By Dr Christian Thoma
Have you ever wondered how different ‘experts’ in nutrition can have such differing opinions, or why the news headlines seem to constantly go from saying something is bad one week to saying the same thing is good the next? Some of this is sensationalist, or even just plain bad reporting. Some of it is human nature. But some of it is because far from being simple, the science of nutrition is pretty complicated.
Healthy eating doesn’t have to be hugely complex, at least not for most of us, but nutrition as a science is far from straightforward. For example, there are three main macronutrients that provide energy and building blocks for our body: protein; fat (including oils); and carbohydrates. Add to that alcohol, which isn’t really a nutrient, but provides more calories per gram than protein or carbohydrates, and dietary fibre, which, among other things, provides some energy indirectly via the bacteria living in our bowels. For most people, only two of these are essential to life: proteins and fats/oils.
Carbohydrates are mostly made up of the units: glucose; fructose; and galactose. These units combine in different ways, e.g. sucrose or table sugar made up of one glucose unit and one fructose unit, lactose or milk sugar is made up of one galactose unit and one glucose unit, and starch is made up of very long strings of glucose units. All of the variations have subtly and sometimes not so subtly differing effects on our bodies (1). For example, glucose has a big effect on how much insulin our bodies produce and need (2), whereas fructose has little to no immediate effect (3). Insulin being a hormone that helps control blood glucose and regulate how we use or store fat (4). The carbohydrates in some foods and meals are rapidly digested and absorbed, and can cause a spike in blood glucose (5). High average concentrations of glucose in the blood are the hallmark of type 2 diabetes. Whereas excess dips in blood glucose can make it hard to concentrate (6).
Proteins are made of amino acids of which there are 20/21, 8 of which are essential, but some are conditionally essential – essential in certain circumstances (7). The specific amino acids in the proteins we eat have specific effects on our bodies, for example the amino acid leucine causes an increase in blood insulin (8,9), and typtophan is the base on which our bodies produce the neurotransmitters serotonin and melatonin, which affects our mood, mental function, and sleep (10,11). Not all food proteins are made up of the same amino acids or in the same proportions (12). The type of protein can make a difference to our appetite and health (12-16). Making claims about ‘high’ or ‘low’ protein diets without considering some of the specifics of those proteins is missing a potentially important part of the nutrition puzzle, yet this is almost always what reports about nutrition do.
Fats can be divided into saturated, monounsaturated, polyunsaturated, and trans-fats. The last can be further divided into conjugated and non-conjugated. Polyunsaturated fats are divided on the basis of specifically where they have the double-bonds that make them ‘unsaturated’; hence we get omega-3, omega-6, and omega-9 varieties. Fats also consist of different fatty acids and can come as triacylglycerols, diacylglycerols, mono-acylglycerols, non-esterified fatty acids, cholesterol esters, phospholipids, and a few other varieties. Fatty acids can be short, medium, long and very long chain. And it goes on. These aren’t trivial distinctions either as they have effects on everything from digestion to influencing inflammation (17,18), and many aspects of health in-between (19,20). For example, nutritionists have been aware for some time that the effect of saturated fat on cholesterol depends on the specific length of that fat (21). Defining a diet purely on the totality of fat, or even just dividing the types of fat into broad categories, misses the importance of ratios and specifics.
Another problem with comparing dietary approaches in studies is that, at least when it comes to the macronutrients (that’s fats/oils, proteins, and carbohydrates) we can’t isolate a single quantity. If you reduce carbohydrates you either reduce total energy (calories), or you have to replace the carbohydrate with fat, protein, or both to make up the energy difference. It’s impossible to just make one change. This has caused no end of problems when it comes to labelling diets in studies.
There are no clearly defined universally excepted dietary labels. You’ll read headlines like High fat diet increases risk of heart disease and High fat low carbohydrate diet reduces heart disease risk factors, and you’ll either be cynical and just keep doing what you’re doing because ‘these scientists can’t make up their minds’, or you’ll ask yourself ‘what is going on here?’ Chances are the two diets in these studies were quite different. The study showing a high fat diet was linked with higher risk of heart disease probably looked at diets where 35-50% of the calories came from fats, whereas the second study probably looked at a diet where 70-80% of calories were coming from fat and only 10% or so from carbohydrates. Both studies are higher in fat than common recommendations, but one brings about a very different shift in physiology and therefore risk of disease than the other.
Another way in which diets vary, beyond ratios and specifics is how much energy (calories) they provide. Diets can provide just enough to meet a person’s needs, provide less (for weight loss), or provide more (for weight gain). The body responds quite differently to getting too much than it does to getting too little. For example, weight reduction is often accompanied by cholesterol reduction irrespective of diet, but when the weight reduction stops, the cholesterol may well go back up on that same diet (22).
