Impact of Feeding Mode on Survival and 28-day Hospital Readmissions in Patients with Advanced Cognitive Impairment with Feeding Problems: A Retrospective Cohort Study

Abstract

Background: Nasogastric feeding tubes are commonly inserted in patients with advanced cognitive impairment for enteral feeding when feeding problems arise. Careful hand feeding has been advocated as an alternative due to uncertain benefits of enteral tube feeding weighed against its potential burdens. The impact of feeding mode on survival and hospital readmissions have not been previously studied in this patient population in Hong Kong. Objective: To compare the impact of nasogastric tube feeding versus oral feeding on survival and 28-day hospital readmissions in hospitalized patients with advanced cognitive impairment with feeding problems. Methods: A retrospective study of 454 patients with advanced cognitive impairment ≥ 60 years old admitted to two convalescent hospitals from February 2017 to June 2019 who meet indication for nasogastric tube insertion due to feeding problems, with a minimum follow up of 1 year. Patients initiated on nasogastric tube feeding are compared with those continued on oral feeding. Chart reviews collected data on the feeding mode and the outcomes, as well as other demographic and clinical information. Survival analysis was performed using the Kaplan-Meier method. Prognostic factors for mortality were identified using multivariate Cox proportional hazards regression models. Results: Of the 454 patients, 228 (50.2%) were initiated on nasogastric tube feeding and 226 (49.8%) were continued on oral feeding after rejecting nasogastric tube feeding recommendation. The mean age was 89.2 ± 7.4 years, 60.4% were women, and 72.7% were residents in residential care homes for the elderly. Median survival was 138 days (95% CI = 84-191 days). After multivariate analysis using Cox regression, there was no difference in survival between the two groups (AHR = 1.08, 95% CI =0.85-1.37). The prognostic factors for mortality were age (p=.033, HR = 1.75), albumin level (p=.028, HR = 1.63), and Norton score (p=.035, HR = 1.90). There was no difference in 28-day hospital readmission rates (47.8% vs 41.6%, p=.183). Conclusion: The use of nasogastric tube feeding compared with oral feeding was not associated with improved survival in patients with advanced cognitive impairment with feeding problems. No difference in 28-day hospital readmissions was observed between the feeding modes.