Introduction
Caregiver burden has been defined by Platt as “the presence of problems, difficulties or adverse events which affect the life (lives) of the psychiatric patient's significant others (e.g. members of the household and / or the family)”.1 Over the past 5 or 6 decades, research on caregiver burden has progressed from simple enumeration of problems faced to more elaborative and conceptually driven exploratory studies.2 Newer conceptual models such as the stress- appraisal-coping paradigm, the role theory, and the systems approach have been applied to examine the experience of caregiving and different facets of the caregiver burden.3
The number of different instruments to assess caregiver burden has also increased several folds in recent years. While earlier reviews on this subject had identified about 2 dozens of such scales,1,3 a recent systematic review yielded 74 different instruments with adequate psychometric properties, which had been used to measure caregiver burden.4 These scales differed in terms of their format (interview vs. self-report), purpose (generic vs. specific scales for individual disorders), and content (covering all areas of burden vs. focusing on 1 area, e.g. economic burden).1,3-5 Although many of these scales have been translated into other languages, 2 of them, viz. the Zarit Burden Interview and the Involvement Evaluation Questionnaire (IEQ) have undergone the most extensive testing for their cross-cultural applicability.4,6-8 The Zarit Burden Interview is a self-report scale that has more commonly targeted caregivers of patients with dementia, while the IEQ has more often been used in caregivers of those with schizophrenia or mood disorders.4,9
India has a long history of burden research, with several studies carried out among family caregivers in different parts of the country at various points in time.10 Devised by Pai and Kapur,11 the Family Burden Interview (FBI) Schedule has been the instrument most commonly used to assess caregiver burden. This scale was based on the traditional distinction between objective and subjective burden, wherein objective burden refers to the objective / tangible consequences of the patient's illness on the family, while the caregiver's own perception of the impact of caring is referred to as subjective burden.12 However, the categorisation of burden into objective and subjective aspects has been criticised on the grounds that these overlapping dimensions are difficult to operationalise and assess.13 Moreover, the primary focus of the FBI Schedule is on objective burden, and its assessment of subjective burden is relatively inadequate.1 Finally, although the FBI Schedule has been used in developing countries quite often, its psychometric properties have seldom been examined.1,14,15
In contrast to the FBI, the IEQ is based on the stress- appraisal-coping model of caregiver burden, which is one of the dominant paradigms for conceptualising such burden.4,7,9 The IEQ was developed to assess caregiver burden or the ‘consequences of caregiving' (a more neutral term for burden preferred by the authors). It relies on the caregiver's perceptions to assess burden without distinguishing between its objective and subjective aspects. The IEQ has been extensively tested and has proved to be a comprehensive, conceptually valid, and reliable means of determining the consequences of providing care for a mentally ill family member.7,9 The instrument has also been translated into several European languages. Cross-cultural applicability of some of these versions was examined in 5 different European sites as a component of the European Psychiatric Services: Inputs Linked to Outcome Domains and Needs (EPSILON) study.7,9,16 Although the internal consistency, test-retest reliability, and factorial validity of the instrument were satisfactory in a majority of these sites,9 the results often varied (possibly due to sociocultural differences).7 More recently, a study from Hong Kong found the psychometric properties of a Chinese version of the IEQ to be adequate for use among the local population.8 Apart from a Hong Kong study among caregivers of patients with schizophrenia,17 the IEQ has also been used among corresponding populations in Malawi18 and in Kuwait.19 This makes it ideally suited for adaptation and use in other Asian or African languages. An Indian adaptation of the IEQ would also fill the current local void for a newer instrument to measure caregiver burden, because of problems with the FBI. These considerations prompted us to develop a version of the IEQ in Hindi (the local language) [Hindi-IEQ], and examine its psychometric properties, particularly its factor structure, among Indian caregivers of persons with severe mental illness.
Methods
This study was conducted in the psychiatry department of a multi-specialty hospital catering to a large section of the population of North India.
