{ "$schema_version": "5.4", "review_mode": "fresh", "claim_restatement": { "index_wording": "Dietary carbohydrate drives insulin secretion that promotes fat storage and increases hunger, and this mechanism is the primary cause of obesity rather than excess energy intake as such", "entry_form": "causal chain with primacy claim", "gradeable_restatement": "In common human obesity, a diet high in glycemic load is the primary modifiable upstream cause: its hormonal response, centred on a raised insulin-to-glucagon ratio, shifts fuel partitioning toward adipose storage and lowers fuel availability to other tissues, which increases hunger and lowers energy expenditure, so that fat storage drives positive energy balance rather than following from it; after an adaptation period of weeks, a lower-glycemic-load diet should therefore raise energy expenditure at equal energy and protein intake and reduce ad libitum intake.", "restatement_required": true, "split_into": [ "SC1: carbohydrate/glycemic-load-to-insulin (and insulin-to-glucagon) endocrine response", "SC2: partitioning sufficient to drive net fat accumulation (storage preceding positive balance)", "SC3: insulin-mediated late-postprandial fuel reduction, hunger, and intake", "SC4: post-adaptation isocaloric energy-expenditure or fat-loss prediction", "SC5: insulin-secretion phenotype as effect modifier", "SC6: primacy for common human obesity" ], "parent_claim": null, "scope_qualifiers": [ { "qualifier": "common human obesity", "determinable_before_outcome": true, "determined_from": "species, phenotype, and population eligibility fixed before results; a responder category counts only if defined before randomization and independently reproduced", "action": "retained" }, { "qualifier": "at equal energy and protein intake", "determinable_before_outcome": true, "determined_from": "controlled-feeding design and nutrient prescription", "action": "retained as the decisive isocaloric prediction" }, { "qualifier": "after an adaptation period of weeks", "determinable_before_outcome": true, "determined_from": "study duration, fixed by protocol before outcomes; stated by proponents (PMID 34515299; PMID 33274750) before the trials adjudicated here were scored", "action": "added in the red-team synthesis so that duration is a prespecified boundary rather than a post hoc explanation" } ], "restatement_note": "Revised in the red-team synthesis to match the proponents' current formulation (glycemic load; insulin-to-glucagon ratio; fat storage driving positive energy balance; adaptation over weeks) per PMID 34515299 and PMID 37661740. The earlier SC2 wording 'independently of total energy balance' described a position no proponent defends and was removed." }, "lay_summary": { "paragraphs": [ "The claim is that eating a lot of quickly digested carbohydrate raises insulin, insulin pushes fuel into fat cells, the rest of the body is left short of fuel, hunger rises and metabolism slows, and that this chain, rather than simply eating too much, is the main cause of common obesity.", "This review assigns Γ, meaning mechanistically plausible, partial system support. The first step is firmly established: carbohydrate raises insulin. Experiments in mice and human genetic studies suggest that high insulin can add to fat gain in some settings. Some human studies found that low-glycemic-index meals left people less hungry hours later, and one large European trial found slightly less weight regain on low-glycemic-index diets. But carefully controlled human diets that sharply lower insulin have not produced the extra fat loss the model predicts. Studies of energy use disagree with one another, and in one inpatient study people ate less on a high-carbohydrate diet that raised their insulin more.", "Both sides of this debate accept that stored fat can only grow if more energy comes in than goes out. What they disagree about is what drives that imbalance. Current evidence supports carbohydrate-driven insulin as one factor that can amplify fat storage or appetite in some people and situations. It does not show that this factor is the primary cause of common obesity. A large feeding trial designed to test one key prediction has been completed but its main results are not yet published." ], "grade_restated_faithfully": true, "claims_referenced": [ "C1", "C2", "C4", "C6", "C23", "C26", "C27", "C30", "C31" ], "jargon_glossed": [ "quickly digested carbohydrate: plain-language gloss for high glycemic load", "energy balance: stored fat can only grow if more energy comes in than goes out" ], "no_practical_advice": true, "written_for": "general reader, no specialist nutrition background" }, "hypothesis": { "name": "carbohydrate-insulin model as the primary cause of obesity — Red-Team Synthesis Edition (2026-09-30)", "central_claim": "High-glycemic-load diets drive insulin secretion (a raised insulin-to-glucagon ratio), adipose storage, hunger and lower energy expenditure, and this pathway is the primary cause of common obesity, with fat storage driving positive energy balance rather than following it.", "proposed_mechanism": "Rapidly digestible carbohydrate raises insulin relative to glucagon (with incretin contributions), shifts substrate partitioning toward adipose storage, lowers circulating or oxidizable fuels available to other tissues in the late postprandial period, elicits hunger and lower expenditure, and thereby causes further intake and fat gain; effects on expenditure are said to require weeks of adaptation.", "claimed_scope": "broad common-human-obesity primacy", "claimed_level": [ "initiation", "progression", "primacy" ], "is_conjunctive": true, "conjunctive_note": "The claim requires a linked endocrine, partitioning, appetite, energy-expenditure, adiposity, and primacy chain; a true acute endocrine response (SC1, graded Η) cannot stand in for the later links." }, "rubric_routing": { "primary_rubric": "06-metabolism-aging.md", "secondary_rubrics": [], "rubrics_checked": [ "01-neurodegeneration.md", "02-cancer-biology.md", "03-epigenetics-gene-regulation.md", "04-microbiome-systemic.md", "05-immunology-inflammation.md", "06-metabolism-aging.md", "07-psychiatry-neuromodulation.md" ], "index_notes_applied": [ "No carbohydrate-insulin-model entry exists in hypothesis-index.md; the index's 'Microbiome in obesity' entry (rubric 04) is a different hypothesis and was not applied." ], "rubric_fit_note": "Rubric 06 scope: 'aging drivers, caloric restriction mechanisms, mTOR, NAD/sirtuins, free radicals, senescence as cause, mitochondrial decline, proteostasis aging, and metabolic master-regulator models.' It is the closest fit (metabolic master-regulator model), but its thresholds are written for lifespan/healthspan (e.g. Δ: 'Requires at least one intervention changing lifespan or healthspan in a way consistent with the hypothesis'). They are applied here with adiposity/weight regulation as the organismal phenotype; this adaptation is declared per SKILL.md Step 5 ('If no rubric fits, say so in the review rather than forcing one') and flagged as a rubric gap: no rubric in the packet covers energy or body-weight regulation. Rubric 06's Γ hard rules ('Biomarker-only improvement without organismal benefit'; 'Single-species or single-strain evidence') and confounder 'Caloric intake confounding' apply without adaptation. Only one rubric governs, so the cross-field lowest-ceiling rule does not arise." }, "subclaim_matrix": { "subclaims": [ { "id": "SC1", "statement": "Increasing digestible carbohydrate or glycemic load relative to an appropriate comparator increases postprandial insulin secretion and the insulin-to-glucagon ratio in humans.", "subclaim_type": "mechanism", "governing_rubric": "06-metabolism-aging.md", "claimed_scope": "human postprandial physiology", "scope": "Acute and day-long insulin/C-peptide and glucagon responses to foods or diets differing in carbohydrate amount or glycemic quality, with protein, fat, food form and insulin sensitivity recorded as modifiers; people with functioning β cells.", "key_evidence_anchor": "Controlled human feeding with insulin/C-peptide measurement (several laboratories) and human monogenic perturbation of β-cell glucose sensing (GCK, KCNJ11) with pharmacological rescue.", "hsef_level": "Η", "confidence": "High", "governing_cap_or_threshold": "Η met ('Strong convergent causal evidence with intervention-sensitive trajectory change and no major unresolved upstream/downstream contradiction within the claimed scope'); Θ not met ('High predictive accuracy with minimal unresolved exceptions') because carbohydrate content predicts food insulin demand only modestly (r = 0.39 across foods; protein-rich foods disproportionately insulinogenic).", "rationale": "Two independent causal lines converge: dietary perturbation (22.3% and 47% falls in 24-h C-peptide with carbohydrate restriction; GI-graded meals move insulin as predicted) and human genetics of β-cell glucose sensing (GCK mutations right-shift the glucose–insulin-secretion curve; KCNJ11 activating mutations abolish glucose coupling and sulfonylureas restore it). The response is intervention-sensitive and has no contradiction within scope. The original Ε reflected a decision not to test upward, not the evidence (red-team RT-02)." }, { "id": "SC2", "statement": "The carbohydrate-driven insulin response shifts fuel partitioning toward adipose storage enough to cause net fat accumulation in humans, so that storage precedes and drives positive energy balance.", "subclaim_type": "mechanism", "governing_rubric": "06-metabolism-aging.md", "claimed_scope": "human fat-mass change under energy- and protein-controlled conditions", "scope": "Net human fat-mass change over days to months, distinguished from acute inhibition of lipolysis, tested at controlled energy and protein intake, with study duration recorded.", "key_evidence_anchor": "Human controlled-feeding fat-balance studies and their meta-analyses; engineered-mouse insulin gene dosage and human MR only at their own scope.", "hsef_level": "Γ", "confidence": "Moderate", "governing_cap_or_threshold": "Universal cap 4: 'Biomarker change without functional/phenotypic change caps at Γ.' Rubric 06 Γ hard rules: 'Biomarker-only improvement without organismal benefit'; 'Single-species or single-strain evidence'. Boundary-check Q5 negative at human scope.", "rationale": "Mouse Ins1 gene dosage crosses Δ within the mouse system and MR supports a bounded contribution, but the direct human prediction (less fat accumulation at equal intake on lower glycemic load) is negative in short studies and in a 32-study synthesis, and no long-duration human fat-balance test exists; confidence is Moderate because the proponents' adaptation argument has not been tested for fat balance." }, { "id": "SC3", "statement": "Carbohydrate-driven insulin responses lower late-postprandial fuel availability, which increases hunger and subsequent voluntary energy intake.", "subclaim_type": "mechanism", "governing_rubric": "06-metabolism-aging.md", "claimed_scope": "human hunger and ad-libitum intake across common dietary contexts", "scope": "Randomized human meals or diets with measured insulin, postabsorptive fuels, appetite ratings and subsequent ad libitum intake; acute, weight-maintenance and ad libitum contexts kept separate from chronic weight gain.", "key_evidence_anchor": "Randomized GI/GL manipulations with matched macronutrients measuring fuels, hunger and next-meal intake, and a direct mediator test.", "hsef_level": "Γ", "confidence": "Moderate", "governing_cap_or_threshold": "Boundary check Step 9 Q5 ('Is that perturbation's result positive? A study of exactly the right design that returns a null or contradicts the hypothesis does not cross Δ.') answered 'unclear' for the mechanism as stated: exposure-level GI effects are positive in defined contexts, the one direct mediator test failed, and the largest insulin contrast produced lower intake.", "rationale": "Positive: adolescent high- vs medium-GI intake (+53%), adult blinded crossover (lower glucose and greater hunger at 4 h), DIOGENES (0.95 kg less regain on low GI), group-level lower intake after low-GI meals in 2025. Negative: hunger did not follow insulin, individual-level intake unrelated to fuels or insulin (weak association opposite to prediction), and higher insulin with lower intake in the 2021 inpatient trial. One of the two subclaims closest to Δ." }, { "id": "SC4", "statement": "After an adaptation period of weeks, lowering carbohydrate (glycemic load) at equal energy and protein intake produces a sustained, physiologically important increase in energy expenditure or greater fat loss.", "subclaim_type": "sufficiency", "governing_rubric": "06-metabolism-aging.md", "claimed_scope": "human isocaloric energy expenditure and fat regulation, duration recorded", "scope": "Controlled-feeding comparisons with energy and protein held equal and validated expenditure and body-composition measures, separating studies shorter and longer than about 2.5 weeks.", "key_evidence_anchor": "NIH ward studies, two opposing meta-analyses of controlled feeding, the 2012 expenditure crossover, and the Framingham State Food Study with its reanalysis, response and energy-requirement analysis.", "hsef_level": "Γ", "confidence": "Moderate", "governing_cap_or_threshold": "Universal cap 4: 'Biomarker change without functional/phenotypic change caps at Γ.' Boundary check Step 9 Q5 answered 'contested': the only randomized parallel test is positive by the revised analysis plan and null by the registered one, and no measured fat-loss correlate exists.", "rationale": "Short isocaloric studies favour lower-fat diets for expenditure and fat balance; longer studies (6 trials) show higher expenditure on lower-carbohydrate diets, supported by the 2012 crossover and the Framingham energy-requirement analysis (one research group), but few (three of six by doubly labelled water), analysis-dependent, and without a fat-loss correlate; the independent 2016 ward study found a small chamber increase that waned and slowed fat loss. The subclaim closest to Δ; a completed residential isocaloric trial (results unpublished) bears on it directly." }, { "id": "SC5", "statement": "Baseline insulin secretion identifies people who benefit preferentially from lowering glycemic load.", "subclaim_type": "generalization", "governing_rubric": "06-metabolism-aging.md", "claimed_scope": "prespecified effect modification in adults with overweight or obesity", "scope": "Baseline metabolic phenotype measured before randomization and tested as a prespecified diet-by-phenotype interaction on weight or fat outcomes, with independent reproduction.", "key_evidence_anchor": "Ebbeling 2007 effect modification, Framingham insulin-secretion subgroup, DIETFITS prespecified interaction and its exploratory reanalysis.", "hsef_level": "Γ", "confidence": "Moderate", "governing_cap_or_threshold": "Boundary check Step 9 Q2 not satisfied: no prospectively defined, independently reproduced interaction; the one large prespecified independent test was null.", "rationale": "Positive: INS-30 effect modification in a 73-person trial (P = .02) and a larger expenditure contrast in the Framingham high-secretion third, plus an exploratory DIETFITS reanalysis on achieved (non-randomized) glycemic load. Negative: DIETFITS prespecified diet × INS-30 interaction null (P = .47)." }, { "id": "SC6", "statement": "The carbohydrate-insulin pathway is the primary, necessary, or rate-limiting cause of common human obesity rather than one contributor among multiple regulators of energy intake, expenditure, and partitioning.", "subclaim_type": "primacy", "governing_rubric": "06-metabolism-aging.md", "claimed_scope": "common human obesity across ordinary food environments", "scope": "Population-relevant human weight gain and obesity, assessed against food processing, energy density, eating rate, palatability, protein, activity, sleep, medication, neuroendocrine, and genetic alternatives without outcome-fitted qualifiers.", "key_evidence_anchor": "Concordance of direct human diet perturbations, long-term trials, genetic evidence, and alternative dietary perturbations that change intake and weight at similar presented macronutrients.", "hsef_level": "Γ", "confidence": "High", "governing_cap_or_threshold": "Rubric 06 Γ hard rule 'Single-species or single-strain evidence' (the only in vivo causal demonstration of insulin-driven adiposity is one engineered mouse model); boundary check Q2 and Q5 fail at human scope; framework role step 3 requires removal-prevents, engagement-initiates and rate-limiting, none shown.", "rationale": "Carbohydrate-insulin signalling can contribute in defined contexts, but lowering glycemic load does not produce greater fat loss at equal intake and yields at most a small long-term advantage; a higher-glycemic-load, higher-insulin diet produced lower intake in the only inpatient test; excess dietary fat is stored efficiently; and food architecture changes intake at matched macronutrient proportions (though glycemic load was not matched there, so that trial is not treated as a decisive falsifier)." } ], "aggregation_rule": "weakest-link", "aggregation_preregistered": true, "aggregation_rationale": "The