Self-esteem of out-patients with schizophrenia: The association of self-esteem and socio-demographic and clinical characteristic.
A.B. Shamsunnisah** & C. I. Hasanah*
* Associate Professor, Department of Psychiatry, School of Medical Sciences, 16150, Kubang Kerian, Kelantan, Malaysia
** Consultant Psychiatrist, General Hospital Sultan Abdul Halim, Sungai Petani, Kedah, Malaysia
Objective: Self-esteem is an important component of psychological health. The study aims to determine the association of self-esteem with socio-demographic and clinical characteristics of patients with schizophrenia.
Methods: The validated Malay version of Rosenberg Self-esteem scale (RSES) was used concurrently with Shamshunnisah Self-esteem Scale (SSES) in assessing level of self-esteem of out-patients with schizophrenia. Other assessments were BPRS (Brief Psychiatric Rating Scale), GAF, socio-demographic details, simple assessment on the acknowledgment of illness, perceived self-devaluation, social support and overall feeling of quality of life and health satisfaction by using the Q1 and Q2 of WHOQOL-BREF. The SSES provided the main data for this study and the overall score of SSES were used as dependent variables for data analysis.
Results: Using SSES, 15(9.1%) patients scored in the low self-esteem range, 84 (50.9%) patients scored in the moderate self-esteem range and 66 (40.0%) patients scored in the high self-esteem range. Under multiple linear regression analysis, the socio-demographic factors associated significantly with high level of self-esteem were being married and perceived social support. Those who perceived self-devaluation was associated with lower self-esteem. Clinical factors associated with higher self-esteem were denial of mental illness, perceived good overall health and quality of life, and being free from psychotic and depressive symptoms.
Conclusions: Majority of patients with schizophrenia had a relatively good self-esteem. High level of self-esteem in schizophrenia was associated with more subjective factors rather than clinical, occupational and functional factors.
Keywords: self-esteem, SSES (Shamshunnisah self-esteem scale), schizophrenia, RSES (Rosenberg self-esteem scale)
Studies on subjective experience, especially on self-esteem were lacking, and none so far in Malaysia. The subjective experience of schizophrenia consisted of self-esteem, satisfaction of life and subjective distress (Brekke et al, 1995). Self-esteem refers to an individual’s sense of value or worth, or the extent to which a person appreciates, or likes himself (Taylor et al, 2000).
Previous researches indicated that lowered self-esteem frequently accompanies or became an etiological factor in many psychiatric conditions (Robson, 1998; Silverstone et al, 2003). Wright et al (2000) stated that the mentally ill people who experienced negative self-appraisals performed badly in community, and were more likely to relapse. Patients with schizophrenia and low self-esteem are expected to have a compromised quality of life and poor psychosocial functioning (Breeke et al., 2001; Gureje et al., 2004). Link et al (2001), using RSES (Rosenberg Self-esteem scale) found 24% of the people with schizophrenia scored below the mid-score (reflecting low score and low self-esteem). Silverstone et al., (2003) found that patients with schizophrenia had intermediate levels of self-esteem. However, when compared to other psychiatric conditions, this group of patients had significantly low self-esteem.
Torrey et al (2000) examined self-esteem using RSES in adults with chronic psychotic illness in vocational rehabilitation program and found that at baseline, majority of the participants had above or near the average self-esteem. Many studies failed to show correlation between self-esteem in people with schizophrenia with any socio-demographic factors, except in one study, where female patients were found to have higher self-esteem than their male counterpart and those who had at least one close friend was associated with positive self-esteem (Sorgaard. et al, 2002). Torrey et al (2000) specifically looked at the correlation of employment status and self-esteem, and had not found any. Positive and high self-esteem among people with schizophrenia were those who were not depressed in the BPRS affective symptom sub-score and those who perceived good quality of life or satisfied with their life (Torrey et al, 2000; Sorgaard et al, 2002; Gureje et al, 2004).
