 calories per day for all of the groups from low carb to high carb.
I would say 1,600 calories is very, very low - being 600-700 calories below recommended levels for weight maintenance. The recommended level of calorie intake for an adult in the UK is 2,000 for a woman and 2,500 for a man (NHS data here) so a population average of 2,250. Mainstream nutritional advice is to bring calorie levels per day down by around 500 from this figure to lose weight, and although I'm not convinced that is great advice per se, I don't think anyone in the mainstream of this debate would go round suggesting that it is healthy or that people can cope for long only eating 1,600 calories. So these participants, who had a mean BMI of 27.7 (i.e. already quite overweight) were eating 30% fewer calories each day than recommended for normal folk - and six years later had managed to put on a point in their BMI! This 1,600 calories per day figure just doesn't seem right to me. We already know from alcohol studies that people under-report their drinking to a significant degree - perhaps people also under-report their eating - particularly of foods which they think are unhealthy?
Something was definitely not right with the questionnaire - or rather the responses to it. If people were either deliberately or accidentally not telling the researchers what they were eating, and specifically their total amount of consumption, how can we really know what % of their diet was carb vs fat vs protein etc?
I'd also really like to know what constituted animal fat and protein for these people - was it grilled steaks at home with a salad, or fast food burgers in a bun with french fries at a greasy spoon?
The study goes on to discuss animal vs plant protein & fats for this low carb group, and how more plant and less animal helps ameliorate some of the claimed negative effects of the carb reduction. However, as I am quite skeptical about the whole study, this element seems more than a step too far, trying to unpick second order effects on what started out as an investigation into carbs v protein v fat in general.
If this 15,000 study were representative of the US population as a whole, I doubt many of them were at all aware of the benefits of particular structured diets back in the 80s when their food habits were being examined. Throwing that lack of knowledge forward 25 years, to castigate a particular approach to diet today, seems to me at least to be a stretch.
Another recent study, published last year also in The Lancet, and based again on validated food frequency questionnaires on 135,000 individuals enrolled from 2003-2013, concluded that "......High carbohydrate intake was associated with higher risk of total mortality, whereas total fat and individual types of fat were related to lower total mortality. Total fat and types of fat were not associated with cardiovascular disease, myocardial infarction, or cardiovascular disease mortality, whereas saturated fat had an inverse association with stroke. Global dietary guidelines should be reconsidered in light of these findings....."
In this study, levels of mortality rose throughout the carbohydrate groups from lowest to highest (the bands were a little higher overall here, with the low group eating 46%, and the high group 77% of their food as carbs) and fats from animals (saturated fats mainly come from dairy, fatty meats etc) appeared to be beneficial. Quite a different set of conclusions - although high carb was definitely bad for longevity in both studies.
Talk to anyone pursuing a considered Low Carb / High Fat approach today (including those who are Keto or full carnivore), and their animal intake will be fresh meat, fresh fish, all well cooked, and with plenty of veg (for most), and using mostly olive oil or coconut oil for cooking and dressings. Many will supplement this with nuts, and some (like me) who are just plain "low carb" will try to ensure they eat fewer carbs overall, and their carbs are mainly whole grain or low GI as a matter of choice. I suspect this is a long, long way from the dietary habits of those people who initially started in the low carb group in this study, and I'm not convinced therefore this most recent study has much to offer in terms of dietary advice today, except to confirm really high levels of carbs are not good for you.