Predictive Models for Undocumented Comorbidity Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional healthcare systems based on a fee-for-service model lack financial incentives for efficient service management and patient health outcomes, leading to inefficiencies, spiraling costs, and miscommunication among healthcare entities, which can result in life-threatening conditions and billing inaccuracies due to undocumented comorbidities.

Innovation Solution

A method and system using predictive models to identify undocumented comorbidities by generating patient risk scores from aggregated and individual data, correcting billing inaccuracies, and providing clinical decision-making support through an integrated care system.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional fee-for-service model is used, then healthcare providers receive compensation per treatment or service, but this leads to increased healthcare costs and inefficient service management

Engineering Contradiction:
Improveservice management efficiencyVSAvoidhealthcare costs
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The system implements feedback mechanisms by continuously monitoring patient data, comorbidity documentation, and billing information. Predictive models analyze this feedback to identify undocumented comorbidities, which then triggers corrections to billing and treatment plans, creating a closed-loop system that continuously improves efficiency and reduces costs

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system enables self-service through automated predictive modeling and risk score generation. The AI-driven platform automatically identifies undocumented comorbidities and generates billing corrections without requiring manual review of every patient record, allowing the system to serve itself in detecting and correcting errors

Inventive Principle:
Principle #25Self-service

2Adaptability or versatility

If patient data is spread across multiple healthcare entities, then comprehensive care can be provided, but this leads to miscommunication and lack of coordination among providers

Engineering Contradiction:
Improvecare coordinationVSAvoidcommunication accuracy
Core Design Contradiction:
Adaptability or versatilityVSLoss of information

Solution Approach 1:

The system merges data from multiple healthcare entities into a unified analysis framework. By aggregating patient information across different providers and facilities, the predictive models can identify patterns and undocumented comorbidities that would be invisible in isolated records, thereby improving communication and coordination

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The AI-driven platform acts as an intermediary that processes and standardizes data from various healthcare entities. It translates diverse data formats into unified risk scores and comorbidity identifiers, facilitating accurate information exchange and coordination among providers

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If comprehensive patient evaluation is performed to identify undocumented comorbidities, then billing accuracy improves, but this increases time and resource consumption

Engineering Contradiction:
Improvebilling accuracyVSAvoidevaluation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary action by continuously analyzing patient data in the background to generate risk scores and identify potential undocumented comorbidities before billing occurs. This proactive approach allows billing corrections to be made automatically without requiring time-consuming manual evaluations

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system applies partial action by focusing evaluation resources only on patient records with high risk scores for undocumented comorbidities. Rather than reviewing all patient records equally, the predictive models identify and prioritize only those cases most likely to contain errors, reducing overall evaluation time while maintaining high billing accuracy

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3803887B1Systems and methods for identifying comorbidities
Publication Date: 2025.09.10 FRESENIUS MEDICAL CARE HOLDINGS INC
  • EP3803887B1 patent drawingFigure 1A
  • EP3803887B1 patent drawingFigure 1B
  • EP3803887B1 patent drawingFigure 1C

AI summary

A method and system for identifying an undocumented comorbidity condition in a patient. In one embodiment, the method includes extracting patient data from one or more databases corresponding to a pool of patients receiving treatment; using one or more predictive models with the extracted patient data to generate, for each of the patients in the pool of patients, a respective patient risk score for having an undocumented comorbidity condition; identifying a subset of the pool of patients having a respective patient risk score that is higher than a predetermined threshold value; and based on the identified subset of the pool of patients, identifying one or more patients for additional evaluation, additional review, or combinations thereof.