Link et al (2001) studied the impact of stigma on self-esteem of people with severe mental illness. 36% of the subjects were schizophrenia. He found that those who had no intervention to cope with the stigma had low self-esteem than those who had intervention to cope with the stigma. Upward social comparison and high goal for living in people with schizophrenia placed them at risk of depression and low self-esteem (Sing-Fai, 2000). Brekke et al (2001) studied executive functioning as a moderator of the relationship between psychosocial functioning and the self-esteem. They found that patients with impaired executive functioning displayed a positive and statistically significant association between psychosocial functioning and self-esteem but those with intact executive functioning showed negative association.
The specific scale or tools used in the above studies were not developed to measure self-esteem in people with schizophrenia, and most relied on RSES. This study measured self-esteem by the established RSES and Shamshunnisah Self Esteem Scale (SSES), the locally designed self-esteem scale for patients with schizophrenia.
The study aims to determine the level of self-esteem and the factors that are associated with the level of self-esteem in patients with schizophrenia attending outpatient psychiatric clinics of two major hospitals in Kota Bharu, Malaysia.
Study sample and procedure
165 patients with schizophrenia attending outpatient psychiatric clinics of HUSM (Hospital Universiti Sains Malaysia) and HKB (General Hospital Kota Bharu) were included. The diagnosis of schizophrenia was based on DSM-IV classification of Mental Disorder. They were stable patients, scoring 0 or 1 on BPRS (Brief Psychiatric Rating Scale) psychotic symptom subscale (Bech et al. 1986). Patients with co-morbid substance misuse or dependence or Axis II and III diagnoses were excluded (DSM-IV).
The study protocol was approved by the Research & Ethics Committee, Universiti Sains Malaysia. A single researcher trained in psychiatric interview and examination, interviewed all the subjects individually.
Two self-esteem scales were used. The Malay version of RSES (10 items) has been validated by Mahmood et al (1999) with Cronbach alpha value of 0.63. SSES has 25 items, conceptualized locally in a population consisted of healthy subjects and subjects with schizophrenia. The questionnaire developed from initial qualitative research and validated for validity and reliability in the same population. The scale has two domains: Self-worth domain with 15 items (Cronbach alpha 0.88) and social relation or functioning domain with 10 items (Cronbach alpha= 0.81), and with fairly good test re-test reliability (ICC ranged from 0.44 to 0.87). Its construct validity was confirmed by confirmatory factor analysis and the concurrent validity of SSES and RSES using Pearson correlation was 0.77.
On completion, the participants filled up the second set of questionnaires which consisted of questions on: the socio-demographic details, the acknowledgment of illness, perceived self-devaluation and social support; and overall feeling of quality of life and health satisfaction by using the Q1 and Q2 of WHOQOL-BREF Malay (Hasanah et al, 2003).. After the completion of these questionnaires, the subjects were assessed for Global Assessment of Functioning (GAF).
The collected data was entered and analyzed initially using the Statistical Program for Social Science (SPSS version 10). Intercooled Stata program, version 8.0 (Stata Corp, Texas) was used for multiple linear regression analyses to eliminate confounding effects among the significant independent variables.
A total of 235 outpatients with schizophrenia were consecutively approached by the researcher. 60 patients met the exclusion criteria. From 175 patients who were given the self-esteem questionnaires, 10 failed to complete the questionnaire and self-withdrew. The final subject number was 165 patients. 81 patients were male and 84 were female. The participants’ age, ranged from 17 to 58 years with the mean of 33.4 year. 96 of the 165 patients were single, 51 of them were married and 18 of them were divorced. Majority of the patients were Malays (97.6%), educated up to secondary school (72.7%), denied history of substance abuse (96.9%), stayed with family (98.8%), and reported perceived overall support (72.7%). Only 32.1% of the subjects were unemployed and 10.9% of the subjects were housewives. About half (48.5%) of the subjects did not have monthly personal income and 21.2% of the subjects reported perceived self-devaluation Refer to table 1.
