US 2013/0085773 A1 · METHOD FOR GENERATING HEALTHCARE-RELATED VALIDATED PREDICTION MODELS FROM MULTIPLE SOURCES

Abstract

Provided is a method for generating prediction models from multiple healthcare centers. The method allows a third party to use data sets from multiple sources to build prediction models. By entering the data sets in a Model Deconstruction and Transfer (MDT) platform, a healthcare center may provide data to a third party without the need to de-identify data or to physically transfer any identifying or de-identified data from the healthcare center. The MDT platform includes a variable library, which allows the healthcare center to select variables that will be used to generate and validate the prediction model. Also provided is a method for compensating sources that contribute data sets based upon the percentage of clinical data that is used to generate a prediction model.

Claims76

Dependent claims

A method of generating a prediction model for a health outcome of interest comprising the steps of:

(a) providing more than one healthcare centers with a Model Deconstruction and Transfer (MDT) platform comprised of a variable library (VL), wherein each healthcare center enters at least one data set relevant to the health outcome of interest into the MDT platform and selects variables from the VL that are relevant to the health outcome of interest;
(b) generating at least one prediction model (PM0) for each healthcare center from the MDT platform, wherein the at least one PM0 is based upon the selected variables.

A method of generating a prediction model for a health outcome of interest comprising the steps of:

(a) obtaining at least one data set relevant to the health outcome of interest from more than one healthcare centers;
(b) selecting variables from the at least one data set that are relevant to the health outcome of interest;
(c) generating at least one first prediction model (PM0) for each healthcare center based upon differential weighting of the at least one data set variables, wherein the at least one PM0 is generated using a computing device;
(d) generating a Model Component Library (MCL) from each PM0, wherein the MCL comprises components that result from deconstruction of each PM0;
and
(e) generating at least one second prediction model (PM1) from the MCL based upon differential weighting of the MCL components relevant to the health outcome of interest.

3 more elements

A method of compensating a healthcare center based upon data provided by the healthcare center for use in the generation of a prediction model for a health outcome comprising the steps of:

(a) obtaining at least one data set relevant to the health outcome of interest from more than one healthcare centers;
(b) selecting variables from the at least one data set that are relevant to the health outcome of interest;
(c) generating at least one first prediction model (PM0) for each healthcare center based upon the differential weighting of the variables relevant to the health outcome of interest, wherein the at least one PM0 is generated using a computing device;
(d) generating a Model Component Library (MCL) from each PM0, wherein the MCL comprises components that result from deconstruction of each PM0;
and
(e) generating at least one second prediction model (PM1) for the health outcome of interest based upon differential weighting of the MCL components relevant to the health outcome of interest;
(f) compensating the healthcare center according to the MCL components that contribute to the generation of the at least one PM1.

4 more elements

Dependencies are inferred from the claim text.

Method for generating healthcare-related validated prediction models from multiple sources

Life of this patent

Method for generating healthcare-related validated prediction models from multiple sources

Life of this patent

Abstract

Claims76