limitations
Documents caveats, constraints, and boundaries of a claim, comparison, or body of evidence. Critical for maintaining intellectual honesty and helping readers understand what conclusions can and cannot be drawn.
Why This Property Exists
Evidence-based health content requires intellectual honesty. Claims should be accompanied by their caveats, and comparisons should acknowledge their boundaries. This is particularly important when:
- Evidence is emerging: New research areas where studies are limited or methodology is developing
- Compositional claims don't guarantee clinical outcomes: "Similar fatty acid profile" doesn't prove "works equally well"
- Mechanistic reasoning exceeds empirical validation: The theory is sound but trials are lacking
- Population generalisability is uncertain: Studies in one group may not apply to others
Schema.org has no standard property for expressing what a claim does NOT establish. The limitations property fills this gap, enabling:
- Transparent communication: Readers understand the strength and boundaries of claims
- Scientific integrity: Authors demonstrate awareness of evidence gaps
- Appropriate reliance: Decision-makers know what conclusions are and aren't supported
- Future research direction: Gaps are explicitly documented for addressing
This property embodies a core principle: being honest about what we don't know is as important as stating what we do.
Domain and Range
Used On (Domain)
| Type | Context |
|---|---|
| hs:CompositionComparison | Caveats about why compositional similarity may not predict clinical equivalence |
| hs:CausalPathway | Boundaries of the mechanistic model; conditions where it may not apply |
| hs:PhysiologicalRequirement | Constraints on the evidence for stated requirements |
Expected Values (Range)
| Type | Notes |
|---|---|
| schema:Text | Free text describing limitations. Should be specific about what isn't established, not vague disclaimers. |
Distinction From Related Properties
Several properties relate to evidence quality and constraints:
| Property | Purpose | Example |
|---|---|---|
| limitations | What this claim/evidence does NOT establish | "Compositional similarity does not prove clinical equivalence; no RCTs exist comparing these substances" |
| evidenceLevel | Quality grade of supporting evidence | "Moderate" or "Limited" |
| evidenceSource | Links to actual supporting research | ScholarlyArticle with DOI |
| contraindications | When something should NOT be used | "Not suitable for patients with wool allergy" |
The Key Distinction
limitations— "What we can't conclude from this"evidenceLevel— "How strong the evidence is"contraindications— "When not to use this"
These are complementary. A claim might have strong evidence (evidenceLevel: "Strong") but significant limitations on generalisability. Or it might have weak evidence but no particular safety contraindications.
Usage Guidance
Do
- Be specific about what isn't established—name the gap precisely
- Distinguish types of limitations—is it evidence quantity, quality, generalisability, or something else?
- Quantify when possible—"only 19 studies met inclusion criteria" is more informative than "limited research"
- Connect limitations to claims—explain how the limitation affects what can be concluded
- Use when making comparisons—compositional similarity doesn't prove functional equivalence
Don't
- Use as a generic disclaimer—"this is not medical advice" is not a limitation in this sense
- Be vague—"more research is needed" without specificity is unhelpful
- Omit when relevant—if evidence is genuinely limited, say so
- Undermine valid claims—limitations should be proportionate to actual uncertainty
Common Limitation Categories
| Category | Description | Example Phrasing |
|---|---|---|
| Evidence quantity | Few studies exist | "Only 19 studies met inclusion criteria in a 2024 scoping review" |
| Evidence quality | Studies exist but have methodological issues | "No RCTs exist; evidence limited to observational studies and case series" |
| Specificity gap | Evidence exists for related but not identical question | "Studies examine facial skin application; lip-specific evidence lacking" |
| Generalisability | Evidence may not apply to all populations | "Studies conducted in adults aged 25-45; applicability to elderly unclear" |
| Mechanism vs outcome | Theoretical rationale exceeds empirical validation | "Compositional similarity suggests potential but doesn't prove clinical equivalence" |
| Confounding factors | Alternative explanations not ruled out | "Improvement may reflect overall formulation quality rather than the specific ingredient" |
Examples
Example 1: Compositional Comparison Limitations
Tallow-sebum similarity claim with appropriate caveats:
{
"@type": "CompositionComparison",
"@id": "#CompositionComparison/TallowSebumSimilarity",
"name": "Tallow-sebum fatty acid composition comparison",
"substance": {"@id": "#ChemicalSubstance/BeefTallow"},
"comparedTo": {"@id": "#AnatomicalStructure/Sebum"},
"similarityBasis": "Fatty acid profile composition",
"similarityDegree": {
"@type": "PropertyValue",
"name": "Shared fatty acid percentages",
"description": "Tallow contains oleic acid (37-47%), palmitic acid (26-28%), and stearic acid (17-25%)—all major components of human sebum in similar proportions"
},
"clinicalRelevance": "Because lips cannot produce sebum, providing a substance with similar fatty acid composition may better address the physiological gap than petroleum products or plant oils with different profiles",
"limitations": "Compositional similarity does not guarantee clinical superiority. Only 19 studies met inclusion criteria in a 2024 Cureus scoping review; no RCTs exist specifically comparing tallow to other lip treatments. The individual fatty acids have Cosmetic Ingredient Review safety approval, but tallow-specific clinical evidence remains limited."
