Duration of untreated psychosis and pathways to psychiatric services in first-episode schizophrenia


Dr Masafumi Mizuno, Department of Neuropsychiatry, School of Medicine, Keio University, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan. Email: mizuno@sc.itc.keio.ac.jp


Abstract  The aim of the present study was to examine the duration of untreated psychosis (DUP) in first-episode schizophrenia patients in Japan and to investigate the available pathways to psychiatric services. Eighty-three patients who visited Keio University Hospital (n = 54) or Oizumi Mental Hospital (n = 29) were evaluated retrospectively with regard to their DUP, living situation, social participation level, referral pathway, reason for seeking treatment, and their global assessment of functioning (GAF) score. The mean DUP was 13.7 months (median, 5.0 months) overall. No significant difference in DUP was found between subjects living alone and those living with others; however, employed patients had a significantly shorter DUP (8.1 months) than unemployed patients (18.7 months). Pathways to psychiatric services were totally different between the two institutions. Fifty-two subjects (62.7%) came to the services directly: 40 patients (74.1%) came to the university hospital and 12 patients (41.4%) came to the mental hospital. At the mental hospital, nine patients (31.0%) had been admitted because of a legal obligation, and six (20.7%) had been referred through public health centers. None of the patients had been referred to either of the services by general practitioners. The main reason for seeking treatment was psychiatric symptom aggravation (59.3%) at the university hospital and acting out (64.3%) at the mental hospital. Some universal psychosocial factors appear to influence the DUP but the characteristics of specific psychiatric services may also affect treatment delays.


Although numerous studies on the duration of untreated psychosis (DUP) in several countries have been reported, the DUP can also be influenced by the various characteristics of psychiatric services. With the exception of a few reports,1,2 recent research suggests that the early treatment of schizophrenia may be correlated with a more favorable outcome3–6 and an improvement in the natural course of the disorder.7 These findings have drawn attention to the need for early recognition and intervention in first-episode psychosis; several findings regarding the DUP have been reported, mainly by studies conducted in Europe and North America. According to McGlashan's review, the mean DUP for studies conducted in Western countries was between 1 and 2 years, while the median DUP was approximately 6 months.8 The DUP has also been focused on as a key variable in longitudinal studies, but reports from countries with different attitudes towards psychiatric services are still rare, especially from east-Asian countries; the DUP in Japan has not been previously reported.

The specificity and differences in the Japanese psychiatric service system have been discussed elsewhere9–11 but the main characteristics of the Japanese system are long hospital stays, a large number of psychiatric beds per unit population, of which more than 90% are in private hospitals, and the strong stigma associated with mental illness. These circumstances make access to psychiatric services difficult and complicate efforts to reduce the DUP. Nevertheless, shortening the DUP would reduce unnecessary suffering and may even reduce the direct negative consequences of psychotic episodes. To develop effective strategies for reducing the DUP, the factors contributing to treatment delay must be identified. These factors may be related to the patients, the social environment, or the professional services that are available.

The present study focuses on variations in the DUP among patients in different institutions and investigates the pathways through which patients with first-episode psychosis may seek treatment in Japan.



Two series of patients with first-episode schizophrenia were examined. Both groups were consecutive outpatients who visited psychiatric services between 1 April 1999 and 31 March 2001, and were treated by a psychiatrist for the first time. The first group consulted the Department of Neuropsychiatry at Keio University Hospital in central Tokyo, a general hospital with approximately 1000 beds, including 31 beds in the neuropsychiatry service. The second group consulted the Oizumi Mental Hospital, which specializes in psychiatry  and  has  approximately  400  beds;  the  hospital is located in the suburbs, approximately 20 km from central Tokyo. Oizumi Mental Hospital has been designated by the Tokyo government as a psychiatric hospital that also accepts and admits emergency and involuntary psychotic patients. Diagnoses were made according to the International Statistical Classification of Diseases and Related Health Problems (10th revision; ICD-10) criteria,12 by the treating psychiatrist at the time of the patient's first visit to the hospital. The diagnoses were then confirmed by two independent principal investigators using the information contained in the medical records. When the three psychiatrists agreed unanimously that the diagnosis was schizophrenia (F20), the case was included in the present study.

The age of the subjects in some previous DUP studies was limited between 15 and 54 years.13,14 We adhered to this criteria and, as a result, 83 subjects (university hospital patients, 54; mental hospital patients, 29) were enrolled in the present study. Thirty-five of the patients were male (university hospital patients, 24; mental hospital patients, 11). The mean age of these subjects at the time of their first consultation was 29.8 years (SD = 8.9 years); the difference in the mean age of the university hospital patients (30.4 years) and the mental hospital patients (28.6 years) at the time of their first consultation was not statistically significant.

Several methods have been used to measure DUP. We defined the DUP as the interval (in months) between the onset of psychotic symptoms and the first prescription of neuroleptics for psychosis. All of the subjects were neuroleptic-naïve at the time of their first consultation. We defined the onset of schizophrenia as follows: (i) the presence of at least one of the first rank symptoms of Schneider15 or (ii) at least one of the four ICD-10 criteria in G1. (1) (a) to (d). Negative symptoms and a reduction in social functioning were not considered in the assessment of DUP.

