A study by Seena Fazel and colleagues in the September issue of the Archives of General Psychiatry is generating a bit of a buzz among mental health professionals - and it provides a good lesson on how the results of a good study can be viewed depending on the interpretative lens one wants to look through.
The study examined arrests for violence among people with bipolar disorder in Sweden. The study is notable because Sweden maintains a national mandatory health registrar which makes large epidemiological population studies possible. This is in contrast to the approach taken in most other studies which rely on sampling which is prone to sampling bias and other errors.
And what did Fazel and colleagues discover?
For people with bipolar disorder alone, the risk of violence was comparable to the general population. But for those with bipolar disorder and substance abuse, the link was strong: the adjusted odds ratio was 6.4%. That's a substantial increase in the risk of violence compared to the general population. So, what's the take home message?
One might conclude that the association between bipolar disorder and violence is negligible and this study helps prove that fact. But with the prevalence of substance abuse among those with bipolar disorder around 50% such conclusions seem overly simplistic. Indeed as the authors themselves conclude: "Available data suggest a common familial etiology for bipolar disorder, violent criminality, and substance abuse." Untangling that etiology has proven difficult but claims that "people with a severe mental illness are no more likely to be violent than anyone else - unless they abuse drugs or alcohol, a study has suggested" seems to understate matters a bit.
The study examined arrests for violence among people with bipolar disorder in Sweden. The study is notable because Sweden maintains a national mandatory health registrar which makes large epidemiological population studies possible. This is in contrast to the approach taken in most other studies which rely on sampling which is prone to sampling bias and other errors.
And what did Fazel and colleagues discover?
For people with bipolar disorder alone, the risk of violence was comparable to the general population. But for those with bipolar disorder and substance abuse, the link was strong: the adjusted odds ratio was 6.4%. That's a substantial increase in the risk of violence compared to the general population. So, what's the take home message?
One might conclude that the association between bipolar disorder and violence is negligible and this study helps prove that fact. But with the prevalence of substance abuse among those with bipolar disorder around 50% such conclusions seem overly simplistic. Indeed as the authors themselves conclude: "Available data suggest a common familial etiology for bipolar disorder, violent criminality, and substance abuse." Untangling that etiology has proven difficult but claims that "people with a severe mental illness are no more likely to be violent than anyone else - unless they abuse drugs or alcohol, a study has suggested" seems to understate matters a bit.
Leave a comment