Preferences for Life-Sustaining Treatments and Associations With Accurate Prognostic Awareness and Depressive Symptoms in Terminally Ill Cancer Patients' Last Year of Life

• Published in: Journal of pain and symptom management 2016-01, Vol.51 (1), p.41-51.e1
• Main Authors: Tang, Siew Tzuh, DNSc, Wen, Fur-Hsing, PhD, Hsieh, Chia-Hsun, MD, Chou, Wen-Chi, MD, Chang, Wen-Cheng, MD, Chen, Jen-Shi, MD, Chiang, Ming-Chu, MS
• Format: Article
• Language: English
• Subjects: Adult; Aged; Anesthesia & Perioperative Care; Awareness; Depression; Disease Progression; end-of-life care; Female; Humans; Life Support Care - methods; Life Support Care - psychology; life-sustaining treatments; Linear Models; Logistic Models; Longitudinal Studies; Male; Middle Aged; Neoplasms - diagnosis; Neoplasms - psychology; Neoplasms - therapy; Pain Medicine; Patient Preference - psychology; Preferences; Prognosis; stability; Taiwan; Terminal Care - methods; Terminal Care - psychology; Terminally Ill - psychology; terminally ill cancer patients; Treatment Refusal - psychology
• ISSN: 0885-3924; 1873-6513
• DOI: 10.1016/j.jpainsymman.2015.08.006

Abstract Context The stability of life-sustaining treatment (LST) preferences at end of life (EOL) has been established. However, few studies have assessed preferences more than two times. Furthermore, associations of LST preferences with modifiable variables of accurate prognostic awareness, physician-patient EOL care discussions, and depressive symptoms have been investigated in cross-sectional studies only. Objectives To explore longitudinal changes in LST preferences and their associations with accurate prognostic awareness, physician-patient EOL care discussions, and depressive symptoms in terminally ill cancer patients' last year. Methods LST preferences (cardiopulmonary resuscitation, intensive care unit [ICU] care, intubation, and mechanical ventilation) were measured approximately every two weeks. Changes in LST preferences and their associations with independent variables were examined by hierarchical generalized linear modeling with logistic regression. Results Participants ( n  = 249) predominantly rejected cardiopulmonary resuscitation, ICU care, intubation, and mechanical ventilation at EOL without significant changes as death approached. Patients with inaccurate prognostic awareness were significantly more likely than those with accurate understanding to prefer ICU care, intubation, and mechanical ventilation than to reject these LSTs. Patients with more severe depressive symptoms were less likely to prefer ICU care and to be undecided about wanting ICU care and mechanical ventilation than to reject such LSTs. LST preferences were not associated with physician-patient EOL care discussions, which were rare in our sample. Conclusion LST preferences are stable in cancer patients' last year. Facilitating accurate prognostic awareness and providing adequate psychological support may counteract the increasing trend for aggressive EOL care and minimize emotional distress during EOL care decisions.