The Thyroid Cancer Modified Anxiety Scale (TC-MAX): Development and External Validation of a Thyroid Cancer Distress Instrument.

BACKGROUND: Patient distress associated with thyroid cancer greatly influences treatment choice, yet is underestimated by clinicians. We developed and validated the Thyroid Cancer Modified Anxiety Scale (TC-MAX) to measure patient-reported thyroid cancer-related worry. METHODS: Development entailed (1) conceptualization (systematic literature review, expert panel [n = 4], patient focus groups [n = 10]); (2) qualitative revision (pilot testing [n = 59], expert panel [n = 4], patient feedback [n = 4]); (3) internal validation (n = 148); and (4) external validation (n = 1002), with papillary thyroid cancer patients. Internal consistency, test-retest reliability, exploratory/confirmatory factor analysis, and construct validity were evaluated. Severity categories and minimal clinically important differences (MCIDs) were determined. RESULTS: An 18-question scale was constructed, encompassing three subscales (Collective Anxiety, Ultrasound Anxiety, and Fear of Recurrence/Progression). Internal validation revealed high internal consistency (Cronbach's α = 0.93), excellent test-retest reliability (intraclass correlation coefficient [ICC] 0.86), and significant construct validity (Distress Thermometer [Spearman's ρ = 0.56, p < 0.001], Hospital Anxiety Depression Scale [HADS] [Spearman's ρ = 0.52, p < 0.001], and Functional Assessment of Cancer Therapy-General [FACT-G] [Spearman's ρ = -0.50, p < 0.001]). Exploratory factor analysis showed expected loadings for respective subscales, and confirmatory factor analysis confirmed that the structure was parsimonious. External validation exhibited similar psychometrics, including internal consistency (Cronbach's α = 0.94), test-retest reliability (ICC 0.87), and construct validity (Distress Thermometer [Spearman's ρ = 0.54, p < 0.001], HADS [Spearman's ρ = 0.67, p < 0.001], and FACT-G [Spearman's ρ = -0.62, p < 0.001]). Proposed ranges (low = 0-26, moderate = 26-43, and severe = 43-100) were in excellent agreement between internal and external validation cohorts (weighted κ = 0.84, p < 0.001). Anchor-based MCID was found to be 16.7 (active surveillance vs. surgery) (25.9 [standard deviation (SD) ± 16.4] vs. 42.6 [SD ± 23.3], p < 0.001) or 16.9 (hemithyroidectomy vs. thyroidectomy) (33.6 [SD ± 17.4] vs. 50.5 [SD ± 24.1], p = 0.005). CONCLUSIONS: TC-MAX represents the first dedicated measure of anxiety in thyroid cancer. It exhibits robust psychometric properties in a vulnerable population, accounting for complex viewpoints. The scale effectively assesses preference and informs patients, including determining fit for active surveillance or extent of surgery.