Background: What is the 3 Months 3 Ways experiment?
Between September 2018 and December 2018, I experimented with a different diet each month while undertaking intensive strength training. The diets were: Paleo, Keto, and Carnivore. Throughout the experiment I shared updates and insights on Instagram [https://www.instagram.com/damcgrath/]. The dietary effects on my body were monitored using a range of tools including the FreeStyle Libre [https://www.freestylelibre.com.au/] for 24/7 blood glucose, Oura ring [https://ouraring.com] for sleep, Sweatbeat app [http://www.sweetwaterhrv.com], Polar H10 chest strap [https://www.polar.com/au-en] for Heart Rate Variability (HRV), an extensive panel of blood work through iMedical [https://imedical.com.au], and strength and body composition assessments from Recomp HQ [https://recomp.com.au], with dexascans from MeasureUp [http://www.measureup.com.au/melbourne/]. My food consumption, macro and micronutrients were captured through Cronometer [https://cronometer.com]. I trained under a strength coach at Recomp HQ. Training involved two intense weightlifting sessions a week with strength gains and calculated one rep max tracked for each exercise each session. Changes in body composition were measured fortnightly using a 9- site calliper pinch test. Baseline body composition and strength assessments were done before commencing the program, throughout, and at the end of each of the experiment periods. No other exercise is done for the three months. Sauna and infrared protocols were used throughout to support recovery.
I have eaten a variety of diets over the years. Some for weight loss, some for energy, some for optimising my performance and other health reasons. When I dived into the data at times and experienced different levels of ‘success’, I felt they all worked to some capacity, and I realised how adaptive my body is. I was curious to learn which of these different ways of eating were ‘optimal’ for me by applying a comparative experimental protocol. Naturally, ‘optimal’ is subjective depending on what outcomes I sought. I have shared some results from my first two months: Paleo [https://www.diannemcgrath.com.au/3-months-3-ways-paleo/], and Keto [https://www.diannemcgrath.com.au/keto-month-2-of-3months3ways/]. For month 3 I ate a Carnivore diet. From mid-November to mid-December 2018 I ate a Carnivore diet starting at 1938 calories, gradually titrating down to 1584 calories.
What did I eat?
With the exception of coffee, not a plant-based food passed my lips, even spices. Figure 1 lists the most commonly consumed food.
Some insights from month three’s end
Table 1. Body composition and strength changes
End Month 1 (%change vs baseline)
End Month 2(%change vs baseline)
End Month 3(%change vs baseline)
% body fat
Lean muscle mass
Strength (calculated 1RM)
Carnivore definitely saw good strength gains, and while Paleo (month 1) saw the largest change in a single month, whichever diet I would have tried first would likely to have seen the largest gains. I was impressed with the strength changes from Keto. In Keto I managed to maintain or increase strength with a low-calorie intake. And Carnivore illustrated that carbohydrates consumed are not needed to obtain strength increases, as my carbohydrates were lowest out of the 3 months. I put on weight over the month: an extraordinary 2 kg of lean muscle and surprising 1 kg of fat. My final weight and calliper measurements were taken the day after returning from an interstate flight. During the 3-month experiment I travelled a bit, and found that the day after any air travel, my % body fat was up, with calliper measurements were slightly higher around the mid-section, and the skin folds measured were ‘watery’ – something that disappeared within a few days. So potentially my final measurements may have in reality been slightly lower in body fat than this table indicates. It is not surprising that I had such an increase in lean muscle mass. The research showed that increased protein and energy intake led to increased muscle cross-sectional area of the quadriceps femoris in ‘older’ women doing resistance training. Eat more, eat meat, grow muscle.
