Food · lesson 6 of 11
The major diets, compared
Mediterranean, DASH, keto, plant-based, paleo, carnivore — the real evidence, even-handed.
8 min read · reviewed October 2026
Every few years a new diet becomes The Answer, and the one before it becomes The Enemy. It's noisy and tribal — but underneath the marketing there's actual evidence, and it varies a lot from one approach to the next.
This lesson lines up the big six honestly: what each one gets right, where it overreaches, and who it tends to suit. The goal isn't to crown a winner. It's to see that most of the good ones overlap heavily — and that 'best diet' usually means 'the high-quality pattern you can actually keep.'
Adherence — the hidden variable behind every diet claim
Head-to-head diet trials often end in a near-tie on weight and metabolic markers — not because the diets are identical, but because the biggest predictor of results is whether people actually stuck to it. A 'perfect' diet you quit in a month loses to a merely good one you keep for years.
How it works →
Any diet that creates a sustained calorie deficit and improves food quality will produce results — but only while it's followed. Restrictive or extreme approaches often work impressively at first, then fail on adherence: the more a diet fights your preferences, social life, and budget, the harder it is to maintain, and the gains reverse. So 'how effective is this diet?' is in practice inseparable from 'how livable is it for this person?' — which is why the same diet can be a triumph for one person and abandoned by the next.
What the studies show →
In long diet-comparison trials, dropout and poor adherence are pervasive, and within every group the people who stick closest to the plan get the best results regardless of which plan it is. Reviews repeatedly conclude that for body weight, adherence and overall diet quality explain more of the variation than the specific macronutrient pattern. Read every diet claim through that lens: how sustainable is this for a real life?
The big six, at a glance
What it is: abundant vegetables, fruit, legumes, whole grains, nuts; olive oil as the main fat; fish regularly; modest dairy/poultry; little red and processed meat. Gets right: the deepest evidence base in nutrition — trials and cohorts point the same way (heart, metabolic, cognitive). Flexible, food-positive, sustainable. Watch: it's a pattern, not a rulebook, so 'Mediterranean' on a label means little. Best fit for almost anyone.
What it is: a Mediterranean cousin engineered to lower blood pressure — heavy on vegetables, fruit, whole grains, legumes, low-fat dairy; deliberately low sodium, added sugar, saturated fat. Gets right: randomized trials show real blood-pressure drops; great if hypertension is your concern. Watch: the low-fat-dairy emphasis is a 1990s artifact; the core works regardless. Suits people managing blood pressure.
What it is: cuts carbohydrate sharply (keto: very low, enough to shift fuel toward ketones); more protein and fat. Gets right: effective for short-to-medium-term weight loss and can markedly improve blood sugar and triglycerides — genuinely useful for many with type 2 diabetes or insulin resistance. Watch: long-term advantage over other diets shrinks once calories/adherence are matched; quality matters (keto can be all bacon or all avocado-and-fish). Hard to sustain for some; LDL response varies.
What it is: mostly or entirely plants; vegan excludes all animal foods. Gets right: consistently linked to lower heart-disease risk, lower LDL, and a smaller environmental footprint when built from whole plants. Watch: quality split is huge — whole-food vegan is excellent; 'junk-food vegan' (refined carbs, vegan UPF) is not. Needs deliberate B12 (and attention to omega-3, iron, zinc, iodine, calcium).
What it is: built around foods presumed available before agriculture — meat, fish, eggs, vegetables, fruit, nuts; excludes grains, legumes, dairy, refined sugar, and processed food. Gets right: by cutting ultra-processed food and added sugar, it's a real upgrade over a typical Western diet. Watch: the 'cavemen ate this' rationale is shaky, and excluding legumes and whole grains drops two of the most evidence-backed longevity foods. The benefit is mostly the UPF cut, not the ancestral story.
What it is: animal foods only — meat, organs, eggs, sometimes dairy; zero plants. Gets right: some people report short-term symptom relief (often an elimination-diet effect) or better appetite control, though very little long-term evidence exists. Watch: no fiber, no plant compounds, no long-term safety data, and it discards essentially everything the strongest evidence supports. The most extreme and least-studied of the six — approach with real skepticism.
What the head-to-head trials actually show
When researchers pit popular diets against each other, the differences in weight loss and heart markers are usually small — and they mostly come down to how well people stick to the diet, not which one it is. The clear standout for long-term health outcomes (not just weight) is the Mediterranean pattern.
How it works →
Most diets that 'work' share the same active ingredients: they cut ultra-processed food and added sugar, raise food quality, and create a modest calorie deficit — often without counting, because protein, fiber, and whole-food structure blunt appetite. Keto adds a specific metabolic shift (burning fat/ketones for fuel) that helps blood sugar and triglycerides, but for body weight the deficit still does most of the work. So the 'macro religion' people fight about matters less than the shared move toward whole foods.
What the studies show →
Large randomized comparisons of named diets (for example trials comparing low-carb, low-fat, and Mediterranean-style approaches, and the DIETFITS trial comparing healthy low-fat vs healthy low-carb) tend to find modest, similar average weight loss with wide individual variation — and adherence predicting results more than diet type. The standout for hard outcomes is PREDIMED, which randomized higher-risk adults to a Mediterranean diet (extra olive oil or nuts) vs a low-fat control and found fewer major cardiovascular events; it was re-analyzed after a randomization issue and the main conclusions held. For type 2 diabetes specifically, low-carb approaches have shown meaningful improvements in blood sugar and the ability to reduce medication in trials, though long-term durability and LDL responses vary. The honest summary: several diets can be healthy; the Mediterranean pattern has the broadest, most consistent evidence; extreme exclusion diets have the least.
In big head-to-head trials of named diets, how much does the AVERAGE one-year weight loss typically differ between, say, low-carb and low-fat groups?
Somewhere between 0 and 30 kg difference.
Reveal the answer →
usually only ~0–2 kg apart
Despite fierce online debate, average one-year weight loss between well-designed diet arms is usually within a couple of kilograms of each other — often statistically indistinguishable. The spread within each group (some people lose a lot, some none) dwarfs the difference between diets. Translation: the macro split is a smaller lever than diet quality and adherence.
What's the most evidence-aligned takeaway when comparing the major diets?
- One diet is objectively best for everyone and the rest are harmful
- Several can be healthy; Mediterranean has the broadest evidence, and adherence + quality matter more than the macro split
- Whichever cuts the most carbs always wins
- They're all equally good, so the choice is purely random
Show the answer →
B.Several can be healthy; Mediterranean has the broadest evidence, and adherence + quality matter more than the macro split
The even-handed reading: most well-built diets share the same core (more whole foods, fewer ultra-processed ones) and perform similarly when adherence is matched. Mediterranean stands out for the breadth and consistency of its evidence on real health outcomes, while extreme exclusion diets like carnivore have the least support. Sustainability and food quality beat dogma.
Whatever label you lean toward, the common ground is the same — find whole-food, plant-rich recipes you'll actually want to repeat.