Computerized system and method for identifying members at high risk of falls and fractures

Abstract

Systems and methods for automated interventions to persons identified as being of risk of falling are provided. A subset of members is identified which are associated with at least one of a plurality of falls predictors. At least one falls prediction algorithm is applied to a subset of said medical claims data associated with the subset of members to generate a falls risk score for each of member of the subset. At least one intervention is assigned to each of member of the subset having an assigned risk score above any of several predetermined risk score thresholds which are automatically and electronically initiated based, at least in part, on member data.

What is claimed is: 
     
         1 . A system for automatically assigning risk-level appropriate interventions to persons identified as being of risk of falling, said system comprising:
 one or more databases, said databases affiliated with a medical insurance provider and comprising medical claims data for a plurality of members associated with said medical insurance provider;   one or more computing devices in electronic communication with said one or more databases and comprising one or more processors and one or more electronic storage devices, said one or more electronic storage devices comprising software instructions, which when executed, configure said one or more processors to:
 identify a subset of said members associated with at least one of a plurality of falls predictors; 
 apply at least one falls prediction algorithm to a subset of said medical claims data associated with said subset of said members to generate a falls risk score for each of said subset of said members; 
 assign at least one of a plurality of interventions to each of said subset of said members having an assigned risk score above one of a number of predetermined risk score thresholds (the “identified at-risk members”); and 
 automatically and electronically initiate the assigned interventions based, at least in part, on member data stored at said one or more databases for said identified at-risk members. 
   
     
     
         2 . The system of  claim 1  wherein:
 said interventions comprise a phone assessment, an in-home assessment, electronic contact, and enrolment in a clinical program. 
 
     
     
         3 . The system of  claim 1  wherein:
 said one or more electronic storage devices comprise additional software instructions, which when executed, configure said one or more processors to generate a graphical report comprising identifying information and said risk score for each of said identified at-risk members. 
 
     
     
         4 . The system of  claim 3  wherein:
 said graphical report stratifies said identified at-risk members by said risk score. 
 
     
     
         5 . The system of  claim 1  wherein:
 said risk score is specific to at least age and gender. 
 
     
     
         6 . The system of  claim 1  wherein:
 said one or more electronic storage devices comprise additional software instructions, which when executed, configure said one or more processors to evaluate said medical claims data for said plurality of members associated with said medical insurance provider to develop said at least one falls prediction algorithm. 
 
     
     
         7 . The system of  claim 6  wherein:
 said at least one falls prediction algorithm is developed using a modeling technique comprising one or more of: decision tree, logistic regression, artificial neural networks, and ensemble. 
 
     
     
         8 . The system of  claim 6  wherein:
 said at least one falls prediction algorithm is developed, at least in part, by:
 analyzing said medical claims data for an initial subset of said members having at least one falls model trigger present; 
 pre-processing said medical claims data for said initial subset of said members; 
 extracting features from the medical claims data for said initial subset of said members to generate at least one initial falls prediction algorithm; and 
 providing a plurality of training conditions to said at least one initial falls prediction algorithm to develop said at least one falls prediction algorithm. 
 
 
     
     
         9 . The system of  claim 7  wherein:
 said features are extracted using temporal feature extraction, wherein said falls predictors are selected to correspond with the extracted features; 
 said pre-processing of said medical claims data comprises one or more of: variable selection, principle component analysis, and clustering; and 
 said training conditions comprise one or more of: an unintentional fall, a skull fracture, a neck fracture, a trunk fracture, an upper limb fracture, a lower limb fracture, and a dislocation of bones. 
 
     
     
         10 . The system of  claim 7  wherein:
 said at least one falls model trigger comprises one or more of: alcohol abuse, Alzheimer's disease, blood vessel injury, cognitive dysfunction, concussion, dementia, dialysis, difficulty walking, epilepsy and convulsion, face, eye, or neck contusion, face, neck, or scalp injury, fracture, hallucinations, hip, knee or joint pain, hypotension, motor problems, muscle weakness, nerve or spinal injury, obesity, one or more previous falls, other injury, Parkinson's disease, and stroke. 
 
