A history of pneumonia is one of the top risk factors for predicting COVID-19 mortality, a Harvard analysis of electronic health record data finds.
The investigation looked at risk factors across age groups in 17,000 people. In patients aged 65 to 85 years, diseases of the pulmonary system were highly predictive of a COVID-19 patient’s death. Aside from pneumonia, these included chronic obstructive pulmonary disease and lung cancer, reported Shawn N. Murphy, M.D., Ph.D., and colleagues in npj Digital Medicine.
For patients between the ages of 45 and 65 years, notable risk factors included a history of diabetes with complications and breast and prostate cancer.
Across all age groups, the top risk factors spanned several different organ systems, the researchers said. After age (by far the most outstanding predictor of COVID-19 death) and history of pneumonia, were chronic kidney disease, diabetes mellitus type 2 with complications, and heart failure.
The study’s model of COVID-19 risk based on historic EHR data mirrored the findings of more complex models and supports the ability to compute precise risk scores exclusively based on the EHR, the authors contend. This may be “crucial for effectively allocating and distributing resources, such as prioritizing vaccination among the general population,” they said.
Currently, for example, vaccination recommendations by the Centers for Disease Control and Prevention rely on generalizations based on age, occupation and the living environment. “Such recommendations are imprecise and do not leverage the wealth of information in the EHR,” according to the authors.
In contrast, an EHR model such as the one used in the study could assess a single individual’s risk within a healthcare system based on demographics, diagnoses and medications, Murphy proposed.
A history of pneumonia, for example, rarely is asked about in typical epidemiology studies, they noted.
