Electronic Message Modeling for Prescription Adherence Estimation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing systems lack an efficient method to estimate target quantitative measures for binary categorical indicators, such as prescription drug adherence, based on historical electronic messages, which are crucial for optimizing healthcare transactions and adherence rates.

Innovation Solution

A system and method that utilizes historical electronic messages to parse, partition, and model data by geographic areas and demographic indicators, estimating a quantitative predictor of adherence levels and adjusting measures based on entity criteria to achieve target adherence rates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If historical electronic messages are used to estimate target quantitative measures, then prediction accuracy of adherence levels is improved, but system complexity increases due to data parsing, partitioning, and modeling requirements

Engineering Contradiction:

Improveprediction accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent partitions historical electronic message data into multiple segments based on geographic areas, allowing the system to process and analyze data in manageable portions rather than as a single large dataset. This segmentation enables parallel processing and reduces the computational burden on individual processing units while maintaining overall prediction accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a service provider switch as an intermediary component that collects, stores, and processes historical electronic messages between various network entities. This intermediary handles the complex data parsing and modeling operations, shielding end users from system complexity while enabling accurate predictions through centralized data management.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If data is partitioned by geographic areas and modeled with multiple variables, then adherence prediction precision is improved, but processing time increases

Engineering Contradiction:

Improveadherence prediction precisionVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary actions by pre-processing historical electronic messages during off-peak periods, parsing and partitioning data into geographic segments in advance. This allows the system to have pre-computed models ready for rapid deployment during peak usage times, reducing real-time processing requirements while maintaining high prediction precision.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements a multi-level modeling approach where basic adherence predictions are generated using partial data sets for quick results, while more comprehensive models with full geographic and demographic variables are used selectively when higher precision is required. This partial action strategy balances processing time constraints with prediction accuracy needs.

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If multiple quantitative measures are estimated for different entity criteria, then optimization capability is improved, but computational resource consumption increases

Engineering Contradiction:

Improveoptimization capabilityVSAvoidcomputational resource consumption

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent applies local quality by determining second quantitative measures tailored to specific entity criteria such as plan deductibles, sponsor types, and geographic regions. Instead of computing all possible measures for all entities uniformly, the system selectively estimates measures only where needed based on entity-specific characteristics, reducing overall computational resource consumption while maintaining optimization capability.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent dynamically adjusts computational parameters based on entity criteria, changing the level of modeling detail and data granularity required for different optimization scenarios. For example, simpler models are used for routine queries while more complex multi-variable models are activated only when entity-specific optimization is required, thereby managing computational resource consumption efficiently.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12518304B1Method, apparatus, and computer program product for estimating a target quantitative measure based upon historical electronic messages

Publication Date: 2026.01.06 MCKESSON CORPORATION

AI summary

A method, apparatus and computer program product are provided to estimate at least one target quantitative measure and a second quantitative measure based upon historical electronic messages. The historical electronic messages may be partitioned based on geographic areas. The target quantitative measure may be estimated based upon respective binary categorical indicators, and respective quantitative measures of the historical electronic messages. The target quantitative measure may be adjusted dependent upon a quantitative criterion of an entity. The target quantitative measure may be used to price prescription drugs and/or the like, and may impact prescription adherence and/or revenue of a payer and/or a drug manufacturer.