BackgroundMedical distrust in patients has been related with poor compliance to medications and suboptimal clinical outcomes. Effective interventions may improve medical distrust in patients which is warranted specific studies.ObjectiveTo evaluate the effect of health literacy and exercise interventions on medical mistrust in patients with type 2 diabetes (T2DM) .MethodsThis study was based on a cluster randomized-controlled trial conducted during February 2015 to March 2016. A total of 800 T2DM patients were recruited from four community health service centers in Minhang and Changning Districts of Shanghai by using a multi-stage sampling method. All patients were randomly divided into the control group and 3 intervention groups in the community. Routine care was provided to all the patients, and on this basis, health literacy intervention, exercise intervention and health literacy + exercise intervention (comprehensive intervention) were performed to the three intervention groups based on partnership to improve diabetes education (PRIDE) toolkit, respectively. Information was collected using the Chinese versions of Health Literacy Management Scale (c-HeLMS) , the 5-item Diabetes Numeracy Test Scale (c-DNT-5) , and Medical Mistrust Index (c-MMI) at baseline, 3-, 6-, 12- (end of intervention) and 24-months follow-up (post-intervention) . The generalized estimating equation was used to evaluate the effect of the interventions at each time point.ResultsA total of 780 patients were recruited in this study. The c-MMI was a reliable and valid scale to measure medical mistrust in our subjects, with Cronbach's α of 0.826. The median score and interquartile range of c-MMI was 31 (7) at baseline while the rate of medical mistrust (scores≥30) was 65.9% (514/780) , both of which decreased at almost each follow-up survey in four groups. Compared with the control group, a lower risk of medical mistrust was observed at the 3-months〔OR (95%CI) =0.42 (0.23, 0.78) 〕 and 6-months of follow-up〔OR (95%CI) =0.46 (0.24, 0.88) 〕 for the health literacy group, at the 6-months〔OR (95%CI) =0.50 (0.25, 0.99) 〕, 12-months〔OR (95%CI) =0.43 (0.22, 0.86) 〕 and 24-months of follow-up〔OR (95%CI) =0.37 (0.19, 0.72) 〕 for the exercise group, and at the 6-months〔OR (95%CI) =0.30 (0.16, 0.56) 〕 for the comprehensive group.ConclusionBoth health literacy and exercise intervention may effectively decrease the scores of c-MMI and reduce the risk of medical mistrust in diabetes patients.