Early prodromal symptoms and diagnoses before first psychotic episode in 219 inpatients with schizophrenia

Authors


Toshiki Shioiri, MD, PhD, Department of Psychiatry, Niigata University Graduate School of Medical and Dental Sciences, 1-757 Asahimachi, Niigata 951-8510, Japan. Email: tshioiri@med.niigata-u.ac.jp

Abstract

Abstract  The authors examined the diagnosis before the onset of schizophrenia and retrospectively evaluated the presence/absence of early prodromal symptoms (EPS) and their types (such as depressive symptoms, anxiety symptoms, and obsessive-compulsive [OC] symptoms) and the period from the onset of these symptoms to that of schizophrenia in 219 inpatients with schizophrenia diagnosed according to the DSM-IV(-TR). A diagnosis was made before the onset of schizophrenia in 53 patients (24.2%). The diagnoses were mood disorder in 39 patients, anxiety disorder in seven, obsessive-compulsive disorder (OCD) in three, adjustment disorder in two, and eating disorder in two. EPS were present in 65 (29.7%) of all patients, slightly more frequent in female patients (male : female = 1:1.41). In the group with EPS, depressive symptoms (61.5%) were most frequently observed, followed by anxiety symptoms (23.1%) and OC symptoms (9.2%). The age at onset for each type of symptom was significantly lower for OC symptoms (14.5 ± 2.4 years) than for the other symptoms (approx. 20 years). The mean period from the onset of each symptom to that of schizophrenia was the shortest for depressive symptoms (2.7 ± 3.1 years) and the longest (>4 years) for OC symptoms. These results as well as previous studies in Western countries showed that more non-specific and general symptoms are frequently present for some years before the onset of schizophrenia. With consideration of this point, efforts toward early detection of schizophrenia are important.

INTRODUCTION

Schizophrenia is a psychiatric disorder that develops after puberty and the prevalence is approximately 1%.1 Schizophrenia is treated by comprehensive approaches including drug therapy, but the untreated period from onset to the initiation of treatment has been reported to be closely associated with the time until remission and the degree of remission.2 Therefore, early detection and treatment are important,3 and the early detection of schizophrenia during the prodromal phase before the development of psychotic episodes is a major topic.4 One study found that approximately 75% of patients with schizophrenia were found to have passed through three stages of prodromal symptoms in a fixed order.5 That is, patients reported subthreshold psychotic symptoms in the year preceding onset, prominent negative symptoms in the 2 years preceding onset, and non-specific affective and anxiety symptoms earlier.5,6 Recently, Maier et al. reviewed the transition to schizophrenia and related disorders and discriminated the following four phases in a stage model of the progression to schizophrenia:6 (i) a premorbid phase, without psychosocial impairment but with risk factors and vulnerability traits present; (ii) an early prodromal phase, consisting of negative and unspecific, mainly affective symptoms as well as psychosocial impairment; (iii) a late prodromal phase, consisting of attenuated psychotic symptoms or brief, limited intermittentpsychosis; and (iv) a psychosis phase, which might progress to schizophrenia.

Symptoms in late prodromal phase and early stage symptoms of schizophrenia have been extensively studied.4,7–11 In Japan, Nakayasu proposed ‘early stage schizophrenia’, which belongs to psychotic disorders as a disorder unit but which should be regarded as a clinical unit distinguished from schizophrenia, and described its tetralogy (autochthonous idea, obscure feeling of observation, promotion of notice, and tension/perplexity mood).12 In contrast, there are few reports for early prodromal symptoms (EPS),4,13–15 which are more non-specific and general16 such as depression and anxiety6 before illness onset.9 Ruhrmann et al. referred to an importance of the detection in the earlier prodromal stage because functional decline often set in before the late prodromal phase.17 In the present study we performed a survey of EPS and diagnosis before the onset of schizophrenia in 219 inpatients during the past 5-year period.

METHODS

Subjects

The subjects were 219 patients who were admitted to the Psychiatry Unit of the Niigata University Medical and Dental Hospital between April 1999 and March 2004 and who were diagnosed as having schizophrenia at the time of discharge. For diagnosis, a structured interview for approximately 60 min was performed using the DSM-IV(-TR).1Table 1 lists the profiles of the subjects. There were 98 men and 121 women (male : female = 1:1.23), and their mean age was 33.9 ± 14.6 years (men, 35.2 ± 15.2 years; women, 32.8 ± 14.0 years). The subtypes of schizophrenia and age at onset are also given in Table 1.

