UCL Institute of Health Informatics


HDR UK Launches New Phenotype Library for Improved Research using Electronic Health Records

11 October 2021

IHI and HDR UK are delighted to announce the launch of the HDR UK Phenotype Library, a new resource supporting the sharing of electronic phenotypes (definitions of how health data can be used to measure real-world concepts relevant to research and clinical care).

CALIBER database is a comprehensive and searchable phenotype library with 570 phenotypes, 941 concepts, 82843 clinical codes and 24 data sources. There are 5 computer generated blue figures on a black background.

“This substantial upgrade in the HDR UK Phenotype Library is a clear step forward in providing patients and clinicians with useful data-driven definitions of the diseases and conditions that matter. Now with more than 700 phenotypes, there is a much wider coverage of diseases, as well as interoperability with tools to generate further phenotypes using electronic health records for better healthcare and research.”

(Professor Harry Hemingway, Director of Health Data Research London)

“Understanding disease and developing new treatments starts with a shared understanding of what each disease looks like. To unleash the power of health data we need that shared understanding to be in terms of things we find in health records. This catalogue is growing and this latest launch is a huge step forward towards a full catalogue of human disease. Its potential is enormous.”

(Colin Wilkinson, patient and member of the Phenomics for Patient Action Group)

"When people, unfortunately, become ill, the first and most important challenge is to receive the correct diagnosis, because without that important information a person's treatment plan will be flawed and could result in unnecessary treatment or lack of treatment, increased stress and ultimately failure to cure the problem or untimely death. The HDR UK Phenotype Library of phenotyping algorithms for defining human health conditions for research could help improve this potential scenario leading to improved care of patients."

(Stella Macpherson, patient and member of the Phenomics for Patient Action Group)

The Phenotype Library curates over 100,000 clinical ontology terms into 753 phenotypes from numerous contributing organisations across the UK, spanning critical disease areas including heart disease, cancer, COVID-19 and others. Phenotypes are defined against 28 different research datasets and 14 coding systems, with more being added frequently. New contributions are welcome from anyone working in the field.

The Library is interoperable with other tools and resources, making it part of a broader ecosystem driving the next generation of health research methods. It is currently integrated with the metadata catalogue of the HDR UK Innovation Gateway, as well as Phenoflow, a tool enabling workflow-based computable phenotype definitions. Further integrations with other tools and analysis workflows are possible using the REST API and client R package. We look forward to collaborating with others to use this resource in research and clinical settings with the potential to improve patient health and well-being.

 For more information, visit the Phenotype Library at https://phenotypes.healthdatagateway.org.