Date of Award

Spring 2018

Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Guttormson, Jill L.

Second Advisor

Garnier-Villarreal, Mauricio

Third Advisor

Schroeter, Kathryn


The proportion of older adults in the U.S. is rapidly increasing. One-third of Medicare expenditures occur in the final year of life, with nearly half resulting from acute exacerbations of chronic, progressive diseases(Riley & Lubitz, 2010). Older adults prefer comfort over life-sustaining care, and decreased intensity of care is associated with improved quality of life at the end-of-life (EOL). Advance directives (ADs) have been proposed as mechanisms to improve congruence between patient wishes and EOL care; however, the impact of ADs on care delivered in the acute care setting at the EOL for this population is unclear. A retrospective, correlation design framed by the Quality Health Outcomes Model was used to describe (a) the relationship between ADs and the intensity of care received by older adults in the acute care setting at the EOL, and (b) the congruence between patient preferences within ADs and actual care received. Four hundred and ninety-six patients, aged 65 and older who died while admitted to a large, academic medical center, were identified using electronic health records. Regression analyses, to determine the association between ADs and indicators of intensity of care, and content analysis, to describe congruence of care, were conducted. Advance directives were not independently associated with any indicators of high-intensity (i.e., high-cost, high-technology) care. While ADs were independently associated with palliative and hospice referrals, effect sizes were small, and referral timing was late. In a subset of one hundred patients with ADs, less than half received care that was congruent with documented preferences. In approximately one-fourth, patient preferences were vague, and congruence could not be determined. Advance directives may be ineffective, in their current form, to decrease aggressive care in the acute care setting. Further research is necessary to determine whether this is a function of how ADs are used within acute care or ambiguous preferences within the document. A shift in the approach of healthcare providers may be necessary to promote engagement in advanced care planning discussions, with patients and family members, with an AD serving as the product of those discussions rather than a document completed out of context.