Table 1. Sociodemographic characteristics of the patients with schizophrenia (n=165)
|Sociodemographic characteristics||Frequency||% of total sample|
|Age (years) Mean 33.44 Range 17-58|
Separated / divorced
|History of substance abuse||
|Personal income (Ringgit Malaysia)||
Less than 500 Less than 1000 More than 1000
|Body Mass Index||
Normal (20-24) Overweight (25-29)
|Perceived overall support||
Socio-demographic variables and self-esteem
Statistical analysis showed that sex, age, educational level, smoking status, body mass index (BMI) and personal income did not have any significant association with the level of self-esteem. Marriage, perceived support and denial of self-devaluation are three characteristics that remained significantly associated with self-esteem after multiple linear regression analysis (MLR), Table 2.
Table 2: Significant factors associated with the level of self esteem (by using multiple linear regression).
|Factors||b (95%CI)||t stat||P value|
Marital Status c
Separated or divorced
Perceived overall support d
Perceived self-devalutaion e
|-2.33 -2.63||0.021 0.009|
Acknowledge illness f
Total Q1 and Q2 of
Psychotic symptom subscore h
Mild or doubtful
a= There is no significant interaction; no multicolinearity problem; and linearity, normality and equal variance assumptions were made. b=adjusted regression coefficient. r2=0.51. CI= confidence interval
c reference level is married, d reference level is denied perceived overall support, e reference level is denied self devalued, f reference level is acknowledged illness, g reference level is scored fair on total score of Q1 and Q2 of WHOQOL-BREF, h reference level is no psychotic symptom, i reference level is possible depression.
No significant difference found in the self-esteem score between the single and married patients or between single and separated patients. Housewives, being employed and students were found to have significantly higher self-esteem than those who were jobless in the simple linear regression (SLR). The p values for each of the criteria were 0.028, 0.024 and 0.045 respectively. However, the associations were not significant in the MLR.
Clinical variables and the self-esteem
There was no significant association between the age of onset of illness, duration of illness, frequency of admission, duration under remission, family history of psychiatric illness, types of antipsychotic medication, side-effect of medication and self-report on compliance to treatment and the level of self-esteem. Having even minimal impairment in socio-occupational functioning significantly reduced self-esteem but in MLR analysis, this association was not significant.
Four factors that remained significantly associated with the level of self-esteem in the MLR are denial of psychiatric illness, giving a response of “good” in their subjective feeling of quality of life and health satisfaction (total score of the Q1 and Q2 of WHOQOL-BREF of within 8 to 10), and on BPRS, free from psychotic symptoms and depression. (Table 2).
Results of this study is comparable to the study by Torrey et al (2000), who found above or near average level of self-esteem in remitted chronic psychotic disorder patients. Other studies (Link et. al, 2001; Gureje et al, 2004) reported about 24%-43% of the patients having low self-esteem, much higher than (9.1%) in our study. Our study also used RSES, which actually designed for normal subjects, but almost always used in previous studies. In this study, we add another self-esteem scale (SSES), which was locally designed and validated. Self esteem from both scales showed high correlation and the overall results is comparable to other similar international studies.
There are several probable explanations to our relatively small number of patients with low self-esteem. The subjects studied seemed to have ideal characteristic of patient in remission. Majority of the subjects lived with family and almost half of them were employed. These could reflect good general social support and self-worth which had been shown to be positively related with self-esteem (Taylor et al, 2000; Sorgaard et al, 2002). People who abuse drugs are associated with low self-esteem (Silverstone, et al., 2003; Mahmood, 1999), and our study excluded patients with concomitant drug use.
Western culture values independent self, but our local culture values interdependent self (Wylie, 1974; Taylor et al, 2000). Having severe psychiatric illness increased one’s dependence to others, which has the potential to harm one’s self esteem. However, in our local setting, being dependent to others might have a lesser impact to one’s self-esteem as compared to western counterpart.
