Factors influencing duration of untreated psychosis in first episode schizophrenia in a Malay community in Malaysia


* Rohayah H & ** Hasanah C



Background: Duration of untreated psychosis (DUP) in first episode schizophrenics patients in average was very long (one to two years). However which factors influenced DUP in first episode Schizophrenia in the Malay population has not been fully evaluated.
Aims: To determine factors associated with DUP in first episode Malay Schizophrenic patient admitted to psychiatric ward at Hospital Kota Bharu, Kelantan
Methodology: Forty consecutive first episode schizophrenia who fulfill the criteria for the study were selected. At hospitalization the sociodemographic and DUP were determined from patients and relatives. Patients’ symptoms were assessed using the Positive and Negative Syndrome Scale (PANSS) and level of functioning were assessed using Global Assessment of Functioning (GAF) scale. An assessment of patient social network done by a semistructured questionnaire.
Results: The mean DUP was 159 weeks ( nearly three years). Patient with long DUP (>26 weeks) had more negative symptoms and socially withdrawn compared to the short DUP (<26 weeks). Most of them are severely ill and brought by family members to hospital after failed traditional treatment. The long DUP group of patients have a smaller number of friends.
Conclusion: The DUP of our patients was longer than previous study in the west. The fact that they have more negative symptoms at hospitalization might delay the help seeking behaviour of their family. The prolong DUP may disrupt their social network, especially in socializing with friends.

Key words:

first episode schizophrenia, duration of untreated psychosis, negative symptoms, social network



DUP is the time interval between onset of psychotic symptoms and first effective treatment. Studies of first episode psychosis worldwide found that the average DUP is about one to two years. (Johnstone et al, 1986, Beiser et al, 1993). The delay in treatment will results in poor outcome (Birchwood et al, 1998). Loebel in a first episode study in New York also noted that those patients who had psychiatric symptoms for long periods without antipsychotic medication had a slower and less complete recovery, and increased subsequent risk of relapses than those who received prompt treatment (Loebel et al, 1992). Carbone et al. (1999) in his study of first episode psychosis showed that 1-year outcome could be significantly enhanced by more intensive treatment only if the DUP was between one and 6 months.
Larsen et al (1996) found that the long DUP group of patients had more negative symptoms at hospitalization. DUP is partly determined by concern about the illness by sufferers or people in their social networks. This concern is influenced by the poor insight and negative symptoms (social deficits) of the sufferers. Some researchers reported that the delay was because the patient did not want treatment (Helgason, 1990, Amador et al, 1994), or were not offered treatment ( Lincoln et al, 1998, McGorry et al, 1996).
In Malaysia, from our experience the DUP varies from days to many years. However to my knowledge, there is no research done to evaluate factors influencing DUP in first episode schizophrenia in a solely Malay community in this country. This study was conducted at Department of Psychiatry, Hospital Kota Bharu, Kelantan, Malaysia which has 122 beds psychiatric ward. The Kelantan state is located at the east-coast of peninsular Malaysia and 93% of its population are Malay (Statistic, 1998) . In Malay community, traditional treatment is the preferred choice of psychiatric treatment since ancient times and still being the popular practice till today.

Aims and methods


This study is designed to look at the association between DUP in first episode Malay Schizophrenic patients with the demographic variable, clinical presentation, severity of illness at first admission and patient social network. We hypothesized that patients with long DUP are male, unemployed, have low educational level, low family income and from rural areas; had more negative symptoms, more severe symptoms and poorer GAF and are those who had no spouse, not staying with parents or relatives and had smaller number of close friends. This study was approved by the local research and ethical committee.

Study design and sampling

The study consists of forty people with schizophrenia, admitted for the first time to Psychiatric ward, Hospital Kota Bharu, kelantan. Inclusion criteria were patient and/or relatives give verbal consent, fulfill DSM-IV criteria for Schizophrenia or Schizophreniform Disorder, no previous hospitalisation for psychosis and no previous neuroleptic treatment. We exclude patient with underlying organic pathology (e.g. mental retardation, dementia etc.), evidence of neurological or serious medical disorder or being intoxicated with narcotics.

