LB049 - IMPACT OF A NATIONAL REIMBURSEMENT INCENTIVE ON EARLY ENTERAL NUTRITION IN CRITICALLY ILL PATIENTS: AN INTERRUPTED TIME SERIES ANALYSIS

LB049

IMPACT OF A NATIONAL REIMBURSEMENT INCENTIVE ON EARLY ENTERAL NUTRITION IN CRITICALLY ILL PATIENTS: AN INTERRUPTED TIME SERIES ANALYSIS

M. Yoshida^1,2,*, S. Shimizu², N. Kanazawa¹, Y. Hashimoto¹, R. Sugi¹, N. Inoue³, K. Fushimi², T. Goto²

¹Clinical Research Center, National Hospital Organization Headquarters, Tokyo, ²Department of Health Data Science, Graduate School of Data Science, Yokohama City University, Yokohama, ³Department of Clinical Epidemiology , Showa Medical University, Tokyo, Japan

Rationale: Early enteral nutrition (EN) is recommended by international guidelines for critically ill patients. In Japan, a national reimbursement incentive introduced in April 2020 rewards initiation of early EN within 2 days of intensive care unit (ICU) admission. We evaluated the effectiveness of this policy in increasing the proportion of early EN at a national level. 

Methods: We used an administrative database covering approximately 90% of acute care hospitals in Japan. We included patients urgently admitted to an ICU between April 2018 and March 2024 who stayed in the ICU for at least three days. The outcome was initiation of early EN within 2 days of admission. April 2020, was set as the start of the intervention period, with the period before this set as baseline, and an interrupted time series (ITS) analysis was performed. After multiple imputations, we used modified Poisson regression with cluster-robust variance, adjusting for patient and hospital characteristics, treatments including catecholamine use, and the proportion of COVID-19. 

Results: A total of 223,569 patients were included (74,343 before and 149,226 after the incentive). The median age was 73 years; 28% received mechanical ventilation. Noradrenaline and dobutamine were used in 24.5% and 13.7%, respectively. The unadjusted rate of early EN was 47.8% before and 52.4% after the incentive (p <0.001). In the adjusted ITS analysis, the level change at intervention was a risk ratio of 1.05 (95% confidence interval 1.017–1.085), while the pre-intervention trend (1.00, 0.999–1.002) and post-intervention slope change (1.00, 0.999–1.002) showed no significant change. 

Image:

Conclusion: Among urgently admitted critically ill patients, introduction of the EN reimbursement incentive was associated with a modest immediate increase in initiation of early EN, without evidence of a sustained increase thereafter.

Disclosure of Interest: None declared