}Why this works: The limitation is specific (19 studies, no RCTs), explains what isn't established (clinical superiority), and notes what IS established (safety of components). It doesn't undermine the compositional claim—it contextualises what that claim can support.
Example 2: Mechanistic Pathway Limitations
Causal pathway with generalisability caveats:
{
"@type": "CausalPathway",
"name": "PRP growth factor cascade",
"description": "How platelet-derived growth factors trigger tissue regeneration",
"pathwaySteps": [
{
"@type": "CausalStep",
"stepNumber": 1,
"mechanism": "Platelet activation triggers alpha granule degranulation",
"result": "Release of PDGF, TGF-β, VEGF, and other growth factors"
}
],
"evidenceLevel": "Strong for mechanism; Moderate for clinical outcomes",
"limitations": "Pathway established primarily in wound healing and orthopaedic contexts. Hair follicle response may differ from dermal fibroblast response. Individual variation in platelet count and function affects growth factor concentration achieved. Optimal protocols (centrifugation speed, activation method, injection depth) not standardised."
}Why this works: Acknowledges that mechanistic understanding (strong) doesn't automatically transfer to specific clinical applications. Lists specific factors that create uncertainty.
Example 3: Physiological Requirement Limitations
Requirement claim with evidence caveats:
{
"@type": "PhysiologicalRequirement",
"name": "Lip tissue lipid requirement",
"tissue": {"@id": "#AnatomicalStructure/VermilionBorder"},
"requirement": "External lipid source to compensate for absent sebaceous glands",
"endogenousCapacity": "Zero—vermilion border contains no sebaceous glands",
"clinicalSignificance": "Explains why lips require ongoing external care whilst facial skin largely self-maintains",
"limitations": "The anatomical absence of sebaceous glands is well-established; the optimal external lipid composition is not. Individual tolerance varies, and some people maintain lip health with minimal intervention despite the anatomical limitation. Environmental factors (humidity, wind exposure) significantly modify the practical requirement."
}Why this works: Distinguishes between what's firmly established (anatomical fact) and what isn't (optimal intervention). Acknowledges individual variation.
Example 4: Multiple Limitation Categories
When several types of limitation apply:
{
"@type": "CompositionComparison",
"name": "Jojoba oil-sebum structural similarity",
"substance": {"@id": "#ChemicalSubstance/JojobaOil"},
"comparedTo": {"@id": "#AnatomicalStructure/Sebum"},
"similarityBasis": "Wax ester composition rather than triglyceride structure",
"similarityDegree": "Jojoba is technically a liquid wax; its wax ester structure more closely resembles sebum than true plant oils",
"limitations": "Three distinct limitations apply: (1) Evidence quantity—clinical studies specifically comparing jojoba to animal-derived emollients for lip care are sparse; (2) Specificity—most jojoba research concerns facial skin or hair, not vermilion border specifically; (3) Compositional nuance—while structurally more similar to sebum than other plant oils, jojoba still differs significantly from animal-derived options in fatty acid ratios."
}Why this works: Explicitly categorises different types of limitation, making it clear the claim faces multiple distinct challenges rather than one vague concern.
Implementation Notes
Relationship to Credibility
Counter-intuitively, including limitations often increases content credibility rather than decreasing it. Readers recognise intellectual honesty and are more likely to trust claims that acknowledge their boundaries than those that don't.
This property is designed to support evidence-based content that maintains scientific integrity whilst still making useful claims.
Granularity
Consider where to place limitations:
- On individual CausalSteps: When a specific mechanism is less certain than others in the pathway
- On entire CausalPathway: When limitations apply to the overall model
- On CompositionComparison: When compositional claims shouldn't be over-interpreted
Validation
Standard schema.org validators will not recognise this property. This is expected for Health Schema extensions. The property is defined in the Health Schema context and will be correctly interpreted by systems that load this context.