Overview of tool
Johns Hopkins used an on-prem, machine learning tool to predict post-acute care need. It enables clinicians to predict the likelihood of post-acute placement need for inpatients within 24-48 hours of admission. As a result, the model is well-positioned to reduce overall LoS by allowing faster, more proactive planning for patient discharge.