The Bradford Hill viewpoints.
Nine considerations — weighed, never tallied.
Bradford Hill's nine viewpoints help assess whether an observed association may be causal. They are not a checklist, formula, or scoring system. A causation opinion should not count how many viewpoints are present and then declare a conclusion. Instead, each viewpoint is weighed in context, with attention to the diagnosis, exposure, timing, dose, competing explanations, and quality of the scientific evidence.
In occupational causation analysis, temporality, strength of association, and dose–response often carry the greatest practical weight. Temporality is essential because exposure must precede the outcome. Strength and dose–response help determine whether the association is large enough and biologically patterned enough to support causation rather than coincidence, background risk, or confounding.
Temporality
The exposure must occur before the outcome. This is the one essential viewpoint: an exposure cannot cause a condition that already existed before the exposure occurred. However, temporality is never sufficient by itself. The fact that symptoms or a diagnosis followed an event may raise a causation question, but timing alone does not establish cause.
An individual develops lateral elbow pain after several weeks of forceful repetitive gripping at work. The timing is compatible with causation because the exposure came first. But the analysis must still determine whether the diagnosis is correct, whether the exposure was sufficient, and whether other explanations are more likely.
Strength of association
The larger the measured association between an exposure and a condition, the more weight it may carry in a causal analysis. A strong association is less likely to be explained entirely by chance, bias, or confounding. A weak association may still be relevant, but it requires more careful evaluation of study quality, alternative explanations, and individual exposure facts.
If high-quality studies show that a particular exposure is associated with a substantially increased risk of a specific disease, that evidence carries more causal weight than a small or inconsistent increase in risk.
Dose–response
Dose–response asks whether more exposure produces more risk, and whether less exposure produces less risk. This viewpoint is especially important in occupational analysis because a recognized risk factor matters only if the individual's exposure was sufficient in intensity, frequency, duration, and timing.
If the risk of a condition increases with greater cumulative vibration exposure, longer duration of forceful gripping, or higher levels of a chemical exposure, that pattern supports a causal interpretation. A minimal, brief, or remote exposure may be biologically plausible but still insufficient.
Consistency
Consistency asks whether the association has been replicated across different studies, populations, settings, and methods. A finding is more credible when it appears repeatedly in well-designed studies rather than in a single isolated report.
An association between a workplace exposure and a disease is more persuasive when it is observed in multiple industries, different worker populations, and studies using different research designs.
Plausibility
Plausibility asks whether the association fits a credible biological, biomechanical, toxicologic, or pathophysiologic mechanism. The proposed cause should make sense in light of what is known about the body, the exposure, and the disease process.
A high-force twisting injury followed by immediate knee pain and swelling may be biomechanically plausible for certain internal knee injuries. By contrast, a trivial exposure with no plausible mechanism for the claimed structural injury carries less causal weight.
Alternative explanations
A credible causation analysis must consider other possible causes and competing explanations. The question is not simply whether the claimed exposure could have contributed, but whether it is more probable than other explanations when the full record is considered.
In an individual with knee osteoarthritis, the analysis should consider age, BMI, prior injury, prior surgery, alignment, genetics, and non-occupational activity before attributing the condition to work exposure alone.
Experimental evidence
Experimental evidence asks whether altering the exposure changes the outcome. In occupational settings, this may include improvement after exposure reduction, worsening with re-exposure, or prevention of disease when exposure controls are introduced. This can be supportive, but it must be interpreted cautiously because symptoms may improve for many reasons.
If symptoms improve when a specific exposure is removed and recur when the exposure resumes, that pattern may support causation. If symptoms persist unchanged despite removal of the exposure, the causal inference may be weaker.
Specificity
Specificity asks whether a particular exposure is associated with a particular disease. It is generally the weakest Bradford Hill viewpoint because many diseases are multifactorial and many exposures can produce more than one effect. Lack of specificity does not rule out causation, but high specificity can strengthen the analysis.
A rare disease strongly linked to a specific chemical exposure may have greater specificity. By contrast, low back pain has many occupational and non-occupational causes, so specificity is usually limited.
Coherence
Coherence asks whether the proposed causal relationship is compatible with the existing body of knowledge. The conclusion should fit, rather than contradict, established clinical, epidemiologic, biologic, and occupational evidence.
A proposed work-related cause is more credible when it aligns with known disease mechanisms, accepted exposure science, clinical course, and diagnostic findings. A proposed cause that conflicts with established knowledge requires strong supporting evidence.
No causal hypothesis is proven with absolute certainty. Scientific analysis tests proposed explanations against the available evidence and is often strongest when excluding unsupported explanations. Sometimes the most defensible conclusion is not a positive causal finding, but a disciplined exclusion: the evidence does not support the proposed explanation.