What physicians, attorneys, and claims professionals need to know

A diagnosis answers what condition is present. A causation analysis asks why it occurred – and whether the evidence supports linking it to a claimed event or exposure. Those are different questions, and a defensible opinion must answer both in the correct order.

At a glance: Diagnosis is not causation. Timing is not causation. Impairment is not causation. Each requires a separate analysis supported by the clinical facts, exposure evidence, scientific literature, and applicable law.

Why Causation Analysis Matters

Workers’ compensation claims, personal injury matters, disability determinations, and occupational disease cases often turn on whether a diagnosed condition is medically related to a particular event or exposure. The answer cannot rest on intuition, chronology alone, or an expert’s credentials. It should be transparent enough that another reader can follow the evidence from the diagnosis to the conclusion.

Start With Three Different Questions

Level	Core question	What the analysis does
General causation	Can this type of exposure cause this type of condition?	Critically appraises the population-level scientific evidence.
Individual causation	Did the exposure probably cause, exacerbate, or aggravate this person’s diagnosed condition?	Applies the general evidence to verified exposure, timing, clinical findings, and competing factors.
Legal causation	Does the governing law attach liability or benefits?	Applies statutes, case law, presumptions, and the jurisdiction’s burden of proof. This is not decided by medical science alone.

Practical point: the physician provides a medical opinion; the adjudicator or other authorized decision-maker applies the law.

Cause in fact is often described as the ‘but-for’ inquiry: would the outcome have occurred without the event or exposure? Depending on the jurisdiction and claim type, other factual-causation formulations may apply. Proximate or legal cause asks a different question – whether the factual connection is close enough, under the governing law, for liability or benefits to attach.

Statutory or judicial presumptions may alter who must prove what. Some presumptions are rebuttable; others may be conclusive once specified facts are established. Their existence, scope, and effect are legal questions that should be confirmed before the medical opinion is framed.

Association Is Not Causation

A statistical association shows that two variables occur together more often than expected. It does not, by itself, prove that one caused the other. The AMA Guides to the Evaluation of DISEASE and INJURY Causation, Second Edition distinguishes among direct, indirect, and noncausal relationships:

1. Direct causal association. In the AMA Guides’ terminology, the event is necessary and sufficient for the outcome. These situations are less common than multifactorial cases.
2. Indirect causal association. The event contributes to the outcome in combination with other factors. Many diseases and musculoskeletal presentations fit this multifactorial model.
3. Noncausal association. A third factor, bias, chance, or another explanation accounts for the observed relationship. Age, for example, can confound associations between two age-related findings.

The classic post hoc fallacy – ‘after this, therefore because of this’ – is especially important in medicolegal work. A symptom that begins after an incident satisfies temporal sequence, but temporal sequence alone does not establish mechanism, dose, biological plausibility, or probability of causation.

Use a Two-Stage Medical Analysis

A strong opinion moves from population evidence to the individual case. Skipping either stage creates a gap in the reasoning.

Stage 1: Establish General Causation

General causation asks whether the type and level of exposure at issue can cause the diagnosed condition. A disciplined review should:

1. Define the question. Specify the exposure, dose or task characteristics, outcome, and relevant population before searching.
2. Search comprehensively. Use reproducible terms and seek the full body of relevant epidemiologic evidence, not only supportive studies.
3. Appraise study design and methods. Consider exposure measurement, diagnostic ascertainment, bias, confounding, statistical precision, and whether results are consistent across higher-quality studies.
4. Evaluate causal coherence. Use the updated Bradford Hill considerations – including temporality, strength, dose-response, consistency, plausibility, coherence, experiment, and related considerations – as aids to reasoning, not as a mechanical checklist.[2]
5. State the strength and limits of the evidence. Distinguish evidence of causation from evidence of association, insufficient evidence, and evidence suggesting no material association.

The Guides also describes a formal five-step scoring method used to develop its evidence tables: literature search, quality scoring, study-design classification, weighting, and overall strength-of-evidence determination. An evaluator should understand that method, but a case report need not pretend to recreate a multidisciplinary panel score when it has not done so.

Stage 2: Determine Individual Causation

Individual causation asks whether the exposure caused or contributed to this person’s condition to the degree required by the governing standard. The analysis should address six questions:

1. Is the diagnosis sound? Separate symptoms from diagnoses and identify the objective, clinical, imaging, laboratory, electrodiagnostic, or pathologic evidence that supports each diagnosis.
2. What was the actual exposure? Describe frequency, intensity, duration, posture, force, repetition, vibration, concentration, or other relevant dose metrics. A job title is not an exposure assessment.
3. Does the timing fit? Assess onset, latency, progression, dechallenge or recovery, recurrence, and the condition’s expected natural history.
4. Does the mechanism fit the clinical pattern? Compare the claimed mechanism with anatomy, physiology, biomechanics, toxicology, and the distribution and severity of findings.
5. What other factors matter? Identify relevant preexisting disease, constitutional and genetic factors, comorbidities, nonoccupational exposures, prior injuries, medications, and psychosocial factors when supported by the literature.
6. What conclusion does the total evidence support? Weigh supporting and contrary evidence, apply the jurisdiction’s required terminology, and state uncertainty or missing information plainly.

