The reuse of data from electronic medical records (EMRs) and other clinical data systems holds tremendous promise for improving the efficiency and effectiveness of health research. Clinical data in the EMR is a potential source of rich longitudinal data for research, and the recent government efforts to promote the use of EMRs in the clinical setting may further promote the use of such systems in the US healthcare system. As the use of EMRs expands, the demand for usable data from these systems for research has also expanded.
One such effort by the Electronic Medical Records and Genomics Network (eMERGE) has investigated whether data captured through routine clinical care using EMRs can identify disease phenotypes with sufficient positive and negative predictive values for use in genome-wide association studies (GWAS). Most EMRs captured key information (diagnoses, medications, laboratory tests) used to define phenotypes in a structured format; in addition, natural language processing has also been shown to improve case identification rates.*
PheKB is an outgrowth of that validation effort and provides a collaborative environment of building and validating electronic phenotype algorithms. On this site you can:
- View existing algorithms
- Enter or create new algorithms
- Collaborate with others to create or review algorithms
- View implementation details for existing algorithms
Phenotype algorithms can be viewed by data modalities or methods used:
- CPT codes
- ICD 10 codes
- ICD 9 codes
- Laboratories
- Medications
- Vital Signs
- Natural Language Processing
Algorithms can also be viewed by:
- Implementation results (positive predictive value, sensitivity, publications)
- Institution
- Work Group
- Network affiliation (e.g., eMERGE, PGPop, PGRN)
*Kho, Sci Transl Med 20 April 2011: Vol. 3, Issue 79, p. 79re1