PreCare: Designing AI assistants for Advance Care Planning (ACP) to enhance personal value exploration, patient knowledge, and decisional confidence
Computers in Human Behavior Reports · 2026-12 · 24 · 101388
DOI: 10.1016/j.chbr.2026.101388 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computers in Human Behavior Reports, 2026-12. Verfasst von Yu Lun Hsu und 11 weiteren Autor:innen. Die Arbeit nennt 105 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Council, Ministry of Education.
Themen: Ethik in der Pflege
Abstract (Originalsprache)
Advance Care Planning (ACP) is a process that enables individuals to clarify their values, define goals and preferences for future medical care, discuss them with family and health-care providers, and record and review them. Despite high reach, today's ACP websites and apps are static : presenting information and collecting responses without using user input to (i) deepen exploration of personal values, (ii) ensure sufficient ACP knowledge, or (iii) help users reflect on potential consequences of their decisions for themselves (e.g., quality of life) and others (e.g., caregiver burden) before recording their preferences. This research was conducted in Taiwan. We first conducted two formative studies: (1) shadowing five ACP group consultations delivered by three multidisciplinary clinical teams (each comprising a physician, a nurse, and a social worker) at a tertiary medical center, followed by semi-structured interviews with six ACP professionals (three physicians, two social workers, one nurse); and (2) interviewing 14 users of a static ACP website. Leveraging these insights, we developed PreCare in collaboration with ACP professionals, with a three-assistant conversational workflow: a semi-structured value-elicitation assistant; an in-context knowledge assistant based on a clinician-validated corpus; and a non-directive consequence-review assistant. A usability study (n = 12; 5 male, 7 female; age range = 22–64 years, mean = 46, SD = 17) showed that PreCare achieved a usability rating (SUS) of excellent. In an exploratory comparative study (n = 12; 5 male, 7 female; age range = 26–73 years, mean = 52, SD = 17), interactive assistance showed preliminary evidence of improved exploration of personal values, perceived knowledge, and decisional confidence, and was preferred by 92% of participants. We contribute design-oriented recommendations for interactive, non-directive support for reflection, knowledge, and consequence review for ACP and other decisional support systems.
Förderung
- National Science and Technology Council
- Ministry of Education
Zitieren
APA
Hsu, Y. L., Chou, Y., Chang, C., Chang, Y., Lee, Z., Gipiškis, R., Li, R., Shih, C., Peng, J., Huang, H., Tsai, J., & Chen, M. Y. (2026). PreCare: Designing AI assistants for Advance Care Planning (ACP) to enhance personal value exploration, patient knowledge, and decisional confidence. Computers in Human Behavior Reports, 24, 101388. https://doi.org/10.1016/j.chbr.2026.101388
Vancouver
Hsu YL, Chou Y, Chang C, Chang Y, Lee Z, Gipiškis R, et al. PreCare: Designing AI assistants for Advance Care Planning (ACP) to enhance personal value exploration, patient knowledge, and decisional confidence. Computers in Human Behavior Reports. 2026;24:101388. doi:10.1016/j.chbr.2026.101388
BibTeX
@article{Hsu2026,
author = {Hsu, Yu Lun and Chou, Yun-Rung and Chang, Chiao-Ju and Chang, Yu-Cheng and Lee, Zer-Wei and Gipiškis, Rokas and Li, Rachel and Shih, Chih-Yuan and Peng, Jen-Kuei and Huang, Hsien-Liang and Tsai, Jaw-Shiun and Chen, Mike Y.},
title = {PreCare: Designing AI assistants for Advance Care Planning (ACP) to enhance personal value exploration, patient knowledge, and decisional confidence},
journal = {Computers in Human Behavior Reports},
year = {2026},
volume = {24},
pages = {101388},
doi = {10.1016/j.chbr.2026.101388}
}
EUVIMED