Table 1.  Schizophrenia subject profile (n = 219)
Schizophrenia
subtype
Men (n = 98)Women (n = 121)
n (%)Age at onset (years)
(mean ± SD)
n (%)Age at onset (years) (mean ± SD)
Disorganized type25 (25.5)21.7 ± 9.257 (47.1)20.9 ± 8.6
Catatonic type3 (3.1)21.7 ± 1.54 (3.3)21.3 ± 8.2
Paranoid type13 (13.3)29.2 ± 9.613 (10.7)27.1 ± 7.3
Undifferentiated type52 (53.1)24.0 ± 7.844 (36.4)22.2 ± 10.6
Residual type5 (5.1)25.8 ± 4.93 (2.5)23.0 ± 5.3

Methods

A survey was performed based on treatment records including reports by family for the 219 subjects. Corresponding to the Beiser et a. study,13 the record included an anamnestic interview focusing on premorbid history and details about the progression of the illness, which was conducted by two senior psychiatrists (T.S. and T.S.). The survey items were: (i) presence/absence of EPS and their types; (ii) diagnosis before the onset of schizophrenia (diagnoses other than schizophrenia including those not based on the DSM-IV); (iii) age at onset of EPS; (iv) period from the onset of EPS and that of schizophrenia; and (v) Global Assessment of Functioning (GAF) score of DSM-IV(-TR)1 at the time of admission and discharge as a parameter of global function.

According to the definition of EPS in previous studies, that is, ‘more non-specific and general such as depression and anxiety before the illness onset’,6,9,16 as aforedescribed, EPS were classified into four categories here for convenience: (i) depressive symptoms (such as depressive mood, decreased appetite, and insomnia); (ii) anxiety symptoms (such as anxiety/irritation, fear, and autonomic symptoms); (iii) obsessive-compulsive (OC) symptoms (these symptoms are originally classified as anxiety symptoms but were evaluated as an independent item in the present study because schizophrenia patients sometimes show OCD-like symptoms not only during its chronic course but also during the prodromal phase preceding the emergence of obvious psychotic symptoms18–21); and (iv) others (symptoms such as those suggesting somatoform disorders and eating disorders). In patients with only decreased volition, EPS were considered to be absent because its differentiation from the negative symptoms of schizophrenia is difficult.

Statistical analysis was performed using the t-test and χ2 test. A probability level of P < 0.05 was regarded as statistically significant. The data were analyzed using statistical SPSS software (release 10.07J, SPSS, Chicago, IL, USA). The present study was approved by the ethical committee of Niigata University Graduate School of Medical and Dental Sciences.

RESULTS

Presence/absence and types of EPS

Table 2 lists the presence/absence and types of EPS. EPS were present in 65 patients (29.7%), consisting of 27 men and 38 women (male : female = 1:1.41). In the group with EPS, depressive symptoms were most frequently observed (40 patients, 61.5%), followed by anxiety symptoms (15, 23.1%), OC symptoms (6, 9.2%), and others (4, 6.2%). The incidence of anxiety symptoms was slightly higher in the female patients.

Table 2.  Comparison of presence and absence of EPS
 Group with EPSGroup without
EPS
DepressionAnxietyObsessionWhole
  1. EPS, early prodromal symptoms; GAF, Global Assessment of Functioning.

Males16532771
Females241033883
Mean age at onset of EPS (years)22.1 ± 8.319.7 ± 6.114.5 ± 2.420.8 ± 7.6
Mean age at onset of schizophrenia (years)24.6 ± 9.723.1 ± 8.718.8 ± 6.423.8 ± 8.922.7 ± 8.9
Mean period until onset of schizophrenia (years) 2.7 ± 3.1 3.6 ± 5.0 4.4 ± 6.6 3.1 ± 4.0
GAF on admission28.5 ± 8.427.0 ± 8.7 34.8 ± 10.828.9 ± 8.628.2 ± 10.4
GAF at discharge 39.4 ± 11.237.7 ± 9.742.3 ± 4.6 40.2 ± 10.437.8 ± 11.9
Improvement degree in GAF 10.9 ± 12.9 10.8 ± 11.8 8.6 ± 9.9 11.4 ± 12.5 9.4 ± 12.9

Diagnosis before the onset of schizophrenia

A diagnosis was made before the onset of schizophrenia in 53 patients (24.2%). The diagnoses were mood disorder in 39 patients, anxiety disorder in seven, OCD in three, adjustment disorder in two, and eating disorder in two. Diagnoses made in other hospitals included those not based on the DSM-IV such as ‘psychogenic reaction’, ‘school refusal’, ‘abnormal experience’, and ‘puberty crisis’. These diagnoses were excluded from those made before the onset of schizophrenia.

