Table 1.

Phenotypes used in the study. Additional concept unique identifiers (CUIs) for each phenotype are listed in Supplementary Table 1. The last column indicates whether validation was performed against unlabeled subjects in the biobank whose charts were manually reviewed or against a registry cohort

Phenotype	Phecode	CUIs a (Primary and Secondary)	Type of Validation Data
Asthma	495.	C0004096, C0038218	Manually reviewed charts
Crohn disease	555.1	C0010346, C0156147	Manually reviewed charts
Ulcerative colitis	555.2	C0009324, C2937222	Manually reviewed charts
Cardiomyopathy	425.	C0878544, C0007194	Manually reviewed charts
Congestive heart failure; nonhypertensive	428.	C0018802, C0018801	Manually reviewed charts
Epilepsy, recurrent seizures, convulsions	345.	C0014544, C0009951	Manually reviewed charts
Juvenile rheumatoid arthritis	714.2	C0553662, C0157917	Manually reviewed charts
Chronic pulmonary heart disease	415.2	C0152171, C0238074	Manually reviewed charts
Type 1 diabetes	250.1	C0011854, C0375114	Manually reviewed charts
Cardiac congenital anomalies	747.1	C0041207, C0018818	Registry cohort & Manually reviewed charts
Inflammatory bowel disease	555.	C0010346, C0009324	Registry cohort

^aUnified Medical Language System Release 2012AA.