The aim of this large cross-sectional population-based study was to examine a possible positive or negative association between thyroid dysfunction and chronic musculoskeletal complaints (MSC). Between 1995 and 97, all 94,197 adults in Nord-Trøndelag County in Norway were invited to participate in a health survey. A total of 64,787 (69%) responded to questions related to MSC, whereof thyroid-stimulating hormone (TSH) was measured in 34,960 individuals. These included a 5% random sample of women and men 20-40 years of age (n=2165), nearly all women above 40 (n=19,308), a random sample which included 50% of men older than 40 years (n=9983), and 3504 (97%) with self-reported thyroid dysfunction. Among the 64,787 participants, 30,158 (47%) who reported MSC continuously for at least 3 months during the past year where defined as having chronic MSC. Associations between thyroid dysfunction and chronic MSC were assessed in multivariate analyses, estimating prevalence odds ratios (ORs) with 95% confidence intervals (CIs). High TSH values were associated with lower prevalence of chronic MSC at ten anatomical sites among women with no history of thyroid dysfunction. Among these, chronic MSC was less likely (OR=0.6, 95% CI 0.4-0.8) if TSH >or=10 mU/L than in women with normal TSH (0.2-4 mU/L). Chronic MSC was less likely among women with high TSH values. The mechanism is unclear and, theoretically, may reflect a fundamental gender-specific relationship between TSH and pain perception in the central nervous system.