Sleep was challenging the first 2 weeks of carnivore. I woke frequently during the night, had a warmer body temperature and slightly higher resting heart rate. A large study in middle-aged Japanese workers showed that when protein consumption was >19% people experienced greater difficulty maintaining sleep. To prevent sarcopenia (muscle wastage and strength loss) 25-30 grams of high-quality protein per meal is recommended – i.e.75-90g/day.. I was eating roughly 114g of protein/day. So, with 30% of my diet being protein, I was very likely to experience wakefulness. There seems to be a sweet spot for quality sleep with protein consumption sitting between 16-19%. Below 16% of energy consumption, and people suffer from difficulty falling asleep. My strength coach and I dropped my total calories marginally but did not alter the protein ration any significant amount. Instead, I experimented with a pescatarian-Carnivore week, which improved my sleep. While I was in low ketosis early in the Carnivore month, the pescatarian week accidentally pushed me into the nutritional ketosis range. This was likely due to the big jump in omega 3’s from fish and seafood. Oooops. Research showed that when obese women consumed higher amounts of high Omega-3 PUFAs, ketone (beta-hydroxybutyrate) production increased by 50% in 1-3 weeks. While I’m not obese, it could be expected in most people, regardless of gender or age. A number of studies in epileptic patients and in animal models show similar results.[e.g. 4, 5] Interestingly, when I returned to predominantly red-meat based Carnivore eating the next week, I not only remained in ketosis, but continued to see my ketones rise marginally. Becoming ketogenic on an almost zero carbohydrate diet is not surprising and can be an aim of this way of eating. When the body has insufficient glucose (from carbs) for fuel, it both produces its own and looks to fat, and if fat consumption doesn’t provide that energy, hello body fat and bring on the ketone party! I was consuming only 2% of my diet as carbs (8-10g of carbs/day). The body breaks carbs down to glucose units, which are broken down further before being in an energy form. At any one time, the blood stream of a 70kg person has roughly 4g of glucose circulating in it; and two hormones, insulin and glucagon, act to keep that number the same, no matter how many carbs you eat. More? It gets shunted out of the blood stream and stored in muscle tissue and then as fat. Less? The body will produce its own glucose to top up the tank. Free fatty acids from body fat fuel this activity,  and fatty acids can also be turned into glucose. So bye bye body fat with low carb-eating.
The most significant change during Carnivore was LDL particle size. What are LDL particle sizes? LDL can typically be large (not damaged LDL – all good here!) or small (damaged LDL – demons!!!). I was happy to see that the number of my small dense LDL was 44% lower than my baseline measurements. And I saw a dramatic change in the sizes of my small dense LDL (there are 5 different sizes from small dense LDL 3 (the ‘largest’ of these) – small dense LDL 7 (the smallest and most concerning)). During the month of Carnivore, I reduced all my small dense LDL 4 to zero and dropped the amount of small dense LDL 3 by 30%. I had no small dense LDL 5-7. What does this mean? In one month of eating a purely carnivorous diet I decimated my coronary risk. We need cholesterol, HDL and LDL, for producing hormones, cell membranes, bile production and more. If we do not eat cholesterol, our body makes it. In fact, we rarely eat enough cholesterol, and our body makes 75% of our cholesterol needs in the liver. We top up the tank with what we eat, and if we eat a bit more than the 25%, our body makes a little less. We also only absorb half of the cholesterol that ends up in the small intestine (the rest passes in the faeces), so our body does a lot to make sure we have enough cholesterol for our needs. This also goes to show the ‘high cholesterol/low cholesterol’ mindset some folks have about food may be unfounded. The common myth is that HDL (high density lipoprotein) is the ‘good’ cholesterol, and LDL (low density lipoprotein) is the ‘bad’ cholesterol. I say ‘myth’, as the story is not that simple, and not necessarily supported by causation scientific research, mostly by correlation or association. HDL does a heap of useful stuff, including removing cholesterol from built up deposits in blood vessels, taking cholesterol that’s in cells back to the liver for excretion, and more. However, the benefits of HDL on reducing atherosclerotic cardiovascular disease has come from epidemiology studies, and not from studies that have tested whether it really helps. In fact, a 2014 study in the New England Journal of Medicine showed that increasing people’s HDL does nothing to improve their cardiovascular health outcomes; and that it is something called ‘cholesterol efflux capacity’ that matters, not how high your HDL number is. ‘Cholesterol efflux capacity’ is how well HDL removes cholesterol from cells. I.e. stop chasing numbers – start chasing effect.
