The desire to die in palliative care: a sequential mixed methods study to develop a semi-structured clinical approach

Table 3 Clinical approach domains and Delphi survey importance ratings

	Mean (Standard Deviation)			Consensus^a
	Round 1^b	Round 2^c	p	Round 1	Round 2	Increase
A – Usage Notes
1. Usage notes^d	–	4.32 (0.91) [5]	–	–	92.6%	–
B – Conversation Aspects
2. Actively building the relationship	4.64 (0.85)	–	–	92.6%	–	–
3. Proactively addressing desire to die	4.01 (0.94)	4.16 (0.92)	< 0.05	74.5%	83.2%	8.7%
4. Closure of discussion	4.62 (0.74)	–	–	92.6%	–	–
5. After discussion	4.64 (0.65)	–	–	94.0%	–	–
C – Classification, Meaning and Functions
6. Classification of desire to die	4.26 (1.0)	4.37 (0.80)	0.10	85.2%	90.6%	5.4%
7. Background and meanings of desire to die	4.81 (0.50)	–	–	97.3%	–	–
8. Functions of desire to die	4.31 (1.07)	4.64 (0.73)	< 0.01	83.9%	95.3%	11.4%
D – (Self-)Reflection
9. Conscious engagement with own attitudes and emotions	4.77 (0.53)	–	–	97.3%	–	–
10. Self-protection	4.74 (0.53)	–	–	96.0%	–	–
E – Further Recommended Action
11. Further recommended action	4.53 (0.85)	4.68 (0.56)	0.07	87.9%	95.3%	7.4%

^a Likert scale items were labeled ‘5’ (‘very important’) to ‘1’ (‘not important at all) with the option to report ‘don’t know’ (exclusion from analysis). Consensus was assumed if participants rated domains with ‘4’ (‘quite important’) or ‘5’ (‘very important’). Percentages are quotas of all participants who answered a respective question, not of the entire sample
^b For all ratings the full range of possible answers was used except for ‘conscious engagement with own attitudes and emotions’ (Min = 2, Max = 5) and ‘self-protection’ (Min = 3, Max = 5)
^c For all ratings the full range of possible answers was used expect for ‘further recommended action’ (Min = 2, Max = 5)
^d Domain added after round 1

ISSN: 1472-684X