Dying to talk : methods and tools for conversations about dying, death and advance end-of-life care planning

Abstract

Introduction: Advance care planning has been described as a process of reflection and discussion about values and preferences for future end-of-life care. In Sweden, advance care planning is rarely systematically practiced and virtually unheard of among the general public. Furthermore, the Swedish context provides no judicial support for appointing proxy decision- makers or documenting legally binding advance directives. Thus, Sweden presents a context for advance care planning which is quite different from that in other countries where advance care planning has generally been studied. We therefore conceptualized an approach to advance care planning built on three cornerstones, as it involved early initiation of a conversation-based intervention targeting mostly older adults in the community.

Aims: The overarching aim of this thesis was to explore a structured intervention for advance care planning conversations about individual values and preferences for future end-of-life care among community-dwelling, older adults.

Methods: The project was planned with inspiration from participatory action research and data were mainly derived from qualitative interviews. A first database consists of initial interviews with 65 actively volunteering individuals (sub-studies I-III) and follow-up interviews with 59 of them 5-12 months later (III). Interviews at both time-points explored who and what individuals thought would be important to them at the end-of-life. This was explored initially with open questions followed by use of EcoMaps to map individuals’ social networks and a Swedish version of the GoWish cards, the DöBra cards, to discuss values and preferences for future end-of-life care. The translated and adapted DöBra cards contain 37 statements about that which may be important at the end-of-life, and wild cards to freely formulate any other individual preferences. While discussing their choices, participants were encouraged to sort and finally rank their 10 most important card statements. The initial process of establishing the intervention, along with data from EcoMaps and card rankings as well as participants’ reactions to the tools were described and analyzed in sub-study I. In sub- study II, participants’ reasoning about their values and preferences were analyzed, inspired by interpretive description. In sub-study III, a mixed-methods approach was employed to explore if and how participants’ reasoning and card rankings changed over time, using descriptive statistics, non-parametric analyses, and longitudinal qualitative analysis. In sub- study IV we were inspired by Ripple Effects Mapping as we analyzed data from additional interviews with 20 individuals who had used or promoted the DöBra cards outside of the research study, as they had become available to the general public due to popular demand. Starting with three known cases of card use in different settings, i.e. a national interest organization for older people, a patient organization, and a formal health care organization, we explored three ‘ripples’ of further dissemination and use of the DöBra cards.

Results: We found an openness to discussing topics of death, dying and future end-of-life care among the participants and engagement to promote discussions about these topics in other contexts (I, IV). EcoMaps could be valuable for highlighting social networks taken for granted, however participants were not enthusiastic about this exercise and sometimes even reluctant (I). In contrast, the DöBra cards were found to be feasible for stimulating both advance care planning conversations and more general discussions about death and dying (I, IV). Main benefits included offering a means to initiate these potentially difficult conversations by having pre-formulated statements to consider, and easing conversations by having something to do, rather than just talking (I, IV). Large individual variations between participants in card rankings as well as reasoning about card choices were noted, and we also found variations in the same individual’s ranking and reasoning over time (II-III). Wild cards were used to formulate individual preferences, although these formulations could also change over time (I-III). Issues related to assisted dying were common topics for wild cards (I-III). In sub-study IV we found use of the DöBra cards in varied community setting, e.g. study circles, high schools, book clubs, and in continued education for health care staff. Some resistance towards use and promotion of the cards was also noted. Impact of using the cards on an individual level included personal development and strengthening of relationships, private as well as professional. On a policy level, dissemination and use of the DöBra cards in new settings contributed to raising topics of death and dying in new contexts, where they had not previously been discussed (IV).

Conclusions: These findings indicate that the DöBra card deck is a feasible tool for those who wish to have conversations about death, dying and future end-of-life care. It should however be cautioned that ranking of cards alone is not sufficient to understand individuals’ underlying values and motivations for card choices. Thus, the card ranking exercise is when possible optimally accompanied by concurrent conversation about end-of-life values and preferences. The high degree of individual variation and also complex pattern in changes of reasoning and card choices over time found, supports conceptualizing advance care planning as a process of repeated conversations.