Thyroid nodules (TNs) are among the most common endocrine disorders. With the widespread use of high-resolution ultrasonography, their detection rate has risen sharply, making them one of the most frequently identified findings during routine health examinations (1). Recent studies estimate that TNs affect more than half of the adult population, although the majority of thyroid nodules are benign, a subset carries a risk of malignancy (2–4). Therefore, early identification of high-risk individuals is crucial for effective prevention and intervention.

In recent years, increasing attention has been directed toward the role of metabolic syndrome and lifestyle factors in the development of TNs. Metabolic syndrome is a pathological condition characterized by central obesity, insulin resistance, hypertension, and dyslipidemia, and is closely associated with both endocrine and cardiovascular disorders. Several studies have reported associations between metabolic syndrome components, including obesity, dyslipidemia, hyperglycemia, and hyperuricemia, are associated with an increased risk of TNs (5–9). Lifestyle behaviors such as smoking and alcohol consumption have also been linked to TNs (10). However, findings across different populations and regions remain inconsistent, and the specific contributions of individual factors are still not fully understood. Moreover, many existing studies are limited by small sample sizes, insufficient evaluation of metabolic variables, and inadequate control of potential confounders.

To address these gaps, we conducted a large-scale cross-sectional study based on 23,305 adults undergoing physical examinations in northern China. We aimed to: (1) assess the prevalence of TNs in a large adult population undergoing routine health screening; (2) explore their associations with a wide spectrum of metabolic factors using rigorous statistical methods; and (3) construct a supplementary logistic regression-based prediction model as a secondary exploratory analysis. Our findings are intended to enrich the epidemiological understanding of TNs and support refined screening strategies in general health management.

In this large cross-sectional analysis of 23,305 adults undergoing routine health examinations, we observed a high prevalence of thyroid nodules (64.7%), which is higher than that reported in many earlier population-based studies but comparable to rates reported in recent ultrasound-based surveys and meta-analyses (6). Several factors may explain the relatively high prevalence observed in this study. First, the study population was derived from individuals undergoing routine health examinations rather than a randomly selected community-based sample. Health examination populations tend to be older and more health-conscious, which has been associated with higher detection rates of thyroid nodules. Second, high-resolution thyroid ultrasonography was routinely used and performed by experienced sonographers, enabling the detection of small or subclinical nodules that might be missed by palpation-based screening or lower-resolution imaging. According to the 2015 American Thyroid Association Management Guidelines, high-resolution ultrasound can detect thyroid nodules in 19%–68% of randomly selected individuals, with higher detection rates observed in women and older adults (26). Third, regional characteristics, including demographic structure and iodine nutrition patterns in Northern China, may also contribute to variability in thyroid nodule prevalence (27).

TN prevalence demonstrated a clear age-related trend, increasing from 38.8% in participants aged ≤30 years to 87.8% in those aged >70 years. Women consistently exhibited a higher prevalence than men across all age groups, with the sex difference most pronounced in the 40–70 age range. These findings align with prior reports and reflect the well-established role of aging and sex hormones in thyroid physiology (27).

Our comparative analysis revealed that individuals with TNs displayed a distinct metabolic profile characterized by higher proportions of overweight/obesity, hypertension, abnormal fasting glucose, hyperuricemia, and elevated creatinine. These patterns are in line with accumulating evidence linking metabolic syndrome components to thyroid nodular disease (28, 29). In particular, obesity has been associated with increased thyroid volume and nodule formation, possibly through insulin resistance and chronic low-grade inflammation (7, 30). Insulin and insulin-like growth factors may exert mitogenic effects on thyrocytes, stimulating proliferation and angiogenesis (31). Hypertension and dysglycemia may also contribute via endothelial dysfunction and oxidative stress, promoting thyroid tissue remodeling (32).

Sex-stratified analyses showed that elevated triglycerides were more prevalent among TN-positive women, whereas smoking and alcohol consumption were significantly associated with TNs in men. Due to the extremely low prevalence of smoking and alcohol use among females in this cohort, reliable assessment of these behavioral factors in women was not feasible, and therefore, no conclusions regarding their association should be drawn. These results suggest potential sex-specific metabolic and behavioral influences on TNs development, likely reflecting differences in hormonal milieu, fat distribution, and lifestyle exposures (33).

