The real cons of a vegan diet: an honest accounting

Commercial livestock feed is built on vitamin and mineral premixes: vitamin A, D3, E, selenium, and a full B-complex including B12. The animal eats the supplement. Then you eat the animal. By the time a steak reaches a fridge it has been fortified for months.

So the supplement question was never about plant based diets. It is about where in the chain the tablet goes in. For animal foods it goes in upstream, at the feed mill, billed to the farmer and printed on nobody's packet. Someone eating plants does the same job by hand, in full view, and gets told they are on a deficient diet.

Hold that thought and the standard list of downsides reads differently. Here is the list anyway, with the evidence, and what each item asks of you.

On words: veganism is an ethical position; the eating pattern is a plant-exclusive diet. This piece is about the diet.

Is a plant-exclusive diet adequate?

Yes, and the largest professional body in the field says so. The Academy of Nutrition and Dietetics' 2016 position holds that appropriately planned vegetarian and vegan diets are nutritionally adequate and appropriate for all stages of the life cycle, including pregnancy, lactation, infancy, childhood, adolescence, older adulthood and for athletes. Their 2025 update narrows its scope to nonpregnant, nonlactating adults, and reaches the same conclusion for that group.

The operative word is planned, and it carries the whole article. Everything below is about what planning involves, which comes down to about five things.

Why does eating plants seem to need more planning?

Because the planning for animal foods already happened, somewhere you cannot see it.

Take B12 specifically. A dairy cow does not eat it. She builds it in her rumen out of cobalt, and the cobalt goes into her feed at a specified rate: the requirement is set at 0.2 mg per kilo of dietary dry matter, only 3-15% of it ever becomes B12, and how much she makes depends on the ration she is given. None of that is natural or accidental. It is dosed, measured and paid for.

Someone eating plants does that job by hand: a tablet on the counter instead of a premix in a silo. Same molecule, visible instead of hidden, paid for by the eater rather than the farmer. The work is identical. Only the accounting moves.

So the comparison this question implies, a diet needing planning against one that runs itself, was never real. Nobody in an industrial food system eats unplanned. What varies is who does the planning, and whether you can see it happen.

Is B12 deficiency a real risk?

Yes, if nothing in your week supplies it, whether that is a tablet or reliably fortified food. B12 comes from bacteria rather than plants, and prolonged deficiency causes nerve damage that can become permanent.

Food can cover it. Fortified nutritional yeast runs roughly 2-8 mcg per tablespoon, fortified cereals around 0.6-2.1 mcg per serving, and most fortified plant milks carry a dose per cup. Against an adult requirement of 2.4 mcg a day, a quarter-cup of fortified yeast clears it several times over. The catch is tracking: fortification varies by brand, country and batch, and nobody audits their own intake week to week. For most people the tablet is the lower-effort route, at pennies a day and with no upper limit set, because B12 is toxic at essentially no dose and the excess is excreted.

A 2019 study of 151 Czech vegans found lab signs of deficiency in 52.9% of those taking no supplement, against 5.6% of those supplementing regularly. Those two numbers are the whole argument. Same diet, same population; the variable is the tablet.

General-population deficiency runs at roughly 5-20%, and 3-43% among older adults, in a food system where the animals are already fortified. Millions of people eating meat every week are quietly deficient, and it still gets discussed as a vegan problem.

Two things decide whether you run short: whether you take it, and whether your stomach still absorbs it well, which fades with age no matter what you eat. Steak has very little to do with it.

Do plant-exclusive eaters break more bones?

The honest answer needs two studies, not one, and the order matters.

Start with bone density, which is what predicts fracture. Comparisons of plant-based and omnivorous diets find no significant difference in bone mineral density once body weight and food quality are accounted for. On the measure that matters most, the diets look the same.

Then the fracture data, which is where the alarming number comes from. The EPIC-Oxford study followed nearly 55,000 UK adults recruited between 1993 and 2001, including 1,982 vegans. Hip fracture risk came out more than twice as high in vegans, at 14.9 extra cases per 1,000 people over ten years. Total fracture risk was 43% higher.

That deserves to be taken seriously and also read properly. The vegans in this cohort were lighter than the meat eaters, and ate less calcium and less protein. All three weaken bone in anybody. The associations were stronger before body weight was accounted for, and adjusting further for calcium and protein shrank the result again without eliminating it. The authors closed by saying vegan bone health needs more research, which is the correct place to leave it.

What the two studies together say is this: eat too little, skimp on calcium and protein, and your bones suffer. That is true on any diet. It shows up in this cohort because these particular vegans were doing it, not because plants do it.

The assumption underneath the headline is worth testing too. If dairy were straightforwardly protective, the populations drinking most of it would break fewest bones. They do not. A Swedish cohort of more than 61,000 women followed for around twenty years found each daily glass of milk associated with slightly higher fracture risk in women, while cheese and fermented milk went the other way at 10-15% lower risk. That study is observational too, and its own authors cautioned against reading causation into it. What neither supports is the tidy story that milk builds bones and its absence breaks them.

