To search for risk factors that could predict progression in latent autoimmune diabetes in adults (LADA) and compare them with those for type 2 diabetes. This study included 175 participants with LADA (autoantibody positive, without insulin treatment ≥1year after diagnosis) and 2331 participants with type 2 diabetes (autoantibody negative, without insulin treatment ≥1year after diagnosis) from the HUNT2 and HUNT3 surveys. We used Cox regression models and receiver operating characteristic curve analysis to identify predictive factors for progression to insulin dependency within 10 years. Low C-peptide levels (<0.3nmol/L) predicted progression to insulin dependency within 10 years in both LADA (hazard ratio [HR] 6.40 [95% CI, 2.02-20.3]) and type 2 diabetes (HR 5.01 [95% CI, 3.53-7.10]). In addition, a high glutamic acid decarboxylase autoantibody (GADA) level (HR 5.37 [95% CI, 1.17-24.6]) predicted progression in LADA. Together, these two factors had a discriminatory power between non-progressors and progressors of area under the curve (AUC) of 0.86 (95% CI, 0.80-0.93). In type 2 diabetes, younger age at diagnosis (<50 years: HR 2.83 [95% CI, 1.56-5.15]; 50-69 years: HR 2.11 [95% CI, 1.19-3.74]), high HbA1c levels (≥53 mmol/mol, HR 2.44 [95% CI, 1.72-3.46]), central obesity (HR 1.65 [95% CI, 1.06-2.55]) and a body mass index of more than 30 kg/m2 (HR 1.73 [95% CI, 1.23-2.41]) were independent predictors. Together with C-peptide they reached an AUC of 0.79 (95% CI, 0.76-0.82). Factors predicting progression to insulin dependence are partly similar and partly dissimilar between LADA and type 2 diabetes. A constellation of low C-peptide and high GADA levels identifies LADA patients who are probable to progress to insulin dependence.