# Scientific audit of equinefeedroom.com

**Reviewed: 27 September 2026**  
**Website version displayed: v1.2, updated 22 August 2026**

## Assessment

**Useful introductory guidance, but important scientific corrections and calculator repairs are needed before this should be relied on for ration formulation.**

The practical emphasis on forage, measured feeding, body-condition monitoring and veterinary involvement is generally sensible. Reliability is less consistent in the advanced physiology, disease-specific recommendations and numerical decision support. In particular, the calculator can give reassuring messages despite important nutritional problems. Breeding and growth calculations are not a faithful implementation of all the NRC equations.

This is an independent evidence appraisal, not a veterinary certification. It assesses the website rather than prescribing a ration for an individual horse.

## Scope and method

I read the public homepage, all **29 educational modules**, **11 mineral panels**, the **14 feed entries**, requirement descriptions and reference tables. I inspected the embedded JavaScript and reproduced selected calculations locally. All navigation links on the retrieved page led to sections of that page; searches did not reveal further indexed content. Both `/robots.txt` and `/sitemap.xml` returned 404 during retrieval. This covers the discoverable public content, not unlinked or private pages.

The site was compared with equine experiments, observational studies, clinical reports and professional consensus documents. The NRC companion program was used as the numerical reference [1]. Research access varied: full text was available for some studies, whereas other judgements use publisher or PubMed abstracts. This was a targeted audit, not a preregistered systematic review or a claim that every relevant publication has been retrieved. Where direct evidence was insufficient, that is stated explicitly.

Calculator checks used functions extracted from the retrieved source with controlled inputs. They were not a full browser, security or usability test. No enquiry was sent to the website.

**Evidence labels:**

- **Supported direction:** the main practical message is consistent with relevant evidence; exact wording may still need qualification.
- **Needs qualification:** reasonable principle, but scope, certainty or implementation is too broad.
- **Correction required:** an identifiable scientific error, misleading inference or numerical problem.
- **Not established:** the stated specificity is not justified by the evidence located; this does not prove the opposite.

## Highest-priority findings

### 1. The ration builder is incomplete and can misclassify feeds

Every custom entry is assigned the forage category. A concentrate entered using its actual analysis therefore counts towards forage intake. Custom feeds also receive an assumed magnesium concentration and a lysine estimate derived from crude protein, even though neither is entered by the user.

The displayed balance covers energy, crude protein, estimated lysine, calcium, phosphorus and magnesium. It does **not** establish adequacy of sodium, chloride, potassium, copper, zinc, selenium, iodine or vitamins. A green result cannot mean that the complete ration is balanced.

**Correction:** allow feed category selection; accept analysed magnesium and lysine; explicitly label estimates and unknowns; state which nutrients have not been assessed.

### 2. Breeding calculations underestimate several NRC reference values

For a 500 kg reference mare in the first month of lactation, the website calculates approximately **65 g lysine, 48 g calcium and 30.8 g phosphorus/day**. The NRC companion equations produce approximately **84.8 g, 59.1 g and 38.3 g**, respectively [1]. These are meaningful differences, not rounding errors.

The lactation lysine discrepancy arises because the site uses the same crude-protein-to-lysine factor as for maintenance. The reference uses a separate milk-production contribution. Later lactation is compressed into one preset despite changing milk yield.

**Correction:** implement and test stage-specific equations, and document the body-weight convention and milk-production assumptions.

### 3. The vitamin D explanation is outdated

Equine vitamin D physiology should not be explained using the usual human sunlight model. An experimental study found no detectable vitamin D3 production in UVB-exposed horse skin, while sheep skin produced it [24]. A separate study of grazing ponies and supplemented Thoroughbreds supports a predominantly dietary source and absent or minimal skin synthesis [25].

**Correction:** explain the importance of dietary vitamin D, including forage-derived D2, and the uncertainties in equine status and requirements. These findings do not justify indiscriminate supplementation.

### 4. Salt safety is stated too absolutely

Free water is essential, but it does not make arbitrary salt doses harmless. An equine feeding experiment found acid–base changes with additional sodium chloride; the higher dose produced mild metabolic acidosis under the study's low-hay conditions [46]. That experiment does not demonstrate that ordinary maintenance supplementation is dangerous, and the clinical importance depends on context.

**Correction:** match supplementation to the whole ration and sweat losses. Avoid statements implying unlimited safety whenever water is available. Salt-block access also does not establish how much an individual horse consumes.

### 5. Hay soaking and sugar targets need tighter interpretation

Soaking can reduce soluble carbohydrate, but the reduction is variable. In Longland and colleagues' experiment, even prolonged soaking left several hays above the proposed low-carbohydrate threshold [9]. Other work found deterioration in microbial hygiene following soaking [10].

WSC and ESC are different laboratory fractions. WSC includes a broader fructan fraction; ESC predominantly measures simple sugars but can include shorter fructans. Their sums with starch are not interchangeable [11].

**Correction:** specify the assay and dry-matter basis, use an appropriate forage analysis, and avoid treating a fixed soaking time or a single percentage as a guarantee of safety. Meal size, total intake and the horse's insulin response also matter [7,8,12].

