De-identified Mortality Indicator System Using Encrypted Person Tokens
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional healthcare data systems are limited in identifying deceased individuals due to the presence of protected health information (PHI) in mortality data, which is a HIPAA violation, preventing the incorporation of mortality data into healthcare data sets.
Innovation Solution
A system and method that de-identifies mortality data by removing PHI and personal identification information, assigning unique encrypted person tokens, and merging it with de-identified healthcare data sets, allowing for the matching of mortality data with healthcare data without exposing protected information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If mortality data is incorporated into healthcare data sets, then data completeness and analytical value are improved, but data privacy and security are compromised due to exposure of PHI
Solution Approach 1:
The patent extracts and removes protected health information (PHI) and personal identification information from mortality data records. This extraction process separates the identifying elements from the mortality indicators, allowing the mortality data to be incorporated into healthcare data sets without exposing sensitive personal information.
Solution Approach 2:
The patent introduces unique encrypted person tokens as intermediary elements that enable matching between mortality data and healthcare data sets without exposing underlying personal information. These tokens act as mediators that preserve data linkages while maintaining privacy, allowing analytical value to be retained without direct exposure of PHI.
2Reliability
If PHI is removed from mortality data to ensure compliance, then data privacy is protected, but the ability to match and merge data sets is lost
Solution Approach 1:
The patent employs unique encrypted person tokens as intermediary elements that enable matching between mortality data and healthcare data sets without exposing underlying personal information. These tokens preserve the ability to link records across data sets while maintaining HIPAA compliance, as they do not constitute PHI themselves.
Solution Approach 2:
The patent creates encrypted token copies that represent individual identifiers without containing the actual personal information. These token copies can be used for matching and merging operations, preserving data linkages while the original PHI remains protected and separate from the merged data sets.
3Adaptability or versatility
If conventional data merging techniques are used, then data integration is achieved, but data security requirements cannot be met due to reproduction of protected information
Solution Approach 1:
The patent extracts protected health information from mortality data before integration, removing the elements that would violate security requirements. This extraction enables data integration to proceed with non-PHI elements while the removed PHI is not reproduced in the merged data sets.
Solution Approach 2:
The patent uses encrypted person tokens as intermediary elements that enable data matching and integration without requiring the reproduction of protected health information. These tokens facilitate versatile data integration while maintaining security compliance, as they can be used for merging operations without containing actual PHI.
Data Source
AI summary
A method and system for identifying individuals in a healthcare dataset who are likely deceased without exposing protected health information to users. The present invention assembles mortality data from different sources, de-identifies this data by removing or modifying all elements regarded as protected health information, and adds a unique encrypted person token to each record. The tokenized mortality data is merged with other healthcare data sets that have likewise been de-identified and tokenized by matching the unique person tokens in data sets against each other. The resulting merged data sets include an indicator of mortality, a uniqueness score giving the likelihood that the person token is unique, and a death validity score giving a measure of confidence that the person is actually deceased.