everyday life · Curated Lever · 7 min
Why can vegans and high-protein meat eaters both feel better?
Opposite diet labels can hide shared improvements, different starting points and outcomes measured on different clocks.
The intuitive answer
If two opposing diets produce positive testimonials, nutrition science must be arbitrary or one community must be imagining its improvement.
In 30 seconds
Opposite diets can both help when they replace different problems, improve shared habits and trigger different short-term responses in different people.
Read the full explanation ↓The unseen lever
A diet label bundles composition, food quality, routine, expectations and identity, so different bundles can improve different constraints while personal testimony attributes the whole result to one headline rule.
triggerDifferent baselines meet a new diet ↓determines what the new diet replacesmechanismFoods, routines and expectations change together ↓can improve common behavioursmechanismPlanning and food quality may improve ↓produces different feedback across peopleconstraintBiology and tolerance modify the response ↓turns personal feedback into a general explanationoutcomeOpposing groups both report success
The deeper explanation
The short answer is the start, not the whole story.
Both groups can genuinely feel better because they may start from different baselines and change many things besides animal-food intake. Either diet can increase meal planning, protein or fibre adequacy, reduce particular highly processed foods, alter energy intake and remove foods a person tolerates poorly. Short-term wellbeing is real evidence about that person, but it does not by itself identify the active ingredient or establish long-term health superiority.
The forces underneath
Starting point
Replacing a deficient, excessive or highly processed baseline can help through very different diets.
Shared behaviour
Planning, cooking and monitoring often improve when someone adopts a named diet.
Distinct composition
Protein source, fibre, fats, micronutrients and energy density still differ materially.
Personal response
Physiology, tolerances, medication, activity and goals modify outcomes.
Time horizon
Appetite and energy can change in days while deficiencies or disease risks may take much longer to appear.
Incentives
What each actor is trying to do
Follower
Maintain the pattern that produces noticeable feedback and a sense of control.
Online community
Highlight vivid success stories that reinforce shared rules and identity.
Researcher or clinician
Separate the label into actual foods, adherence, outcomes and time horizons.
Better afternoon energy
The change could reflect meal timing, energy intake, protein, fibre, sleep or expectation rather than the presence or absence of meat alone.
Improved blood lipids
Different lipid components can move in different directions, so one favourable marker should not stand in for the entire risk profile.
Who can gain
- People whose new diet fixes a real constraint and remains nutritionally adequate
- Communities that make difficult behaviour change easier to sustain
Who can bear the cost
- People who generalize another person's response to their own medical context
- Followers whose short-term improvement hides a later nutrient or risk tradeoff
Second-order effects
- Positive feedback can turn a practical experiment into a rigid identity, raising the social cost of changing course.
- Opposing communities may argue about one headline nutrient while overlooking shared improvements in food quality and routine.
Use the lever elsewhere
The mechanism travels.
Vegan transition from fast food
More legumes, vegetables and home cooking may improve fibre and diet structure while animal foods are removed.
High-protein transition from snack-heavy eating
More structured meals and protein may change satiety while refined snacks and alcohol fall.
Digestive symptom experiment
Removing several foods can reduce symptoms without identifying which ingredient mattered or whether the effect will persist.
Common overstatements
The diets are not equivalent simply because followers of both report improvement. Nutrient adequacy, LDL cholesterol, glycaemic control, fibre, food processing and long-term disease risk can move differently and require separate evidence.
Individual variation is real but can be exaggerated into 'anything works for anyone.' Shared human requirements and population-level risk patterns still constrain which diets are plausible and safe.
Expectations and community support can influence subjective experience, but calling every improvement placebo would ignore genuine changes in intake, symptoms, body weight and measured biomarkers.
Where the answer stops
Self-reported energy, digestion and mood are valuable experiences but are vulnerable to expectation, selective memory and concurrent lifestyle changes.
Short-term improvement cannot establish long-term cardiovascular, renal, bone or micronutrient outcomes.
Vegan, high-protein and animal-based labels each contain many different diets; food quality and composition within a label can matter as much as the label itself.
This Lever is explanatory education, not personal medical or dietary advice.
Vegans and high-protein meat eaters can both describe real improvement because they are not running mirror-image laboratory experiments. They start from different baselines, adopt different bundles and may share improvements in planning, diet quality or reduced ultra-processed food. Their biology, goals and time horizon also differ. Personal experience is useful evidence for generating a hypothesis, but comparing long-term health claims requires controlled outcomes, nutrient adequacy and the actual foods eaten, not only the diet label.
Concepts that unlock it
Baseline dependence
The effect of a change depends on what a person was eating and experiencing before it.
Intervention bundle
A diet simultaneously changes foods, nutrients, routines, expectations and social behaviour.
Individual heterogeneity
People can respond differently to the same food because their physiology, context and goals differ.
Short-term proxy
An early measure such as appetite, energy, weight or a blood marker that may not predict every long-term outcome.
Attribution error
Assigning a bundled result to one visible component without isolating the alternatives.
What if?
What if a vegan and a high-protein omnivore both replace a highly processed baseline with planned meals rich in minimally processed foods?
Check your understanding
Can you move the mechanism?
Question 1 of 2
What is the best first explanation when followers of opposing diets both feel better?
Choose the best explanation.
Question 2 of 2
Two people remove different foods but both start cooking, planning meals and eating fewer packaged snacks. What should a careful analysis test?
Choose the best explanation.
Evidence and limits
What supports this answer?
In the 12-month DIETFITS randomized trial, healthy low-fat and healthy low-carbohydrate groups achieved similar average weight loss while both were taught to emphasize diet quality.
Limit: Low-fat and low-carbohydrate are not the same as vegan and animal-based diets, and average weight loss does not capture every health outcome or individual response.JAMA - Effect of Low-Fat vs Low-Carbohydrate Diet on 12-Month Weight Loss in Overweight Adults ↗In a controlled inpatient crossover trial, participants consumed more energy and gained weight during an ultra-processed diet than during a minimally processed diet matched for presented macronutrients, sugar, sodium and fibre.
Limit: The trial involved 20 adults over four weeks and does not prove that processing alone explains every real-world response or testimonial.Cell Metabolism - Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain ↗The PREDICT 1 study observed substantial between-person variation in glucose, insulin and triglyceride responses to identical meals.
Limit: Post-meal biomarker variation does not establish that every person needs a unique diet or that subjective wellbeing will follow the same pattern.Nature Medicine - Human postprandial responses to food and potential for precision nutrition ↗An eight-week randomized trial in 22 pairs of identical twins found better cardiometabolic outcomes on a healthy vegan diet than on a healthy omnivorous diet for several measured endpoints.
Limit: The trial was small and short, compared two specifically designed healthy diets, and did not test every vegan or high-animal-protein pattern.JAMA Network Open - Cardiometabolic Effects of Omnivorous vs Vegan Diets in Identical Twins ↗WHO guidance notes that a healthy diet can vary with individual and cultural context and that protein can come from both plant and animal sources.
Limit: Population guidance does not replace individualized clinical advice and does not imply that all plant or animal foods are nutritionally interchangeable.World Health Organization - Healthy diet ↗