Approval and Recruitment
The study protocol was approved by the research and ethics committees of the institute where it was conducted. All participants were recruited from the outpatient clinic of the department during the period of January 2010 to August 2010. Purposive sampling was used, in which all participants fulfilling the selection criteria were inducted after obtaining proper written informed consent. Other ethical safeguards were also maintained during the conduct of the study.
Involvement Evaluation Questionnaire — European Union Version
Originally developed in the Netherlands, the IEQ consisted of a series of interconnected domains that represent the possible consequences of caring for a family member with severe mental illness.4,6,7 The instrument has been refined over several versions and translated into a number of languages. The psychometric properties of the scale in terms of internal consistency (Cronbach's alpha = 0.74-0.90), test-retest reliability (intraclass correlation coefficients [ICCs] = 0.83-0.90), coverage and comprehensiveness, and ease of use had all proved to be satisfactory across different studies.7,9 Factor analyses have revealed 4 factors or caregiving domains, namely tension, supervision, worrying, and urging.16 The cross-cultural applicability of the IEQ European Union version (IEQ-EU) [in English] also appears to be largely adequate, although some reservations have been expressed in this regard.7,8,16
The version chosen for this study was the one used in the EPSILON study, which is the one usually referred to when using the term IEQ-EU.9 Its core module, which assesses caregiving consequences, consists of 31 items. Among these, 27 were grouped into 4 subscales (2 items appeared in [greater than] 1 subscale), namely: tension (9 items referring to the strained interpersonal atmosphere between patients and relatives), worrying (6 items relating to concerns about the patient's safety and future), urging (8 items that covered activational and motivational aspects of caregiving), and supervision (6 items that included tasks related to managing the patient's behavioural problems). The remaining 4 items are: (1) the ability to pursue their own activities; (2) getting used to the patient's problems; (3) the ability to cope with the patient's problems; and (4) change in emotional relationship. For research use, all items were scored on 5-point Likert scales. In addition to the 4 subscale scores, a 27-item total IEQ-EU score is usually calculated. The scale is meant to cover a 4-week period prior to the assessment, and is primarily used among caregivers who had more than 1 hour's contact with the patient during that period. It is meant to be used as a self-administered questionnaire to be completed by the caregiver, but can also be administered as a structured telephone interview. The core module took about 10 minutes to complete.
Involvement Evaluation Questionnaire — Hindi Translation
The IEQ-EU was translated into Hindi by bilingual mental health professionals, using standard translation[-]back- translation methodology.8,20 Following this, the initial Hindi version was reviewed for accuracy of translation, cultural appropriateness, and the relevance of items and constructs by 5 primary caregivers of patients with schizophrenia, 3 psychiatrists, 1 clinical psychologist, and 1 social worker. All the suggestions made by this group were incorporated into a pre-final version of the Hindi-IEQ, which was administered to a second set of 5 primary caregivers to test its acceptability. This exercise resulted in the final version, which was subsequently used to evaluate the psychometric properties and factor structure of the Hindi-IEQ.
Participants
The Hindi-IEQ was administered to 162 primary caregivers of patients with severe mental illnesses (schizophrenia, other psychotic disorders, bipolar affective disorder, recurrent depressive disorder, and obsessive-compulsive disorder).
All the patients were diagnosed according to the ICD-10 criteria after detailed semi-structured assessments by a trainee and a consultant psychiatrist. In addition, they had to be aged 18 to 60 years, had an illness duration of [greater than] 1 year, and were living with relatives. Patients with organic brain disorders, substance abuse / dependence (except nicotine), or those with chronic physical disorders were excluded. Primary caregivers were healthy adult relatives who were living with the patients for [greater than or equal to] 1 year prior to assessment. Only those judged to be intimately involved in the patient's care were included. This involved looking after the patient's daily needs, supervising medications, bringing the patient to the hospital, staying with the patient during hospitalisation, and liaising with the hospital staff.
