Abstract

A complex age-malnutrition interaction and the risk for 7-day unplanned pediatric readmission

Author: Louis Ehwerhemuepha

Background: This study provides an analysis of the role of malnutrition and nutritional deficiencies on the risk of 7-day unplanned pediatric readmission adjusting for patient demographics and socioeconomic status. Methods: We obtained data from all inpatient encounters within a 2-year period at CHOC Children’s Hospital, Orange excluding all neonates and patients older than 21. The corresponding data retrieved include demographics (gender, age, race and ethnicity), type of health insurance as proxy for socioeconomic status (SES), acuity of admission, prior ED visits, length of stay, and indicator of whether an inpatient encounter was planned or not. While adjusting for demographics and SES, we evaluated all possible models consisting of malnutrition/nutritional deficiencies and the LACE readmission variables (length of stay, acuity of admission, Charlson Comorbidity Index Score (CCIS), and number of ED visits within prior six months) in addition to all possible 2-way statistical interactions between demographics/SES and malnutrition. The final model was selected using best subset regression. Results: The best fitting model indicate that race and ethnicity, all LACE readmission variables, and a complex 2-way interaction between age and malnutrition are significantly associated with a patient’s odds of being readmitted within 7-days of a previous hospitalization. The main effects of the patient’s age and malnutrition status, together with the corresponding 2-way interactions between these variables indicate that the risk of being readmitted differs by age. For all pediatric age ranges, malnourished patients have greater odds of being readmitted with the effect size larger in younger patients. This 2-way interaction is statistically significant only for 31-day to 16-year old patients. Conclusion: Malnourished patients have greater odds of being readmitted and can be identified for interventions to improve general health outcomes and reduce the odds of being readmitted within 7-days of a previous hospitalization leading to better care for patient and reduced cost for the hospital.

Co Author/Co-Investigator Names/Professional Title: William Feaster, MD, MBA, CMIO Duncan Yeung Neil Garde