RESUMO
CONTEXT: Increased thyroid cancer incidence is at least partially attributed to increased detection and shows considerable regional variation. OBJECTIVE: We investigated whether regional variation in cancer incidence was associated with variations in thyroid disease management. DESIGN: We conducted a retrospective population-based cohort study that involved linking data from the Belgian Health Insurance database and the Belgian Cancer Registry to compare thyroid-related procedures between regions with high and low cancer incidence. MAIN OUTCOME MEASURES: Primary outcome measures were rates of TSH testing, imaging, fine-needle aspiration cytology (FNAC), and thyroid surgery. Secondary study outcomes were proportions of subjects with thyrotoxicosis and nodular disease treated with surgery, of subjects treated with surgery preceded by FNAC or with synchronous lymph node dissection, and of thyroid cancer diagnosis after surgery. RESULTS: The rate of TSH testing was similar, but the rate of imaging was lower in the low incidence region. The rate of FNAC was similar, whereas the rate of surgery was lower in the low incidence region (34 [95% CI 33; 35 ] vs 80 [95% CI 79; 81 ] per 100,000 person years in the high incidence region; P < .05). In the low incidence region compared to the high incidence region, surgery represented a less chosen therapy for euthyroid nodular disease patients (47% [95% CI 46; 48] vs 69% [95% CI 68; 70]; P < .05), proportionally more surgery was preceded by FNAC, more cancer was diagnosed after total thyroidectomy, and thyroid cancer patients had more preoperative FNAC and synchronous lymph node dissection. CONCLUSION: Regional variation in thyroid cancer incidence, most marked for low-risk disease, is associated with different usage of thyroid imaging and surgery, supporting variable detection as a key determinant in geographic variation.
