Switching to a vegan diet changes measurable things on a fairly predictable schedule. The gut microbiome shifts within a day. Cholesterol, weight and blood pressure move over weeks to months. And a short list of nutrients — vitamin B12 first among them — becomes a planning problem that does not go away.
What follows is a timeline drawn from the trials and the large cohorts, including the findings that do not flatter plant-based eating: a fracture signal, a stroke signal, and several effects smaller than the headlines suggest. The B12 section is worth reading twice — that is the only mistake here that can genuinely hurt someone.
The first two weeks: the gut moves first
The fastest documented change is bacterial. In a 2014 crossover study in Nature, David and colleagues moved 10 subjects between an entirely animal-based and an entirely plant-based diet for five days each, and gut communities diverged within a single day — bile-tolerant genera up on the animal arm, fermenters such as Roseburia up on the plant arm. Ten people is a very small study, and the shift reversed within days of normal eating resuming.
Fiber is the engine, and it is also what makes the first fortnight uncomfortable. Adequate intake is 25 g/day for women and 38 g/day for men; US intake averaged 17 g/day in 2017–2020, and about 94% of adults fall short. EPIC-Oxford recorded 28.9 g/day of non-starch polysaccharides in vegans against 21.7 g in meat eaters. StatPearls describes the predictable result of closing that gap too fast — bloating, gas and cramping — and the fix: increase by 2–3 g every few days. One honest note on the scale: some of the early loss is glycogen and its bound water, not fat. We have not found a well-sourced figure for how much, so we will not invent one.
The twin trial that held genetics constant
The cleanest piece of evidence was built to remove the confounder that muddies most diet research. In a randomized trial published in JAMA Network Open in November 2023, Landry, Gardner and colleagues at Stanford randomized one twin in each of 22 identical pairs — 44 adults — to a vegan diet and the other to an omnivorous one for eight weeks.
Because each pair shared genetics and largely shared upbringing, the comparison is unusually tight. LDL cholesterol in the vegan twins fell from 110.7 to 95.5 mg/dL, against 118.5 to 116.1 mg/dL in their siblings. Fasting insulin dropped about 20%, and the vegan twins lost 4.2 lb more. The detail most coverage skipped is the comparator: the omnivorous arm ate a healthy diet including meat, eggs and dairy. This was vegan measured against good, not against fast food.
One to six months: what the controlled trials found
Over a longer horizon the pooled randomized evidence is consistent, and the effects are modest.
- Weight. Huang and colleagues, Journal of General Internal Medicine 2015: 12 randomized trials, 1,151 participants, median 18 weeks. Vegetarian diets −2.02 kg; the vegan subset −2.52 kg.
- Lipids. Koch and colleagues, European Heart Journal 2023: 30 randomized trials. Total cholesterol −7%, LDL-C −10%, apolipoprotein B −14%. Triglycerides did not change (p=0.11), a null result that rarely survives into summaries.
- Blood pressure. Yokoyama and colleagues, 2014: across seven controlled trials and 311 participants, systolic pressure ran 4.8 mmHg lower and diastolic 2.2 mmHg lower.
- Inflammation. Bell and colleagues, 2026: seven randomized trials, 541 participants, mean CRP difference −1.13 mg/L.
Where the evidence is thinner than the headlines
Some widely repeated findings are weaker than they look. A 2022 meta-analysis reporting a large improvement in insulin resistance drew on six studies and 303 participants, and measured change within-group against baseline in an obese population rather than against a control arm — a design that cannot separate the diet from the weight loss.
The big cohorts cut both ways, and EPIC-Oxford published both sides. Across 48,188 participants followed 18.1 years, Tong and colleagues reported in the BMJ in 2019 a lower rate of ischemic heart disease among vegetarians and vegans (HR 0.78; about 10 fewer cases per 1,000 over ten years) and a higher rate of total stroke (HR 1.20, roughly 3 more per 1,000), driven by hemorrhagic stroke. All-cause mortality did not differ significantly.