Classify the Exposure’s Role Precisely

Once a medical relationship is supported, describe the role of the event or exposure consistently. The Guides uses the following concepts, although statutes and case law may define them differently:

1. Direct cause. The event or exposure is sufficient to produce the condition without another contributing variable.
2. Aggravation. A preexisting condition is permanently worsened beyond the level expected from its natural history.
3. Exacerbation. Symptoms, signs, disability, or impairment worsen temporarily, with eventual return to baseline or to the level expected from natural progression.
4. Acceleration. A condition already progressing along its natural course advances more rapidly because of the exposure.
5. Recurrence. Prior signs or symptoms reappear, often with minimal or no new provocation; recurrence does not necessarily establish a new work cause.
6. Multiple causation. Two or more occupational or nonoccupational factors probably contribute to the condition or its consequences.

Do not use aggravation and exacerbation interchangeably. The distinction between permanent change and temporary symptom increase can determine both the medical characterization and the legal consequence.

State the Legal Threshold Carefully

Causation standards are not uniform. Common formulations include ‘more probable than not,’ ‘preponderance of the evidence,’ ‘reasonable degree of medical probability,’ and ‘reasonable degree of medical certainty.’ Courts do not interpret these phrases identically in every jurisdiction.

‘More probable than not’ is commonly described as greater than 50%, but it is a decision threshold, not a demand for a mathematically calculated patient-specific probability. ‘Preponderance of the evidence’ is a legal burden that often expresses a similar more-likely-than-not concept, but the controlling authority should define its use. ‘Medical certainty’ and ‘medical probability’ may also carry jurisdiction-specific meanings.

Drafting rule: Use the exact legal formulation supplied for the case. Avoid ‘possibly,’ ‘could have,’ or ‘cannot rule out’ when the question requires probability. At the same time, do not imply scientific certainty when the evidence supports only a probabilistic conclusion.

Treat Apportionment as a Separate Analysis

Apportionment allocates responsibility among multiple probable causes. It should not be confused with identifying risk factors or listing every possible contributor. Under the AMA Guides’ framework, only causes that meet the applicable probability threshold are included; percentages should then be tied to the medical literature and the facts of the case.

A sound apportionment opinion should identify what is being apportioned – the condition, permanent impairment, disability, treatment need, or another outcome – because these are not interchangeable. It should explain why each included cause qualifies, how the allocation was derived, and what uncertainty remains. A confident percentage without a disclosed method is still speculation. Whether apportionment is permitted, required, or legally relevant is jurisdiction-specific.

What a Defensible Report Should Contain

A high-quality report lets a nonphysician trace the reasoning without guessing. At minimum, it should include:

1. Referral question and legal context. Identify each question, the jurisdiction, and the requested causation standard.
2. Information reviewed. List records, imaging, tests, testimony, job materials, exposure data, and unavailable information that could change the opinion.
3. Supported diagnoses. Distinguish symptoms from diagnoses and subjective reports from objective findings.
4. Exposure and chronology. Provide a specific, sourced timeline and quantify exposure when possible.
5. General-causation evidence. Summarize the most relevant literature, including important contrary or limiting evidence.
6. Individual-causation analysis. Compare the patient’s facts with the evidence and address competing causes rather than merely naming them.
7. Classification and apportionment. State whether the opinion concerns direct cause, aggravation, exacerbation, acceleration, recurrence, or multiple causation, and explain any allocation.
8. Conclusion. Answer every referral question for each diagnosis using the required legal language, with limitations and the information that would change the opinion.

Three Common Failure Modes

1. Chronology as proof. ‘Symptoms began after work’ establishes sequence, not causation.
2. The population-to-patient leap. A published association does not prove that the exposure was sufficient, relevant, or causal in the individual case.
3. Ipse dixit. ‘Based on my experience’ is not a substitute for a documented method, reliable facts, and scientific support.

Practical Takeaways by Role

For physicians: Start with a defensible diagnosis, establish causation, quantify the actual exposure, weigh alternative explanations, and answer in the jurisdiction’s language. If the evidence is insufficient, say so.

For attorneys: Confirm the governing standard and presumptions before the opinion is written. Test whether the expert separated general from individual causation, used reliable exposure facts, addressed contrary evidence, and explained any apportionment.

For claims professionals: A useful report shows its work. Look for a supported diagnosis, a specific exposure assessment, a literature-based causal bridge, a fair analysis of competing factors, and a conclusion that directly answers the referral questions.

The Bottom Line

Causation analysis is a reproducible reasoning process, not a conclusion added to the end of a clinical note. The strongest opinions separate science from law, association from causation, general evidence from individual application, temporary symptoms from permanent change, and causation from apportionment. When each step is explicit, physicians can explain the opinion, attorneys can test it, and claims professionals can evaluate it fairly.

Selected References

1. Melhorn JM, Talmage JB, Ackerman WE III, Hyman MH, eds. AMA Guides to the Evaluation of Disease and Injury Causation. 2nd ed. American Medical Association; 2014.
2. Hill AB. The environment and disease: association or causation? Proceedings of the Royal Society of Medicine. 1965;58:295-300.

Educational notice: This article is a general educational overview, not legal or medical advice for a specific case. Causation standards, presumptions, and apportionment rules vary by jurisdiction. Consult current scientific literature, applicable law, and appropriately qualified professionals for case-specific analysis.