Table 1

Proportion (%) of physicians deciding to withdraw a treatment in the given patient scenarios according to specialty

			Speciality
Treatment	Scenario	All	Surgery (N=175)	Medicine (N=184)	GP (N=316)	Oncology (N=54)
Withdraw, 4–5 on Likert scale 1–5 “definitely would not withdraw to definitely would withdraw”. Statistical significance by χ² test in cross-tabulation to three decision categories (Likert 1–2, 3 and 4–5). Differences between original scenario, alternative 1, and alternative 2 were all statistically significant (p<0.001) by Friedman's test.
Antibiotics	Original scenario (OS)	60	67	57	57	68
	A1 (OS + family's appeal)	40	54	38	33	36	p<0.001
	A2 (OS + advance directive)	88	89	87	88	85

Iv hydration	Original scenario (OS)	29	21	23	31	60	p<0.001
	A1 (OS + family's appeal)	18	16	16	17	34	p<0.001
	A2 (OS + advance directive)	56	44	53	60	81	p<0.001

Nasogastric tube	Original scenario (OS)	54	58	55	49	70	p=0.008
	A1 (OS + family's appeal)	43	51	45	35	56	p<0.001
	A2 (OS + advance directive)	74	74	74	71	89	P=0.043

Thrombos prophylaxis	Original scenario (OS)	81	82	87	74	93	p=0.002
	A1 (OS + family's appeal)	72	79	75	63	85	p=0.002
	A2 (OS + advance directive)	93	94	95	89	100

Mechanical ventilation	Original scenario (OS)	79	79	82	79	93	p=0.017
	A1 (OS + family's appeal)	64	65	67	58	85	p=0.017
	A2 (OS + advance directive)	92	90	91	93	98

Supplementary oxygen	Original scenario (OS)	13	18	10	10	23	p=0.034
	A1 (OS + family's appeal)	9	12	7	7	11	p=0.003
	A2 (OS + advance directive)	33	39	36	27	40	p=0.048