These are observational findings in a self-selected population, and adjustment matters: the same cohort’s diabetes difference looked large before BMI was accounted for and came to about 1% after.
The fracture finding, and the calcium behind it
The result most often left out of articles like this one deserves the opposite treatment. In a 2020 analysis in BMC Medicine, Tong and colleagues followed 54,898 EPIC-Oxford participants — 1,982 vegan — for a mean of 17.6 years. After BMI adjustment, vegans had a higher total fracture rate (HR 1.43, 1.20–1.70; +19.4 per 1,000 over ten years) and a markedly higher hip fracture rate (HR 2.31, 1.66–3.22). Vegetarians sat close to meat eaters.
The resolution comes from an earlier analysis of the same cohort. In Appleby and colleagues, European Journal of Clinical Nutrition 2007, the vegan excess fell from IRR 1.30 to 1.00 (0.69–1.44) among participants consuming at least 525 mg/day of calcium — the authors attributed the gap to vegans’ lower mean calcium intake, 848 mg/day against 1,083 mg in meat eaters. That makes it a planning problem rather than a property of plant-based eating itself. Source matters as much as total: absorption runs near 50% from kale and 30% from fortified plant milks, but about 5% from spinach.
B12: the one thing that cannot be improvised
Plant foods contain no reliable active vitamin B12. The Academy of Nutrition and Dietetics puts it without hedging: vegans must regularly consume reliable sources — B12-fortified foods or B12-containing supplements — or they could become deficient. Fermented foods and unfortified nutritional yeast do not qualify.
The data show what happens when that step is skipped. In EPIC-Oxford, mean serum B12 in men ran 281 pmol/L in meat eaters, 182 in vegetarians and 122 in vegans, with 52% of vegans classed as deficient against 7% of vegetarians. A 2013 review of 18 studies by Pawlak and colleagues found deficiency in 62% of pregnant vegetarian and vegan women. The NIH Office of Dietary Supplements sets the adult RDA at 2.4 µg/day; oral absorption is inefficient, which is why supplement doses dwarf it.
Spirulina is where this goes wrong most often. Its cobalamin is overwhelmingly pseudovitamin B12, inactive in humans, and Watanabe and colleagues concluded outright that spirulina is not suitable as a B12 source, especially in vegetarians. The same goes for chlorella and seaweed; what spirulina does and does not do is worth reading separately.
A greens powder is not a B12 solution either, unless its Supplement Facts panel declares genuine cyanocobalamin or methylcobalamin at a stated dose. If the panel does not name the form and the microgram amount, assume no usable B12 and take a supplement. Start on day one rather than waiting for symptoms: neurological damage can precede any change in the blood count.
The rest of the nutrient plan
Everything else here is manageable with forethought, and none of it is a reason not to switch.
| Nutrient | Why it becomes a gap | What the guidance says |
|---|---|---|
| Iron | Non-heme iron absorbs at 5–12% versus 14–18% from mixed diets; requirements run 1.8x higher | Pair with vitamin C; keep tea, coffee and calcium supplements away from the meal. Absorption rose almost 40% after ten weeks of adaptation |
| Zinc | Bioavailability swings from 5% to over 50% with phytate load | Intakes and serum levels run lower, though the Academy notes vegetarian serum zinc stays within the normal range |
| Iodine | A 2023 review of 11 studies and 4,421 adults found vegan median urinary iodine of 12.2–44.0 µg/L against the WHO’s optimal 100–200; no diet group reached the optimal range | Iodized salt or a 150 µg supplement — not seaweed, whose content varies wildly and can overshoot the 1,100 µg upper limit |
| Omega-3 | EPA and DHA are typically absent; ALA converts at under 15% | Algal oil supplies 100–300 mg DHA, with bioavailability equal to cooked salmon |
| Calcium | 848 mg/day in EPIC-Oxford vegans versus 1,083 mg in meat eaters | Treat 525 mg/day as a floor; favor kale and fortified plant milks over spinach |
| Vitamin D | Plasma 25(OH)D 55.8 nmol/L in vegans versus 77.0 in meat eaters; intake 0.7 versus 3.1 µg/day | Sunlight, fortified foods or a supplement, depending on latitude and season |
Protein is the gap people worry about most. Vegans in EPIC-Oxford took 13.1% of energy from protein against 17.2% in meat eaters, yet the Academy holds that protein needs at all ages, athletes included, are well met by balanced vegetarian diets. Variety and adequate total intake handle it.
