Asian Journal of Psychiatry 6 (2013) 380–388

Relationship of caregiver burden with coping strategies, social support,psychological morbidity, and quality of life in the caregivers ofschizophrenia

Natasha Kate, Sandeep Grover *, Parmanand Kulhara, Ritu Nehra

Post Graduate Institute of Medical Education and Research, Chandigarh, India

A R T I C L E I N F O

Article history:

Received 23 December 2012

Received in revised form 9 March 2013

Accepted 30 March 2013

Keywords:

Caregiver

Burden

Coping

Quality of life

A B S T R A C T

Aim: To evaluate the relationship of caregiver burden as assessed by using Hindi Involvement Evaluation

Questionnaire (IEQ) with coping strategies, social support, psychological morbidity, and quality of life of

caregivers of patients with schizophrenia. Additionally, the relationship of caregiver-burden with

sociodemographic variables, and clinical variables, including severity of psychopathology and level of

functioning of patients, was studied.

Methodology: The study included 100 patients with schizophrenia and their caregivers recruited by

purposive random sampling.

Results: Among the four domains of IEQ, highest number of correlations emerged with tension domain.

Tension domain had positive correlation with the caregiver being single, time spent in caregiving per

day, and use of avoidance, collusion, and coercion as coping strategies. Additionally, tension domain was

associated with poor quality of life in all the domains of WHO-QOL Bref and was associated with higher

psychological morbidity. Worrying urging-I domain of IEQ correlated with frequency of visits, higher use

of problem focused coping and poor physical health as per the WHO-QOL Bref. Worrying urging-II

domain of IEQ had positive correlation with higher level of positive symptoms, lower level of functioning

of the patient, younger age of caregiver, caregiver being unmarried, and higher use problem focused and

seeking social support as coping strategies. Supervision domain of IEQ correlated positively with lower

income, being an unmarried caregiver, from an urban locality and non-nuclear family. Supervision

domain was associated with poor physical health as assessed by WHO-QOL Bref.

Conclusion: Caregiving burden, especially tension is associated with use of maladaptive coping

strategies, poor quality of life and higher level of psychological morbidity in caregivers.

� 2013 Elsevier B.V. All rights reserved.

Contents lists available at SciVerse ScienceDirect

Asian Journal of Psychiatry

jo u rn al h om epag e: ww w.els evier .c o m/lo cat e/a jp

1. Introduction

Schizophrenia is a severe mental disorder, which places aconsiderable burden on the caregivers. Families experience a lot ofphysical, emotional and financial distress due to a patient’sabnormal behavior, and social and occupational dysfunction. Incountries like India, where rehabilitation services are almost non-existent and there is no social security system, the role of thefamily becomes more important.

Many studies from India have evaluated burden perceived bythe caregivers of schizophrenia and these show that schizophreniais a burdensome illness. Studies which have compared the burden

* Corresponding author at: Department of Psychiatry, Postgraduate Institute of

Medical Education & Research, Chandigarh 160012, India. Tel.: +91 172 2756807;

fax: +91 172 2744401/2745078.

E-mail address: drsandeepg2002@yahoo.com (S. Grover).

1876-2018/$ – see front matter � 2013 Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.ajp.2013.03.014

of caregiving of schizophrenia with other psychiatric disorders,suggest that schizophrenia is more burdensome than recurrentdepressive disorder (Chakrabarti et al., 1995), bipolar disorder(Chakrabarti et al., 1995; Chakrabarti and Gill, 2002), obsessivecompulsive disorder (Thomas et al., 2004) and other neuroticdisorders (Chakrabarti and Kulhara, 1999).

However, previous studies from India have mostly used theFamily Burden Interview (FBI) (Pai and Kapur, 1981) or BurdenAssessment Schedule (BAS) (Thara et al., 1998), which areinterviewer based instruments for assessment of burden incaregivers. The primary focus of the FBI is on objective burden,and it is relatively inadequate with regards to assessment of thesubjective consequences of caregiving, which is assessed by onlyone item of the instrument. On the other hand, BAS assesses bothobjective and subjective burden, but still the major emphasis is onthe objective burden.

Over the years, several new burden-assessment instrumentshave been developed. One such instrument, the Involvement

N. Kate et al. / Asian Journal of Psychiatry 6 (2013) 380–388 381

Evaluation Questionnaire (IEQ) was developed to assess theconsequences of caregiving, and is based on the stress-apprai-sal-coping model (Schene et al., 1996, 1998). Being a self reportscale, the IEQ evaluates the ‘consequences of caregiving’ andreflects the caregivers’ perception of burden without distinguish-ing between the objective and subjective aspects. Recently thisinstrument was adapted and evaluated for the psychometricproperties in the Indian setting and has been shown to have goodinternal consistency and reliability (Grover et al., 2011).

Studies suggest that the level of caregiver burden is commonlyassociated with levels of dysfunction or disability, severity ofsymptoms and duration of illness (Raj et al., 1992; Chakrabartiet al., 1995; Kataria et al., 2002; Rammohan et al., 2002a; Saldanhaet al., 2002; Creado et al., 2006; Thomas et al., 2004; Kalra et al.,2009). Findings with regard to association of burden withsociodemographic variables have been inconsistent. The generaltrend is that the caregiver burden is higher when the patient ismale, caregiver is female, caregiver is less educated and in thosefrom a low-socioeconomic background (Roychaudhuri et al., 1995;Rammohan et al., 2002a,b; Kumar and Mohanty, 2007; Nehra et al.,2006; Chakrabarti, 2010). Studies evaluating the relationship ofburden with coping suggest that uses of denial and other emotion-focused strategies are associated with higher burden (Sekharanet al., 2001; Rammohan et al., 2002a; Creado et al., 2006). However,as mentioned, most of the Indian research is based on instrumentswhich mainly assess objective burden, while the subjective burdenis rarely being the major focus.

As there is lack of data on subjective burden as a consequence ofcaregiving on schizophrenia from India, the present study, aimedto (1) Evaluate the relationship of caregiver-burden and copingstrategies, social support, psychological morbidity and quality oflife of caregivers of patients with schizophrenia. (2) Study therelationship of caregiver burden with sociodemographic variablesand clinical variables including severity of psychopathology andlevel of functioning of patients with schizophrenia.

2. Methods

2.1. Setting and study population

The study was conducted in the psychiatry department of amulti-specialty hospital catering to a large section of thepopulation of north-India. The study was approved by the EthicsCommittee of the Institute. All participants were recruited afterobtaining proper written informed consent. A cross-sectionalstudy design was employed. The patients were assessed only onceat the time of intake into the study. The study included 100patients with schizophrenia along with their caregivers. Thesample was selected by purposive random sampling, i.e. to getthe study sample of 100 patients, 200 patients fulfilling theinclusion and exclusion criteria were approached and of these 100were included in the study based on the pre-generatedrandomization number.

