| Authors (year) | Outcome | Study result | Conclusions | Predictors in OCD of suicide | Limitations | Possible bias |
| Alonso et al. 2010 | 11 suicide attempts, 2 completed suicides | 2/ 218 (0.9%) committed, 11/218 (5%) attempted suicide within the observation period | Suicide is not highly common but should not be disregarded. | Suicide behaviour associated with being unmarried, history of affective disorders and symmetry/ ordering obsessions. | Selected intensively treated clinic sample | Selection bias |
| Angst et al. 2005, re-reported by Fineberg et al. 2013 | 1 patients with OCD/OCS and 3 control subjects attempted suicide in 20 years | 3.1% of with subjects OCD or OCS (without comorbidity) and 2.3% control subjects had attempted suicide | Suicide rates not significantly increased in subjects with pure, non-comorbid OCD/ OCS. | Comorbid OCD/ OCS with was significantly associated higher levels of treatment seeking, impairment,distress and suicidality compared with pure OCD/OCS. | Selected healthy subjects at intake. | Minor selection bias, some study attrition. |
| Bolton et al. 2008 | 3 suicide (6.3%) attempts in OCD subjects over 13 years, 16 in 1229 subjects without anxiety disorders | OCD did not significantly increased in subjects who attempted suicide. | Any anxiety disorders at baseline was significantly associated with subsequent onset suicide attempts (adjusted odds ratio 2.20. | Predictors not individually assessed for OCD. | No specific assessment for OCD. | Low power for OCD subsample |
| Fernandez de la Cruz et al. 2017 | 4297 suicide attempts (11.7%) and 545 suicides (1.5%) according to Swedish Cause of death Register | Increased risk of suicide and suicide attempts (OR 9.8 resp. 5.5) in comparison with controls. | Increased risk of suicide attempts and completed suicide in OCD even after accounting for psychiatric comorbidities. | A previous suicide attempt, comorbid substance abuse or personality disorder increased the risk of death by suicide | Changes of diagnostic criteria over time, some attrition due to loss to register | Minor diagnostic imprecision but no major bias likely |
| Rueck et al. 2017 | 2 out of 29 deceased subject had died of suicide (plus one suspected suicide) | There is a relevant suicide risk even though there is no comparison with control subjects. | Attention needs to be given to possible suicide in OCD. | Predictors impossible to assess due to low sample size. | Highly selected sample | Selection bias likely |
| Sareen et al. 2005. update reported ten Have et al. 2009 | 2 out of 31 OCD subjects (3.3%) and 39/ 4670 of all subjects at risk (0.83%) attempted suicide during follow up. | Non-significant increase in suicide attempts in subjects wth OCD. | No clear conclusion possible | Predictors not assessed for OCD. | Limited power due to the rarity of the possible event. | Lack of statistical power |