Up-ward and unfavorable social comparison harms individual’s self-esteem (Sing Fai et al, 2000; Brekke et al, 2001; Silverstone et al, 2003). Compared to other places in Malaysia, there is a general understanding that life in Kelantan is less stressful, less demanding and has minimal up-ward social comparison. Economic wise, majority of the people still depend on agriculture and self-employment, and females are known for their business and entrepreneur activities. Generally, the people in Kelantan have more self-acceptance and are satisfied with what they have, irrespective of the impact of the illness to one-self and life in general. Kelantan is said to be the veranda of Mecca and governed by the Islamic political party.
In accordance with previous studies, there was no association found between sex, age, educational level, smoking status, body mass index or personal income and the level of self-esteem. This study however, found significant association between marital status and level of self-esteem. Being married could reflect many things: being a mother or a father, being responsible, relating and contributing to others, having support and some degree of social network, which are all associated with good self-appraisal and social comparison. In addition, being married also could be due to less severe illness, feeling better about one-self and being less stigmatized.
The association between the social support and level of self-esteem was shown by Sorgaard et al (2002), who found that schizophrenic patients who had at least one close friend were associated with better self-esteem than those who did not have any friend. Social supports which include economic, household or psychological could act as a buffer to both emotional and material deficiency resulted from the illness. Almost all of the subjects in this study live with their families. However, only 72.7% acknowledged perceived overall support and this group of subjects reported higher self-esteem than those who claimed otherwise.
Housewife, students, or those in employment, reported a better self-esteem than those who were jobless. Previous studies had not shown this association (Torrey et. al, 2000; Gureje et al, 2004). However, the association of working status and self-esteem disappeared with MLR analysis. It seemed that the association between these two is only superficial, probably complicated by many confounding factors such as degree of social support and functioning level. Further studies in future should take into account the nature or amount of the activity or work undertaken and the patients’ satisfaction.
Beck (1967) has suggested that an individual’s level of self-esteem is determined, in part, by judgments made by others. Stigma of mental illness leads to a substantial proportion of people with the illness to conclude that they are failures or that they have little to be proud of. Two known aspect of stigma is: perceptions of devaluation/discrimination and social withdrawal because of perceived rejection (Link et al, 2001). Previous studies found that stigma of mental illness harmed the self-esteem of many people with serious mental illnesses (Link et al, 2001; Lai et al, 2001). This study also attempted to look at the possible demonstrable relation between stigma and the level of self-esteem in schizophrenia. Using a standard scale such as the Internalized Stigma of Mental Illness (Ritsher et al, 2004) would add too many questions to the subjects. The authors decided to resort to a simple question on perceived self-devaluation as a proxy measure of stigma in this study. The proxy question as mentioned above asked whether one ever felt being treated or seen differently by people in general and the responses was either “yes” or “no”. This study found that those patients who denied perceived self-devaluation have significantly higher self-esteem than those who claimed being devalued. This finding remained to be significant with MLR analysis... This simple question on perceived self-devaluation was rather simplified and poor in specificity in evaluation perceived stigma.
Similar to previous studies (Sing Fai et al, 2000; Torrey et al, 2000; Brekke et al, 2001; Link et al, 2001), this study found no association between the level of self-esteem and general clinical information. However, patients’ denial of psychiatric illness and their subjective good feeling on their quality of life and health satisfaction were found to be significantly associated to the level of self-esteem with MLR analysis.
Insight is patient’s degree of awareness and understanding about being ill (Kaplan and Sadock, 1997). The insight to mental illness has intellectual and emotional component, where the former refers to patients’ admission of being ill and the later refers to patients’ awareness of their feelings which leads to a change in their personality and behavior (Kaplan and Sadock, 1997). Asking the patient self-acknowledgement of psychiatric illness is definitely too simplified, but it did served as a superficial assessment of insight, which could be used to relate to self-esteem in this study. The question asked whether the patient agree that he or she has mental illness and the response was either “yes” or “no”. Patients who answered “no” had significantly higher self-esteem than those who acknowledged having psychiatric illness.
