@murphai/murph 1.3.0 → 1.3.2

package/node_modules/@murphai/assistant-engine/skills/nutrition-strategy/references/named-diets/low-carbohydrate.md

@@ -0,0 +1,167 @@
+# Low-carbohydrate eating
+
+Evidence checked: **2026-08-09**
+
+Use this reference only after the parent `nutrition-strategy` skill routes a low-carbohydrate question here. This file adds diet-specific evidence and execution detail; the parent still controls universal safety, routing, data reuse, tracking burden, clinical boundaries, and answer shape.
+
+## What the label means
+
+Low-carbohydrate eating is an umbrella for patterns that reduce carbohydrate relative to a person's usual intake or a comparison diet. It is broader than ketogenic eating: carbohydrate reduction alone does not establish nutritional ketosis. The label does **not** identify one food list, macronutrient split, protein intake, fat quality, calorie level, or clinical purpose.
+
+Published thresholds conflict. Examples include:
+
+- about 26-45% of energy from carbohydrate in one diabetes-education scheme;
+- less than 130 g/day or less than 26% of energy in many trial reviews; and
+- roughly 20-50 g/day or less than 10% of energy for some very-low-carbohydrate or ketogenic protocols.
+
+Treat those as study conventions, not universal cutoffs. When precision matters, state the actual foods, approximate grams or percentage, and whether fiber is included instead of relying on the label. [2][7][12]
+
+Meaningful variants include:
+
+- **Flexible lower-carbohydrate:** fewer sugary drinks, sweets, refined-grain snacks, or oversized starch portions while retaining fruit, legumes, dairy, whole grains, or starchy vegetables as useful.
+- **Plant-forward lower-carbohydrate:** more nonstarchy vegetables, soy foods, legumes in workable portions, nuts, seeds, avocado, and unsaturated oils.
+- **Higher-protein lower-carbohydrate:** more protein replaces some carbohydrate; satiety and adequacy depend on total intake and food choices.
+- **Higher-fat lower-carbohydrate:** more fat replaces carbohydrate; the balance of unsaturated and saturated fats can materially change lipid response.
+- **Very-low-carbohydrate:** a stricter range that may produce ketosis, but the label alone does not prove sustained nutritional ketosis.
+- **Ketogenic:** deliberately formulated to induce and maintain nutritional ketosis. It is a narrower pattern with different implementation and safety issues and should not be inferred from a general low-carbohydrate request.
+
+## Evidence bottom line
+
+### Glycemia
+
+In randomized trials enrolling adults with type 2 diabetes, low- and very-low-carbohydrate diets can produce modest short-term improvements in A1C, triglycerides, weight, and medication use over about six months. A 2021 systematic review rated much of this evidence moderate to low certainty, and many comparative advantages were smaller or absent by 12 months. These findings are condition-specific and do not establish a cure, a universal best diet, or a reason to change medication without clinical supervision. [1][2]
+
+Eating less carbohydrate at a meal can reduce the immediate glucose load. That plausible mechanism and a lower post-meal reading are surrogate observations, not proof of better long-term health for every person.
+
+### Triglycerides and other lipids
+
+Triglycerides often fall and HDL cholesterol may rise, especially over shorter periods. LDL cholesterol, non-HDL cholesterol, and apoB responses are heterogeneous: they may fall, remain similar, or rise, occasionally substantially. A scientific statement found mixed LDL effects, and a 2024 meta-analysis found larger average LDL increases in lower-BMI trial groups than in higher-BMI groups. Neither body size nor food choice predicts an individual's response with certainty. [3][4]
+
+Do not treat lower triglycerides or higher HDL as canceling a meaningful rise in LDL-related markers. What replaces carbohydrate matters: olive or canola oil, nuts, seeds, avocado, fish, tofu, and legumes create a different pattern from one dominated by butter, coconut oil, fatty or processed meats, and large amounts of cheese.
+
+### Weight and satiety
+
+Some people find lower-carbohydrate meals more filling or simpler and may spontaneously eat less. This is not universal. Effects can reflect more protein, less refined or highly palatable food, fewer choices, or—in variants that actually achieve nutritional ketosis—ketosis. Appetite suppression is not proof of adequate intake or a unique fat-loss mechanism. [3]
+
+Initial scale change may include glycogen and water. Across randomized weight-reduction trials, low-carbohydrate and balanced-carbohydrate diets probably produce little to no meaningful difference in weight loss or common cardiovascular risk factors by one to two years. The sustainable energy and food pattern matters more than the label. [5]
+
+### Cardiovascular outcomes
+
+Comparative trials mostly measure weight and risk factors rather than heart attacks, strokes, or cardiovascular death. A category-wide causal cardiovascular benefit from low-carbohydrate eating has not been established.
+
+A 2026 analysis of three long-running cohorts linked higher-quality low-carbohydrate patterns—emphasizing plant foods, whole grains, and unsaturated fats—with lower coronary heart disease incidence, while lower-quality versions emphasizing refined carbohydrates and animal fats and proteins were linked with higher incidence. This is direct health-outcome evidence but remains observational, based on dietary reports and pattern scores; it cannot prove that prescribing a low-carbohydrate diet prevents events. [6]
+
+### Guideline and consensus position
+
+Diabetes guidance treats carbohydrate reduction as one individualized eating-pattern option, especially for short-term glycemic goals, while emphasizing food quality, nutritional adequacy, and medication safety. Lipid and cardiovascular statements do not endorse carbohydrate restriction as a universal goal; they prioritize sustainable pattern quality, unsaturated fats in place of saturated fats, fiber-rich foods, and response monitoring. [1][3][7][8]
+
+### What the evidence does not support
+
+- Carbohydrate is not inherently harmful, and carbohydrate avoidance is not a purity test.
+- Insulin or ketosis explanations alone do not prove superior long-term fat loss or disease prevention.
+- Short-term changes in weight, A1C, triglycerides, hunger, or glucose do not establish long-term cardiovascular benefit.
+- Evidence from adults with type 2 diabetes, overweight, or obesity does not automatically generalize to children, pregnancy, frailty, eating-disorder recovery, high-performance athletes, or people with complex disease.
+- Evidence for very-low-carbohydrate or ketogenic diets should not be presented as evidence for every flexible lower-carbohydrate pattern, or vice versa.
+
+## Why users may choose it
+
+Common jobs include reducing sugary drinks or refined snacks, simplifying meals, seeking steadier perceived energy or appetite, supporting a clinician-managed glycemic goal, lowering triglycerides, losing weight, or matching a food preference. Ask which job matters when it changes the recommendation; do not assume weight loss or diabetes from the label. Popularity is also reinforced by simple rules, visible early scale changes, testimonials, and commercial branding; that shows demand, not comparative efficacy.
+
+A useful core may be much smaller than the branded rule package. For many users, the lowest-risk version is simply fewer low-fiber discretionary carbohydrates, better meal structure, and carbohydrate portions matched to activity—not a fixed gram ceiling.
+
+## Diet-specific tradeoffs and adequacy
+
+Restriction is not automatically harmful, but risk rises as more food groups disappear without replacement.