Not only does what and how much we eat make a difference, but when and how frequently we eat does (23-30). Even if the carbohydrate, protein, and fat ratios as well as energy (calories) are kept the same overall, how these are distributed across the day, e.g. big breakfast/small dinner vs. small breakfast/big dinner (31), or carbohydrates mostly in the evening vs. throughout the day (32), can make a noticeable difference to our bodies in terms of weight and health. So what we consume, and when is going to make a noticeable difference.
This all assumes that the people in the study did exactly what they were told, or reported exactly what they did, and that the methods used to measure the outcomes of interest were the same, or at least readily comparable. In reality, some of us are really strict when we do something and will carefully weigh out and consider every morsel. Others are of the view that ‘near enough is good enough’, and so may be out by quite a bit. Also, some studies ask people to live on a hospital ward where they can be observed and provided specific food at specific times, while others might just provide some recipe ideas and basic instructions. These different circumstances will have an affect on adherence, outcomes, and therefore conclusions.
Another difference that won’t usually make it into a news report and that you would need specialised knowledge to spot, is use of different methods in different studies. For example, one of the most commonly reported outcomes in nutrition studies is LDL-cholesterol. This is also commonly done in standard medical checks. The thing is the LDL-cholesterol is rarely measured, instead it’s estimated based on other things that are measured, so the accuracy may not always be that good (33-35). In addition, LDL-cholesterol and HDL-cholesterol come in different forms, with some more harmful or beneficial than others (36,37). So a study reporting no change, or a potentially adverse change, in estimated total LDL-cholesterol may be masking the fact that there was a shift in the type of LDL-cholesterol. Research costs a lot of money, and yet is poorly funded, so researchers often have to go with the cheaper and/or easier option.
If you’re confused by all this, then you get the point. All this complexity is confusing. Although much is known about nutrition, there is a lot that isn’t known. The day-to-day practice of nutrition is therefore as much an art as a science, and much of the information in popular books and throughout the internet is opinion. The best results in terms of health come to those willing to learn with great coaches combined with some self-experimentation.
For more insight into the how even similar looking diets can differ in important ways, and how that affects the results of nutrition studies, please read Part 2 and Part 3. Also check out the EAT101 series to build your understanding of nutrition from the basics up.
References: click here to see the full list of references cited.
Nutrition Consultations: want coaching on your journey to better health and greater happiness? Check out our Nutrition or Comprehensive Health Coaching services or organise a seminar for your work, group, or team.
Have you ever wondered how different ‘experts’ in nutrition can have such differing opinions, or why the news headlines seem to constantly go from saying something is bad one week to saying the same thing is good the next? Some of this is sensationalist, or even just plain bad reporting. Some of it is human nature. But some of it is because far from being simple, the science of nutrition is pretty complicated.
Healthy eating doesn’t have to be hugely complex, at least not for most of us, but nutrition as a science is far from straightforward. For example, there are three main macronutrients that provide energy and building blocks for our body: protein; fat (including oils); and carbohydrates. Add to that alcohol, which isn’t really a nutrient, but provides more calories per gram than protein or carbohydrates, and dietary fibre, which, among other things, provides some energy indirectly via the bacteria living in our bowels. For most people, only two of these are essential to life: proteins and fats/oils.
Carbohydrates are mostly made up of the units: glucose; fructose; and galactose. These units combine in different ways, e.g. sucrose or table sugar made up of one glucose unit and one fructose unit, lactose or milk sugar is made up of one galactose unit and one glucose unit, and starch is made up of very long strings of glucose units. All of the variations have subtly and sometimes not so subtly differing effects on our bodies (1). For example, glucose has a big effect on how much insulin our bodies produce and need (2), whereas fructose has little to no immediate effect (3). Insulin being a hormone that helps control blood glucose and regulate how we use or store fat (4). The carbohydrates in some foods and meals are rapidly digested and absorbed, and can cause a spike in blood glucose (5). High average concentrations of glucose in the blood are the hallmark of type 2 diabetes. Whereas excess dips in blood glucose can make it hard to concentrate (6).
Proteins are made of amino acids of which there are 20/21, 8 of which are essential, but some are conditionally essential – essential in certain circumstances (7). The specific amino acids in the proteins we eat have specific effects on our bodies, for example the amino acid leucine causes an increase in blood insulin (8,9), and typtophan is the base on which our bodies produce the neurotransmitters serotonin and melatonin, which affects our mood, mental function, and sleep (10,11). Not all food proteins are made up of the same amino acids or in the same proportions (12). The type of protein can make a difference to our appetite and health (12-16). Making claims about ‘high’ or ‘low’ protein diets without considering some of the specifics of those proteins is missing a potentially important part of the nutrition puzzle, yet this is almost always what reports about nutrition do.
