Euthanasia and other end-of-life decisions: a mortality follow-back study in Belgium

BMC Public Health

Table 2 Frequency of end-of-life decisions in non-sudden deaths in Belgium, according to community

	Dutch-speaking community		French-speaking community				Belgium total
	N	%	N	%	OR^b	95%CI	N	%	95%CI^d
Most important end-of life decision that possibly hastened death (ELD)
Number of deaths studied ^a	1007		637				1644
Euthanasia/assisted suicide^e	16	1.6	6	0.9	1.70	[0.66–4.36]	22	1.3	[0.8–2.0]
Administering life-ending drugs without explicit patient request	17	1.7	9	1.4	1.20	[0.53–2.71]	26	1.6	[1.0–2.3]
Intensified alleviation of symptoms	295	29.3	160	25.1	1.21	[0.96–1.53]^c	455	27.7	[25.5–29.9]
Withholding or withdrawing of life-prolonging treatment	177	17.6	89	14.0	1.33	[0.999–1.77]^c	266	16.2	[14.4–18.1]
Total ELD	505	50.1	264	41.4	1.43	[1.16–1.77]	769	46.8	[44.3–49.2]
Total ELD with partly or explicitly life-shortening intent	153	15.2	70	11.0	1.42	[1.05–1.93]	223	13.6	[11.9–15.3]
Continuous deep sedation ^f
Number of deaths studied ^a	999		630				1629
All sedation	82	8.2	95	15.1	0.50	[0.37–0.69]	177	10.9	[9.4–12.5]
forgoing artificial food/fluid	36	3.6	37	5.9	0.60	[0.37–0.96]	73	4.5	[3.5–5.6]
administering food/fluid	46	4.6	58	9.2	0.42	[0.28–0.64]	104	6.4	[5.2–7.7]

^a Missing data: all questions concerning ELDs unanswered n = 46 (2.7%) (of which 43 cases had died in the hospital: 24 Dutch and 19 French-speaking); question sedation unanswered n = 61 (of which 56 had died in a hospital: 29 Dutch and 27 French-speaking)
^b Odds ratios based on logistic regression with community as predictor (French = ref cat) and controlling for patients' age, sex, educational level, cause of death, and place of death
^c The probability was significant if alpha = .10 (i.e. level of the CI is 90%)
^d Binomial Confidence Interval; exact method
^e Physician-assisted suicide was limited to one case
^f Sedation was provided in conjunction with any possibly death-hastened decision for 7.9%, and Dutch- and French-speaking communities did not differ.

ISSN: 1471-2458