Long‐term outcomes of schizophrenia in relation to course of social recovery and prescribed antipsychotics: A 10‐year follow‐up study at a single university hospital in Japan

Representative meta‐analyses on the long‐term prognosis of schizophrenia, as well as studies on the long‐term outcomes of various therapeutic interventions, have been conducted (Table S1). Molstrom et al. conducted a systematic review and meta‐regression analysis, revealing that 24.2% of the patients achieved recovery over 20 years. ¹ In Japan, Ogawa et al. used Eguma's Social Adjustment Scale (ESAS) to evaluate social outcomes, revealing varying levels of recovery (Table S2). ² , ³ Despite advances in pharmacotherapy and psychosocial approaches, the long‐term prognosis in Japan has remained unclear since this report. This study aimed to update this information by clarifying the long‐term prognosis of schizophrenia, focusing on social recovery and the effects of prescribed antipsychotics over a 10‐year follow‐up at a single Japanese facility.

We retrospectively reviewed the medical records of 114 schizophrenia patients treated at Fukushima Medical University Hospital and followed their progress over 10 years (Table S3). Social recovery was assessed using the ESAS, and data on antipsychotic medications were collected. The primary objectives were to compare findings with Ogawa et al.'s results and to examine associations between antipsychotic treatments—including first‐generation antipsychotics (FGAs), second‐generation antipsychotics (SGAs), and long‐acting injections (LAIs)—and social outcomes. The χ ² test was used to analyze differences in patient distributions and relationships between antipsychotic use and social recovery. This study was approved by the Ethics Committee of Fukushima Medical University School of Medicine (REC2023‐055) and was conducted in accordance with the Declaration of Helsinki. Because this was a retrospective medical record survey, the requirement for informed consent was waived; however, information about the study was made available to allow patients the opportunity to freely opt out.

Figure 1a summarizes the 10‐year social prognosis of patients over time. The results showed a significant shift in the distribution of patients across social recovery categories. The proportion of patients classified as fully self‐supportive or hospitalized decreased compared to that in earlier studies (self‐supportive: 47%–10%, hospitalized: 31%–2%), while the proportion of those in intermediate social recovery stages increased (semi‐self‐supportive, socially adjusted to family or community, maladjusted: 22%–88%) (Table S4). This change may reflect a redefinition of the recovery concept, as the medical model for schizophrenia has expanded to include rehabilitation and recovery, making full recovery more challenging. ⁴ Additionally, recent mental health policies reducing hospitalization durations and promoting community‐based care may have also contributed to this shift. ⁵ , ⁶

Fig. 1.

(a) Course of social adjustment in each patient. This figure shows the monthly social adjustment status of individual patients over a 10‐year period. Each row represents an individual patient, and the horizontal axis represents consecutive months. Color coding in the chart corresponds to the medical conditions. The patterns in this figure represent the following statuses of social adjustment: white for self‐supportive, gray diagonal for semi‐self‐supportive, black diagonal for socially adjusted to family or community, gray for maladjusted, and black for hospitalized. (b) This figure shows the changes in the types of prescribed antipsychotic medications for each patient over 10 years. The vertical axis represents each of the 114 patients, and the horizontal axis represents time. The sequence of patients is sorted based on trends in the duration of pharmacotherapy use and does not correspond to the order shown in Figure (a). The colors in this figure represent the following types of antipsychotics: red for FGA, blue for SGA, purple for FGA + SGA, yellow for LAI, orange for FGA + LAI, green for SGA + LAI, gray for FGA + SGA + LAI, and light green for no medication. (c) Rate of each antipsychotic medication at each social adjustment level. This figure shows the proportion of patients based on the duration of use of each antipsychotic medication for each ESAS assessment level. The proportion of patients receiving a single type of antipsychotic (FGA or SGA) was larger among those with better social functional outcomes. However, as social functional outcomes worsened, the rates of antipsychotic polypharmacy and use of LAIs increased. ESAS, Eguma's Social Adjustment Scale; FGA, first‐generation antipsychotic; LAI, long‐acting injection; SGA, second‐generation antipsychotic.

Regarding pharmacotherapy, patients with better social outcomes were treated with a single type of antipsychotic, regardless of whether they received FGAs or SGAs. In contrast, patients with poorer social recovery were significantly more likely to receive polypharmacy, including FGAs, SGAs, and LAIs (Fig. 1b,c, Table S5). There was no significant evidence that SGAs or LAIs led to better social outcomes than FGAs. Swartz et al. reported that while atypical antipsychotics have improved efficacy and tolerability, their effects on psychosocial recovery in schizophrenia are similar to those of conventional drugs. ⁷ Our results suggest that while advanced pharmacotherapy is effective in reducing acute symptoms and preventing relapse, it is not sufficient to promote long‐term social recovery. Alternatively, differences in pharmacological treatments may not affect patients with an inherently good prognosis.

Although clozapine is recognized as effective for treatment‐resistant schizophrenia, ⁸ its use was limited in this study. Only four patients received clozapine, which may have influenced the findings, as clozapine is known to improve outcomes in patients with refractory symptoms. Nonetheless, its impact on long‐term social recovery remains unclear and warrants further research. ⁹

This study had some limitations. First, the single‐institution setting and small sample size limit generalizability to the broader Japanese population. Second, this retrospective study relied on medical records, and it was not possible to track patients who transferred to other facilities or died during the study period, introducing potential biases. The retrospective design limits the ability to establish causality between prescription patterns and social recovery outcomes. Additionally, the use of the ESAS, a scale developed in the 1960s specific to Japan, may limit the comparability of these results with those of international studies that use more widely recognized measures of social functioning. Future research would benefit from a multi‐center, prospective approach with larger samples and internationally recognized scales to improve generalizability and align findings with global schizophrenia research.

In conclusion, the long‐term social outcomes of schizophrenia patients in Japan appear to have shifted over recent decades. Fewer patients are achieving full recovery or requiring long‐term hospitalization, while more fall into intermediate categories of social recovery. This trend is likely influenced more by changes in the concept of recovery and mental health policies aimed at reducing hospitalization and promoting community integration than by pharmacotherapy advances. The findings highlight the need for an integrated approach to treating schizophrenia, incorporating not only pharmacological interventions but also psychosocial support systems to enhance long‐term social recovery. Details regarding the introduction, methods, results, and discussion are provided in the Supporting information.

Disclosure statement

Y.M. received speaker honoraria for lectures from Sumitomo. I.M. received speaker honoraria from Sumitomo, Eisai, Janssen, Meiji Seika Pharma, Otsuka, Takeda, Towa, Tanabe‐Mitsubishi, and Yoshitomi. Y.K. received speaker honoraria for lectures from Sumitomo Dainippon, Janssen, Meiji Seika Pharma, MSD, Otsuka, Pfizer, and Yoshitomi. The other authors declare no conflicts of interest with respect to the research, authorship, or publication of this article.

Author contributions

Conception and design of the study: Y.M. and Y.K. Data acquisition: Y.M., R.S., A.S., A.N., M.H., Y.H., and Y.K. Data analysis: Y.M., R.S., H.H., M.H., and Y.K. Drafting the manuscript and figures: Y.M., R.S., M.H., and Y.K. All authors reviewed the manuscript and approved its submission.

Supporting information

Data S1. Supporting information.

Table S1. Representative studies related to the prognosis of schizophrenia.

Table S2. Eguma's social adjustment scale (ESAS).

Table S3. Demographic characteristics of study participants at the end of the study period.

Table S4. Eguma's Social Adjustment Scale in our study and in a previous study.

Table S5. Duration of use of each antipsychotic medication at each social adjustment level and results of residual analysis.