Two worries you can set aside
Carnitine is not a deficiency risk in healthy vegans, despite how often it is listed as one. A vegan diet supplies roughly 1.2 mg/day against 24–145 mg for omnivores, but the body synthesizes about 14.4 mg/day against a requirement near 15 mg/day — sufficient, per the NIH, for healthy people.
Creatine is more nuanced than the usual telling. Kaviani and colleagues reported in 2020 that vegetarian plasma creatine ran about 50% lower, but muscle total creatine was only 10–15% lower, with brain creatine similar. Vegetarians did respond more strongly to supplementation, gaining about 20% over five days against 10% in omnivores — which matters to a competitive lifter and little to anyone else. Training and body composition get their own article.
A workable first 90 days
- Start B12 on day one — a supplement or genuinely fortified foods.
- Raise fiber by 2–3 g every few days, not all at once.
- Treat 525 mg/day of calcium as a floor, from sources that absorb.
- Pair non-heme iron with vitamin C, away from tea, coffee and calcium supplements.
- Source iodine deliberately: iodized salt or a 150 µg supplement.
- Consider algal DHA at 100–300 mg, especially in pregnancy and lactation.
- At six months, test B12 with methylmalonic acid, ferritin and 25(OH)D.
- Build from whole foods — every trial above tested a planned, healthy diet.
Where a greens powder fits, and where it does not
The honest case is narrow. A greens powder can deliver named nutrients at declared doses while cooking habits are in flux, and it can carry iodine or B12 if — and only if — those are formulated in and printed on the panel. The case against leaning on one is stronger. The NIH Office of Dietary Supplements is direct that supplements cannot take the place of the variety of foods important to a healthy diet, and ConsumerLab benchmarks these products at 2–10 g of powdered produce per serving — a scoop that size cannot close a 10 g/day fiber shortfall. Our guide to reading a label covers what the panel terms mean.
One safety note belongs here rather than in the small print. Greens powders are variable, often undisclosed sources of vitamin K, and vitamin K can interact seriously and dangerously with anticoagulants such as warfarin. Anyone taking one should speak to their prescriber first.
Common questions
How long before I notice anything?
Digestive changes come first, often within days, and not all of them are pleasant while fiber intake climbs. The cardiometabolic changes took eight weeks to several months; in the Stanford twin trial, LDL cholesterol had moved by week eight.
Can I rely on fortified foods instead of a B12 supplement?
Yes, if the fortification is genuine, regular and adequate. Fortified plant milks, some nutritional yeasts and fortified cereals count when the label declares B12 and the amount. Fermented foods, unfortified yeast, spirulina, chlorella and seaweed do not.
Is a greens powder enough to cover the gaps?
No, and nobody should sell one on that basis. B12, iodine, calcium, vitamin D and omega-3 each need a deliberate source, and a blended powder rarely delivers meaningful amounts unless they are added and declared.
A well-planned vegan diet has decent randomized evidence behind it for lipids, weight and blood pressure, an unflattering fracture signal that looks like a calcium problem, and one nutrient that has to be supplemented rather than hoped for. Plan the B12, the iodine, the calcium and the omega-3, and the rest largely follows.
If a greens powder is part of how you get more vegetables into the day while the cooking catches up, see what is in ours and read the panel before you buy.
About the Author