### 6. Muscle disorders should not share one assumed diet

Evidence supports low-starch, higher-fat approaches for selected RER and PSSM1 cases [16,17]. That does not establish the same treatment for all PSSM2 or myofibrillar myopathy cases. Primary studies distinguish these disorders; early MFM nutritional recommendations were explicitly preliminary [14,15].

**Correction:** separate diagnoses and distinguish demonstrated effects from proposed management. Added fat should reflect energy needs and body condition. Chronic exercise programmes must also be distinguished from management of an acute painful episode.

### 7. Several biochemical explanations overstate certainty

The thiomolybdate explanation imports a ruminant mechanism into horses too confidently. An equine experiment did not demonstrate the expected copper interference or persistent circulating thiomolybdates under its study conditions [30]. This is insufficient to rule out every possible interaction, but it contradicts presenting the mechanism as established routine equine physiology.

The hindgut discussion should also distinguish Gram-positive acid producers from **Gram-negative lipopolysaccharide endotoxin**, and should not present one simplified toxin sequence as a complete account of laminitis [47].

## Calculator audit

### Numerical comparison

Values below are **website / NRC companion program**. Units are Mcal/day for digestible energy and g/day for the other nutrients. Calculations used the downloaded source, before display rounding [1].

| Scenario | Energy | Crude protein | Lysine | Calcium | Phosphorus |
|---|---:|---:|---:|---:|---:|
| Average maintenance, 500 kg | 16.65 / 16.65 | 630 / 630 | 27.1 / 27.1 | 20 / 20 | 14 / 14 |
| Pregnancy, month 9 | 18.48 / 19.25 | 819 / 796.7 | 35.2 / 34.3 | 28 / 36 | 19.6 / 26.3 |
| Pregnancy, month 11 | 19.98 / 21.39 | 894.6 / 892.9 | 38.5 / 38.4 | 32 / 36 | 22.4 / 26.3 |
| Lactation, month 1 | 31.64 / 31.73 | 1512 / 1535 | 65.0 / 84.8 | 48 / 59.1 | 30.8 / 38.3 |
| Lactation, month 4 | 25.81 / 29.44 | 1071 / 1397.5 | 46.1 / 75.7 | 36 / 41.7 | 23.8 / 26.2 |
| Moderate work, 500 kg | 23.31 / 23.31 | 882 / 767.9 | 37.9 / 33.0 | 32 / 35 | 22.4 / 21 |

For pregnancy, the reference uses **500 kg mature/prepregnancy weight**, then estimates pregnancy-associated weight gain. The website uses the entered 500 kg directly and does not make this convention sufficiently explicit. Month 4 is compared with the website's broad late-lactation preset; it is not a claim that one reference value applies to every month in that preset.

Agreement with NRC is a test of implementation, not proof that the NRC estimate is the correct intake for every individual.

### Reproduced defects and limitations

| Finding | What the check demonstrated | Recommended repair |
|---|---|---|
| Custom-feed classification | An 8 kg custom concentrate entry counted as 100% forage. | Require forage/concentrate/supplement classification; distinguish fibre replacers where appropriate. |
| Hidden nutrient assumptions | Custom magnesium is fixed; lysine is a fraction of crude protein. | Accept real values or display “estimated/unknown”; do not award an unqualified adequacy result. |
| Forage minimum is not enforced | For 500 kg, 8 kg of the site's hay plus 0.2 kg oil supplied 7.04 kg forage DM, below the displayed 7.5 kg minimum, without a low-forage warning. | Compare forage DM directly with the chosen minimum. |
| Zero-forage warning gap | An all-oats test generated other warnings but no specific low-forage flag. | Remove the condition that suppresses that flag when forage equals zero. |
| Unlimited ratio approval | A constructed Ca:P ratio of 100:1 received a healthy-ratio message, despite separate nutrient warnings. | Assess absolute amounts and appropriate ratio bounds together. |
| Contradictory energy messages | A test ration at 120% energy was both flagged as overfeeding and described as close to requirement. | Use one consistent tolerance policy. |
| Under-target values can appear adequate | Nutrients at 90–99% of the calculated requirement receive the acceptable classification. | Explain the tolerance and uncertainty; distinguish “near estimate” from “meets requirement.” |
| Meal starch is not calculated | The tool knows daily starch but not meals or their individual composition. | Ask for meal allocation; daily totals cannot certify meal safety. |
| Age does not drive growth | Changing entered age from 0.8 to 1.5 years, holding the yearling category and other inputs constant, left every requirement unchanged. | Calculate with age in months and an appropriate growth/gain model. |
| Growth nutrients use fixed anchors | Protein and bone-mineral values scale with mature weight, rather than the entered current weight and actual gain. | Make model assumptions visible and calculate consistently. |
| Two-year-old work adjustment differs | The website multiplies the entire growth-energy expression by the work factor. The reference applies it to the maintenance component. | Separate maintenance, growth and exercise terms. |
| Pregnancy-stage gap | Intermediate gestation months lack dedicated presets. | Cover the full pregnancy trajectory. |
| Low/high maintenance protein | Energy changes between maintenance types, but protein and lysine remain fixed. | Match the intended model or label the deliberate simplification. |
| Water/salt model | Fixed multipliers substitute for measured environmental and sweat-related needs. | Present as rough starting estimates, never water allowances or maximums. |
| Mineral “upper” colours | Multiples of requirement are used for colour bands, without nutrient-specific toxicity calculations. | Do not portray these bands as tested safe upper limits. |
| Small-feed unit precision | kg/lb conversion rounds amounts to one decimal place. | Retain precision, especially for supplements and small oil quantities. |

