3043 — Concurrent Validity of the Global Assessment of Function (GAF) for PTSD
Jackson JC (VA TVHS), Marx B
(VA National Center for PTSD), Sayer N
(VA Minneapolis), Greevy R
(Vanderbilt Department of Biostatistics), Nwosu S
(Vanderbilt Department of Biostatistics), Owen R
(VA Mental Health Quality Enhancement Research Initiative (MH QUERI)), Weathers F
(Department of Psychology, Auburn University), Schnurr P
(Departments of Psychiatry, Dartmouth Medical School), Sinnott P
(Health Economics Resource Center VA Palo Alto Health Care System), Speroff T
Disability examination for PTSD requires the clinician to assign a GAF, a global subjective rating that characterizes functional impairment from a score of 1 to 100 (superior functioning). Our objective was to compare PTSD examiner GAF scores to those independently assigned by PTSD experts and to a severity of functional assessment derived from clinical interviews using the World Health Organization Disability Assessment Schedule (WHODAS-II).
The intervention arm of the QUERI-funded Enhancing Equitable and Effective PTSD Disability Assessment Clinical Trial included standardized functional assessment using the WHODAS-II. Subsequently, PTSD expert raters from the National Center-PTSD performed blinded review of audio recordings of the disability examination and assessed functional impairment in seven domains comprising the WHODAS-II, while independently assigning a GAF score to each interview. We tested the correlation among the examiner GAF scores, the PTSD expert GAF scores, and the expert WHODAS-derived functional impairment severity ratings (from 1 to 5, higher is worse). Statistical significance was tested using mixed effects multivariate regression models to account for clustering of patients within examiners within sites.
Preliminary results include expert review of 106 exams. The examiner GAF scores had a median of 53 (IQR: 50 to 57) compared to GAF scores determined by PTSD experts of 55 (50 to 60). The correlation between examiner and PTSD expert GAF scores was 0.37 (p = 0.002). Examiner GAF ratings were inversely correlated with functional impairment (r = -0.40, p = 0.001). For veterans with a PTSD diagnosis, the median examiner GAF was 50 (48 to 55) and the functional impairment scores was 3.2 (2.8 to 3.4). For veterans not receiving a diagnosis of PTSD, scores were 60 (53 to 65) and 2.7 (2.3 to 3.0), respectively.
WHODAS clinical interviews during the disability examination result in GAF scores corresponding to degree of functional impairment reported by the veteran. The level of functional impairment was “moderate”, with veterans who met criteria for a diagnosis of PTSD demonstrating more functional difficulty than their counterparts without PTSD.
Incorporating the WHODAS clinical interview into the disability assessment may anchor the GAF scoring to an evidence-based approach that increases the validity of the GAF Axis-V diagnosis.