2.2. Inclusion and exclusion criteria

Inclusion criteria for the patients were a diagnosis of schizophre-nia as per DSM-IV (American Psychiatric Association, 1994), agebetween 20 and 60 years and with an illness of 2–10 years duration.Patients with comorbid chronic physical, psychiatric, substancedependence (except tobacco dependence) disorders and organicbrain syndromes were excluded. Patients who had a family memberwith a diagnosed chronic physical illness or comorbid psychiatricdisorder staying in the same dwelling unit were also excluded.

For this study, a primary caregiver was defined as a ‘‘personliving with the patient and intimately involved in the care of

the patient for at least one year, i.e., looking after her/his dailyneeds, supervising the medications, bringing the patient to thehospital, staying with the patient during inpatient stay andmaintaining liaison with the hospital staff’’. To be included inthe study, the caregivers were required to be older than 18 yearsof age, involved in the continuous care of patient for at least thelast one year, free from any diagnosed physical or psychiatricdisorder (other than tobacco dependence) and were required tobe able to read Hindi.

2.3. Instruments

Severity of psychopathology of the patients was assessed onPositive and Negative Syndrome Scale for Schizophrenia (PANSS)(Kay et al., 1987).

2.3.1. Hindi Version of Involvement Evaluation Questionnaire (Grover

et al., 2011)

Caregiver burden was assessed by using IEQ. One of the versionsof IEQ, commonly used in Europe in the EPSILON study, known asIEQ-EU, has a core module which assesses ‘‘caregiver-burden’’ andconsists of 31 items, each rated on 5-point Likert scales (vanWijngaarden et al., 2000). In our previous study, we adapted andtranslated the IEQ-EU into Hindi and evaluated its psychometricproperties (Grover et al., 2011). The Hindi-IEQ has been found tohave significant Pearson’s correlation coefficients (p < 0.05) andintra-class correlation coefficients for each item (0.76–1.0) andsubscales, indicating a satisfactory level of agreement between theHindi and English versions. Test–retest reliability for all items ofthe Hindi-IEQ was adequate, and internal consistency (Cronbach’salpha – 0.89) and split-half reliability (Spearman–Brown coeffi-cient – 0.68) of the Hindi-IEQ were also satisfactory. Factor analysisof the scale yielded four subscales covering 29 items. These fourfactors were labeled as tension (10 items), worrying-Urging-I (10items), worrying-Urging-II (6 items) and supervision (3 items)(Grover et al., 2011).

2.3.2. Global Assessment of Functioning Scale (GAF) (Endicott et al.,

1976)

The GAF provides measures of overall functioning relating topsychiatric symptoms. It is a clinician-rated, 100 point scale basedon all available information with clear description of each 10-pointintervals.

2.3.3. Social Support Questionnaire-Hindi Adaptation (Nehra et al.,

1996)

Social support was assessed by using a Hindi adaptation (Nehraet al., 1996) of the Social Support Questionnaire (SSQ) by Pollackand Harris (1983).

2.3.4. The Coping Checklist-Hindi version (Nehra et al., 2002)

This was used to assess the coping strategies used by thecaregivers. It is a modified Hindi version of the coping check list ofScazufca and Kupiers (1999). It was modified by Nehra et al.(2002). It consists of 14 items divided into 5 domains/subscales:problem focused (3 items), seeking social support (4 items),avoidance (5 items), collusion (1 item), coercion (1 item). The itemsare rated on a three point scale (0 – never; 1 – sometimes and 2 isalways used). There is good correlation between most of the itemsin the English and Hindi version. Cronbach’s alpha of the scale is0.62 indicating acceptable level of internal consistency

2.3.5. WHO Quality of Life – Bref Version (Hindi Version) (Saxena

et al., 1998)

This scale was used to assess the Quality of Life (QOL) of thecaregivers. It is the only QOL instrument that has been designed as

N. Kate et al. / Asian Journal of Psychiatry 6 (2013) 380–388382

a multilingual instrument and a self-administered Hindi version ofthe instrument is available. It places emphasis on subjectiveevaluation of respondent’s health and living conditions. The scalehas shown good discriminant validity, concurrent validity, internalconsistency and test–retest reliability (Saxena et al., 1998)

Hindi translated version (Gautam et al., 1987) of General Health

Questionnaire-12 (GHQ-12) (Goldberg, 1972) was used to assesspsychological morbidity in the caregivers. A score of <2 on GHQ-12indicates that the subject is free from any psychiatric illness (Canoet al., 2001)

2.4. Procedure

Patients with a diagnosis of schizophrenia, attending theoutpatient services with a caregiver or admitted to the inpatientunit were approached. They were explained the nature of thestudy and requested to participate in the investigation. Patientswith schizophrenia and their caregivers meeting the inclusionand exclusion criteria were recruited by purposive randomsampling. The patients were assessed on PANSS and GAF. Thecaregivers were requested to complete the Hindi IEQ, CopingChecklist, SSQ, WHO-QOL Bref and GHQ-12. Those caregiversscoring more than two on GHQ were advised to seek a formalpsychiatry consultation.

For this study, 239 patients and their caregivers wereapproached, of which two refused to participate in the study.Seventeen patient and caregiver pairs were excluded becausethe caregivers were unable to read Hindi and/or English.Another 15 patients and caregivers were excluded because ofduration of illness was more than 10 years and five patients hadpsychiatric comorbidity. This resulted in a sample of 200patients and caregivers. Of this, 100 were selected based on therandomization table generated using the SPSS prior to start ofthe study.

2.5. Statistical analysis

Data were analyzed by using SPSS-14 (Statistical Package forthe Social Sciences, 2005, Chicago, IL, USA). Mean and standard

Table 1Sociodemographic profile of the patient and caregivers.

Socio-demographic variables

Sex – male

Marital status – currently single

Occupation – paid Employment

Religion – Hindu

Family type – nuclear

Locality – urban

Age in years

Education (in number of years)

Family income (in rupees/month)

Distance of place of residence from the hospital in kilometers

Relationship with the patients

Parents (mother/father)

Spouses (husband/wife)

Siblings/others

Caregivers supervising the medications of the patient

Hours spent in caregiving/day

Duration since the time the family member has taken up the caregiver’s role (in yea

% of visits in last 3 months when the caregivers accompanied the patient to the hosp

deviation with range were calculated for continuous variablesincluding sociodemographic, clinical, PANSS scores, SSQ, CopingCheck List, Hindi IEQ, WHO-QOL Bref and GHQ-12. Frequency andpercentages were determined for discontinuous sociodemographicand clinical variables. Association between the various domains ofHindi IEQ and continuous variables was assessed by using Pearson’sproduct moment correlation analysis. Spearman’s rank correlationwas used to assess the association between the various domains ofHindi IEQ and discontinuous variables. Comparisons were done byusing chi-square test, t-test and Mann–Whitney U test.

3. Results

3.1. Socio-demographic profile

The sociodemographic profile of the patients and caregiversis displayed in Table 1. About half of the caregivers were parents(51%), while 22% were spouses and 20% were siblings. Caregivershad been performing the role of a primary caregiver for the last5.2 years (range 1–10 years) and on an average spent about 3 heach day in providing care to the patient. Caregivers accompa-nied their patients during the 93% of follow up visits in theprevious three months and 73% were supervising the patient’smedications.

