Why half-life varies between people and studies

Half-life numbers are often reported as if they are fixed properties. In reality, they are measurements under specific conditions.

Study design variables

• Population: age range, sex distribution, health status, co-medications, and inclusion criteria.
• Dose and formulation: immediate vs extended release, route of administration, and dose level.
• Sampling window: how long samples were collected after the dose and how dense the sampling was.
• Analytic method: assay sensitivity, whether metabolites are measured separately, and how “terminal” phase was defined.
• Modeling approach: one-compartment vs multi-compartment fits and which parameters were reported.

Person-to-person variables

• Metabolism and transport: genetic differences, enzyme induction or inhibition, and drug-drug interactions.
• Liver and kidney function: changes in clearance shift persistence.
• Body composition: distribution can differ with fat mass and total body water.
• Timing and food: meals, circadian rhythms, and hydration can influence absorption and clearance in some cases.

How to use HalfLifeDB given this variability

HalfLifeDB chooses one representative half-life value per substance to illustrate a curve. That curve is not an individualized prediction. Use the curve to learn the behavior of exponential decay and to compare “short” vs “long” half-life substances.

If you need to verify a number, use the source links on each substance page and read the original study context.