+
+- **Fiber:** Removing whole grains, legumes, fruit, and starchy vegetables can lower fiber and worsen constipation. Preserve workable sources such as nonstarchy vegetables, berries or other whole fruit, avocado, nuts, seeds, soy foods, legumes, and any whole grains that fit the chosen version.
+- **Micronutrients:** Common pressure points can include folate, thiamin, magnesium, potassium, and vitamin C; calcium and vitamin D may also become harder to cover if dairy or fortified alternatives disappear. Risk depends on the foods removed, not the carbohydrate number alone, and a carefully constructed pattern can still be adequate. [7][11]
+- **Fat quality:** A low-carbohydrate label can hide a large rise in saturated fat. Prefer unsaturated fats and varied protein sources rather than making butter, coconut oil, processed meat, or cheese the default energy replacement. [3][8]
+- **Protein and total energy:** More protein can help meal structure, but protein alone does not correct under-fueling. Higher-fat foods are energy dense, while appetite suppression can also produce unintentionally low intake.
+- **Training fuel:** Easy or low-volume training may be compatible with a lower-carbohydrate pattern, but repeated high-intensity work, long endurance sessions, tournaments, and high weekly volume rely more on carbohydrate availability. Preserve or reintroduce carbohydrate around key sessions rather than treating the label as a ban. [9]
+- **Ordinary life:** Strict rules can increase cost, preparation burden, social friction, or food preoccupation. A pattern that only works in controlled weeks is not yet a durable plan.
+
+Do not recommend supplements automatically. Route supplement evidence, dose, label, and safety questions to `micronutrients-supplements`.
+
+## Lowest-risk practical execution
+
+Use the least restrictive version that serves the user's actual goal.
+
+1. **Name the change, not just the diet.** Start with one high-leverage source: sugary drinks, frequent sweets, low-fiber snack foods, or a starch portion that is routinely larger than wanted. Do not automatically remove whole fruit, legumes, milk or yogurt, whole grains, or starchy vegetables.
+2. **Keep a flexible meal scaffold.** Combine a protein option, nonstarchy vegetables or another fiber-rich food, enough fat or sauce for satisfaction, and a deliberate carbohydrate portion matched to preference and demand.
+3. **Choose replacement quality deliberately.** Options include fish, eggs, tofu or tempeh, plain yogurt or a fortified alternative, poultry, lean meat, beans or lentils in workable portions, olive or canola oil, nuts, seeds, and avocado.
+4. **Keep useful carbohydrate available.** Fruit, oats, beans, lentils, potatoes, rice, whole-grain bread or tortillas, and milk or yogurt can remain flexible options—especially before, during, or after demanding training.
+5. **Use an ordinary-life fallback.** At a restaurant, family meal, or travel stop, choose a satisfying protein and produce when available, take the carbohydrate portion that fits the situation, and resume the usual pattern at the next meal. Do not compensate.
+
+A minimally restrictive example is “replace the daily sweet drink, add vegetables and protein at lunch, and keep rice or potatoes around hard training,” not “ban all starch and fruit.” Concrete foods are options, not requirements.
+
+## Common failure modes and repairs
+
+- **The label is doing all the work.** Repair by naming the actual foods reduced, retained, and added. Two people at the same carbohydrate percentage can have very different diets.
+- **Unplanned drift into ketogenic intake.** Repair by checking whether fruit, legumes, dairy, grains, and training fuel have disappeared. Treat very-low-carbohydrate intent as a separate decision with added safety review.
+- **Carbohydrate is replaced mostly with saturated fat.** Repair by shifting toward olive or canola oil, nuts, seeds, avocado, fish, tofu, legumes, and leaner protein options.
+- **Fiber falls sharply.** Add tolerated fiber-rich foods gradually. Persistent constipation, pain, or other digestive symptoms belong with `gut-digestion`.
+- **Training output or recovery declines.** Add carbohydrate before, during, or after the demanding sessions first; do not explain persistent decline away as required “adaptation.”
+- **Appetite suppression becomes under-fueling.** Restore regular meals and accessible carbohydrate or energy. Falling performance, persistent fatigue, cold intolerance, menstrual changes, recurrent injury or illness, food preoccupation, or unintended weight loss outrank the diet goal.
+- **Early water loss is treated as proof of superior fat loss.** Route intentional body-change strategy and interpretation to `body-composition`.
+- **One higher-carbohydrate meal becomes a lapse.** Return to the next normal meal. Do not fast, overexercise, or tighten rules in response.
+- **The plan requires full-time tracking.** Step down to meal structure or one decision-changing target. Do not require ketone testing, continuous glucose monitoring, daily weighing, or full macro logging for ordinary execution.
+
+## Monitoring, stop conditions, and off-ramps
+
+For a low-risk fit test, define one goal and one or two signals before adding tracking. Useful signals include hunger and satisfaction, daytime energy, training output and recovery, bowel pattern, food preoccupation, social fit, and whether the meals remain practical.
+
+When a clinical marker, disease, or medication is central, hand off rather than inventing a lab schedule or treatment target. If lipids are already being monitored in usual care, a material rise in LDL cholesterol, non-HDL cholesterol, or apoB warrants clinical review rather than assuming the triglyceride response makes it harmless.
+
+Pause the restriction, restore adequate food or carbohydrate as appropriate, and escalate when any of these appear:
+
+- symptomatic or recurrent low glucose, especially in a person using insulin or an insulin-releasing medicine;
+- persistent dizziness, weakness, fatigue, poor concentration, low mood, or impaired training and recovery;
+- substantial unintended weight loss, very low appetite, recurrent illness or injury, menstrual disruption, or increasing cold intolerance;
+- escalating food fear, rigid rules, binge-restrict cycles, compensatory behavior, or social withdrawal;
+- persistent constipation, vomiting, abdominal pain, or other meaningful digestive change; or
+- a marked adverse lipid response or conflict with a clinician-directed eating plan.
+
+Use the parent's urgent-care rules for acute deterioration. Very-low-carbohydrate or ketogenic intake combined with an SGLT2 inhibitor adds ketoacidosis risk; nausea or vomiting, abdominal pain, rapid breathing, unusual drowsiness, or confusion needs urgent assessment even when glucose is not dramatically elevated. Do not advise medication changes. [1][10]
+
+An off-ramp is not failure. Reintroduce one or more tolerated carbohydrate sources at the meal or training window where energy, performance, fiber, or flexibility is most constrained. A moderate-carbohydrate, quality-first pattern may preserve the useful core with less burden.
+
+## Groups needing extra caution
+
+Apply the parent `Clinical Boundaries` section. Diet-specific caution is especially important for:
+
+- anyone using insulin, sulfonylureas, meglitinides, or an SGLT2 inhibitor;
+- type 1 diabetes or any diabetes plan where medication and carbohydrate are tightly linked;
+- known markedly elevated LDL-related markers, a prior large lipid response to carbohydrate restriction, or a clinician-managed lipid disorder;
+- pregnancy or breastfeeding, people under 18, frailty, low intake, or recent substantial weight loss;
+- kidney disease, advanced liver disease, significant heart disease, relevant endocrine disease, or post-bariatric surgery;
+- current or past eating-disorder symptoms, compulsive exercise, or strong number sensitivity; and
+- high-volume or high-intensity training, multiple daily sessions, or competition periods.