3.2. Clinical profile of the patient group

The clinical details of the study group are provided in Table 2.

3.3. Scores on IEQ, SSQ, coping checklist and WHO-QOL Bref

Details of the scores on various domains of IEQ, total IEQ score,SSQ, coping checklist and different WHO-QOL Bref domains andtotal WHO-QOL Bref are shown in Table 3.

3.4. Coping strategies used by caregivers

Table 3 shows the mean scores obtained for various types ofcoping strategies used by the caregivers. Additionally, depending

Patient group [N = 100] Caregivers [N = 100]

N N

56 65

57 10

20 73

85

49

53

Mean (SD) Mean (SD)

30.8 (9.7) [range18-58] 45.9 (11.6) [range 24–71]

13.2 (4.3) 14.7(3.0)

[range 6–22] [range 6–22]

15,840 (11,420) [range 3000–70,000]

120.5 (99.97)

N

23/29

17/5

20/6

73

Mean (SD)

3.1 (2.2)

[range 1–12]

rs) 5.2 (3.0)

[range 1–10]

ital 93

Table 2Clinical Profile of the patient group.

Clinical variable Mean (S.D.)/N

Schizophrenia subtype

Paranoid 50

Non-paranoida 50

Age of onset (in years) 25.1 (8.8) [range 13–53]

Duration of illness (in years) 5.7 (3.0) [range 2–10]

Number of visits in past three month 1.9 (1.7) [range 1–9]

Duration of treatment (in years) 4.50 (3.1) [range 1–10]

Duration of current treatment (in years) 1.47 (1.0) [range 0.1–6]

Approximate cost of medication for a

month (in rupees) to the patient/family

619.50 (496.8) [range 0–3000]

Severity of psychopathology

Positive subscale total 13.0 (5.0)

Negative subscale total 14.4 (5.2)

General psychopathology subscale total 26.5 (6.0)

Positive and negative syndrome scale

for schizophrenia TOTAL

54.0 (12.7)

Global Assessment of Functioning

Scale score

51.73 (14.61) [range 15–80]

a In the non-paranoid group 41 patients were diagnosed with undifferentiated

schizophrenia, six with catatonic schizophrenia, two with heberphrenic schizo-

phrenia and one with residual schizophrenia

Table 4Coping strategies used by caregivers.

Variables Ever

used

Never

used

Problem-focused

1. Thought about and tried to understand the problem 91 9

2. Made plans and took practical steps to

deal with the problem

79 21

3. Worked out problems calmly with family/friends 79 21

Seeking social support

4. Consulted doctors about the problem 88 12

5. Talked with friends/relatives, asked for practical help 74 26

6. Talked with someone about how I felt,

sought reassurance

63 37

7. Talked with people facing similar problems 67 33

Avoidance

8. Tried to take my mind off things by

smoking/drinking/taking pills

24 76

9. Got angry at others, broke or threw things about 32 68

10. Hoped for miracles/prayed for good times 51 49

11. Reduced interaction with people 46 54

12. Tried not to think about the problem 54 46

Collusion

13. Agreed/gave in/did not force when patient

did things I disliked

79 21

Coercion

14. Got angry/shouted/used force when patient

did things I disliked

63 37

N. Kate et al. / Asian Journal of Psychiatry 6 (2013) 380–388 383

on the use of coping skills, caregivers were categorized into twogroups, i.e., those who remembered to use the particularcoping strategy (either sometimes or always) and those neverused the particular coping strategy. As shown in Table 4, thefrequency of ‘ever use’ was high for problem focusedstrategies and overall avoidance strategies were least commonlyused.

3.5. Psychological morbidity in caregivers

Psychological morbidity in caregivers was assessed using GHQ-12. A score of I2 was taken as an indicator of the presence ofpsychological morbidity. Out of the 100 caregivers, about one-third(36%) had I2 score and the mean GHQ-12 score for the caregiverswith score of I2 was 3.72 � 2.55 (range 2–10), indicating a fair levelof morbidity.

Table 3Scores on IEQ, Coping checklist, SSQ and WHO-QOL Bref.

Variables Mean (SD) Range

Scores on IEQ

Tension 7.13 (6.7) 0–28

Worrying-urging-I 25.76 (9.0) 4–40

Worrying-urging-II 9.7 (5.4) 0–21

Supervision 2.67 (3.55) 0–12

Total IEQ score 45.73 (20.43) 6–101

Total SSQ score 45.0 (6.9) 24–60

Coping checklist

Problem focused 3.75 (1.7) 0–6

Seek social support 4.34 (2.0) 0–8

Avoidance 2.7 (2.2) 0–10

Collusion 1.2 (0.7) 0–2

Coercion 0.75 (0.6) 0–2

Coping check list total 12.7 (4.9) 0–27

WHO-QOL Bref

WHOQOL General health Score 7.1(1.54) 2–10

WHOQOL Physical health Score 24.69 (4.90) 9–35

WHOQOL Psychological Score 20.75 (4.68) 7–39

WHOQOL Social relationships Score 10.39 (2.71) 3–15

WHOQOL Environment Score 26.85 (5.98) 9–40

Total WHOQOL Score 89.99 (17.04) 36–129

SSQ: social support questionnaire; IEQ: involvement evaluation questionnaire.

3.6. Association of IEQ with other variables

3.6.1. Association of sociodemographic and caregiving variables with

IEQ

There were very few sociodemographic correlates of IEQ.Significant correlations are shown in Table 5. Those caregivers whospent more time with the patient experienced more tension. Single(unmarried) caregivers experienced more tension. Youngercaregivers and those caregivers who were single had significantlyhigher score on the worrying urging II domain of IEQ. Significantlyhigher score in the supervision domain of IEQ was associated withlower income, being a single (unmarried) caregiver, from an urbanbackground and from a non-nuclear family. Total IEQ score hadsignificant correlation with being a single (unmarried) caregiver

3.6.2. Association of coping and quality of life of caregivers with IEQ

Tension domain of IEQ had significant positive correlation withcoping strategies of avoidance, collusion, coercion and total copingchecklist score. Tension domain of IEQ also had significant negativecorrelation with all the domains of WHO-QOL Bref and total WHO-QOL Bref score. Total GHQ score also had significant positivecorrelation with tension domain of IEQ. Worrying urging-I and IIdomains had significant positive correlation with problem focusedcoping strategies. Additionally worrying urging-II domain hadsignificant positive correlation with seeking social support as acoping mechanism and total coping checklist score. Total IEQ scorehad positive correlation with problem focused, seeking socialsupport, avoidance and total coping checklist score. Physical healthdomain of WHO-QOL Bref had positive correlation with worryingurging-I domain and negative correlation with supervision domainof IEQ

The relationship of burden experienced by the caregivers andcoping was further evaluated by comparing the level of burdenexperienced by those who used the particular coping (‘ever’ usedgroup) and those who ‘never’ used the specific coping skill. Thedata were compared using‘t’ test or Mann Whitney ‘U’ testdepending on the distribution of scores. When the standarddeviation exceeded the mean scores, Mann–Whitney test was

Table 5Correlates of consequences of caregiving.

