Sources

Each substance page includes a small list of source links. These links are provided so you can verify information and explore the underlying literature.

HalfLifeDB treats source context as part of the content, not as decoration. A half-life number is only useful when readers can understand where it came from, what was measured, and why another study might report a different value.

What we link to

• Primary research: peer-reviewed pharmacokinetic studies (e.g., concentration-time curves, half-life estimates, metabolites).
• Reviews and textbooks: clinical summaries that describe typical ranges and important caveats.
• Official labeling: prescribing information and regulatory documents when available.
• Scientific databases: records from sources such as PubMed, PubChem, NCBI Bookshelf, DailyMed, and PMC.

Source quality hierarchy

Direct links to a specific study, label, or textbook chapter are stronger than a database search page. Search-result pages are not used as citations. When evidence is broad, mixed, or changing, the profile links the specific studies, labels, chapters, public-health records, or identity pages that support its discussion.

1. Specific pharmacokinetic study or official labeling that reports a relevant value.
2. Reputable review, textbook, or government database record that summarizes pharmacokinetic data.
3. Compound identity or public-health record used only for the background it directly supplies.

Indirect and preclinical sources

Some substances have limited public human pharmacokinetic data. When a profile includes an animal, in vitro, toxicology, postmortem, or otherwise indirect source, that source should be read as context, not as a direct estimate for typical human exposure. Those entries should also carry extra caveats in the notes or reference text.

If a stronger human pharmacokinetic study or official label is later identified, the indirect source may be replaced or moved lower in the reference list.

What does not count as support for a number

A PubChem identity record, a general toxicology overview, or a paper about a related compound can add useful background, but it does not by itself support a numerical human half-life. The same is true of a database search-results page: it can help with research, but it is not a citation for a displayed value.

If the available evidence is indirect, the profile publishes without a calculator and labels the missing human estimate directly. This preserves useful source context without giving a placeholder the appearance of authority.

How to interpret half-life values

Half-life values depend on study conditions such as dose, route, sampling window, analytic method, and the population studied. For many substances, different studies report different estimates. HalfLifeDB uses a single value only when the source set supports a clearly labeled model; otherwise, the substance remains an unmodeled evidence note.

See the research standards and disclaimer for more context.

For the calculator assumptions themselves, read the methodology page.