Diagnosis and procedure

The diagnosis was made using the DSMIV criteria (APA, 1994). The researcher and the hospital team including a consultant psychiatrist reviewed the diagnosis in a consensus meeting. The same investigator conducts all interviews for this study. All consented patients will be interviewed within the first 48 hours after admission.
The investigator conducted a semi-structured interview with patient and their relatives to get demographic data and DUP. Patient symptoms were evaluated with the Positive and Negative Syndrome Scale (PANSS) which is an established scale to assess symptomatology of Schizophrenia (Kay et al, 1987). Patient level of functioning in the last week before hospitalization was assessed from the patient and relative’s history using the Global Assessment of Functioning (GAF) scale. The scale varies from 0 to 100, 0 being the lowest and 100 the highest score.
Onset measures
The onset of psychosis is defined as follows: a score of four or more on PANSS positive sub-scale and/ or manifestation of psychotic symptoms such as delusion, hallucination, thought disorder or inappropriate bizarre behaviour in which the symptoms are not apparently due to organic cause; these symptoms must have lasted throughout the day for several days or several times a week, not only limited to a few brief moment
( Larsen et al, 1998).
DUP is defined as the time interval between the onset of psychotic symptoms and hospitalisation for psychosis or initiation of adequate treatment.
Onset of the illness was ascertained by semistructured interviews with patient and family members. Multiple informants were available and were used to obtain these data for all study subjects. The onset will be determined by two ways:
a) Asking the patient and family members when the patient (or family member) first experienced ( or noticed) behavioural changes which in retrospect, appear to have been related to the patient becoming ill
b) After explaining psychosis in clear language as defined by the above criteria of onset, the patient (or family members) will be asked when the patient first experience (or notice) psychotic symptoms.
When differences between patients’ and family members` occurred, the date given by the reliable patient will be taken because most of the time the exact onset of illness had been over looked by the relatives (Beiser et al, 1993, Loebel et al, 1993). For the unreliable patient a consensus was made using information from multiple informants.
Social network were determine by asking the patients and/or relatives regarding their circle of friends using a semi structured questionnaires.
The data were analysed using the Statistical Programme for Social Science (SPSS) version 10.0 for windows. The DUP was categorized into two groups using 6 months (26 weeks) as a cut off point for short and long DUP. Logistic regression analysis was used to test the association between DUP and demographic data, illness variable and social network.


For the total sample the mean value of DUP was 159 weeks, med= 40 weeks, and s.d. =290.29.

Patient sociodemographic data
The total number of participants was forty. All of them were Malays. Table 1 showed the short and long DUP sample characteristic.

DUP and Clinical variables

Table 2 showed psychopathological measures in the DUP. DUP was significantly associated with negative sub-scale of PANSS (OR = 1.122, p = 0.01, 95% CI=1.027 – 1.226). The long DUP group of patients had more negative symptoms at hospitalisation. DUP was not associated with positive and general subscale of PANSS neither the GAF score (table 3). Logistic regression analysis using DUP as a dependent variable revealed no statistically significant association between DUP and the length of hospital stays or the total PANSS score. Further analysis of the PANSS item revealed that items related to social withdrawal were significantly associated with DUP including emotional withdrawal, poor rapport, passive/apathetic social withdrawal, mannerism/posturing and active social avoidance.
DUP and Social network
The data collected for social network component were continuous data by giving score to reflect the quantity and quality of social network the patient has at the time of hospitalization. Variables measure include either patient having a spouse or not, employment status (unemployed to fulltime job), living status( alone to staying with spouse/relatives) and number of friends. Number of friends the patient has at the onset of illness was significantly associated with the DUP (OR=0.515, p=0.024, CI (0.289 – 0.915). The long DUP group of patients have fewer friends in which eight subjects in this group did not have any friends. The other variables did not show any association with DUP(table 4). It was noted that majority of the patients (77.5%) stayed either with their spouse or parents before admission to the hospital.

Multiple logistic regression
The entire significant finding from the simple logistic regression analysis was entered consecutively using a forward stepwise logistic regression method, all of the variables analysed were not associated with DUP except negative symptoms score (OR=2.43; CI 1.06 – 6.15; p=0.03) and item active social avoidance (OR= 2.65, CI=1.13 –6.26; p=0.026).