Tension Worrying-urging-I Worrying-urging-II Supervision Total IEQ score

Income �0.241 (0.016)#

Hours spent per day 0.254 (0.011)

Age of caregivers �0.247 (0.013)

Single (unmarried) caregiver 0.212 (0.034)# 0.245 (0.014)# 0.217 (0.030)# 0.260 (0.009)

Locality – urban 0.216 (0.031)#

Family type – non nuclear 0.220 (0.028)#

No of visits in last 3 months 0.219 (0.029)

PANSS positive score 0.206 (0.04)

GAF score �0.192 (0.05)

Coping

Problem focused 0.282 (0.005) 0.228 (0.022) 0.283 (0.004)

Seek social support 0.220 (0.028) 0.228 (0.023)

Avoidance 0.333 (0.001) 0.208 (0.038)

Collusion 0.298 (0.003)

Coercion 0.327 (0.001)

Coping check list total 0.369 (<0.001) 0.290 (0.003) 0.331 (0.001)

WHO-QOL Bref

WHOQOL general health score �0.251 (0.012)

WHOQOL physical health score �0.222 (0.027) 0.234 (0.019) �0.231 (0.021)#

WHOQOL psychological score �0.299 (0.002)

WHOQOL social relationships score �0.283 (0.004)

WHOQOL environment score �0.296 (0.003)

Total WHOQOL score �0.343 (<0.001)

Total GHQ score 0.270 (0.007)

PANSS: positive and negative syndrome scale for schizophrenia; GHQ: general health questionnaire; WHOQOL-Bref: who quality of life – bref version# Spearman Rank correlation values.

Table 6Comparison of IEQ scores between GHQ positive and GHQ negative caregivers.

Variables GHQ positive

N = 36

GHQ negative

N = 64

t-Test value

Tension 10.47 (7.5) 5.25 (5.39) 684 (0.001)#

Worrying-urging-I 26.11 (8.56) 25.56 (9.4) 0.289

Worrying-urging-II 11.55 (5.26) 8.67 (5.22) 2.642 (0.01)

Supervision 3.75 (3.70) 2.06 (3.34) 795 (0.006)#

Total IEQ score 54.55 (23.04) 40.76 (17.07) 3.40 (0.001)

Parents

N = 52

Spouses

N = 22

Tension 5.96 (6.1) 8.86 (7.35) 422.5 (0.07)#

Worrying-urging-I 23.57 (9.96) 28.95 (7.44) 2.27 (0.026)

Worrying-urging-II 8.38 (5.25) 11.36 (5.54) 2.19 (0.032)

Supervision 1.57 (2.93) 3.68 (4.22) 381.5 (0.012)#

Total IEQ score 39.88 (20.5) 53.04 (21.33) 2.49 (0.015)

GHQ: general health questionnaire; IEQ: involvement evaluation questionnaire.# Mann Whitney value.

N. Kate et al. / Asian Journal of Psychiatry 6 (2013) 380–388384

used. Few interesting differences emerged. Caregivers who usedthe problem solving coping skill of ‘‘Thought about and tried to

understand the problem’’ were found to perceive more burden inthe domains of tension (mean score 6.7 � 0.7 vs 1.2 � 1.4; Mann–Whitney U value 151.5; p < 0.001), worrying-urging II (mean score10.1 � 5.4 vs. 5.4 � 2.8; t value – 2.55; p = 0.012) and total IEQscores (mean score 47.1 � 20.4 vs. 31.3 � 15.5; t value – 2.26;p < 0.026) compared to those who never used this strategy. Thosewho used the avoidance coping strategy of ‘‘Tried to take my mind

off things by smoking/drinking/taking pills’’, had significantly higherscore in the IEQ domains of tension (mean value 10.9 � 6.5 vs.5.9 � 6.3; Mann Whitney I value – 511; p = 0.001), supervision(3.8 � 4.0 vs. 2.3 � 3.3; Mann Whitney U value – 674; p = 0.04) andtotal IEQ scores (53.0 � 20.3 vs. 43.4 � 20.1; p = 0.04) than thosewho never used this coping skill. Significantly higher scores werein the tension domain in those who ever used the avoidance copingskills of ‘‘Reduced interaction with people’’ (9.4 � 7.3 vs. 5.2 � 5.5;Mann Whitney U Value – 836.5; p = 0.005) and ‘‘Tried not to think

about the problem’’ (9.1 � 7.1 vs. 4.8 � 5.3; Mann Whitney U value– 781; p = 0.001). Caregivers who used collusion as coping in theform of ‘‘Agreed/gave in/did not force when patient did things I

disliked’’ had significantly higher score in the IEQ domain of tension(8.3 � 6.8 vs. 2.6 � 3.7; Mann Whitney U value – 392.5; p < 0.001)and worrying-urging-II (10.4 � 5.3 vs. 7.1 � 4.8; p < 0.013) thanthose who never used collusion. Among those who ever usedcoercion in the form of ‘‘Got angry/shouted/used force when patient

did things I disliked’’ to deal with situations, had significantly higherscore in the domain of tension than those who never used this formof coping (8.6 � 6.8 vs. 4.6 � 5.7; Mann Whitney U value – 756.5;p = 0.003).

3.6.3. Association of relationship of caregivers with IEQ

When the association of various domains of IEQ and relation-ship of caregiver with patient was assessed it was seen thatparents had significantly lower scores on all the domains (excepttension domain) and total score of IEQ compared to spouses (seeTables 5 and 6)

3.7. Comparison of IEQ scores between GHQ positive and GHQ

negative caregivers

Comparison of IEQ scores between GHQ positive and GHQnegative caregivers revealed significantly higher scores for theGHQ positive caregivers in all the domains except the worryingurging-I domain. Total IEQ score was also significantly higher forGHQ positive caregivers.

3.8. Regression analysis evaluating the relationship of IEQ with other

variables

A stepwise regression analysis was carried out using total IEQas the dependent variable and the total WHO-QOL Bref score,total social support score, total coping scale score and GHQpositive versus GHQ negative as independent variables. Onlytotal coping checklist score and GHQ status emerged as thesignificant predictors of IEQ. It was seen that total copingchecklist score explained 10.1% (F value = 12.07; p = 0.001) of the

N. Kate et al. / Asian Journal of Psychiatry 6 (2013) 380–388 385

variance of IEQ. Taken together total coping checklist score andGHQ status explained 16.5% (F value = 10.78; p < 0.001) of thevariance of IEQ.

