PPubMed28 Feb 2025Cited 5×

Exploring beyond diagnoses in electronic health records to improve discovery: a review of the phenome-wide association study

Nicholas C Wan, Monika E Grabowska, Vern Eric Kerchberger, Wei-Qi Wei

Abstract

We queried the PubMed database from 2010 (year of the first PheWAS publication) to December 31, 2023. The search terms included keywords relating to PheWAS and informatics approaches found in titles and abstracts (ie, phenome wide association[Title/abstract] OR pheWAS[Title/abstract] OR pheWAS[Title/abstract] OR conceptWAS[Title/abstract] OR medWAS[Title/abstract] OR ddiWAS[Title/abstract] OR proWAS[Title/abstract] OR visitWAS[Title/abstract]). All identified free full-text publications underwent a manual review (performed by N.W.). Our objective was to encompass studies that utilized PheWAS a

A figure from Exploring beyond diagnoses in electronic health records to improve discovery: a review of the phenome-wide association study — fig. from the paper

§ The Valyu brief

The paper, in five minutes

Read from the full text via Valyu Contents

EHR‑driven PheWAS mostly rely on diagnosis codes, often lack replication, and need richer data types like clinical notes to improve discovery.

What’s new

i.Systematic review of 291 PheWAS articles published 2010‑2023, summarizing methods, exposures, and cohorts.
ii.Quantified that 55.6% of studies use phecodes, 72.8% rely on genetic exposures, and 69.4% lack replication.
iii.Identified gaps such as limited phenotyping depth, ancestry diversity, and need for data interoperability.

How they did it

We queried PubMed for PheWAS‑related articles from 2010 to 31 December 2023, screened titles/abstracts, and performed a manual full‑text review by a single extractor. Data on exposures, phenotypes, cohort size, terminology, replication status, and ancestry were extracted and tabulated.

What they found

690 articles retrieved; 291 met inclusion criteria.
162 (55.6%) defined phenomes with phecodes.
72.8% of studies used genetic variants or polygenic risk scores as exposures.
69.4% of studies did not perform replication analyses.
Diagnosis codes were the primary phenotyping data in 71.5% of articles.

Caveats

✕The literature search was limited to PubMed, potentially missing relevant studies from other disciplines.
✕The review focused only on EHR‑driven PheWAS, excluding studies using non‑EHR phenotyping sources.
✕Data extraction was performed by a single reviewer, introducing possible bias.

§ The Valyu brief

Reading the full paper and taking notes. This takes a few seconds…

§ Ask this paper

Ask a question about this paper

Valyu reads the full text and answers from what the paper actually says.

§ Related work in other archives

Exploring beyond diagnoses in electronic health records to improve discovery: a review of the phenome-wide association study

The paper, in five minutes

Ask a question about this paper

Similar papers in all four archives

Similar papers in all four archives

The paper, in five minutes