| CPC G16H 40/20 (2018.01) [G06Q 10/067 (2013.01); G06Q 10/0639 (2013.01); G16H 10/60 (2018.01); G16H 50/70 (2018.01); G16H 80/00 (2018.01)] | 17 Claims |
|
1. A storage media comprising: a non-transitory computer-readable medium having computer-executable instructions embodied thereon that when executed, facilitate performance of a method of predicting, via a decision support system comprising a distributed computing architecture, an estimate of conformity to a medical care performance measure, the method comprising:
identifying a population of patients who received a particular therapy;
identifying a set of patients who received the particular therapy, wherein the set of patients is different than the population of patients;
determining an Electronic Clinical Quality Measures (eCQM) performance measure that identifies a threshold performance based on:
a first number identifying how many of the set of patients experienced a particular outcome in response to the particular therapy; and
a second number of patients in the set of patients;
obtaining a set of patient data for the set of patients, at least a portion of the set of patient data being stored in real-time, wherein the set of patient data includes sets of attribute variables that includes at least one of: an age, gender, treatment location, treating clinician, comorbid condition, prior treatment received, prescribed treatment frequency, prescribed treatment duration, or estimated adherence to a prescribed treatment;
identifying a set of clinicians that treated the set of patients;
receiving a set of clinician data corresponding to the set of clinicians;
distributing and performing parallelized determinations, among a plurality of processors at multiple locations included in the distributed computing architecture, to estimate conformity to the medical care performance measure, wherein the parallelized determinations among the plurality of processors at the multiple locations comprises:
accessing the set of attribute variables associated with the set of patients;
analyzing the set of clinician data and the sets of attribute variables using Bayesian Markov Chain Monte Carlo (MCMC) one-inflated beta regression until model convergence is reached;
from the converged model, extracting a set of one or more regression coefficients and beta distribution parameters;
from the set of one or more regression coefficients and beta distribution parameters, determining those regression coefficients and beta distribution parameters having a statistical significance thereby forming a subset of statistically significant regression coefficients and beta distribution parameters, wherein each of the subset of statistically significant regression coefficients corresponds to a specific attribute variable of the sets of attribute variable; and
based on the subset of statistically significant regression coefficients and beta distribution parameters, estimating, without having to actively monitor the set of patient data, a measure of conformity with the eCQM performance measure; and
automatically displaying, via a display of a user device of the distributed computing architecture, a graphical user interface page that comprises the eCQM performance measure, the measure of conformity with the eCQM performance measure, information characterizing the set of patients, and a recommendation comprising instructions indicating that each of the specific attribute variable has an effect on patients conforming to the eCQM performance measure for the set of patient data, wherein the recommendation further includes policies or methods of other entities that the decision support system has identified as statistically conforming to the eCQM performance measure for the set of patient data.
|