The Personalised Diet by Eran Segal and Erin Elinav
High-level thoughts
A great resource for people looking to understand their diet and its impact on their mood, health, etc. First half of the book gives the background science on how our responses to food vary - written by the authors who were involved in some of the landmark research in the area. Second half gives practical advice on how to find your personal diet profile and create an optimised diet. I haven’t done it yet (it involves checking your blood glucose), but planning to in the future.
My notes
Blood sugar is the window into our response to food
We wanted a metric that would be relevant for weight loss but would also be relevant for metabolic (diet-related) disease. The metric needed to be measured easily and accurately across a large study group. All of these parameters led us to consider blood glucose levels, or more precisely, blood glucose levels following a meal.
We chose blood sugar as a primary measure, because:• postmeal blood sugar provides an immediate, measurable response to food;• blood sugar fluctuation is a good marker for weight and health issues;• blood sugar monitoring has good technology—we could measure blood sugar in our study subjects every 5 minutes for an entire week, resulting in glucose responses for approximately 50,000 meals and snacks overall.
Of course, we knew that there were many factors beyond blood glucose levels that influence weight and health, but we also knew it was an important one, and using it as a metric to determine food responses seemed promising and potentially informative.
Diabetes isn’t the only problem that can develop from poor blood sugar control and blood sugar that is too high. When your blood sugar remains too high for too long, or regularly goes too high after eating, you will also be at a higher risk for the following:
- weight gain
- hunger and food cravings
- overall mortality
- heart disease
- cancer
- dementia
- nerve damage
Responses to diet are personal
For every single medical or nutritional finding that came up in this study, there were many people whose results were very different from it.
For example, for every food that was likely to cause a high postmeal blood sugar response on average (such as pita bread), there were people who had a low postmeal blood sugar response to it, and for every food that was likely to cause a low postmeal blood sugar response on average (like chocolate, probably due to its high fat content), there were those people who had a high postmeal blood sugar response to it.
Glycaemic Index doesn’t predict blood glucose response
The Glycaemic Index (GI) is a system for ranking foods according to how much they influence blood sugar. It is based on a scale from 1 to 100. Foods with no carbohydrate content, such as olive oil or steak, do not have a GI because they don’t contain any carbohydrates and should not therefore directly influence blood sugar. Pure glucose is ranked 100 because it should increase blood sugar (blood glucose) more than any other food. Every food containing carbohydrates is ranked somewhere between 1 and 100.
Theoretically, consumption of high-GI foods will cause a blood sugar spike, and consumption of low-GI foods will keep blood sugar more stable. However, most GI values you see in books or on the Web are based on an experiment conducted by one company (there is no official body that does this or sanctions the publication of GI values).
The blood glucose responses to a food vary widely, so the average of those responses will not be informative to any of these tested people. For example, the responses to apples range from about 45 all the way up to nearly 90. An average for an apple would also be 65, the same as for a banana. But the individual response to an apple could be anywhere in the rage of 45 to 90 (and possibly outside of either of those results). It’s likely that you would respond to a banana like others in the study, but not so with the apple.
An additional problem is that GI values are not additive. You can’t eat broccoli with carrots, and predict the response by adding the GI of each component together
People have different sensitivities to carbs. Some people respond very strongly, others do not, so a single value for a given amount of carbs predicting the response for all people would not work.
Context also matters. If meals are eaten before or after exercise or at different times of day, these factors also have an effect—that is, two meals that are identical in calories but that vary in fat or protein, proximity to exercise, or time of day have been shown to have quite variable blood sugar responses in different people. For these reasons, a model using only carb content will not be a good predictor of response.
Generalised trends in dietary responses
Carbohydrate Content: There was also a gradient of people between these two extremes, from very high to very low sensitivity, as well as variability in the response to specific carbohydrate-rich foods (such as a high/low or low/high response to fructose solution compared to white bread or ice cream compared to cookies).
Fat content: In general, the more fat added to a meal, the lower the postmeal blood sugar response. This may sound surprising but is actually consistent with previous studies that have shown that adding fat to meals can reduce the postmeal blood sugar response
Fiber: In general, higher fiber within a meal tended to increase postmeal blood sugar response for that meal but had a longer-term positive lower sugar spike effect in future meals. In other words, 24 hours after eating a high-fiber meal, most people’s blood sugar response improved even when the fiber-containing meal had a higher postmeal blood sugar response at the time.
Sodium and water: In general, a higher sodium content in food was associated with a higher postmeal blood sugar response, but a higher water content was associated with a lower postmeal blood sugar response.
Time of meal: The more time that had elapsed from waking, the higher the postmeal blood sugar response, so the breakfast postmeal blood sugar response was generally lower than the blood sugar response after dinner.
Age: The older you are, the more likely your postmeal blood sugar response will be elevated—but again, this was not always the case.
Microbiome: Having higher levels of a bacterium called Parabacteroides distasonis was associated with higher postmeal blood sugar responses, whereas higher levels of the bacteria Bacteroides dorei was associated with lower postmeal blood sugar responses.
How to find your personal diet profile
The key to discovering your personalized diet is to test your blood sugar before and after eating the specific foods you enjoy or that you wish you could enjoy more often.
