Abstract 2: African American Patients Without Pre-existing Cognitive Impairment Are More Likely To Develop Post Stroke Cognitive Impairment After First-ever Stroke: An Analysis Of Electronic Health Record Data

Background: Post stroke cognitive impairment (PSCI) can be as high as 15-70% after stroke depending on the patient population and diagnostic tool. Few studies on PSCI have utilized large administrative or electronic health records (EHR) to evaluate trends in PSCI in the current population. Methods: We analyzed Cerner Health Facts® EMR database, which is comprised of de-identified EHR data from over 700 hospitals and clinics in the US from 2000-2018. We evaluated patients ≥40 years of age with a first time acute ischemic stroke (AIS) diagnosis using ICD9/10 codes. PSCI was defined as pts with ICD 9/10 codes for dementia, mild cognitive impairment, or on medications for dementia. Pts with first stroke in the Cerner database and no pre-existing cognitive impairment were included; those with no follow-up visits in the system were excluded. We compared hazard ratios for developing PSCI for patient characteristics. Results: A total of 211,622 AIS pts were evaluated, of which 153,078 had follow-up data in the system and no prior diagnosis of PSCI. Among these, the rate of PSCI was 9.29%. Most pts qualified under PSCI using dementia ICD codes (15,280) compared to mild cognitive impairment codes (4,321) or medication usage (1,032). Median time to PSCI diagnosis was at their first follow-up visit within the first year after stroke. Pts ≥65 years of age at time of stroke (HR 2.95, 95% CI 2.83,3.07) and of African American race (HR 1.37, 95% CI 1.31,1.46) were more likely to develop dementia. There were no disparities for developing PSCI between Hispanic and non-Hispanic pts (HR 0.89, 95% CI 0.69, 1.16). Male pts were less likely to develop PSCI than female counterparts (HR 0.84, 95% CI 0.81,0.87). Conclusion: Our analyses highlight the viability of utilizing large administrative databases to assess trends in PSCI diagnosis. The number of pts with PSCI may be underestimated however, given the importance of EHR records in patient care, this suggests PSCI is under-diagnosed in the community. Previously described racial disparities for black survivors persisted but male patients had less PSCI. Further study into other administrative databases is necessary to assess if these data are consistent in other EHR systems and to evaluate new trends in PSCI diagnosis and treatment.