Date of Award

2018

Embargo Period

8-1-2024

Document Type

Dissertation

Degree Name

Doctor of Health Administration

College

College of Health Professions

First Advisor

Kit N. Simpson

Second Advisor

Annie N. Simpson

Third Advisor

Scott W. Goodspeed

Fourth Advisor

James S Zoller

Abstract

The study objective is to develop a predictive model for pediatric inpatient days based on ambulatory outpatient visits and emergency department visits. This model aims to study the relationship between ambulatory visits and inpatient days, and determine if in-patient days can be predicted based on sub-specialty practice. Such a model does not currently exist, and when created and validated such a model could be utilized for various important management decisions, including refined insight into inpatient capacity and operational efficiency for self-governing children’s hospitals with large sub-specialty practices. The data set was a sample of convenience from one health system in the PEDSnet database. The requested data set yielded 3,832,428 distinct records, inclusive of all billed encounters for January through December 2017. Multi-regression analysis was used to predict variations in weekly occupied days over time. Ordinary least squared regression model results were used to examine the predictive power of outpatient variables. This enabled comparison of beta values for as many combinations of predictors as possible, in an efficient manner and yielded 80 models. The conclusion was that big data from one children’s hospital within a children’s health system was able to predict in-patient occupancy for greater than 50% of the variance.

Rights

All rights reserved. Copyright is held by the author.

Download

Share

COinS