Temporal Relationship and Event Timing

The temporal relationship is the requirement that an adverse event follow, rather than precede, the drug exposure and occur within a biologically plausible interval. Because a cause must precede its effect, timing is the one criterion that, if violated, can by itself make drug causation untenable, and it anchors every method of causality assessment.

Definition

The temporal relationship is the criterion that an adverse event must begin after the drug is administered and within an interval compatible with the drug's pharmacology and the reaction's expected mechanism, providing the necessary (though not sufficient) timing condition for attributing the event to the drug.

Scope

This entry covers the role of time-to-onset, the idea of a plausible latency window for different reaction types, and how timing interacts with dose, duration, and dechallenge. It is a methodological reference on a causality criterion and not guidance for interpreting timing in any individual patient's care.

Core questions

• Why is temporal precedence a necessary condition for drug causation?
• What counts as a biologically plausible time-to-onset for different types of adverse reaction?
• How does timing interact with dechallenge, dose, and duration of exposure?
• Why is plausible timing necessary but not sufficient to establish causation?

Key concepts

• Temporal precedence (cause before effect)
• Time to onset / latency
• Plausible latency window
• Immediate vs. delayed reactions
• Time-course classification of reactions
• Interaction of timing with dose and duration
• Necessary but not sufficient condition

Mechanisms

An event that begins before the drug was taken cannot have been caused by it, so temporal precedence is a logical prerequisite for attribution. Beyond mere precedence, the interval between exposure and onset must be consistent with how the reaction is thought to arise: immediate hypersensitivity may appear within minutes, dose-dependent toxic effects accumulate over a course of treatment, and some reactions are characteristically delayed by days to weeks or appear only after withdrawal. Frameworks that classify reactions by their time-course, such as the dose, time-course, and susceptibility (DoTS) approach, make these expectations explicit so that an observed onset can be judged plausible or implausible (Aronson & Ferner 2003). Plausible timing, reinforced by a consistent dechallenge, strengthens causal inference, but because many non-drug causes can also follow exposure by chance, satisfactory timing is necessary rather than sufficient (Hill 1965; Naranjo 1981).

Clinical relevance

Time-to-onset is recorded in every individual case safety report and scored in structured algorithms, so judging temporal plausibility is central to appraising pharmacovigilance evidence. The entry explains how timing is interpreted as evidence and is not a basis for individual diagnostic or treatment decisions.

Evidence & guidelines

Bradford Hill listed temporality as the one consideration on which he would insist for a causal interpretation, since the cause must precede the effect (Hill 1965). In drug-reaction assessment this principle is operationalised through time-to-onset items in algorithms such as the Naranjo scale (Naranjo 1981) and through time-course classification in the DoTS framework (Aronson & Ferner 2003), which together make timing a routine, explicit element of attribution.

History

The primacy of timing in causal reasoning was articulated for epidemiology by Bradford Hill in 1965, who singled out temporality as indispensable. Drug-safety methods then embedded the criterion: Naranjo's 1981 algorithm scored whether the event followed administration, and later classifications such as DoTS organised reactions explicitly by their time-course, refining the notion of a plausible latency window.

Debates

How wide should the plausible time-to-onset window be?

Reactions range from immediate to long-delayed, so a fixed latency window risks excluding genuine delayed reactions or admitting coincidental ones; time-course classifications attempt to set type-specific expectations, but judging plausibility for an unusual reaction remains a matter of pharmacological reasoning.

Key figures

• Austin Bradford Hill
• Cesar A. Naranjo
• Jeffrey K. Aronson
• Robin E. Ferner

Seminal works

• hill-1965
• naranjo-1981

Frequently asked questions

Why does timing matter so much in causality assessment?

Because a cause must precede its effect, an adverse event that started before the drug was given cannot be attributed to it, and an onset outside a biologically plausible interval weakens the case for drug causation.

Is plausible timing enough to prove a drug caused a reaction?

No. Plausible timing is necessary but not sufficient, because non-drug causes can also follow exposure by chance; timing must be combined with dechallenge, rechallenge, and exclusion of alternative causes.