Two independent expert psychiatrists who were unaware of the patients’ characteristics determined the DUP retrospectively by reviewing the first-rank symptoms and the ICD-10 criteria for schizophrenia described in the available clinical charts. The interrater reliability was good (kappa = 0.886). The main demographic and clinical variables of the subjects were also noted and analyzed: gender, onset age, age at first treatment for psychosis, premorbid severity  of  impairment  in  social  functioning  as rated by the global assessment of functioning (GAF) score,16 presence of family members living together in the same residence, social participation level at the time of the first consultation, pathway by which psychiatric services were accessed, and comorbidity. ‘Social participation’ refers to attending school or having a non-sheltered job. Due to non-normal distribution of the data, the Man-Whitney's U-test was used for all analyses. Significance levels were set at P < 0.05.


Duration of untreated psychosis

The mean DUP (±SD) for all subjects attending the two psychiatric services was 13.7 ± 20.2 months; the mean DUP for the university hospital patients was 13.4 ± 21.5 months, while that for the mental hospital patients was 14.3 ± 17.9 months. The difference between the mean DUP of the patients attending the two psychiatric services was not significant. These results were similar to those obtained in previous studies.4,13 The median DUP for all the subjects attending the two psychiatric services was 5.0 months: 3.75 months for the university hospital patients and 5.0 months for the mental hospital patients. Further data are shown in Table 1.

Table 1.  Mean DUP based on the onset ages
nMean DUP
nMean DUP
  1. DUP, duration of untreated psychosis.

16–19 6 5.0 432.0
20–2411 3.3 734.4
25–291114.612 8.6
30–34 617.3 8 5.8
35–39 120.0 8 9.8
40–44 0 278.0
45–49 0 3 5.4
50–54 0 4 5.0

Effects of living with others and social participation level

No significant difference in the DUP of subjects living alone (10.7 ± 18.8 months) and of subjects living with others (14.5 ± 20.6 months) was observed (z = −0.277, P = 0.7818). Patients who were employed, however, had a significantly shorter DUP (8.1 ± 14.2 months) than patients who were unemployed (18.7 ± 23.6 months; Z = −2.806, P = 0.005).

Pathways to psychiatric services

Fifty-two subjects (62.7%) came to the services directly: 40 patients (74.1%) came to the university hospital and 12 patients (41.4%) came to the mental hospital (Fig. 1). The remaining patients were referred by outpatient clinic services (5, 9.3%), by way of other services in the same hospital (4, 7.4%), and to the university hospital by way of company clinics (3, 5.6%). At the mental hospital nine patients (31.0%) came to the hospital because of a legal commitment and six (20.7%) were referred by public health centers.

Figure 1.

Pathways to the psychiatric services. The numbers show the number of cases using each pathway; the mean durations of untreated psychosis (DUP) are in parentheses.

Reasons for seeking treatment

The most common reasons for seeking treatment in both groups were aggravation of psychiatric symptoms (48.2%) and acting out caused by psychopathology (48.2%), including an aggressive attitude towards the patient's family or an unsuccessful attempt to commit suicide. Treatment was sought because of social factors, such as complaints from neighborhoods, in 3.6% of the patients. Among the university hospital patients, the most common reason for seeking treatment was psychiatric symptom aggravation (59.3%), while in the mental hospital patients, the most common reason was acting out (62.1%). Further data are shown in Table 2.

Table 2.  Reasons for seeking treatment
n (%)
Keio University Hospital
n (%)
Oizumi Mental Hospital
n (%)
  1. DUP, duration of untreated psychosis.

Aggravation of psychiatric symptoms40 (48.2)32 (59.3)8 (27.6)
Mean DUP (months)13.913.316.2
Acting out caused by psychopathology40 (48.2)22 (40.7)18 (62.1)
Mean DUP (months)13.513.413.6
Social factors3 (3.6)0 (0.0)3 (10.3)
Mean DUP (months)5.75.7

As a pretreatment measurement of the severity of social  functioning  impairment,  the  mean  GAF  score of the subjects attending the university hospital (32.4 ± 7.5) was significantly higher (z = 3.354, P = 0.0001) than that of the subjects in the mental hospital (20.4 ± 10.6).


In previous studies the DUP was measured in units of weeks. In the present retrospective study, however, we calculated the DUP in terms of months based on the patient medical records. The mean DUP of the subjects (who were between the ages of 15 and 54 years) was 13.7 months; this value is similar to previously reported DUP determined by studies performed outside of Japan.4,13,17–23 The large standard deviation means that some individual values deviated widely from the mean. The person with the largest SD was a 48-year-old woman with an 84 month DUP who contacted the university hospital.

