Why did you develop the CHOSEN Score? Was there a particular clinical experience or patient encounter that inspired you to create this tool for clinicians?
As the COVID-19 pandemic hit Boston in March 2020 and hospitals were overrun, clinicians were forced to make difficult decisions regarding discharge from a monitored setting. There simply was no evidence base. The typical risk scores, such as CURB-65 or q-SOFA, did not seem appropriate. Our hospital was bursting, yet the decision to discharge a patient was highly variable and based simply on gestalt. Our team wanted to pitch in with data to help inform discharge decisions, both in the emergency department and general medical ward.
What pearls, pitfalls and/or tips do you have for users of the CHOSEN Score? Do you know of cases when it has been applied, interpreted, or used inappropriately?
Users should be aware that our dataset was based solely on patients in the early days of the pandemic during the first surge. Therefore, CHOSEN might best be applied during a surge when a hospital is beyond its capacity. Users should also keep in mind that our COVID-19 treatments and pathways have in some ways changed since the beginning of the pandemic. We’ve found that having our small figure for reference can make it easy to remember CHOSEN’s inputs and interpret its outputs.
What recommendations do you have for practitioners once they have applied the CHOSEN Score?
If you don’t have a recent albumin level available, consider using the modified CHOSEN score that can be calculated without laboratory values. This model does not perform quite as well but can be useful particularly in the emergency department where a clinician may not obtain laboratory testing. As always, clinical decision rules do not replace clinical judgment.