**What worked in the checks:** the average-maintenance calculations above; adult work-energy multipliers; conversion of custom dry-matter analyses to as-fed supply; summing known feed nutrients; and conversion of whole-ration carbohydrate to a dry-matter percentage. Those useful components do not remove the limitations above.

## Topic-by-topic evidence assessment

The numbered topics follow the order of the site's educational modules. These are assessments and proposed clarifications, rather than reproductions of the website.

| # | Topic | Assessment | Evidence-based interpretation |
|---:|---|---|---|
| 1 | Digestion and forage | Supported direction | Fibre fermentation and avoiding starch overload are sound principles. Adaptation varies with the starting diet, replacement diet and individual; 1–2 weeks is a practical transition convention, not a universal biological deadline [22,44,47]. |
| 2 | Essential nutrients | Correction required, minor | A standard classification is water, carbohydrate, fat, protein, vitamins and minerals. Energy is a property supplied by nutrients; fibre is predominantly carbohydrate. The site's teaching categories can be useful, but should not be presented as the conventional six classes [1]. |
| 3 | Body condition and weight | Supported direction | Henneke scoring and repeated measurements are useful [2,3]. Targets depend on purpose and health. Stable weight establishes neither micronutrient adequacy nor absence of insulin dysregulation [7]. |
| 4 | Water | Needs qualification | Continuous access is sound. Intake varies with forage moisture, weather, work and lactation. Cold-weather research supports offering warmer water, but not a universal precise temperature optimum for all horses [4]. The displayed litres must never be interpreted as a limit. |
| 5 | Fibre, starch and fat | Supported direction | Substituting digestible fibre or fat for some cereal can be useful. Fat adaptation changes substrate use during low-intensity exercise [43]; it is not a universal performance or temperament guarantee. Starch guidance must be matched to disease risk [5,16]. |
| 6 | Protein and amino acids | Needs qualification | Protein quality matters and lysine is commonly limiting. There is no universally demonstrated ranking in which threonine and methionine always follow in fixed order; response depends on the diet [19,20]. More protein is not automatically better. |
| 7 | Mineral balance | Needs qualification | Calcium/phosphorus imbalance can cause disease [40]. Both absolute adequacy and balance matter: an attractive ratio can exist when both nutrients are deficient. Likewise, a Cu:Zn ratio cannot certify adequate amounts. |
| 8 | Vitamin supplementation | Mixed; vitamin D correction required | Assess supply and context. The vitamin D skin-production explanation needs updating [24,25]. Vitamin E deficiency has relevant equine disease evidence [26]. Broad B- or C-vitamin supplementation rules for all working, sick or old horses were not established by this search. |
| 9 | Electrolytes | Needs qualification | Sweat replacement is appropriate; water and total sodium/chloride intake matter. Free-choice salt intake is variable. Product value should be assessed by the delivered electrolytes, dose and intended use, rather than dismissing glucose-containing formulations solely by ingredient order [46,51]. |
| 10 | Feeding routine | Supported direction | Weighing feeds and reducing long forage gaps are useful. Meal limits should incorporate starch content and horse weight, not just kg of concentrate. An observational study linked longer forage gaps and larger starch meals with ulcer risk [5]. Transition guidance is sensible but not an exact microbial timetable [44]. |
| 11 | Feeding myths | Mostly supported direction | Avoiding unnecessary concentrates and overlapping supplements is sensible. Bran can be part of a properly formulated ration; disease risk concerns the whole ration and sustained imbalance, not its mere presence [40]. Claims ranking the “most common” error or the cheapest intervention lack supporting comparative data here. |
| 12 | Forage analysis | Needs qualification | Analysis is central. Keep as-fed and dry-matter units separate; identify the carbohydrate assay [11]. DE may be predicted from laboratory composition. Generic fibre indices guide interpretation but are not perfect predictors of intake or digestibility across all forages. |
| 13 | NSC and laminitis | Supported mechanism; tighter practical wording needed | Experimental hyperinsulinaemia can induce laminitis [8]. Carbohydrate-overload and endocrine pathways must remain distinct [47]. Soaking and time of grazing do not guarantee safety [9,10,12]. WSC plus starch and ESC plus starch must be identified separately [11]. |
| 14 | Broodmares | Correction required in numerical guidance | Requirements vary across pregnancy and lactation; see the worked model comparison [1]. Copper supply matters, but prenatal supplementation does not guarantee lifelong joint soundness: equine studies have differing results [28,29]. |
