A longitudinal study of the health and wellbeing of culturally and linguistically diverse caregivers of people with psychosis in Australia

Abstract

Background:

Despite Australia being an ethnically diverse country, there are limited longitudinal studies of the health and wellbeing of culturally and linguistically diverse (CALD) carers of people with psychosis. An Australian population-based survey found that 22.7% of carers of people with mental disorders were born overseas.

Aims:

This study aimed to explore the demographic profile, social connectedness, psychological health, quality of life, grief and caregiving consequences of CALD carers of people with psychosis over 12 months.

Method:

This longitudinal study utilised a prevalence survey design and recruited 52 CALD carers who were born in 24 countries outside Australia. A series of validated instruments were administered to assess carer health and wellbeing.

Results:

Descriptive and inferential analysis of the results suggested that CALD carers experienced social isolation (34.7%), psychological distress (28.9%), moderate grief and caregiving consequences and poorer quality of life than the general population. Younger CALD carers and spouse carers experienced greater negative caregiving consequences than other kinship or friend carers. There were no significant changes in these outcomes over the 12-month period of the study.

Conclusion:

Ethnic-sensitive interventions are needed to better support CALD carers experiencing negative caregiving consequences, especially younger carers or spouses, within the current level of service provision.

Keywords

Caregivers psychotic disorders health cultural diversity prevalence

Introduction

Studies of carers of people living with mental illness have provided descriptions of the caregiving experience for all carers in Australia. In the 2007 Australian National Survey of Mental Health and Wellbeing, 15% of the population were carers of people with mental disorders and 22.7% of them were born overseas (Pirkis et al., 2010). However, there are limited studies which specifically examine the health and wellbeing of culturally and linguistically diverse (CALD) carers of people with mental illness in multi-cultural Australia (Kokanovic, Petersen, & Klimidis, 2006; Minas et al., 2013; Rooney, Wright, & O’Neil, 2006), and these carers are largely neglected in service delivery in Australia (Kokanovic et al., 2006) and elsewhere (Barrio & Yamada, 2010; Horwitz & Reinhard, 1995). In addition, most caregiver studies excluded carers who had difficulty with English language which resulted in a significant gap in the literature (Minas et al., 2013).

Studies conducted elsewhere have demonstrated the poor health and challenges faced by ethnic minority caregivers within a Western-dominant culture. These findings are relevant not only for the wellbeing of these carers but have also been shown to have a negative impact on their relative with mental illness by diminishing their capacity to provide care and support (Guada, Hoe, Floyd, Barbour, & Brekke, 2012; Pearce, McGovern, & Barrowclough, 2006). Latino family members of people with schizophrenia were found to experience a high level of depression in the United States (Magaña, García, Hernández, & Cortez, 2007). Indian carers in the United Kingdom were more distressed than White carers concerning psychotic behaviours of their relative which may have been due to closer proximity of living together with the person with mental illness in the same household (Lloyd et al., 2011). Studies of African American people with mental illness showed that family factors such as caregiver burden and family dysfunction had an adverse effect on the psychosocial functioning of the person with psychiatric disorder (Guada et al., 2012). The issues faced by caregivers from ethnic minority backgrounds will likely escalate in countries with Western-dominant cultures and in the current globalisation context as supported by a review that found a high relative risk of developing schizophrenia for first-generation and second-generation migrants internationally (Cantor-Graae & Jean-Paul, 2005).

Most Australian studies of CALD carers were qualitative in nature and cross-sectional (Cross & Bloomer, 2010; Endrawes, O’Brien, & Wilkes, 2007; Kokanovic et al., 2006; Rooney et al., 2006). The limited studies on CALD caregivers of people with mental illness conducted in Australia reported that CALD carers experienced social isolation (Endrawes et al., 2007; Kokanovic et al., 2006; Morse & Messimeri-Kianidis, 2001) and psychological distress (Poon, Joubert, & Harvey, 2013). They also faced barriers in seeking help from mainstream services (Endrawes et al., 2007; Kokanovic et al., 2006; Poon et al., 2013). It was common for some cultural groups to cope with the responsibilities and impact of caregiving on their own and at the expense of their health and wellbeing (Morse & Messimeri-Kianidis, 2001; Poon et al., 2013). The challenge of dealing with the stigma of mental illness was common in CALD families (Endrawes et al., 2007; Poon et al., 2013). In addition, CALD carers reported receiving insufficient support and communication from health professionals (Endrawes et al., 2007; Rooney et al., 2006). However, the main limitation of these studies was the lack of longitudinal results to provide understanding of the changes in health and wellbeing of CALD carers over time. The other limitation was most caregiving studies used convenience sampling in recruiting carers which is problematic in generalising the results to other carers (Maurin & Boyd, 1990). This longitudinal study adds to the existing knowledge regarding CALD carers in Australia. Based on the limited studies in this area (Cross & Bloomer, 2010; Endrawes et al., 2007; Kokanovic et al., 2006; Morse & Messimeri-Kianidis, 2001; Poon et al., 2013; Rooney et al., 2006), this study is likely the first Australian longitudinal study of caregivers of people with psychosis using professional interpreters to assist with recruitment and interviewing of carers in diverse languages and using survey instruments in carer languages where available.