Onset of EPS

The age of the patients at the onset of each EPS is shown in Table 2. The age at onset for OC symptoms varied because of the low number of patients (= 6), but the mean age was 14.5 ± 2.4 years, which was significantly lower than the age at onset for the other symptoms (approx. 20 years; t = 2.01, P < 0.05).

Period from the onset of EPS to that of schizophrenia

The age at the onset of schizophrenia did not differ between the presence and absence of EPS. However, the mean age at onset for schizophrenia was slightly lower for OC symptoms (18.8 ± 6.4 years) than for the other EPS or the group without EPS. The mean period until the onset of schizophrenia was the shortest for depressive symptoms (2.7 ± 3.1 years) and the longest for OC symptoms (4.4 ± 6.6 years).

Presence/absence of EPS and the degree of improvement in the GAF score

Table 3 shows the degree of improvement in the GAF score in the presence/absence of EPS according to schizophrenia subtypes. The degree of improvement in the GAF score (the GAF score at admission subtracted from that at discharge) was slightly higher in the group with EPS (11.4 ± 12.5) than in those without EPS (9.4 ± 12.9). According to schizophrenia subtypes, the degree of improvement in EPS was marked for the catatonic type, but no significant differences were observed because of the low number of subjects. The degree of improvement in the GAF score also did not differ among the types of EPS (Table 2).

Table 3.  Changes in GAF in the presence or absence of EPS for each schizophrenia subtype
Schizophrenia
subtype
 nGAF
on admission
GAF
at discharge
Improvement degree in GAF
  1. EPS, early prodromal symptoms; GAF, Global Assessment of Functioning.

Disorganized typeEPS(+)2124.8 ± 8.133.9 ± 11.99.1 ± 11.8
EPS(–)6125.8 ± 9.132.0 ± 11.16.8 ± 11.9
Whole8225.5 ± 8.832.5 ± 11.37.5 ± 11.9
Catatonic typeEPS(+)212.0 ± 0.046.0 ± 15.634.0 ± 15.6
EPS(–)522.8 ± 15.635.8 ± 7.713.0 ± 13.0
Whole719.7 ± 13.838.7 ± 10.219.0 ± 16.1
Paranoid typeEPS(+)528.1 ± 11.441.8 ± 11.513.7 ± 14.7
EPS(–)2128.8 ± 9.143.9 ± 10.915.2 ± 12.4
Whole2628.5 ± 9.743.2 ± 11.014.7 ± 12.9
Undifferentiated typeEPS(+)3630.3 ± 6.642.6 ± 8.112.1 ± 11.5
EPS(–)6031.7 ± 11.541.2 ± 11.58.7 ± 14.3
Whole9631.1 ± 9.641.8 ± 10.110.3 ± 13.2
Residual typeEPS(+)142475
EPS(–)732.1 ± 8.441.4 ± 10.49.3 ± 7.1
Whole833.4 ± 8.542.1 ± 9.88.8 ± 6.8
WholeEPS(+)6528.9 ± 8.640.2 ± 10.411.4 ± 12.5
EPS(–)15428.2 ± 10.437.8 ± 11.99.4 ± 12.9
Whole21928.4 ± 9.838.5 ± 11.510.0 ± 12.8

DISCUSSION

In the present study we evaluated EPS (such as depressive symptoms, anxiety symptoms, and OC symptoms) in 219 inpatients with schizophrenia diagnosed according to the DSM-IV(-TR).1 As aforedescribed, there have been few studies on EPS in Western countries,4,13–15 in particular no study has been conducted to evaluate EPS based on international diagnostic criteria in Japan.