Assessment of Psychometric Properties
All 162 caregivers completed the Hindi-IEQ on their own; of these, 18 also completed the English version to check the level of agreement between these 2 versions. Another 27 completed the Hindi-IEQ twice (1 week apart) to evaluate the test-retest reliability of this version. All items were rated on a 5-point scale (0 = never; 1 = sometimes; 2 = regularly; 3 = often; 4 = always). The factor structure of the Hindi- IEQ was examined among all 162 caregivers using an exploratory, principal components and factor analysis.
Statistical Analyses
Statistical analyses were carried out using the SPSS 14 (SPSS, Chicago [IL], US). Frequencies, percentages, means, and standard deviations were used to describe the data. Agreement between the Hindi and English versions was examined using the ICCs and Pearson's correlation coefficients (r). The test-retest reliability of the Hindi- IEQ was evaluated by estimating Cohen's kappa values and ICCs. Cronbach's alpha was used to examine the internal consistency, and the Spearman-Brown coefficient to determine the split-half reliability of the entire Hindi- IEQ scale. Principal components analysis using orthogonal rotation with the varimax technique was employed to explore the factor structure of the Hindi-IEQ. To determine the optimal number of the factors, initially only those with an eigenvalue of [greater than] 1 were included and loadings that were [greater than or equal to] 0.4 were identified as significant to render the extracted
Table 1. Profile of participants.*
* Data are shown as mean ± standard deviation and No. (%).
† p [lesser than] 0.0001.
‡ p [lesser than] 0.01.
§ p [lesser than] 0.001.
factors meaningful and interpretable. When any of the items loaded [greater than or equal to] 0.4 on [greater than or equal to] 2 factors, they were assigned to the factor with the highest loading. The aim was to obtain a factor structure, which retained the maximum number of variables and explained the maximum variance, with each factor having a loading of at least 3 items. The final factor solution was worked out by conducting multiple analyses based on differing numbers of factors, by constructing scree plots and by referring to previous literature.
Results Profile of the Sample
Demographic and clinical characteristics of the 162 patients and their primary caregivers are summarised in Table 1. The patients were in their late thirties when assessed, and had been ill for a mean (standard deviation [SD]) of about 11 (9) years. The sample consisted of an equal proportion of those with schizophrenia and mood disorder (43% each); this diagnostic distribution was probably the main reason for the almost equal numbers of males and females. A certain number of the patients were married (62%), educated, and either employed (46%) or engaged in household duties (38%). As expected, caregivers were significantly different from patients in several aspects, being older; and more likely to be male, better educated, married, and working. They had cared for the patients for almost the entire duration of the illness, and spent a mean of 4.6 (SD = 3.7) hours per day in doing so. Regarding the caregivers in schizophrenia group, 28 (41%) were spouses, 26 (38%) parents, and 15 (22%) other relatives; the corresponding figures in the mood disorder group were 35 (50%), 16 (23%), and 19 (27%). These differences between the groups were not significant (Chi-square value = 4.41; p = 0.11). Nevertheless, this distribution could account for the fact that though spousal caregivers formed the largest single category (44%), one- third of the total sample of caregivers were parents of the patients.
Psychometric Properties Agreement between Hindi and English Versions
Pearson's correlation coefficients and ICCs for scores on the Hindi and English versions among the 18 caregivers who completed both were statistically significant (p [lesser than] 0.05) for most items. This was the case for 24 out of the 31 items for correlation coefficients, and 28 out of the 31 items for ICCs. This indicated that there was a satisfactory level of agreement between the English and Hindi versions. Results are shown in Table 2.
Test-retest Reliability
For all 31 items, both the kappa values (range, 0.46-0.95) and ICC values (range, 0.76-1.00) were significant in the sample of 27 caregivers who completed the Hindi-IEQ questionnaire on 2 occasions, 1 week apart, suggesting adequate test-retest reliability.
Internal Consistency and Split-half Reliability
Cronbach's alpha (as a measure of internal consistency) was 0.89, and the Spearman-Brown coefficient (for assessing split-half reliability) was 0.68; both being more than adequate for the Hindi-IEQ.
