Agreeing to Disagree: Inter-Subject Variability of Mood Rating Scales During Rtms Treatment of Depression

Background: Outcomes of repetitive Transcranial Magnetic Stimulation (rTMS) for treatment-resistant depression vary widely across published studies. Several different mood rating scales have been used to assess response and remission. Objective: To compare the performance of three self-report scales (Inventory of Depressive Symptomatology 30-item [IDS], Patient Health Questionnaire 9-item [PHQ], and Profile of Mood States 30-item [POMS]) and the observer-rated Hamilton Depression Rating Scale 17-item (HDRS) in detecting symptom change during rTMS treatment. Results: 805 subjects with nonpsychotic MDD completed a six-week course of rTMS treatment with weekly self-report and observer ratings. All scales detected significant improvement during treatment, with correlations of Kendall’s τ=0.71-0.80 among self-report instruments and τ=0.52-0.56 between self-report and observer-rated instruments (p<0.001). Response/remission rates varied widely across instruments: highest was PHQ (54%/34%), followed by POMS (49%/27%), IDS (41%/30%), and HDRS (36%/20%). Higher baseline severity was associated with lower likelihood of remission across all scales (Hazard Ratio 0.86-0.98) and response across self-rated scales except PHQ (HR 0.98-0.99) (p<0.05). Greater improvement by session 10 predicted remission across all scales (HR 1.51-3.24, p<0.001). Defining outcome based on the minimum number of scales meeting response/remission criteria yielded rates of 56%/37% for at least one scale, 42%/27% for two scales, and 22%/18% for three scales. Using any single scale conferred 5.2-27% risk for not detecting response/remission detected by another scale (lowest with PHQ). Conclusion: All instruments detected improvement (including early improvement) with rTMS treatment, but degree of improvement varied notably. Use of any single scale conferred a significant risk of not detecting response/remission detected on another. Early improvement appeared more strongly predictive of eventual response/remission on self-rated scales. These findings suggest caution in comparing degree of improvement across studies using different scales, and that accurate assessment of symptom burden over treatment may require multiple instruments. Research Category and Technology and Methods Clinical Research: 10. Transcranial Magnetic Stimulation (TMS) Keywords: rTMS, measurement-based care, depression