Date of Award


Degree Name

PhD Nursing

Dissertation Committee

Patricia A. Roth, EdD, RN, Chairperson; Robert V. Topp, PhD, RN, Member; Brenda N. Boone, PhD, RN, Member


Dementia, Palliative Care, Readmissions


Background: Dementia represents a global health priority; there were 46.8 million diagnosed older adults reported worldwide in 2015, a number will likely double every 20 years. The costs of treatment and care for older adults with dementia are expected to rise since patients often require frequent hospitalizations to treat symptoms related to comorbid conditions. The increased frequency of hospitalization results in more care transitions that create strain upon patients, their families, and the overall health care system with a poorer recovery outlook for rehospitalizations.

Purpose: The purpose of this study is to identify biological and other factors that contribute to 30-day hospital readmissions (HR) among older adults with any type of diagnosed dementia.

Methods: This descriptive cross-sectional correlation study used an existing electronic health record (EHR) database of adults aged 50 years or older who received care from a 444-bed acute care community hospital in southern California. Demographic factors included age, gender, and race/ethnicity. Co-morbidities, fall risk, skin risk, discharge destination, length of stay (LOS), payer source, primary language, and Palliative Care (PC) consultations were also considered. Relationships between study factors and 30-day hospital readmissions (HR) were calculated. Variables that showed statistically significant correlation were included in a logistic regression model analysis to calculate odds ratios related to 30-day HR.

Results: Readmitted older adults were 84.9 years old, vs. 85.7 years for those not readmitted. Among the readmitted, a majority were female (75.5%) and white (89.7%). Nominal statistical significance in fall risk (p < .05), skin risk (p < .05), LOS (p < .05), and number of comorbidities (p < .05) predicted 30-day HR. However, PC consultation use decreased odds of 30-day HR by 41.4%, 95% CI [0.43, 0.80]. The overall model showed an 89.4% accuracy rate for predicting 30-day HR.

Conclusions: Identifying factors associated with 30-day HR for older adults with dementia may lead to nursing care delivery model innovations to improve care transitions and prevent 30-day hospital readmissions. Implications for further research include how to optimize EHR functionality and documentation strategies in order to identify patients at risk for 30-day HR.

Document Type

Dissertation: USD Users Only