The mean DUP for our sample was 159 weeks (about three years) which is much longer compared to findings in developed countries (Crow et al, 1986 ; Rabiner et al, 1986 ; Loebel et al, 1992; Liberman et al, 1993; Larsen et al, 1996 ; Larsen et al, 1998). The prolonged DUP could reflect our samples, which were represented by Malay ethnicity and living in a developing country where the psychiatric services are still lacking compared to developed world. Data gathered from the carers of the same samples found that those with long DUP had visited one or more traditional healers more frequently, from a few visits to up to 40 visits.( Rohayah, 2001). In fact carers claimed that they came late to hospital mainly because they had tried traditional treatment first. This finding is consistent with the previous study showing the preferences of traditional healer treatment in Malaysia, especially the Malay community ( Yeoh,1980, Razali et al, 1996, Razali, 1998). Most of them view the symptoms of mental illness as a form of disturbance by evil spirits and this assumption was reinforced by the traditional healers. Though Malaysia has developed significantly over the last twenty years, there is not much changed in sociocultural beliefs about and attitudes towards traditional healers and their role in many diseases.

There were more males (n=28) than females (n=12) in the ratio of 7 to 3. Males outnumbered females in both short and long DUP group of patients, but the difference was not statistically significant. Females were older on admission, about 6 years. The lower incidence of female admission taken in conjuction with the older age at admission might be due to their nature of presentation; usually less aggressive. This is true as the reason for seeking treatment in males were mainly aggressive behaviour.

From the simple logistic regression analysis DUP was significantly associated with age at hospitalization (OR = 1.137, p = 0.0019, CI=1.021 – 1.266). Thus, the long DUP group of patients were hospitalized later; though their age of onset of illness does not have any significant differences. (OR=1.062, p = 0.266, CI=0.956 – 1.179).

Most of the patients only came for treatment when the symptoms were intolerable. We found that at hospitalization patients were moderate to severely ill as reflected by the total PANSS score (ranged 69-148) and the mean total PANSS score was 111.65. This is consistent with the GAF score that ranged from 30-50, reflecting serious symptoms or impairment and poor judgment. As expected the long DUP group of patients had more negative symptoms at hospitalization compared to the short DUP group of patients (OR=2.561, p=0.132, CI=1.067 – 6.148). This finding was consistent with previous studies of first episode Schizophrenia (Larsen et al, 1996; Larsen et al, 1998, Waddington et al, 1995). This can be explained by the chronicity of the illness as well as the negative symptoms were much tolerated by the carers, thus delaying treatment. On the other hand, in Malay culture, patients with negative symptoms might not be recognized as having any illness and the social withdrawal, lack of initiative and emotional bluntedness could be misinterpreted or even accepted as female passiveness and adolescence acquiesence within the family. For the patients themselves, the negative symptoms such as social withdrawal might lead them to deviate from the society thus reducing their social network and therefore reducing concern by others. As they themselves have poor insight and were not concerned about the illness, they were left out. Multiple logistic regression also revealed that active social avoidance item of PANSS was significantly associated with DUP. (OR=2.562, P=0.0352, CI=1.067 – 6.148) which replicates the finding by Larsen et al, in 1998.
For the severity of the illness, though there was no statistically significant association between DUP and the total PANSS score and length of hospital stays, we feel that the length of hospital stays in this hospital might not be a good predictor of illness severity. Since this is the only gazzeted mental unit which received compulsory admission for the state of Kelantan , the pattern of admission is still that of revolving door with fast turn over of patients. Average length of hospital stays for first episode patients is between ten to fourteen days. The shortage of beds also occurred make a longer hospital stays not possible. We also found that the most common reasons why the relative brought them for treatment were aggressive behaviour, regardless their DUP. On the other hand DUP was not associated with the positive symptoms which had been shown in previous study (Larsen et al, 1996; Larsen et al, 1998).

