Severity

Severity is a graded rating assigned to a harm — a drug interaction, an adverse effect, or a toxicity finding — that describes how serious the outcome is if it occurs. It measures magnitude, not probability: a severe rating means the outcome is dangerous when it happens, not that it is likely to happen.

Different classification systems divide the scale differently. One framework might use four tiers — mild, moderate, severe, life-threatening — while another compresses or expands them. Because there is no universal standard, the same interaction can carry different labels in different databases, depending on which scheme the source applied.

How it is done

A severity rating is produced by evaluating what a harm actually does to the body. The core dimensions are how far physiology is disrupted, whether the effect reverses on its own or requires intervention, and how serious that intervention would need to be.

At the lower end sit outcomes that resolve without treatment. At the upper end are outcomes that require emergency response, cause irreversible damage, or can be fatal. Most severity frameworks are anchored at these poles — spontaneous recovery versus clinical necessity — and subdivide the space between them.

On a substance page or interaction record, severity is typically displayed as a tier label alongside a brief mechanism note. That pairing — label and mechanism — is intended to give the reader both a summary judgment and the reasoning behind it.

What it cannot tell you

Severity alone does not establish how likely a harm is to occur. An interaction rated severe might be rare under ordinary conditions; one rated mild might be near-certain. A low rating is easy to misread as a clean bill of health for a combination — that is the failure mode worth naming plainly.

Risk is the product of both severity and probability. Without probability data, severity tells only half the story: the half about what happens if things go wrong, not how often they do.

Individual variation complicates the picture further. Pre-existing conditions, genetic differences in how drugs are metabolised, and the presence of other substances can all shift where a given person lands on the outcome distribution, regardless of a population-level severity label.

Severity ratings are also context-dependent in ways that are often invisible. A rating drawn from studies at therapeutic doses in healthy adults does not transfer automatically to higher doses, more frequent use, or populations with greater vulnerability. The label carries the context of its source whether or not that context is printed beside it.