Published in:
01-02-2020 | Post-Traumatic Stress Disorder | Original Research
The Relationship of Post-traumatic Stress Disorder to End-of-life Care Received by Dying Veterans: a Secondary Data Analysis
Authors:
Kathleen E. Bickel, MD, MPhil, MS, Richard Kennedy, MD, PhD, Cari Levy, MD, PhD, Kathryn L. Burgio, PhD, F. Amos Bailey, MD
Published in:
Journal of General Internal Medicine
|
Issue 2/2020
Login to get access
Abstract
Background
Post-traumatic stress disorder (PTSD) can be exacerbated by subsequent trauma, but it is unclear if symptoms are worsened by impending death. PTSD symptoms, including hyperarousal, negative mood and thoughts, and traumatic re-experiencing, can impact end-of-life symptoms, including pain, mood, and poor sleep. Thus, increased symptoms may lead to increased end-of-life healthcare utilization.
Objectives
To determine if veterans with PTSD have increased end-of-life healthcare utilization or medication use and to examine predictors of medication administration.
Design
Secondary analysis of a stepped-wedge design implementation trial to improve end-of-life care for Veterans Affairs (VA) inpatients. Outcome variables were collected via direct chart review. Analyses included hierarchical, generalized estimating equation models, clustered by medical center.
Subjects
Veterans, inpatient at one of six VA facilities, dying between 2005 and 2011.
Main Measures
Emergency room (ER) visits, hospitalizations, and medication administration in the last 7 days of life.
Key Results
Of 5341 veterans, 468 (8.76%) had PTSD. Of those, 21.4% (100/468) had major depression and 36.5% (171/468) had anxiety. Veterans with PTSD were younger (mean age 65.4 PTSD, 70.5 no PTSD, p < 0.0001) and had more VA hospitalizations and ER visits in the last 12 months of life (admissions: PTSD 2.8, no PTSD 2.4, p < 0.0001; ER visits: 3.2 vs 2.5, p < 0.0001). PTSD was associated with antipsychotic administration (OR 1.52, 95% CI 1.06–2.18). Major depression (333/5341, 6.2%) was associated with opioid administration (OR 1.348, 95% CI 1.129–1.609) and benzodiazepines (OR 1.489, 95% CI 1.141–1.943). Anxiety disorders (778/5341, 14.6%) were only associated with benzodiazepines (OR 1.598, 95% CI 1.194–2.138).
Conclusions
PTSD’s association with increased end-of-life healthcare utilization and increased antipsychotic administration in the final days of life suggests increased symptom burden and potential for terminal delirium in individuals with PTSD. Understanding the burden of psychiatric illness and potential risks for delirium may facilitate the end-of-life care for these patients.
Trial Registration
NCT00234286