Domain Concordance: Tech Enhancement Pilot Data

Some diagnosis concept sets used to identify relevant cohorts will need to be adjusted, as some couplets (like Fracture Diagnosis / Imaging Procedure) did not have any patients meeting the diagnosis criteria despite that being an unlikely reality. The Outpatient Visit / Outpatient Condition Type couplet demonstrates that there is an unusually high incidence of outpatient visits not being associated with a diagnosis with the same enc_type. Inpatient and ED visits do not display this same level of incongruity. The CKD Diagnosis / Antihypertensive Drug couplet also displays some unusual behavior, as we generally observe a higher proportion of patients only in cohort 2 due to the myriad of treatment targets for antihypertensive medication not related to CKD. This could be related to the concept set used for the drug cohort or be reflective of the sampling method used to retrieve the 10% dataset. Please note that concept sets used in this check may need to be refined for a full rollout. These results may not reflect the full availability of some variables due to site-specific coding practices that may not have been accounted for.