Previous studies on the relationship between plasma levels of new-generation antipsychotics (NGAs) and clinical response did not account for inter- and intra-individual variability in drug levels. Therefore, the present study calculated the ratio of observed versus expected NGA plasma levels and investigated its relationship with changes in the Positive and Negative Syndrome Scale (PANSS). Data of patients starting monotherapy with a NGA were collected 2, 4, 8, and 12 weeks after initiation of treatment. Next to the assessment of changes in psychopathology (PANSS) the ratio of observed versus expected plasma level was calculated. A total number of 221 ratios were eligible for analysis. About half of them ranged from 0.5-2 and were considered "normal", whereas the others were considered either "too low" or "too high". Psychopathological symptoms improved over the course of treatment, but changes in PANSS from baseline did not correlate significantly with the ratios of observed versus expected plasma levels at any assessment. The lack of linear correlation can be explained by the fact that 92% of the observed NGA plasma levels were at ≥ 50% of the lower limit of the therapeutic reference range, i.e., within the asymptote of the logistic plasma level-effect relationship. Accordingly, our findings indicate that the great majority of patients were treated with NGA doses that led to optimal plasma levels, based on the clinical impression of the treating psychiatrist only. Thus, calculating the ratio of observed versus expected plasma level may not be necessary in a routine clinical setting.