Factor Structure
Bartlett's test of sphericity was used to test the null hypothesis that the variables were uncorrelated in the population; the Chi-square statistic for the current data was 2323.27, which was highly significant (p [lesser than] 0.001). The Kaiser-Meyer-Olkin measure of sampling adequacy was 0.87, suggesting that factor analysis was an appropriate method for analysing the current data. In the initial analyses, factor solutions with 3, 4 and 5 factors were quite similar and all provided good fit for the data. The scree plot tailed off at the 3- to 4-factor level, indicating 3- or 4-factor solutions to be optimal. A 4-factor solution was also in keeping with all the previous factor analytic studies of the IEQ.4,6-8 While the 4-factor model accounted for 29 of the 31 items, only 1 item was left unaccounted for in the 5-factor solution. The 4-factor model explained 51% of the variance, while the 5-factor solution explained 55%. In view of the tailing of the scree plot and existing literature on the factor structure of the IEQ-EU, only the 4-factor solution was eventually retained.
The rotated factor matrix and item loading of the 4- factor model is depicted in Table 3. It shows that the factor structure of the Hindi-IEQ differed from the IEQ-EU in several important ways. First, there were more than 27 items loaded onto the 4-factor version of the Hindi-IEQ. Second, differences in loading from the IEQ-EU were noted for 10 items in the 4-factor model of the Hindi-IEQ. Finally, and perhaps most significantly, there was a substantial overlap between the worrying and urging factors in terms of items that loaded onto these factors. Hence, they were re-labelled as worrying-urging factors I and II in the Hindi-IEQ.
Scores of Subscales
To allow comparison with the IEQ-EU, total and subscale scores were estimated using the 27-item total score of the original IEQ-EU format. Mean (SD) scores among caregivers of this study were as follows: total IEQ = 38.3 (19.8), urging subscale = 16.5 (1.0), worrying subscale = 10.6 (7.0), tension subscale = 7.0 (6.6), and supervision subscale = 5.8 (5.1). Comparisons of these scores with pooled estimates of IEQ-EU scores from the EPSILON study9 revealed that total IEQ scores of the latter study were significantly higher than those obtained in the present study (p [lesser than] 0.001). Similarly, scores on the subscales of worrying, tension, and supervision were also significantly higher in the EPSILON study (p [lesser than] 0.001), while there were no significant differences in the urging subscale scores between the 2 studies.
Another difference between the 2 versions was the significant associations obtained between scores on various subscales of the Hindi-IEQ. These included significant correlations between scores on the tension subscale and the other 3 subscales, i.e. the 2 worrying-urging subscales (r = 0.25, p [lesser than] 0.01; r = 0.47, p [lesser than] 0.001) and the supervision subscale (r = 0.35, p [lesser than] 0.001). Significant correlations were also observed between scores on the 2 worrying-urging subscales (r = 0.58, p [lesser than] 0.001), and between scores on 1 worrying-urging subscale and the supervision subscale (r = 0.35, p [lesser than] 0.01).
Demographic and Clinical Correlates of Total Scores
Only 2 variables were significantly associated with the total Hindi-IEQ scores. Younger patients demonstrated a significant association with higher total scores (r = 0.17; p [lesser than] 0.05), while the time spent with the patient each day by the caregiver was positively associated with higher total scores (r = 0.24; p [lesser than] 0.01). None of the other variables demonstrated any consistent associations with the Hindi-IEQ scores. In particular, there were no significant differences between total and subscale scores of the 2 main diagnostic groups (schizophrenia and mood disorders). For comparison, Hindi-IEQ scores in the 2 diagnostic groups are depicted in Table 4.
Discussion
The IEQ was chosen for adaptation because it is a simple, brief, and an easy-to-use self-administered measure of caregiver burden. It has a sound conceptual basis and has demonstrated adequate psychometric properties on repeated testing.7,9,16 It has also been used in developing countries, though its validity, reliability and factor structure in such settings have only been evaluated in a single study among Chinese caregivers of patients with schizophrenia.8
The EPSILON study was conducted to examine the cross-cultural validity and applicability of 5 key research instruments across 5 European sites located in different countries. Results regarding the IEQ-EU demonstrated that versions in different languages covered the same
Table 2. Psychometric properties of the Hindi version of Involvement Evaluation Questionnaire.