This study was conducted partially using the sampling framework of the second Australian prevalence study of psychosis (Morgan et al., 2012) with additional recruitment conducted within the same catchment areas. This study aimed to better understand the demographic profile, quality of life, grief, health and wellbeing and caregiving consequences of CALD carers of people with psychosis. The second aim of this project was to study changes in the carers’ experiences over time, within the context of routine mental health service delivery for their relatives with psychosis. There have been several policy initiatives with the aim of improving CALD-related mental health (Department of Health and Ageing, 2011; Rural and Regional Health and Aged Care Services, 2009) so that the catchment areas included in the study engaged in initiatives such as provision of materials in diverse CALD languages, use of interpreters and employment of a culturally diverse workforce. Furthermore, some family-focussed services were available. Therefore, the hypothesis was that CALD carers’ health and wellbeing would improve over time. As there are many definitions for CALD communities, the definition for CALD carers for this study was caregivers who had particular cultural or linguistic affiliations by virtue of their place of birth outside Australia (Rural and Regional Health and Aged Care Services, 2009).

Method

Sampling and procedure

The study was conducted within the framework of the second Australian prevalence study of psychosis, Survey of High Impact Psychosis (SHIP) (Morgan et al., 2012). In March 2010, the SHIP conducted a census of all people in contact with mental health services and non-governmental services providing mental health services. The SHIP interviewed a random sample of people who were screened positive for psychotic episodes during the census. Specifically, in the state of Victoria, SHIP was conducted in St Vincent’s Hospital Mental Health Service (STVMHS) and North West Area Mental Health Service (NWAMHS). Both STVMHS and NWAMHS provide clinical mental health services and case management to people with mental illness, within a culturally responsive framework of service delivery. In addition, both services employed carers as consultants, ran family information sessions and support groups and NWAMHS offered family psycho-education to a limited number of families (Hayes, Harvey, & Farhall, 2013). The local government areas (LGAs) of Hume, Moreland, Yarra and Boroondara serviced by STVMHS and NWAMHS were among the top 20 most diverse LGAs in Victoria and around 40% of the people in these LGAs were born in a country other than Australia (Office of Multicultural Affairs and Citizenship, 2013). This carer study utilised three sources to recruit CALD carers of people with psychiatric illness. These were as follows:

The research project contacted all the participants residing in Victoria of the SHIP (Morgan et al., 2012) and attempted to recruit carers who were not born in Australia.

The SHIP excluded people who had limited English-speaking ability (Morgan et al., 2012). The current study contacted all those who were excluded from the prevalence study but were screened positive for psychosis. The study then attempted to recruit their carers who were not born in Australia.

In order to achieve a reasonable sample size for this longitudinal study, CALD carers were also recruited from the same mental health clinics which were included in the prevalence study sites in Victoria.

The study recruited 28 carers from sampling source 1, 13 carers from source 2 and 11 carers from source 3. The diagnoses of the people from the first recruitment source were obtained from the SHIP study which had utilised the Diagnostic Interview for Psychoses (Castle et al., 2006). The diagnoses of the people from the second and third recruitment sources were obtained from the carers. The people in the second source were screened positive for psychosis using the validated Psychosis Screener (Morgan et al., 2012). The people in the third recruiting source were prescribed a type of antipsychotic medication and were receiving case management services from the clinic.

The first author made initial phone contacts with the people diagnosed with psychosis and their carers. When both carers and their relatives with psychosis agreed, the first author conducted the research interviews. For English, Cantonese and Mandarin, he conducted the interviews on his own. For other languages, professional interpreters were engaged to assist with the interpretation. The other languages were Arabic, Assyrian, Greek, Italian, Turkish and Vietnamese. These languages reflected the common languages, other than English, being spoken at home in the two geographical sites (Australian Bureau of Statistics, 2014). Written consents were obtained from all the people with psychosis and their carers who agreed to participate in the study.

When the survey instrument was available in a language other than English, it was used for the interview. The time duration of the survey was 30 to 60 minutes, and the carers were reimbursed AUD$50 for each interview for their time and effort. A total of 1 year after the first interview, the first author recontacted the CALD carers for a follow-up interview. The purpose of the follow-up was to assess for changes in the carers’ health and wellbeing after 1 year. The procedures for the follow-up interview were similar to the baseline interview. The study was approved by St Vincent’s Hospital (Melbourne) (Project no: 111/11), Melbourne Health (Project no: 2010.011) and the University of Melbourne Human Research Ethics Committees (Project no: 1034494.1).

