Comorbid medical illness in bipolar disorder

Forty et al ¹ investigated whether the presence of physical comorbidity in individuals with bipolar disorder is associated with a more severe bipolar illness course that may contribute to the worsening of the mortality gap between individuals with bipolar disorder and the general community. Forty et al claimed this to be the first study on this issue in a UK clinical sample: that is, the first study that assessed rates of physical comorbidity in individuals with bipolar disorder and made direct comparisons with unipolar and control samples. The study is done statistically carefully.

While the results concerning the self-remembered physical comorbidities over the lifespan in the unipolar sample and the bipolar sample are clear, we have reservations concerning the composition of the control sample. Specifically, Forty et al used mixed samples of treatment-seeking individuals with unipolar and bipolar disorders that were originally recruited in genetic studies from psychiatric clinics, hospitals, general medical practices and self-help groups, as well as volunteers responding to media advertisements. ² The observed odds may be falsely calculated as the younger control sample is chosen, at least to a substantial proportion, from a specialised community sample that is representative neither of the general population nor of treatment-seeking individuals of the general community.

The reader might be interested to know that we recently published a paper on this issue. ³ Between 1 January 2000 and 30 June 2012, 621 individuals with bipolar disorder were admitted to three general Manchester hospitals. All mental and physical comorbidities with a prevalence >1% were compared with those of 6210 randomly selected and group-matched hospital controls of the same age and gender, regardless of priority of diagnoses. Comorbidities that increased the risk for hospital-based mortality (but not mortality outside of the hospitals) were identified using multivariate logistic regression analyses. The study was intended to determine which specific mental and physical comorbidities contribute to later in-hospital deaths in individuals with bipolar disorder, and whether the risk factors for hospital-based mortality differ for individuals with bipolar disorder in comparisonwith hospital controls.

In our view, our study has the advantage that a more representative and more relevant control sample was used and that the most relevant outcome from comorbidity – mortality – was addressed. In partial agreement with Forty et al, we found that excess comorbidity in individuals with bipolar disorder was caused by asthma and type 2 diabetes mellitus (T2DM). In addition, T2DM in individuals with bipolar disorder represented a major risk factor for general hospital-based mortality with excess mortality due to acute T2DM, as well as other diabetes-related complications. Our study gives support for an aggressive multidisciplinary approach to identify and treat T2DM to prevent diabetic, respiratory and vascular complications in all individuals with bipolar disorder.