* p [lesser than] 0.001.
† p [lesser than] 0.01.
‡ p [lesser than] 0.05. 0.83* 0.93* 0.73* 0.93*
Table 3. Factor structure of the Hindi version of Involvement Evaluation Questionnaire: the 4-factor solution.*
* Principal components analysis with varimax rotation; only items with factor loading of [greater than or equal to] 0.4 are shown. Values in bold represent item loadings of each factor.
Table 4. Subscale scores of the Hindi version of Involvement Evaluation Questionnaire.
caregiving domains in all the 5 sites.7 Internal consistency and reliability of the IEQ-EU were adequate in 4 of these 5 sites.9 The underlying 4-factor structure of the IEQ also proved to be stable across all sites, and was identical to the original Dutch IEQ. At the same time, there was considerable variation in total and subscale scores across the sites. These differences persisted despite adjusting for patient and caregiver characteristics known to influence IEQ scores. Although other confounding influences could not be completely ruled out, the authors considered the possibility that the differences between sites could be attributed to cultural factors, or differences in provision of mental health care services. They therefore cautioned that other translated versions of the IEQ need to develop their own norms suited to their local culture.7
The results of the current study appear to echo these concerns by revealing patterns of similarities and differences between the Hindi and the EU versions of the IEQ. Thus, much like its progenitor, the Hindi-IEQ had sound psychometric properties. There was a satisfactory level of agreement between the translated and original versions on a majority of the items. Test-retest reliability statistics (kappa values and ICCs) of individual items of the Hindi-IEQ were significant, and well within the range found in the EPSILON sites,9 as well as those found in the sole Chinese study on psychometric properties of the IEQ.8 Similarly, internal consistency determined by Cronbach's alpha was equivalent to earlier studies,7,16 whereas split-half reliability of the scale was similar to that found in the Chinese study.8 Thus, the benchmark for substantial reliability21 was attained by the Hindi-IEQ in all these instances. Moreover, these results were obtained for both psychosis (mainly schizophrenia) and mood disorder groups. Notably, the patient sample of the current study also contained a substantial number of patients with mood disorders. This was a somewhat unique feature of this study, since the IEQ-EU has been primarily validated among caregivers of patients with schizophrenia, and only the original Dutch version of the IEQ was validated among caregivers of patients with depressive disorders.2 Another similarity was the positive association between hours of caregiving and the Hindi-IEQ scores, which was akin to the finding of higher IEQ-EU scores among caregivers with greater contact with patients in the EPSILON study.7 However, contrary to previous reports of higher IEQ scores among older patients,7,16 in the current study younger patient's age was linked to higher Hindi-IEQ scores. Then again, studies of caregiver burden, which employed different measures of burden, have highlighted the inconsistent effect of patient age on burden, with both higher and lower levels of caregiver burden being found in younger patients.12 Accordingly, other differences between the Hindi-IEQ and the IEQ-EU with regard to distribution of scores and underlying factor structure were more notable. Comparisons of mean scores with some of the earlier studies of the IEQ-EU revealed several important differences. Firstly, in the present study scores on the urging subscale were greater than those of the worrying subscale. Consequently, the rank order of mean scores (urging [greater than] worrying [greater than] tension [greater than] supervision) was somewhat different from most of the earlier studies with the IEQ.2,9,7,17 Scores on many of the subscales of the Hindi- IEQ demonstrated significant correlations with each other, which was unlike the IEQ, in which only the 2 interpersonal domains correlated with each other, as did the 2 behavioural domains.16 Finally, total and subscale IEQ scores obtained in this study were significantly lower than the pooled estimates found in the EPSILON study,7 except in the case of the urging subscale scores. However, these differences did not immediately imply a European-Asian divide, since there were also considerable variations in scores across different sites in the EPSILON study, as well as between the few Asian studies of the IEQ. A Kuwaiti study19 using the IEQ-EU obtained scores that were significantly higher than that in the current study, though the rank order of subscale scores was similar. On the other hand, in the study from Hong Kong17 the rank order of subscale scores was different from the current