submitted statement is a serial causal chain ending in primacy. A robust carbohydrate-to-insulin response cannot rescue a failed hunger, isocaloric, or primacy link.", "primary_subclaim_id": "", "aggregation_derivation": "SC1=Η; SC2=Γ; SC3=Γ; SC4=Γ; SC5=Γ; SC6=Γ. Weakest-link therefore returns Γ.", "aggregation_alternatives": { "weakest_link_would_return": "Γ", "primary_subclaim_would_return": "Γ", "scope_matched_would_return": "Γ", "note": "SC6 is the primary and the only full-scope subclaim and is Γ, so all three rules return Γ. SC1's Η does not bear on the composite under any rule because its scope (the endocrine link) does not match the central claim's scope." }, "index_rule_deviation": "" }, "competing_models": { "models_evaluated": [ { "rank_in_tournament": 1, "model_id": "M2", "is_nominal": false, "name": "Integrated energy-homeostasis model", "core_claim": "Brain, endocrine, autonomic, adipose, gut, muscle, genetic, and environmental signals jointly regulate intake, expenditure, and partitioning; insulin is one signal within that system.", "explains_what_nominal_does_not": "It accommodates acute insulin physiology, similar long-term results from different macronutrient prescriptions, and intake changes caused by food architecture at matched presented macronutrients.", "strongest_supporting_evidence": [ "Short-term isocaloric ward trials and the 32-study controlled-feeding synthesis [1,2,21]", "DIETFITS and long-term trials [11-13]", "Ultra-processed diet perturbation and independent replication [14,44]", "Lower intake despite higher insulin in the 2021 inpatient trial [3]" ], "strongest_contradictory_evidence": [ "Longer-duration expenditure signals on lower-carbohydrate diets [9,22-24]", "GI effects on late-postprandial hunger and weight regain [32,33]", "Model breadth makes it harder to falsify (reviewer assessment, applied symmetrically to the CIM responder claims)" ], "hsef_level_for_this_model": "Ε", "rationale": "Graded on its testable core: food-environment and energy-prescription perturbations change intake and adiposity in defined contexts [3,14,18,44]. As a broad framework it is harder to falsify than the CIM; the letter applies to the testable core only, not to the whole framework." }, { "rank_in_tournament": 2, "model_id": "M3", "is_nominal": false, "name": "Food-form and dietary-architecture model", "core_claim": "Energy density, eating rate, texture, palatability, protein leverage, fiber matrix, and processing can alter voluntary intake and weight independently of a uniquely carbohydrate-driven insulin pathway.", "explains_what_nominal_does_not": "It explains the large intake and weight difference between ultra-processed and unprocessed diets with matched presented macronutrients and several nutrients.", "strongest_supporting_evidence": [ "Hall et al. ultra-processed inpatient crossover [14]", "Independent 8-week free-living UPF vs minimally processed crossover [44]", "Lower intake on the minimally processed high-carbohydrate low-fat diet [3]" ], "strongest_contradictory_evidence": [ "Glycemic load was not matched or reported in the UPF inpatient trial, so a glycemic-load route is not excluded [14]", "Mediators (energy density, eating rate, texture) not isolated" ], "hsef_level_for_this_model": "Ε", "rationale": "Two randomized feeding trials in different settings establish context-specific causal effects of dietary architecture on intake and weight; mediators remain incompletely separated." }, { "rank_in_tournament": 3, "model_id": "M4", "is_nominal": false, "name": "Context-specific carbohydrate sensitivity", "core_claim": "High glycemic load or high insulin secretion worsens appetite or adiposity in identifiable people or dietary contexts, without being the universal primary cause of obesity.", "explains_what_nominal_does_not": "It preserves the positive adolescent, mouse, genetic, and Framingham signals while allowing null results in DIETFITS and metabolic-ward studies.", "strongest_supporting_evidence": [ "Ludwig 1999 and Lennerz 2013 GI crossovers [8,32]", "DIOGENES low-GI weight-regain effect [33]", "Longer-duration expenditure signals [9,22-24]", "Insulin-secretion effect modification [9,34,35]", "Mehran mouse genetics and Astley MR [15,16]" ], "strongest_contradictory_evidence": [ "DIETFITS prespecified INS-30 interaction null [11]", "Failed mediator test and hunger null in 2025 [7]", "Higher insulin with lower intake in 2021 [3]" ], "hsef_level_for_this_model": "Γ", "rationale": "Plausible with partial-system and several context-limited human signals, but no prospectively defined, independently reproduced responder rule or mediated chain." }, { "rank_in_tournament": 4, "model_id": "M1", "is_nominal": true, "name": "Primary carbohydrate-insulin model", "core_claim": "Carbohydrate-induced insulin diverts fuel into adipose tissue, causes hunger and lower expenditure, and is the dominant cause of common obesity independently of energy intake as such.", "explains_what_nominal_does_not": "As the nominal model, it supplies a clear directional chain from dietary exposure to storage, appetite, and obesity if every link holds.", "strongest_supporting_evidence": [ "Endocrine link at Η [1,2,7,8,42,43]", "GI effects on hunger, intake and regain [8,32,33]", "Framingham published analysis and energy-requirement analysis [9,24]", "Mouse and human genetics [15,16]" ], "strongest_contradictory_evidence": [ "Isocaloric ward tests and 32-study synthesis [1,2,21]", "Lower intake despite higher insulin [3]", "2025 mediator test [7]", "DIETFITS effect-modification null [11]", "Registered-anchor reanalysis of Framingham [10]" ], "hsef_level_for_this_model": "Γ", "rationale": "The first link is at Η and several downstream links have context-limited positive signals, but the linked human hunger, isocaloric and primacy predictions are negative, mixed or contested in direct tests." } ], "tournament_summary": "The integrated model best explains why insulin changes predictably while appetite, expenditure and fat loss do not move in one fixed direction; the context-specific glycemic-load model (a special case of it) accounts for the positive signals; the primary-cause formulation ranks last because its defining human predictions are negative, mixed or contested.", "study_separating_rank1_from_rank2": "A long-duration inpatient or residential randomized trial that factorially varies carbohydrate/glycemic load and food processing or food form while clamping protein and energy density, with ad-libitum intake, insulin, fuel flux, expenditure, and fat mass measured prospectively.", "nominal_wins_only_by_funding_artifact": false, "level_unchanged_by_tournament": true }, "evidence_inventory": [ { "citation": "Hall et al. 2015, Cell Metabolism", "source_pointer": "PMID 26278052; PMCID PMC4603544; DOI 10.1016/j.cmet.2015.07.021", "relates_to_subclaims": [ "SC1", "SC2", "SC4" ], "favors_models": [ "M2" ], "evidence_class": "Strong", "system": "human metabolic ward", "decisive_perturbation": true, "endogenous": true, "direction": "contradicts", "what_it_shows": "Carbohydrate restriction lowered 24-h C-peptide by 22.3% but calorimetry-measured fat loss was greater with fat restriction at equal prescribed energy and protein; DXA did not detect a difference and the carbohydrate-restricted diet produced more weight loss.", "what_it_does_not_show": "Six-day periods do not settle long-term adherence or very-low-carbohydrate ketogenic effects.", "confounders_present": [ "short duration", "reduced-carbohydrate diet was not ketogenic", "small inpatient sample" ], "verification": "verified", "verification_note": "PMC full text re-read in red-team pass (2026-09-30).", "source_tier": "A" }, { "citation": "Hall et al. 2016, American Journal of Clinical Nutrition", "source_pointer": "PMID 27385608; PMCID PMC4962163; DOI 10.3945/ajcn.116.133561", "relates_to_subclaims": [ "SC1", "SC2", "SC4" ], "favors_models": [ "M2", "M4" ], "evidence_class": "Strong", "system": "human metabolic ward", "decisive_perturbation": true, "endogenous": true, "direction": "null", "what_it_shows": "An isocaloric ketogenic diet lowered insulin secretion by 47% ± 3%; chamber EE rose modestly and waned, DLW EE rose, and body-fat loss slowed rather than accelerated.", "what_it_does_not_show": "The sequential design and short duration do not establish long-term weight outcomes.", "confounders_present": [ "fixed sequence", "unintentional energy deficit", "small male-only sample" ], "verification": "verified", "verification_note": "PMC full text re-read in red-team pass (2026-09-30).", "source_tier": "A" }, { "citation": "Hall et al. 2021 with 2024-2025 reanalyses, replies and secondary analyses", "source_pointer": "PMID 33479499; PMID 38128881; PMID 38286396; PMID 38316213; PMID 39163976; PMID 39113385; PMID 40846772; PMID 40846773; NCT03878108", "relates_to_subclaims": [ "SC1", "SC3", "SC6" ], "favors_models": [ "M2", "M3", "M4" ], "evidence_class": "Strong", "system": "human inpatient randomized crossover", "decisive_perturbation": true, "endogenous": true, "direction": "contradicts", "what_it_shows": "The high-glycemic-load low-fat diet produced greater urinary C-peptide yet lower ad libitum intake than the ketogenic diet, without significant differences in self-reported appetite; the no-washout order effect is disputed between two analyses.", "what_it_does_not_show": "Because the diets differed in many food properties and there was no washout, it does not isolate insulin as the only mediator or settle chronic obesity.", "confounders_present": [ "no washout", "food-source and diet-architecture differences", "short two-week periods", "order dispute" ], "verification": "verified", "verification_note": "Parent at abstract depth (paywalled); C-peptide and appetite statements verified in open full text of same-study reports (PMID 39163976, PMID 39113385); both order analyses verified.", "source_tier": "A" }, { "citation": "Liu et al. 2025, Cell Metabolism", "source_pointer": "PMID 40043690; DOI 10.1016/j.cmet.2025.01.015; erratum PMID 40306286", "relates_to_subclaims": [ "SC1", "SC3" ], "favors_models": [ "M2", "M4" ], "evidence_class": "Strong", "system": "human randomized meal trial", "decisive_perturbation": true, "endogenous": true, "direction": "null", "what_it_shows": "Mixed: glucose and insulin followed the predicted GI pattern; at group level low-GI meals led to lower next-meal intake (CIM-consistent); hunger did not follow; individual-level intake was unrelated to fuels or insulin, with a weak association opposite to prediction.", "what_it_does_not_show": "A single breakfast in lean healthy adults cannot determine chronic weight change or obesity incidence.", "confounders_present": [ "single-meal horizon", "lean healthy participants", "parallel groups", "published erratum (Figure 2F only)" ], "verification": "verified", "verification_note": "Abstract, PubMed erratum record, publisher erratum text and ClinicalTrials.gov registry read; full text not accessible in red-team pass.", "source_tier": "A" }, { "citation": "Ludwig et al. 1999, Pediatrics", "source_pointer": "PMID 10049982; DOI 10.1542/peds.103.3.e26", "relates_to_subclaims": [ "SC3" ], "favors_models": [ "M1", "M4" ], "evidence_class": "Moderate", "system": "human acute crossover", "decisive_perturbation": true, "endogenous": true, "direction": "supports", "what_it_shows": "A high-GI condition increased later voluntary intake (53% vs macronutrient-matched medium GI; 81% vs low GI) with the predicted postabsorptive fuel and hormone pattern in 12 adolescent boys with obesity.", "what_it_does_not_show": "It does not show chronic weight gain, broad adult generalization, or that insulin rather than correlated meal features mediated intake.", "confounders_present": [ "n=12", "acute horizon", "adolescent male sample", "abstract-only audit" ], "verification": "verified", "verification_note": "PubMed abstract and metadata read; only abstract-reported results used.", "source_tier": "A" }, { "citation": "Ebbeling et al. 2018 with reanalysis, response and energy-requirement analysis", "source_pointer": "PMID 30429127; PMID 31548574; PMID 31586125; PMID 32470981; NCT02068885", "relates_to_subclaims": [ "SC4", "SC5" ], "favors_models": [ "M1", "M4" ], "evidence_class": "Strong", "system": "human randomized weight-loss-maintenance trial", "decisive_perturbation": true, "endogenous": true, "direction": "supports", "what_it_shows": "By the published (post-weight-loss baseline) analysis, TEE rose as carbohydrate fell, more so with high insulin secretion; energy requirement to hold weight was also higher on low carbohydrate.", "what_it_does_not_show": "The registered pre-weight-loss anchor was null, unaccounted energy was large, and no fat-loss advantage was measured; the parties dispute the appropriate baseline and exclusions.", "confounders_present": [ "analysis-anchor dispute", "unaccounted energy", "same-dataset reanalysis", "provided-food adherence uncertainty" ], "verification": "verified", "verification_note": "Parent, reanalysis and response read in full text; energy-requirement analysis at abstract depth; grouped as one study.", "source_tier": "A" }, { "citation": "Gardner et al. 2018, DIETFITS", "source_pointer": "PMID 29466592; PMCID PMC5839290; NCT01826591", "relates_to_subclaims": [ "SC5", "SC6" ], "favors_models": [ "M2", "M3" ], "evidence_class": "Strong", "system": "human 12-month randomized trial", "decisive_perturbation": true, "endogenous": true, "direction": "null", "what_it_shows": "Healthy low-fat and low-carbohydrate approaches produced similar mean weight loss, and baseline insulin secretion did not modify the diet effect.", "what_it_does_not_show": "Free-living adherence convergence prevents a strict test of sustained extreme carbohydrate separation.", "confounders_present": [ "free-living adherence", "moderate 12-month macronutrient separation", "behavioral cointervention" ], "verification": "verified", "verification_note": "Complete corrected PMC article read.", "source_tier": "A" }, { "citation": "Sacks et al. 2009 and Foster et al. 2010", "source_pointer": "PMID 19246357; PMID 20679559", "relates_to_subclaims": [ "SC3", "SC4", "SC6" ], "favors_models": [ "M2", "M3" ], "evidence_class": "Strong", "system": "human two-year randomized trials", "decisive_perturbation": true, "endogenous": true, "direction": "null", "what_it_shows": "Long-term weight loss was not consistently superior with lower carbohydrate, and appetite ratings did not show the proposed general advantage where measured.", "what_it_does_not_show": "The trials cannot separate physiology from adherence because achieved diets converged over time.", "confounders_present": [ "attrition", "adherence convergence", "behavioral support" ], "verification": "verified", "verification_note": "Complete PMC primary articles read.", "source_tier": "A" }, { "citation": "Hall et al. 2019, ultra-processed diet trial", "source_pointer": "PMID 31105044; PMCID PMC7946062; NCT03407053", "relates_to_subclaims": [ "SC6" ], "favors_models": [ "M2", "M3" ], "evidence_class": "Strong", "system": "human inpatient randomized crossover", "decisive_perturbation": true, "endogenous": true, "direction": "null", "what_it_shows": "Food processing and correlated consumed-food properties changed intake and weight despite matching presented diets for macronutrients and several nutrients.", "what_it_does_not_show": "Glycemic index/load was neither matched nor reported and C-peptide trended higher on the ultra-processed diet, so the trial does not exclude a glycemic-load route and is not a direct test of CIM necessity.", "confounders_present": [ "glycemic load not matched", "consumed energy density differed", "eating rate differed", "short two-week periods" ], "verification": "verified", "verification_note": "Complete corrected PMC article read.", "source_tier": "A" }, { "citation": "Mehran et al. 2012, Cell Metabolism", "source_pointer": "PMID 23217255; DOI 10.1016/j.cmet.2012.10.019", "relates_to_subclaims": [ "SC2", "SC6" ], "favors_models": [ "M1", "M4" ], "evidence_class": "Moderate", "system": "engineered mouse", "decisive_perturbation": true, "endogenous": true, "direction": "supports", "what_it_shows": "Reduced insulin gene dosage protected mice from diet-induced obesity, establishing causal capacity of hyperinsulinemia in that engineered high-fat-diet context.", "what_it_does_not_show": "It does not establish high dietary carbohydrate, human hunger, or common-human obesity primacy.", "confounders_present": [ "engineered genotype", "mouse physiology", "high-fat diet context", "abstract-level audit" ], "verification": "verified", "verification_note": "PubMed and publisher abstracts read; no abstract-external number used.", "source_tier": "A" }, { "citation": "Astley et al. 2018, Clinical Chemistry", "source_pointer": "PMID 29295838; PMCID PMC5937525; DOI 10.1373/clinchem.2017.280727", "relates_to_subclaims": [ "SC2", "SC6" ], "favors_models": [ "M1", "M4" ], "evidence_class": "Moderate", "system": "human Mendelian randomization", "decisive_perturbation": false, "endogenous": true, "direction": "supports", "what_it_shows": "Genetically predicted glucose-stimulated insulin