Predicting and preventing caregiver musculoskeletal injuries
AssigneeCerner Innovation, Inc.
InventorDouglas S. McNair
Life of this patent
Estimating the term…
Abstract
Systems, methods and computer-readable media are provided for determining the modality for transferring, lifting, or repositioning (TLR) a human patient in a health care setting contexts. In some cases, a model-based recursive partitioning and Bradley-Terry regression is applied, which may be optionally parallelized so as to determine statistical associations with various factors, such as caregiver attributes, care venue, and patient attributes. One embodiment determines a Bradley-Terry regression model from the recursive partitioning which may be incorporated into a TLR selection decision-support tool or otherwise utilized to identify the optimal modality.
Claims19
A system for providing an optimal transferring, lifting, and/or repositioning (TLR) modality option for a particular context to a computing device with a user interface, the system comprising:
- one or more processors;
- and
- computer-readable media having computer-usable instructions embodied thereon that, when executed by the one or more processor, cause the one or more processor to:
- receive raw count TLR usage information and context attributes associated with the TLR usage information, at least a portion of the context attributes including an explanatory analysis of context associated with the variable, wherein the explanatory analysis of context associated with the variable includes a history of previous injury;
- determine one or more statistical distributions that fit the TLR usage information;
- estimate a set of parameters for the one or more statistical distributions;
- determine a set of distribution quantiles for the set of parameters;
- based on the quantiles, transform the TLR usage information into a paired-comparison of TLR modalities;
- determine one or more regression models by performing model-based recursive partitioning on the context attributes, each model having a set of model coefficients;
- determine model convergence and model stability;
- for each model of the one or more regression models, determine a statistical significance for the set of model coefficients;
- based at least on the statistical significance for each of the one or more regression models, determine a model from the one or more regression models to be utilized in a decision-support application for determining an optimal TLR modality option for the particular context;
- utilize the decision-support application for automatically determining the optimal TLR modality option for the particular context;
- and
- provide to the computing device having the user interface the optimal TLR modality option, wherein the computing device is associated with the decision-support application.
One or more computer-readable media having computer-usable instructions embodied thereon that, when executed, enable a processor to perform a method for determining an optimal transferring, lifting, and/or repositioning (TLR) modality option for a particular context, the method comprising:
- receiving TLR usage information and context attributes for a historical population of patients and caregivers, at least a portion of the context attributes including an explanatory analysis of context associated with the variable, wherein the explanatory analysis of context associated with the variable includes a history of previous injury;
- determining one or more statistical distributions that fit the TLR usage information;
- estimating a set of parameters for the one or more statistical distributions;
- determining a distribution grouping for the set of parameters;
- determining one or more regression models, each model of the one or more regression models having a set of model coefficients;
- for each of the model of the one or more regression models, determining a statistical significance for the set of model coefficients;
- based on at least on the statistical significance for each model of the one or more regression models, utilizing a model from the one or more regression models in a decision-support application for automatically determining an optimal TLR modality option for a particular context;
- and
- causing to be presented on a user interface of a device associated with the decision-support system the optimal TLR modality option.
A computerized method for providing a transferring, lifting, and/or repositioning (TLR) option for a particular context to a user device having a user interface, the method comprising:
- receiving raw count TLR usage information and context attributes having a set of explanatory attribute variables that provide an explanatory analysis of context associated with the variable, wherein the TLR usage information and context attributes are received for a historical population of caregivers that have performed TLR methods on the historical population of patients, and further wherein the explanatory analysis of context associated with the variable includes a history of previous injury;
- determining one or more statistical distributions that fit the TLR usage information;
- estimating a set of parameters for the one or more statistical distributions;
- determining a distribution grouping for the set of parameters;
- based on the distribution grouping, transforming the TLR usage information into a paired-comparison of TLR modalities, the paired-comparison of TLR modalities having more than one possible outcome;
- determining one or more regression models by performing model-based recursive partitioning on the context attributes and the set of explanatory attribute variables, each model of the one or more regression models having a set of model coefficients;
- determining model convergence and model stability for each of the one or more regression models, wherein the model convergence and model stability is determined using a structural change method;
- for each model of the one or more regression models, determining a statistical significance for the set of model coefficients;
- based on at least on the statistical significance for each model of the one or more regression models, determining a model from the one or more regression models to be utilized in a decision-support application for determining an optimal TLR modality option for the particular context;
- utilizing the decision-support application for automatically determining the optimal TLR modality option for the particular context;
- and
- providing to the computing device having the user interface the optimal TLR modality option, wherein the computing device is associated with the decision-support application.
Dependencies are inferred from the claim text.