Abstract

Optimizing end‐of‐life (EOL) care involves integrating decisions at various levels for patients and palliative care (PC) physicians. Measuring “quality” in PC delivery is therefore challenging and often not assessed. One focus of quality is to measure features of communication that help us understand patients’ personal values and beliefs that influence choices fostering patient‐centered treatment. These can be derived from the vocabulary patients use, and we can use this knowledge to differentiate and respond more accurately to the needs of patients. The objective of this study was to relate deep moral values to concrete moral actions, and we mine a large‐scale, human‐generated dataset to investigate morally relevant phenomena in a high‐stake real‐world decision context.We used the Moral Foundations Dictionary, an established list of moral words in the English language, to extract moral words from the text using Natural Language Processing. With a latent class analysis, we explored if there were qualitatively different underlying patterns in the PC patient population. We used count models to analyze different types of morality in the conversations and explore if heterogeneous patterns of morality terms in patients.We are using data from the PCCRI which was designed to understand the relation between clinical communication and patient‐centered outcomes. The 6‐month cohort data include directly observed and audio‐recorded palliative care consultations (up to first 3 visits); patient/proxy and clinician self‐report questionnaires (such as patient perceptions and preferences; clinician perceptions) both before and the day after consultation; post‐consultation in‐depth interviews; and medical/administrative records (such as disease status; treatment utilization).We found two latent classes: class one in which patients did not use many expressions of morality (about two thirds) in their PC consultations and class two in which patients did (one third). Age, race, education, spiritual needs, and whether a patient was affiliated with Christianity or another religion were all associated with class membership. Gender, financial hardship, and preference for longevity‐focused over comfort‐focused treatment near EOL did not affect class membership. In our count models, we also found that some patient characteristics were associated with the use of different moral terminology.This study is among the first to use text data from a real‐world situation to extract information regarding individual moral expressions and the relation with patient characteristics, attitudes, and emotions.The results of this study are relevant to those who seek to improve the quality of communication in order to achieve better values‐concordant treatment at EOL. From text data, this study identifies moral values among patients in a palliative care setting. The effect of moral values on EOL care preference resulting in PC treatment utilization illustrates the contribution to empirical investigation of human motivations related to morality. It highlights the importance of differentiating prognosis communication with respect to patients’ different moral values influencing their EOL preferences. This study also emphasizes the importance of using text data for health services research in addition to existing quantitative and qualitative HSR methods.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call