Fats can be divided into saturated, monounsaturated, polyunsaturated, and trans-fats. The last can be further divided into conjugated and non-conjugated. Polyunsaturated fats are divided on the basis of specifically where they have the double-bonds that make them ‘unsaturated’; hence we get omega-3, omega-6, and omega-9 varieties. Fats also consist of different fatty acids and can come as triacylglycerols, diacylglycerols, mono-acylglycerols, non-esterified fatty acids, cholesterol esters, phospholipids, and a few other varieties. Fatty acids can be short, medium, long and very long chain. And it goes on. These aren’t trivial distinctions either as they have effects on everything from digestion to influencing inflammation (17,18), and many aspects of health in-between (19,20). For example, nutritionists have been aware for some time that the effect of saturated fat on cholesterol depends on the specific length of that fat (21). Defining a diet purely on the totality of fat, or even just dividing the types of fat into broad categories, misses the importance of ratios and specifics.
Another problem with comparing dietary approaches in studies is that, at least when it comes to the macronutrients (that’s fats/oils, proteins, and carbohydrates) we can’t isolate a single quantity. If you reduce carbohydrates you either reduce total energy (calories), or you have to replace the carbohydrate with fat, protein, or both to make up the energy difference. It’s impossible to just make one change. This has caused no end of problems when it comes to labelling diets in studies.
There are no clearly defined universally excepted dietary labels. You’ll read headlines like High fat diet increases risk of heart disease and High fat low carbohydrate diet reduces heart disease risk factors, and you’ll either be cynical and just keep doing what you’re doing because ‘these scientists can’t make up their minds’, or you’ll ask yourself ‘what is going on here?’ Chances are the two diets in these studies were quite different. The study showing a high fat diet was linked with higher risk of heart disease probably looked at diets where 35-50% of the calories came from fats, whereas the second study probably looked at a diet where 70-80% of calories were coming from fat and only 10% or so from carbohydrates. Both studies are higher in fat than common recommendations, but one brings about a very different shift in physiology and therefore risk of disease than the other.
Another way in which diets vary, beyond ratios and specifics is how much energy (calories) they provide. Diets can provide just enough to meet a person’s needs, provide less (for weight loss), or provide more (for weight gain). The body responds quite differently to getting too much than it does to getting too little. For example, weight reduction is often accompanied by cholesterol reduction irrespective of diet, but when the weight reduction stops, the cholesterol may well go back up on that same diet (22).
Not only does what and how much we eat make a difference, but when and how frequently we eat does (23-30). Even if the carbohydrate, protein, and fat ratios as well as energy (calories) are kept the same overall, how these are distributed across the day, e.g. big breakfast/small dinner vs. small breakfast/big dinner (31), or carbohydrates mostly in the evening vs. throughout the day (32), can make a noticeable difference to our bodies in terms of weight and health. So what we consume, and when is going to make a noticeable difference.
This all assumes that the people in the study did exactly what they were told, or reported exactly what they did, and that the methods used to measure the outcomes of interest were the same, or at least readily comparable. In reality, some of us are really strict when we do something and will carefully weigh out and consider every morsel. Others are of the view that ‘near enough is good enough’, and so may be out by quite a bit. Also, some studies ask people to live on a hospital ward where they can be observed and provided specific food at specific times, while others might just provide some recipe ideas and basic instructions. These different circumstances will have an affect on adherence, outcomes, and therefore conclusions.
Another difference that won’t usually make it into a news report and that you would need specialised knowledge to spot, is use of different methods in different studies. For example, one of the most commonly reported outcomes in nutrition studies is LDL-cholesterol. This is also commonly done in standard medical checks. The thing is the LDL-cholesterol is rarely measured, instead it’s estimated based on other things that are measured, so the accuracy may not always be that good (33-35). In addition, LDL-cholesterol and HDL-cholesterol come in different forms, with some more harmful or beneficial than others (36,37). So a study reporting no change, or a potentially adverse change, in estimated total LDL-cholesterol may be masking the fact that there was a shift in the type of LDL-cholesterol. Research costs a lot of money, and yet is poorly funded, so researchers often have to go with the cheaper and/or easier option.
If you’re confused by all this, then you get the point. All this complexity is confusing. Although much is known about nutrition, there is a lot that isn’t known. The day-to-day practice of nutrition is therefore as much an art as a science, and much of the information in popular books and throughout the internet is opinion. The best results in terms of health come to those willing to learn with great coaches combined with some self-experimentation.
For more insight into the how even similar looking diets can differ in important ways, and how that affects the results of nutrition studies, please read Part 2 and Part 3. Also check out the EAT101 series to build your understanding of nutrition from the basics up.
References: click here to see the full list of references cited.
Nutrition Consultations: want coaching on your journey to better health and greater happiness? Check out our Nutrition or Comprehensive Health Coaching services or organise a seminar for your work, group, or team.