Study subjects

The total population in the rural area of Xinjin County (framework populations) was 235 546. The six townships (including 149 231 people in the rural community) were randomly selected from all 14 townships of Xinjin County in the south of Chengdu. An epidemiological investigation of all members of rural communities was conducted in the six townships of Xinjin County in March 1994. Methods used were the same as those reported by Shen et al (Reference Shen, Chen and Zhang1986) and Cooper & Sartorius (Reference Cooper and Sartorius1996) in a 1982 survey. We carried out face-to-face interviews with the head of each household (together with the key informant method) in order to identify potential cases of severe mental disorder. The Screening Schedule for Psychoses (Reference Shen and WangShen & Wang, 1985) was completed via interview with the head of the household, and discussion with village doctors and neighbours comprising the survey sample. All the patients who met the schizophrenia criteria of the Chinese Classification and Diagnostic Criteria of Mental Disorder (CCMD-2-R) (Reference YangYang, 1989) and the ICD-10 (World Health Organization, 1982) were included in the study. Then all patients with schizophrenia were followed up for 2 years.

Measurement

Widely used national epidemiological schedules and rating scales were used to assess symptoms and social abilities. Standard instruments including the Screening Schedule for Psychoses, Present State Examination (PSE-9, Chinese translation), the General Psychiatric Interview Schedule and Summary Form (including the project diagnosis), and the Social Disability Screening Schedule (SDSS) were used (Reference Shen and WangShen & Wang, 1985; Reference Shen, Chen and ZhangShen et al, 1986; Reference Cooper and SartoriusCooper & Sartorius, 1996). When positive answers were obtained for a subject on the screening procedures for psychosis, a comprehensive general psychiatric interview was then completed with that subject by an investigator who was a member of the survey team. All the investigators were lecturers, associate professors and professors of psychiatry. The senior professor trained investigators, reviewed all ratings and completed ratings for many of the patients. The mean percentage agreement on the Screening Schedule for Psychoses, PSE-9 and SDSS on the ratings for 10 patients ranged from 80.5 to 99.0%. Kappa values between pairs of investigators ranged from 0.74 to 1.0. Two main assessments were conducted, the first at the beginning of the study, and the second at the end of the follow-up period.

Number of cases identified and prevalence rates

Of the 123 572 people aged 15 years and over assessed using the Screening Schedule for Psychoses, a total of 510 were identified as either having schizophrenia at the time of the survey (n=367), or as having recovered but had a history of schizophrenia in their past (n=143). This gives a point prevalence rate of 2.97 per 1000 population aged 15 years and over, and a total prevalence rate of 4.13 per 1000.

Treatment and management

Thirty of the 510 patients (5.9%) maintained regular antipsychotic treatment (regular treatment group) for more than one year, and 156 (30.6%) received no treatment at all (never-treated group). The others were treated by a variety of agencies, including: trained practitioners of traditional Chinese medicine (such as acupuncture and herbal remedies - traditional treatment group), psychiatric hospitals, psychiatric out-patient clinics, and local healers, but received irregular treatment or discontinued treatment (brief and/or irregular treatment group), as shown in Table 1.

Table 1

Treatment received by 510 patients with schizophrenia

Treatment	n (%)
Maintained regular antipsychotic treatment for more than one year (regular treatment group)	30 (5.9)
Hospitalised once or maintained irregular or discontinued antipsychotic treatment (brief and/or irregular treatment group)	218 (42.7)
Traditional Chinese method (traditional treatment group)	106 (20.8)
No treatment (never-treated group)	156 (30.6)

All the 510 patients were in the age range 15-95 years; the mean age, age of onset and duration of illness for the four groups of patients are shown in Table 2.

Table 2

Patients' age, age of onset and duration of illnesss

	Regular treatment group (n=30)	Brief and/or irregular treatment group (n=218)	Traditional treatment group (n=106)	Never-treated group (n=156)	P
Mean age (s.d.)	35.9 (13.4)	42.3 (14.5)	47.3 (16.8)	48.2 (15.2)	> 0.05
Age of onset (s.d.)	27.9 (10.0)	28.9 (12.0)	32.3 (13.8)	34.2 (13.4)	> 0.05
Duration (years)	7.8 (7.5)a	12.7 (10.4)b	15.0 (11.6)c	13.6 (12.1)d

The results indicated that the duration of illness in the regular treatment group was significantly shorter than that in the brief and/or irregular treatment, traditional treatment and never-treated groups.

