| CPC G16H 50/50 (2018.01) [G16H 50/30 (2018.01)] | 20 Claims |
|
1. A computer-implemented process for performing a clinical assessment, comprising:
implementing a data abstraction process that receives as input, electronic patient data about a patient of interest from a first data source, and transforms at least a portion of the received input to standardized format patient data for the patient of interest, which is stored in a second data source;
matching the standardized format patient data for the patient of interest to a set of risk variables, where the risk variables include at least one variable classified into a baseline group composed of risk variables that are non-modifiable based on provided medical care of a corresponding patient, and at least one variable classified into a dynamic group composed risk variables that are modifiable based on the provided medical care of the corresponding patient, where the set of risk variables are selected to be predictive of an adverse outcome of interest associated with a physiological condition;
extracting by a processor, a set of outcome models, the set of outcome models corresponding to the adverse outcome type associated with the physiological condition of interest, the set of outcome models including a baseline outcome likelihood model extracted from a first computer memory and a dynamic outcome likelihood model extracted from a second computer memory, wherein a model of the set of outcome models is generated using a first training data set comprising electronic patient training data transformed from a proprietary format to a standardized generic data format and extracted from a computer storage device, where the selection process automatically selects risk variables that are classified into the baseline group and the dynamic group;
applying the set of risk variables to the set of outcome models to generate a baseline outcome likelihood and a dynamic outcome likelihood;
aggregating patient histories for groups of patients similar to the patient of interest;
determining whether at least one of the baseline outcome likelihood and dynamic outcome likelihood exceeds a predetermined threshold;
facilitating an intervention in treating the physiological adverse outcome type associated with the physiological condition of interest by including an attribution associated with at least one risk factor causing the predetermined threshold to be exceeded;
verifying an effectiveness of the intervention based on the aggregated patient histories for the groups of patients similar to the patient of interest; and
outputting an electronic alert to a remote electronic device to alert a clinician if at least one of the baseline outcome likelihood and dynamic outcome likelihood exceeds the predetermined threshold, the alert providing the intervention.
|