2039-P: Virtual Glucose Management Service Diminished Longer Length of Stay Associated with High A1C

Résumé fourni par la source

Introduction and Objective: Virtual glucose management service (VGMS) was found to decrease the rate of hyperglycemia and hypoglycemia in the hospital setting. In our study, we looked at VGMS’s effect on length of stay (LOS) for hospitalized people with diabetes (PWD). Methods: The study involved non critically ill hospitalized PWD between June 2022 and October 2024. A General Linear Model including LOS, VGMS consults and most recent A1c was fitted to assess if PWD’s most recent A1c would moderate the impact of VGMS consults on their LOS. Data was fit to model 1, with A1c as a continuous variable and as an interaction term with VGMS. In model 2, A1c was mean centered to assess the impact of VGMS on PWD’s LOS. LOS was also transformed using its log linear expression, as LOS was skewed right. Results: Among 13,816 PWD admitted to the hospital during the study period, 511 received VGMS consults (2.78% of the sample). The average raw LOS for PWD with VGMS consult and without VGMS were 10.07 and 5.79 days respectively. The results of an Independent Sample t-test indicated LOS for the VGMS group (log linear LOS mean = 1.98, SD = 0.030) was significantly greater than the non-VGMS Group (log linear LOS mean = 1.337, SD = 0.007; [t(730.130)=20.917, p < 0.01]). The general linear model fit the data well and demonstrated a significant interaction exists between VGMS and A1c (p <.05). This indicates the association between VGMS consults and LOS varies by A1c. The model showed PWD with average A1c levels had 0.61-day shorter lengths of stay than did patients with higher A1c. However, VGMS diminished the longer LOS seen at higher A1c. For example, for PWD whose A1c is 1 standard deviation above the mean (A1c value = 8.846), the length stay would decrease by 0.599 days, bringing their LOS very close to the average LOS of patients with A1c under 7.07. Conclusion: Our study showed that patients who had VGMS consult have significantly higher LOS as VGMS is indicated only in patients who have severe hyperglycemia and hypoglycemia. However, the model shows that VGMS diminished the projected longer LOS associated with high A1c. Disclosure N. Aung: Advisory Panel; Current; Novo Nordisk. M.E. Lape: None.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.