Violent and non-violent methods of attempted and completed suicide in Swedish young men: the role of early risk factors

Table 4 Methods of attempted suicide as predictor of subsequent completed suicide

	Suicide
	Model I^a	Model II^b	Model III^c
Suicide attempt methods	HR (95 % CI)	HR (95 % CI)	HR (95 % CI)
Poisoning (yes vs no)	11.61, 9.38-14.38	7.17, 5.54-9.28	4.27, 3.12- 5.84
Hanging, strangulation or suffocation (yes vs no)	27.42, 14.18-53.03	22.41, 10.59- 47.45	18.28, 8.58- 38.97
Firearm (yes vs no)	11.72, 2.92-47.00	8.09, 2.00- 32.77	5.18, 1.27- 21.24
Cutting or piercing (yes vs no)	13.44, 7.58-23.82	8.86, 4.83- 16.24	5.74, 3.07- 10.73
Jumping from a height (yes vs no)	16.25, 6.07-43.48	12.88, 4.77- 34.84	8.74, 3.20-23.85
Other methods (yes vs no)	4.54, 1.70-12.34	3.44, 1.28-9.26	2.18,0.80-5.94

Multivariate Cox proportional regression analyses. HR, 95 % CI
^aMultivariate model: including all methods in the model: poisoning, hanging, firearm, cutting, jumping and other methods.
^bAdjusted for early confounders measurement at conscription (nervous problems in the family, social class, fathers’ alcohol habits, medication for own psychiatric problems, psychiatric diagnoses at conscription, intelligence, emotional control, conduct problems, sniffing, alcohol use, drug use)
^cadjusted for early confounders measured at conscription (see above) and additional hospitalisation for alcohol and substance use

ISSN: 1471-244X