Psychology researchers have developed an algorithm that can predict a patient’s risk of transitioning from a diagnosis of major depression to a diagnosis of bipolar disorder. These findings, published in the journal Translational Psychiatry, may help practitioners identify patients with bipolar disorder who have been misdiagnosed with major depression.
Major depressive disorder (MDD) and bipolar disorder (BD) are common mood disorders with overlapping symptomology. While both conditions are characterized by depressive episodes, a bipolar diagnosis is only met when patients present with at least one manic or hypomanic episode — an episode of extreme energy or elevated feelings.
Researchers have noted that many patients with MDD go on to receive a diagnosis of BD, suggesting they were initially misdiagnosed. There is also evidence that many people with BD wait years before receiving a correct diagnosis, delaying their treatment and increasing their risk of negative outcomes.
Notably, although MDD and BD share clinical features, their recommended treatment differs. While MDD is often treated with antidepressants, there is evidence that these medications can provoke mania among patients with BD. With the aim of helping clinicians identify BD as early as possible, a team of researchers set out to develop a risk assessment algorithm that can help predict which patients with an MDD diagnosis will go on to receive a diagnosis of BD.
Study authors Anastasiya Nestsiarovich and her team designed a model to predict change in diagnosis from MDD to BD over a one-year period. The model was tested across a network of five US patient databases, which included 2,687,578 patients with MDD. Patients who went on to have their diagnosis changed to BD within one year made up the case group, and patients who did not have their diagnosis changed made up the control group.
The final regression model identified seven variables that positively predicted patient transition from MDD to BD within one year — younger age, more severe depression at baseline, the presence of psychosis, the presence of anxiety, thoughts or acts of self-harm, substance misuse, and prior diagnosis with a mental disorder. Three variables negatively predicted transition from MDD to BD — older age, less severe depression at baseline, and pregnancy.
The positive predictor of younger age is in line with evidence that BD tends to present at an earlier age than depression. The positive predictor of depression severity corresponds with research suggesting that patients with BD experience more severe depressive episodes compared to patients with MDD.
The model was then validated across additional datasets in multiple countries. The researchers suggest this algorithm might be a helpful tool for clinicians, allowing them to assign their patients risk scores for diagnosis conversion to BD. “Considering the identified predictors (early onset of depression, presence of severe depression, and psychotic features) might help to promote vigilance among clinicians regarding the possibility of underlying BD diagnosis in depressed patients and encourage them to ask clarifying questions about manic/hypomanic episodes,” Nestsiarovich and her colleagues report.
A notable limitation to the study was that it relied on electronic health records — these records can be limited by missing data and by variations in the criteria used for diagnosis.
The study, “Predictors of diagnostic transition from major depressive disorder to bipolar disorder: a retrospective observational network study”, was authored by Anastasiya Nestsiarovich, Jenna M. Reps, Michael E. Matheny, Scott L. DuVall, Kristine E. Lynch, Maura Beaton, Xinzhuo Jiang, Matthew Spotnitz, Stephen R. Pfohl, Nigam H. Shah, Carmen Olga Torre, Christian G. Reich, Dong Yun Lee, Sang Joon Son, Seng Chan You, Rae Woong Park, Patrick B. Ryan, and Christophe G. Lambert.