study, while mean scores appeared to be lower than in the present study. The differences in factor structure of the Hindi-IEQ were even more pronounced than the variations in distribution of scores. Rather than the consistently replicated 4-factor structure of the IEQ,7,8,16 in the current study 3-, 4- and 5-factor solutions all seemed to provide good fit for the data. Even though a 4-factor solution was eventually retained for the present study, there were quite a few differences between the Hindi-IEQ and the IEQ-EU. These included differences in many of the item loadings, and more importantly, a great degree of overlap between the worrying and urging factors observed in the current study. All these differences indicated that in the present study, the urging dimension of caregiver burden was much more prominent than in most of the earlier studies.2,7-9,17
The reasons for these differences were not clear. Although the scores were not adjusted for patient and caregiver characteristics, it was unlikely these factors had contributed much to the different pattern of scores of the Hindi-IEQ, as they had very little influence on the scores in the present study, and in most earlier ones.7,17 Many of the other studies had been conducted among caregivers of patients with schizophrenia,8,17,19 whereas a large proportion of the current patient sample had mood disorders. In this study, however, scores of the mood disorder group did not differ significantly from those of the schizophrenia group, indicating that diagnostic distribution was not the reason for the differences observed. Therefore, the role of sociocultural differences needs to be considered, lest they account for the variations in scores and factor structure of the Hindi-IEQ. One of the mediating factors proposed by the EPSILON study investigators was the disparity in the quality and quantity of mental health services between countries.7 However, according to this hypothesis,22 inadequate levels of service provision (the expected norm in low-income countries like India) should lead to comparatively higher levels of caregiver burden, rather than the lower levels that were actually found. Consequently, other cultural factors, particularly culturally influenced differences in perception of caregiver burden, may need to be invoked to explain these variations.
Historically, and in contrast to their western counterparts, Indian caregivers have never been systematically excluded from the treatment of people with mental illness. Thus, a large part of the mental health care has always taken place in the community with the family as the primary care provider. Long-term studies have shown that caregivers are called upon to provide support to their patients not only during the period of the acute illness, but also in the aftercare and recovery process, and that such care extended to housing and financial support to those not employed.10,23,24 Moreover, as noted by several authors, Indian patients not only tend to live with natural caregivers, but both research evidence and clinical practice suggest that the primary kin network has a pre-eminent role in virtually all decisions regarding treatment.10 In common with other ethnic groups, filial obligations (e.g. filial piety) and other forms of family responsibility (e.g. familism) also appear to influence the process of caregiving among Indian families.25,26 Apart from these expectations regarding relations and obligations of kinship, other factors that keep the Indian family involved and committed to its caregiving role have also been proposed. These include a culturally determined view of the world, an explanatory model of mental illness that is benign, as well as cultural variations in the expression of emotions.27 Moreover, this natural preference of families to be involved in the care of their mentally ill is further reinforced by the woefully inadequate mental health infrastructure and the lack of facilities and personnel, which virtually compel families to become sole caregivers of the mentally ill.10,23 Accordingly, it is not surprising that ethnographic studies have noted that relatives of Indian patients have to perform a number of care-related tasks, ordinarily carried out by mental health professionals elsewhere. By assuming many of the roles fulfilled by mental health personnel, Indian families appear to maintain considerable control over many aspects of the treatment process, e.g. defining the disorder, outpatient consultations, record-keeping, admissions, inpatient care, discharge, and aftercare.28 It is possible that this over- involvement of Indian caregivers in the treatment process could explain the predominance of the urging dimension, and its overlap with other dimensions as found in this study. However, such explanations are essentially speculative, and remain so till studies specifically designed to test them are conducted.
