Course of bereavement over 8-10 years in first degree relatives and spouses of people who committed suicide: longitudinal community based cohort study

The authors of this Research paper have alerted us to some errors in table 4 (BMJ 2013;347:f5519, doi:10.1136/bmj.f5519). They reported a regression coefficient (B) and 95% CI of 4.0 (1.5 to 6.4), P=0.002, for the association between mastery and depression, when this should have read 0.4 (0.2 to 0.5), P<0.001. In addition, the B (95% CI) for the association between mutual support use and complicated grief was erroneously reported to be 6.4 (1.8 to 11.0), P=0.006, when it should have read 5.9 (1.6 to 10.3), P=0.007. Finally, the B (95% CI) for the association between complicated grief and depression is missing in table 4. This association should have been 0.9 (0.7 to 1.0), P<0.001. The authors state that these altered values do not affect the general findings or overall conclusions of their study.