The mean age of the subjects at the time of their first consultation was 29.8 years; previous reports have described an earlier mean age at the time of first consultation, such as 24.7 years in the article by Loebel et al.4 These results suggest that patients with severe mental illness at the time of their first consultation with a psychiatric service may be older in Japan than in other countries. However, because the DUP in our study was similar to the DUP reported by previous studies, this finding does not suggest a high tolerance or acceptance of mental illness in Japanese society.

Exclusion criteria for substance-related disorders was not defined in the present study, but none of the subjects had a history of substance abuse or substance-related disorders. This finding is not surprising because drug abuse is still an implicit problem in Japan; public interest is very low, and no investigations of the actual state of affairs have been published in English. One study funded by the Ministry of Health and Welfare reported the results of a household survey on substance abuse.24 The survey asked subjects (n = 5000) ‘How many people do you know who have abused drugs within this past year?’ The results supported the common perception, with the percentage of respondents who answered ‘more than one person’ being 0.05% for cocaine use, 0.05% for heroin use, and 0.2% for marijuana use. Birchwood et al. reported that 21% of their subjects were suspected of having a history of drug use and that this figure was likely an underestimate.25 Recently, Bühler et al. showed that the age of schizophrenia onset is significantly lower for drug abusers than for alcohol abusers and markedly lower than for patients without any type of abuse.26 Because the number of subjects analyzed in the present study is small, we cannot discuss the lateness of onset of severe mental illness in Japanese people with regard to drug use or other factors. Nevertheless, it is noteworthy that the comorbidity rate of the subjects in the present study was very low. In the present study we incidentally had a larger number of female subjects. Gender is related to a longer DUP and may be partially responsible for the higher onset age.27 In fact, nine female subjects more than 40 years old were included in the present study (Table 1).

Social participation, rather than living at home with family, appears to increase the accessibility of social services, including psychiatric professionals. The stigma of mental illness in Japanese society may prevent early access to psychiatric services10,11 and the families of patients with mental illness may be unwilling to encourage a psychiatric consultation.

None of the subjects in the present study were referred to the psychiatric services by general practitioners (GPs). In Japan, GPs unfortunately do not perform screening for the early detection of mental illness. The pathways of access to psychiatric services were very different for the university and mental hospitals. A large number of involuntary admissions to the mental hospital were made with police assistance; these admissions obviously depend on the mental health system and mental health and welfare laws and are influenced by the severity of the psychopathology. The provision and modification of psychiatric services for easy access and a system for the early recognition and detection of mental illness is more important in Japan than a further increase in the number of psychiatry clinics.

Although the pathways and the reasons for seeking -treatment were totally different between the two groups and the mean GAF score was significantly lower in the mental hospital patients, the patients who visited the mental hospital had more difficulties in their daily lives than those who visited the university clinic, even though the mean DUP of the two groups were not significantly different. According to the mean GAF scores, the patients with the severest cases visited the mental hospital. In fact, 26 out of 29 patients who visited the mental hospital were admitted after their first consultation, including nine involuntary hospitalizations ordered by the prefectural governor.28 Thirteen patients were hospitalized for medical care and protection under Article 33. Our data suggest that it is the level of danger or difficulty in coping with symptoms and other social factors, rather than the severity of the psychopathological symptoms, that determines whether or not an individual may be required to consult a mental hospital. Delaying treatment until the patients themselves become aware of the need for treatment may increase the DUP. The duration of untreated psychosis may partially reflect the level of tolerance of the people living or working close to the patient. This level of tolerance may, in turn, be affected by the social attitude towards psychosis. Recently, de Haan et al. pointed out that some awareness of the presence of a mental disorder at the time of onset was related to a shorter DUP.29 Early intervention may help prevent the often significant biological, social and psychological deterioration that can occur in the early years following the onset of a psychotic disorder.30

The present study has several limitations. First, the onset of schizophrenia was defined by referring to medical records. However, even if structured interviews such as the Interview for the Retrospective Assessment of the Onset of Schizophrenia31 are used, defining the onset of psychosis is difficult and can be influenced by many factors. Thus, the DUP scores should be considered as an index of public mental health, not as a precise measure of duration. Furthermore, schizophrenia patients in the present study were partially diagnosed retrospectively and were not interviewed using a structured format. Only the schizophrenia cases diagnosed unanimously by three psychiatrists who had not performed a structured interview were included in the present study. Therefore, patients suspected of having other conditions belonging to the schizophrenia spectrum of disorders, such as schizoaffective disorder, acute and transient psychotic disorder psychosis were excluded, and the patients with schizophrenia included in the study were very narrowly selected in spite of the difficulties referring to the subtypes due to the retrospective study design. In addition, for investigating the pathways to the psychiatric services, the visits to the so-called traditional therapists such as shamans were not investigated thoroughly.

The DUP in the Japanese patients in the present study was approximately 1 year, similar to that described in previous reports from Western countries. Various efforts to shorten the DUP have been made and are ongoing in several countries. Shortening the DUP in Japan should be an urgent challenge for Japanese public health officials.


Preliminary data described in the present study were presented at the First Japan International Conference on Early Intervention and Prevention in Psychiatric Disorders, held in Okinawa by Professor C. Ogura, Japan, June 2001.