In the present study EPS were limited to non-specific symptoms (such as depressive symptoms and anxiety symptoms) observed in mood disorder and anxietydisorder, not all symptoms observed in the prodromal phase of schizophrenia for the following reasons: (i) symptoms such as lack of self-activation, fatigability, and decreased thinking/concentration ability are difficult to distinguish from the negative symptoms of schizophrenia; and (ii) these symptoms have been long considered to be the symptoms of pure defect,22 endogenous juvenile–asthenic failure syndrome,23 or simple type schizophrenia.

The major results of the present study are as follows: (i) EPS were present in 65 patients (29.7%), slightly more frequent in women (male : female = 1:1.41); (ii) depressive symptoms were most frequently observed (61.5%), followed in order by anxiety symptoms (23.1%) and OC symptoms (9.2%); (iii) mean age at the onset of EPS was significantly lower for OC symptoms (14.5 ± 2.4 years) than for depressive or anxiety symptoms (approx. 20 years); and (iv) the mean period from the onset of EPS to that of schizophrenia was the shortest (2.7 ± 3.1 months) for depressive symptoms and the longest (>4 years) for OC symptoms. Each result is discussed in the following section (Table 4).

Table 4.  Comparison of study results
 Present studyKlosterkötter
et al. (2001)4
Häfner et al. (1999)14Beiser et al.
(1993)13
Huber et al. (1980)15
  1. BSABS, Bonn Scale for Assessment of Basic Symptoms; IRAOS, Instrument for the Retrospective Assessment of the Onset of Schizophrenia; OC, obsessive-compulsive; PSE9, Ninth version of the Present State Examination.

Subjects21916023272502
Male/female (ratio)98/121
(1:1.23)
84/76
(1:0.90)
108/124
(1:1.15)
56/16
(1:0.28)
209/293
(1:1.40)
Age at survey (years)33.9 ± 14.639.6 ± 11.9
Age at onset (years)
All subjects23.8 ± 8.9  
Male24.1 ± 8.4 Male: 26.7  
Female22.1 ± 9.3 Female: 30.9  
Percentage of patients with EPS, n (%)  
Whole65/219 (29.7)77/79 (97.5%)43/232 (61.6)  
Depressive symptoms40/65 (61.5) 60/143 (42.0)
Anxiety symptoms15/65 (23.1) 52/143 (36.4)  
OC symptoms6/65 (9.2)   
Others4/65 (6.2) 31/143 (21.7)  
Mean age at onset of EPS (years)
Total group20.8 ± 7.629.3 ± 10.0Male: 22.5
Male: 20.2 ± 5.6 Female: 25.4  
Female: 21.3 ± 8.9    
Depressive symptoms22.1 ± 8.3    
Male: 19.2 ± 4.2    
Female: 24.2 ± 9.9    
Anxiety symptoms19.7 ± 6.1    
Male: 19.4 ± 4.4    
Female: 19.8 ± 7.0    
OC symptoms14.5 ± 2.4    
Male: 16.0 ± 2.6    
Female: 13.0 ± 1.0    
Period from onset of EPS and that of schizophrenia (years)
Total group3.1 ± 4.05.6 ± 6.1Male: 4.22.2 ± 2.73.3
Male: 3.5 ± 4.0 Female: 5.5Male: 1.8 ± 2.5 
Female: 2.8 ± 4.0  Female: 3.4 ± 3.4 
Depressive symptoms2.7 ± 3.1    
Male: 2.5 ± 2.3    
Female: 2.9 ± 3.5    
Anxiety symptoms3.6 ± 5.0    
Male: 6.2 ± 6.9    
Female: 2.4 ± 3.5    
OC symptoms4.4 ± 6.6    
Male: 2.1 ± 3.4    
Female: 6.7 ± 9.0    
Survey methods (assessment scales used)RetrospectiveProspective
(BSABS/PSE9)
Retrospective
(IRAOS)
RetrospectiveRetrospective

EPS were present in approximately 30% of the patients in the present study. Häfner et al. retrospectively surveyed 232 patients with first-episode of schizophrenia and observed a prodromal phase for some years in 73%.14 Klosterkötter et al. evaluated symptoms before onset using the Bonn Scale for Assessment of Basic Symptoms (BSABS) and the ninth version of the Present State Examination (PSE9) and subsequently performed a prospective survey. They observed development of schizophrenia in 77 of 110 patients (70%) with EPS and two of 50 (4%) without EPS.4 Assuming that the base rate of those with EPS in the general population is 5%, 47.9% [=5 × 0.7/(5 × 0.7 + 95 × 0.04)] of schizophrenia patients would have shown EPS before the diagnosis of schizophrenia. Assuming that the base rate is 10%, 66.0% [=10 × 0.7/(10 × 0.7 + 90 × 0.04)] would have shown EPS. The slight difference between the present results and those of these previous studies may be because the definition of EPS differed (thought disorder, disorder of language understanding, visual distortions were also included in their study). The present study was a preliminary study and showed only a considerable incidence of EPS. A prospective study is necessary to evaluate the types of EPS and their course until onset.

