Invariance of the Bifactor Structure of Mild Traumatic Brain Injury (mTBI) Symptoms on the Rivermead Postconcussion Symptoms Questionnaire Across Time, Demographic Characteristics, and Clinical Groups: A TRACK-TBI Study
Stephanie Agtarap, Mark D. Kramer, Laura Campbell‐Sills, Esther L. Yuh, Pratik Mukherjee, Geoffrey T. Manley, Michael McCrea, Sureyya Dikmen, Joseph T. Giacino, Murray B. Stein, Lindsay D. Nelson, Opeolu Adeoye, Neeraj Badjatia, Kim Boase, Yelena G. Bodien, M. Ross Bullock, Randall M. Chesnut, John D. Corrigan, Karen Crawford, Ramon Diaz‐Arrastia, Ann-Christine Duhaime, Richard Ellenbogen, V. Ramana Feeser, Adam R. Ferguson, Brandon Foreman, Raquel C. Gardner, Etienne Gaudette, Dana Goldman, Luis González, Shankar P. Gopinath, Rao P. Gullapalli, J. Claude Hemphill, Gillian Hotz, Sonia Jain, Frederick K. Korley, Joel H. Kramer, Natalie Kreitzer, Harvey S. Levin, Christopher J. Lindsell, Joan Machamer, Christopher J. Madden, Alastair J. Martin, Thomas W. McAllister, Randall E. Merchant, Laura B. Ngwenya, Florence Noël, David O. Okonkwo, Eva Palacios, Daniel P. Perl, Ava M. Puccio, Miri Rabinowitz, Claudia S. Robertson, Jonathan Rosand, Angelle M. Sander, Gabriella Satris, David M. Schnyer, Seth A. Seabury, Mark Sherer, Sabrina R. Taylor, Nancy Temkin, Arthur W. Toga, Alex B. Valadka, Mary J. Vassar, Paul Vespa, Kevin Wang, John K. Yue, Ross Zafonte
Abstract
This study aimed to elucidate the structure of the Rivermead Postconcussion Symptoms Questionnaire (RPQ) and evaluate its longitudinal and group variance. Factor structures were developed and compared in 1,011 patients with mild traumatic brain injury (mTBI; i.e., Glasgow Coma Scale score 13-15) from the Transforming Research and Clinical Knowledge in TBI study, using RPQ data collected at 2 weeks, and 3, 6, and 12 months postinjury. A bifactor model specifying a general factor and emotional, cognitive, and visual symptom factors best represented the latent structure of the RPQ. The model evinced strict measurement invariance over time and across sex, age, race, psychiatric history, and mTBI severity groups, indicating that differences in symptom endorsement were completely accounted for by these latent dimensions. While highly unidimensional, the RPQ has multidimensional features observable through a bifactor model, which may help differentiate symptom expression patterns in the future.