Abstract
Advance care planning (ACP) is the process of communicating about end-of-life (EOL) care with loved ones. Due to the deadly nature of COVID-19, ACP is vital. Unfortunately, fewer than 30% of American patients engage in ACP. In addition to low motivation, people experiencing death anxiety (DA) similarly avoid ACP. This finding coincides with predictions from terror management theory (TMT) that people avoid DA-arousing behaviors. Guided by the theory of planned behavior (TPB) and the health belief model (HBM), we posited COVID-19 risk perceptions would be positively associated with determinants of health behavior, including intention to share and ask loved ones about EOL wishes, as well as the associated attitudes, norms, and level of perceived behavioral control regarding ACP. Guided by TMT, we posited that DA negatively mediated relationships between COVID-19 risk perceptions and these behavioral determinants. An MTurk participant sample (N = 522) completed a survey about COVID-19 risk perceptions, DA, and health behavior determinants in the context of ACP. Results indicate COVID-19 risk perceptions were rarely directly related to these determinants. However, results of PROCESS mediation models examining the role of DA in these relationships reveal a different picture. While direct relationships were rarely significant, DA negatively mediated most relationships between COVID-19 risk perceptions and behavioral determinants. Our results indicate DA demotivates EOL communication during the COVID-19 pandemic; a concerning yet important finding due to the increased importance of ACP in the context of a deadly disease like COVID-19.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Supplementary material
Supplemental data for this article can be accessed on the publisher’s website.
Notes
1 Preliminary analyses showed no difference in death anxiety scores between those who answered the scale early vs. late. However, those who took the scale early reported more favorable attitudes toward both discussing with family members one’s own EOL wishes (M = 4.74 vs. 4.53, p = .043) and discussing with family members their EOL wishes (M = 4.49 vs. 4.28, p = .045). We thus decided to include early vs. late measurement as a control variable in all analyses to control for potential confounds.