A latent variable framework for modeling dyadic measures in research on shared decision-making.

Area of Science:

• Psychometrics
• Health Services Research
• Biostatistics

Background:

• Assessing patient and physician agreement is crucial for effective healthcare.
• Traditional methods may not fully capture the nuances of dyadic perceptions.
• Shared decision-making (SDM) is a key component of patient-centered care.

Purpose of the Study:

• To present a novel statistical framework for analyzing correspondence between patient and physician perceptions.
• To apply this framework to shared decision-making (SDM) in chronic disease consultations.
• To combine multitrait-multimethod and measurement invariance methodologies for robust analysis.

Main Methods:

• Latent variable modeling was employed to investigate perception correspondence.
• Multitrait-multimethod and measurement invariance methodologies were integrated.
• Data from 285 primary care consultations using the SDM-Q-9 and SDM-Q-Doc questionnaires were analyzed.

Main Results:

• Patients and physicians agree on the fundamental aspects of SDM.
• Discrepancies were observed in the perceived extent of SDM presence during consultations.
• The statistical framework offered unique insights beyond traditional approaches.

Conclusions:

• The proposed statistical framework enhances understanding of stakeholder perception correspondence.
• The approach provides valuable data for improving SDM in clinical practice.
• Generalizability may be limited by the data-driven nature and sample size requirements (≥200 cases).