| 15 | Youngstock | Supported direction; model needs repair | Appropriate growth and adequate amino acids/minerals matter. DOD is multifactorial, and no single growth curve or supplement guarantees prevention [20,28,29]. Use actual age, weight, expected mature size and an explicit gain assumption. |
| 16 | Performance horses | Needs qualification | Fuel choice depends on intensity and recovery. Fat adaptation can alter fuel oxidation [43]. RER trial results cannot be generalised to say every horse has substantially less tying-up risk on a particular ration [16]. |
| 17 | Seniors | Needs qualification | Dentition, intake, weight trend and endocrine status should drive changes. A controlled comparison did not find lower nutrient digestibility simply because healthy horses were older [18]. Extra amino acids also did not universally improve protein synthesis in an aged-horse trial [19]. |
| 18 | Weight gain/loss | Supported direction | Investigating unexplained weight loss and avoiding abrupt restriction are appropriate. A 1.5% forage-DM rule is a conservative owner-facing starting point, not an absolute clinical floor: supervised obesity studies and consensus protocols sometimes use less [7,50]. Emaciation needs a distinct refeeding plan [42]. |
| 19 | EMS/PPID/laminitis | Needs qualification | PPID, obesity and insulin dysregulation overlap but are not interchangeable. Lean animals can need carbohydrate control without energy restriction. Do not assume every PPID horse needs weight loss; insulin assessment matters, and pergolide does not reliably correct ID [7,13]. |
| 20 | Gastric ulcers | Supported distinction; some claims too certain | Separating squamous and glandular disease is important [52]. Low-starch feeding and avoidance of long fasting have support [5]. Alfalfa has experimental evidence, but benefits depend on the diet and disease [6]. A few handfuls before work are not a validated universal protective dose; turnout alone is not a guaranteed treatment. |
| 21 | RER/PSSM/MFM | Correction required in grouping/certainty | Keep diagnosis-specific advice. Trials support selected RER/PSSM1 approaches [16,17], while PSSM2/MFM evidence and disease definitions differ [14,15]. Do not recommend routine selenium increases without assessing intake/status [32,33]. |
| 22 | Hindgut health/probiotics | Needs qualification | Avoiding excessive fermentable substrate is sound. Live-yeast effects depend on strain, dose, diet and outcome; some experiments show benefits, others do not demonstrate improved digestibility or clear pH changes [22,23]. Nonspecific behaviour or droppings do not diagnose acidosis. |
| 23 | Digestive biochemistry | Correction required | Fructans are not handled like starch by mammalian digestive enzymes. Distinguish lactate-producing organisms, endotoxin origin and other inflammatory pathways [47]. VFA proportions are diet- and sampling-dependent, not fixed constants [22]. |
| 24 | Energy metabolism | Supported framework; qualify outcomes | DE/ME/NE distinctions and different exercise fuels are useful. Demonstrating altered substrate oxidation after fat adaptation is not equivalent to proving improved competition performance in every discipline [43]. Avoid deterministic “cool energy” explanations for behaviour. |
| 25 | Nitrogen metabolism | Supported direction | Amino acid supply and disposal of excess nitrogen matter [19–21]. High protein increased water/urinary outputs in one forage study, but another eventing-horse trial found no change in water intake; use “can increase,” not “always increases” [21]. |
| 26 | Mineral homeostasis | Mixed; antagonism correction required | Blood calcium alone cannot certify adequate nutrition [40]. The molybdenum/sulphur mechanism needs equine-specific qualification [30]. Quantitative antagonisms should not be asserted solely from mechanisms established in ruminants. |
| 27 | Antioxidants | Mostly supported direction; update vitamin D | Vitamin E and selenium have distinct, overlapping biological roles, but one cannot simply substitute for the other [26,32]. Biotin has controlled evidence for selected poor-quality hooves over long periods [27]. Avoid converting these findings into blanket supplement advice. |
| 28 | Blood tests | Supported direction; clinical interpretation required | CK/AST, selenium measures and endocrine tests can be useful, but sample timing, assay, exercise and other disease affect interpretation [13,32,41]. CK/AST cannot alone establish the cause of muscle disease. Normal routine minerals do not validate a ration [40]. |
| 29 | Endocrine and therapeutic feeding | Supported principles; specialist detail needed | Insulin-mediated laminitis is experimentally supported [8]. Do not reduce all ID to a single insulin-resistance mechanism [7]. PPID management needs separate endocrine assessment [13]. Refeeding schedules are patient-dependent; the early days require monitoring, not a fixed guaranteed-safe timetable [42]. |

## Mineral-panel assessment

The stage-by-stage lists sometimes imply stronger diagnostic specificity than the research supports. Coat changes, poor growth, weakness, pica or nervousness are not nutrient-specific diagnoses.