Despite the observed univariate associations between TNs and various metabolic indicators, only three factors—age, female sex, and higher BMI—remained independent predictors in the multivariable logistic regression model. This implies that the contributions of other metabolic parameters may be mediated through these primary factors or confounded by them. The biological plausibility of these associations is supported by prior research: aging leads to cumulative oxidative and hormonal changes that affect thyroid morphology, estrogen may stimulate thyroid epithelial cell growth, and obesity-related hyperinsulinemia can enhance thyrocyte proliferation via insulin/insulin-like growth factor 1 signaling pathways (34–36). The non-independence of other metabolic indicators in multivariable analysis likely reflects their secondary or downstream nature relative to these core determinants.

We developed a simplified prediction model for TNs risk based on positive findings from multivariable regression analysis, incorporating age, sex, and BMI. The model demonstrated moderate discriminative ability (AUC = 0.706) and good calibration (MAE = 0.007), indicating solid internal validity. While inclusion of additional variables such as thyroid function tests, iodine status, or inflammatory markers could potentially improve predictive accuracy, the simplicity and accessibility of these three variables make the model particularly suitable for large-scale preliminary risk stratification. Nevertheless, it cannot replace comprehensive, thyroid-specific evaluations, including ultrasonography or detailed biochemical assessments.

Although most thyroid nodules are benign, their extremely high prevalence in health examination populations poses challenges related to over-screening, unnecessary follow-up, and inefficient use of ultrasound resources. In this context, the simplified prediction model proposed in this study is not intended for diagnostic purposes, nor to replace thyroid ultrasonography. Instead, the model is designed as a preliminary risk stratification tool to support large-scale health check-up programs. By relying solely on readily available and cost-free variables—age, sex, and body mass index—the model may help identify individuals with a higher probability of thyroid nodules who could benefit from prioritized or more frequent ultrasonographic evaluation, while potentially reducing unnecessary imaging in low-risk groups. This risk-stratified approach may contribute to more rational allocation of medical resources in population-based health management. Importantly, the simplicity of the model enhances its feasibility and scalability in real-world screening settings, where more complex models may be impractical.

This study makes several important contributions. First, the large and diverse sample, encompassing a wide range of ages, sexes, and metabolic statuses, enables precise prevalence estimates and enhances external validity. Second, beyond well-established factors such as age, sex, and BMI, the study simultaneously evaluated a broad spectrum of metabolic indicators—including hyperuricemia, elevated SCR, and dyslipidemia—allowing for a more comprehensive understanding of their interplay in TNs risk. Third, the construction of a simple three-variable prediction model balances usability with predictive performance, offering a practical tool for large-scale clinical and community screening. Collectively, these findings provide evidence to inform precision screening strategies and early interventions, and lay the groundwork for future longitudinal and multi-center studies aimed at improving TNs prevention and management.

Several limitations of this study should be acknowledged. Most importantly, data on thyroid function parameters, including thyroid-stimulating hormone (TSH), thyroid hormones, thyroid autoantibodies, and urinary iodine concentration, were not available in this health examination database. TSH is a well-established independent predictor of thyroid nodule development and malignancy risk, and iodine status plays a critical role in thyroid morphology and nodular disease. Variations in TSH levels have been reported to be associated with metabolic abnormalities such as obesity, insulin resistance, and dyslipidemia, which are also key components of metabolic syndrome (37). Therefore, the inability to adjust for TSH may have resulted in residual confounding and could have led to partial overestimation or underestimation of the observed associations between metabolic factors and thyroid nodules. Similarly, unmeasured variation in iodine status, which is known to influence thyroid size and nodular formation and may differ across regions and populations, could have modified the relationship between metabolic syndrome and thyroid nodules in this cohort (38). Because this study was based on a retrospective health examination dataset, these variables could not be retrieved.

Additionally, smoking and alcohol history were self-reported, which may have introduced recall or social desirability bias. Finally, our cohort was derived from a single center in northern China, potentially limiting generalizability to other regions or ethnic populations.

In conclusion, our study demonstrates a high prevalence of thyroid nodules in the general adult population of northern China and identifies age, female sex, and elevated body mass index as the primary independent risk factors. Although other metabolic disturbances—including hypertension, hyperglycemia, hyperuricemia, and lipid or renal abnormalities—were more frequent among individuals with thyroid nodules, their effects appear largely mediated through these core factors.

We developed a practical clinical prediction model based solely on age, sex, and body mass index, which showed moderate discriminatory ability and excellent calibration. This model may facilitate early identification of high-risk individuals and support risk stratification for thyroid nodules in both clinical practice and public health initiatives.

Future studies should aim to validate this model longitudinally across diverse populations and investigate the potential inclusion of thyroid function parameters, iodine status, and inflammatory markers to improve predictive performance and elucidate the mechanistic links between metabolic health and thyroid nodular disease.