     
     
         11 . The system of  claim 7  wherein:
 said falls predictors comprise one or more of: injury/poisoning incidence count, a lower limb fracture, head, neck, or spine trauma incidence count, previous falls, an upper limb fracture, a neck or trunk fracture, a dislocation, a narcotic prescription, age, a hospital emergency room visit, obesity, governmental agency health score, a bone disorder, gender, race, an anti-depressant prescription, and an anti-hypertensive prescription. 
 
     
     
         12 . The system of  claim 6  wherein:
 said one or more databases comprise consumer data for each of said plurality of said members associated with said medical insurance provider; and 
 said at least one falls prediction algorithm is developed, at least in part, with said consumer data for said subset of said members. 
 
     
     
         13 . The system of  claim 12  wherein:
 said consumer data comprises one or more of: demographic data, geographic data, and financial data. 
 
     
     
         14 . The system of  claim 6  wherein:
 said one or more databases comprise pharmacy claims data for each of said plurality of said members associated with said medical insurance provider; and 
 said at least one falls prediction algorithm is developed, at least in part, with said pharmacy claims data for said subset of said members. 
 
     
     
         15 . The system of  claim 1  wherein:
 said risk score is configured to reflect a risk of falling within a specified period; and 
 said medical claims data is limited to a predetermined historical period. 
 
     
     
         16 . The system of  claim 15  wherein:
 the specified period is 12 months; and 
 said predetermined historical period is one month prior to a specific date. 
 
     
     
         17 . A system for automatically assigning risk-level appropriate interventions to persons identified as being of risk of falling, said system comprising:
 one or more databases, said databases affiliated with a medical insurance provider and comprising medical claims data for a plurality of members associated with said medical insurance provider;   one or more computing devices in electronic communication with said one or more databases and comprising one or more processors and one or more electronic storage devices, said one or more electronic storage devices comprising software instructions, which when executed, configure said one or more processors to:
 identify a subset of said members associated with at least one of a plurality of falls predictors within a predetermined historical period; 
 develop at least one falls prediction algorithm based, at least on part, on a subset of said medical claims data associated with said subset of said members within the predetermined historical period using a modeling technique comprising one or more of: decision tree, logistic regression, artificial neural networks, and ensemble, wherein said at least one falls prediction algorithm is configured to reflect a risk of falling within a specified future period; 
 identify a second subset of said members associated with at least one of a plurality of falls predictors within a second predetermined historical period forward in time from said predetermined historical period; 
 pre-process a second subset of said medical claims data associated with said second subset of said members utilizing one or more of: variable selection, principle component analysis, and clustering; 
 apply said at least one falls prediction algorithm to the second subset of said medical claims data associated with said second subset of said members to generate a falls risk score for each of said second subset of said members reflecting a risk that one of said members in said second subset falls within the specified future period; 
 assign at least one of a plurality of interventions to each of said second subset of said members having an assigned risk score above one of a number of predetermined risk score thresholds (the “identified at-risk members”); 
 generate a graphical report comprising identifying information and said risk score for each of said identified at-risk members where said identified at-risk members are stratified into various risk groups by said risk score; and 
 automatically and electronically initiate the assigned interventions based, at least in part, on member data stored at said one or more databases for said ide identified at-risk members, wherein said interventions comprise one or more of: a phone assessment, an in-home assessment, electronic contact, and enrolment in a clinical program 
   
     
     
