Abstract

AbstractBackgroundAdult Changes in Thought (ACT) is a prospective cohort study of older adults from Seattle, Washington, USA. Participants are enrolled from Kaiser Permanente Washington membership and may consent for autopsy. Many participants have decades of medical records prior to ACT enrollment. We describe an ambitious project to abstract medical records to facilitate epidemiological investigation.MethodAbstractions include medication data preceding 1977 (the start of automated pharmacy data), laboratory data preceding 1988 (the start of automated laboratory data), women’s health history elements, blood pressures, several cardiovascular conditions, several neurological conditions, physical injuries, psychiatric conditions, and other medical conditions, among many others. Data are entered in a Microsoft Access front‐end interface linked to a Microsoft SQL Server database engine. The database comprises 15 tables in a relational structure. We summarize chart review data and other data according to dementia status.ResultWe have completed abstractions on 1329 individuals, of whom 908 had an autopsy. For the first 380 people with an autopsy, it took an average of 6.2 hours (SD 3.8 hours) to abstract >30 years of medical data. More recent electronic medical record data take longer to abstract. In this sample, 236 (62%) died without Alzheimer’s Disease or Related Dementias (ADRD) dementia, and 144 (38%) died with an ADRD dementia diagnosis. Demographic and clinical data are summarized in the Table. Over 60% had pharmacy data preceding 1977 with a median of more than a half decade’s worth. Over 90% of the cohort had additional laboratory data preceding 1988, with a median of a decade’s worth. Medical conditions were similarly common across people without and with ADRD, except cerebrovascular accidents, diabetes, depression, and confusion during hospitalization were more common among people with dementia. We will update all numbers for AAIC.ConclusionThe ACT Medical Records Abstraction Project transforms medical records into data ready for analyses of associations with outcomes derived from other data from the ACT study. For the autopsy cohort in particular, the Medical Records Abstraction Project provides unmatched detailed longitudinal clinical data to support a wide variety of epidemiological research on clinical‐pathological correlations.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call