Williams, Amy W.

Temporal inference from laboratory testing results and triangulation with clinical outcomes extracted from unstructured electronic health record (EHR) provider notes is integral to advancing precision medicine. Here, we studied 246 SARS-CoV-2 PCR-positive (COVID pos ) patients and propensity-matched 2460 SARS-CoV-2 PCR-negative (COVID neg ) patients subjected to around 700,000 lab tests cumulatively across 194 assays. Compared to COVID neg patients at the time of diagnostic testing, COVID pos patients tended to have higher plasma fibrinogen levels and lower platelet counts. However, as the infection evolves, COVID pos patients distinctively show declining fibrinogen, increasing platelet counts, and lower white blood cell counts. Augmented curation of EHRs suggests that only a minority of COVID pos patients develop thromboembolism, and rarely, disseminated intravascular coagulopathy (DIC), with patients generally not displaying platelet reductions typical of consumptive coagulopathies. These temporal trends provide fine-grained resolution into COVID-19 associated coagulopathy (CAC) and set the stage for personalizing thromboprophylaxis.

Understanding temporal dynamics of COVID-19 symptoms could provide fine-grained resolution to guide clinical decision-making. Here, we use deep neural networks over an institution-wide platform for the augmented curation of clinical notes from 77,167 patients subjected to COVID-19 PCR testing. By contrasting Electronic Health Record (EHR)-derived symptoms of COVID-19-positive (COVID pos ; n = 2,317) versus COVID-19-negative (COVID neg ; n = 74,850) patients for the week preceding the PCR testing date, we identify anosmia/dysgeusia (27.1-fold), fever/chills (2.6-fold), respiratory difficulty (2.2-fold), cough (2.2-fold), myalgia/arthralgia (2-fold), and diarrhea (1.4-fold) as significantly amplified in COVID pos over COVID neg patients. The combination of cough and fever/chills has 4.2-fold amplification in COVID pos patients during the week prior to PCR testing, in addition to anosmia/dysgeusia, constitutes the earliest EHR-derived signature of COVID-19. This study introduces an Augmented Intelligence platform for the real-time synthesis of institutional biomedical knowledge. The platform holds tremendous potential for scaling up curation throughput, thus enabling EHR-powered early disease diagnosis.

Background and objectives Withdrawal from maintenance hemodialysis before death has become more common because of high disease and treatment burden. The study objective was to identify patient factors and examine the terminal course associated with hemodialysis withdrawal, and assess patterns of palliative care involvement before death among patients on maintenance hemodialysis. Design, setting, participants, & measurements We designed an observational cohort study of adult patients on incident hemodialysis in a midwestern United States tertiary center, from January 2001 to November 2013, with death events through to November 2015. Logistic regression models evaluated associations between patient characteristics and withdrawal status and palliative care service utilization. Results Among 1226 patients, 536 died and 262 (49% of 536) withdrew. A random sample (10%; 52 out of 536) review of Death Notification Forms revealed 73% sensitivity for withdrawal. Risk factors for withdrawal before death included older age, white race, palliative care consultation within 6 months, hospitalization within 30 days, cerebrovascular disease, and no coronary artery disease. Most withdrawal decisions were made by patients (60%) or a family member (33%; surrogates). The majority withdrew either because of acute medical complications (51%) or failure to thrive/frailty (22%). After withdrawal, median time to death was 7 days (interquartile range, 4–11). In-hospital deaths were less common in the withdrawal group (34% versus 46% nonwithdrawal, P =0.003). A third (34%; 90 out of 262) of those that withdrew received palliative care services. Palliative care consultation in the withdrawal group was associated with longer hemodialysis duration (odds ratio, 1.19 per year; 95% confidence interval, 1.10 to 1.3; P <0.001), hospitalization within 30 days of death (odds ratio, 5.78; 95% confidence interval, 2.62 to 12.73; P <0.001), and death in hospital (odds ratio, 1.92; 95% confidence interval, 1.13 to 3.27; P =0.02). Conclusions In this single-center study, the rate of hemodialysis withdrawals were twice the frequency previously described. Acute medical complications and frailty appeared to be driving factors. However, palliative care services were used in only a minority of patients.