Your body will respond well to certain meals and foods with a gradual blood sugar rise and fall within a narrow range, or it will respond poorly to certain meals and foods with a dramatic blood sugar rise and fall, or it will have blood sugar that stays elevated for longer than it should.
You can find out all of this information with a simple finger prick. Although a finger prick may not sound appealing, it is relatively simple and the only way for you to see in real time how your body responds to specific foods or meals.
While we understand that pricking your finger may seem like a sticking point (pardon the pun), we assure you that the personalized results you will get from this experiment will be well worth the effort. You’ll be glad you tried it.
One alternative to a finger prick is a continuous glucose monitor (CGM) which are becoming more available for consumer purchase.
If you want to get your blood sugar under control, but you really don’t want to test your own blood sugar, or cannot do so for some reason, there is another way you can get a fairly good, if less precise, idea of how specific meals and foods affect your blood sugar. That is to track your hunger levels and weight. In general, hunger after eating, when you should be satiated, is a sign of blood sugar going too high and then dropping too low due to an insulin surge.
In this personalized diet program, you will be:
- Planning what foods you want to test and shopping for what you will need.
- Organizing your blood sugar testing schedule. You will be take a baseline test in the mornings, then test the foods you eat before and at intervals after eating them.
- Testing your blood sugar in response to specific meals and foods.
- Tracking your results on our app, on your own, or on a chart
- Analyzing your results to determine which foods and meals work for you and which give you a blood sugar spike.
- Enjoying your own personal “good” foods without guilt and eliminating your “bad” foods, or determining ways to modify your spike with the ideas in the next chapter.
- Watching your weight and health measures normalize… and enjoying your life!
How to further optimise your diet
Once you’ve finished your week of testing, you have some valuable information. You know which meals and foods make your blood sugar go too high, and you know which meals and foods cause only a gentle rise or no rise at all. You may also still want to test new meals and foods you discover from time to time, and that’s fantastic.
Using what we discovered, you can try out these strategies to see if you can get your spikes to come down. In some cases, the food itself may not have been the primary cause of the spike. It might have been that you didn’t get enough sleep, or added too much salt, or perhaps the spike would come down if you changed the type of carb or added fat.
You may have already discovered from your blood sugar testing that some carbohydrate-rich foods tend to give you a blood sugar spike and others don’t. Here are your options:
Isolate the offender. If the spiking meal contains multiple carbs, the first thing you could do is test it several times, eliminating one of the carbs each time you test, so you can figure out which carb (or carbs) is the offender.
Reduce the portion. If you love to eat a big bowl of oatmeal (or pasta or rice or whatever grain you like), would you be happy with a small bowl?
Break up the meal. If you increase the duration of your meal, you may decrease your blood sugar rise. You could try eating more slowly or breaking a large meal into portions or courses and waiting a short while between courses (this is what dinner conversation is for).
Increase fiber. Try including more fiber to your meal, such as using whole grains instead of refined grains, or adding wheat bran, oat bran, wheat germ, or another fiber-rich addition to a fruit smoothie or yogurt, for example.
Manipulate your fruit. If you discover or suspect that fruit is the problem, try a different fruit. Dried fruits contain concentrated sugars, so you may find that substituting fresh fruit (such as adding blueberries to your oatmeal instead of raisins) makes a difference.
Manipulate your added sugar. If you always use cane sugar on your hot cereal, try a different added sugar, like honey, real maple syrup, coconut sugar, date sugar, or a little molasses.
Manipulate your milk. Milk is a source of protein, calcium, and fat (except for nonfat milk), but what some people don’t realize is that milk is also a source of carbohydrates because of the high levels of lactose (milk sugar) it contains.
Add more fat. In many cases, adding fat to a carbohydrate-rich food decreased the blood sugar spike, sometimes dramatically.
Reduce processed foods. Many processed foods contain food additives that tend to spike blood sugar in many people. The most obvious is artificial sweeteners, which we have talked about at length and may spike blood sugar in some consumers.
Modify your lifestyle. Sleep more, exercise more, adjust meal timing, less salt and more water, consider your hormonal shifts, try relaxation or meditation.
What the future of dieting could look like
There are continuous blood glucose monitors (CGMs) that do not require finger pricking. “Minimally invasive” CGMs are already available, which require a tiny insertion of a miniature needle into the skin but does not require finger pricking for continuous calibration.
Several companies are working on entirely non-invasive CGMs, which will meausre blood glucose through the skin. None have yet been announced as successful, though.
Microbiome analysis: The bacteria in your gastrointestinal tract are being increasingly realised as relevant for how you respond to food and many aspects of your health.
Wearable sensors: We are seeing more hi-tech companies developing self-monitoring devices of various kinds. Many are already available. Some monitor number of steps, calories burned, and heart rate (like Fitbit and Apple Watch), and each new release seems to add more features, like sleep tracking and blood pressure.
“Omics” approaches: One of the most exciting revolutions occurring in science and medicine includes the use of advanced computational platforms that are now able to analyze endless quantities of individualized big data and apply them to multiple facets and aspects of human health and disease. Some examples are the sequencing of the human genome (genomics) and measuring levels of gut bacteria (microbiomics), bacterial product levels (metabolics), and gene activity by RNA level (transcriptomics).