Cross-specialty PROMIS-global health differential item functioning

Purpose To investigate the functioning of the PROMIS-Global Health (PROMIS-GH) across clinical setting, patient age, and medical complexity by evaluating differential item functioning (DIF) within the Global Physical Health (GPH) and Global Mental Health (GMH) domains. To our knowledge, no study demonstrates lack of differential item functioning (DIF) for PROMIS-GH across these populations. We hypothesize that the PROMIS-GH domains of GMH and GPH will perform similarly when compared across these populations. Methods Seven thousand nine hundred and seventy four complete PROMIS Global Health measures were retrospectively analyzed using the 'Lordif' package on the R platform. DIF was investigated for both GMH and GPH across clinical environment (Orthopedic Surgery, Family Medicine, & Internal Medicine), age group (<= 53, > 53-66, > 66), and Charlson Comorbidity Index (CCI:0, CCI:1, CCI:2 +) using quasi Monte Carlo estimation. To assess the significance of DIF, Wald tests were used with the Benjamini & Hochberg procedure. Results No items contained in the GMH or GPH demonstrated DIF across age groups, medical complexity, or clinical environment. Conclusion Items assessing the domains of GMH and GPH within the PROMIS-GH function comparably across treatment setting, age category, and medical comorbidities. The PROMIS-Global Health holds potential to facilitate interdisciplinary patient care and patient optimization prior to surgical intervention.

Automated summary

This study investigated the functioning of PROMIS-GH across different clinical settings, patient ages, and medical complexities, finding no differential item functioning (DIF) within the GMH and GPH domains across age groups, medical complexity, or clinical environment. Items assessing GMH and GPH within the PROMIS-GH were found to function comparably across treatment setting, age category, and medical comorbidities.