+
+General education and flexible food ideas may still help, but condition-specific targets and therapeutic versions belong with a clinician or registered dietitian.
+
+## Murph handoffs
+
+- `food-journal`: exact labels, capture, or retrospective links among food, glucose, symptoms, energy, or performance.
+- `body-composition`: intentional fat loss, muscle gain, weight or waist trends, plateaus, and calorie tradeoffs.
+- `gut-digestion`: constipation, diarrhea, reflux, bloating, elimination, or reintroduction.
+- `cardiometabolic-health`, `chronic-illness-support`, and care navigation: glucose, lipids, blood pressure, disease, medication, therapeutic diets, and lab interpretation.
+- `micronutrients-supplements`: supplement need, product evidence, dose, label, and safety.
+- `experiment-onboarding`: a bounded test after the user chooses one specific safe change.
+- `behavior-followthrough`: recurring support after the plan itself is sound.
+
+## Source register
+
+Checked **2026-08-09**. Sources support the evidence distinctions above; none makes one low-carbohydrate threshold or implementation universally correct.
+
+1. **Standards of Care in Diabetes—2026, Section 5: Facilitating Positive Health Behaviors and Well-being.** Individualized eating patterns, short-term low-carbohydrate evidence, nutritional adequacy, and SGLT2 caution. DOI: [10.2337/dc26-S005](https://doi.org/10.2337/dc26-S005).
+2. **BMJ systematic review and meta-analysis (2021), “Efficacy and safety of low and very low carbohydrate diets for type 2 diabetes remission.”** Twenty-three randomized trials; moderate-to-low-certainty short-term benefits with attenuation by 12 months and population limits. DOI: [10.1136/bmj.m4743](https://doi.org/10.1136/bmj.m4743).
+3. **National Lipid Association scientific statement (2019), “Review of current evidence and clinical recommendations on the effects of low-carbohydrate and very-low-carbohydrate diets.”** Appetite, triglycerides, mixed LDL response, adherence, food quality, and long-term evidence gaps. DOI: [10.1016/j.jacl.2019.08.003](https://doi.org/10.1016/j.jacl.2019.08.003).
+4. **American Journal of Clinical Nutrition meta-analysis (2024), “Increased low-density lipoprotein cholesterol on a low-carbohydrate diet in adults with normal but not high body weight.”** Short-duration randomized trials showing heterogeneous LDL response and effect modification by trial-group BMI; not a cardiovascular-outcomes study. DOI: [10.1016/j.ajcnut.2024.01.009](https://doi.org/10.1016/j.ajcnut.2024.01.009).
+5. **Cochrane Review CD013334 (2022), “Low-carbohydrate versus balanced-carbohydrate diets for reducing weight and cardiovascular risk.”** Little to no comparative difference in weight loss or common risk factors through one to two years in adults with overweight or obesity, with and without type 2 diabetes. DOI: [10.1002/14651858.CD013334.pub2](https://doi.org/10.1002/14651858.CD013334.pub2).
+6. **JACC cohort analysis (2026), “Effect of Low-Carbohydrate and Low-Fat Diets on Metabolomic Indices and Coronary Heart Disease in U.S. Individuals.”** Direct coronary outcomes showing strong food-quality dependence; observational and not proof of causal benefit. DOI: [10.1016/j.jacc.2025.12.038](https://doi.org/10.1016/j.jacc.2025.12.038).
+7. **Scientific Advisory Committee on Nutrition report (2021), “Lower carbohydrate diets for adults with type 2 diabetes.”** No agreed universal definition; short-term, population-specific findings; fiber, micronutrient, saturated-fat, and medication considerations. Stable report page: [GOV.UK](https://www.gov.uk/government/publications/sacn-report-lower-carbohydrate-diets-for-type-2-diabetes).
+8. **American Heart Association scientific statement (2026), “Dietary Guidance to Improve Cardiovascular Health.”** Pattern quality, vegetables and fruit, whole grains, protein sources, unsaturated-fat substitution, and minimally processed foods. DOI: [10.1161/CIR.0000000000001435](https://doi.org/10.1161/CIR.0000000000001435).
+9. **Joint position statement (2016), “Nutrition and Athletic Performance.”** Carbohydrate availability matched to training and competition demand rather than a universal daily intake. DOI: [10.1249/MSS.0000000000000852](https://doi.org/10.1249/MSS.0000000000000852).
+10. **Standards of Care in Diabetes—2026, Section 6: Glycemic Goals, Hypoglycemia, and Hyperglycemic Crises.** Hypoglycemia risk with insulin, sulfonylureas, and meglitinides, and the need to review treatment when lows recur. DOI: [10.2337/dc26-S006](https://doi.org/10.2337/dc26-S006).
+11. **Obesity Reviews systematic review (2019), “Impacts of carbohydrate-restricted diets on micronutrient intakes and status.”** Variable definitions and limited reporting, but recurring risks when carbohydrate-rich food groups are removed without adequate replacement. DOI: [10.1111/obr.12857](https://doi.org/10.1111/obr.12857).
+12. **American Diabetes Association professional nutrition overview, “Nutrition & Wellness” (accessed 2026-08-09).** Examples of one education scheme using 26-45% of energy for low-carbohydrate and 20-50 g non-fiber carbohydrate or less than 26% of energy for very-low-carbohydrate patterns; these are not universal cutoffs. Stable page: [professional.diabetes.org](https://professional.diabetes.org/clinical-support/nutrition-wellness).

package/node_modules/@murphai/assistant-engine/skills/nutrition-strategy/references/named-diets/mediterranean.md

@@ -0,0 +1,141 @@
+# Mediterranean-style eating
+
+Use this reference only for a narrow Mediterranean-style eating question routed by the parent `nutrition-strategy` skill. The parent remains authoritative for routing, tracking burden, clinical boundaries, and answer shape. This is ordinary food-pattern guidance, not a therapeutic-diet protocol.
+
+## Working definition
+
+Mediterranean-style eating is a flexible pattern, not one menu, macro split, national cuisine, or list of imported foods. Common versions emphasize:
+
+- a variety of vegetables and fruit;
+- beans, lentils, peas, whole grains, nuts, and seeds when tolerated and safe;
+- unsaturated fats used mainly in place of saturated fats;
+- fish or seafood in many, but not all, versions;
+- optional poultry, eggs, yogurt, and cheese in amounts that fit the person; and
+- less processed meat, red meat, refined grain, sweets, and sugar-sweetened drinks than in the person's starting pattern.
+
+Extra-virgin olive oil and nuts were supplied in landmark trials, so those trials most directly support the tested packages rather than every possible substitution. For ordinary implementation, another acceptable unsaturated oil, seed, seed butter, or other tolerated local source of unsaturated fat can preserve the broader substitution principle when olive oil or nuts do not fit cost, taste, access, culture, or allergy constraints. Do not imply that the substitute has identical direct trial evidence.
