Predictors of High Hospital Costs among Patients with Chronic Illness who develop Acute Respiratory Failure (GP753)

Objectives Identify predictors of high hospital costs for patients with chronic illness who develop acute respiratory failure Recognize the complexity of identifying patients who may benefit most from targeted interventions to improve value-based care. Importance. Patients with acute respiratory fai...

Full description

Saved in:
Bibliographic Details
Published in:Journal of pain and symptom management Vol. 60; no. 1; p. 277
Main Authors: Khandelwal, Nita, Downey, Lois, Engelberg, Ruth, Curtis, J. Randall, May, Peter
Format: Journal Article
Language:English
Published: Madison Elsevier Inc 01.07.2020
Elsevier Limited
Subjects:
Anesthesia
Chronic illnesses
Costs
Dementia
Generalized linear models
Health services
Health services utilization
High risk
Hospital costs
Hospitalization
Inflation
Inpatient care
Intensive care
Intervention
Kidney diseases
Length of stay
Linear analysis
Mechanical ventilation
Medical diagnosis
Medical records
Morbidity
Objectives
Pain Medicine
Patient admissions
Patients
Proximity
Survivor
Value
Ventilation
ISSN:0885-3924, 1873-6513
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Objectives Identify predictors of high hospital costs for patients with chronic illness who develop acute respiratory failure Recognize the complexity of identifying patients who may benefit most from targeted interventions to improve value-based care. Importance. Patients with acute respiratory failure and underlying chronic illness are at high risk for burdensome care. Objective(s). Our objective was to identify predictors of high-cost, potentially burdensome care among patients hospitalized with acute respiratory failure. Method(s). Data were derived from a multi-hospital system electronic health record linked to hospital financial records (2011-2017). Eligible patients were age $18 years with at least one of nine Dartmouth Atlas chronic conditions and received mechanical ventilation for >48 hours in the intensive care unit (ICU). Our primary outcome was total hospital costs for the hospitalization with the qualifying ICU admission, adjusted for inflation and analyzed using generalized linear models. Predictors were number and type of chronic conditions, and prior healthcare utilization. Results. Our sample included 4,892 patients. Mean age was 59 (SD: 15); 36% were female. Mean number of chronic conditions was 2.3 (SD 1. 38). Median ICU length of stay was 12 days (IQR: 7, 22); median inpatient length of stay was 20 days (IQR: 11, 33). Median Apache II score was 22 (IQR: 12, 28). Median hospital costs were $135k (IQR: 78k, 228k). Hospital mortality was 35%. For survivors and decedents, hospital costs in the prior year were associated with higher costs during the admission of interest (b¼0.0004, p<0.001) and dementia was associated with lower costs (survivors: b¼-23.8, p¼0.018; decedents: b¼- 45.7, p<0.001). Among decedents, number of diagnoses was positively correlated with costs (b¼24.0, p<0.001); among survivors, chronic renal failure was positively correlated (b¼37.4, p<0.001). Conclusion(s). Higher hospital costs in the year prior to a hospitalization with acute respiratory failure was predictive of higher costs during the current admission. Individual conditions and multi-morbidity predict costs, but had a differential effect depending on survival. Impact. Patients at risk for high-cost hospitalizations may allow targeted interventions to improve value-based care. Identification efforts must address a complex web of determinants including diagnosis, proximity to death, multi-morbidity burden and prior healthcare use.
Bibliography:ObjectType-Conference Proceeding-1
SourceType-Scholarly Journals-1
content type line 14
ISSN:0885-3924
1873-6513
DOI:10.1016/j.jpainsymman.2020.04.180