The results of the current study need to be viewed with some caution because of its methodological limitations. These included the fact that all patients were hospital attendees, which restricts the ability to generalise the results to other groups, e.g. community-based patients. The predominance of male caregivers was also unusual, and could be an artefact of treatment-seeking. The Indian literature in this regard is somewhat inconsistent; though female predominate as caregivers in most of the studies, there is a preponderance of male caregivers in some of the others.29 Although the sample size (n = 162) was adequate for factor analysis, being more than 5 times the number of items, sample sizes for agreement between English and Hindi versions (n = 18), and test-retest reliability (n = 27) were rather small. Thus, results on the latter parameters need replication among larger samples. Finally, the effects of several potential confounders such as social support, caregiver coping, patient and caregiver characteristics were neither explored nor controlled for.
Despite these problems, cultural variations in the IEQ as apprehended by its authors6 appear to have been replicated in this study. Accordingly, the need to modify adapted versions of the instrument to suit local cultural norms is further reiterated. Future research, which pays particular attention to this aspect, is more likely to yield a greater understanding of the caregiver burden in different cultures.
Declaration
This study was funded by a research grant of the Post Graduate Institute of Medical Education and Research, Chandigarh, India.
Platt S. Measuring the burden of psychiatric illness on the family: an evaluation of some rating scales. Psychol Med 1985;15:383-93.
van Wijngaarden B, Schene AH, Koeter MW. Family caregiving in depression: impact on caregivers’ daily life, distress, and help seeking. J Affect Disord 2004;81:211-22.
Schene AH. Objective and subjective dimensions of family burden. Towards an integrative framework for research. Soc Psychiatry Psychiatr Epidemiol 1990;25:289-97.
Whalen KJ, Buchholz SW. The reliability, validity and feasibility of tools used to screen for caregiver burden: a systematic review. JBI Libr Syst Rev 2009;32:1372-429.
Schene AH, Tessler RC, Gamache GM. Instruments measuring family or caregiver burden in severe mental illness. Soc Psychiatry Psychiatr Epidemiol 1994;29:228-40.
Chan TS, Lam LC, Chiu HF. Validation of the Chinese version of the Zarit Burden Interview. Hong Kong J Psychiatry 2005;15:9-13.
van Wijngaarden B, Schene A, Koeter M, Becker T, Knapp M, Knudsen HC, et al. People with schizophrenia in five countries: conceptual similarities and intercultural differences in family caregiving. Schizophr Bull 2003;29:573-86.
Tang VW, Leung SK, Lam LC. Validation of the Chinese version of the Involvement Evaluation Questionnaire. Hong Kong J Psychiatry 2008;18:6-14.
van Wijngaarden B, Schene AH, Koeter M, Vázquez-Barquero JL, Knudsen HC, Lasalvia A, et al. Caregiving in schizophrenia: development, internal consistency and reliability of the Involvement Evaluation Questionnaire — European version. EPSILON Study 4. European Psychiatric Services: Inputs Linked to Outcome Domains and Needs. Br J Psychiatry Suppl 2000;(39):S21-7.
Shankar R, Rao K. From burden to empowerment: the journey of family caregivers in India. In: Sartorius N, Leff J, López-Ibor JJ Jr, Okasha A, editors. Families and mental disorder: from burden to empowerment. Chichester, England: John Wiley & Sons Ltd; 2009: 259-90.
Pai S, Kapur RL. The burden on the family of a psychiatric patient: development of an assessment scale. Br J Psychiatry 1981;138:332-5.
Awad AG, Voruganti LN. The burden of schizophrenia on caregivers: a review. Pharmacoeconomics 2008;26:149-62.
Szmukler GI, Burgess P, Herrman H, Benson A, Colusa S, Bloch S. Caring for relatives with serious mental illness: the development of the Experience of Caregiving Inventory. Soc Psychiatry Psychiatr Epidemiol 1996;31:137-48.
Chien WT, Norman I. The validation and reliability of a Chinese version of the family burden interview schedule. Nurs Res 2004;53:314-22.