+
+Meaningful variants include ad-libitum patterns rich in unsaturated fat; energy-reduced versions bundled with activity and behavior support; pescatarian, vegetarian, and omnivorous versions; and locally adapted versions. "Green Mediterranean," MIND, branded score-based plans, and other hybrids are distinct interventions; do not automatically transfer all Mediterranean-diet findings to them.
+
+Wine is optional, not a defining nutrient or treatment. Do not advise a non-drinker to start alcohol for this pattern or for heart health. For someone who drinks, answer the food question without presenting alcohol as beneficial and route risk-specific alcohol guidance appropriately.
+
+"Mediterranean" is research shorthand derived from selected food patterns in parts of a large, diverse region. Do not imply that communities around the Mediterranean share one timeless cuisine, grade a user's food as culturally authentic, or require Greek- or Italian-coded dishes. Preserve the useful pattern with the user's own staples, flavors, cooking methods, household meals, and access.
+
+## Evidence calibration
+
+Mediterranean-style eating has a relatively mature cardiometabolic evidence base compared with many named diets, especially for cardiovascular prevention in adults at increased risk. That does not make every version equivalent or establish it as uniquely best.
+
+| Question | Current evidence read |
+| --- | --- |
+| **Guideline and consensus support** | Strong. Current cardiovascular guidance favors the same pattern-level priorities: varied plant foods, whole grains, healthy protein sources, unsaturated fats replacing saturated fats, and fewer highly processed foods, added sugars, and excess sodium. Mediterranean-style and DASH-like patterns are examples, not the only valid routes. |
+| **Direct cardiovascular outcomes** | Randomized trials and a 2023 network meta-analysis support fewer major cardiovascular events, and probably lower mortality for some increased-risk populations, versus minimal intervention. The clearest trials followed older or higher-risk adults for roughly five to seven years and included repeated counseling and, sometimes, supplied foods. |
+| **Comparison with other high-quality patterns** | Less certain. The 2023 network meta-analysis found no convincing mortality or nonfatal-heart-attack advantage over well-implemented low-fat programs. Comparator quality, support intensity, and the exact rule package matter; do not call Mediterranean eating universally superior. |
+| **Weight, glycemia, lipids, and blood pressure** | Average changes are often favorable but usually modest and heterogeneous. The pattern is not inherently calorie restricted and does not guarantee weight loss. A 2025 trial found lower diabetes incidence with an energy-reduced Mediterranean pattern plus physical activity and intensive behavior support, but that bundle does not isolate the diet label. |
+| **Cancer, cognition, mood, longevity, or other broad claims** | Much of the support is observational, surrogate-based, or secondary. Associations can be useful context but do not justify prevention, treatment, reversal, or lifespan promises. |
+
+Important limits:
+
+- The direct-outcome randomized evidence is concentrated in Spanish, older, and increased-risk populations; generalization to younger, lower-risk, and culturally different populations is plausible but not equally direct.
+- The republished PREDIMED analysis addressed departures from individual randomization. Its direction was similar after reanalysis, but the design history and intensive intervention warrant explicit caution.
+- CORDIOPREV was a single-center secondary-prevention trial whose participants were mostly men with established coronary disease. It supports a clinician-managed population, not condition-specific instructions from Murph.
+- Trials tested packages of foods, substitutions, counseling, and social support. They do not prove that olive oil, nuts, fish, wine, polyphenols, or any supplement reproduces the whole pattern's effect in isolation.
+- Plausible mechanisms include higher fiber and micronutrient density, unsaturated-for-saturated-fat substitution, lower intake of some highly processed foods, and easier long-term food variety. Mechanistic plausibility helps explain coherence; it is not comparative-outcome proof.
+
+## Why users may choose it
+
+Common reasons include wanting a flexible heart-health-aligned pattern, more plant foods without excluding all animal foods, fewer calculations, meals that can work for a household, or an evidence-supported alternative to a highly restrictive plan. Clarify the user's actual job before expanding the pattern: definition, comparison, one meal change, grocery defaults, cultural adaptation, or troubleshooting.
+
+## Tradeoffs and adequacy
+
+A Mediterranean label does not guarantee diet quality or adequacy.
+
+- **Under-fueling:** A "salad and fish" interpretation can provide too little energy, carbohydrate, or protein, especially with high training demand, low appetite, frailty, or recovery needs. Keep useful grains, potatoes, bread, rice, corn, legumes, fruit, and other tolerated carbohydrate sources.
+- **Fiber tolerance:** A rapid increase in legumes, whole grains, vegetables, nuts, or seeds can cause discomfort. Increase gradually and use `gut-digestion` when persistent symptoms, elimination, or reintroduction is the actual problem.
+- **Energy-dense additions:** Oils, nuts, seeds, cheese, and spreads can be useful foods, but simply adding them to an unchanged pattern may not serve an intentional body-change goal. Do not frame them as free foods; hand that goal to `body-composition`.
+- **Food-pattern quality:** Pastries, refined grains, processed meats, salty convenience foods, and alcohol do not become protective because they are marketed as Mediterranean. Judge the whole pattern and the substitution, not the label.
+- **Allergens and exclusions:** Nuts, sesame, fish, shellfish, dairy, and gluten-containing grains can be central in some versions but are never mandatory. Use exact-label and cross-contact checks when material.
+- **Vegetarian or vegan adequacy:** Legumes, soy foods, peas, nuts or seeds when safe, and whole grains can anchor protein. Depending on exclusions and total intake, iron, vitamin B12, calcium, iodine, vitamin D, and omega-3 adequacy may need focused review; do not invent supplement doses here.
+- **Cost and access:** Fresh fish, imported oil, specialty cheese, or branded products can make an otherwise flexible pattern expensive. Frozen or canned produce, dry or canned legumes, oats, rice, corn, potatoes, local whole-grain foods, eggs, tofu, peanut or seed butter, canned fish when wanted, and an affordable unsaturated oil can carry the pattern.
+
+## Lowest-risk practical execution
+
+Start with one recurring meal or shopping default rather than an adherence score.
+
+1. **Add a plant anchor.** Use a vegetable or fruit plus beans, lentils, peas, or a whole-grain or other high-fiber staple that already fits the household.
+2. **Make one replacement, not a purity rule.** At some meals, replace processed meat or a saturated-fat-heavy default with beans, soy foods, fish, poultry, eggs, yogurt, nuts, seeds, or an unsaturated oil that fits the user.
+3. **Protect satisfaction and adequacy.** Keep enough familiar starch, protein, fat, sauce, seasoning, and total food for appetite, training, recovery, and ordinary life.
+
+Flexible meal structures include:
+
+- beans, lentils, tofu, fish, chicken, eggs, or yogurt with rice, barley, potatoes, bread, or corn tortillas plus vegetables and a flavorful unsaturated-fat-based sauce;
+- oats or whole-grain toast with fruit and yogurt, soy yogurt, nuts, seeds, or seed butter;
+- a familiar soup, stew, curry, grain bowl, pasta dish, sandwich, or family meal shifted toward more legumes or vegetables and less processed meat; or
+- a fallback such as canned beans with frozen vegetables and a microwave grain, lentil soup with bread and fruit, hummus or a seed spread sandwich, or canned fish with potatoes or crackers when those foods fit.