| Mineral | Assessment and necessary qualification | Evidence |
|---|---|---|
| Copper | Adequacy is important, especially during development. Prenatal copper and joint outcomes are more complex than a direct guarantee. Do not diagnose copper deficiency from coat colour or use a ratio alone. | Supplementation studies and a study finding no relationship between liver copper and later OC [28,29]. |
| Zinc | Extreme environmental exposure can cause serious skeletal disease, including with copper disturbance. These exposures do not establish that ordinary deviations from a preferred ratio cause the same outcome. | Equine exposure reports [31]. |
| Selenium | Deficiency and toxicity warnings are justified. Differentiate nutritional need, supplemental intake, chronic toxicity and acute overdose; route matters. “Blind staggers” is problematic historical terminology and should not label acute poisoning as though it were a precise equine diagnosis. | Status experiment, acute equine poisoning investigation and historical critique [32,33,49]. |
| Iron | Routine supplementation needs justification. Neonatal toxicity is documented. The iron–insulin relationship remains uncertain: observational association does not establish causation, and a trainer survey cannot definitively rule it out. Avoid extrapolating a universal iron:Cu:Zn rule. | Neonatal study, hyperinsulinaemic-horse study and survey [34–36]. |
| Iodine | Maternal excess and foal goitre are well documented; seaweed/kelp can deliver excessive iodine. Both dose and complete intake matter. A clinical sign alone does not distinguish excess from deficiency. | Equine cases and maternal-exposure research [37]. |
| Manganese | Biological roles are plausible, but strong equine evidence for the panel's specific age-linked clinical syndromes was not located. Those entries should be marked extrapolated or uncertain rather than presented as established diagnostic patterns. | Available equine supplementation research measured blood concentrations, not proof of the listed syndromes [48]. |
| Calcium | Adequacy and phosphorus/oxalate context matter. A normal serum result does not exclude a chronic dietary problem. High ratios are not automatically acceptable merely because calcium exceeds phosphorus. | Clinical equine nutritional hyperparathyroidism evidence [40]. |
| Phosphorus | Prolonged imbalance can cause skeletal disease. Evaluate total amounts and ratio across the ration. An occasional bran meal is not equivalent to the extreme or prolonged diets in clinical reports. | Clinical equine reports [40]. |
| Magnesium | Genuine deficiency exists, but nervousness or muscle signs alone are insufficient evidence. Studies on diagnosed headshaking do not validate magnesium as a general-purpose calmer or establish every claimed adult deficiency pattern. | Foal deficiency experiment and controlled headshaking study [38]. |
| Sodium/chloride | Supplementation is often useful, particularly with sweat losses. Free water is necessary but not proof that any dose is harmless; measure supply and actual intake. | Equine supplementation and hydration experiments [46,51]. |
| Potassium | HYPP dietary sensitivity is supported. Describing HYPP as the only context in which potassium can be problematic is too absolute; renal/excretory and systemic clinical disorders also matter. | HYPP experiment and kidney-disease consensus [39,45]. |

## Feed tables and reference values

The listed feeds are **illustrative compositions**, not verified analyses of a product or forage batch. The site provides no linked analytical dataset that would allow every value to be independently traced. Therefore, the 14 entries cannot be “verified” as the actual nutrient content of feeds someone buys.

Specific improvements:

- Identify the provenance, analytical basis and variability of each value.
- Distinguish dry beet-pulp weight from weight after soaking. Added water changes as-fed concentration without adding nutrients.
- Make pasture intake an explicit estimate. A fixed moisture content is particularly uncertain for grazing.
- Show that a small as-fed carbohydrate percentage in a wet feed can represent a much larger dry-matter percentage.
- Use product-specific data for compound feeds and balancers. Generic protein/mineral values cannot certify a particular commercial product.
- Track micronutrients and supplement amounts before claiming full balance. A generic “balancer” name is not a nutrient guarantee.

The maintenance mineral targets broadly resemble standard reference values, but they should be accompanied by body-weight/intake assumptions and units [1]. They are not a substitute for a full-ration analysis or nutrient-specific upper-limit assessment.

## Editorial changes needed for credibility

1. Add a named author, qualifications, veterinary/nutrition review and a linked reference list. Neither a named scientific reviewer nor a claim-linked bibliography was visible in the retrieved material.
2. Replace categorical wording with statements that identify population, dose, outcome and uncertainty.
3. Separate clinical emergencies from routine management, and clearly distinguish an acute episode from a long-term prevention programme.
4. Validate calculator outputs against documented cases and publish test cases, including breeding, growth, no forage, custom concentrates and extreme mineral ratios.
5. Display missing-data warnings. Unknown is not zero and estimated is not measured.
6. Clarify that body-condition trends guide energy adjustment but cannot establish vitamin/mineral adequacy.

**Use recommendation:** retain the general education after corrections. Treat calculator results as provisional illustrations until the issues above are repaired and independently reviewed. For youngstock, broodmares, muscle disorders, ulcers or endocrine disease, formulate the ration with appropriate equine veterinary/nutrition input and actual feed data.

## Evidence register

Links below point to the original paper, its PubMed record, an author/institutional copy, or the issuing scientific body. Most are original equine studies; consensus documents are identified. A paper supporting part of a topic does not validate every sentence in that topic. Bibliographic titles are supplied for traceability, not as substitutes for appraisal.

**[1] NRC requirements and official model.** National Research Council (2007), *Nutrient Requirements of Horses*, sixth revised edition. [Publication and DOI](https://doi.org/10.17226/11653); [official companion calculator](https://webassets.nationalacademies.org/nrh/). Source-code calculations inspected. A requirement framework, not an individual clinical validation.