Concerning the types of EPS, depressive symptoms were most frequently observed. This finding was consistent with that of previous reports.3,24

In the present study, interestingly, the age at onset for OC symptoms was significantly lower than the age at onset for other symptoms because (i) some schizophrenia patients often suffer from OC symptoms, (ii) schizophrenia patients with OC symptoms have more severe clinical symptoms, and (iii) OC symptoms are induced by atypical antipsychotic medications.18,28–35 Unfortunately, with regard to the frequency of OC symptoms, we could not compare the present results with those of previous similar studies given in Table 4 because there was no description of this in those studies. Iida et al., who examined the clinical features of childhood-onset schizophrenia with OC symptoms during prodromal phase, noted that the group with OC symptoms was characterized by a higher ratio of male patients, higher incidences of perinatal factors, fewer hereditary factors, longer duration of the prodromal phase, and a higher incidence of insidious onset and negative symptoms compared with the groups without such prodromal symptoms and suggested the possibility of subtype categorization.18

Byerly et al. reported that OC symptoms were developed prior to the onset of schizophrenia in only 28% (8/29) of patients.32 Therefore, patients may be more likely to develop symptoms of their psychotic disorder prior to the onset of OC symptoms.32 Moreover, they indicated that the mean time between onset of the schizophrenia and onset of OC symptoms was 0.5 ± 1.2 years. In contrast, Ohta et al. investigated the prevalence of OCD among patients who were primarily diagnosed with schizophrenia and found that the patients with OCD had significantly more severe motor symptoms than the non-OCD patients.29 Lysaker et al. also found more severe positive symptoms and greater impairment of cognitive executive functioning.36 Some of the findings from both studies such as longer duration of the prodromal phase and a higher incidence of insidious onset may be consistent with the present results, but not those of other studies. Further studies are needed.

The mean period until the onset of schizophrenia was 3.1 years in the present study, which is similar to that in the Huber et al. study.15 This period was slightly shorter than that found in the Klosterkötter et al.4 or Häfner et al. studies.14,24–27 However, Häfner et al. also reported that definite psychotic episodes such as hallucination and delusion develop 2–6 years after the onset of EPS,25 which was consistent with our results.

In the present preliminary study there were a few limitations, as follows. First, the results may not be general because the entire sample used in the present study consisted only of inpatients at the Psychiatry Unit of Niigata University Medical and Dental Hospital. Häfner et al. used a population-based sample14 and Beiser et al. also recruited the resident who was experiencing a first episode of functional psychosis.13 In other studies, however, the study sample was composed of patients referred to outpatient or inpatient units of German psychiatric university departments because of diagnostic problems.4,15 Second, it is possible that the lack of a special instrument for the assessment of potentially prodromal symptoms or onset of illness, such as the Instrument for the Retrospective Assessment of Schizophrenia (IRAOS) had an influence on our results. Third, as for help-seeking behavior, the majority of patients with first-episode psychosis may not seek professional help prior to the first psychotic episode despite experiencing mental problems/symptoms at pre-onset.24,28 To confirm the present findings therefore additional prospective studies in Japanese subjects with the aforementioned instrument are needed. Finally, there is also a limitation to interpretation of the present data because a retrospective survey method was used, although almost previous studies used the same methodology (Table 4).13–15,25–27

CONCLUSION

We evaluated EPS (such as depressive, anxiety, and OC symptoms) in 219 inpatients with schizophrenia diagnosed according to the DSM-IV(-TR), and showed that non-specific EPS are often present for some years before the onset of schizophrenia in Japan, as has also been reported in Western countries. Approximately 30% of the patients developed non-specific and general prodromal symptoms before the onset of schizophrenia, which is important in the understanding of pathology and for early detection and treatment. These results also indicate the necessity for both further symptomatological studies and the understanding of pathology from viewpoints other than symptomatology.

Ancillary