Characteristics of patients who had never received treatment

Of the total 156 patients who had never been treated at all, 39.7% were single, 41.7% were married, 8.3% were divorced, and 10.3% were widows or widowers. This divorce rate is very much higher than that of the overall Chinese sample in the Nationwide Epidemiological Survey, 1982 (<1%) (Reference Cooper and SartoriusCooper & Sartorius, 1996). The gender ratio (male:female) was 1:0.93, and no significant difference was found between the proportion of males (n=81, 51.9%) and females (n=75, 48.1%; P>0.05). The marital status indicated that the sample included more unmarried male patients (42.0%) than female patients (4.0%; P<0.01); the mean (s.d.) number of people per household was 3.45 (1.50), and 34 patients (21.8%) lived alone. The rate of illiteracy or those who had never attended school was 25.0%, 57.1% had attended primary school and 17.9% had attended middle school. The socio-economic level of each household was assessed: 5.8% were high, 32.0% were middle and 62.2% were low. This indicated that most patients who had never received treatment were from a poorer family background.

The reasons for not receiving treatment included: lack of money (n=55, 35.3%); relatives' uncertainty about the patients' illness (or not thinking the patient was suffering from mental illness) (n=46, 29.5%); patients' refusing to accept drug treatment (n=29, 18.6%); no relatives to care for patient (n=23, 14.7%); and local unavailability of medical facilities (n=3, 1.9%).

The 10 most frequent symptoms of schizophrenia in 156 patients (PSE-9 definitions) were: delusions of persecution (39.7%), apathy (34.0%), incongruency of affect (28.2%), splitting of thought (26.3%), withdrawal (24.4%), affectively neutral auditory hallucinations (21.2%), delusion of observation (21.2%), auditory hallucinations (commentary) (19.2%), poverty of thought (17.9%), and other delusions (15.4%).

Patients' ability to work

Among the patients who had never received treatment, the percentages of patients who could do full- or part-time farm- or housework, or could do no work were 32.1% (n=50), 45.5% (n=71) and 22.4% (n=35), respectively.

Social disability

Among the patients with schizophrenia who had never received treatment, there were 129 (82.7%) whose social ability was rated as being more than a mild disability. Of the 129 patients, 69 (53.5%) were rated as having the most seriously impaired social functioning, 19 (14.7%) were rated as serious, 14 cases (10.9%) were rated moderately impaired, and 27 cases (20.9%) were rated as mildly impaired.

Suicidal thoughts and behaviour

Of the 156 patients who had not received treatment, seven had experienced suicidal thoughts or behaviour during their illness history. The rate of suicidal thought and behaviour was 4.5%. The prevalence of suicidal thoughts and behaviour found in this study was not consistent with some previously published reports in Western countries (Reference Black, Warrack and WinokurBlack et al, 1985; Reference Landmark, Cernovsky and MerskyLandmark et al, 1987). Between 20% and 42% of patients with schizophrenia attempt suicide, and 10-15% are successful in these studies. The results of our study were the same as that found in a study in India (Reference Verghese, John and RajkumarVerghese et al, 1989). This might be caused by cultural differences and differences in the groups of patients studied. In a rural area of China, the patients with schizophrenia might endure less social pressure and discrimination than in other places.

Baseline clinical status

The baseline clinical status of the 510 patients with schizophrenia is shown in Table 3. The clinical status in the regular treatment group was significantly better than that in the brief and/or irregular treatment, traditional treatment, and never-treated groups (P<0.05). No significant difference was found between the brief and/or irregular treatment and traditional treatment groups in clinical status. This result might be caused by the fact that none of the patients maintained regular drug treatment in the brief and/or irregular treatment group. Meanwhile, this result also supports the theory that the Chinese traditional treatment group will have a slightly improved outcome. The clinical status of the nevertreated group was the worst of the four groups (P<0.001).