secretion was associated with higher BMI and the reverse direction was not supported in that analysis.", "what_it_does_not_show": "The instruments do not assign diet, exclude all pleiotropy, explain much exposure variance, or establish hunger and primacy.", "confounders_present": [ "instrument heterogeneity driven by a QPCTL/GIPR-region variant", "possible pleiotropy", "about 0.9% of insulin-30 variance explained in validation" ], "verification": "verified", "verification_note": "Complete PMC article and sensitivity analyses read.", "source_tier": "A" }, { "citation": "Horton et al. 1995 and Bray et al. 2012", "source_pointer": "PMID 7598063; PMID 22215165; NCT00565149", "relates_to_subclaims": [ "SC2", "SC6" ], "favors_models": [ "M2" ], "evidence_class": "Moderate", "system": "human controlled overfeeding", "decisive_perturbation": true, "endogenous": true, "direction": "contradicts", "what_it_shows": "Excess carbohydrate and fat can both be stored, and imposed excess energy increased body fat across protein levels.", "what_it_does_not_show": "Imposed-surplus designs cannot test whether glycemic load drives the surplus; the protein trial is not a carbohydrate-versus-fat comparison.", "confounders_present": [ "small samples", "short controlled overfeeding", "different primary contrasts" ], "verification": "verified", "verification_note": "Horton abstract and complete corrected Bray full text read.", "source_tier": "A" }, { "citation": "Ludwig et al. 2021, Hall et al. 2022, Ludwig 2023, Flier 2023, Magkos et al. 2024 perspectives", "source_pointer": "PMID 34515299; PMID 35134825; PMID 37661740; PMID 37086719; PMID 39164418", "relates_to_subclaims": [ "SC6" ], "favors_models": [ "M1", "M2" ], "evidence_class": "Weak", "system": "model framing", "decisive_perturbation": false, "endogenous": false, "direction": "supports", "what_it_shows": "Both camps accept energy balance and dispute causal direction upstream of it; the currently defended CIM names glycemic load and the insulin-to-glucagon ratio; neutral and joint-workshop perspectives describe the field as unresolved.", "what_it_does_not_show": "Perspectives cannot establish which model is causally correct.", "confounders_present": [ "advocacy framing", "selective evidence synthesis", "not primary causal evidence" ], "verification": "verified", "verification_note": "Both complete PMC perspectives read for framing only.", "source_tier": "B" }, { "citation": "Hall & Guo 2017, Gastroenterology (meta-analysis of controlled feeding)", "source_pointer": "PMID 28193517; PMCID PMC5568065", "relates_to_subclaims": [ "SC2", "SC4" ], "favors_models": [ "M2" ], "evidence_class": "Moderate", "system": "human controlled-feeding synthesis", "decisive_perturbation": false, "endogenous": true, "direction": "contradicts", "what_it_shows": "Across 32 controlled-feeding studies with isocaloric carbohydrate-for-fat substitution, expenditure and fat loss were modestly greater with lower-fat diets.", "what_it_does_not_show": "Most included studies were short; the synthesis cannot address post-adaptation effects.", "confounders_present": [ "short median duration", "heterogeneous designs" ], "verification": "verified", "verification_note": "Full text (PMC) read for inclusion criteria and direction.", "source_tier": "B" }, { "citation": "Ludwig et al. 2021, Journal of Nutrition (updated and reanalysed meta-analysis)", "source_pointer": "PMID 33274750", "relates_to_subclaims": [ "SC4" ], "favors_models": [ "M1", "M4" ], "evidence_class": "Moderate", "system": "human controlled-feeding synthesis", "decisive_perturbation": false, "endogenous": true, "direction": "supports", "what_it_shows": "Study duration modified the carbohydrate effect on TEE: reduced in shorter trials, increased in the few longer trials.", "what_it_does_not_show": "Only six longer trials, three measured by doubly labelled water (validity under macronutrient change disputed); no fat-balance correlate.", "confounders_present": [ "few long trials", "high heterogeneity in reanalysis" ], "verification": "verified", "verification_note": "Abstract read.", "source_tier": "B" }, { "citation": "Ebbeling et al. 2012, JAMA", "source_pointer": "PMID 22735432; PMCID PMC3564212; NCT00315354", "relates_to_subclaims": [ "SC4" ], "favors_models": [ "M1", "M4" ], "evidence_class": "Moderate", "system": "human controlled-feeding crossover after weight loss", "decisive_perturbation": true, "endogenous": true, "direction": "supports", "what_it_shows": "TEE decline after weight loss was greatest on low-fat, intermediate on low-GI and least on very-low-carbohydrate diets (trend by glycemic load P < .001).", "what_it_does_not_show": "Four-week periods; DLW under macronutrient change; no fat-mass outcome.", "confounders_present": [ "crossover without long washout", "DLW assumptions" ], "verification": "verified", "verification_note": "Abstract read.", "source_tier": "A" }, { "citation": "Lennerz et al. 2013, American Journal of Clinical Nutrition", "source_pointer": "PMID 23803881; PMCID PMC3743729; NCT01064778", "relates_to_subclaims": [ "SC3" ], "favors_models": [ "M1", "M4" ], "evidence_class": "Moderate", "system": "human randomized blinded crossover", "decisive_perturbation": true, "endogenous": true, "direction": "supports", "what_it_shows": "A high-GI meal matched for calories, macronutrients and palatability lowered plasma glucose and increased hunger 4 h later in overweight or obese men.", "what_it_does_not_show": "No ad libitum intake measure; n = 12; insulin not isolated as mediator.", "confounders_present": [ "small sample", "single-meal horizon" ], "verification": "verified", "verification_note": "Abstract read.", "source_tier": "A" }, { "citation": "Larsen et al. 2010, DIOGENES, NEJM", "source_pointer": "PMID 21105792; PMCID PMC3359496; NCT00390637", "relates_to_subclaims": [ "SC3", "SC6" ], "favors_models": [ "M4" ], "evidence_class": "Strong", "system": "human multicentre randomized ad libitum maintenance trial", "decisive_perturbation": true, "endogenous": true, "direction": "supports", "what_it_shows": "Low-GI ad libitum maintenance diets reduced weight regain after weight loss compared with high-GI diets (ITT).", "what_it_does_not_show": "Effect is modest and does not isolate insulin; protein arm also mattered.", "confounders_present": [ "free-living adherence", "attrition" ], "verification": "verified", "verification_note": "Abstract read.", "source_tier": "A" }, { "citation": "Ebbeling et al. 2007, JAMA", "source_pointer": "PMID 17507345; NCT00130299", "relates_to_subclaims": [ "SC5" ], "favors_models": [ "M4" ], "evidence_class": "Moderate", "system": "human randomized diet trial", "decisive_perturbation": true, "endogenous": true, "direction": "supports", "what_it_shows": "Baseline insulin-30 modified weight and body-fat response to low-glycemic-load versus low-fat diets.", "what_it_does_not_show": "n = 73; overall diets did not differ; not independently reproduced in a prespecified test.", "confounders_present": [ "small sample", "imputation of missing data" ], "verification": "verified", "verification_note": "Abstract read.", "source_tier": "A" }, { "citation": "Soto-Mota et al. 2023, AJCN (DIETFITS reanalysis)", "source_pointer": "PMID 36811468; NCT01826591", "relates_to_subclaims": [ "SC5" ], "favors_models": [ "M4" ], "evidence_class": "Weak", "system": "secondary analysis of randomized trial", "decisive_perturbation": false, "endogenous": true, "direction": "supports", "what_it_shows": "Achieved glycemic-load reduction tracked weight loss, with effect modification by baseline insulin secretion at 3 and 6 months.", "what_it_does_not_show": "Exposure (achieved GL) was not randomized; authors describe the analysis as exploratory.", "confounders_present": [ "post hoc", "non-randomized exposure" ], "verification": "verified", "verification_note": "Abstract read; grouped with DIETFITS.", "source_tier": "A" }, { "citation": "Tobias et al. 2015, Lancet Diabetes & Endocrinology (meta-analysis)", "source_pointer": "PMID 26527511; PMCID PMC4667723", "relates_to_subclaims": [ "SC6" ], "favors_models": [ "M2", "M4" ], "evidence_class": "Moderate", "system": "meta-analysis of long-term randomized trials", "decisive_perturbation": false, "endogenous": true, "direction": "supports", "what_it_shows": "In weight-loss trials of at least one year, low-carbohydrate interventions produced a small, consistent advantage over low-fat interventions.", "what_it_does_not_show": "Magnitude is small and mechanism-agnostic (protein, ketosis, adherence).", "confounders_present": [ "intervention-intensity differences" ], "verification": "verified", "verification_note": "Abstract read.", "source_tier": "B" }, { "citation": "Byrne et al. 1994 and Pearson et al. 2006 (human β-cell glucose-sensing genetics)", "source_pointer": "PMID 8132752; PMID 16885550", "relates_to_subclaims": [ "SC1" ], "favors_models": [ "M1", "M2", "M4" ], "evidence_class": "Strong", "system": "human monogenic perturbation with pharmacological rescue", "decisive_perturbation": true, "endogenous": true, "direction": "supports", "what_it_shows": "Glucokinase mutations right-shift the glucose–insulin-secretion dose–response; KCNJ11 activating mutations abolish glucose coupling and sulfonylureas restore insulin secretion, more with oral glucose or meals than intravenous glucose.", "what_it_does_not_show": "Addresses the endocrine link only, not adiposity or primacy.", "confounders_present": [ "small samples", "monogenic populations" ], "verification": "verified", "verification_note": "Abstracts read.", "source_tier": "A" }, { "citation": "Holt et al. 1997, AJCN (insulin index of foods)", "source_pointer": "PMID 9356547", "relates_to_subclaims": [ "SC1" ], "favors_models": [ "M2", "M3" ], "evidence_class": "Moderate", "system": "human acute food-response study", "decisive_perturbation": false, "endogenous": true, "direction": "null", "what_it_shows": "Glucose and insulin scores correlated, but total carbohydrate correlated only modestly with insulin demand and protein-rich foods were disproportionately insulinogenic.", "what_it_does_not_show": "Acute 2-h responses in healthy subjects; not a test of adiposity.", "confounders_present": [ "single-food portions", "finger-prick sampling" ], "verification": "verified", "verification_note": "Abstract read.", "source_tier": "A" }, { "citation": "Dicken et al. 2025, Nature Medicine", "source_pointer": "PMID 40760353; PMCID PMC12532614; NCT05627570", "relates_to_subclaims": [ "SC6" ], "favors_models": [ "M3", "M2" ], "evidence_class": "Moderate", "system": "human randomized crossover feeding, free-living", "decisive_perturbation": true, "endogenous": true, "direction": "null", "what_it_shows": "Minimally processed diets following national guidelines produced greater weight loss than ultra-processed diets following the same guidelines over 8 weeks.", "what_it_does_not_show": "Glycemic load not isolated; does not test the CIM chain directly.", "confounders_present": [ "free-living intake", "no washout reported in abstract" ], "verification": "verified", "verification_note": "Abstract read.", "source_tier": "A" } ], "verification_summary": { "load_bearing_claims_checked": 31, "verified": 31, "unverified": 0, "contradicted": 0, "note": "Red-team pass 2026-09-30: every identifier resolved by NCBI esummary (44 sources + 8 attached corrections; 0 mismatches). Every numeric claim was checked against an opened Tier A record at the stated access depth. One original number ('2.7-fold' C-peptide, source 3) could not be verified and was removed; the original '130 adults' (source 7) was replaced by the registry enrollment. Tier B syntheses carry direction only." }, "factual_issues_found": [ { "quoted_text": "rather than excess energy intake as such", "problem": "This sets a proposed biological mechanism against a conservation relationship and can create a false either-or comparison.", "source_says": "Both the carbohydrate-insulin and integrated energy-homeostasis perspectives accept energy balance; they disagree about the upstream causes that regulate intake, expenditure, and partitioning.", "source_pointer": "PMID 34515299; PMID 35134825", "affects_score": true, "action": "scored-on-corrected-understanding" }, { "quoted_text": "insulin secretion that promotes fat storage", "problem": "Acute inhibition of lipolysis or increased postprandial storage is not equivalent to sustained net fat gain independent of whole-body energy and substrate balance.", "source_says": "Human isocaloric studies lowered insulin substantially without the predicted preferential body-fat loss, while excess dietary fat was also efficiently stored.", "source_pointer": "PMID 26278052; PMID 27385608; PMID 7598063", "affects_score": true, "action": "scored-on-corrected-understanding" }, { "quoted_text": "increases hunger", "problem": "The generalized direction is not reproduced across controlled human settings.", "source_says": "Glycemic-index manipulations increased late-postprandial hunger or intake in some controlled settings (adolescent crossover; adult blinded crossover; group-level intake in 2025), but hunger did not track insulin in 2025, the one direct mediator test failed, and higher insulin coincided with lower intake in the 2021 inpatient trial.", "source_pointer": "PMID 10049982; PMID 23803881; PMID 40043690; PMID 33479499", "affects_score": true, "action": "scored-on-corrected-understanding" }, { "quoted_text": "primary cause of obesity", "problem": "Primary-driver status requires evidence that engaging the pathway initiates the phenotype, removing it prevents or materially delays the phenotype, and the pathway is rate-limiting across the claimed scope.", "source_says": "Controlled human studies identify multiple routes to higher intake and fat gain, and substantial carbohydrate or insulin reduction is not consistently sufficient for greater fat loss or lower hunger.", "source_pointer": "PMID 26278052; PMID 28193517; PMID 29466592; PMID 33479499", "affects_score": true, "action": "scored-on-corrected-understanding" }, { "quoted_text": "Dietary carbohydrate drives insulin secretion", "problem": "The currently defended model names glycemic load (rapidly digestible, processed carbohydrate) and the insulin-to-glucagon ratio, not total carbohydrate or insulin alone; carbohydrate content is also only a modest predictor of food insulin demand.", "source_says": "'Among modifiable factors, dietary glycemic load (GL) has central importance' (PMID 34515299); protein-rich foods elicit disproportionately high insulin responses (PMID 9356547).", "source_pointer": "PMID 34515299; PMID 37661740; PMID 9356547", "affects_score": false, "action": "scored-on-corrected-understanding" } ], "cap_rules_applied": [ { "subclaim_id": "SC2", "cap": "Γ", "cap_source": "universal", "triggered": true, "trigger_text_quoted": "Biomarker change without functional/phenotypic change caps at Γ.", "specific_trigger": "Insulin secretion changes substantially under controlled feeding without the predicted human adiposity outcome.", "resulting_ceiling": "Γ" }, { "subclaim_id": "SC2", "cap": "Γ", "cap_source": "rubric", "triggered": true, "trigger_text_quoted": "Single-species or single-strain evidence", "specific_trigger": "The only in vivo causal demonstration of hyperinsulinemia-driven adiposity is one engineered mouse model (Ins1+/-:Ins2-/-, high-fat diet).", "resulting_ceiling": "Γ" }, { "subclaim_id": "SC4", "cap": "Γ", "cap_source": "universal", "triggered": true, "trigger_text_quoted": "Biomarker change without functional/phenotypic change caps at Γ.", "specific_trigger": "Large falls in insulin secretion and RQ with carbohydrate restriction without a measured fat-loss advantage.", "resulting_ceiling": "Γ" }, { "subclaim_id": "SC4", "cap": "Γ", "cap_source": "rubric", "triggered": true, "trigger_text_quoted": "Biomarker-only improvement without organismal benefit", "specific_trigger": "Endocrine and substrate-oxidation shifts are not accompanied by adiposity change in isocaloric human studies.", "resulting_ceiling": "Γ" }, { "subclaim_id": "SC6", "cap": "Γ", "cap_source": "rubric", "triggered": true, "trigger_text_quoted": "Single-species or single-strain evidence", "specific_trigger": "Causal demonstration of insulin-driven adiposity is confined to one engineered mouse model; human evidence for primacy is absent.", "resulting_ceiling": "Γ" }, { "subclaim_id": "SC3", "cap": "Γ", "cap_source": "rubric", "triggered": false, "trigger_text_quoted": "Mechanistic plausibility plus one-species or in vitro support caps at Γ.", "specific_trigger": "Not triggered: SC3 rests on human perturbations; its Γ ceiling comes from boundary check Q5 (mediation unclear), recorded in boundary_check_BGD.", "resulting_ceiling": "none from cap" }, { "subclaim_id": "SC5", "cap": "Γ", "cap_source": "rubric", "triggered": false, "trigger_text_quoted": "Mechanistic plausibility plus one-species or in vitro support caps at Γ.", "specific_trigger": "Not triggered: SC5's Γ ceiling comes from boundary check Q2 (no independently reproduced prospectively defined interaction).", "resulting_ceiling": "none from cap" } ], "cap_precedence_resolutions": [ { "collision": "Direct human dietary