The stigma of suicide

Abstract

comparison with a statistical model based comparison with a statistical model based on baseline variables.I would like to express on baseline variables.I would like to express three concerns about the technical details of three concerns about the technical details of this article.this article.First, the kappas they report are First, the kappas they report are Cohen's kappas whereby the disagreement Cohen's kappas whereby the disagreement between ``full recovery within 6 months'' between ``full recovery within 6 months'' and ``partial recovery'' is penalised equally and ``partial recovery'' is penalised equally to disagreement between ``full recovery to disagreement between ``full recovery within 6 months'' and ``no recovery''.within 6 months'' and ``no recovery''.Clinically, however, the former is apparently Clinically, however, the former is apparently less grave than the latter.More appropriate less grave than the latter.More appropriate statistics would be weighted kappas, which statistics would be weighted kappas, which are 0.31 (95% CI 0.15±0.46)for GP prog-are 0.31 (95% CI 0.15±0.46)for GP prognosis for depression, 0.35 (95% CI 0.16± nosis for depression, 0.35 (95% CI 0.16± 0.54) for GP prognosis for anxiety, 0.56 0.54) for GP prognosis for anxiety, 0.56 (95% CI 0.43±0.70)for model prognosis (95% CI 0.43±0.70)for model prognosis for depression and 0.51 (95% CI 0.33±0.69)for depression and 0.51 (95% CI 0.33±0.69)for model prognosis for anxiety.These for model prognosis for anxiety.These figures are appreciably larger than those figures are appreciably larger than those originally reported.originally reported.Moreover, regardless of whether we use Moreover, regardless of whether we use Cohen's kappas or weighted kappas, the Cohen's kappas or weighted kappas, the authors did not examine whether the GP authors did not examine whether the GP prediction is indeed statistically signifi-prediction is indeed statistically significantly cantly worse than the model's.The reported worse than the model's.The reported 95% confidence intervals overlap, and we 95% confidence intervals overlap, and we do not know whether the clinicians are actu-do not know whether the clinicians are actually performing worse than the maximally ally performing worse than the maximally attainable model.