Where plant eaters get calcium, and where they lose it

Getting it is arithmetic, and it fits inside an ordinary day. Here is 1,021 mg without a supplement:

Breakfast, lunch, dinner. Nothing bought from a supplement aisle and nothing anyone would call a special diet.

Absorption is where plants do better than their reputation. Kale absorbs at roughly 49-53%, bok choy 52%, broccoli 48%, all ahead of milk's 32%. Calcium-set tofu runs 31%, fortified soy milk 21%. Spinach and chard are the trap: oxalate binds most of their calcium and only about 5% gets absorbed, which is why they look impressive on a label and do almost nothing.

Greens do a second job too. Kale and other leaves carry the vitamin K that bone mineralisation runs on, plus magnesium and potassium, and roughly 60% of the body's magnesium sits in bone.

Keeping it is the other half. Calcium leaves in urine, and diet drives how much:

• Salt is the big one. Sodium and calcium compete for reabsorption in the kidney, and every 2,300 mg of sodium takes roughly 40 mg of calcium out with it. That bites hardest for people already low on calcium.
• Caffeine adds a smaller, dose-proportional loss for a few hours after drinking, and the body never adapts to it.
• Protein raises urinary calcium as well, though the loss is offset once calcium intake is adequate, which puts the fix back on intake.

So the practical picture is plain: eat enough calories, take calcium from the low-oxalate list, keep salt moderate, and the fracture signal in that cohort stops looking like a plant problem.

Which nutrients need attention?

Iron. Heme iron from meat absorbs at roughly 25-30%. Non-heme iron from plants absorbs at around 3-5% eaten alone. That gap is why recommended intake for vegetarians is set at 1.8 times the standard figure: about 14 mg a day for men and postmenopausal women, and about 32 mg for menstruating women.

Vitamin C helps, and it is worth being exact about how much. In a controlled study using a liquid formula meal containing 4.1 mg of non-heme iron, ascorbic acid raised absorption from 0.8% to 7.1% across a dose range running from 25 mg up to 1,000 mg. At a realistic 25 mg, a squeeze of lemon, absorption improved by about 1.65 times. Useful, free, and some way short of turning lentils into steak.

So the lever that actually moves the number is total intake. Absorption also climbs on its own when body stores run low, which means plant-heavy eaters are less linearly disadvantaged than the raw percentages imply. Here is what 14 mg looks like on a plate:

Any three of those clears 14 mg. Oats at breakfast, lentils at lunch, a handful of seeds in the afternoon. Take the vitamin C where it is free: lemon over the lentils, tomato in the sauce, berries on the oats.

The 32 mg target for menstruating women is genuinely harder. Reaching it from food alone means eating most of that list most days. This is the one place besides B12 where fortified foods or a supplement earn their keep, and where a ferritin test tells you more than any amount of meal planning.

Zinc. Zinc runs immune function, wound healing and your sense of taste. The mechanics match iron. Phytate in grains, beans and seeds binds zinc and holds on, so plant zinc absorbs at roughly 15-26% against 40-60% from animal foods. Hence the standard advice to aim about 50% above the usual target: roughly 16 mg a day for men rather than 11, and 12 mg for women rather than 8.

Then take the phytate back off. Soaking beans overnight, choosing sourdough or sprouted bread over flatbread, and cooking rather than eating raw all break phytate down and release zinc that was already sitting there. This is how these foods have been prepared for centuries, and this is why.

Iodine. Iodine builds thyroid hormone, and running short of it caused one of the great public health disasters of the twentieth century: goitre, and in severe cases permanent developmental damage in children. Which is exactly why salt was iodized. That is the entire reason the programme exists, and it worked. Salt iodization now runs in more than 120 countries, and the number classified as iodine deficient fell from 113 in 1993 to 19 by 2019.

So for most people the answer is already in the cupboard. Iodized table salt carries about 45 mcg of iodine per gram, so a quarter teaspoon delivers roughly 70 mcg against a 150 mcg daily target. Half a teaspoon across a day of cooking covers it. Two things break that, and neither has anything to do with eating plants: the salt in your kitchen may not be iodized, since sea salt, pink Himalayan and most finishing salts are not; and most dietary sodium comes from processed and restaurant food, which is typically made with non-iodized salt.

That sits awkwardly against keeping sodium moderate for calcium retention, and the resolution is that these are different salt. The sodium problem is packaged food. The iodine solution is the shaker you cook with. Note the scale of who this affects: the WHO has estimated around 2 billion people worldwide with insufficient iodine intake, roughly 30% of the planet. The NIH does list people eating little or no dairy, seafood and eggs among at-risk groups. But iodine deficiency was never a vegan problem. It is a geology problem, because iodine is scarce in soil far from the sea, and the fix was fortification for everybody.