**[2] Body condition scoring.** Henneke et al. (1983), *Relationship between condition score, physical measurements and body fat percentage in mares*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/6641685/).

**[3] Weight estimation.** Carroll & Huntington (1988), *Body condition scoring and weight estimation of horses*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/3366105/). Predictive relationship, not a substitute for a weighbridge in every breed/age group.

**[4] Water temperature.** Kristula & McDonnell (1994), *Drinking water temperature affects consumption of water during cold weather in ponies*. [Publisher](https://www.sciencedirect.com/science/article/pii/0168159194900205); [university paper](https://www.vet.upenn.edu/docs/default-source/research/equine-behavior-laboratory/94drinku.pdf?sfvrsn=325ee0ba_0).

**[5] Ulcer risk factors.** Luthersson et al. (2009), *Risk factors associated with equine gastric ulceration syndrome (EGUS) in 201 horses in Denmark*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/19927579/). Observational evidence; association is not a guaranteed threshold for every horse.

**[6] Alfalfa and gastric conditions.** Nadeau et al. (2000), *Evaluation of diet as a cause of gastric ulcers in horses*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/10895901/). Experimental comparison of specific diets.

**[7] EMS consensus.** Durham et al. (2019), *ECEIM consensus statement on equine metabolic syndrome*. [Full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC6430910/). Clinical recommendations, including individual monitoring and uncertainty.

**[8] Insulin-induced laminitis.** Asplin et al. (2007), *Induction of laminitis by prolonged hyperinsulinaemia in clinically normal ponies*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/17719811/). Experimental causation under controlled exposure; not a direct test of every feeding scenario.

**[9] Hay-soaking variability.** Longland, Barfoot & Harris (2011), *Effects of soaking on the water-soluble carbohydrate and crude protein content of hay*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/21652661/). Laboratory forage study; reductions varied between hays.

**[10] Soaking and hygiene.** Moore-Colyer et al. (2014), *The effect of five different wetting treatments on the nutrient content and microbial concentration in hay for horses*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/25426729/).

**[11] Carbohydrate assays.** *Chromatographic profiles of nonstructural carbohydrates contributing to the colorimetrically determined fructan, ethanol-soluble, and water-soluble carbohydrate contents of five grasses*. [Original analytical study](https://www.sciencedirect.com/science/article/abs/pii/S0377840113003039). Also [original USDA analytical research](https://www.ars.usda.gov/research/publications/publication/?seqNo115=260646).

**[12] Grazing and insulin.** *Insulin and incretin responses to grazing in insulin-dysregulated and healthy ponies* (2019). [PubMed](https://pubmed.ncbi.nlm.nih.gov/30506731/). Also *Effects of grazing system, season, and forage carbohydrates on glucose and insulin dynamics of the grazing horse* (2019): [PubMed](https://pubmed.ncbi.nlm.nih.gov/30911753/).

**[13] PPID clinical guideline.** *BEVA primary care clinical guidelines: Diagnosis and management of equine pituitary pars intermedia dysfunction* (2023). [PubMed](https://pubmed.ncbi.nlm.nih.gov/37795557/). Evidence-based clinical guideline, not an individual feeding trial.

**[14] MFM.** *Clinical and histopathological features of myofibrillar myopathy in Warmblood horses* (2017). [Full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC5640499/). Management suggestions in this study were preliminary, not a controlled dietary efficacy trial.

**[15] PSSM2 disease distinction.** *Absence of myofibrillar myopathy in Quarter Horses with a histopathological diagnosis of type 2 polysaccharide storage myopathy and lack of association with commercial genetic tests*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/35288976/). Also *Type 2 polysaccharide storage myopathy in Quarter Horses is a novel glycogen storage disease causing exertional rhabdomyolysis*: [PubMed](https://pubmed.ncbi.nlm.nih.gov/36102343/).

**[16] RER feeding experiment.** McKenzie et al. (2003), *Effect of dietary starch, fat, and bicarbonate content on exercise responses and serum creatine kinase activity in equine recurrent exertional rhabdomyolysis*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/14529137/). Small crossover study of affected horses.

**[17] PSSM1 feeding experiment.** *Effect of dietary fats with odd or even numbers of carbon atoms on metabolic response and muscle damage with exercise in Quarter Horse-type horses with type 1 polysaccharide storage myopathy* (2010). [PubMed](https://pubmed.ncbi.nlm.nih.gov/20187835/).

**[18] Age and digestion.** *Comparison of Nutrient Digestibility Between Adult and Aged Horses* (2014). [Publisher](https://www.sciencedirect.com/science/article/abs/pii/S0737080614002664). Findings in healthy horses do not exclude impaired intake/digestion in diseased seniors.

**[19] Amino acid limitations.** *Effects of threonine supplementation on whole-body protein synthesis and plasma metabolites in growing and mature horses*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/26670331/). Also *Effects of dietary amino acid supplementation on measures of whole-body and muscle protein metabolism in aged horses* (2018): [PubMed](https://pubmed.ncbi.nlm.nih.gov/30284316/).