         18 . A computer-implemented system for identifying a member of a health insurance market-based member population at risk for falling within a predetermined time period, the system comprising:
 (a) one or more computing devices storing:
 (1) falls model triggers comprising one or more of: alcohol abuse, Alzheimer's disease, blood vessel injury, cognitive dysfunction, concussion, dementia, dialysis, difficulty walking, epilepsy and convulsion, face, eye, or neck contusion, face, neck, or scalp injury, fracture, hallucinations, hip, knee or joint pain, hypotension, motor problems, muscle weakness, nerve or spinal injury, obesity, one or more previous falls, other injury, Parkinson's disease, and stroke; 
 (2) falls predictors comprising one or more of: injury/poisoning incidence count, a lower limb fracture, head, neck, or spine trauma incidence count, previous falls, an upper limb fracture, a neck or trunk fracture, a dislocation, a narcotic prescription, age, a hospital emergency room visit, obesity, governmental agency health score, a bone disorder, gender, race, an anti-depressant prescription, and an anti-hypertensive prescription; and 
   (b) one or more computing devices executing instructions to:
 (1) receive member data and consumer data for the entire health insurance market-based member population, wherein said member data comprises data selected from the group consisting of: medical claims data and pharmacy claims data, wherein said consumer data comprises one or more of: demographic data, geographic data, and financial data; 
 (2) analyze said received member data for the entire health insurance market-based member population to identify a subset of members within said health insurance market-based member population having one or more of said falls model triggers present in said member's data; 
 (3) process the member data and said consumer data for said subset of members using an algorithm utilizing one or more of: variable selection, principle component analysis, and clustering; 
 (4) extract features from the member data for said subset of members by temporal feature extraction, wherein said falls predictors are selected to correspond with the extracted features; 
 (5) provide a plurality of training conditions to a computing device that comprises a falls predictive model, said training conditions comprising one or more of: an unintentional fall, a skull fracture, a neck fracture, a trunk fracture, an upper limb fracture, a lower limb fracture, and a dislocation of bones; 
 (6) develop the falls predictive model using a modeling technique comprising one or more of: decision tree, logistic regression, artificial neural networks, and ensemble; 
 (7) provide said member data and said consumer data for said subset of members to the computing device that comprises said falls predictive model; 
 (8) receive a calculated falls risk score from the computing device that comprises said falls predictive model, wherein said falls risk score represents the likelihood that the respective member of said subset of members will visit an emergency room as a result of experiencing a fall within the predetermined time period, and wherein said calculated falls risk score is determined at least in part based on the presence or absence of each of said falls predictors in said member data for the respective member; 
 (9) sort the received calculated falls risk score into one of a plurality of groups according to a severity level indicated by the received calculated falls risk score; 
 (10) assign a clinical program or intervention for each of the members in said subset of members, wherein said assignment is determined based on the group into which said member's calculated falls risk score has been sorted, where said intervention is adapted to reduce the member's calculated falls risk score; and 
 (11) enroll said member in said assigned clinical program or intervention. 
   
     
     
         19 . The computer-implemented system of  claim 18  wherein:
 the temporal feature extraction is accomplished by executing software instructions which cause the one or more computing devices to:
 gather member data for each feature for a time period prior to a date in question, wherein said member data comprises a number of events, each of which is associated with a particular time; 
 divide the time period into a number of equal intervals spanning the time period; 
 sort the gathered data such that each event is sorted into the interval corresponding with the particular time for the respective event; 
 sum the data falling within each interval; 
 assign a weighting to each interval in decreasing fashion such that the interval temporally closest to the date in question gets the highest weight and the interval temporally farthest from the date in question gets the lowest weight; 
 multiply the summed value for each interval by the weighting for the respective interval to determine a weighted sum for each interval; and 
 sum the weighted sums to determine a cumulative sum, wherein the cumulative sum is utilized to determine the calculated falls risk score. 
 
 
     
     
         20 . The computer-implemented system of  claim 18  wherein:
 the temporal feature extraction is accomplished by executing software instructions which cause the one or more computing devices to:
 gather member data for each feature for a time period prior to a date in question, wherein said member data comprises a number of events, each of which is associated with a particular time; 
 divide the time period into a number of equal intervals spanning the time period; 
 sort the gathered data such that each event is sorted into the interval corresponding with the particular time for the respective event; 
 fit a predictive model to determine a temporal feature value for each extracted feature; and 
 weight each extracted feature with the respective temporal feature value, wherein the weighted values are utilized to determine the calculated falls risk score.