Melbourne, VIC (03) 9005 8035

What is the 3 Months 3 Ways experiment?

Between mid-September 2018 and mid-December 2018, I ‘m experimenting with a different diet each month while undertaking intensive strength training. The diets are: Paleo, Keto, and Carnivore. Throughout the experiment I share updates and insights on Instagram []. The effects on my body from the different diets are being monitored using a range of tools including the FreeStyle Libre [] for 24/7 blood glucose, Oura ring [] for sleep, Sweatbeat app [], Polar H10 chest strap [] for Heart Rate Variability (HRV), an extensive panel of blood work through iMedical [], and strength and body composition assessments from Recomp HQ [], with dexascans from MeasureUp []. My food consumption, macro and micronutrients are being captured through Cronometer []. My strength training is under a coach at Recomp HQ. It involves two very intense weightlifting sessions a week with strength gains and calculated one rep mass tracked for each exercise each session. Changes in body composition are 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 are used throughout to support recovery.

  • Text Hover
  • Text Hover

Why am I doing it?

I have eaten a variety of diets over the years. Some for weight loss, some for energy, some for optimising my performance or 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 comparing them through an experimental science-based protocol. Naturally, ‘optimal’ is subjective depending on what outcomes I may seek. I started with the ‘higher’ carbohydrate containing diet in the trio: Paleo.

  • Text Hover

Some insights from month one’s end


Eating more carbs than usual effected my sleep. I found myself waking in the early morning more frequently than usual. This was confirmed by Oura data, where waking time periods were identifiable, and verified with Freestyle Libre data where larger dips in overnight blood glucose occurred. I ended up adjusting how I spread my macros across my day’s meals, so I had a higher fat proportion at night and loaded more of the carbs earlier in the day where feasible. This helped reduce how often I woke during the night. It is not a surprising outcome. Research in normal weight women who consume higher carb loads, especially processed carbs, have worse sleep than those who do not. [1,2]


I had no difficulty exercising on this diet and had some very large strength gains and alterations to my body composition (see Table 1). The more personal challenges were: a) the shift to exercising only twice a week, b) only exercising in the afternoon after eating multiple times (I usually exercise fasted in the morning), and c) eating so much protein – all very different to my usual fitness approach. However, as my goal was strength gains, the dietary approach of a higher protein diet with low-ish carbs makes sense. Female strength athletes are less glycogen reliant during training, and don’t respond to carbs post-training as effectively as men, meaning that strength gains for women may be more likely to come from protein and fat.[3]

Blood panel/biomarkers:

I saw many changes in my blood panel. Before I share the two most remarkable changes, I want to preface that when baseline bloods were drawn, I had just completed 3-4 months of a carnivore diet with exceptionally low carbohydrates (<10g total daily) and daily intermittent fasting (IF) with a 4-6-hour eating window. In some instances, the changes in biomarkers may reflect ceasing this diet or the IF. Estradiol: My estradiol (E2) increased by a whopping 1,190%. This is remarkable as usually estradiol declines for women quite dramatically in the late 40’s and beyond (menopause).[4] But this result is also not surprising due to a few possible factors associated with the month’s activity, including but not limited to: I ended a 3-4-month period of intense intermittent fasting. Fasting for lean women with 20% body fat or less can lead to reduced E2, and my baseline levels at the start of this experiment were around 18%, where I sit naturally.[5] My body fat percentage tends to sit at around 18% naturally. So ending fasting may have reduced the pressure on my E2. The following have been shown to significantly increase estradiol:Intensive strength (or endurance) training.[6] A significant increase in animal protein consumption – my protein consumption was titred up to over 2:1 g protein/kg lean body mass across the month. Women who consume proportionally higher vegetable diets have a lot less plasma estrogen[7] Sauna can lead to a significant spike in prolactin expression in women (but not significantly in men);[8] and moderate amounts of prolactin are associated with higher estrogen during the follicular phase of menstruation (when my blood was drawn for analysis),[9] and Yoga/meditation can increase serum estradiol in perimenopausal women – which I do daily and increased over the month.[10] Growth Hormone: Like my E2, my Growth Hormone (GH) saw an extraordinary increase over baseline of 1,500%. Similar to the change in E2, a number of factors may have contributed to this remarkable change, including but not limited to increasing: Estrogen levels (see above)[11] Protein consumption[12] Resistance training[12,13] Sauna[14] Cold exposure,[15] and Achieving an even lower percentage of body fat (very strong trend, but not statistically significant).[16]

  Baseline End Month 1 (%change)
Body comp    
Weight 61.6kg 62.7kg (1.8%)
% body fat 18.3% 16.1% (-12.0%)
Lean muscle mass 50.3kg 52.6kg (4.6%)
Fat 11.3kg 10.1kg (-10.6)
Strength (calculated 1RM)
Squat 60kg 69kg (15%)
Bench 30kg 50kg (66.7%)
Dead 100kg 110kg (10%)
  • Text Hover
Mental health:

I was curious to see how having a slightly higher carb-load than usual would affect my mental health. I tracked this subjectively through using a simple app called HappyTracker that presents 3 different emojis to select that best reflect my mood at the time. I did this at the same time every day: first thing in the morning after checking my waking blood glucose, HRV and sleep. I experienced 5 mornings out of the month where my mood was ‘meh’. Every other morning my mood was ‘happy’. I did not experience any ‘sad’ mornings. My partner also commented that I seemed quite happy on a paleo diet. As I was still relatively low carbohydrates, this was not a large surprise. My past experience eating ‘normal’ diets that are much higher in carbs often resulted in a lot more ‘down’ days, anxiety and other mental health challenges. But whether that was due to lower carbs or higher cholesterol is up for debate. Research seems to suggest that healthy middle-aged women with lower cholesterol have a higher prevalence of depression.17 What was interesting to note was that my diet on the ‘meh’ days was much higher proportionally in green vegetables (kale, spinach, etc) compared to non-green vegetables (sweet potato) than on other days. A potential coincidence? TBD.

What’s next?

Month 2: Keto! It will be interesting to see how my body responds to dropping the carb load considerably, and the calories.


  1. MP St-Onge, A Mikic and CE Pietrolungo 2016, ‘Effects of Diet on Sleep Quality’, Advances in Nutrition, 7(5): 938-949 [URL:].
  2. MP St Onge et al. 2016, ‘Fiber and Saturated Fat are Associated with Sleep Arousals and Slow Wave Sleep’, Journal of Clinical Sleep Medicine, 12(1): 19-24 [URL:].
  3. JS Volek, CE Forsythe and WJ Kraemer 2006, ‘Nutritional aspects of women strength athletes’, British Journal of Sports Medicine, 40(9): 742-748 [URL:].
  4. BL Lasley et al. 2002, ‘The relationship of circulating dehydroepiandrosterone, testosterone, and estradiol to stages of the menopausal transition and ethnicity’, The Journal of Clinical Endocrinology & Metabolism, 87(8), pp.3760-7 [URL:].
  5. R Alvero et al. 1998, ‘Effects of fasting on neuroendocrine function and follicle development in lean women’, The Journal of Clinical Endocrinology & Metabolism, 83(1), pp.76-80 [URL:].
  6. JL Copeland, LA Consitt and MS Tremblay 2002, ‘Hormonal responses to endurance and resistance exercise in females aged 19-69 years’, The Journals of Gerontology Series A, 57(4), pp.B158-65 [URL:].
  7. SL Gorbach and BR Goldin 1987, ‘Diet and the excretion and enterohepatic cycling of estrogens’, Preventative Medicine, 16(4), pp.525-31 [URL:].
  8. D Ježová, R Kvetňanský and M Vigaš 1994, ‘Sex differences in endocrine response to hyperthermia in sauna’, Acta Physiologica Scandinavica, 150(3), pp.293-298 [URL:].
  9. KP McNatty 1979, ‘Relationshop between plasma prolactin and the endocrine microenvironment of the developing human antral follicle’, Fertility and Sterility, 32(4), pp.433-8 [URL:].
  10. R Khadka et al. 2013, ‘Yogic practices can increase estrogen and progesterone levels and heart rate variability in perimenopausal women’, 37th Congress of IUPS (Birmingham, UK), Proc 37th IUPS, PCA264 [URL:].
  11. AG Frantz and MT Rabkin 1965, ‘Effects of estrogen and sex difference on secretion of human growth hormone’, The Journal of Clinical Endocrinology & Metabolism, 25(11), pp.1470-80 [URL:].
  12. DS Rushdy et al. 2018, ‘Impact of resisted exercises and whey protein on growth hormones and testosterone in normal subjects’, Journal of Medical Sciences, 18, pp.27-33 [URL:].
  13. BW Craig, R Brown and J Everhart 1989, ‘Effects of progressive resistance training on growth hormone and testosterone levels in young and elderly subjects’, Mechanisms of Ageing and Development, 49(2), pp.159-169 [URL:].
  14. K Kukkonen-Harjula and K Kauppinen 1988, ‘How sauna affects the endocrine system’, Annals of Clinical Research, 20(4), pp.262-6 [URL:].
  15. G Gerra et al. 1992, ‘Sex-related responses of beta-endorphin, ACTH, GH and PRL to cold exposure in humans’, Acta Endocrinologica, 126, pp.24-8 [URL:].
  16. A Weltman et al. 1994, ‘Relationship between age, percentage body fat, fitness, and 24-hour growth hormone release in healthy young adults: effects of gender’, The Journal of Clinical Endocrinology & Metabolism, 78(3), pp.543-548 [URL:].
  17. M Horsten et al. 1997, ‘Depressive symptoms, social support , and lipid profile in healthy middle-aged women’, Psychosomatic Medicine, 59(5), pp.521-528 [URL:].