Investigate and characterize the occurrence of encephalopathy in COVID-19 patients.

Using the Cerner COVID-19 de-identified dataset, we identified patients with COVID-19 infection and encephalopathy using ICD-10-CM. We compared the mean length of hospitalization stay in days (LoS) and outcomes (rate of mortality, discharge home, discharge to skilled nursing facilities (SNF), and discharge to a hospice) in these patients: COVID-19 with encephalopathy (COVID-19E), COVID-19 without encephalopathy (COVID-19WE), non-COVID-19 with encephalopathy (non-COVID-19E), non-COVID-19 without encephalopathy (non-COVID-19WE).

Among a total of 117,496 patients, 25,261 were diagnosed with COVID-19 infection. There was a significantly higher number of encephalopathies in COVID-19 compared to non-COVID-19 patients, respectively 4% vs. 2.6%. The LoS and mortality rate were significantly higher in COVID-19E compared to COVID-19WE, non-COVID-19E, and non-COVID-19WE, respectively, 14.1, 28.4%; 5.0, 7.7%; 4.3, 18.8%; 2.4, 3.1% (p<0.05). The discharge rate to SNF and hospice were significantly higher in COVID-19E compared to COVID-19WE, non-COVID-19E and in non-COVID-19WE patients, respectively 21.2, 12.3%; 6.3%, 2.8%; 5.4%,1%; 1.64%, 0.31% (p<0.05).

Our study demonstrated an increased incidence of encephalopathy in COVID-19 compared to non-COVID-19 patients. Encephalopathy in COVID-19 patients is characterized by an increased length of hospitalization, discharge to hospice and SNF, and mortality rate. Work is in progress to investigate specific factors for poor prognosis of encephalopathy in COVID-19 infection using risk stratification machine learning (ML) models.