4. Discussion

4.1. IEQ as an instrument to assess burden

Many studies from India have evaluated burden on thecaregivers of patients with schizophrenia or other disorders. Mostof these studies have used the FBI and others have relied uponinstruments like the BAS. Both these instruments primarily assessobjective burden. However, in recent times, burden is understoodin the stress-coping appraisal model, which is one of the dominantparadigms of conceptualizing caregiver-burden (Whalen andBuchholz, 2009; van Wijngaarden et al., 2000, 2003). The IEQwas developed in Europe to assess caregiver-burden or the‘‘consequences of caregiving’’ in the stress-coping appraisal model.It evaluates the caregiver’s perceptions of burden. The IEQ has beenextensively tested in studies from the West (van Wijngaardenet al., 2000, 2003) and other developing countries (Tang et al.,2008; Sefasi et al., 2008; Zahid and Ohaeri, 2010) and has proved tobe a comprehensive, conceptually valid and reliable instrument todetermine the consequences of providing care for a mentally illfamily member (van Wijngaarden et al., 2000, 2003).

In this background, the present study aimed to study the‘caregiving burden’ as assessed by using IEQ on the primarycaregivers of patients with schizophrenia. Additionally, the studyexplored the relationship of caregiver-burden with sociodemo-graphic variables of patients and caregivers, clinical variables andcoping strategies and QOL of caregivers.

4.2. Sociodemographic and burden profile

The mean age of the patients was about 31 years. More than halfof the patients were males; the majority of them were unemployedand Hindu by religion. There was nearly equal distribution ofsubjects from urban and rural background and those belonging tonuclear and non-nuclear families. Previous studies from Indiawhich have evaluated burden as perceived by caregivers havestudied patients with a higher mean age (more than 35 years)(Rammohan et al., 2002a; Srivastava, 2005; Nehra et al., 2005).However, the mean age of patients in our sample is similar to arecent study from the same region (Aggarwal et al., 2011) wherethe mean age was less than 35 years. Similar to our results, otherstudies from India also suggest predominance of male patientswith a significant percentage of the patients being unmarried(Rammohan et al., 2002a; Aggarwal et al., 2011) and unemployed(Srivastava, 2005; Aggarwal et al., 2011; Nehra et al., 2005). One ofthe limitations of the previous studies is overinclusion of patientsfrom an urban background and lower socioeconomic status(Srivastava, 2005; Aggarwal et al., 2011; Nehra et al., 2005). Incontrast, in the present study, there was nearly equal representa-tion of patients from urban and rural backgrounds. However, themean family income of the patients was rupees 15,840, indicatinghigher representation of patients from middle and highersocioeconomic status.

The mean age of the caregivers was about 46 years and abouttwo-third of them were males. A majority of the caregivers weremarried and employed and the mean duration of education was14.7 years. About half of the caregivers were parents (51%), while22% were spouses and 20% were siblings. This profile is similar tothe characteristic profile of caregivers who have participated in theprevious studies from India which have evaluated the caregiverburden (Aggarwal et al., 2011; Nehra et al., 2005; Jayakumar et al.,2002).

4.3. Clinical profile

In terms of clinical variables, paranoid schizophrenia was themost common subtype, which is commensurate with the globalliterature (Kirkpatrick and Tek, 2004) and previous studies fromIndia (Aggarwal et al., 2011; Grover et al., 2005; Arora et al., 1997).The mean age of onset was 25.12 years which is in the range oftypical age of onset reported in standard textbooks and otherstudies (Aggarwal et al., 2011; Grover et al., 2005). The meanduration of illness (5.67 years) was similar to the previous studies(Aggarwal et al., 2011; Grover et al., 2005; Arora et al., 1997), whichmeant that the chronic nature of the illness was not totallycompromised, despite restriction of the duration of illness to 2–10years. The PANSS score of the patients suggested moderate severityof psychopathology and akin to that reported in many of studiesevaluating burden in patients with schizophrenia.

4.4. Psychological morbidity in caregivers

As evaluated by GHQ, findings of the present study suggest thatat least one-third (36%) of the caregivers had psychologicalmorbidity, which is the range (30–40%) reported in previousstudies from West (Tennakoon et al., 2000; Addington et al., 2005;Harvey et al., 2002) and India (Aggarwal et al., 2011).

4.5. Burden experienced by the caregivers

When one tries to understand the IEQ profile in the light ofstudies done in other countries using the same instrument,certain similarities and differences are evident. In the presentstudy, the rank order of various domains of IEQ was worryingurging I > worrying urging II > tension > supervision. However,studies from the West suggest slightly different rank order (vanWijngaarden et al., 2000, 2003). Further the total and scores onthe various domains of IEQ in the present study weresignificantly lower than the pooled estimates reported in theEPSILON study (van Wijngaarden et al., 2000), except in the caseof ‘‘urging’’ subscale scores. However, these differences shouldnot be interpreted as difference in caregiver-burden in the Westand East, because studies suggest wide variations in scoresacross different sites in the EPSILON study, as well as betweenthe few Asian studies of the IEQ. There is only one previousstudy from India which used IEQ in patients with severemental disorders (Grover et al., 2011). The rank order of scoresobtained in the various domains (Worrying urging I > worryingurging II > tension > supervision) in current study is similarto that seen in the previous study and the scores arecomparable.

4.6. Correlates of burden

In the present study, there were very few sociodemographiccorrelates of caregiver burden. Previous studies from India havealso not found many consistent sociodemographic correlates ofburden in patients with schizophrenia (Tennakoon et al., 2000). Inthe current study, distance of place of residence from the hospitalcorrelated positively with worrying urging-I domain. This isunderstandable considering the fact that traveling for longdistances for medical care can be very stressful and hencecontributes to worry. The tension domain correlated positivelywith time spent in caregiving in hours. Similar findings have beenpreviously reported in studies from the West which found thatfamily members living with the patient or spending a lot of timewith their ill relative perceive more burden (Schene et al., 1998;Roick et al., 2006; Parabiaghi et al., 2007). Previous studies fromIndia have not focused on this variable.

N. Kate et al. / Asian Journal of Psychiatry 6 (2013) 380–388386

Previous studies from the West suggest that parents perceivemore burden than spouses (Caqueo-urizar et al., 2009). However,findings of the present study are contrary to the findings reportedfrom the West and in accordance with other Indian studies(Rammohan et al., 2002a; Jayakumar et al., 2002).

The effects of caregivers’ ages, marital status, education andemployment on burden are highly variable. Findings of the presentstudy suggest that in general single caregivers (i.e., currentlyunmarried) and those from urban backgrounds, belonging to non-nuclear families and those having to visit the hospital morefrequently perceive more caregiver burden. Some of these findings,for example, higher caregiver burden by the single caregivers (i.e.,currently unmarried) can be understood from the perspective thatmost of these would be either widowed parents or unmarriedsiblings and hence having lower social support and therebyperceive more burden. Urbanized lifestyle brings in its ownstressors and urbanization has been shown to lead to stress (Paykelet al., 2003; Verheij, 2006). Also, other Indian studies have showngreater caregiver burden in caregivers from urban backgrounds(Ranga Rao, 1988; Srivastava, 2005). Frequency of hospital visitsalso increases caregiver burden as the average distance of thepatient’s home from the hospital was quite large (mean 120.5 km)and hence transportation and supervision of a moderately illpatient over such a large distance would not be an easy affair.

