Géraud Dautzenberg

Chapter 7

pitfall. How useful is the proposed theoretical cut-off in practice, especially for cognitive impairment? As much as the MoCA will produce a solid score and thus appear to be an arbiter between cognitive impairment and no impairment, clinical reality, again, involves many other factors that can affect this MoCA score. This is exemplified in our study. The mean MoCA scores differ significantly from each other per cognitive group. However, the range (or SD) is too wide and shows us that the individual score must be weighted with this knowledge. The ‘best’ mean cut-off as suggested by a study is a mean and does not fit all individuals due to personal demographic and clinical factors. These individual factors range from intoxication, (lack of) motivation, or anxiety to disabilities, such as poor eyesight. The MoCA score itself does not correct for these factors. All of these clinical data must be considered and many will be considered in the clinical reality, but on the other hand, many factors that are (or appear to be) influential may remain unknown and may not be included or even be excluded in a study. This can range from the level or years of attended education (MoCA corrects for this), through alcohol use (corrected by our study) up to literacy and ethnicity of the patient (not corrected in our study, but some US-based studies do). For research, you have to create study groups, and you need to translate these results to individual patients. However, these groups can never fully match the unique patient since the study outcomes are averages of several factors, whereas the individual patient consists of many factors. This is clarified in our study by using education as a factor. Although the MoCA tries to correct this with one extra point for education of 12 years or less to reduce the number of false negatives, it will never be able to offer a custom-made correction. Stratification is sometimes suggested as a solution (Oren et al. , 2014; Wong et al. , 2015; Borland et al. , 2017), but in practice, it seems impractical, given the many parameters that may be affected and need to be stratified. Through years of education ( < 12; + 1 point), an attempt is made to capture the lower baseline values of an individual patient. However, certain groups do not seem to be corrected well with this, and there are suggestions to use literacy (Sisco et al. , 2015). Low baseline values can play a role, as can high baseline values. Thus, in our study, we see that a very high level of education can also lead to false negatives. Other factors, especially in old age psychiatry, that can greatly influence the MoCA score but seem to differ from individual to individual, are psychiatric diagnoses. This is clearly illustrated in Figures 2a and b of Chapter 4. The mean total MoCA scores are significantly different among the three cognitive groups. If we distinguish more on an aetiological level, a different situation appears, and the use of the MoCA in clinical reality becomes less distinct (sharp). First, Figure 2b confirms that the MoCA score corresponded well with

168