Subjects in this study could be divided to those who were completely free from psychotic symptoms and those who were either having doubtful or mild psychotic symptoms. Thus, those who were free from symptoms might be more comfortable with the denial as it decreased self-stigmatized feeling. This is consistent with finding of higher self esteem in patients free of psychotic symptoms. Patients who showed mild or doubtful psychotic symptoms had significantly lower self-esteem than the patients who were actually free from any psychotic symptoms.
Denial to the self-acknowledgement of mental illness might not necessarily mean “no insight at all” as majority of these subjects come to the follow-up clinic regularly on their own. Another issue to consider is whether accepting one’s illness is harmful to self-esteem. This finding should be highlighted to mental health care providers, as there is a need to be more “patient sensitive” and organize a psycho-education program that is not only factual but carefully tailored and presented in a step wise manner. It should focus not only on information of one’s illness but also consideration of their inner subjective experience, such as their self-esteem. This preliminary finding should lead to future research looking into association of self-esteem and insight in patients with remission from schizophrenia.
Quality of life or life satisfaction and depressive states are other factors which had been shown to be significantly related to self-esteem. Schizophrenic patients who reported better perceived quality of life have higher level of self-esteem than those who had poor quality of life. (Torrey et al, 2000; Sorgaard et al, 2002; Gureje et al, 2004). Depression was associated with reduced level of self-esteem and the low self-esteem was a risk factor for impaired quality of life (Gureje et al, 2004). Hasanah & Razali (2002) studied quality of life in schizophrenia using the Malay version of WHOQOL-BREF, and found that people with this illness have most impairment in the social relationship domain.
Depressive subscale of BPRS was used in several studies to detect depression in schizophrenia (Torrey et al, 2000; Brekke et al, 2001; Sorgaard et al, 2002). This study detected 28.5% subjects with possible depression and none with definite depression. Those with possible depression had significantly lower self-esteem than those who were not. Low self-esteem is among the well known causes of depression in general but low self-esteem can also be the symptoms of depression (Kaplan and Sadock, 1997). Incidence of Post-psychotic depression in schizophrenia is about 25% and is associated with poor prognosis (Kaplan and Sadock, 1997). Patients who were possibly depressed and had a lower self-esteem in this study could be having post-psychotic depression but in reality the etiology of depression in schizophrenia is multi-factorial. Being stigmatized, having poor social support, presence of psychotic symptoms and poor quality of life are among the possible associations with either low self-esteem or depression in schizophrenia.
Regardless of losing its’ significant in the multiple linear regression analysis, patient functional assessment using the GAF scale was found to be significantly associated with self-esteem in the linear regression analysis. The impairment in socio-occupational function of patients significantly reduced level of self-esteem. Sorgaard et al. (2002) and Brekke et al (2001) found no relationship between the psychosocial functioning and level of self-esteem in patients with schizophrenia. Brekke et al (2001) emphasized the role of executive functioning as a moderator of the relationship between psychosocial functioning and self-esteem. Patients with impaired executive functioning displayed a positive and statistically significant association between psychosocial functioning and self-esteem but those with intact executive functioning showed negative association.
There are numerous limitations in this study. Even though this study gave the picture that patients generally felt good about themselves, caution should be exercised in interpreting this finding. Factors such as executive functioning and social comparison which played important role as moderators for one’s self-esteem were not assessed. Other limitation is the cross-sectional assessment and absence of comparison group of similarly stigmatized illness.
Majority of subjects in this study are Malays from Kota Bharu, and its nearby sub-urban towns. Results of this study may not be representative of multiracial and multicultural Malaysian population. The convenient sampling method employed might not represent the general view of the target population. Usage of single item questionnaires which assessed patients’ awareness of illness, social support and perceived stigma is subjected to criticism and one would query its reliability. However, they were intended to act as screening tools to see possible association with the self-esteem. This deficiency should offer guidance for improvement in future research.
This study showed that patients with schizophrenia that attended the follow-up clinic generally have average to good self-esteem. The desirable level of self esteem seemed to be associated with marriage, good social support, denial of illness or devaluation by other, satisfaction with life and health and being free from psychotic and depressive symptoms.