Earlier studies from India have reported higher burden withincrease in duration of illness, however findings of the presentstudy do not support this. However, while interpreting thisassociation it is important to remember that most of the previousstudies from India focused only on the objective burden, whereasthe lack of associations seen in the present study indicate lack ofrelationship between duration of illness and caregiver-burden,which is more akin to the subjective burden. It is possible that withincrease in duration of illness and duration of treatment, the illnessstabilizes and the caregivers also develop adequate copingmechanisms to handle the stress of illness.

In the present study, there was negative correlation between apatient’s GAF score and worrying-urging-II domain of IEQ andpositive correlation between PANSS positive symptom score andworrying-urging-II domain of IEQ, indicating that better func-tioning of the patient was associated with lesser burden on thecaregivers and vice versa. This association is also supported byprevious studies which indicate that with increase in severity ofpsychopathology, perceived burden increases (Chakrabarti et al.,1995; Roick et al., 2006; Pai and Kapur, 1982; Rammohan et al.,2002b; Creado et al., 2006; Kalra et al., 2009). However, previousstudies which likewise have evaluated the relationship of burdenwith specific symptom clusters of schizophrenia have beeninconclusive. Some studies suggest that caregivers experiencemore burden in dealing with positive symptoms (Raj et al., 1991),while others have reported negative symptoms (Raj et al., 1991;Chakrabarti et al., 1995) to be more burdensome. Our findingssupport the relationship of burden with the positive symptoms ofschizophrenia.

We also found that higher scores in the tension domain wereassociated with poor QOL of caregivers in all the domains. Previousstudies also suggest that emotional burden on caregivers as aconsequence of their role is associated with a significant impact onthe QOL of caregivers. Poor QOL in the physical health domain wasalso associated with less urging – worrying (urging-worrying-Idomain) but with higher level of supervision. These findingssuggest that overall tension as a caregiving consequence impairsthe QOL of the caregivers. Further, the relationship between poorphysical health and less urging– w orrying (urging-worrying-Idomain) may be reciprocal and a causative role cannot beconcluded. It can be hypothesized that individuals with poorphysical health possibly do not have the physical ability to invest

time in urging an ill relative. Also, possibly these individuals wouldneed to focus their resources on their own care and hence may notworry so much about the patient. With regard to higher perceivedneed for supervision and poor QOL in the physical health domain, itcan be hypothesized that the caregivers tend to supervise thepatients more often so as to avoid crisis situations which may betaxing to their impaired physical health. The other way around therelationship can be understood as a higher level of supervisionleads to neglect of personal health and resultant poor QOL in thisdomain.

Presently, the stress-appraisal-coping paradigm is most fre-quently used to understand the evaluation of an individual towarda stressful experience. As per this paradigm, the experience of astressful situation is modified by an individual’s coping skills.There is more than adequate literature available that shows thatcaregiving for patients with schizophrenia is stressful (Lee et al.,2006; Magana et al., 2007; Laidlaw et al., 2002). In concordancewith the stress-appraisal-coping paradigm, our results show thatthe use of certain coping skills influences the burden experiencedby the caregiver. When correlation were studied betweencaregiving burden and various coping strategies, it was foundthat use of problem focused strategies and seeking social supportare associated with burden in the worrying-urging domains,whereas other coping strategies like avoidance, collusion andcoercion are associated with tension domain. This pattern ofassociation between caregiver-burden and differential use ofadaptive (i.e., problem focused and seeking social support) ormaladaptive (avoidance, coercion) is understandable. It is gener-ally suggested that problem solving strategies are used to modifydifficult situations and are associated with behaviors like seekinginformation and positive methods of communications; many ofsuch components are covered by worrying-urging domains. It isdifficult to compare these findings with previous studies due todifferences in the type of instruments used for assessment ofburden and coping. Some of the studies suggest that use ofavoidance coping has positive correlation with burden (Chaddaet al., 2007), others have suggested a relationship between burdenand ‘social interests’ and ‘resignation’ as coping strategies(Hanzawa et al., 2010). Earlier studies from India also suggestthat use of maladaptive, emotion-focused strategies is frequentlyassociated with higher burden (Chakrabarti and Kulhara, 1999).

When we studied the relationship of each coping strategy usedby the caregivers, it was seen that the caregivers who used coercionto deal with problems (‘‘Got angry/shouted/used force when patientdid things I disliked’’) were found to have more tension as comparedto those who did not. Tension seemed to increase in the caregiver, asdid the amount of urging and worrying by the caregiver whencollusion as a coping skill was used (‘‘Agreed/gave in/did not forcewhen patient did things I disliked’’). Avoidance coping too had asignificant influence on burden, the caregiver who used avoidanceexperienced more tension and had to put more effort intosupervision. Further certain avoidance strategies were associatedwith overall higher total IEQ score. This is in accordance withIndian literature wherein use of avoidance, resignation and coercioncoping have been reported to lead to higher perceived burden(Sekharan et al., 2001; Chandrasekaran et al., 2002; Chadda et al.,2007). As these are believed to be maladaptive coping skills, it ispossible that these do not adequately mitigate the stress ofcaregiving. These relationships also possibly reflect over-involve-ment of family members in the care of their sick relatives, which isquite frequently seen in Indian culture. Among the problem solvingcoping skills, caregiver burden was more when caregivers used theskill involving ‘‘Thought about and tried to understand the problem’’.This is not in accordance with the literature that shows that use ofproblem solving coping leads to decreased burden (Chakrabarti,2010). It is possible that problem solving skill are dependent on the

N. Kate et al. / Asian Journal of Psychiatry 6 (2013) 380–388 387

caregivers knowledge, information source, provided psychoeduca-tion, and cognitive prowess and hence, if these were deficient orinappropriate, this coping strategy would not be able to adequatelytone down the stress and burden experienced by the caregiver.

Findings of the present study suggested higher psychologicalmorbidity as assessed by GHQ scores and tension domain of IEQ.This relationship is understandable considering the fact thathigher level of distress is associated with higher level ofpsychological morbidity.

5. Conclusion

To conclude, findings of the present study suggest that caregiversof schizophrenia experience considerable level of burden. Thepattern of caregiver burden is slightly different from that reported instudies from the West. The lower total IEQ scores seen in studiesfrom India compared to that reported in studies from the Westsuggests that there are cultural differences in as to how familymembers react to the mental illness of their relatives. Due to therelatively close knit family structure in India, family membersconsider caregiving as their moral obligation and as a part of theirfamily life. Due to this, they report lesser burden, despite beingintimately involved in the care of their relatives. About one third ofcaregivers experience psychological morbidity as a result of being inthe caregiver’s role and those with psychological morbidity reporthigher level of caregiver-burden. The caregiver burden is not relatedto sociodemographic variables of patients and the clinical profile to alarge extent, except for few correlations between numbers of visitsin last three months and time spent per day in taking care of the illrelatives. Among the various dimensions of psychopathology, onlythe severity of positive symptoms showed some association withworrying urging-II domain. Similarly lower level of functioning ofpatients as assessed by GAF was associated with worrying urging-IIdomain. However many correlations were seen between tensiondomain of IEQ and different coping strategies and various domains ofQOL of caregivers.