+
+Cultural adaptation is expected, not a compromise. Keep local legumes, grains, vegetables, fruits, spices, sauces, and shared dishes. Organic produce, fresh seafood, imported olive oil, wine, and specialty products are not entry requirements.
+
+For common constraints:
+
+- **Vegetarian or vegan:** Use legumes, tofu, tempeh, textured vegetable protein, peas, nuts or seeds when safe, and fortified alternatives as needed; fish is not required.
+- **Nut allergy:** Use seeds or seed butter when safe, legumes, soy foods, eggs, dairy, fish, poultry, or other tolerated proteins and fats.
+- **Fish or shellfish allergy, dislike, or avoidance:** Use legumes, soy foods, eggs, dairy, poultry, nuts, or seeds as appropriate; do not treat fish as mandatory.
+- **Dairy-free:** Use tolerated fortified alternatives and other calcium-containing foods; exact needs belong with the focused adequacy owner.
+- **Gluten-free:** Use rice, corn, potatoes, quinoa, buckwheat, and certified gluten-free oats as appropriate, with exact-label and cross-contact checks when needed.
+- **Olive oil does not fit:** Use another affordable unsaturated oil or tolerated whole-food fat while being clear that this is a practical adaptation, not the exact landmark-trial intervention.
+
+## Common failure modes
+
+- Reducing the pattern to olive oil, wine, cheese, or one "superfood."
+- Adding oil and nuts without replacing anything when the user's actual goal depends on overall energy intake.
+- Turning a flexible pattern into unnecessary bans on bread, pasta, potatoes, dairy, or all red meat.
+- Buying costly imported foods or preparing separate "Mediterranean" meals instead of adapting familiar dishes.
+- Using an authenticity score or moral language that makes ordinary variation feel like failure.
+- Increasing fiber so quickly that the plan becomes uncomfortable or unsustainable.
+- Building meals around produce alone and missing energy, protein, or carbohydrate needs.
+- Treating observational associations, supplements, or a biomarker change as proof of disease prevention or treatment.
+
+## Monitoring, stop conditions, and off-ramps
+
+Use the lowest-burden signals that can change the next decision: meal satisfaction, hunger, energy, training and recovery, digestive tolerance, time, cost, access, and social fit. Clinician-ordered markers may inform a clinical plan, but this reference does not set condition-specific targets or medication changes.
+
+Adjust, pause, or hand off when the pattern is followed by persistent inadequate intake, unintended weight loss, fatigue, dizziness, declining performance or recovery, ongoing digestive symptoms, a suspected allergic reaction, increasing food fear or rigidity, compensatory behavior, or a concerning change in a clinician-monitored marker or medication response. Apply the parent's urgent and clinical-boundary rules rather than creating a Mediterranean-specific emergency protocol.
+
+A useful off-ramp is to drop the label while keeping one or two workable elements: a bean-based meal, more produce, an unsaturated-fat substitution, or a reliable emergency meal. Slow the fiber increase, restore tolerated foods, or move to another evidence-aligned pattern when cost, symptoms, culture, access, or preference makes this one a poor fit. Do not compensate after an off-pattern meal; return to the next ordinary meal.
+
+## Extra caution
+
+The parent's clinical boundaries control. Mediterranean-specific implementation needs additional care when:
+
+- a condition, medication, therapeutic diet, pregnancy, growth, frailty, post-bariatric status, or low intake makes changes in carbohydrate, protein, potassium-rich foods, sodium, alcohol, or total intake clinically material;
+- severe allergy or celiac disease makes ingredient identity or cross-contact decisive;
+- eating-disorder risk, recent major restriction, or unintentional weight loss makes a named-diet score or further restriction unsafe; or
+- the user plans to replace prescribed care, alter medication, or use the pattern to treat a disease.
+
+Keep general meal ideas low risk while routing the condition-specific decision to the appropriate clinician or registered dietitian.
+
+## Murph handoffs
+
+- Keep ordinary definition, comparison, cultural adaptation, grocery defaults, meal structure, and adequacy-preserving execution in `nutrition-strategy`.
+- Use `food-journal` for capture, retrospective pattern finding, or exact ingredient, allergen, and label provenance.
+- Use `body-composition` when intentional weight, waist, fat loss, muscle gain, or calorie strategy is the primary job; return here only for meal execution.
+- Use `gut-digestion` for persistent symptoms or elimination and reintroduction work.
+- Use `cardiometabolic-health`, `chronic-illness-support`, care navigation, and the user's clinician for marker-first, disease-specific, medication-linked, or therapeutic-diet decisions.
+- Use `micronutrients-supplements` when supplement evidence, dose, safety, or focused adequacy is the actual question.
+- Use `experiment-onboarding` only after the user chooses one specific safe change, and `behavior-followthrough` when reminders or repeated support becomes the main job.
+
+Do not expose these internal route labels in the user-facing answer.
+
+## Source register
+
+Evidence reviewed through **2026-08-09**. Prefer the pattern-level conclusions and stated population limits over isolated component claims.
+
+| Source | Evidence type and use |
+| --- | --- |
+| American Heart Association, ["2026 Dietary Guidance to Improve Cardiovascular Health"](https://doi.org/10.1161/CIR.0000000000001435), *Circulation* (2026) | Current scientific statement supporting food-pattern focus, plant variety, whole grains, healthy protein, unsaturated-for-saturated-fat substitution, and not initiating alcohol. |
+| American Heart Association, ["Popular Dietary Patterns: Alignment With American Heart Association 2021 Dietary Guidance"](https://doi.org/10.1161/CIR.0000000000001146), *Circulation* (2023) | Consensus comparison placing Mediterranean-style patterns among the closest alignments while preserving multiple valid heart-healthy patterns. |
+| ["Comparison of seven popular structured dietary programmes and risk of mortality and major cardiovascular events in patients at increased cardiovascular risk"](https://doi.org/10.1136/bmj-2022-072003), *BMJ* (2023) | Systematic review and network meta-analysis supporting direct cardiovascular outcomes versus minimal intervention in increased-risk adults; head-to-head superiority over low-fat programs remained unconvincing. |
+| ["Primary Prevention of Cardiovascular Disease with a Mediterranean Diet Supplemented with Extra-Virgin Olive Oil or Nuts"](https://doi.org/10.1056/NEJMoa1800389), *New England Journal of Medicine* (2018) | Republished PREDIMED randomized analysis; direct outcomes over a median 4.8 years in older Spanish adults at high cardiovascular risk, with randomization departures and intensive support limiting transfer. |
+| ["Long-term secondary prevention of cardiovascular disease with a Mediterranean diet and a low-fat diet (CORDIOPREV)"](https://doi.org/10.1016/S0140-6736(22)00122-2), *The Lancet* (2022) | Randomized direct-outcome comparison over about seven years in people with established coronary disease; single-center, mostly male, clinician-managed population. |
+| ["Mediterranean-style diet for the primary and secondary prevention of cardiovascular disease"](https://doi.org/10.1002/14651858.CD009825.pub3), *Cochrane Database of Systematic Reviews* (2019) | Systematic review documenting heterogeneity and earlier uncertainty in intervention definitions and clinical endpoints; predates CORDIOPREV and the 2023 network meta-analysis. |
+| ["Comparison of an Energy-Reduced Mediterranean Diet and Physical Activity Versus an Ad Libitum Mediterranean Diet in the Prevention of Type 2 Diabetes: A Secondary Analysis of a Randomized Controlled Trial"](https://doi.org/10.7326/ANNALS-25-00388), *Annals of Internal Medicine* (2025) | Long-term randomized evidence for a bundled energy-reduction, activity, and behavior-support intervention; does not isolate a Mediterranean label or justify condition-specific treatment advice. |
+| ["Culturally adapting the Mediterranean Diet pattern - a way of promoting more sustainable dietary change?"](https://doi.org/10.1017/S0007114522001945), *British Journal of Nutrition* (2022) | Review supporting adaptation around local foods, preferences, access, and barriers while preserving core pattern characteristics; not a direct-outcome trial. |

package/node_modules/@murphai/assistant-engine/skills/nutrition-strategy/references/named-diets/vegan-plant-based.md

@@ -0,0 +1,175 @@
+# Vegan and plant-based patterns
+
+Use this reference only after the parent `nutrition-strategy` skill routes a named-diet question here. It is progressive-disclosure guidance, not a registered skill, and it does not replace the parent's safety, tracking-burden, clinical-boundary, or handoff rules.