Table 3

Clinical status among the four groups

	Cases	Complete remission (%)	Partial remission (%)	Marked symptoms (%)	Deteriorated (%)
Regular treatmenta	30	30.0	36.7	26.7	6.6
Brief and/or irregular treatmentb	218	31.2	14.2	45.9	8.7
Traditional treatmentc	106	31.1	13.2	44.4	11.3
No treatmentd	156	9.6	8.3	75.7	6.4
Total	510	24.5	13.4	53.7	8.4

Factors affecting outcome

For the 156 patients with schizophrenia who had never received treatment, the study showed that patients' mean age, age of first onset, family history of mental illness, marital status, psychosocial factors and their family care state were not significantly related to the clinical outcome (P>0.05) (see Table 4). There were 58.3% patients who had no partner or had not married, and 37.8% of patients whose family care status was poor. All these results showed that their social support state was not good.

Table 4

Factors affecting clinical outcome of never-treated patients

Factor	Cases n	Complete remission (%)	Partial remission (%)	Marked symptoms (%)	Deterioration of illness (%)	P
Gender
Male	81	3.70	6.17	80.25	9.88	< 0.05
Female	75	16.00	10.67	70.67	2.66
Age (years)
< 40	40	7.50	12.50	70.00	10.00	> 0.05
≥ 40	116	10.34	6.90	77.59	5.17
Age of first onset (years)
< 40	106	8.49	9.43	73.59	8.49	> 0.05
≥ 40	50	12.00	6.00	80.00	2.00
Duration of illness (years)
< 1	14	35.72	7.14	57.14	0	< 0.05
1-5	37	5.40	8.11	78.38	8.11
≥ 5	105	7.62	8.57	77.14	6.67
Family history of mental illness
Yes	53	11.32	3.78	75.47	9.43	> 0.05
No	103	8.74	10.68	75.73	4.85
Marital status
Partner	65	6.15	3.08	83.08	7.69	> 0.05
No partner	91	12.09	12.09	70.33	5.49
Family care state
Good	97	12.37	9.28	72.16	6.19	> 0.05
Insufficient	59	5.08	6.78	81.36	6.78
Psychosocial factors
Yes	51	17.65	5.88	72.55	3.92	> 0.05
No	105	5.72	9.52	77.14	7.62

There was a significant difference in the natural clinical outcome between male and female patients (P<0.05). The female patients' clinical prognosis and rate of complete remission were better and higher than those of male patients. More male patients had marked symptoms and deteriorated.

Among those patients whose duration of illness was less than 1 year, there were five cases (35.7%) who recovered fully, and none deteriorated. For those patients whose duration of illness was more than 1 year, the percentage of complete remission was less than 7.6%, and deterioration more than 6.7%. The results strongly indicate that the longer the duration of the illness, the worse the patient's prognosis (P<0.05).

Outcome at follow-up

We followed up all the patients who had never received treatment for 2 years. We provided community mental health services in the pilot area, during the follow-up period, and 55 people (35.3%) accepted treatment. Six patients (3.8%) had died: one in an accident, two from physical diseases and three because of poor care. Ninety-five (60.9%) still remained without treatment.

Of 95 cases, 12 had recovered, while two patients relapsed during the 2-year follow-up. The relapse rate was 8.3% per year.

At the end of follow-up period, the clinical outcome of 95 patients with schizophrenia who had not received treatment showed that there were 10 patients (10.5%) whose illness was in complete remission, 11 patients (11.6%) whose illness was in partial remission, 68 patients (71.6%) whose illness still maintained marked symptoms, and six patients (6.3%) whose illness had deteriorated. There was no significant difference in clinical status between baseline (Time 1) and follow-up (Time 2) (P>0.05).

CLINICAL IMPLICATIONS

• ▪ The natural course of schizophrenia, especially clinical outcome, was poor in this rural Chinese area.

• ▪ For people with schizophrenia, some form of treatment, especially regular antipsychotic medication, is better than no treatment at all.

• ▪ Occupational functioning in relation to schizophrenia did not correspond with symptom severity.

LIMITATIONS

• ▪ More detailed measures should be conducted at 2-year follow-u.

• ▪ Measurement scales should be more comprehensive.

• ▪ Longer-term follow-up of these subjects is desirable.