Siu BW, Yeung TM. Validation of the Cantonese version of Family Burden Interview Schedule on caregivers of patients with obsessive- compulsive disorder. Hong Kong J Psychiatry 2005;15:109-17.
Schene AH, van Wijngaarden B, Koeter MW. Family caregiving in schizophrenia: domains and distress. Schizophr Bull 1998;24:609-18.
Tang VW, Leung SK, Lam LC. Clinical correlates of the caregiving experience for Chinese caregivers of patients with schizophrenia. Soc Psychiatry Psychiatr Epidemiol 2008;43:720-6.
Sefasi A, Crumlish N, Samalani P, Kinsella A, O’Callaghan E, Chilale H. A little knowledge: caregiver burden in schizophrenia in Malawi. Soc Psychiatry Psychiatr Epidemiol 2008;43:160-4.
Zahid MA, Ohaeri JA. Relationship of family caregiver burden with quality of care and psychopathology in a sample of Arab subjects with schizophrenia. BMC Psychiatry 2010;10:71.
Knudsen HC, Vázquez-Barquero JL, Welcher B, Gaite L, Becker T, Chisholm D, et al. Translation and cross-cultural adaptation of outcome measurements for schizophrenia. EPSILON Study 2. European Psychiatric Services: Inputs Linked to Outcome Domains and Needs. Br J Psychiatry Suppl 2000;(39):S8-14.
Schene AH, Koeter M, van Wijngaarden B, Knudsen HC, Leese M, Ruggeri M, et al. Methodology of the multi-site reliability study. EPSILON Study 3. European Psychiatric Services: Inputs Linked to Outcome Domains and Needs. Br J Psychiatry Suppl 2000;(39):S15- 20.
Magliano L, Fadden G, Madianos M, de Almeida JM, Held T, Guarneri M, et al. Burden on the families of patients with schizophrenia: results of the BIOMED I study. Soc Psychiatry Psychiatr Epidemiol 1998;33:405-12.
Leggatt M. Families and mental health workers: the need for partnership. World Psychiatry 2002;1:52-4.
Shankar R, Menon MS. Development of a framework of intervention with families in the management of schizophrenia. Psychosoc Rehabil J 1993;16:75-91.
Gupta R, Pillai VK. Caregiver burden in South Asian families: a systems theory perspective. J Ethnic Cult Divers Soc Work 2000;9:41- 53.
Gupta R, Rowe N, Pillai VK. Perceived caregiver burden in India: implications for social services. J Women Soc Work 2009;24:69-79.
Lefley HP. Discussion on family and social support system in the care of the mentally ill. In: Murthy RS, Burns BJ, editors. Community Mental Health. Bangalore: National Institute of Mental Health and Neurosciences; 1992: 289-95.
Nunley M. The involvement of families in Indian psychiatry. Cult Med Psychiatry 1998;22:317-53.
Chakrabarti S. Impact of schizophrenia on the family: the Indian perspective. In: Kulhara P, Avasthi A, Grover S, editors. Schizophrenia: the Indian scene. Chandigarh, India: Postgraduate Institute of Medical Education and Research; 2010: 215-62.
Dr Sandeep Grover, MD, Department of Psychiatry, Postgraduate Institute of Medical Education and Research, Chandigarh, India.
Prof. Subho Chakrabarti, MD, FRCPsych, Department of Psychiatry, Postgraduate Institute of Medical Education and Research, Chandigarh, India. Dr Deepak Ghormode, MD, Department of Psychiatry, Postgraduate Institute of Medical Education and Research, Chandigarh, India.
Dr Alakananda Dutt, MD, Department of Psychiatry, Postgraduate Institute of Medical Education and Research, Chandigarh, India.
Dr Natasha Kate, MBBS, Department of Psychiatry, Postgraduate Institute of Medical Education and Research, Chandigarh, India.
Prof. Parmanand Kulhara, MD, FRCPsych, Department of Psychiatry, Postgraduate Institute of Medical Education and Research, Chandigarh, India.
© 2011. Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the associated terms available at https://www.easap.asia/index.php/journal-menu/about-the-journal