Skip to main content
. 2005 Feb 19;330(7488):389. doi: 10.1136/bmj.330.7488.389

Table 5.

Risk of non-fatal self harm and suicide in relation to time since starting tricyclic antidepressant monotherapy or SSRI monotherapy (all ages)*

Non fatal self harm
Completed suicides
Exposure Cases (n=1173) Controls (n=18 537) Crude odds ratio (95% CI) Adjusted odds ratio*(95% CI) Cases (n=29) Controls (n=607) Crude odds ratio (95% CI) Adjusted odds ratio*(95% CI)
≤14 days:
Tricyclic antidepressants 118 1643 1 1 5 48 1 1
SSRIs 242 4214 0.76 (0.60 to 0.97) 0.77 (0.61 to 0.98) 5 117 0.51 (0.12 to 2.08) 0.45 (0.10 to 2.06)
15-30 days:
Triyclic antidepressants 67 1231 1 1 1 36
SSRIs 208 3482 1.10 (0.82 to 1.48) 1.11 (0.82 to 1.49) 4 69
31-60 days:
Tricyclic antidepressants 77 1183 1 1 5 87 1 1
SSRIs 233 3395 1.10 (0.84 to 1.44) 1.17 (0.88 to 1.54) 4 144 0.49 (0.12 to 1.95) 0.44 (0.10 to 2.01)
>60 days:
Tricyclic antidepressants 57 844 1 1 4 30 1 1
SSRIs 171 2545 1.08 (0.79 to 1.49) 1.08 (0.78 to 1.50) 4 76 0.39 (0.08 to 1.86) 0.33 (0.07 to 1.66)
*

Participants exposed to more than one antidepressant or prescribed non-SSRI, non-tricyclic antidepressants (see table 4) included in models but, for presentational purposes, data are not given in table.

Adjusted for severity of depression; time depression was diagnosed in relation to start of therapy; referral to psychiatrist or psychologist before index day; history of self harm; diagnosis of, or treatment for, anxiety or panic disorder; schizophrenia; antipsychotic drugs; drug misuse; and alcohol misuse.