and genetic perturbations establish the endocrine link at Η, while downstream appetite, expenditure, adiposity and primacy predictions do not cross Δ.", "subclaim_scope_cap_binds_on": "SC2-SC6 at sustained human adiposity, appetite, effect-modification, and primacy scope.", "subclaim_scope_threshold_binds_on": "SC1 at context-mapped human postprandial endocrine physiology.", "resolution": "SC1 is scored Η within its endocrine scope. Universal cap 4 and rubric 06 Γ hard rules govern the later links; the first link's success is not inherited by the chain.", "note": "A causal endocrine response and a negative downstream obesity test can coexist without inconsistency." }, { "collision": "Mouse insulin-gene manipulation establishes context-specific causality, while the nominal claim concerns dietary carbohydrate as the primary cause of common human obesity.", "subclaim_scope_cap_binds_on": "SC2 and SC6 at typical-human dietary and population scope.", "subclaim_scope_threshold_binds_on": "The engineered-mouse high-fat-diet context tested by Mehran et al.", "resolution": "The mouse result is retained as positive causal evidence in its own system; the cross-species and exposure change prevents it from lifting the human primacy subclaim.", "note": "No animal result is discarded, and none is silently promoted to human primacy." } ], "boundary_check_BGD": [ { "subclaim_id": "SC1", "field_valid_perturbation_quote": "Genetic or pharmacologic intervention producing reproducible lifespan/healthspan extension [applied with adiposity/weight regulation substituted for lifespan/healthspan; declared rubric adaptation]", "qualifying_evidence_present": "yes", "qualifying_evidence_citation": "Controlled human diet and meal perturbations [1,2,7,8]; human GCK and KCNJ11 genetics with sulfonylurea rescue [42,43].", "gamma_hard_rules_triggered": [], "disease_phenotype_relevance": "no", "perturbation_result_positive": "yes", "perturbation_result_note": "Positive and reproduced for the scoped endocrine response; the obesity phenotype is not part of SC1.", "ceiling_from_boundary_check": "Δ or higher" }, { "subclaim_id": "SC2", "field_valid_perturbation_quote": "Genetic or pharmacologic intervention producing reproducible lifespan/healthspan extension [applied with adiposity/weight regulation substituted for lifespan/healthspan; declared rubric adaptation]", "qualifying_evidence_present": "no", "qualifying_evidence_citation": "Human isocaloric fat-balance studies and synthesis [1,2,21] do not show the predicted consequence; the positive mouse study [15] binds only within the mouse system.", "gamma_hard_rules_triggered": [ "Biomarker-only improvement without organismal benefit", "Single-species or single-strain evidence" ], "disease_phenotype_relevance": "yes", "perturbation_result_positive": "no", "perturbation_result_note": "Step 9 Q5: 'Is that perturbation's result positive? A study of exactly the right design that returns a null or contradicts the hypothesis does not cross Δ.' Answer: no at human scope.", "ceiling_from_boundary_check": "Γ" }, { "subclaim_id": "SC3", "field_valid_perturbation_quote": "Genetic or pharmacologic intervention producing reproducible lifespan/healthspan extension [applied with adiposity/weight regulation substituted for lifespan/healthspan; declared rubric adaptation]", "qualifying_evidence_present": "unclear", "qualifying_evidence_citation": "GI perturbations positive for hunger, intake or regain in defined contexts [7,8,32,33]; mediator test failed [7]; higher insulin with lower intake [3].", "gamma_hard_rules_triggered": [], "disease_phenotype_relevance": "yes", "perturbation_result_positive": "unclear", "perturbation_result_note": "Step 9 Q5 answered 'unclear' for the mechanism as stated (insulin-mediated); 'Any negative or unclear answer → the ceiling is Γ for that subclaim.'", "ceiling_from_boundary_check": "Γ" }, { "subclaim_id": "SC4", "field_valid_perturbation_quote": "Genetic or pharmacologic intervention producing reproducible lifespan/healthspan extension [applied with adiposity/weight regulation substituted for lifespan/healthspan; declared rubric adaptation]", "qualifying_evidence_present": "unclear", "qualifying_evidence_citation": "Framingham positive by revised analysis and null by registered anchor [9,10]; energy-requirement analysis and response [24,25]; longer-trial synthesis [22] versus short-trial synthesis [21]; no fat-loss correlate [1,2].", "gamma_hard_rules_triggered": [ "Biomarker-only improvement without organismal benefit" ], "disease_phenotype_relevance": "yes", "perturbation_result_positive": "unclear", "perturbation_result_note": "Step 9 Q5 answered 'contested'; 'Any negative or unclear answer → the ceiling is Γ for that subclaim.' Closest subclaim to Δ; pending trial [37].", "ceiling_from_boundary_check": "Γ" }, { "subclaim_id": "SC5", "field_valid_perturbation_quote": "Genetic or pharmacologic intervention producing reproducible lifespan/healthspan extension [applied with adiposity/weight regulation substituted for lifespan/healthspan; declared rubric adaptation]", "qualifying_evidence_present": "no", "qualifying_evidence_citation": "Positive signals [9,34,35] are small, from one group, or post hoc; the prespecified independent test was null [11].", "gamma_hard_rules_triggered": [], "disease_phenotype_relevance": "yes", "perturbation_result_positive": "no", "perturbation_result_note": "Step 9 Q2 not satisfied: no prospectively defined, independently reproduced interaction.", "ceiling_from_boundary_check": "Γ" }, { "subclaim_id": "SC6", "field_valid_perturbation_quote": "Genetic or pharmacologic intervention producing reproducible lifespan/healthspan extension [applied with adiposity/weight regulation substituted for lifespan/healthspan; declared rubric adaptation]", "qualifying_evidence_present": "no", "qualifying_evidence_citation": "Human ward, long-term and inpatient ad libitum interventions [1-3,11-13,21,36] show neither necessity nor rate limitation.", "gamma_hard_rules_triggered": [ "Single-species or single-strain evidence" ], "disease_phenotype_relevance": "yes", "perturbation_result_positive": "no", "perturbation_result_note": "Q2 and Q5 fail at the claimed scope.", "ceiling_from_boundary_check": "Γ" } ], "thresholds_tested": [ { "subclaim_id": "SC1", "threshold": "Δ", "requirement_quoted": "At least one endogenous in vivo (or equivalent human interventional / natural-experiment) perturbation shows the mechanism contributes causally to the claimed phenotype.", "satisfied": "yes", "supporting_evidence": [ "Controlled human feeding alters insulin and C-peptide [1,2,7,8]." ], "what_would_cross_it": "Crossed." }, { "subclaim_id": "SC1", "threshold": "Ε", "requirement_quoted": "Context-specific causality established. The mechanism is causally operative in a defined context, subtype, stage, or model — but generalization to the full claimed scope fails.", "satisfied": "yes", "supporting_evidence": [ "Satisfied and exceeded: generalization within SC1's scope does not fail." ], "what_would_cross_it": "Crossed." }, { "subclaim_id": "SC1", "threshold": "Ζ", "requirement_quoted": "Multiple independent causal lines converge. Replication across at least two relevant systems, with major alternative models narrowed but not excluded.", "satisfied": "yes", "supporting_evidence": [ "Dietary perturbation line [1,2,7,8] and human monogenic β-cell glucose-sensing line with pharmacological rescue [42,43]; independence recorded in independence_basis." ], "what_would_cross_it": "Crossed." }, { "subclaim_id": "SC1", "threshold": "Η", "requirement_quoted": "Strong convergent causal evidence with intervention-sensitive trajectory change and no major unresolved upstream/downstream contradiction within the claimed scope.", "satisfied": "yes", "supporting_evidence": [ "Insulin secretion falls rapidly and persistently with carbohydrate restriction [2]; sulfonylurea restores secretion in KCNJ11 carriers [43]; no contradiction within scope." ], "what_would_cross_it": "Crossed." }, { "subclaim_id": "SC1", "threshold": "Θ", "requirement_quoted": "High predictive accuracy with minimal unresolved exceptions. The mechanism accounts for most major observations in its claimed scope.", "satisfied": "no", "supporting_evidence": [ "Carbohydrate content correlates only r = 0.39 with food insulin demand; protein-rich foods are disproportionately insulinogenic [41]." ], "what_would_cross_it": "A validated dietary-load metric that predicts insulin demand across foods and people with high accuracy." }, { "subclaim_id": "SC2", "threshold": "Δ", "requirement_quoted": "At least one endogenous in vivo (or equivalent human interventional / natural-experiment) perturbation shows the mechanism contributes causally to the claimed phenotype.", "satisfied": "no", "supporting_evidence": [ "Mouse gene dosage crosses within the mouse system only [15].", "Human isocaloric fat-balance results are negative [1,2,21]." ], "what_would_cross_it": "A long-duration human energy- and protein-controlled trial in which lower glycemic load produces greater net fat loss with insulin shown to mediate." }, { "subclaim_id": "SC3", "threshold": "Δ", "requirement_quoted": "At least one endogenous in vivo (or equivalent human interventional / natural-experiment) perturbation shows the mechanism contributes causally to the claimed phenotype.", "satisfied": "no", "supporting_evidence": [ "Exposure-level GI effects positive in defined contexts [7,8,32,33].", "Mediation failed [7]; higher insulin with lower intake [3]." ], "what_would_cross_it": "Independent adult randomized feeding reproducing the insulin → fuel → hunger → intake chain with prospective mediation at matched food architecture." }, { "subclaim_id": "SC4", "threshold": "Δ", "requirement_quoted": "At least one endogenous in vivo (or equivalent human interventional / natural-experiment) perturbation shows the mechanism contributes causally to the claimed phenotype.", "satisfied": "no", "supporting_evidence": [ "Positive by revised analysis, null by registered anchor [9,10,24,25].", "Longer-trial synthesis positive, short-trial synthesis negative [21,22]; no fat-loss correlate." ], "what_would_cross_it": "A replicated long-duration isocaloric protein-clamped trial showing sustained expenditure increase with a corresponding fat-balance advantage, robust to baseline choice (e.g. publication of [37] with independent replication)." }, { "subclaim_id": "SC5", "threshold": "Δ", "requirement_quoted": "At least one endogenous in vivo (or equivalent human interventional / natural-experiment) perturbation shows the mechanism contributes causally to the claimed phenotype.", "satisfied": "no", "supporting_evidence": [ "Positive: [9,34,35]; prespecified independent null: [11]." ], "what_would_cross_it": "Two independent trials using the same pre-registered INS-30 definition reproducing a meaningful diet interaction." }, { "subclaim_id": "SC6", "threshold": "Δ", "requirement_quoted": "At least one endogenous in vivo (or equivalent human interventional / natural-experiment) perturbation shows the mechanism contributes causally to the claimed phenotype.", "satisfied": "no", "supporting_evidence": [ "No human evidence of necessity, initiation or rate limitation [1-3,11-14,21,36]." ], "what_would_cross_it": "A population-relevant human perturbation programme showing that selectively removing the glycemic-load-insulin route prevents or materially delays obesity while engaging it initiates obesity under controlled alternatives." } ], "generalization_check": { "generalization_tested": "yes", "context_boundary_mapped": "yes", "internal_negative_controls": [ "Insulin fell during carbohydrate restriction without greater calorimetry-measured fat loss in the 2015 ward trial.", "A ketogenic diet lowered insulin 47% while body-fat loss slowed in the 2016 ward trial.", "The 2021 high-glycemic-load diet produced greater C-peptide but lower ad libitum intake without appetite differences.", "In 2025, hunger and individual-level intake did not track insulin, although group-level intake followed GI.", "DIETFITS did not reproduce baseline insulin secretion as a prespecified effect modifier." ], "note": "The endocrine link generalizes within its scope (SC1 Η). Downstream effects vary by duration, food architecture, population and analysis; exposure-level GI effects appear in defined contexts but the insulin-mediated chain has not been shown. Mapping a failed broad generalization does not establish context-specific causality for the full chain." }, "falsification_patterns_met": [ { "subclaim_id": "SC2", "pattern": "Pathway manipulation changes biomarkers but not lifespan/healthspan [rubric 06; applied as: insulin changes without the predicted adiposity change]", "evidence": "Carbohydrate restriction lowered insulin secretion 22-47% without greater fat loss [1,2]; 32-study synthesis favours lower-fat diets for fat loss [21].", "downgrade_applied": false, "downgrade_note": "Same observations set the Step 9 Q5 answer and the Γ ceiling; a further tier deduction would double-count them (literal application would give Β)." }, { "subclaim_id": "SC4", "pattern": "Pathway manipulation changes biomarkers but not lifespan/healthspan [rubric 06; applied as: substrate and insulin shifts without adiposity change]", "evidence": "RQ and insulin shifted markedly without a fat-loss advantage [1,2]; the positive expenditure trial had no fat-loss correlate [9].", "downgrade_applied": false, "downgrade_note": "Already reflected in the contested Q5 answer; not double-counted." }, { "subclaim_id": "SC3", "pattern": "Proposed mediator does not track the outcome [review-specific pattern; no rubric 06 analogue]", "evidence": "Individual-level intake unrelated to insulin and fuels, weak association opposite to prediction [7]; higher insulin with lower intake [3].", "downgrade_applied": false, "downgrade_note": "Already the basis of the 'unclear' Q5 answer." }, { "subclaim_id": "SC5", "pattern": "Effects disappear across strains or species [rubric 06; applied as: effect modification disappears in an independent prespecified trial]", "evidence": "DIETFITS prespecified diet × INS-30 interaction P = .47 [11].", "downgrade_applied": false, "downgrade_note": "Already the basis of the Q2 failure." }, { "subclaim_id": "SC6", "pattern": "Necessity fails when the proposed pathway is removed [rubric 06]", "evidence": "Lowering glycemic load produces at most a small long-term weight advantage [11-13,36] and no isocaloric fat-loss advantage [1,2,21]; excess dietary fat is stored efficiently [17].", "downgrade_applied": false, "downgrade_note": "Already the basis of the Q2/Q5 failure; the UPF trial is not used here because glycemic load was not matched [14]." } ], "pitfalls_flagged": [ { "subclaim_id": "SC2", "pitfall": "10. Mechanism conflation", "where": "Acute insulin-mediated suppression of lipolysis is kept separate from sustained net fat gain and common-obesity primacy." }, { "subclaim_id": "SC6", "pitfall": "7. Selective generalization", "where": "A mouse high-fat-diet genotype, 12 adolescent boys, and a disputed human subgroup are not generalized to all common obesity." }, { "subclaim_id": "SC4", "pitfall": "17. Negative field-valid perturbation read as a crossed threshold", "where": "The rigorous metabolic-ward experiments are negative for the nominal fat-loss prediction and therefore do not lift it to Δ." }, { "subclaim_id": "SC4", "pitfall": "18. Causal-evidence tallies that do not survive an audit", "where": "Parent trials, reanalyses, rebuttals, and corrections share study IDs and are not counted as independent causal lines." }, { "subclaim_id": "SC6", "pitfall": "13. Therapeutic effect misread as mechanistic proof", "where": "The clinical usefulness of a lower-carbohydrate diet for some people would not by itself prove insulin-mediated primacy." }, { "subclaim_id": "SC3", "pitfall": "14. Missing-control interpretation", "where": "Glycemic-index and macronutrient comparisons are interpreted with food form, energy density, protein, fiber, and washout limitations visible." }, { "subclaim_id": "SC5", "pitfall": "9. Replication confused with reproduction", "where": "Multiple analyses of Framingham and the 2021 crossover are one dataset each, not independent reproduction." }, { "subclaim_id": "SC1", "pitfall": "8. Confirmation by biomarker", "where": "Insulin and C-peptide confirm endocrine exposure but do not confirm hunger, obesity, or primacy." }, { "subclaim_id": "SC1", "pitfall": "16. Failed generalization mistaken for weak evidence", "where": "The original held SC1 at Ε without testing upward; within its scope generalization does not fail. Corrected to Η (red-team RT-02)." }, { "subclaim_id": "SC3", "pitfall": "19. Unlinked claims", "where": "An unsourced sentence about 'later analyses' of the 1999 crossover was removed (red-team RT-17)." }, { "subclaim_id": "SC6", "pitfall": "14. Missing-control interpretation", "where": "The ultra-processed-food trial did not match glycemic load, so it is not used as a decisive falsifier of the CIM (red-team RT-10)." } ], "causal_tally_audit": { "tally_claimed_by_document": 0, "tally_after_audit": 0, "items_rejected": [ "Insulin-lowering diet trials whose