Using Rasch modeling to re-evaluate rapid malaria diagnosis test analyses

Area of Science:

• Medical Diagnostics
• Biostatistics
• Public Health

Background:

• Malaria rapid diagnostic tests (RDTs) are crucial for malaria control.
• Assessing the validity of scoring methods for RDTs is essential for accurate diagnosis.
• The Rasch model offers a framework for evaluating measurement properties.

Purpose of the Study:

• To demonstrate the application of the Rasch model in evaluating malaria RDT scoring.
• To assess the appropriateness of demographic, socioeconomic, and geographic factors in RDT scoring.
• To evaluate RDT data against Rasch model expectations.

Main Methods:

• Utilized data from a baseline malaria indicator survey conducted in Ethiopia (Amhara, Oromiya, SNNP regions) in 2007.
• Applied the Rasch model to analyze the performance of malaria RDT items.
• Assessed reliability, threshold ordering, and differential item functioning.

Main Results:

• The Rasch model analysis indicated high reliability for the malaria RDT.
• Results showed minimal disordering of thresholds, suggesting consistent item performance.
• No evidence of differential item functioning was found, indicating fairness across subgroups.

Conclusions:

• The Rasch model is a suitable tool for assessing the psychometric properties of malaria RDTs.
• Demographic, socioeconomic, and geographic factors align with Rasch model expectations for RDT scoring.
• The findings support the reliability and validity of malaria RDT scoring in the studied Ethiopian regions.