What about LDL – isn’t that’ bad’?
LDL is actually important and necessary for our bodies. Let’s take a step back – VLDL (very low-density lipoprotein) transports cholesterol from where it is produced in the liver to the different parts of the body needing it (e.g. brain cells!). It also carries the fatty acid energy packs (triglycerides) and the fat-soluble vitamins (A, D, E and K) around the body at the same time – kind of a ride share arrangement. Once the VLDL has delivered all the energy (triglycerides) to where the body needs it, VLDL changes shape to be LDL. A bit like HDL, LDL numbers may not necessarily be an issue. As you can see, it may be indicating how much fatty acid energy has been delivered around the body. Older medical thought was that a high LDL number was ‘bad’ and associated with atherosclerosis. However, LDL ‘targets’ are starting to disappear from health guidelines and ratios, or other more relevant markers are being used, in particular LDL particle size. Large fluffy (or large buoyant) LDL particles seem to have a protective function, whereas small dense LDL are the ones associated with atherosclerosis. We all have a mix of the different sizes of LDL at any time. Small dense LDL have been damaged due to processes such as oxidation and seem to be able get into damaged arterial walls and accumulate to form the plaques we think of for cardiovascular conditions. Large fluffy LDL doesn’t accumulate in arterial walls – only these little blighters. Of course, this means the arterial walls need to be damaged first – that happens due to high blood pressure, and high blood glucose. Neither of these are a problem for me. As you’ll see below, my blood glucose during Carnivore was almost non-existent.
This image was a typical Carnivore day for my blood glucose. It was so stable all month. Not dips or peaks, except tiny blips when I used the sauna. Unsurprisingly my energy levels throughout the month were constant. The fact that the levels were below the ‘normal’ levels were not a concern. As I was in mild ketosis during Carnivore, and then in moderate ketosis when I switched in that week of seafood, I had more than enough energy from fat. Last month’s Keto blood glucose looked very similar in that it was a flatline near the bottom of the range, dipping below the line on occasion. However, Carnivore’s flatline was so low, it was striking, with many days where I the line did not go into the normal threshold at all.
What happened next?
I ate whatever I wanted the month after this experiment AND TRACKED NOTHING! What diet have I adopted after all this? I’ve taken a strategic approach to how I eat. Two months on, and I am predominantly ketogenic with a strong carnivore leaning and add a good amount of sweet potato into the mix on days I lift heavy. So, in 3 months 3 ways I worked out what a balanced diet looks like for me based on a range of solid and defensible metrics. What does a balanced diet look like for you? #testitdontguessit
E Tanaka et al. 2013, ‘Associations of protein, fat, and carbohydrate intakes with insomnia symptoms among middle-aged Japanese workers’, Journal of Epidemiology. 23(2): 132–138 [URL: https://www.ncbi.nlm.nih.gov/pubmed/23419282].
D Paddon-Jones & BB Rasmussen 2009, ‘Dietary protein recommendations and the prevention of sarcopenia. Protein, amino acid metabolism and therapy’, Current Opinion in Clinical Nutrition and Metabolic Care. 12(1):86–90 [URL: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2760315/].
Kunes?ova´ et al. 2006, ‘The influence of n-3 polyunsaturated fatty acids and very low calorie diet during a short-term weight reducing regimen on weight loss and serum fatty acid composition in severely obese women’, Physiological Research. 55(1):63-72 [URL: https://www.ncbi.nlm.nih.gov/pubmed/15857162].
BS Fuehrlein et al. 2004, ‘Differential metabolic effects of saturated versus polyunsaturated fats in ketogenic diets’, The Journal of Clinical Endocrinology & Metabolism. 89(4):1641-5 [URL: https://www.ncbi.nlm.nih.gov/pubmed/15070924].