UK phenomics platform for developing and validating EHR phenotypes: CALIBER

Abstract

Objective Electronic Health Records (EHR) are a rich source of information on human diseases, but the information is variably structured, fragmented, curated using different coding systems and collected for purposes other than medical research. We describe an approach for developing, validating and sharing reproducible phenotypes from national structured EHR in the United Kingdom (UK) with applications for translational research. Materials and MethodsWe implemented a rule-based phenotyping framework, with up to six approaches of validation. We applied our framework to a sample of 15 million individuals in a national EHR data source (population-based primary care, all ages) linked to hospitalization and death records in England. Data comprised continuous measurements e.g. blood pressure, medication information and coded diagnoses, symptoms, procedures and referrals, recorded using five controlled clinical terminologies: a) Read (primary care, subset of SNOMED-CT), b) International Classification of Diseases 9th/10th Revision (ICD-9, ICD-10, secondary care diagnoses and cause of mortality), c) OPCS Classification of Interventions and Procedures (OPCS-4, hospital surgical procedures), and d) DM+D prescription codes. Results Using the CALIBER phenotyping framework, we created algorithms for 51 diseases, syndromes, biomarkers and lifestyle risk factors and provide up to six validation approaches. The EHR phenotypes are curated in the open-access CALIBER Portal (https://www.caliberresearch.org/portal) and have been used by 40 national/international research groups in 60 peer-reviewed publications.

The genomics of heart failure: design and rationale of the HERMES consortium

“…Subsequent studies will follow a stepwise strategy for phenotype definition to address HF subtypes based on aetiology, LVEF, and disease progression ( Figure ). Mobilizing HF subtype data from electronic health records, leveraging large genomic biobanks, will be necessary to ensure sufficient statistical power for subtype analysis, and this will be achieved through the deployment of validated multi‐modal rule‐based phenotyping algorithms 25 …”

“…Mobilizing HF subtype data from electronic health records, leveraging large genomic biobanks, will be necessary to ensure sufficient statistical power for subtype analysis, and this will be achieved through the deployment of validated multi-modal rule-based phenotyping algorithms. 25 Given the mortality associated with HF, inclusion of incident and prevalent cases in analyses may lead to attenuation of effect estimates, due to survivorship or collider bias and increased heterogeneity 26,27 ; however, this bias is partially mitigated by the increased power associated with a larger sample size that can be achieved when prevalent cases are included.…”

Healthcare resource utilisation and mortality outcomes in international migrants to the UK: analysis protocol for a linked population-based cohort study using Clinical Practice Research Datalink (CPRD), Hospital Episode Statistics (HES) and the Office for National Statistics (ONS)

“…This study will be carried out as part of the CALIBER programme. CALIBER, led from the UCL Institute of Health Informatics, is a research resource consisting of anonymised, coded variables extracted from linked electronic health records, methods and tools, specialised infrastructure, and training and support 11,12 .…”

“…Previously established methods by CALIBER will be used for the development of a migration phenotype 12 . The CPRD code browsers will be searched for diagnostic terms relating to migration using the following search terms: *migrant*, *migrat*, *countr*, *asylum*, *refugee*, *visa*, *abroad*, *born in*, *origin*, *illegal*, *language*.…”

Machine Learning Approaches for Electronic Health Records Phenotyping: A Methodical Review

“…Historically, phenotypes have been inferred from rule-based algorithms using expert defined criteria from care guidelines and/or knowledge of clinical practice [11]. Because documentation varies substantially across phenotypes, providers, and institutions, developing a sufficient set of rules that encode how a condition may be documented in patient records is prohibitively resource intensive [12,13]. Moreover, the scope of rule-based algorithms is severely limited, particularly for more complex phenotypes recorded primarily in clinical notes [14][15][16].…”