Working status or psychosocial functioning was associated with self-esteem only on simple linear regression analysis. Thus, future rehabilitation program should put greater emphasis on patient subjective experience by enhancing rehabilitation modalities that correct feeling of self-stigma and self-worth.
Beck, A.T. (1967). Depression: clinical, experimental and therapeutic aspects. New
York: Harper and Row.
Bech, P., Kastrup, M., Rafaelsen, O.J. (1985). Mini-compendium of rating scales.
Acta Psychiatrica Scandinavia 326, 7-37.
Brekke, J.S., Kohrt, B. & Green, M.F. (2001). Neuropsychological functioning as a
moderator of the relationship between psychosocial functioning and the subjective
experience of self and life in schizophrenia. Schizophrenia Bulletin 27, (4): 697-
Gureje, O., Harvey, C. & Herrman, H. (2004). Self-Esteem in patients who have
recovered from psychosis: profile and relationship to quality of life. Australian
and New Zealand Journal of Psychiatry 38, 334-338.
Hasanah, CI & Razali MS. (2002). Quality of life: an assessment of the state of
psychosocial rehabilitation of patients with schizophrenia in the community. The
Journal of The Royal Society for the Promotion of Health 4, 251-255.
Hasanah, CI, Naing, L & Rahman, ARA. (2002). World Health Organization Quality of Life Assessment: Brief Version in Bahasa Malaysia. Med J Malaysia Vol 58 No 1 March 2003
Kaplan and Sadock (1997). Synopsis of Psychiatry, Behavioural Sciences/Clinical
Psychiatry. 8th edition, William and Wilkins.
Lai,Y.M., Hong C.P., Chee C.Y.,(2001). Stigma of mental illness. Singapore Medical
Journal 42, (3):111-4.
Link, B.G., Struening, E.L., Neese-Todd, S., Asmussen, S., Phelan, J.C., (2001).
The consequences of stigma for the self-esteem of people with mental illness. Psychiatric Service 52: 1621-1626.
Mahmood, N.M. (1999). Penagihan Dadah dan Residivisme. Aspek-aspek Psikology dan
Persekitaran. Pusat Penyelidikan dan Perundingan,University Utara Malaysia.
Ritsher, J.B., Otilingam, P.G., Grajales, M., (2003). Psychiatry Res 121, (1):31-49.
Robson, P.J. (1998). Self-Esteem- A Psychiatric View. British Journal of Psychiatry
Rosenberg, M. (1986). Rosenberg scale of Self-Esteem. The Morris Rosenberg
Foundation, Department of Sociology, University of Maryland.
Silverstone, P.H.& Salsali, M., (2003). Low Self-Esteem and Psychiatric Patients. Annals
of General Hospital Psychiatry 2:2.
Sing-Fai Tam., Yee-Chiu Ip, Cap Siu-Ching Chan. (2000). Self-concepts of hospitalized
and day-care psychiatric patients. The Journal of Social psychology: 140, (2):202.
Sorgaard, K.W., Heikkila, J., Hansson, L.,. (2002). Self-esteem in persons with
schizophrenia. A Nordic multicenter study. Journal of Mental Health 11, 405-415.
Taylor, S. E., Peplau, L. A., & Sears, D. O. (2000). Social psychology .10th Ed. Upper
Saddle River, NJ: Prentice-Hall.
Torrey,W.C., Mueser, K.T., McHugo, G.H., Drake, R.E., (2000) Self-esteem as an outcome measure in studies of vocational rehabilitation for adults with severe mental illness. Psychiatric Services 51, 229-233.
Wright, E.R., Gronfein, W.P., Owens T.J., (2000). Deinstitutionalization, social rejection,
and the self-esteem of former mental patients. J of Health and Social Behavior 41,
Wylie, R. C. (1974). The self-concept. Lincoln: The University of Nebraska Press.
Assoc. Prof. Dr. Hasanah Che Ismail
Department of Psychiatry
School of Medical Sciences,
16150, Kubang Kerian, Kelantan, Malaysia.
First Published October 2008
Copyright Priory Lodge Education Limited 2008