Larsen et al 1998, showed that patients with long DUP has poor social network. The lack of friends the patient has at the onset of illness in the long DUP group of patient might be due to the illness chronicity which cause them to have smaller social network thus leaving them to be more with their house contact only. The finding is consistent with study by Hamilton, whereby schizophrenia with negative symptoms had significantly smaller social network (Hamilton et al, 1989). It was noted that all patients were brought in for admission by their relatives who stays with them. This might reflect the role of available social network ( spouse and relatives ) in Malay community in bringing patient for treatment. Malay community was known as having a strong bonding to each other especially within their family members. However, the strong belief in traditional healers made them to seek traditional treatment first.


We conclude that the mean DUP found in this study was 159 weeks (about three years), which is very long compared to other studies. Negative symptoms at hospitalisation were significantly associated with long DUP with active social avoidance being the most common symptoms. Reducing the DUP and early detection is important to prevent further deterioration and undesirable complications. Several strategies can be applied;
a) Implementation of psycho-education program about mental illness specifically Schizophrenia in the community,
b) To emphasize that Schizophrenic patients with negative symptoms need early treatment eventhough they are managable/ not aggressive
c) The concern among individual social network about any peculiarity or behavioural changes of others need to be emphasized
d) The Malay community belief and preference towards traditional healers as a first contact person for treatment of psychosis need to be addressed.



I would like to thank all the staffs from Psychiatric Department Hospital Kota Bharu, Kelantan for their encouraging help to complete this study.


Table 1: Short and long duration of untreated psychosis (DUP) sample characteristic



Short DUP


Long DUP





Educational status


   Form 5 (11years)

   Form 3 (9 years)

   Not schooling   









Marital status












Domicile area







Age at hospitalisation, mean (sd)



Age at onset of psychosis, mean (sd)



Table 2: Psychopathological measures in the different DUP types

Total sample

Short DUP

Long DUP

GAF score, mean (sd)




*PANSS at hospitalization

          mean (sd)

   Positive symptoms

   Negative symptoms

   General symptoms

   Total score

25.6 (6.26)

25.6 (9.18)

60.2 (11.38)

111.7 (20.48)









Length of hospitalisation, mean (sd)

16.3 (11.63)



*PANSS= Positive and Negative Syndrome Scale


Table 3: Logistic regression analysis between DUP and clinical variables ( level of significant=95%, P<0.05)

PANSS item

Odd Ratio ( OR)

95% CI for OR

P value

PANSS positive


0.890 – 1.089


PANSS negative


1.027 – 1.226

0.011 *

PANSS general


0.986 – 1.113


PANSS total


0.996 – 1.065


GAF score


0.812 – 1.032


Length of hospitalisation


0.982 – 1.133


*significant p < 0.05

Table 4: Logistic regression analysis with the short and long DUP and social network scores ( level of significant=95%, P<0.05)

Social network variable

Odd Ratio

95% CI for OR


Having a spouse


 0=no spouse,

 1= having spouse)


0.296 – 13.507


Employment status

(score: 0=not working, 2=part time,  4=full time)


0.884 – 2.003


Living status at the onset (score:

6=staying alone, 4=friends, 2=siblings/relatives, 0=spouse)


0.534 – 1.097


Number of friends


0.289 – 0.915


* stastically significant p< 0.05


Amador, X F, Flaum, M, Andreason et al. (1994) Awareness of illness in schizophrenia, schizoaffective and mood disorder. Archive of General Psychiatry. 51, 826-836.

American Psychiatric Association. ( 1994) Diagnostic and Statistical Manual of Mental Disorders. Fourth Edition (DSM- IV), 4th. Edn. Washington, DC.

Beiser, M, Erickson, D, Flemming JAE. (1993) Establishing the onset of psychotic illness. American Journal of Psychiatry.150, 1349-1354.
Birchwood, M, Todd, P, Jackson C. (1998) Early intervention in psychosis: the critical period hypothesis. British Journal of Psychiatry. 172 (Suppl 33), 53-9.
Carbone, S, Harrigan, S, McGorry, P, Curry, C, Elkins, K .(1999) Duration of untreated psychosis and 12-month outcome in first-episode psychosis: the impact of treatment approach. Acta Psychiatr Scand. 100, 96-104.
Crow, TJ, MacMillan, J F, Johnson, A L, et al. (1986) The Northwich Park study of first episodes of schizophrenia : II. A randomised controlled trial of prophylactic neurotic treatment. British Journal of Psychiatry. 148, 120-127.

