Written and maintained by CASRAI Editorial Board
Published
On this page
Short answer: The PhenX Toolkit is a free online catalogue of recommended standard protocols for collecting phenotype and exposure data in biomedical research. Expert working groups choose the protocols by consensus, so that different studies measure the same characteristics the same way and their data can be compared. It is funded by the National Human Genome Research Institute (NHGRI) and maintained by RTI International. This page is general information about a research resource, not medical advice.
Definition and scope
PhenX stands for consensus measures for Phenotypes and eXposures. The original description in the American Journal of Epidemiology (Hamilton et al., 2011) calls it a core set of high-quality, well-established, low-burden measures intended for large-scale genomic studies. The toolkit’s own About page says it provides recommended standard data collection protocols for biomedical research, and that common protocols facilitate cross-study analysis.
The programme has been funded by NHGRI since 2007; the About page lists the current award under the Biomedical Knowledgebase (U24HG012556) and names co-funding from NHLBI, OD/OBSSR, NINDS and NIMHD, a NIDA supplement, and earlier support from NIMH, NCI and the NIH Tobacco Regulatory Science Program. The toolkit is freely available to the scientific community.
How the toolkit is organised and used in practice
The About page defines a small vocabulary. Knowing it is the quickest way to read the site, and it is where most newcomers get confused.
| Term | What the toolkit means by it | What you use it for |
|---|---|---|
| Measure | A standardized way of capturing data on a characteristic of a study subject | Deciding what to record |
| Protocol | The standard procedure a Working Group recommends for collecting and recording a measure | Deciding how to record it |
| Domain | A research field with a unifying theme | Browsing by field |
| Collection | A group of measures sharing a characteristic, target population or topic; may span domains | Finding measures for a specific study type |
| Data Collection Worksheet (DCW) | Lists each data item a protocol requires | Building case report forms and data capture |
| Data Dictionary (DD) | Lists each protocol variable with its attributes and identifiers | Defining variables, and preparing data for submission to dbGaP |
A study team typically picks a measure, adopts the recommended protocol unchanged, and uses the worksheet and dictionary to build its forms and variable definitions. The toolkit states that its data dictionaries are designed to fully support submission of data to dbGaP; see also dbGaP submission and controlled-access requests. For how a dictionary differs from a codebook, see Data Dictionary in Research Data Management and Codebook: What It Is and How to Build One.
How protocols are chosen
According to Hamilton et al. (2011), each domain is handled by a Working Group of 6 to 9 domain experts. The process typically took 7 to 9 months, with one in-person meeting and four to six conference calls, and up to 25 preliminary measures were put to the wider scientific community for input before the group reached consensus. The stated goal was about 15 measures per domain, and the selection criteria were that a measure be well established, usable by non-specialists, and low burden. That last criterion explains the toolkit’s character: protocols are designed to be added to a study without much extra cost to participants or staff.
Where it matters for labs and clinics
For teams that collect data from participants, the practical value is harmonisation. If two studies use the same protocol for the same characteristic, their data can be combined or compared without re-deriving definitions. Because the data dictionaries map to dbGaP submission, the toolkit also fits funder expectations for sharing data; see NIH Data Management and Sharing Plan for the planning side. Teams that capture data in REDCap can read REDCap data dictionary structure alongside the worksheets.
What PhenX does and does not promise
The 2011 paper is explicit about limits, and these are worth knowing before citing the toolkit as a justification for a measurement choice:
- The recommended measures are not necessarily gold-standard measures, which are often burdensome to collect.
- Promising newer measures may be absent.
- The toolkit does not modify existing protocols; it recommends them as published.
The toolkit has grown since its 2011 launch, which began with 21 domains, so check the toolkit itself for the current list of domains and collections.
Common misunderstandings
- “PhenX is a questionnaire.” It is a catalogue of protocols; a protocol may use a questionnaire, a physical measurement or another procedure.
- “A protocol is the best possible measure.” It is the consensus low-burden choice, not necessarily the most precise one.
- “Using PhenX is the same as submitting to dbGaP.” The dictionaries are designed to support dbGaP submission; submission is a separate process with its own requirements.
- “The toolkit is a clinical guideline.” It supports research data collection and is not clinical guidance.
FAQ
What does PhenX stand for?
Consensus measures for Phenotypes and eXposures (Hamilton et al., 2011).
Who runs and funds the PhenX Toolkit?
RTI International maintains it; NHGRI has funded it since 2007, with co-funding from several other NIH institutes and offices (toolkit About page, read 2026-11-08).
Is the PhenX Toolkit free?
Yes. The original paper describes it as freely available to the scientific community.
What is the difference between a measure and a protocol?
A measure is the characteristic to capture; a protocol is the recommended procedure for capturing it.
Sources
- PhenX Toolkit, About page, https://www.phenxtoolkit.org/about (read 2026-11-08).
- Hamilton CM et al. The PhenX Toolkit: Get the Most From Your Measures. Am J Epidemiol 2011;174(3):253-260. doi:10.1093/aje/kwr193, https://pmc.ncbi.nlm.nih.gov/articles/PMC3141081 (read 2026-11-08).