**[20] Protein quality and growth.** *Growth of thoroughbreds fed a low-protein supplement fortified with lysine and threonine* (2001). [PubMed](https://pubmed.ncbi.nlm.nih.gov/11518223/).

**[21] Protein and excretion.** *Effects on exercise response, fluid and acid-base balance of protein intake from forage-only diets in standardbred horses*: [PubMed](https://pubmed.ncbi.nlm.nih.gov/17402499/). Contrasting water-intake result in *The impact of dietary protein levels on nutrient digestibility and water and nitrogen balances in eventing horses*: [PubMed](https://pubmed.ncbi.nlm.nih.gov/25568371/).

**[22] Yeast and fermentation.** Medina et al. (2002), *Effect of a preparation of Saccharomyces cerevisiae on microbial profiles and fermentation patterns in the large intestine of horses fed a high fiber or a high starch diet*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/12413082/). Small experimental study; product/diet-specific results.

**[23] More recent yeast trial.** Carter et al. (2025), *Influence of Saccharomyces cerevisiae CNCM I-1077 on the fecal pH, markers of gut permeability, fecal microbiota, and markers of systemic inflammation in sedentary horses fed a high-starch diet*. [Full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC12010697/). Thirty horses, 32 days; microbial differences did not translate into improved apparent digestibility.

**[24] Equine skin and vitamin D.** Azarpeykan et al. (2022), *Undetectable vitamin D3 in equine skin irradiated with ultraviolet light*. [Institutional paper record](https://mro.massey.ac.nz/items/31682157-ed72-46f1-9f10-71b9888ce1af); DOI 10.1294/jes.33.45. Ex vivo skin experiment, supported by the independent status study below.

**[25] Vitamin D status.** Dosi et al. (2023), *The effect of season, management and endocrinopathies on vitamin D status in horses*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/36054781/); [publisher](https://beva.onlinelibrary.wiley.com/doi/10.1111/evj.13873). Observational cohort; does not define an optimal supplementation dose.

**[26] Vitamin E and neuromuscular disease.** *Vitamin E deficiency and risk of equine motor neuron disease* (2007). [Full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC1950518/).

**[27] Biotin.** Josseck et al. (1995), *Hoof horn abnormalities in Lipizzaner horses and the effect of dietary biotin on macroscopic aspects of hoof horn quality*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/7556044/). Includes a controlled double-blind trial; selected horses and long follow-up.

**[28] Maternal copper and development.** Pearce et al. (1998), *Effect of copper supplementation on the evidence of developmental orthopaedic disease in pasture-fed New Zealand Thoroughbreds*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/9622322/). Also the associated copper-status study: [PubMed](https://pubmed.ncbi.nlm.nih.gov/9622320/).

**[29] Contrasting copper evidence.** *Influence of liver copper status of mare and newborn foal on the development of osteochondrotic lesions* (2003). [PubMed](https://pubmed.ncbi.nlm.nih.gov/12553465/).

**[30] Molybdenum interaction.** Strickland et al. (1987), *Dietary molybdenum as a putative copper antagonist in the horse*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/3691460/). Small experimental study; no demonstrated ruminant-like effect under its conditions.

**[31] Zinc exposure.** *Skeletal abnormalities in young horses associated with zinc toxicity and hypocuprosis* (1984). [PubMed](https://pubmed.ncbi.nlm.nih.gov/6497803/). Also *The effects of natural exposure to high levels of zinc and cadmium in the immature pony as a function of age*: [PubMed](https://pubmed.ncbi.nlm.nih.gov/3732202/).

**[32] Selenium status.** *Effect of selenium source and dose on selenium status of mature horses* (2008). [PubMed](https://pubmed.ncbi.nlm.nih.gov/18791154/).

**[33] Selenium poisoning.** Desta et al. (2011), *Acute selenium toxicosis in polo ponies*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/21908304/). Acute compounded injectable overdose; should not be used to infer an oral dietary toxicity threshold.

**[34] Neonatal iron toxicity.** *Iron toxicity in neonatal foals* (1988). [PubMed](https://pubmed.ncbi.nlm.nih.gov/3371313/).

**[35] Iron–insulin association.** *Possible dysmetabolic hyperferritinemia in hyperinsulinemic horses* (2020). [Full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC6971364/). Observational evidence, not causal proof.

**[36] Iron–insulin counterpoint.** *Dietary Iron Unlikely to Cause Insulin Resistance in Horses* (2022). [PubMed](https://pubmed.ncbi.nlm.nih.gov/36230253/). Trainer survey; important limitations preclude treating it as definitive exclusion of a relationship.

**[37] Iodine.** *Equine goiter associated with excess dietary iodine* (2004): [PubMed](https://pubmed.ncbi.nlm.nih.gov/15171493/). *Nodular goiters of three mares and their foals, induced by feeding excessive amount of seaweed*: [PubMed](https://pubmed.ncbi.nlm.nih.gov/723109/).

**[38] Magnesium.** *Influence of magnesium deficiency on horse foal tissue concentration of Mg, calcium and phosphorus*: [PubMed](https://pubmed.ncbi.nlm.nih.gov/1148153/). *Effects of magnesium with or without boron on headshaking behavior in horses with trigeminal-mediated headshaking* (2019): [PubMed](https://pubmed.ncbi.nlm.nih.gov/30990929/). Different populations and endpoints.

