A Dashboard for Latent Class Trajectory Modeling: Application in Rheumatoid Arthritis

Area of Science:

• Biomedical Informatics
• Clinical Research Methods
• Rheumatology

Background:

• The shift towards precision medicine necessitates tools for identifying patient subgroups benefiting from specific interventions.
• Clinicians require accessible methods to explore patient data and derive population-specific insights.
• Rheumatoid arthritis (RA) management benefits from tailored treatment approaches.

Purpose of the Study:

• To present a novel dashboard application for exploring patient subgroups based on longitudinal treatment response.
• To enable clinicians to identify patient segments for precision treatment strategies.
• To apply latent class mixed modeling for analyzing treatment effectiveness in rheumatoid arthritis.

Main Methods:

• Development of an interactive dashboard application using R Shiny.
• Application of latent class mixed modeling (LCMM) to analyze patient data.
• Utilizing an observational study of moderate to severe rheumatoid arthritis patients on first-line biologic therapy.

Main Results:

• The dashboard effectively visualizes patient subgroups with varying longitudinal treatment responses.
• Latent class mixed modeling identified distinct patient trajectories under biologic treatment.
• The approach demonstrated utility in an observational study of rheumatoid arthritis.

Conclusions:

• The developed dashboard is a valuable tool for clinicians in the era of precision medicine.
• It facilitates the identification of patient subgroups for targeted therapies, improving treatment efficacy.
• This methodology offers a pathway for generating new inferences in clinical research, particularly in rheumatology.