Method and computer program product for predicting and minimizing future behavioral health-related hospital admissions

摘要

An accurate predictive model that identifies the patient/members within the healthcare system at high risk of hospital admission for a wide range of morbidities, or co-morbidities, and that allows subsequent intervention to manage those patients identified as high risk to an acceptable level. There is a further need for such a predictive model that focuses on specific groupings of conditions, e.g., behavioral health predictive modeling. There is also a need for a method that provides for intervention to manage the risk to the identified patients/members. One embodiment of the present invention discloses and claims a method of high-risk patient identification and management. In one aspect, the inventive method may comprise compiling a listing including all individuals with any primary behavioral health diagnosis over a specified time period; merging the listing with at least one data source to extract at least one behavioral health-related predictive factor; generating, based on at least one predictive model, a predictive output comprising the probability that the individuals listed will require a future behavioral health-related hospital admission; identifying the high-risk individuals from the predictive model output; and intervening with the high risk members to identify and modify, to the extent possible, the risk factors that place the member at high risk.