Can short PROMs support valid factor-based sub-scores? Example of COMQ-12 in chronic otitis media

Area of Science:

• Psychometrics
• Clinical Outcome Measurement
• Audiology

Background:

• Interpretable factor solutions for questionnaires often justify sub-scores.
• However, factor solutions do not guarantee criterion validity for clinical applications.
• The psychometric properties of the 12-item Chronic Otitis Media Questionnaire (COMQ-12) were previously documented.

Purpose of the Study:

• To appraise the criterion validity of COMQ-12 sub-scores in adult chronic otitis media patients.
• To evaluate the clinical utility of factor-based sub-scores derived from the COMQ-12.

Main Methods:

• Compared factor solutions for the COMQ-12 in 246 adult chronic otitis media patients.
• Utilized clinical data (audiometry, oto-microscopic findings) and the SF-36 as external criteria.
• Assessed convergent validity for total and sub-scores against established measures.

Main Results:

• A 3-factor solution was favored but had insufficient item support and dominant first principal component.
• A bi-factor solution was optimal, but only the weighted total score met the convergent validity standard (r=0.50).
• Specific sub-scores ('Ear discharge symptoms', 'Hearing') showed poor correlation with clinical findings and audiometric thresholds.

Conclusions:

• The COMQ-12 total score demonstrates acceptable content validity for general clinical use.
• The limited item set and factor structure do not adequately support criterion-valid factor-based sub-scores.
• Profile sub-scoring is discouraged until robust convergent and divergent criterion validities are established.