Long-Term Effectiveness of the TIME Intervention to Improve Diabetes Outcomes in Low-Income Settings: a 2-Year Follow-Up

Background We previously found that a 6-month multidimensional diabetes program, TIME ( T elehealth-Supported, I ntegrated Community Health Workers, Me dication-Access) resulted in improved clinical outcomes. Objective To follow TIME participant clinical outcomes for 24 months Participants Low-income Latino(a)s with type 2 diabetes Design and Intervention We collected post-intervention clinical data for five cohorts ( n = 101, mean n = 20/cohort) who participated in TIME programs from 2018 to 2020 in Houston, Texas. Main Measures We gathered HbA1c (primary outcome), weight, body mass index (BMI), and blood pressure data at baseline, 6 months (intervention end), and semiannually thereafter until 24 months after baseline to assess sustainability. We also evaluated participant loss to follow-up until 24 months. Key Results Participants decreased HbA1c levels during the intervention ( p < 0.0001) and maintained these improvements at each timepoint from baseline to 24 months ( p range: < 0.0001 to 0.015). Participants reduced blood pressure levels during TIME and maintained these changes at each timepoint from baseline until 18 months (systolic p range < 0.0001 to 0.0005, diastolic p range: < 0.0001 to 0.008) but not at 24 months (systolic: p = 0.065; diastolic: p = 0.85). There were no significant weight changes during TIME or post-intervention: weight ( p range = 0.07 to 0.77), BMI ( p range = 0.11 to 0.71). Attrition rates (loss to follow-up during the post-intervention period) were 5.9% (6 months), 24.8% (12 months), 35.6% (18 months), and 41.8% (24 months). Conclusions It is possible for vulnerable populations to maintain long-term glycemic and blood pressure improvements using a multiple dimensional intervention. Attrition rates rose over time but show promise given the majority of post-intervention timepoints occurred during the COVID-19 pandemic when low-income populations were most susceptible to suboptimal healthcare access. Future studies are needed to evaluate longitudinal outcomes of diabetes interventions conducted by local clinics rather than research teams.