downstream fat-loss predictions are negative are not positive causal lines for primacy.", "The Framingham parent, reanalysis, response and energy-requirement analysis are one trial (NCT02068885).", "The 2021 parent, reanalysis, rebuttal, reply, order analysis, hormone analysis and 2025 correspondence are one participant dataset (NCT03878108).", "DIETFITS and its exploratory reanalysis are one trial (NCT01826591).", "Mouse insulin-gene dosage cannot establish dietary glycemic-load primacy in humans.", "MR of glucose-stimulated insulin secretion does not randomize diet and its heterogeneity traces to a GIPR-region variant.", "Meta-analyses (Tier B) are not additional causal lines; they synthesize the trials already counted." ] }, "steelman": { "case_one_rung_higher": "Δ: the endocrine link is at Η; GI manipulation changes late-postprandial glucose and hunger in a matched blinded crossover, raises next-meal intake, lowered group-level intake in 2025 and reduced weight regain in DIOGENES; six longer controlled-feeding studies, the 2012 crossover and the Framingham published and energy-requirement analyses show higher expenditure on lower-carbohydrate diets; insulin-secretion effect modification appears in two trials and one reanalysis; mouse gene dosage and MR support insulin-driven adiposity.", "case_one_rung_lower": "Β: the direct human tests of the chain are negative or contested, five falsification patterns are triggered, and much of the positive evidence is small, from one research group, or post hoc.", "which_rejected": "Δ is rejected because the preregistered weakest-link rule requires every link to cross: SC6 has no positive necessity, initiation or rate-limitation evidence; SC5's prespecified independent test was null; SC3's mediator test failed; SC4's key trial is null on its registered analysis and lacks a fat-loss correlate. Β is rejected because it describes associational evidence without causal dependency, whereas the endocrine link is causal at Η, insulin-driven adiposity is causal in mice, and several human GI perturbations are positive in defined contexts; the falsifying observations already set the Γ ceiling.", "blocking_cap_threshold_or_missing_experiment": "Missing experiment: a long-duration human trial in which lower glycemic load, at controlled energy, protein and food architecture, raises expenditure and reduces fat accumulation or ad libitum intake, with insulin shown to mediate. Boundary-check Q5 (SC3 unclear, SC4 contested) and Q2 (SC5, SC6) are the binding thresholds." }, "final_score": { "hsef_level": "Γ", "hsef_descriptor": "Mechanistically plausible, partial system support", "confidence": "High", "confidence_rationale": "Confidence reflects evidence density and reproducibility, not belief that the model is false. The composite is bound by SC6, whose Γ placement rests on a dense, multi-laboratory, multi-design base: two opposing syntheses of 29-32 controlled-feeding studies, several large independent long-term RCTs (DIETFITS, POUNDS LOST, Foster 2010, DIOGENES) and a 53-trial meta-analysis, inpatient feeding trials plus an independent free-living replication, and proponents' own statement that rigorous research is still needed to compare the models. SC2-SC5 are individually Moderate (conflicting, single-group or analysis-dependent evidence). High does not mean glycemic load never matters.", "rests_on_non_peer_reviewed_work": false, "rationale": "The endocrine link (carbohydrate/glycemic load → insulin and insulin-to-glucagon ratio) reaches Η on two independent causal lines. The downstream chain remains Γ: isocaloric fat-balance predictions are negative in short studies and untested at long duration; expenditure gains after adaptation are contested and lack a fat-loss correlate; GI effects on hunger, intake and regain are positive in defined contexts but insulin mediation failed its direct test; the responder rule failed its prespecified replication; and primacy has no necessity or rate-limitation evidence. Weakest-link, primary-subclaim and scope-matched aggregation all return Γ." }, "role_of_mechanism": { "classification": "reinforcing / stabilizing factor", "derivation_step_matched": "5", "derivation": "Step 1 (downstream consequence?): no; the dietary GL-to-insulin response exists independently of adiposity and bidirectional MR did not support BMI → insulin-30. Step 2 (epiphenomenon?): no; causal dependency is shown for the endocrine link and, in mice, for adiposity. Step 3 (primary driver: removal prevents AND engagement initiates AND rate-limiting?): no; none shown in humans. Step 4 (permissive?): no at the claimed scope; insulin signalling per se is required for adipose storage, but carbohydrate-driven insulin has not been shown to be required for common obesity, and excess dietary fat is stored efficiently. Step 5 matched: GI effects on late-postprandial hunger and on weight regain, and hyperinsulinemia-driven adiposity in mice, are amplifying or maintaining effects → reinforcing / stabilizing factor.", "tournament_used_as_tiebreaker": false, "note": "The classification concerns the strong common-obesity claim. It does not deny insulin's essential physiological role or the usefulness of carbohydrate restriction for selected outcomes or individuals." }, "extended_scoring_MSD": { "momentum": { "value": "Mechanistic only", "artifact": "2025 randomized GI meal trial (PMID 40043690); 2023-2025 reanalyses, replies and correspondence (PMIDs 36811468, 38128881, 38286396, 38316213, 39163976, 40466618, 40466619, 40846772, 40846773); appetite-hormone analysis (PMID 39113385); residential isocaloric trial design paper (PMID 38458559).", "justification": "Recent work is acute, reanalytic or correspondence; the one purpose-built longer residential test has published only its design and addresses SC4 alone. Nothing underway addresses the composite's binding subclaim (SC6)." }, "stability": { "value": "Stable", "trigger": "Primary publication of the residential isocaloric VLC vs HC-starch vs HC-sugar trial (PMID 38458559; 70 completers, 65% of planned) — direction unknown; bears on SC4 only.", "justification": "Under weakest-link the trigger cannot move the composite while SC3, SC5 and SC6 remain Γ; the trial is also underpowered relative to plan." }, "divergence": { "implied_level": "none (contested)", "implied_level_basis": "Proponents present the CIM as better reflecting the biology (PMID 34515299); opponents present the EBM as 'a more robust theory' (PMID 35134825); a neutral perspective finds 'neither has yet provided a validated mechanistic account' (PMID 37086719); a joint workshop calls for more testing (PMID 39164418).", "assessed_level": "Γ", "gap_in_rungs": null, "value": "Contested", "justification": "The field is genuinely split, so no single implied level applies. For transparency: the submitted wording 'primary cause', read literally, implies Η (Γ→Δ→Ε→Ζ→Η, +4, severe overstatement), while proponents' own wording ('central importance'; 'rigorous research is needed') implies a lower claim." }, "calibration_note": "M/S/D never modify HSEF level or confidence (Rule 10). Each is tied to a checkable artifact." }, "change_record": { "applies": true, "prior_level": "Γ", "prior_confidence": "High", "prior_source": "HSEF v5.4 original edition dated 2026-09-28 (authoring model OpenAI Codex)", "derived_before_consulting_prior": false, "new_level": "Γ", "direction": "unchanged (composite); SC1 raised Ε → Η; SC2-SC5 confidence High → Moderate", "what_moved_it": "Composite unchanged. SC1 regraded after testing thresholds upward with two independent lines; subclaim confidences re-derived on evidence density; divergence set to Contested per framework definition; balance restored by 24 added verified sources.", "divergence_from_other_reviewers": "Cross-model red team agrees on the composite letter and role; differs on SC1 level, subclaim confidence, divergence and several evidence characterizations (see redteam/REDTEAM_REPORT.md)." }, "proposed_decisive_experiments": [ { "name": "Factorial long-duration residential feeding trial", "experiment": "Randomize adults with overweight or obesity to a factorial design crossing high versus low glycemic load with minimally processed versus ultra-processed food architecture for at least 8-12 weeks (past the proposed adaptation window), then include a prespecified ad-libitum phase.", "readout": "Daily intake, hunger, insulin/C-peptide and glucagon, glucose, free fatty acids, ketones, substrate oxidation, doubly labelled water (with food quotient recorded) and chamber expenditure, body fat, and metabolic-flux mediation.", "critical_control": "Clamp protein, offered energy density, palatability targets, fiber type, sodium, meal schedule, and food mass where feasible; blind analysis; specify washout or parallel groups; audit adherence and unaccounted energy.", "falsifying_result": "Large insulin separation without the predicted hunger, intake, expenditure, or fat-mass ordering after food architecture is controlled.", "supporting_result": "A reproducible high-glycemic-load effect on intake and fat gain quantitatively mediated by insulin across both processing conditions." }, { "name": "Insulin-pathway mediation and rescue study", "experiment": "Within a controlled dietary platform, alter carbohydrate-driven insulin exposure using two mechanistically distinct, clinically acceptable perturbations and include a rescue that restores insulin signaling without restoring the carbohydrate exposure.", "readout": "Adipose lipolysis and esterification flux, circulating fuels, hypothalamic or behavioral appetite proxies, ad-libitum intake, expenditure, and fat balance.", "critical_control": "Match glycemia and other counter-regulatory hormones where possible, measure drug-specific off-target effects, and prespecify temporal mediation from insulin to fuels to appetite.", "falsifying_result": "Hunger and intake track carbohydrate or food properties when insulin is dissociated, or the phenotype persists when insulin exposure is selectively reduced.", "supporting_result": "Selective insulin reduction abolishes the fuel, hunger, and intake phenotype, and pathway-specific rescue restores it." }, { "name": "Prospective susceptibility replication", "experiment": "Use one fixed, pre-registered baseline insulin-secretion definition (e.g. INS-30 with a registered cut point) in at least two independent randomized trials with the same low- and high-glycemic-load diets.", "readout": "Prespecified diet-by-phenotype interaction on fat mass and weight, plus appetite and expenditure mediation.", "critical_control": "Register the cut point and interaction model before enrollment, maintain meaningful diet separation, harmonize protein and diet quality, and avoid post-outcome responder definitions.", "falsifying_result": "No replicated interaction or opposite-direction interactions across trials.", "supporting_result": "A consistent clinically meaningful interaction in the same independently defined metabolic subgroup." } ], "knowledge_gaps": [ { "name": "Long-term controlled mediation", "statement": "No long-duration human experiment simultaneously isolates glycemic load, food architecture, insulin exposure, appetite, expenditure, and fat-mass change.", "relevance": "This is the cleanest route to distinguishing the nominal chain from integrated and food-form models." }, { "name": "Quantitative sufficiency", "statement": "It is unresolved whether observed insulin-related expenditure or partitioning effects are large and durable enough to explain meaningful differences in common-obesity incidence.", "relevance": "A statistically detectable mediator is not automatically rate-limiting or population-primary." }, { "name": "Reproducible responder phenotype", "statement": "No pre-outcome insulin-secretion or insulin-resistance rule has been independently reproduced as a robust modifier of carbohydrate-restriction effects on fat loss.", "relevance": "A real responder stratum could raise the context-specific model even if the universal primary-cause claim remains false." }, { "name": "Diet-order adaptation", "statement": "The size and biological source of carryover after abrupt macronutrient changes without washout remain incompletely resolved.", "relevance": "This affects interpretation of short crossover trials but does not create an independent positive line from a reanalysis of the same participants." }, { "name": "Genetic exposure specificity", "statement": "Existing insulin-secretion instruments may affect adiposity through incretin, appetite, or other pleiotropic routes and do not encode dietary carbohydrate exposure.", "relevance": "Better instruments or triangulation are needed before genetics can carry dietary-mechanism primacy." }, { "name": "Unpublished residential isocaloric trial", "statement": "The primary expenditure and body-composition results of a completed 10-week residential VLC vs HC-starch vs HC-sugar trial (PMID 38458559) were not located in PubMed as of 2026-09-30.", "relevance": "Directly tests SC4 past the proposed adaptation window." }, { "name": "Adaptation window", "statement": "Only six isocaloric controlled-feeding studies in the updated synthesis exceed about 2.5 weeks, three of them measured by doubly labelled water.", "relevance": "The duration dependence proposed by proponents is testable but thinly tested." } ], "review_metadata": { "review_date": "2026-09-30", "reviewer": "OpenAI Codex using HSEF v5.4 (original 2026-09-28); red-team synthesis by Claude (Anthropic) 2026-09-30", "evidence_base_note": "Primary human feeding, long-term randomized, animal-genetic, human-genetic, and correspondence literature through 2026-09-30. Full text read where retrievable in the red-team pass; abstract-only and metadata-only records are labelled and carry no number beyond what was read.", "literature_tools_available": true, "literature_tools_note": "Red-team pass: NCBI E-utilities (esummary/efetch), PubMed MCP, PMC full text, Crossref, ClinicalTrials.gov API, publisher pages. Scite not used; citation-context counts omitted rather than encoded as zero.", "skill_version": "5.4", "validator_run": "pass", "validator_warnings_addressed": "Validator 2.2 run on the final pair in the 2026-09-30 red-team pass; validator 2.4 (archived HSEF v5.4) re-run for the consensus-consistency supplement and again on 2026-10-14 after adversarial review; output in validation-report.json.", "edition": "Red-Team Synthesis Edition (2026-09-30)", "consensus_consistency_audit": "2026-10-09; supplementary provenance/source reconciliation; original grading retained; prior grades visible." }, "is_calibration_entry": { "value": false, "entry_role": "not_calibration", "blind_graded": false, "completed_blinded_calibration": false, "expected_band_sealed_at": "", "target_sha256": "", "target_authorship": "No sealed calibration target was supplied.", "target_author_saw_review": null, "future_run_status": "not prepared", "unblinded_reason": "Standalone evaluation; no independently authored sealed expected band was supplied." }, "upper_ladder_check": { "final_level_below_zeta": true, "equivalents_considered": [ { "kind": "mendelian_randomization", "available": true, "why_not_sufficient": "Instruments glucose-stimulated insulin secretion rather than diet, explain about 0.9% of insulin-30 variance in validation, and their heterogeneity traces to a QPCTL/GIPR-region variant; does not test the hunger-to-obesity chain." }, { "kind": "multiple_human_dietary_perturbations", "available": true, "why_not_sufficient": "Several field-valid human trials exist, but for the full chain they are negative, mixed or contested (isocaloric fat balance negative; expenditure contested; mediation failed; responder replication null)." }, { "kind": "dose_response_across_populations", "available": true, "why_not_sufficient": "Dietary dose-response observations and trials are entangled with adherence, food form, energy density, protein, and socioeconomic context and do not establish insulin-mediated primacy." }, { "kind": "withdrawal_rechallenge", "available": true, "why_not_sufficient": "Crossover diet switches are short and lack washout; the order effect is disputed between analyses of the same participants." }, { "kind": "therapeutic_perturbation_hard_endpoint", "available": false, "why_not_sufficient": "No insulin-specific intervention establishes prevention of common obesity through this dietary pathway with a hard long-term endpoint and mechanistic mediation." } ], "convergence_present": false, "cap_reapplied_to_convergent_set": false }, "provenance_header": { "sources_total": 44, "claims_total": 31, "claims_verified": 31, "claims_unverified": 0, "claims_contradicted": 0, "tier_a_sources": 30, "generated_by": "OpenAI Codex with HSEF v5.4 (original); Claude red-team synthesis 2026-09-30; validator 2.2", "gate_passed": true }, "claims": [ { "id": "C1", "text": "Digestible carbohydrate and higher-glycemic meals produce larger postprandial insulin responses and a higher insulin-to-glucagon ratio than lower-carbohydrate or lower-glycemic comparators; protein, fat, food form and insulin sensitivity modify the size of the response.", "carries_number": false, "load_bearing_for": [ "SC1" ], "source_ns": [ 