The prevalence of symptomatic remission in schizophrenia by Andreasen criteria vary widely across reported studies (17% to 88%) [19, 20]. However, for most samples in cross-sectional studies in naturalistic settings approximately one-third of individuals had symptomatic remission [6, 21, 22]. These results were confirmed in our cross-sectional study in an outpatient schizophrenia population in one territorial sector of Moscow city, where 31.5% of patients met the symptomatic remission criterion. It has to be noted that 62% of patients were treated with a first-generation antipsychotic and only 12% with a second-generation antipsychotic. In contrast, a Spanish cross-sectional study using 80.1% second-generation antipsychotics, the rate of symptomatic remission was 44.8% (452 of 1,010 patients) . Maintenance of remission status over a 6-month period without changes to current antipsychotic treatment constituted 82.8% of patients, which is very close to that reported in the previous study with the same follow-up duration . But for total outpatient population full remission criteria were only fulfilled in 26.1% of cases. In a recent German naturalistic study the remission rate over the 1-year follow-up period was even less: 10.3% of patients .
The most significant factors predicting achievement of remission reported by Lambert et al. , using multivariate regression analysis of 12 studies, are (i) shorter duration of untreated psychosis, (ii) better premorbid adjustment, (iii) lower psychopathology or illness severity scores at baseline, (iv) better functioning level at baseline, (v) early improvement in symptoms or functioning, and (vi) medication adherence during treatment. These predicting factors were supported by our observational study. The most important baseline patient characteristics associated with symptomatic remission were symptom severity (PANSS score at baseline), executive functions (WCST, TMT Part B), social and personal functioning (PSP score), and quality of life (physical functioning and social functioning subscales of the SF-36). The relationship between symptomatic remission and cognitive improvement was also confirmed by Buckley et al. . Additionally, we have observed that diagnoses of episodic (with progressive deficit) and remittent courses of schizophrenia were associated with an increased chance of remission as compared to continuous and episodic courses with stable deficit. This observation was also noted previously by Wobrock et al. . They reported that outpatients with different types of schizophrenia by ICD-10 definition had differing chances of achieving symptomatic remission: patients with paranoid schizophrenia had a greater possibility of meeting the international remission criteria than those with other diagnoses, particularly residual schizophrenia. Furthermore, each of the ICD-10 types and courses of schizophrenia in our study differed in the threshold for the eight core PANSS items in our sample of stable patients. When Eberhard et al.  analyzed a 5-year risperidone trial, they found that the core eight PANSS items are quite common in patients with schizophrenia and non-schizophrenic disorders with psychotic symptoms. However, discriminate analysis has highlighted G6 (depression) and G15 (preoccupation) as potentially important symptoms for differentiating other psychotic disorders from schizophrenia. Therefore, our opinion is that the symptom threshold is very strict for chronic patients with schizophrenia and it is necessary to develop specific remission criteria for different clinical forms and courses of schizophrenia. This point of view agrees with previous research performed in Russia using a categorical approach to the psychopathology of schizophrenia and defining remission within an interval from full recovery (symptomatic and functional) to marked deficit with long-term symptomatic stability of patients [28, 29]. This definition is partly supported by findings in our study: 105 (75.5%) of 139 patients who did not meet the symptomatic criterion in the cross-sectional study were stable over 6 months. Moreover, ANCOVA in the RLAI group in our observational study showed that ICD-10 diagnosis was statistically significant for achievement of symptomatic remission. Schizophrenia is heterogeneous in psychopathology and its outcomes, and it is impossible to ignore the different clinical types of the disorder. It is not surprising that patients with schizoaffective disorder and remittent or episodic courses of paranoid schizophrenia had a greater chance to achieve symptomatic remission. In the observational 1-year study, stable patients switched to RLAI had a significant reduction of psychopathological symptoms and non-compliance rate, as well as an improvement in social and personal functioning; however, only 21.4% of patients met symptomatic remission criterion and only 19% achieved full remission. In the routine treatment group, the remission rate at end point was much lower, at 5.7%. This finding agrees with some previous studies. Lasser et al.  found that 82 (20.8%) of the 394 stable patients who did not meet symptomatic remission criterion at baseline achieved it over 1 year of treatment with RLAI. However, Rossi et al.  reported a higher level of sustained remission; 32% of a sample of 347 stable patients that were switched to RLAI met remission at week 52. Generally, patients considered to be stable may not be at their optimal symptomatic and functional levels, and modern therapeutic approaches can improve their outcomes. However, the proportion of stable patients who did not meet remission in all studies is remarkable, and for most chronic patients the symptomatic threshold is unachievable. This statement is also supported by results from the medication phases of the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study. Only 11.7% of patients attained symptomatic remission and then maintained it for at least 6 months, and 55.5% experienced no symptom remission at any stage .
There are limitations to this study. First of all, this is one-site study, but despite this fact as patients from the outpatient psychiatric service in two healthcare districts of a big city were chosen randomly, we believe it can be representative; future research in Russia is necessary however, featuring multicenter studies with sites in different regions, including rural ones. Secondly, the sample size was not specially estimated for the exploratory objectives, and some analyses only included a small number of subjects. Thirdly, the therapeutic study was observational in a naturalistic setting and did not assume randomization or a blinded design; nevertheless the clinical and demographic characteristics of the two district populations of stable patients were comparable. However, the psychotherapeutic influence of switching to a completely new therapeutic approach and additional attention from medical staff with visits every 2 weeks for injections could be important. Therefore, it is necessary to interpret our results cautiously, and obviously further research is needed to confirm our findings and the importance of revising the remission criteria according to the clinical courses and types of schizophrenia and patient functioning level.