Vitamin D. This tracks sun exposure, latitude, skin tone and age rather than diet type, which makes it a winter problem rather than a vegan one. Roughly half the world is short of it. Fortified plant milks carry close to the same 120 IU per cup that dairy milk does, so a cup on cereal does most of the work, and a supplement handles it for anyone not eating fortified foods.

And mushrooms, which are genuinely interesting. Mushrooms contain ergosterol, concentrated in the gills, which converts to vitamin D2 under UV light exactly as skin does. Fifteen minutes of midday sun on sliced mushrooms raises the content to about 704 IU per 100 g. It climbs roughly linearly for the first hour, then plateaus, and declines after about two hours, so longer is not better. Slice them, lay them gill-side up, give them an hour of midday sun. A free source of a nutrient half the planet is short of.

Does "plant-based" automatically mean healthy?

Not automatically, and the popular version of this point is wrong in an interesting way. The usual line is that swapping animal junk food for vegan junk food achieves nothing. The trial evidence disagrees. SWAP-MEAT put 36 adults on plant-based meat for eight weeks and animal meat for eight weeks, holding everything else in the diet constant. On the plant phase, LDL cholesterol came out at 109.9 mg/dL against 120.7, weight was lower, and TMAO improved. Most of those plant products count as ultra-processed by the NOVA classification, and they still won, largely because they contain fibre and animal meat contains none.

The same holds at the whole-food end, faster than people expect. Switching to whole-food plant-based eating typically means a sharp jump in fibre and a drop in LDL within weeks, without anyone tracking a single nutrient.

So the honest version is narrower: a diet of vegan burgers, chips and soft drinks is still a poor diet. It is just measurably less poor than the animal-based version of the same meal. The same logic runs in the other direction and rarely gets applied. Milk is sold on calcium content, yet the cohort evidence above does not show the fracture protection that marketing implies. "Contains calcium" and "improves bone outcomes" are different claims.

What about the costs that never appear in a study?

Eating out is harder. Travelling is harder. Family meals become a negotiation. Some people find the social friction wears them down faster than any nutritional issue does, and leaving it off the list because it does not appear in a journal would be dishonest.

One cost that does not apply, despite being assumed constantly: money. Oxford's modelling study across 150 countries, published in The Lancet Planetary Health, found vegan diets the cheapest option in high-income countries, cutting food costs by 21-34% against the current typical diet. Vegetarian came second at 27-31%. The expensive part of a Western diet is the meat and dairy in it.

So which rules matter?

Five things. Most take seconds a day.

1. Take B12. Daily or weekly, no exceptions. It is the only genuinely non-negotiable item on this page.
2. Eat enough. The fracture signal travelled with low body weight. Under-eating is a far more common failure than any specific nutrient gap.
3. Cook most of your food. Cooking raises the calorie density and digestibility of plants. Raw-dominant patterns are where low body weight and poor bone outcomes cluster.
4. Cover calcium, vitamin D and iodine deliberately. Fortified plant milk can handle all three at once, but check the label; plenty of brands fortify with some and not others.
5. Get bloodwork at six and twelve months, then periodically. It replaces guessing with knowing.

Pair iron-rich foods with vitamin C while you are at it. That is the whole list: a tablet, a full plate, a hot meal and a label check.

Compared to what?

A list of costs with nothing to compare it against is half a ledger. The largest study of its kind, Poore and Nemecek's 2018 analysis in Science, covering roughly 38,700 farms across 119 countries and 40 products, found that moving to a plant-based diet would cut global agricultural land use by around 76%, from roughly 4 billion hectares to 1 billion, and reduce food-related emissions by up to 73% depending on where you live. Even the lowest-impact beef has a higher carbon footprint than most plant foods.

The other cost of the alternative does not come as a statistic. It is what the supply chain looks like from the inside.

Source · Dominion (2018)[ Graphic ]

Standard-practice footage · hidden by default

Show media

You choose whether to view it.

None of this is an emotional appeal. It is the rest of the ledger, documented and sourced and ordinarily kept out of frame.

The honest bottom line

A plant-exclusive diet is nutritionally adequate at every stage of life. That is the position of the Academy of Nutrition and Dietetics, and it carries one condition: that the diet is planned.

Planning sounds like a chore and is mostly a fortnight of learning. Buy the fortified plant milk once and you keep buying it. Put the B12 next to the kettle and it stops being a decision. Learn three meals with tofu or beans in them and you are done, because nobody cooks forty different dinners. The people who find this hard are usually in week two. The people who have been doing it a year are not thinking about it at all.

The real risks are under-eating, eating raw-dominant, and skipping B12. Handle those and the evidence says you will be fine. It costs less than the diet you were eating, and after the first month you stop noticing you are doing anything.