CGM Threshold Alerts Improve Inpatient Glycemic Control, but Predictive Alerts Add No Benefit, finds trial
In a three-arm trial of hospitalized patients with diabetes, continuous glucose monitoring (CGM) threshold alerts increased time in the target glucose range by 3.9%, exceeding the 3% threshold generally considered clinically meaningful. However, adding predictive alerts did not provide additional improvement in glycemic control. The lack of benefit from predictive alarms may reflect frequent alerts and barriers integrating alarms into hospital workflows, suggesting that simpler threshold-based CGM alerts may be more practical for inpatient diabetes management. The study was published in JAMA Network Open by Yaxin W. and colleagues.
In order to study the value added by various sensor alerts, researchers have undertaken a single-center, open-label, randomized clinical trial involving a tertiary hospital from April 15 through December 9, 2024. The study has recruited 533 adult patients with either type 1 or type 2 diabetes admitted to the hospital, and these patients have been randomly allocated to three unique sensor alert interventions in a ratio of 1:1:1: no alerts (n = 178), threshold alerts only (n = 177), and threshold plus predictive alerts (n = 178). The primary clinical endpoint of the study was the percentage of time that these subjects spent within the TIR of 70-180 mg/dL during their stay in the hospital. Intent-to-treat statistical analysis has been applied with strict Holm adjustment for multiple comparisons.
Key findings:
- Among the 533 randomized participants, 331 (62.1%) were male, the mean age was 62 ± 12 years, and the mean baseline hemoglobin A1c was 9.2% ± 2.1%.
- Mean time in target range (70–180 mg/dL) was significantly higher in the threshold alerts group than in the group with all alerts off (75.1% ± 16.6% vs 71.2% ± 18.1%; adjusted mean difference, 3.9%; 95% CI, 0.1%–7.8%; Holm-adjusted P = .04).
- Time spent above range (>180 mg/dL) was significantly reduced in the threshold alerts group compared to the all-alerts-off group (−3.9%; 95% CI, −7.1% to −0.7%).
- Time spent below range (<70 mg/dL) did not differ significantly between the threshold alerts group and the all-alerts-off group (−0.04%; 95% CI, −0.3% to 0.2%).
- No statistically significant differences in primary or secondary outcomes were observed between the threshold plus predictive alerts group and either of the other two study groups.
Reference:
Wang Y, Cao L, Lu J, et al. Threshold and Predictive Alerts of Continuous Glucose Monitoring and Glycemic Control in Hospitalized Adults With Diabetes: A Randomized Clinical Trial. JAMA Netw Open. 2026;9(8):e2629291. doi:10.1001/jamanetworkopen.2026.29291