1, 2, 7, 8 ], "verification": "verified", "verified_against": "full texts of 1-2; abstracts of 7-8", "verified_on": "2026-09-30", "note": "Endocrine input only." }, { "id": "C2", "text": "In Hall et al. 2015, carbohydrate restriction lowered 24-h urinary C-peptide by 22.3% ± 7.0%; calorimetry-measured fat loss was 53 versus 89 g/day (cumulatively 245 ± 21 versus 463 ± 37 g) for carbohydrate versus fat restriction; DXA did not detect a significant difference and the carbohydrate-restricted diet produced more weight loss.", "carries_number": true, "load_bearing_for": [ "SC1", "SC2", "SC4" ], "source_ns": [ 1 ], "verification": "verified", "verified_against": "PMC full text, Results, Table 3, Figure 3", "verified_on": "2026-09-30", "note": "Six-day periods; non-ketogenic." }, { "id": "C3", "text": "In Hall et al. 2016, an isocaloric ketogenic diet lowered daily insulin secretion by 47% ± 3%; chamber EE rose 57 ± 13 kcal/day and sleeping EE rose then waned; DLW EE rose 151 ± 63 kcal/day; body-fat loss slowed.", "carries_number": true, "load_bearing_for": [ "SC1", "SC2", "SC4" ], "source_ns": [ 2 ], "verification": "verified", "verified_against": "PMC full text", "verified_on": "2026-09-30", "note": "Fixed sequence; overall negative energy balance about 300 kcal/day." }, { "id": "C4", "text": "In Hall et al. 2021, intake was 689 ± 73 kcal/day lower over two weeks and 544 ± 68 kcal/day lower in the final week on the high-glycemic-load (85 g/1,000 kcal) low-fat diet than on the ketogenic (6 g/1,000 kcal) diet; 24-h urinary C-peptide was greater on the low-fat diet in both order groups (P < 0.01) and self-reported appetite did not differ significantly.", "carries_number": true, "load_bearing_for": [ "SC1", "SC3", "SC6" ], "source_ns": [ 3, 6, 27 ], "verification": "verified", "verified_against": "abstract of 3; full texts of 6 and 27", "verified_on": "2026-09-30", "note": "No washout; diets differed in food source, fibre, mass and energy density." }, { "id": "C5", "text": "The order effect in the 2021 crossover is disputed: one reanalysis (16 of 20 participants) found a carry-over interaction P = .0004 and a non-significant net diet effect (P = .4); the originating group found the within-subject intake difference not significantly altered by modelling order (143 ± 140 kcal/day, P = .32); replies and 2025 correspondence continue the dispute.", "carries_number": true, "load_bearing_for": [ "SC3" ], "source_ns": [ 4, 5, 6, 26, 28, 29 ], "verification": "verified", "verified_against": "full texts of 4-6; metadata of 26, 28, 29", "verified_on": "2026-09-30", "note": "Same NCT03878108 participants; no independent causal line." }, { "id": "C6", "text": "In the 2025 randomized GI meal trial, glucose and insulin followed the predicted pattern but hunger did not; at group level low-GI meals led to lower next-meal intake changes; at individual level intake changes were unrelated to fuels and hormones, with a weak negative association with insulin at 300 min, opposite to prediction; published correspondence disputes the interpretation.", "carries_number": false, "load_bearing_for": [ "SC1", "SC3" ], "source_ns": [ 7, 30, 31 ], "verification": "verified", "verified_against": "abstract, erratum record and publisher erratum text of 7; ClinicalTrials.gov NCT05804942; metadata of 30-31", "verified_on": "2026-09-30", "note": "Lean healthy adults (registry BMI 18.5 to <24), single breakfast; registry enrollment 135; erratum corrects Figure 2F only." }, { "id": "C7", "text": "In a 12-participant adolescent crossover, voluntary intake after the high-GI meal was 5.8 MJ, 53% above the macronutrient-matched medium-GI meal and 81% above the low-GI meal.", "carries_number": true, "load_bearing_for": [ "SC3" ], "source_ns": [ 8 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "Acute; the low-GI meal is not described as macronutrient-matched." }, { "id": "C8", "text": "The Framingham State Food Study reported a TEE linear trend of 52 kcal/day (95% CI 23 to 82) per 10% decrease in carbohydrate, 209 kcal/day (91 to 326) for low versus high carbohydrate by ITT, and 308 kcal/day in the highest third of pre-weight-loss insulin secretion.", "carries_number": true, "load_bearing_for": [ "SC4", "SC5" ], "source_ns": [ 9 ], "verification": "verified", "verified_against": "PMC full text and abstract", "verified_on": "2026-09-30", "note": "Published analysis used the post-weight-loss baseline." }, { "id": "C9", "text": "A reanalysis of the same trial using the registered pre-weight-loss baseline found no diet effect (P = 0.43) and mean unaccounted energy of 483 ± 46 kcal/day.", "carries_number": true, "load_bearing_for": [ "SC4", "SC5" ], "source_ns": [ 10 ], "verification": "verified", "verified_against": "PMC full text", "verified_on": "2026-09-30", "note": "Shares NCT02068885 with source 9." }, { "id": "C10", "text": "The investigators responded that the post-weight-loss baseline is statistically preferable and that exclusions on unaccounted energy introduce post-randomization bias; a secondary analysis found energy requirement to hold weight higher on low carbohydrate by 181 (95% CI 8 to 353) to 246 (64 to 427) kcal/day across ITT models.", "carries_number": true, "load_bearing_for": [ "SC4" ], "source_ns": [ 24, 25 ], "verification": "verified", "verified_against": "abstract of 24; full text of 25", "verified_on": "2026-09-30", "note": "Same study; partly answers the DLW-validity concern." }, { "id": "C11", "text": "In DIETFITS, 12-month weight change was -5.3 versus -6.0 kg (difference 0.7 kg, 95% CI -0.2 to 1.6); carbohydrate provided 48% versus 30% of energy at 12 months; diet × INS-30 interaction P = .47 and diet × genotype P = .20.", "carries_number": true, "load_bearing_for": [ "SC5", "SC6" ], "source_ns": [ 11 ], "verification": "verified", "verified_against": "PMC full text", "verified_on": "2026-09-30", "note": "Free-living; modest achieved separation." }, { "id": "C12", "text": "An exploratory DIETFITS reanalysis found achieved glycemic-load reduction tracked weight loss, with effect modification by baseline insulin secretion (P = 0.0009 at 3 months, 0.01 at 6 months, 0.07 at 12 months).", "carries_number": true, "load_bearing_for": [ "SC5" ], "source_ns": [ 35 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "Non-randomized exposure; authors advise caution." }, { "id": "C13", "text": "In 73 obese young adults, 30-minute insulin modified the response to low-glycemic-load versus low-fat diets (P = .02); above the median, weight change was -5.8 versus -1.2 kg at 18 months.", "carries_number": true, "load_bearing_for": [ "SC5" ], "source_ns": [ 34 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "Small; no overall diet difference." }, { "id": "C14", "text": "Two large two-year diet trials found no significant weight or body-composition difference between lower-carbohydrate and lower-fat or higher-carbohydrate approaches; satiety, hunger and satisfaction were similar where measured.", "carries_number": false, "load_bearing_for": [ "SC3", "SC6" ], "source_ns": [ 12, 13 ], "verification": "verified", "verified_against": "PMC full texts", "verified_on": "2026-09-30", "note": "Adherence convergence and attrition." }, { "id": "C15", "text": "A meta-analysis of randomized trials of at least one year found a small, consistent weight-loss advantage for low-carbohydrate over low-fat weight-loss interventions.", "carries_number": false, "load_bearing_for": [ "SC6" ], "source_ns": [ 36 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "Tier B synthesis; pooled estimate reported in the red-team report for context only." }, { "id": "C16", "text": "In an inpatient crossover matched for presented energy, energy density, macronutrients, sugar, sodium and fibre, the ultra-processed diet increased intake by 508 ± 106 kcal/day with 0.9 kg gain versus 0.9 kg loss; glycemic index/load was not matched or reported, and 24-h C-peptide was 38.9 ± 2.8 versus 30.9 ± 2.8 nmol/day (p = 0.052).", "carries_number": true, "load_bearing_for": [ "SC6" ], "source_ns": [ 14 ], "verification": "verified", "verified_against": "PMC full text, Table 1 and Results", "verified_on": "2026-09-30", "note": "Does not exclude a glycemic-load route." }, { "id": "C17", "text": "An independent 8-week free-living crossover found greater weight loss on minimally processed than ultra-processed guideline-concordant diets (difference -1.01%, 95% CI -1.87 to -0.14).", "carries_number": true, "load_bearing_for": [ "SC6" ], "source_ns": [ 44 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "Supports the food-architecture model." }, { "id": "C18", "text": "Reducing pancreatic Ins1 gene dosage on an Ins2-null background prevented high-fat-diet hyperinsulinemia and obesity in mice, with increased white-adipose energy expenditure.", "carries_number": false, "load_bearing_for": [ "SC2", "SC6" ], "source_ns": [ 15 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "Engineered mouse, high-fat diet." }, { "id": "C19", "text": "In bidirectional MR, genetically higher insulin-30 was associated with higher BMI but not the reverse; instruments explained about 0.9% of insulin-30 variance in validation and heterogeneity was driven by a QPCTL/GIPR-region variant.", "carries_number": true, "load_bearing_for": [ "SC2", "SC6" ], "source_ns": [ 16 ], "verification": "verified", "verified_against": "PMC full text", "verified_on": "2026-09-30", "note": "Exposure is stimulated insulin secretion, not diet." }, { "id": "C20", "text": "In 14-day overfeeding, 75-85% of excess carbohydrate energy and 90-95% of excess fat energy was stored.", "carries_number": true, "load_bearing_for": [ "SC2", "SC6" ], "source_ns": [ 17 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "Imposed surplus; cannot test whether GL drives surplus." }, { "id": "C21", "text": "During eight weeks of about 40% overfeeding, body fat rose similarly across 5%, 15% and 25% protein diets while protein altered lean-mass gain and resting expenditure.", "carries_number": true, "load_bearing_for": [ "SC2", "SC6" ], "source_ns": [ 18 ], "verification": "verified", "verified_against": "PMC full text and abstract", "verified_on": "2026-09-30", "note": "Tests energy and protein, not carbohydrate versus fat." }, { "id": "C22", "text": "Both camps accept energy balance and dispute causal direction; the currently defended CIM gives glycemic load central importance and posits that fat storage drives positive balance; neutral and joint-workshop perspectives describe the question as unresolved.", "carries_number": false, "load_bearing_for": [ "SC6" ], "source_ns": [ 19, 20, 38, 39, 40 ], "verification": "verified", "verified_against": "full texts of 19-20; abstracts of 38-40", "verified_on": "2026-09-30", "note": "Tier B framing; does not fix the grade." }, { "id": "C23", "text": "A meta-analysis of 32 isocaloric controlled-feeding studies found expenditure and fat loss modestly greater with lower-fat diets.", "carries_number": false, "load_bearing_for": [ "SC2", "SC4" ], "source_ns": [ 21 ], "verification": "verified", "verified_against": "PMC full text", "verified_on": "2026-09-30", "note": "Tier B synthesis; mostly short studies." }, { "id": "C24", "text": "An updated reanalysis of 29 controlled-feeding studies found that duration modified the effect: lower-carbohydrate diets reduced TEE in shorter trials and increased it in the six trials of about 2.5 weeks or longer.", "carries_number": false, "load_bearing_for": [ "SC4" ], "source_ns": [ 22 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "Tier B synthesis; few long trials." }, { "id": "C25", "text": "In a 3-way isocaloric crossover after 10-15% weight loss, TEE fell by 423 (low-fat), 297 (low-GI) and 97 (very-low-carbohydrate) kcal/day versus pre-weight-loss baseline (trend by glycemic load P < .001).", "carries_number": true, "load_bearing_for": [ "SC4" ], "source_ns": [ 23 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "Four-week periods; DLW." }, { "id": "C26", "text": "In a blinded crossover of overweight or obese men, a high-GI meal controlled for calories, macronutrients and palatability produced lower plasma glucose (4.7 vs 5.3 mmol/L) and greater hunger (P = .04) 4 h later.", "carries_number": true, "load_bearing_for": [ "SC3" ], "source_ns": [ 32 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "n = 12; no intake measure." }, { "id": "C27", "text": "In DIOGENES (773 randomized), weight regain was 0.95 kg less (95% CI 0.33 to 1.57) on low- than high-glycemic-index ad libitum maintenance diets by ITT.", "carries_number": true, "load_bearing_for": [ "SC3", "SC6" ], "source_ns": [ 33 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "Protein also mattered; modest effect." }, { "id": "C28", "text": "Glucokinase mutations shift the glucose–insulin-secretion dose–response right (secretion 61% lower between 5 and 9 mM); KCNJ11 activating mutations impair glucose-coupled secretion, which sulfonylureas restore, more with oral glucose or mixed meals than intravenous glucose.", "carries_number": true, "load_bearing_for": [ "SC1" ], "source_ns": [ 42, 43 ], "verification": "verified", "verified_against": "PubMed abstracts", "verified_on": "2026-09-30", "note": "Independent causal line for SC1." }, { "id": "C29", "text": "Across foods, total carbohydrate correlated only r = 0.39 with the insulin score, and protein-rich foods elicited disproportionately high insulin responses relative to glycemia.", "carries_number": true, "load_bearing_for": [ "SC1" ], "source_ns": [ 41 ], "verification": "verified", "verified_against": "PubMed abstract", "verified_on": "2026-09-30", "note": "Θ blocker for SC1." }, { "id": "C30", "text": "A residential controlled-feeding trial randomized 77 participants (70 completed; 65% of planned completion) to isocaloric very-low-carbohydrate, high-starch or high-sugar diets for 10 weeks after weight loss, with expenditure and body-composition outcomes; primary results were not located in PubMed as of 2026-09-30.", "carries_number": true, "load_bearing_for": [ "SC4" ], "source_ns": [ 37 ], "verification": "verified", "verified_against": "PubMed abstract; PubMed search 2026-09-30", "verified_on": "2026-09-30", "note": "Stability trigger for SC4 only." }, { "id": "C31", "text": "The evidence supports carbohydrate-driven insulin signalling as a contributor in defined contexts but does not establish the pathway as necessary, sufficient or rate-limiting for common human obesity.", "carries_number": false, "load_bearing_for": [ "SC2", "SC3", "SC4", "SC5", "SC6" ], "source_ns": [ 1, 2, 3, 7, 9, 10, 11, 14, 15, 16, 17, 18, 21, 22, 33, 36, 44 ], "verification": "verified", "verified_against": "cross-study synthesis of the opened records listed", "verified_on": "2026-09-30", "note": "Score-fixing synthesis preserving positive bounded evidence and direct negative predictions." } ], "citation_ledger": [ { "n": 1, "citation": "Hall KD, Bemis T, Brychta R, et al. Calorie for Calorie, Dietary Fat Restriction Results in More Body Fat Loss than Carbohydrate Restriction in People with Obesity. Cell Metabolism. 2015;22(3):427-436.", "tier": "A", "kind": "rct", "pmid": "26278052", "doi": "10.1016/j.cmet.2015.07.021", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC4603544/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "HALL-2015-ISOCALORIC-RF-RC", "access_depth": "full_text_complete", "access_note": "Complete PMC article, figures, methods, results, and limitations read.", "carries_claims": [ "C1", "C2", "C31" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 2, "citation": "Hall KD, Chen KY, Guo J, et al. Energy expenditure and body composition changes after an isocaloric ketogenic diet in overweight and obese men. American Journal of Clinical Nutrition. 2016;104(2):324-333.", "tier": "A", "kind": "primary", "pmid": "27385608", "doi": "10.3945/ajcn.116.133561", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC4962163/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "HALL-2016-ISOCALORIC-KD", "access_depth": "full_text_complete", "access_note": "Complete PMC article, supplementary-method context, figures, and discussion read.", "carries_claims": [ "C1", "C3", "C31" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 3, "citation": "Hall KD, Guo J, Courville AB, et al. Effect of a plant-based, low-fat diet versus an animal-based, ketogenic diet on ad libitum energy intake. Nature Medicine. 2021;27(2):344-353.", "tier": "A", "kind": "rct", "pmid": "33479499", "doi": "10.1038/s41591-020-01209-1", "url": "https://pubmed.ncbi.nlm.nih.gov/33479499/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT03878108", "access_depth": "abstract_metadata", "access_note": "Red-team pass: Nature Medicine full text paywalled and not in PMC; abstract and publisher summary read. The original '2.7-fold C-peptide' figure could not be verified and was removed; C-peptide and appetite statements carried from open same-study reports (sources 6, 27).", "carries_claims": [ "C4", "C31" ], "redteam_rechecked_on": "2026-09-30", "access_depth_original": "full_text_complete" }, { "n": 4, "citation": "Soto-Mota A, Jansen LT, Norwitz NG, Pereira MA, Ebbeling CB, Ludwig DS. Physiologic Adaptation to Macronutrient Change Distorts Findings from Short Dietary Trials: Reanalysis of a Metabolic Ward Study. Journal of Nutrition. 