5.1. Limitations

Certain limitations of the present investigation must be kept inmind while interpreting our results. The sample comprisedpredominantly of outpatients attending a General HospitalPsychiatry Unit. Hence, the findings cannot be generalized tocommunity samples or inpatients. The predominance of malecaregivers was also unusual, and could be an artifact of treatment-seeking. The Indian literature in this regard is somewhatinconsistent; though women predominate as caregivers in mostof the studies, there is a preponderance of male caregivers in someof the others (Chakrabarti, 2010). Future studies must attempt toovercome these limitations. Clearly more studies are needed in thisarea with larger samples especially with a need to incorporate alongitudinal study design.

Funding

None.

Conflict of interest

None.

Acknowledgement

We are thankful to the patients and their caregivers for theirparticipation.

References

Addington, J., McCleery, A., Addington, D., 2005. Three-year outcome of family workin an early psychosis program. Schizophrenia Research 79, 107–116.

Aggarwal, M., Avasthi, A., Kumar, S., et al., 2011. Experience of caregiving inschizophrenia: study from India. International Journal of Social Psychiatry57, 224–236.

American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Men-tal Disorders ed. 4, Text Revision (DSM-IV) American Psychiatric Association,Washington, DC.

Arora, A., Avasthi, A., Kulhara, P., 1997. Subsyndromes of chronic schizophrenia: aphenomenological study. Acta Psychiatrica Scandinavica 96, 225–229.

Cano, A., Sprafkin, R.P., Scaturo, D.J., et al., 2001. Mental health screening care: acomparison of 3 brief measures of psychological distress. Primary Care Com-panion of the Journal of Clinical Psychiatry 3, 206–210.

Caqueo-urizar, A., Gutiereg-Meldonado, J., Miraid-Castillo, C., 2009. Quality of life incaregivers of patients with schizophrenia—a literature review. Health andQuality of Life Outcomes 11, 84.

Chadda, R.K., Singh, T.B., Ganguly, K.K., 2007. Caregiver burden and coping. Aprospective study of relationship between burden and coping in caregiversof patients with schizophrenia and bipolar affective disorder. Social Psychiatryand Psychiatric Epidemiology 42, 923–930.

Chakrabarti, S., 2010. Impact of schizophrenia on caregivers: the Indian perspective.In: Kulhara, P., Avasthi, A., Grover, S. (Eds.), Schizophrenia: The Indian scene.2nd ed. PsyPROM, PGIMER, Chandika Press, Barwala, Haryana, pp. 215–262.

Chakrabarti, S., Gill, S., 2002. Coping and its correlates among caregivers of patientswith bipolar disorder: a preliminary study. Bipolar Disorders 4, 50–60.

Chakrabarti, S., Kulhara, P., 1999. Family burden of caring for people with mentalillness. British Journal of Psychiatry 174, 463.

Chakrabarti, S., Raj, L., Kulhara, P., Avasthi, A., Verma, S.K., 1995. A comparison of theextent and pattern of family burden in affective disorders and schizophrenia.Indian Journal of Psychiatry 37, 105–112.

Chandrasekaran, R., Sivaprakash, B., Jayestri, S.R., 2002. Coping strategies of therelatives of schizophrenia patients. Indian Journal of Psychiatry 44, 9–13.

Creado, D.N., Parkar, S.R., Kamath, R.M., 2006. A comparison of the level of func-tioning in chronic schizophrenia with coping and burden in caregivers. IndianJournal of Psychiatry 48, 27–33.

Endicott, J., Spitzer, R.L., Fleiss, J.L., et al., 1976. The Global Assessment scale. Aprocedure for measuring overall severity of psychiatric disturbance. Archives ofGeneral Psychiatry 33, 766–771.

Gautam, S., Nijhawan, M., Kamal, P., 1987. Standardization of Hindi version of Gold-berg’s General Health Questionnaire. Indian Journal of Psychiatry 29, 63–66.

Goldberg, D.P., 1972. The Detection of Psychiatric Illness by Questionnaire. Mauds-ley Monograph No. 21. Oxford University Press, Oxford.

Grover, S., Avasthi, A., Chakrabarti, S., et al., 2005. Cost of care of schizophrenia: astudy of Indian out-patient attenders. Acta Psychiatrica Scandinavica 112, 54–63.

Grover, S., Chakrabarti, S., Ghormode, D., et al., 2011. An Indian adaptation of theInvolvement Evaluation Questionnaire: similarities and differences in assess-ment of caregiver burden. East Asian Arch Psychiatry 21, 142–151.

Hanzawa, S., Bae, J.K., Tanaka, H., et al., 2010. Caregiver burden and copingstrategies for patients with schizophrenia: comparison between Japan andKorea. Psychiatry and Clinical Neurosciences 64, 377–386.

Harvey, K., Burns, T., Fiander, M., et al., 2002. The effect of intensive case manage-ment on the relatives of patients with severe mental illness. Psychiatric Services53, 1580–1585.

Jayakumar, C., Jagadheesan, K., Verma, A.N., 2002. Caregiver’s burden: a comparisonbetween obsessive compulsive disorder and schizophrenia. Indian Journal ofPsychiatry 44, 337–342.

Kalra, H., Nischal, A., Trivedi, J.K., 2009. Extent and determinants of burden of care inIndian families: a comparison between obsessive compulsive disorder andschizophrenia. International Journal of Social Psychiatry 55, 28–38.

Kataria, D., Bhargava, S.C., Vohra, A.K., 2002. A correlational study between psychi-atric disability and family burden in schizophrenia. Indian Journal of Psychiatry44, 54.

Kay, S.R., Fiszbein, A., Opler, A., 1987. The Positive and Negative Syndrome Scale(PANSS) for schizophrenia. Schizophrenia Bulletin 13, 261–276.

Kirkpatrick, B., Tek, C., 2004. Schizophrenia: clinical feature and psychopathologyconcepts. In: Sadock, B.J., Sadock, V.A. (Eds.), Comprehensive Textbook ofPsychiatry. 8th ed. Williams and Wilkins, Philadelphia.

Kumar, S., Mohanty, S., 2007. Spousal burden of care in schizophrenia. Journal of theIndian Academy of Applied Psychology 33, 189–194.

Laidlaw, T.M., Coverdale, J.H., Falloon, I.R.H., et al., 2002. Caregivers’ stresses whenliving together or apart from patients with chronic schizophrenia. CommunityMental Health Journal 38, 303–310.

Lee, T.C., Yang, Y.K., Chen, P.S., et al., 2006. Different dimensions of social support forthe caregivers of patients with schizophrenia: main effect and stress-bufferingmodels. Psychiatry and Clinical Neurosciences 60, 546–550.