+
+## Define the pattern before evaluating it
+
+`Plant-based` has no single operational definition. For direct definitions, answer first with a labeled low-stakes assumption when needed. When evaluation, adaptation, off-ramp, or safety depends on it, ask one focused question about what the person excludes or which rules matter.
+
+- **Vegan dietary pattern:** excludes foods of animal origin. Nutritional quality can range from varied and well fortified to narrow and inadequate; the label alone says little about food quality.
+- **Ethical veganism:** a broader commitment concerning animal use, with diet as one expression. Treat this as a legitimate boundary rather than a health claim or a rule to negotiate away. [S11]
+- **Whole-food plant-based (WFPB):** emphasizes legumes, whole grains, vegetables, fruit, nuts, and seeds. It is often vegan, but the term is not standardized. Extra rules such as oil-free, raw, or near-total avoidance of refined or convenience foods are optional variants, not proven requirements.
+- **Mostly plant-based or plant-forward:** most food comes from plants, while some animal-source foods may remain. This can preserve many useful substitutions with less restriction when the motivation is health rather than ethics.
+
+Do not rank these patterns morally or create a moral hierarchy. When the user's evaluation, adaptation, off-ramp, or safety depends on motivation, ask one focused question about whether ethics, environment, religion or culture, health, taste, dairy intolerance, budget, access, family fit, or some combination is central. Motivation changes which adaptations and off-ramps are acceptable.
+
+## Evidence bottom line
+
+Popularity, ethical coherence, identity, and testimonials can explain demand, but they are not comparative-efficacy evidence.
+
+### What is supported
+
+- In adults, an appropriately planned vegetarian or vegan pattern can be nutritionally adequate. "Appropriately planned" is doing real work: adequacy is not automatic, and a reliable vitamin B12 system is essential for a vegan pattern. [S1][S6][S9]
+- Randomized trials in adults with or at high cardiometabolic risk suggest modest average short-term improvements in LDL cholesterol, HbA1c, and body weight versus control diets, with no clear systolic-blood-pressure effect. The trials were heterogeneous and averaged roughly six months, so they mainly support selected surrogate outcomes, not long-term clinical events. [S3]
+- A newer review limited to people with overweight or obesity found little or no difference for most measured outcomes, a possible small LDL reduction, and very-low-certainty evidence for BMI and HbA1c. This tempers claims of broad superiority. [S4]
+- Prospective cohorts associate healthier plant-rich patterns with lower risks of type 2 diabetes, cardiovascular disease, cancer, and mortality, while patterns high in refined grains, sweets, and sugary drinks do not show the same associations and can trend the other way. These data support food quality, not causality or a unique benefit from strict vegan exclusion. [S5]
+- Available sports studies do not show an inherent performance disadvantage when energy and protein are sufficient, but the evidence is small, short, and does not establish a performance advantage. [S8]
+- Current cardiovascular guidance favors a pattern rich in vegetables, fruit, whole grains, legumes, nuts, and plant proteins, but does not require strict vegan exclusion. [S12]
+
+Plausible mechanisms include replacing some saturated fat with unsaturated fat, increasing fiber and micronutrient-rich foods, changing energy density, and displacing foods associated with poorer outcomes. These mechanisms are credible but do not prove that stricter exclusion is better.
+
+### What is not established
+
+- Strong randomized evidence that a vegan label itself prevents major cardiovascular events, cancer, or premature death.
+- That all vegan foods, meat substitutes, plant drinks, or WFPB implementations are nutritionally equivalent.
+- That oil-free, raw, detox, very-low-fat, or highly restrictive variants are necessary for benefit.
+- That vegan eating cures disease, guarantees weight change, or improves athletic performance.
+- That adult findings can be carried over unchanged to pregnancy, infancy, childhood, frailty, or complex disease.
+
+A recent cohort meta-analysis found vegetarian and vegan patterns associated with higher hip-fracture risk, but certainty was low or very low and residual confounding remains possible. Treat this as a reason to protect energy, protein, calcium, vitamin D, and overall bone-health care—not as proof that vegan eating causes fractures. [S7]
+
+## Lowest-risk practical execution
+
+Start with additions and replacements, not a long ban list. Build two repeatable meals and one emergency option before optimizing details.
+
+### 1. Make vitamin B12 explicit
+
+For a vegan pattern, **a reliable B12 system is non-optional**:
+
+- use foods with verified B12 fortification and/or a B12 supplement;
+- check the actual label and serving pattern because fortification varies by product and country;
+- do not count unfortified nutritional yeast, seaweed, fermented foods, or algae products as reliable B12 sources;
+- hand off product choice, dose, interactions, deficiency treatment, and laboratory interpretation to `micronutrients-supplements` or clinical care.
+
+Do not wait for symptoms. Neurologic effects can occur without obvious anemia, and stored B12 can delay symptoms for years. [S9]
+
+### 2. Anchor protein and enough total food
+
+Use a meaningful protein source at roughly three or four eating occasions before reaching for detailed tracking. Flexible anchors include:
+
+- tofu, tempeh, edamame, soy milk, textured vegetable protein, or soy yogurt;
+- lentils, beans, chickpeas, split peas, or pea-based foods;
+- seitan when gluten is suitable;
+- nuts, seeds, nut or seed butter, and higher-protein grains as contributors rather than the only anchor.
+
+A varied plant pattern can supply all indispensable amino acids; foods do not need to be "combined" at each meal. The practical risks are too little total protein, uneven distribution, and low total energy. Athletes, older adults, and people with high needs may need the parent skill's protein target and distribution guidance, but do not prescribe adult targets to children, pregnancy, or clinician-managed conditions.
