Diabetes mellitus is one of the major non-communicable diseases (NCD) with increasing prevalence in China. There is a lack of high-quality research focusing on prevention and management of diabetes in low and middle income countries (LMICs) compared to developed countries. This comparative study aims to describe the characteristics of diabetes research conducted in China and the USA. The study included 800 studies on diabetes mellitus from both countries. Compared with studies in the USA, studies in China were more likely to be laboratory-based primary research (50.5% versus 30.8%), more likely to use animal subjects (47% versus 27.5%), more likely to focused on risk factors (22.7% versus. 14.7%), more likely to be case-controlled studies (17.7% versus 10.0%), and more likely to evaluate pharmacological treatments (36.5% versus 20.7%). Further, compared with studies in the USA, studies in China were less likely to involve patients (42.7% versus 60.7%), less likely to be clinical trials (6.2% versus 14.5%), less likely to be cohort studies (8.8% versus. 26.0%), and less likely to evaluate disease management interventions (3.3% versus 13.3%). Clinical studies in China should be more patient-based to facilitate more effective control and management of diabetes.