Diabetes, a chronic disease necessitating long-term treatment and self-management, presents significant challenges for patients who spend most of their treatment time outside of hospitals. The potential of digital therapeutics for diabetes has garnered recognition from different organizations. Although some prior studies have demonstrated successful reductions in patients' blood glucose levels and body weight through digital diabetes programs, many studies were limited by including patients with prediabetes, including patients treated with mostly premixed insulin, or evaluating user engagement outcomes rather than clinical outcomes. Consequently, limited evidence remains regarding the effectiveness of health management mobile apps specifically designed for patients with type 2 diabetes mellitus (T2DM) initiating basal insulin (BI). Based on this, a data-based and artificial intelligence management system named "TRIO" was developed to provide patients with more personalized intervention methods in stages, in groups, and around the clock. TRIO assists doctors and nurses in achieving better blood glucose controls, truly carries out standardized management around patients, and allows them to have a higher quality of life. TRIO represents the 3 essential pillars in comprehensive diabetes management: physician, nurse, and patient.This prospective observational study evaluated the effectiveness and safety of the TRIO optimal health management program for patients with T2DM initiating BI therapy in a real-world setting.Patients aged 18-85 years with inadequate glycemic control (baseline hemoglobin A1c [HbA1c] ≥7%) starting BI therapy were enrolled in outpatient and inpatient settings. The study lasted 3 months, with health education and phone-based follow-up assessments. Data collected included patient characteristics, medical history, baseline diabetes conditions, treatment compliance, glycemic control, and safety indicators.A total of 199,431 patients were included, and 118,134 patients completed the 3-month follow-up between December 1, 2019, and December 31, 2021, involving 574 hospitals in China. The mean baseline HbA1c was 9.2%, the mean duration of diabetes was 7.3 years, and 80.4% (1,59,930/1,98,969) of patients were using BI with oral antihyperglycemic drugs. After the intervention, mean HbA1c decreased by -2.59% from baseline, with 55.6% (28,858/51,912) achieving the target HbA1c level of <7%. Patients who set lower fasting plasma glucose goals (<6.1 mmol/L) showed more significant HbA1c reductions (P<.001) and higher target achievement than those with fasting plasma glucose goals of ≥6.1 mmol/L. Factors such as complications, diabetes duration, and baseline HbA1c levels influenced the magnitude of HbA1c reduction. The presence of complications, shorter diabetes duration, and higher baseline HbA1c were significantly associated with increased hypoglycemia incidence risk (all P<.05).The TRIO optimal health management program effectively improved glycemic control in patients with T2DM initiating BI therapy. Individualized treatment approaches considering patient characteristics and glycemic goals are vital for optimal outcomes.©Chenxi Li, Lixin Guo, Lixin Shi, Li Chen, Liming Chen, Yaoming Xue, Hong Li, Yuzhen Liang, Jing Yang, Weimin Wang, Dalong Zhu. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 13.01.2025.