**[39] HYPP.** *Hyperkalemic periodic paralysis in horses* (1990). [PubMed](https://pubmed.ncbi.nlm.nih.gov/2243032/). Includes potassium-provoked episodes and management observations.

**[40] Calcium/phosphorus disease.** *Nutritional secondary hyperparathyroidism in two ponies* (2018): [Full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC5943687/). *Clinical and biochemistry findings, and parathyroid hormone concentrations in three horses with secondary hyperparathyroidism*: [PubMed](https://pubmed.ncbi.nlm.nih.gov/1404225/).

**[41] Muscle biomarkers.** *Kinetic evaluation of muscle damage during exercise by calculation of amount of creatine kinase released*: [PubMed](https://pubmed.ncbi.nlm.nih.gov/8141400/). *Pre-race and race management impacts serum muscle enzyme activity in Australian endurance horses*: [PubMed](https://pubmed.ncbi.nlm.nih.gov/34601756/).

**[42] Refeeding.** Witham & Stull (1998), *Metabolic responses of chronically starved horses to refeeding with three isoenergetic diets*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/9524642/).

**[43] Fat adaptation.** *Effects of fat adaptation on glucose kinetics and substrate oxidation during low-intensity exercise* (2002). [PubMed](https://pubmed.ncbi.nlm.nih.gov/12405656/). Metabolic endpoints are not identical to competition outcomes.

**[44] Dietary adaptation.** *Abrupt dietary changes between grass and hay alter faecal microbiota of ponies* (2020): [PubMed](https://pubmed.ncbi.nlm.nih.gov/32810164/). *Resilience of Faecal Microbiota in Stabled Thoroughbred Horses Following Abrupt Dietary Transition between Freshly Cut Pasture and Three Forage-Based Diets* (2021): [PubMed](https://pubmed.ncbi.nlm.nih.gov/34573577/).

**[45] Kidney disease.** *ECEIM consensus statement on equine kidney disease* (2024). [Paper](https://pmc.ncbi.nlm.nih.gov/articles/PMC11256176/). Relevant to exceptions to general mineral/excretion rules.

**[46] Salt supplementation.** Zeyner et al. (2017), *Effects of Different Oral Doses of Sodium Chloride on the Basal Acid-Base and Mineral Status of Exercising Horses Fed Low Amounts of Hay*. [PubMed](https://pubmed.ncbi.nlm.nih.gov/28045916/); [paper](https://pmc.ncbi.nlm.nih.gov/articles/PMC5207637/). Results are specific to the doses and diet studied.

**[47] Hindgut overload.** *Plasma endotoxin levels in horses subjected to carbohydrate induced laminitis* (1987): [PubMed](https://pubmed.ncbi.nlm.nih.gov/3691456/). *Intestinal bacterial overgrowth includes potential pathogens in the carbohydrate overload models of equine acute laminitis* (2012): [PubMed](https://pubmed.ncbi.nlm.nih.gov/22633481/).

**[48] Manganese.** *Effects of oral manganese supplementation with different compounds on the manganese whole blood and serum concentrations of broodmares and their suckling foals* (2021). [PubMed](https://pubmed.ncbi.nlm.nih.gov/34861732/). Does not establish the website's entire deficiency/excess symptom table.

**[49] Historical selenium terminology.** *Selenium-induced “blind staggers” and related myths. A commentary on the extent of historical livestock losses attributed to selenosis on western US rangelands* (1996). [PubMed](https://pubmed.ncbi.nlm.nih.gov/8826017/). Historical appraisal, not a new equine feeding trial.

**[50] Restricted-soaked-hay study.** *Considerations for the use of restricted, soaked grass hay diets to promote weight loss in the management of equine metabolic syndrome and obesity* (2015). [PubMed](https://pubmed.ncbi.nlm.nih.gov/26403956/). Monitored research conditions do not justify unsupervised severe restriction.

**[51] Hydration experiment.** *Drinking salt water enhances rehydration in horses dehydrated by frusemide administration and endurance exercise* (2002). [PubMed](https://pubmed.ncbi.nlm.nih.gov/12405743/).

**[52] Gastric-ulcer consensus.** Sykes et al. (2015), *European College of Equine Internal Medicine Consensus Statement—Equine Gastric Ulcer Syndrome in Adult Horses*. [Paper](https://pmc.ncbi.nlm.nih.gov/articles/PMC4858038/). Used alongside direct feeding and risk-factor studies; diet evidence is stronger for ESGD than for EGGD.

## Remaining uncertainties

- No actual forage, commercial feed or supplement sample was analysed; generic table values remain illustrative.
- Not every clinical sign in the age-specific tables has a direct equine validation study. Manganese, some magnesium claims and blanket B/C-vitamin recommendations particularly need better sourcing.
- The full texts of all cited papers were not available; abstract-level support is narrower than a full methodological appraisal.
- The audit does not establish rates of real-world harm or an overall percentage accuracy score for the site.
- The numerical reproduction used the retrieved version. A later website update may change these findings.