+
+### 3. Plan the nutrients that most often need attention
+
+| Nutrient | Food-first execution | Boundary |
+| --- | --- | --- |
+| **Iron** | Use lentils, beans, tofu, tempeh, seeds, and iron-fortified grains; pair a source with vitamin-C-rich produce when practical. | Plant iron is nonheme and less predictably absorbed. Do not start iron supplements from the diet label alone; symptoms, heavy menstrual bleeding, pregnancy, abnormal labs, or known anemia need clinical assessment. [S9] |
+| **Iodine** | Use an intentional reliable source, commonly iodized salt or an iodine-fortified product where appropriate. | Many plant drinks contain little iodine; seaweed content varies from low to excessive. Thyroid disease, sodium restriction, pregnancy, and supplement dosing need individualized handoff. [S9] |
+| **Calcium** | Rotate calcium-fortified plant drink or yogurt, calcium-set tofu, and lower-oxalate greens such as kale, bok choy, or broccoli; beans and tahini can contribute. | Fortification and tofu coagulants vary by label. Spinach is not an efficient main calcium source. Bone disease, low intake, or supplement questions need handoff. [S7][S9] |
+| **Vitamin D** | Use verified fortified foods when available and account for season, sun exposure, skin coverage, and geography. | Food alone may not close the gap. Route supplement dose and testing rather than guessing. [S9] |
+| **Omega-3 fats** | Include ground flax or chia, walnuts, and canola or soybean oil for ALA. | Conversion of ALA to EPA and DHA is limited. Algae-derived DHA/EPA is a vegan option, but product and dose decisions belong with `micronutrients-supplements`, especially in pregnancy or pediatrics. [S9] |
+| **Zinc** | Use legumes, soy foods, whole grains, nuts, and seeds. Soaking, sprouting, or fermentation can be optional preparation tools. | Do not add high-dose zinc routinely; excess can create other problems. [S6][S9] |
+
+Fortified and convenience foods are tools, not evidence of failure. A fortified soy or pea drink can be more useful than an unfortified low-protein plant drink; a burger analogue can ease a transition without acquiring a health halo. Use `food-journal`'s exact-label workflow when protein, B12, calcium, vitamin D, iodine, allergens, or ingredients would change the answer.
+
+### 4. Protect energy density and tolerance
+
+A high-volume WFPB pattern can unintentionally underfuel children, athletes, older adults, people with low appetite, and anyone adding large amounts of fiber quickly. Useful additions include nut or seed butter, avocado, oils if acceptable, hummus, tofu, dried fruit, granola, smoothies, bread, pasta, rice, and other convenient starches. Raw or oil-free rules are not the default.
+
+Increase beans, bran, and other high-fiber foods gradually if intake was previously low. For persistent bloating, pain, reflux, constipation, or diarrhea, route to `gut-digestion` rather than layering on more exclusions.
+
+### 5. Protect budget and cultural fit
+
+Specialty substitutes are optional. Low-cost foundations can include dried or canned beans and lentils, tofu or textured vegetable protein where available, peanut butter, oats, rice, pasta, potatoes, frozen vegetables, canned tomatoes, and seasonal fruit.
+
+Adapt existing food traditions rather than imposing a separate cuisine: for example, dal or chana with rice or roti; beans with tortillas and salsa; tofu, tempeh, or edamame with noodles and vegetables; peanut or bean stews; pasta with lentil or soy mince. Preserve family, religious, cultural, and social meaning. No cuisine or level of restriction is morally superior.
+
+### Flexible assemblies
+
+These are structures, not a menu:
+
+- oats or fortified cereal + fortified soy or pea drink + nut butter or ground flax + fruit;
+- dal, chana, tofu, tempeh, bean stew, or lentil sauce + rice, roti, tortilla, noodles, pasta, or potatoes + vegetables and a satisfying sauce;
+- emergency options such as canned beans with a microwave grain, a peanut-butter sandwich with fortified plant drink, fortified cereal with soy milk, hummus with pita, or frozen edamame with a convenient starch.
+
+## Common failure modes
+
+- Removing meat, dairy, and eggs without replacing protein, energy, B12, calcium, iodine, or familiar meals.
+- Assuming `vegan` means cardiometabolically favorable regardless of the overall pattern.
+- Treating unfortified plant drinks as nutritional equivalents to dairy milk or infant formula.
+- Relying on seaweed for predictable iodine or on unverified foods for B12.
+- Escalating from vegan to oil-free, raw, gluten-free, sugar-free, or other purity rules without a separate need.
+- Raising fiber faster than appetite and gastrointestinal tolerance can support.
+- Using protein powder or a broad supplement stack as a substitute for a workable food system.
+- Making the pattern dependent on expensive specialty products, elaborate cooking, or social isolation.
+- Framing lapses, convenience foods, or animal-food choices as virtue, betrayal, cleanliness, or failure.
+
+## Higher-caution groups
+
+The parent clinical boundary applies. Raise the threshold for self-directed restriction and arrange qualified dietetic or clinical support for:
+
+- infants, children, and adolescents;
+- pregnancy or breastfeeding;
+- older adults with frailty, sarcopenia, low appetite, or low intake;
+- athletes or others with very high energy demands;
+- eating-disorder history, escalating food rules, or strong number sensitivity;
+- multiple allergies, gastrointestinal exclusions, or very limited food access;
+- bariatric surgery, kidney disease, thyroid disease, anemia, bone disease, malabsorption, or medication-related appetite problems.
+
+For infants and children, current evidence is too limited to assume that a strict vegan pattern supports normal growth without active planning. Regular growth, intake, and nutritional-status monitoring plus specific micronutrient support—including B12—are recommended. Plant beverages are not substitutes for breast milk or appropriate infant formula. [S2]
+
+Pregnancy and breastfeeding require explicit prenatal and clinical planning for B12, iron, iodine, vitamin D, calcium, protein, zinc, and omega-3 fats. Outcome evidence is heterogeneous and does not justify casual reassurance or blanket prohibition. [S10]
+
+## Monitoring, stop conditions, and off-ramps
+
+Use the lowest useful burden. After roughly two to four weeks, review:
+
+- whether the B12 system is active and label-verified;
+- whether most days contain several protein anchors and enough total food;
+- energy, hunger, bowel tolerance, training output and recovery, menstrual function, and unintended weight change;
+- cost, cooking burden, family/social fit, and whether rules are becoming more rigid.
+
+Do not order or imply a universal "vegan lab panel." Clinical history, symptoms, life stage, medication, prior status, and actual intake should determine testing.
+
+Stop escalating restriction and seek assessment for persistent fatigue or weakness, dizziness, numbness or tingling, balance changes, unexplained weight loss, menstrual changes, falling performance, recurrent injury, poor growth, suspected deficiency, or inability to meet intake. Use the parent's urgent-care rules for acute deterioration.
+
+Choose the off-ramp that respects the motivation:
+
+- **Ethical vegan:** keep the vegan boundary while loosening optional WFPB, raw, oil-free, or minimally processed rules; add fortified foods, convenient protein, energy-dense foods, and a supplement handoff.
+- **Health-only:** move to a mostly plant-based pattern when strict exclusion adds burden without a clear benefit and the user's values allow it.
+- **Any motivation:** return to the last adequate, tolerable version while arranging support rather than compensating or adding more bans.
