Indirect Reference Interval Determination From Laboratory Databases
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining reference intervals in clinical laboratory testing are inaccurate due to reliance on outdated studies, lack of large sample sizes, and subjectivity in data manipulation, making it difficult to account for modern testing methodologies and population differences.
Innovation Solution
A method involving data pooling, transformation, and linear regression analysis of existing laboratory databases to determine a reference interval, allowing for a more robust and reliable estimation of reference intervals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If de novo reference interval studies are conducted using conventional direct donor sampling method, then reference intervals can be established, but the studies are expensive and have limitations including difficulty in recruiting healthy subjects and obtaining informed consent
Solution Approach 1:
The patent inverts the conventional approach by using indirect sampling from existing laboratory databases rather than directly recruiting healthy donors. This reversal allows reference intervals to be determined from routine clinical data without the need for subject recruitment, informed consent, or health screening, thereby resolving the contradiction between reliability and ease of implementation
Solution Approach 2:
The patent uses copies of existing laboratory data stored in databases rather than original data from controlled studies. By analyzing pooled data from routine testing, the method replicates the reference interval determination process without requiring actual healthy subject recruitment, eliminating the logistical and ethical barriers while maintaining statistical validity
2Reliability
If healthy reference populations are recruited with defined criteria, then reference intervals can be determined, but the sample sizes are relatively low (about 100-150 individuals) resulting in lacking statistical power
Solution Approach 1:
The patent merges data from multiple sources and populations into a single pooled database. By combining routine laboratory data from many different patients over time, the method achieves sample sizes in the thousands, far exceeding the 100-150 subjects typical of direct sampling studies, thereby resolving the contradiction between reliability and quantity through data aggregation
3Reliability
If manual plotting and visual assessment methods are used for reference interval estimation, then reference intervals can be determined, but the process is subjective and lacks robustness
Solution Approach 1:
The patent replaces the manual mechanical process of plotting data on graph paper and visual assessment with automated computer-based statistical analysis. The system uses software to pool data, plot cumulative frequencies, apply transformations, and calculate reference intervals objectively, eliminating subjectivity while streamlining the complexity through automation rather than increasing it
Data Source
AI summary
The invention relates to methods for indirectly determining clinical laboratory reference intervals. In one aspect, a reference interval is determined using all measurements for a given analyte stored in a large existing database. In other aspects, a characteristic of a subject is used to select a reference population for inclusion in reference interval calculations. In other aspects, the invention provides methods for changing treatment plan, diagnosis, or prognosis for an individual subject based on differences between the new reference interval and a previously utilized reference interval. In other aspects, the invention provides systems and computer readable media for indirectly determining reference intervals.