PCORnet COMPUTABLE PHENOTYPE
Operational Definition for [Condition/Characteristic]
Submitted by / Name, Network AffiliationCollaborating Networks / Other networks collaborating in computable phenotype creation
Date Submitted / Date submitted to PMO [MM/DD/YYYY]
Location / URL for location where document is stored/displayed
BACKGROUND
A computable phenotype refers to a clinical condition or characteristic that can be ascertained via a computerized query to an electronic health record (EHR) system or clinical data repository using a defined set of data elements and logical expressions.
Provide background and/or summary information regarding the proposed computable phenotype (e.g., who developed the definition if known, and its intended purpose or context of use, and whether it has been used in existing projects). In addition, please explicitly mention whether the intent of the computable phenotype is for screening or to identify confirmed cases for research.
Purpose and Audience for this document:
The purpose of this document is to describe the computable phenotype by defining the specific data elements, included and excluded codes, and supporting logic, as well to describe the justification and any previous experience or history for the computable phenotype in its intended context of use.
The intended audience of this document includes researcher investigators or sponsors that plan to use EHR data to identify or define this condition in research or healthcare applications. PCORnet-funded researchers and applicants are encouraged to re-use existing computable phenotypes wherever possible and appropriate.
OPERATIONAL DEFINITION
Condition/Characteristic:
Briefly explain the condition/characteristic of interest, and how it tends to be diagnosed or recorded in EHRs.
Description of Computable Phenotype for [Condition/Characteristic]:
Provide the phenotype definition. Include the phenotype definition source or authors. List data elements, standardized coding system, and specific codes. Describe any inclusion or exclusion criteria – are there logical and temporal relationships between different data elements? If ICD-9 diagnoses codes are used, specify the diagnosis element type (final, admitting, discharge, etc.), ranking (primary, secondary, etc.), and context (technical billing, professional billing, medical record billing, etc.) where possible. Include details regarding the precision of coding (for example, number of decimal places considered for ICD-9 codes) and encounter types (inpatient, outpatient, ED-to-inpatient, etc.).
What required or preferred data domains would need to be added to CDM v3.0 in order for a phenotype to be usable by PCORnet? Please respond.
Implementation Guidance:
Provide notes or guidance to other researchers who may use this definition for future work. Explain any potential limitations or challenges experienced. Also, include any details that would help a researcher implement this definition in another institution.
PRIOR APPLICATION/VALIDATION
Has this computable phenotype definition been used before? Please respond.
Has it been validated externally or used in an internal validation study? Please respond.
Validation Methodology:
Include details regarding methodology of proposed or completed phenotype definition validation.
Validation Results:
Include details regarding validation results (PPV, NPV, sensitivity, specificity, etc.), if the definition has been validated.
REFERENCES
List references uses in this document.