(PS4-4) Confirmatory Factor Analysis of the Tinnitus Severity Short Form

Friday, November 15, 2024

1:00 PM – 2:00 PM EST

Location: Franklin Hall, Level 4, Marriott Downtown

Author(s)

Emma R. Dear, B.S.

Graduate Student
University of North Carolina at Chapel Hill
Chapel Hill, North Carolina, United States

Co-Author(s)

LM

Lauren Mueller, B.A.

Medical Student
Washington University in St. Louis
Jacksonville, Florida, United States
Juan Turcios, Jr., M.S.

Medical student
Meharry Medical College
Nashville, Tennessee, United States
HS

Harrison Smith, M.D.

Otolaryngology Resident
Icahn School of Medicine of Mount Sinai
New York, New York, United States
MF

Madelyn R. Frumkin, Ph.D.

Clinical Fellow
Massachusetts General Hospital
Boston, Massachusetts, United States
TR

Thomas Rodebaugh, Ph.D.

Professor
University of North Carolina at Chapel Hill
Chapel Hill, North Carolina, United States
DK

Dorina Kallogjeri, M.P.H., M.D.

Assistant Professor Otolaryngology—Head & Neck Surgery
Washington University in St. Louis
St. Louis, Missouri, United States
JP

Jay Piccirillo, M.D.

Professor of Otolaryngology Head & Neck Surgery
Washington University in St. Louis
St. Louis, Missouri, United States

Abstract Text: Individuals with tinnitus perceive sound (i.e., ringing) without an external source, leading to high distress in some individuals. Cognitive Behavioral Therapy (CBT) effectively treats tinnitus distress. However, according to one study, only 0.2% of individuals with tinnitus discuss receiving CBT with their physician. The impact of not receiving this treatment can be severe because tinnitus distress is associated with a lower quality of life, anxiety, and depression. Using a self-report measure to screen for tinnitus distress may help increase the number of people receiving CBT for tinnitus distress. A new, brief, 10-item measure, the Tinnitus Severity Short Form (TS-SF) may be the solution. In a preliminary study, the TS-SF showed strong convergent validity with existing measures of tinnitus distress, strong internal consistency (ɑ = 0.96), and a moderate to strong correlation of TS-SF scores across timepoints (r = 0.69; 95% CI: 0.62-0.75) suggesting good test-retest reliability. The confirmatory factor analyses in this study suggested either a unifactorial structure of a distress factor or a bifactorial model with a distress factor and two method factors (CFI = .98, TFI= .96, RMSEA = .92). After these analyses, one of the questions was identified as redundant and was removed from the instrument. An additional factor analysis with a second data set will help further confirm the optimal factor structure of the TS-SF. Participants (N=325) from the American Tinnitus Association completed the TS-SF. Mplus was used to run confirmatory factor analyses using the maximum likelihood with robust standard errors estimator (MLR). Then, a chi square difference test was used to compare competing models to determine which had the optimal fit. An initial model with two method factors revealed that the second method factor had nonsensical loadings, so this method factor was removed. A bifactorial solution with a distress and a single method factor provided acceptable to strong fit for all indices (CFI =.97, TFI=.95, RMESA=.09, SRMR=.02), and all loadings were statistically significant (p < .001) with the exception of the reverse-scored item. With removal of this item, acceptable to strong fit indices were maintained (CFI = .97, TFI= .94, RMSEA=.10, SRMR= .02), and all loadings were statistically significant. We also tested this second model with factor loadings constrained to be equal for the method factor, and fit indices were largely maintained with no significant differences to the unconstrained model (CFI = .96, TFI= .94, RMSEA=.09, SRMR= .03). The TS-SF shows strong promise as a psychometrically sound, brief measure of tinnitus distress. Factor analyses indicate that a bifactorial model for 9 items of the TS-SF is the best fit for assessing distress in clinical practice. Users of the TS-SF should be aware of the underlying method factor that unites items 2-5, likely due to the wording of these items. With a brief measure of tinnitus distress apt for use in clinical settings, we may be able to connect more individuals with tinnitus distress to CBT.