The need for prospective longitudinal cohort studies on the psychopathology and biology early in bipolar disorder: A commentary to the review by Pfennig et al

The narrative review by Pfennig et al. (2017) in the Australian & New Zealand Journal of Psychiatry (ANZJP) summarizes evidence from 19 prospective studies in the early course of bipolar disorder on psychopathological and other clinical and biological changes over time. The review focuses on a highly central and understudied area within psychiatry, prospective long-term studies of episodic disorders such as bipolar disorder. Data suggest that psychopathology and biology change over the long-term course of bipolar disorder and causality is likely bidirectional. First, there is a body of evidence for clinical progression on average of mood disorders as increasing number of affective episodes seem associated with (1) increasing risk of recurrence, (2) increasing duration of episodes, (3) increasing symptomatic severity of episodes, (4) decreasing threshold for developing episodes, and (5) increasing risk of developing dementia (Kessing and Andersen, 2017). In this way, psychopathology of each mood episode changes during the course of illness. Another example is that the prevalence of mixed features in women with bipolar disorder is much more prevalent later than in the first mood episodes in bipolar disorder (Kessing, 2008). Second, the reason for this bidirectional association between psychopathology and biology is that bipolar disorder is partly genetic and biologically driven and partly driven by environmental stress of sociological and psychological origin. It is likely that the genetically related biology of bipolar disorder is changed by environmental factors (i.e. epigenetics) making patients more stress sensitive during the course of illness that results in behavioral sensitization, neuroprogression and changes in the psychopathology of affective disorders (Post, 2016). Thus, the experience of mood episodes may change the biology of bipolar disorder and changed biology may result in changed psychopathology and clinical progression of the disorder. Taking such psychopathological and biological consequences into account it is striking how few prospective longitudinal studies have been undertaken. Although the review by Pfennig et al. did not conduct a systematic literature search, only 19 prospective studies on the early developmental course of bipolar disorder were identified covering antecedent and early stages of bipolar disorder. We have discussed the limitations of prior prospective longitudinal studies on the biology of bipolar disorder in a research protocol concluding that the results from these studies are hampered by a number of limitations (Kessing et al., in press). Only few studies have recruited patients with bipolar illness from onset of the illness, the number of patients included is few, and the prospective follow-up periods are short, mostly less than a few years. In addition, as discussed in the research protocol, the inherent design of prospective longitudinal studies implies some challenges and limitations (Kessing et al., in press). Often extensive initial assessment procedures may potentially result in selection of participants who are intrinsically positive toward clinical research and readily willing to cooperate. Attrition may increase during long-term follow-up and patients who stay in the study may adhere more to treatment in general. Furthermore, due to the observational design, it is not possible to avoid the potential confounding effects of psychotropic medication and other interventions and finally, due to the large number of biological and statistical tests included in such studies, chance findings may occur in relation to the individual biological test. Nevertheless, with a prospective, repeated measures design it is possible to identify associations between biology and depressive and manic symptoms and states, and fortunately a number of prospective longitudinal cohort studies on the psychopathology and biology early in bipolar disorder have started or are about to be initiated internationally.

See Review by Pfennig et al. (2017) 51: 509–523.