Emerson Natacha D, Ferber Christopher, Jackson Nicholas J, Li Joshua, Tsay Eric, Styne Dennis, Gottschalk Michael, Mittelman Steven D, Moscicki Anna-Barbara
Department of Psychiatry and Biobehavioral Sciences, David Geffen School of Medicine, University of California, Los Angeles, 760 Westwood Plaza, Ste 48-241, Los Angeles, CA, 90095, United States, 1 3107948416.
Department of Medicine, Data Management and Statistical Core, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, United States.
JMIR Diabetes. 2025 Jun 3;10:e70380. doi: 10.2196/70380.
Adherence to type 1 diabetes mellitus (T1DM) treatment regimens decreases during adolescence. While comorbid depression and health insurance disparities are individually known to potentiate this risk, technological devices for T1DM appear to be protective.
We examined whether technology use impacted the association between depression and poorer health outcomes in T1DM. Given established insurance-based disparities based on technology access, we also studied whether the protective effects of T1DM technology differed among publicly and privately insured youth.
Data were prospectively collected from pediatric patients with T1DM across 3 California medical centers. We used linear and negative binomial regression analyses to examine whether technology use was related to diabetes outcomes and whether this differed based on depression status (technology-by-depression interaction) and health insurance type (technology-by-insurance interaction).
Across 1573 patients aged 12 to 25 years (mean age 15.9, SD 2.9 years; n=1050, 66.4%, non-Hispanic White; n=745, 47.0% female), those with a depression diagnosis had higher hemoglobin A1c (HbA1c; mean 9.1%, SD 2.1% vs 10.1%, SD 2.2%) and more frequent diabetic ketoacidosis (DKA) events per year (mean 0.10, SD 0.36 vs 0.24, SD 0.66) than those without (P=.003). Patients using both a continuous glucose monitor (CGM) and pump had lower HbA1c levels and fewer DKA events per year (mean HbA1c 8.2%, SE 0.1%; mean DKA events per year 0.05, SE 0.01) than those using one device (mean HbA1c 9.0%, SE 0.1%; mean DKA events 0.08, SE 0.1%) or none (mean HbA1c 10.0%, SE 0.1%; mean DKA events 0.19, SE 0.1%; P<.001). While youth with public insurance had significantly higher HbA1c levels than those with commercial insurance (mean 9.3%, SD 2.1% vs 9.0%, SD 2.0%, P<.001), those using a CGM had no reliable decrease in HbA1c compared to their commercially insured peers (P=.35).
Technology use in pediatric T1DM appears protective for both youth with a history of depression and those who are publicly insured. These data underscore the importance of universal access to technology to mitigate disparities based on comorbid mental health issues and differential access to care.
青少年在1型糖尿病（T1DM）治疗方案的依从性方面有所下降。虽然已知合并抑郁症和健康保险差异会分别增加这种风险，但用于T1DM的技术设备似乎具有保护作用。
我们研究了技术使用是否会影响T1DM患者抑郁症与较差健康结局之间的关联。鉴于基于技术获取情况已确定的基于保险的差异，我们还研究了T1DM技术的保护作用在公共保险和私人保险的青少年中是否存在差异。
前瞻性收集了加利福尼亚州3个医疗中心的儿科T1DM患者的数据。我们使用线性和负二项回归分析来研究技术使用是否与糖尿病结局相关，以及这是否因抑郁状态（技术与抑郁的相互作用）和健康保险类型（技术与保险的相互作用）而有所不同。
在1573名年龄在12至25岁的患者中（平均年龄15.9岁，标准差2.9岁；非西班牙裔白人1050名，占66.4%；女性745名，占47.0%），有抑郁症诊断的患者的糖化血红蛋白（HbA1c）水平更高（平均9.1%，标准差2.1%，而无抑郁症诊断的患者为10.1%，标准差2.2%），且每年糖尿病酮症酸中毒（DKA）事件更频繁（平均0.10，标准差0.36，而无抑郁症诊断的患者为0.24，标准差0.66）（P = 0.003）。同时使用连续血糖监测仪（CGM）和胰岛素泵的患者每年的HbA1c水平更低，DKA事件更少（平均HbA1c 8.2%，标准误0.1%；每年平均DKA事件0.05，标准误0.01），低于仅使用一种设备的患者（平均HbA1c 9.0%，标准误0.1%；平均DKA事件0.08，标准误0.1%）或未使用任何设备的患者（平均HbA1c 10.0%，标准误0.1%；平均DKA事件0.19，标准误0.1%；P < 0.001）。虽然有公共保险的青少年的HbA1c水平显著高于有商业保险的青少年（平均9.3%，标准差2.1%，对比9.0%，标准差2.0%，P < 0.001），但与有商业保险的同龄人相比，使用CGM的患者的HbA1c没有可靠的下降（P = 0.35）。
儿科T1DM患者使用技术似乎对有抑郁症病史的青少年和有公共保险的青少年都有保护作用。这些数据强调了普遍获得技术以减轻基于合并心理健康问题和不同医疗服务可及性的差异的重要性。
