Géraud Dautzenberg

Summary and general discussion

the MoCA total score, more detailed studies are needed. Are certain cognitive domains more prone to impairment than others due to their underlying aetiology? In the literature, there is evidence that some MoCA items do not contribute (as much) to differentiating between different aetiologies or even states of cognitive impairment as other items do. Most surprisingly, orientation is among them. Therefore, there is a shorter MoCA under development, for example, Basic MoCA (Julayanont et al. , 2015), finding the optimal short Mini-MoCA [https://www.mocatest.org/reference/] as well as digital versions (Berg et al. , 2018) under investigation to save time but with the same high reliability. In the literature, subtle differences are mentioned in cognitive impairment, when present, of depression, schizophrenia, and bipolar disorder. These are illuminated in the paragraph ‘definitions’ of Chapter 1. In short, unipolar depression seems to affect cognition less than bipolar depression. The bipolar profile resembles that of a schizophrenic patient, although less severe (Van Rheenen et al. , 2017). However, is the MoCA sensitive enough to detect these differences? Alternatively, can it detect differences in neurodegenerative aetiologies, as there are differences found (at group level) in domains (Freitas, Simões and Santana, 2014; Coleman et al. , 2016)? Figure 2b in Chapter 4 shows that the total score (particularly on an individual basis) cannot be differentiated. However, can different domains or items be helpful? Future studies should attempt to answer this question. In addition to differences between diseases, there is increasing interest in and evidence in the literature regarding the heterogeneity of cognitive impairment in certain psychiatric diseases. It is even suggested that differentiation by nosological aetiology, such as bipolar disorder, can or should be distinguished by different entities based on their cognitive profile (Van Rheenen et al. , 2020). Bipolar patients with more cognitive impairment, resembling schizophrenic cognitive profiles, will have more psychotic features than bipolar patients with less cognitive impairment. This could explain why no specific cognitive profile is found for these psychiatric diseases, as they are categorised by diagnostic categories and less by functional or dimensional features (Van Rheenen et al. , 2020). This problem also accounts for the use of the MoCA. However, as our results show (figure 2b), the MoCA could differentiate between the three suggested groups of no, mild to moderate, and severe cognitive impairment in bipolar patients. The MoCA could therefore be of added value for the ‘large, longitudinally focused studies of cognition’ that are required to better ‘define cognitive trajectories’ in psychiatry and bipolar disorder specifically (Van Rheenen et al. , 2020) as it is easy and inexpensive to implement, but mostly as it is more sensitive for mild cognitive impairment than the still widely used MMSE in (general population-based) large longitudinal studies.

7

175