2024;154(4):1080-1086.", "tier": "A", "kind": "primary", "pmid": "38128881", "doi": "10.1016/j.tjnut.2023.12.017", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11347797/", "status_flags": [ "corrigendum: PMID 38609206; DOI 10.1016/j.tjnut.2024.04.012; Crossref records it as updating this article, source 5 and source 26; attached to this parent row" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT03878108", "access_depth": "full_text_complete", "access_note": "Complete PMC reanalysis and PubMed corrigendum record read.", "carries_claims": [ "C5" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 5, "citation": "Sciarrillo CM, Guo J, Hengist A, Darcey VL, Hall KD. Flawed reanalysis fails to support the carbohydrate-insulin model of obesity. Journal of Nutrition. 2024;154(3):1058-1060.", "tier": "A", "kind": "primary", "pmid": "38286396", "doi": "10.1016/j.tjnut.2024.01.024", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10997902/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT03878108", "access_depth": "full_text_complete", "access_note": "Complete PMC letter/rebuttal read; it reports no new participants.", "carries_claims": [ "C5" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 6, "citation": "Sciarrillo CM, Guo J, Hengist A, Darcey VL, Hall KD. Diet order significantly affects energy balance for diets varying in macronutrients but not ultraprocessing in crossover studies without a washout period. American Journal of Clinical Nutrition. 2024;120(4):953-963.", "tier": "A", "kind": "primary", "pmid": "39163976", "doi": "10.1016/j.ajcnut.2024.08.013", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11473439/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT03878108", "access_depth": "full_text_complete", "access_note": "Complete PMC secondary analysis read; the paper also analyzes NCT03407053, but this review cites its NCT03878108 order analysis.", "carries_claims": [ "C4", "C5" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 7, "citation": "Liu Y, Mei H, Xue L, et al. Testing the carbohydrate-insulin model: Short-term metabolic responses to consumption of meals with varying glycemic index in healthy adults. Cell Metabolism. 2025;37(3):606-615.e3.", "tier": "A", "kind": "rct", "pmid": "40043690", "doi": "10.1016/j.cmet.2025.01.015", "url": "https://pubmed.ncbi.nlm.nih.gov/40043690/", "status_flags": [ "published erratum: PMID 40306286; DOI 10.1016/j.cmet.2025.04.012; corrects Figure 2F (percentage energy from fat; Excel error) per publisher erratum text; attached, no carried number affected" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT05804942", "access_depth": "abstract_metadata", "access_note": "Red-team pass: full text not retrievable (publisher 403); abstract, PubMed erratum record, publisher erratum text and ClinicalTrials.gov NCT05804942 read.", "carries_claims": [ "C1", "C6", "C31" ], "redteam_rechecked_on": "2026-09-30", "access_depth_original": "full_text_complete" }, { "n": 8, "citation": "Ludwig DS, Majzoub JA, Al-Zahrani A, Dallal GE, Blanco I, Roberts SB. High glycemic index foods, overeating, and obesity. Pediatrics. 1999;103(3):e26.", "tier": "A", "kind": "primary", "pmid": "10049982", "doi": "10.1542/peds.103.3.e26", "url": "https://pubmed.ncbi.nlm.nih.gov/10049982/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "LUDWIG-1999-GI-CROSSOVER", "access_depth": "abstract_metadata", "access_note": "PubMed abstract and metadata read; only abstract-reported numbers are used.", "carries_claims": [ "C1", "C7" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 9, "citation": "Ebbeling CB, Feldman HA, Klein GL, et al. Effects of a low carbohydrate diet on energy expenditure during weight loss maintenance: randomized trial. BMJ. 2018;363:k4583.", "tier": "A", "kind": "rct", "pmid": "30429127", "doi": "10.1136/bmj.k4583", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC6233655/", "status_flags": [ "published erratum: PMID 33144344; PMCID PMC7607868; DOI 10.1136/bmj.m4264; table heading corrected and attached" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT02068885", "access_depth": "full_text_complete", "access_note": "Complete corrected PMC full text, protocol history, tables, and supplementary discussion read.", "carries_claims": [ "C8", "C31" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 10, "citation": "Hall KD, Guo J, Speakman JR. Do low-carbohydrate diets increase energy expenditure? International Journal of Obesity. 2019;43(12):2350-2354.", "tier": "A", "kind": "primary", "pmid": "31548574", "doi": "10.1038/s41366-019-0456-3", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8076039/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT02068885", "access_depth": "full_text_complete", "access_note": "Complete PMC reanalysis read; grouped with the parent trial rather than counted independently.", "carries_claims": [ "C9", "C31" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 11, "citation": "Gardner CD, Trepanowski JF, Del Gobbo LC, et al. Effect of Low-Fat vs Low-Carbohydrate Diet on 12-Month Weight Loss in Overweight Adults and the Association With Genotype Pattern or Insulin Secretion: The DIETFITS Randomized Clinical Trial. JAMA. 2018;319(7):667-679.", "tier": "A", "kind": "rct", "pmid": "29466592", "doi": "10.1001/jama.2018.0245", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC5839290/", "status_flags": [ "erratum: PMID 29614160; DOI 10.1001/jama.2018.2977; units corrected", "erratum: PMID 29710144; DOI 10.1001/jama.2018.4854; funding/support corrected" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT01826591", "access_depth": "full_text_complete", "access_note": "Complete corrected PMC article, figures, protocol-relevant outcomes, and interaction analyses read.", "carries_claims": [ "C11", "C31" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 12, "citation": "Sacks FM, Bray GA, Carey VJ, et al. Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates. New England Journal of Medicine. 2009;360(9):859-873.", "tier": "A", "kind": "rct", "pmid": "19246357", "doi": "10.1056/NEJMoa0804748", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2763382/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT00072995", "access_depth": "full_text_complete", "access_note": "Complete PMC article, tables, weight, adherence, and satiety outcomes read.", "carries_claims": [ "C14" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 13, "citation": "Foster GD, Wyatt HR, Hill JO, et al. Weight and metabolic outcomes after 2 years on a low-carbohydrate versus low-fat diet: a randomized trial. Annals of Internal Medicine. 2010;153(3):147-157.", "tier": "A", "kind": "rct", "pmid": "20679559", "doi": "10.7326/0003-4819-153-3-201008030-00005", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2949959/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT00143936", "access_depth": "full_text_complete", "access_note": "Complete PMC article, longitudinal outcomes, adherence, and attrition read.", "carries_claims": [ "C14" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 14, "citation": "Hall KD, Ayuketah A, Brychta R, et al. Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake. Cell Metabolism. 2019;30(1):67-77.e3.", "tier": "A", "kind": "rct", "pmid": "31105044", "doi": "10.1016/j.cmet.2019.05.008", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7946062/", "status_flags": [ "erratum: PMID 31269427; DOI 10.1016/j.cmet.2019.05.020", "erratum: PMID 33027677; DOI 10.1016/j.cmet.2020.08.014; attached and checked" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT03407053", "access_depth": "full_text_complete", "access_note": "Complete corrected PMC article, supplementary diet design, figures, results, and limitations read.", "carries_claims": [ "C16", "C31" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 15, "citation": "Mehran AE, Templeman NM, Brigidi GS, et al. Hyperinsulinemia drives diet-induced obesity independently of brain insulin production. Cell Metabolism. 2012;16(6):723-737.", "tier": "A", "kind": "primary", "pmid": "23217255", "doi": "10.1016/j.cmet.2012.10.019", "url": "https://pubmed.ncbi.nlm.nih.gov/23217255/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "MEHRAN-INS1-MOUSE-2012", "access_depth": "abstract_metadata", "access_note": "PubMed and publisher abstracts plus bibliographic metadata read; no abstract-external number is used.", "carries_claims": [ "C18", "C31" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 16, "citation": "Astley CM, Todd JN, Salem RM, et al. Genetic Evidence That Carbohydrate-Stimulated Insulin Secretion Leads to Obesity. Clinical Chemistry. 2018;64(1):192-200.", "tier": "A", "kind": "primary", "pmid": "29295838", "doi": "10.1373/clinchem.2017.280727", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC5937525/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "ASTLEY-MR-2018", "access_depth": "full_text_complete", "access_note": "Complete PMC full text, genetic instruments, sensitivity analyses, heterogeneity, and limitations read.", "carries_claims": [ "C19", "C31" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 17, "citation": "Horton TJ, Drougas H, Brachey A, Reed GW, Peters JC, Hill JO. Fat and carbohydrate overfeeding in humans: different effects on energy storage. American Journal of Clinical Nutrition. 1995;62(1):19-29.", "tier": "A", "kind": "primary", "pmid": "7598063", "doi": "10.1093/ajcn/62.1.19", "url": "https://pubmed.ncbi.nlm.nih.gov/7598063/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "HORTON-OVERFEED-1995", "access_depth": "abstract_metadata", "access_note": "PubMed abstract and bibliographic metadata read; only abstract-reported numbers are used.", "carries_claims": [ "C20", "C31" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 18, "citation": "Bray GA, Smith SR, de Jonge L, et al. Effect of dietary protein content on weight gain, energy expenditure, and body composition during overeating: a randomized controlled trial. JAMA. 2012;307(1):47-55.", "tier": "A", "kind": "rct", "pmid": "22215165", "doi": "10.1001/jama.2011.1918", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC3777747/", "status_flags": [ "correction: 'Percentage Error', JAMA 2012;307(10):1028; DOI 10.1001/jama.2012.286 (Crossref verified); attached" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "NCT00565149", "access_depth": "full_text_complete", "access_note": "Complete corrected PMC article, figures, body-composition results, and correction notice read.", "carries_claims": [ "C21", "C31" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 19, "citation": "Ludwig DS, Aronne LJ, Astrup A, et al. The carbohydrate-insulin model: a physiological perspective on the obesity pandemic. American Journal of Clinical Nutrition. 2021;114(6):1873-1885.", "tier": "B", "kind": "review", "pmid": "34515299", "doi": "10.1093/ajcn/nqab270", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8634575/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "PERSPECTIVE-CIM-2021", "access_depth": "full_text_complete", "access_note": "Complete PMC perspective read for model definition and divergence framing only.", "carries_claims": [ "C22" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 20, "citation": "Hall KD, Farooqi IS, Friedman JM, et al. The energy balance model of obesity: beyond calories in, calories out. American Journal of Clinical Nutrition. 2022;115(5):1243-1254.", "tier": "B", "kind": "review", "pmid": "35134825", "doi": "10.1093/ajcn/nqac031", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC9071483/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-28", "status_checked_on": "2026-09-30", "study_id": "PERSPECTIVE-EBM-2022", "access_depth": "full_text_complete", "access_note": "Complete PMC perspective read for competing-model definition and category clarification only.", "carries_claims": [ "C22" ], "redteam_rechecked_on": "2026-09-30" }, { "n": 21, "citation": "Hall KD, Guo J. Obesity Energetics: Body Weight Regulation and the Effects of Diet Composition. Gastroenterology. 2017;152(7):1718-1727.e3.", "tier": "B", "kind": "review", "pmid": "28193517", "doi": "10.1053/j.gastro.2017.01.052", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC5568065/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "HALL-GUO-2017-META", "access_depth": "full_text_partial", "access_note": "PMC full text: meta-analysis methods and direction read.", "carries_claims": [ "C23", "C31" ], "added_in_redteam": true }, { "n": 22, "citation": "Ludwig DS, Dickinson SL, Henschel B, Ebbeling CB, Allison DB. Do Lower-Carbohydrate Diets Increase Total Energy Expenditure? An Updated and Reanalyzed Meta-Analysis of 29 Controlled-Feeding Studies. Journal of Nutrition. 2021;151(3):482-490.", "tier": "B", "kind": "review", "pmid": "33274750", "doi": "10.1093/jn/nxaa350", "url": "https://pubmed.ncbi.nlm.nih.gov/33274750/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "LUDWIG-2021-META", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C24", "C31" ], "added_in_redteam": true }, { "n": 23, "citation": "Ebbeling CB, Swain JF, Feldman HA, et al. Effects of dietary composition on energy expenditure during weight-loss maintenance. JAMA. 2012;307(24):2627-2634.", "tier": "A", "kind": "rct", "pmid": "22735432", "doi": "10.1001/jama.2012.6607", "url": "https://pubmed.ncbi.nlm.nih.gov/22735432/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT00315354", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C25" ], "added_in_redteam": true }, { "n": 24, "citation": "Ebbeling CB, Bielak L, Lakin PR, et al. Energy Requirement Is Higher During Weight-Loss Maintenance in Adults Consuming a Low- Compared with High-Carbohydrate Diet. Journal of Nutrition. 2020;150(8):2009-2015.", "tier": "A", "kind": "primary", "pmid": "32470981", "doi": "10.1093/jn/nxaa150", "url": "https://pubmed.ncbi.nlm.nih.gov/32470981/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT02068885", "access_depth": "abstract_metadata", "access_note": "Abstract read; grouped with source 9.", "carries_claims": [ "C10" ], "added_in_redteam": true }, { "n": 25, "citation": "Ludwig DS, Lakin PR, Wong WW, Ebbeling CB. Scientific discourse in the era of open science: a response to Hall et al. regarding the Carbohydrate-Insulin Model. International Journal of Obesity. 2019;43(12):2355-2360.", "tier": "B", "kind": "commentary", "pmid": "31586125", "doi": "10.1038/s41366-019-0466-1", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC6892734/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT02068885", "access_depth": "full_text_complete", "access_note": "PMC full text read; carries argument, no number used.", "carries_claims": [ "C10" ], "added_in_redteam": true }, { "n": 26, "citation": "Soto-Mota A, Jansen LT, Norwitz NG, Pereira MA, Ebbeling CB, Ludwig DS. Reply to C M Sciarrillo et al. Journal of Nutrition. 2024;154(3):1061-1063.", "tier": "B", "kind": "commentary", "pmid": "38316213", "doi": "10.1016/j.tjnut.2024.01.026", "url": "https://pubmed.ncbi.nlm.nih.gov/38316213/", "status_flags": [ "corrigendum: PMID 38609206 (Crossref lists this item among those updated); recorded on parent source 4" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT03878108", "access_depth": "abstract_metadata", "access_note": "PubMed record read; carries only the existence of the reply.", "carries_claims": [ "C5" ], "added_in_redteam": true }, { "n": 27, "citation": "Hengist A, Sciarrillo CM, Guo J, Walter M, Hall KD. Gut-derived appetite hormones do not explain energy intake differences in humans following low-carbohydrate versus low-fat diets. Obesity (Silver Spring). 2024;32(9):1689-1698.", "tier": "A", "kind": "primary", "pmid": "39113385", "doi": "10.1002/oby.24104", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11357890/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT03878108", "access_depth": "full_text_partial", "access_note": "PMC author manuscript read for the appetite statement.", "carries_claims": [ "C4" ], "added_in_redteam": true }, { "n": 28, "citation": "Ludwig DS, Willett WC, Putt ME. Concern for the validity of short-term dietary crossover trials. Nature Medicine. 2025;31(9):2913-2914.", "tier": "B", "kind": "commentary", "pmid": "40846772", "doi": "10.1038/s41591-025-03909-y", "url": "https://pubmed.ncbi.nlm.nih.gov/40846772/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT03878108", "access_depth": "abstract_metadata", "access_note": "PubMed record read; title-level position only.", "carries_claims": [ "C5" ], "added_in_redteam": true }, { "n": 29, "citation": "Hall KD. Reply to: Concern for the validity of short-term dietary crossover trials. Nature Medicine. 2025;31(9):2915-2916.", "tier": "B", "kind": "commentary", "pmid": "40846773", "doi": "10.1038/s41591-025-03911-4", "url": "https://pubmed.ncbi.nlm.nih.gov/40846773/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT03878108", "access_depth": "abstract_metadata", "access_note": "PubMed record read; title-level position only.", "carries_claims": [ "C5" ], "added_in_redteam": true }, { "n": 30, "citation": "Ludwig DS, Brand-Miller JC, Ebbeling CB, Friedman MI, Norwitz NG, Soto-Mota A. Testing the carbohydrate-insulin model: The data are supportive! Cell Metabolism. 