Objective To assess whether a lifestyle intervention starting in early pregnancy can reduce the incidence of gestational diabetes mellitus (GDM) among Chinese women with high risk factors. Methods Between January 2020 and September 2021, women at gestational age of 6-15 weeks with high risk factors [body mass index (BMI) of 24-35 kg/m2 or a family history of diabetes] in 4 urban districts and 3 suburban districts of Tianjin were recruited as an intervention cohort. During the same period, the high-risk pregnant women who received routine maternal management from two of the above districts, one in urban and the other in suburban district, were retrospectively enrolled as a control cohort through the hospital system information. Women in the intervention cohort received a combination of online and offline lifestyle intervention (nutrition, exercise, and weight control). At the baseline survey, information including age, family history of diabetes, gestational age, and fasting plasma glucose (FPG) level were collected at the baseline survey. The height and body weight were recorded, and BMI was calculated. All pregnant women underwent two-step GDM screening at gestational age of 24-28 weeks. The comparison between the two cohorts was performed using the t test or the χ2 test. Logistic regression model was used to evaluate the effect of lifestyle intervention on GDM incidence. Results Finally, 106 and 2 348 patients were included in the intervention and control cohorts, respectively. There were statistically significant differences between the two cohorts in gestational age, family history of diabetes proportion, gestational age and FPG level at the baseline survey (P<0.05). At gestational age of 24-28 weeks, the incidence of GDM in the intervention and control cohorts was 19.8% (21/106) and 29.6% (696/2 348), respectively (P=0.029), with a relative risk reduction of 33%. Multivariate logistic regression analysis showed that the risk of GDM was significantly lower in the intervention cohort than that in the control cohort, with an OR and 95% confidence interval of 0.41 (0.24-0.69). Conclusion Lifestyle intervention (nutrition, exercise, and weight control) starting in early pregnancy can reduce the incidence of GDM in pregnant women with high risk factors. [ABSTRACT FROM AUTHOR]