Introduction: Latent autoimmune diabetes in Adults (LADA) represents up to 14% of adult diabetes. Compared to type 2 diabetes mellitus (T2DM) patients, LADA is believed to be diagnosed more acutely in younger patients with a lower body mass index (BMI), worse hemoglobin A1Cs (HbA1C), and lower triglycerides (TG). This study was done to examine for actual clinical differences between LADA and T2DM.

Methods: A retrospective case-control study was performed on adults with LADA and randomly matched with T2DM adults. Patients were matched based on gender and age (within 5 years) at time of diagnosis at an allocation ratio of 1:3 using the Military Health System Diabetes CarePoint Registry. Retrospective data were pulled at time of diagnosis to included ethnicity, BMI, HbA1C, lipids, TSH, the presence of symptoms, and medical histories. Univariate analyses were done for each variable followed by a multivariate analysis to determine the best overall model and control for confounding.

Results: In univariate analysis, BMI (P < 0.001), TG (P = 0.003), and HDL (P = 0.009) were significant predictors of LADA compared to T2DM at time of diagnosis. None of the acute symptoms, autoimmune disorders, or other medical history was found to be statistically significant. In the multivariate model, Caucasians were more likely to be LADA (OR = 14.1; 95% CI, 1.34 to 146.2, P = 0.027). Low BMI and low TG levels were found to interact inversely (P = 0.049) with the odds of being a LADA. However, low BMI effect was only significant with TG levels under 350 mg/dl, and low TG effect was only significant in those with BMI <28. The final model correctly classified 86% and 94.4% of the case-controls and controls.

Conclusion: Statistically significant differences were found between BMI, TG and HDL levels at time of diagnosis between LADA and T2DM adults, similar to those previously reported. Strongest differences were noted between BMI and TG levels, which may assist in diagnosis of LADA patients.


M.N. Lee: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at