Hamilton, N G, Donzoha,C A, Cutler, D L et al. (1987) Social networks and negative versus positive symptoms of schizophrenia. Scizohrenia Bull. 15(4), 625-633.

Helgason, L. (1990) Twenty years follow up of first psychiatric presentation for schizophrenia : what could have been prevented: Acta Psy. Scandinavica. 81, 231-235.
Johnstone, E C, Crow,T J, Johnson, A L, MacMillian, J F. (1986) The Northwick Park study of first episode schizophrenia: I. Presentation of the illness and problems relating to admission. British Journal of Psychiatry. 148, 115-120.

Kay, Lewis, A, Opler, Lindenmayer , JP. (1987).Reliability and validity of the Positive and Negative Syndrome Scale ( PANNS) , Psychiatry Research. 23, 99-110.

Larsen, T K, Johannsern, J O, Opjordsmoein S. (1998) First episode schizophrenia with long duration of untreated psychosis: Pathways to care, British Journal of Psychiatry. 172 (suppl 33), 45-52.

Larsen, T K, McGlashan, T H, et al.(1996) First-episode schizophrenia: I. Early course parameters. Schizophrenia Bulletin. 22, 241-256.

Lieberman, J A. (1993) Prediction of outcome in first-episode schizophrenia.J Clin Psychiatry. 54(3) (suppl), 13-17.

Lincoln, C, Harrigan, S, Mc Gorry, P. (1998) Understanding the topography of the early psychosis pathways, British Journal of Psychiatry, 172(suppl 33), 21-25.
Loebel, A D, Lieberman, J A, Alvir, J M, et al.(1992) Duration of psychosis and outcome in first-episode schizophrenia. Am J Psychiatry.149, 1183-1188.

McGorry, P D, Edwards, J, Mihalopoulos, C, Harrigan, S M, et al.(1996) EPPIC: an evolving system of early detection and optimal management. Schizophr Bull. 22, 305-326.
Rabiner, C J, Wegner, J T, Kane, J M. (1986) Outcome study of first-episode psychosis: relapse rates after one year. American Journal of Psychiatry. 143, 1155-1158.
Razali, S M, Khan, U A, Hasanah, C I. (1996) Belief in Supernatural causes of mental illness among Malay patients : Impact on Treatment. Acta Psychiatr Scand. 94, 229-33.
Razali, S M. (1989) The Consultation of traditional healers by Malay patients. Medical Journal Malaysia. 44, 3-12.

Rohayah, H. (2001) Factors influencing duration of untreated psychosis in first episode schizophrenia. Dissertation Master of Medicine, Psychiatry. University Sains Malaysia.
Statistic. (1998) Annual Report, Department of Statistic Malaysia
Waddington, J L, Youssef, H A, Kinsella A .(1995) Sequential cross-sectional and 10-year prospective study of severe negative symptoms in relation to duration of initially untreated psychosis in chronic schizophrenia. Psychol Med, 25, 849-857.
Yeoh, O H. (1980) Malay Psychiatric patients and traditional healers (Bomoh). Med. J. Malaysia. 34(4), 349-357.



* Lecturer and Psychiatrist, Department of Psychiatry, School of Medical Sciences, Universiti Sains Malaysia Health Campus, 16150 Kubang Kerian, Kelantan

**Associate Professor and Consultant Psychiatrist, Department of Psychiatry, School of Medical Sciences, Universiti Sains Malaysia Health Campus, 16150 Kubang Kerian, Kelantan

Corresponding author:
Dr. Rohayah Husain, Lecturer and Psychiatrist, Department of Psychiatry, School of Medical Sciences, Universiti Sains Malaysia Health Campus, 16150 Kubang Kerian, Kelantan


Copyright © Priory Lodge Education Limited 2007

First Published June 2007

Home • Journals • Search • Rules for Authors • Submit a Paper • Sponsor us   
Rules for Authors
Submit a Paper
Sponsor Us

Google Search

Advanced Search



Default text | Increase text size