+
+## Murph handoffs
+
+- `food-journal`: exact product, fortification, allergen, ingredient, or retrospective pattern questions.
+- `body-composition`: intentional fat loss, muscle gain, recomposition, calorie strategy, or weight/waist interpretation.
+- `cardiometabolic-health`: marker-first decisions or interpretation of LDL, glucose, blood pressure, or other risk data.
+- `micronutrients-supplements`: supplement necessity, product, form, dose, interactions, labels, and laboratory follow-up.
+- `chronic-illness-support` and care navigation: disease-, medication-, pregnancy-, pediatric-, bariatric-, or therapeutic-diet decisions.
+- `gut-digestion`: persistent symptoms, elimination, or reintroduction.
+- `experiment-onboarding`: one chosen, safe, bounded change after adequacy is protected.
+- `behavior-followthrough`: reminders, recurring support, or adherence repair after the plan is accepted.
+
+## Dated source register
+
+Evidence checked: **2026-08-09**. Prefer newer applicable clinical guidance when it conflicts with this snapshot.
+
+| ID | Source | Relevance and limits |
+| --- | --- | --- |
+| S1 | *Vegetarian Dietary Patterns for Adults: A Position Paper of the Academy of Nutrition and Dietetics* (2025). [DOI](https://doi.org/10.1016/j.jand.2025.02.002); [PMID 39923894](https://pubmed.ncbi.nlm.nih.gov/39923894/); [official summary](https://www.eatrightpro.org/news-center/research-briefs/new-position-paper-on-vegetarian-and-vegan-diets); [corrigendum](https://doi.org/10.1016/j.jand.2025.156226). | Adult consensus: appropriately planned vegetarian and vegan patterns can be adequate and may offer cardiometabolic benefits. Do not extrapolate the adult position to children or pregnancy. |
+| S2 | *Vegan diet and nutritional status in infants, children and adolescents: A position paper based on a systematic search by the ESPGHAN Nutrition Committee* (2025). [DOI](https://doi.org/10.1002/jpn3.70182); [PMID 40819279](https://pubmed.ncbi.nlm.nih.gov/40819279/); [official page](https://www.espghan.org/knowledge-center/publications/Nutrition/2025-Vegan-Diet). | Ten included reports and about 1,500 vegan children; adequacy and growth evidence remains inconclusive. Recommends qualified counselling, monitoring, and specific micronutrient support including B12. |
+| S3 | *Vegetarian Dietary Patterns and Cardiometabolic Risk in People With or at High Risk of Cardiovascular Disease: A Systematic Review and Meta-analysis* (2023). [DOI](https://doi.org/10.1001/jamanetworkopen.2023.25658); [PMID 37490288](https://pubmed.ncbi.nlm.nih.gov/37490288/). | Twenty randomized trials, mostly short and in higher-risk adults; modest surrogate changes with heterogeneous diets and comparators. |
+| S4 | *Health Effects of Plant-Based Diets in People with Overweight or Obesity: A Systematic Review and Meta-Analysis* (2026). [DOI](https://doi.org/10.3390/nu18121987); [PMID 42356372](https://pubmed.ncbi.nlm.nih.gov/42356372/). | Ten randomized trials; most outcomes showed little or no difference, with limited or very-low-certainty evidence for selected benefits. |
+| S5 | *Associations between plant-based dietary patterns and risks of type 2 diabetes, cardiovascular disease, cancer, and mortality—a systematic review and meta-analysis* (2023). [DOI](https://doi.org/10.1186/s12937-023-00877-2); [PMID 37789346](https://pubmed.ncbi.nlm.nih.gov/37789346/). | Large prospective-cohort synthesis; stronger associations for healthier plant-food patterns, but observational confounding and heterogeneity prevent causal or strict-vegan conclusions. |
+| S6 | *Nutrient Intake and Status in Adults Consuming Plant-Based Diets Compared to Meat-Eaters: A Systematic Review* (2021). [DOI](https://doi.org/10.3390/nu14010029); [PMID 35010904](https://pubmed.ncbi.nlm.nih.gov/35010904/). | Broad adult nutrient-status comparison; identifies both strengths and recurring B12, calcium, iodine, iron, zinc, vitamin D, and omega-3 planning concerns. |
+| S7 | *Vegetarian and Vegan Diets and the Risk of Hip Fracture in Adults: A Systematic Review and Meta-analysis* (2025). [DOI](https://doi.org/10.1093/nutrit/nuaf241); [PMID 41275510](https://pubmed.ncbi.nlm.nih.gov/41275510/). | Cohort association only; certainty was low for vegetarian and very low for vegan estimates. Useful as a bone-adequacy signal, not proof of causation. |
+| S8 | *The Impact Plant-Based Diets Have on Athletic Performance and Body Composition: A Systematic Review* (2024). [DOI](https://doi.org/10.1080/27697061.2024.2365755); [PMID 38913935](https://pubmed.ncbi.nlm.nih.gov/38913935/). | No clear performance detriment in available healthy-adult studies; evidence remains limited and does not establish superiority. |
+| S9 | National Institutes of Health Office of Dietary Supplements, health-professional fact sheets: [vitamin B12](https://ods.od.nih.gov/factsheets/VitaminB12-HealthProfessional/), [iron](https://ods.od.nih.gov/factsheets/Iron-HealthProfessional/), [iodine](https://ods.od.nih.gov/factsheets/Iodine-HealthProfessional/), [calcium](https://ods.od.nih.gov/factsheets/Calcium-HealthProfessional/), [vitamin D](https://ods.od.nih.gov/factsheets/VitaminD-HealthProfessional/), [omega-3 fats](https://ods.od.nih.gov/factsheets/Omega3FattyAcids-HealthProfessional/), and [zinc](https://ods.od.nih.gov/factsheets/Zinc-HealthProfessional/) (accessed 2026-08-09). | Primary official nutrient references. Fortification, product composition, individual requirements, and supplement safety vary. |
+| S10 | National Institutes of Health Office of Dietary Supplements, *Dietary Supplements and Life Stages: Pregnancy—Health Professional Fact Sheet* (accessed 2026-08-09), [official page](https://ods.od.nih.gov/factsheets/Pregnancy-HealthProfessional/); *Strict vegetarian diet and pregnancy outcomes: A systematic review and meta-analysis* (2024), [PMID 39758872](https://pubmed.ncbi.nlm.nih.gov/39758872/). | Pregnancy is a high-consequence planning context; evidence is heterogeneous and nutrient or supplement decisions require prenatal care. |
+| S11 | The Vegan Society, *Definition of veganism* (accessed 2026-08-09), [official page](https://www.vegansociety.com/go-vegan/definition-veganism). | Supports the distinction between a vegan diet and ethical veganism as a broader commitment; not a source for comparative health efficacy. |
+| S12 | *2026 Dietary Guidance to Improve Cardiovascular Health: A Scientific Statement From the American Heart Association* (2026). [DOI](https://doi.org/10.1161/CIR.0000000000001435); [PMID 41914202](https://pubmed.ncbi.nlm.nih.gov/41914202/). | Current cardiovascular consensus supports plant-rich food quality and plant proteins; it does not establish that complete animal-food exclusion is necessary. |