2025;37(6):1239-1240.", "tier": "B", "kind": "commentary", "pmid": "40466618", "doi": "10.1016/j.cmet.2025.04.010", "url": "https://pubmed.ncbi.nlm.nih.gov/40466618/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT05804942", "access_depth": "abstract_metadata", "access_note": "PubMed record read; title-level position only.", "carries_claims": [ "C6" ], "added_in_redteam": true }, { "n": 31, "citation": "Speakman JR, Liu Y. Testing the carbohydrate-insulin model: Some aspects are consistent, but overall the data do not support the model. Cell Metabolism. 2025;37(6):1241-1242.", "tier": "B", "kind": "commentary", "pmid": "40466619", "doi": "10.1016/j.cmet.2025.04.011", "url": "https://pubmed.ncbi.nlm.nih.gov/40466619/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT05804942", "access_depth": "abstract_metadata", "access_note": "PubMed record read; title-level position only.", "carries_claims": [ "C6" ], "added_in_redteam": true }, { "n": 32, "citation": "Lennerz BS, Alsop DC, Holsen LM, et al. Effects of dietary glycemic index on brain regions related to reward and craving in men. American Journal of Clinical Nutrition. 2013;98(3):641-647.", "tier": "A", "kind": "rct", "pmid": "23803881", "doi": "10.3945/ajcn.113.064113", "url": "https://pubmed.ncbi.nlm.nih.gov/23803881/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT01064778", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C26" ], "added_in_redteam": true }, { "n": 33, "citation": "Larsen TM, Dalskov SM, van Baak M, et al.; Diogenes Project. Diets with high or low protein content and glycemic index for weight-loss maintenance. New England Journal of Medicine. 2010;363(22):2102-2113.", "tier": "A", "kind": "rct", "pmid": "21105792", "doi": "10.1056/NEJMoa1007137", "url": "https://pubmed.ncbi.nlm.nih.gov/21105792/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT00390637", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C27", "C31" ], "added_in_redteam": true }, { "n": 34, "citation": "Ebbeling CB, Leidig MM, Feldman HA, Lovesky MM, Ludwig DS. Effects of a low-glycemic load vs low-fat diet in obese young adults: a randomized trial. JAMA. 2007;297(19):2092-2102.", "tier": "A", "kind": "rct", "pmid": "17507345", "doi": "10.1001/jama.297.19.2092", "url": "https://pubmed.ncbi.nlm.nih.gov/17507345/", "status_flags": [ "erratum: JAMA 2007;298(6):627 (listed in PubMed record); attached, content not load-bearing here" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT00130299", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C13" ], "added_in_redteam": true }, { "n": 35, "citation": "Soto-Mota A, Pereira MA, Ebbeling CB, Aronica L, Ludwig DS. Evidence for the carbohydrate-insulin model in a reanalysis of the Diet Intervention Examining The Factors Interacting with Treatment Success (DIETFITS) trial. American Journal of Clinical Nutrition. 2023;117(3):599-606.", "tier": "A", "kind": "primary", "pmid": "36811468", "doi": "10.1016/j.ajcnut.2022.12.014", "url": "https://pubmed.ncbi.nlm.nih.gov/36811468/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT01826591", "access_depth": "abstract_metadata", "access_note": "Abstract read; grouped with source 11.", "carries_claims": [ "C12" ], "added_in_redteam": true }, { "n": 36, "citation": "Tobias DK, Chen M, Manson JE, Ludwig DS, Willett W, Hu FB. Effect of low-fat diet interventions versus other diet interventions on long-term weight change in adults: a systematic review and meta-analysis. Lancet Diabetes & Endocrinology. 2015;3(12):968-979.", "tier": "B", "kind": "review", "pmid": "26527511", "doi": "10.1016/S2213-8587(15)00367-8", "url": "https://pubmed.ncbi.nlm.nih.gov/26527511/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "TOBIAS-2015-META", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C15", "C31" ], "added_in_redteam": true }, { "n": 37, "citation": "Wong JMW, Ludwig DS, Allison DB, et al. Design and conduct of a randomized controlled feeding trial in a residential setting with mitigation for COVID-19. Contemporary Clinical Trials. 2024;140:107490.", "tier": "A", "kind": "primary", "pmid": "38458559", "doi": "10.1016/j.cct.2024.107490", "url": "https://pubmed.ncbi.nlm.nih.gov/38458559/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "RESIDENTIAL-VLC-HC-TRIAL", "access_depth": "abstract_metadata", "access_note": "Abstract read; primary outcome publication searched for in PubMed on 2026-09-30 and not found.", "carries_claims": [ "C30" ], "added_in_redteam": true }, { "n": 38, "citation": "Ludwig DS. Carbohydrate-insulin model: does the conventional view of obesity reverse cause and effect? Philosophical Transactions of the Royal Society B. 2023;378(1888):20220211.", "tier": "B", "kind": "review", "pmid": "37661740", "doi": "10.1098/rstb.2022.0211", "url": "https://pubmed.ncbi.nlm.nih.gov/37661740/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "PERSPECTIVE-CIM-2023", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C22" ], "added_in_redteam": true }, { "n": 39, "citation": "Flier JS. Moderating \"the great debate\": The carbohydrate-insulin vs. the energy balance models of obesity. Cell Metabolism. 2023;35(5):737-741.", "tier": "B", "kind": "review", "pmid": "37086719", "doi": "10.1016/j.cmet.2023.03.020", "url": "https://pubmed.ncbi.nlm.nih.gov/37086719/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "PERSPECTIVE-FLIER-2023", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C22" ], "added_in_redteam": true }, { "n": 40, "citation": "Magkos F, Sørensen TIA, Raubenheimer D, et al. On the pathogenesis of obesity: causal models and missing pieces of the puzzle. Nature Metabolism. 2024;6(10):1856-1865.", "tier": "B", "kind": "review", "pmid": "39164418", "doi": "10.1038/s42255-024-01106-8", "url": "https://pubmed.ncbi.nlm.nih.gov/39164418/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "WORKSHOP-2024", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C22" ], "added_in_redteam": true }, { "n": 41, "citation": "Holt SH, Miller JC, Petocz P. An insulin index of foods: the insulin demand generated by 1000-kJ portions of common foods. American Journal of Clinical Nutrition. 1997;66(5):1264-1276.", "tier": "A", "kind": "primary", "pmid": "9356547", "doi": "10.1093/ajcn/66.5.1264", "url": "https://pubmed.ncbi.nlm.nih.gov/9356547/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "HOLT-1997-INSULIN-INDEX", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C29" ], "added_in_redteam": true }, { "n": 42, "citation": "Byrne MM, Sturis J, Clément K, et al. Insulin secretory abnormalities in subjects with hyperglycemia due to glucokinase mutations. Journal of Clinical Investigation. 1994;93(3):1120-1130.", "tier": "A", "kind": "primary", "pmid": "8132752", "doi": "10.1172/JCI117064", "url": "https://pubmed.ncbi.nlm.nih.gov/8132752/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "BYRNE-1994-GCK", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C28" ], "added_in_redteam": true }, { "n": 43, "citation": "Pearson ER, Flechtner I, Njølstad PR, et al. Switching from insulin to oral sulfonylureas in patients with diabetes due to Kir6.2 mutations. New England Journal of Medicine. 2006;355(5):467-477.", "tier": "A", "kind": "primary", "pmid": "16885550", "doi": "10.1056/NEJMoa061759", "url": "https://pubmed.ncbi.nlm.nih.gov/16885550/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT00334711", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C28" ], "added_in_redteam": true }, { "n": 44, "citation": "Dicken SJ, Jassil FC, Brown A, et al. Ultraprocessed or minimally processed diets following healthy dietary guidelines on weight and cardiometabolic health: a randomized, crossover trial. Nature Medicine. 2025;31(10):3297-3308.", "tier": "A", "kind": "rct", "pmid": "40760353", "doi": "10.1038/s41591-025-03842-0", "url": "https://pubmed.ncbi.nlm.nih.gov/40760353/", "status_flags": [ "none_found" ], "opened": true, "opened_on": "2026-09-30", "status_checked_on": "2026-09-30", "study_id": "NCT05627570", "access_depth": "abstract_metadata", "access_note": "Abstract read.", "carries_claims": [ "C17", "C31" ], "added_in_redteam": true } ], "independence_basis": { "required": true, "applies_to": [ "SC1" ], "lines": [ { "label": "Human dietary perturbation of carbohydrate amount or glycemic index with insulin/C-peptide measurement", "sources": [ 1, 2, 7, 8 ], "excludes_error": "Excludes a purely genetic or developmental artifact: within-person dietary switches in adults with normal β-cell genetics move insulin secretion in the predicted direction within days." }, { "label": "Human monogenic perturbation of β-cell glucose sensing (GCK loss of function; KCNJ11 gain of function with sulfonylurea rescue)", "sources": [ 42, 43 ], "excludes_error": "Excludes food-matrix, protein/fat co-ingestion, and immunoassay artifacts of feeding studies: altering the β-cell glucose sensor itself shifts the glucose-insulin-secretion dose-response, and pharmacological channel closure restores secretion." } ], "shared_failure_mode_checked": true, "note": "No single systematic error could produce both lines. Replication of feeding studies across laboratories raises confidence but is not counted as an additional line. No other subclaim or the composite is at Ζ or above." }, "redteam_synthesis": { "date": "2026-09-30", "red_team_model": "Claude (Anthropic), cross-model red team; authoring model(s) differed", "inputs": [ "carbohydrate-insulin-obesity-hsef-v54-review.md", "carbohydrate-insulin-obesity-hsef-v54-scorecard.json", "carbohydrate-insulin-obesity-hsef-v54-validation.txt", "README-carbohydrate-insulin-obesity-hsef-v54.txt" ], "mode": "single-version red team", "external_redteam_document": null, "findings_total": 25, "findings_accepted": 20, "findings_partially_accepted": 1, "findings_rejected": 4, "findings_by_severity_accepted_or_partial": { "Major": 13, "Minor": 7, "Editorial": 1 }, "score_before": { "level": "Γ", "confidence": "High", "role": "reinforcing / stabilizing factor", "subclaims": { "SC1": "Ε/High", "SC2": "Γ/High", "SC3": "Γ/High", "SC4": "Γ/High", "SC5": "Γ/High", "SC6": "Γ/High" }, "divergence": "Severe overstatement" }, "score_after": { "level": "Γ", "confidence": "High", "role": "reinforcing / stabilizing factor", "subclaims": { "SC1": "Η/High", "SC2": "Γ/Moderate", "SC3": "Γ/Moderate", "SC4": "Γ/Moderate", "SC5": "Γ/Moderate", "SC6": "Γ/High" }, "divergence": "Contested" }, "grade_changed": false, "grade_change_rationale": "Composite unchanged: weakest-link, primary-subclaim and scope-matched aggregation all return Γ because SC6 (primacy) lacks any positive necessity, initiation or rate-limitation evidence. Subclaim SC1 raised Ε → Η after upward threshold testing on two independent causal lines (RT-02); SC2-SC5 confidence lowered to Moderate (RT-12).", "citations_checked": 52, "citations_checked_note": "28 original identifiers (20 references + 8 attached corrections) and 24 added references; all resolved by NCBI esummary; 0 confabulated.", "citations_corrected": 7, "citations_removed": 0, "citations_added": 24, "numbers_removed_unverifiable": [ "'about 2.7-fold' 24-h C-peptide ratio (source 3)", "'130 adults' (source 7; replaced by registry enrollment 135)" ], "rubric_routing_checked": "all 7 rubrics; 06-metabolism-aging retained as primary with declared lifespan/healthspan → adiposity adaptation; rubric gap flagged", "audit_trail": "redteam/REDTEAM_REPORT.md; redteam/SYNTHESIS_CHANGE_LOG.md" }, "consensus_validation": { "completed": true, "status": "documented_historical_discovery_and_reconciliation", "provider": "Consensus", "provider_mode": "Historical browser/connector use as recorded in the source artifacts", "local_audit_date": "2026-10-09", "primary_comparison_count": null, "reported_usage": [], "grade_changed": false, "score_before": { "hsef_level": "Γ", "hsef_descriptor": "Mechanistically plausible, partial system support", "confidence": "High", "confidence_rationale": "Confidence reflects evidence density and reproducibility, not belief that the model is false. The composite is bound by SC6, whose Γ placement rests on a dense, multi-laboratory, multi-design base: two opposing syntheses of 29-32 controlled-feeding studies, several large independent long-term RCTs (DIETFITS, POUNDS LOST, Foster 2010, DIOGENES) and a 53-trial meta-analysis, inpatient feeding trials plus an independent free-living replication, and proponents' own statement that rigorous research is still needed to compare the models. SC2-SC5 are individually Moderate (conflicting, single-group or analysis-dependent evidence). High does not mean glycemic load never matters.", "rests_on_non_peer_reviewed_work": false, "rationale": "The endocrine link (carbohydrate/glycemic load → insulin and insulin-to-glucagon ratio) reaches Η on two independent causal lines. The downstream chain remains Γ: isocaloric fat-balance predictions are negative in short studies and untested at long duration; expenditure gains after adaptation are contested and lack a fat-loss correlate; GI effects on hunger, intake and regain are positive in defined contexts but insulin mediation failed its direct test; the responder rule failed its prespecified replication; and primacy has no necessity or rate-limitation evidence. Weakest-link, primary-subclaim and scope-matched aggregation all return Γ." }, "score_after": { "hsef_level": "Γ", "hsef_descriptor": "Mechanistically plausible, partial system support", "confidence": "High", "confidence_rationale": "Confidence reflects evidence density and reproducibility, not belief that the model is false. The composite is bound by SC6, whose Γ placement rests on a dense, multi-laboratory, multi-design base: two opposing syntheses of 29-32 controlled-feeding studies, several large independent long-term RCTs (DIETFITS, POUNDS LOST, Foster 2010, DIOGENES) and a 53-trial meta-analysis, inpatient feeding trials plus an independent free-living replication, and proponents' own statement that rigorous research is still needed to compare the models. SC2-SC5 are individually Moderate (conflicting, single-group or analysis-dependent evidence). High does not mean glycemic load never matters.", "rests_on_non_peer_reviewed_work": false, "rationale": "The endocrine link (carbohydrate/glycemic load → insulin and insulin-to-glucagon ratio) reaches Η on two independent causal lines. The downstream chain remains Γ: isocaloric fat-balance predictions are negative in short studies and untested at long duration; expenditure gains after adaptation are contested and lack a fat-loss correlate; GI effects on hunger, intake and regain are positive in defined contexts but insulin mediation failed its direct test; the responder rule failed its prespecified replication; and primacy has no necessity or rate-limitation evidence. Weakest-link, primary-subclaim and scope-matched aggregation all return Γ." }, "independent_blinded_audit": false, "raw_receipts": [ { "path": "deliverables\\prepublication-enrichment-2026-10-05\\website-series\\carbohydrate-insulin-obesity\\consensus-browser-links.json", "sha256": "182b0835d896706efe43bed2dd451e1d24ed33aa521a34c1c5bbaf4b467874ed" }, { "path": "deliverables\\prepublication-enrichment-2026-10-05\\website-series\\carbohydrate-insulin-obesity\\consensus-browser.txt", "sha256": "60cfc2b6f0887d3e5e009585159b881ca5193e4a676c87ce62995b760a24698e" }, { "path": "deliverables\\prepublication-enrichment-2026-10-05\\website-series\\carbohydrate-insulin-obesity\\consensus-pro-links.json", "sha256": "dd5465daa0af76f7c20aceb6dcf16e6a9033317c481bc0c92fe46ff8f78b5962" }, { "path": "deliverables\\prepublication-enrichment-2026-10-05\\website-series\\carbohydrate-insulin-obesity\\consensus-pro.txt", "sha256": "31420998ef4ded8cee2a6ca90381e7792ebbb465198901fb5d1a399a3424563d" }, { "path": "deliverables\\prepublication-enrichment-2026-10-05\\website-series\\carbohydrate-insulin-obesity\\consensus-targeted-check.txt", "sha256": "c7aca62dd8de4c2af0e3080b3b9121cfef55b3f11dffa70e041094ab6347c49b" } ], "limitations": "This verifies documented historical use and links the record to the final edition. It adds no primary-source, full-text, independence or verification credit. Original access remains limited to the source-specific record.", "existing_primary_reconciliation": { "slug": "carbohydrate-insulin-obesity", "consensus_used": true, "rawproof": [ "deliverables/prepublication-enrichment-2026-10-05/website-series/carbohydrate-insulin-obesity/consensus-browser.txt", "deliverables/prepublication-enrichment-2026-10-05/website-series/carbohydrate-insulin-obesity/consensus-targeted-check.txt", "deliverables/prepublication-enrichment-2026-10-05/website-series/carbohydrate-insulin-obesity/consensus-browser-links.json" ], "reconciliation_existing": "Yes; 8 primary identifiers matched to timeline; update explicitly says abstract-only from Consensus targeted answer and no independently readable body verified.", "primary_sources_matched": [ "10049982", "26278052", "27385608", "29466592", "30429127", "33479499", "39163976", "40043690" ], "limitations": "Abstract/metadata only; no full-text details attributed as independently verified. Secondary cohort analyses distinguished." } } }