Abstract
Diabetes mellitus (DM) is predisposing to the development of latent tuberculosis infection (LTBI). An understanding of the underlying factors of LTBI-DM is important for tuberculosis prevention and control. This study aims to evaluate the association between LTBI and DM among the noninstitutionalized civilian population in the United States, focusing on the impact of serum globulins. We performed a cross-sectional study design using public data from 2011 to 2012 National Health and Nutrition Examination Survey, focusing on participants diagnosed with LTBI who were aged 20 and above. Weighted Wilcoxon rank-sum and weighted chi-square tests were used to compare group differences. A multivariable logistic regression model was constructed to assess the association between serum globulin and DM, with subgroup analyses and evaluations of nonlinear relationships. Receiver operating characteristic curves were used to assess the predictive power of the models. A total of 694 participants (512 DM and 182 nonDM) were included in our study and the incidence of DM was 22%. Higher serum globulin levels were significantly associated with an increased risk of DM, with a 21% increase in risk for each unit increase in serum globulin (odds ratio = 1.21, 95% confidence interval [1.03, 1.43], P < .001). The relationship between serum globulin and DM was linear, and higher serum globulin levels were associated with a higher risk of DM, particularly in males (P = .043) and obese individuals (P = .019). The area under the curve for serum globulin predicting DM was 0.795, with an optimal cutoff value of 2.9. Elevated serum globulin levels are significantly associated with an increased risk of DM among individuals with LTBI, highlighting the potential role of serum globulin as a predictive biomarker for DM in this population. However, the specific mechanism between globulin and LTBI-DM needs to be further investigated.