Magana, S.M., Ramırez Garcıa, J.I., Hernandez, M.G., et al., 2007. Psychologicaldistress among Latino family caregivers of adults with schizophrenia: the rolesof burden and stigma. Psychiatric Services 58, 378–384.

Nehra, R., Kulhara, P., Verma, S.K., 1996. Adaptation of Social Support Questionnairein Hindi: Indian setting. Indian Journal of Clinical Psychology 23, 33–39.

Nehra, R., Chakrabarti, S., Sharma, R., Kaur, R., 2002. Psychometric properties of theHindi version of the coping checklist of Scazufca and Kuipers. Indian Journal ofClinical Psychology 29, 79–84.

N. Kate et al. / Asian Journal of Psychiatry 6 (2013) 380–388388

Nehra, R., Chakrabarti, S., Kulhara, P., Sharma, R., 2005. Caregiver-coping in bipolardisorder and schizophrenia: a re-examination. Social Psychiatry and PsychiatricEpidemiology 40, 329–336.

Nehra, R., Chakrabarti, S., Kulhara, P., Sharma, R., 2006. Family burden and itscorrelates among caregivers of schizophrenia and bipolar affective disorder.Journal of Mental Health and Human Behaviour 11, 78–84.

Pai, S., Kapur, R.L., 1981. The burden on the family of a psychiatric patient:development of an assessment scale. British Journal of Psychiatry 138,332–335.

Pai, S., Kapur, R.L., 1982. Impact of treatment intervention on the relationshipbetween dimensions of clinical psychopathology, social dysfunction andburden on the family of psychiatric patients. Psychological Medicine 12,651–658.

Parabiaghi, A., Lasalvia, A., Bonetto, C., et al., 2007. Predictors of changes incaregiving burden in people with schizophrenia: a 3-year follow-up study ina community mental health service. Acta Psychiatrica Scandinavica 116, 66–76.

Paykel, E., Abbott, R., Jenkins, R., et al., 2003. Urban-rural mental health differencesin Great Britain: findings from the National Morbidity Survey. InternationalReview of Psychiatry 15, 97–107.

Pollack, L., Harris, R., 1983. Measurement of social support. Psychological Reports 3,446–448.

Raj, L., Kulhara, P., Avasthi, A., 1991. Social burden of positive and negativeschizophrenia. International Journal of Social Psychiatry 37, 242–250.

Rammohan, A., Rao, K., Subbakrishna, D.K., 2002a. Burden and coping in caregiversof persons with schizophrenia. Indian Journal of Psychiatry 44, 220–227.

Rammohan, A., Rao, K., Subbakrishna, D.K., 2002b. Religious coping and psycholog-ical wellbeing in carers of relatives with schizophrenia. Acta PsychiatricaScandinavica 105, 356–362.

Ranga Rao, N.V.S.S., 1988. Comparative study of disability and family burden in ruraland urban areas. MD Thesis. Bangalore University, Bangalore.

Roick, C., Heider, D., Toumi, M., et al., 2006. The impact of caregivers’ characteristics,patients’ conditions and regional differences on family burden in schizophre-nia: a longitudinal analysis. Acta Psychiatrica Scandinavica 114, 363–374.

Roychaudhuri, J., Mondal, D., Boral, A., Bhattacharya, D., 1995. Family burden amonglong-term psychiatric patients. Indian Journal of Psychiatry 37, 81–85.

Saldanha, P., Pai, N., Krishnamurthy, K., 2002. A study of family burden and familydistress in schizophrenia. Indian Journal of Social Psychiatry 18, 63–68.

Saxena, S., Chandiramani, K., Bhargava, R., 1998. WHOQOL-Hindi: a questionnairefor assessing the quality of life in health care settings in India. World HealthOrganisation Quality of life. National Medical Journal of India 11, 160–165.

Scazufca, M., Kupiers, E., 1999. Coping strategies in relatives of people withschizophrenia before and after psychiatric admission. British Journal of Psychi-atry 174, 154–158.

Schene, A.H., Tessler, R.C., Gamache, G.M., 1996. Caregiving in severe mentalillness: conceptualization and measurement. In: Knudsen, H.C., Thornicroft,G. (Eds.), Mental Health Service Evaluation. Cambridge University Press, pp.296–316.

Schene, A.H., van Wijngaarden, B., Koeter, M.W.J., 1998. Family caregiving inschizophrenia: domains and distress. Schizophrenia Bulletin 24, 609–618.

Sefasi, A., Crumlish, N., Samalani, P., et al., 2008. A little knowledge: caregiverburden in schizophrenia in Malawi. Social Psychiatry and Psychiatric Epidemi-ology 43, 160–164.

Sekharan, R.C., Jayshree, S.R., Sivaprakash, B., 2001. Family burden and copingstrategies in relatives of schizophrenic patients. Indian Journal of Psychiatry43 (suppl) 36.

Srivastava, S., 2005. Perception of burden by caregivers of patients with schizo-phrenia. Indian Journal of Psychiatry 47, 148–152.

Tang, V.W., Leung, S.K., Lam, L.C., 2008. Clinical correlates of the caregiving experi-ence for Chinese caregivers of patients with schizophrenia. Social Psychiatryand Psychiatric Epidemiology 43, 720–726.

Tennakoon, L., Fannon, D., Doku, V., et al., 2000. Experience of caregiving: relativesof people experiencing a first episode of psychosis. British Journal of Psychiatry177, 529–533.

Thara, R., Padmawati, R., Kumar, S., et al., 1998. Burden assessment schedule—instrument to assess burden on caregivers of chronic mentally ill. Indian Journalof Psychiatry 40, 21–29.

Thomas, J.K., Suresh Kumar, P.N., Verma, A.N., Sinha, V.K., Andrade, C., 2004.Psychosocial dysfunction and family burden in schizophrenia and obsessivecompulsive disorder. Indian Journal of Psychiatry 46, 238–243.

van Wijngaarden, B., Schene, A.H., Koeter, M., et al., 2000. Caregiving in schizo-phrenia: development, internal consistency and reliability of the InvolvementEvaluation Questionnaire—European version. EPSILON Study 4. European Psy-chiatric Services: Inputs Linked to Outcome Domains and Needs. British Journalof Psychiatry 39 (Suppl.) S21–S27.

van Wijngaarden, B., Schene, A., Koeter, M., et al., 2003. EPSILON Study Group.People with schizophrenia in five countries: conceptual similarities and inter-cultural differences in family caregiving. Schizophrenia Bulletin 29, 573–586.

Verheij, R.A., 2006. Explaining urban–rural variations in health: a review of inter-actions between individual and environment. Social Science & Medicine 42,923–935.

Whalen, K.J., Buchholz, S.W., 2009. The reliability, validity and feasibility of toolsused to screen for caregiver burden: a systematic review. JBI Library of System-atic Reviews 32, 1372–1429.

Zahid, M.A., Ohaeri, J.A., 2010. Relationship of family caregiver burden with qualityof care and psychopathology in a sample of Arab subjects with schizophrenia.BMC Psychiatry 10, 71.