Survey instrument

The study used a socio-demographic questionnaire to identify demographic characteristics of CALD carers and some validated survey instruments to measure their health and wellbeing for comparison with available population norms. The Kessler-10 (K10) was used to measure the psychological wellbeing of the carers. The 10-item K10 has good psychometric properties and has been used in population-based studies in Australia. Higher scores indicate greater psychological distress with 1 = None of the time and 5 = All of the time (Kessler et al., 2002). The K10 is available in English, Arabic, Chinese, Turkish, Italian, Greek and Vietnamese (Department of Health, 2014). The perceived quality of life of the carers was measured using the validated World Health Organization Quality of Life BREF instrument (WHOQOL-BREF). The WHOQOL-BREF has 26 questions covering four domains. Higher scores indicate higher perceived quality of life with 1 = Very dissatisfied/Not at all and 5 = Very/Extremely satisfied (WHOQOL Group, 1998). The WHOQOL-BREF is available in Chinese (World Health Organization, 2004). The Friendship Scale (FSS) was used to measure social connectedness in the carers. The FSS has been validated and has been used to measure social connectedness in the general population. There are six items in the FSS and some items are reverse scored. After adjustment for reverse-scored items, overall higher scores indicate greater connectedness with 1 = Not at all and 5 = Almost always (Hawthorne, 2006, 2008). The Mental Illness Version of the Texas Inventory of Grief (MIVTIG) was used to measure the current level of grief in the carers. The 16-item MIVTIG has been validated and has been used to measure the level of grief in caregivers of people with mental illness. Higher scores indicate more prominent grief with 1 = Completely false and 5 = Completely true (Miller, Dworkin, Ward, & Barone, 1990). The Involvement Evaluation Questionnaire (IEQ) was used to measure caregiving consequences. The validated IEQ has good psychometric properties and is available in English and Italian (van Wijngaarden et al., 2000). There are four subscales (Worrying, Urging, Tension and Supervision) and one sum score. Higher scores on IEQ indicate greater caregiving impact with 0 = Never and 4 = Always. The Life Skills Profile–20 (LSP-20) (Rosen, Trauer, Hadzi-Pavlovic, & Parker, 2001) measures the carer’s perception of five areas of functioning of the person with mental illness: ‘self-care’, ‘anti-social’, ‘withdrawal’, ‘bizarre’ and ‘compliance’. It has 20 items and has been shown to be reliable, valid and consistent. Higher scores indicate better functioning levels with 1 = Extreme difficulty and 4 = No difficulty (Rosen et al., 2001).

Data analysis

The data analysis was conducted using SPSS 21 (IBM Corp., 2012). Most of the analyses were descriptive providing frequency, mean scores with standard deviations, and median score with range and percentages. To examine differences in baseline and follow-up scores, the Student t test was used for scores with normal distribution and Wilcoxon Signed Ranks Test for scores with other distributions using list-wise method. The correlations between baseline and follow-up scores were examined using Pearson’s correlation for normal distribution (r) and Spearman Rho (r_s) for other distributions, using list-wise method. The correlations between all baseline caregiving outcome variables were examined using Spearman Rho analysis. One-way analysis of variance (ANOVA) was used to examine differences in mean scores for the categories for caregiving relationships. Significance level was set to at least .05 (two-tailed) for all statistical tests. The results were compared with available population normative data or population-based carer studies to show the severity of the caregiving impact on the health and wellbeing of CALD carers.

Results

Socio-demographic characteristics of the carers

A total of 52 carers were recruited to the study. The median age of the carers was 57.7 years (range = 29.3 to 81.8). There were 28 (53.8%) female carers. Most of them were married (n = 40, 76.9%). The carers were born in 24 countries in Europe (n = 15), Asia (n = 29), Africa (n = 7) and Oceania (n = 1). The largest numbers of carers came from the United Kingdom and Ireland (n = 6), China (n = 5), Italy (n = 5), Turkey (n = 3), Iraq (n = 3), Lebanon (n = 3) and Vietnam (n = 3). The median years lived in Australia was 30.5 years (range = 2 to 62). Just over a third of the carers spoke only English at home (n = 19, 36.5%). Some of them spoke English and another language at home (n = 8, 15.4%). Close to half of them spoke a language other than English at home (n = 25, 48.1%). Most of them lived in their own homes (n = 33, 63.5%). Some of them lived in public rented homes (n = 11), private rented homes (n = 6) and a rented room in a family house (n = 2). Most of them had tertiary or secondary education (Table 1). Around one-third were working and close to half of them were not in the labour force. The median age of those working was 55.0 years old (n = 19, range = 46.0 to 81.8) and median age of those not working was 57.2 years old (n = 8, range = 33.4 to 65.3).

Table 1.

Highest education, employment and marital status of the carers.

	CALD carers		Australia 2011 census (Australian Bureau of Statistics, 2013)
	n	Percentage	Percentage
Highest education attained
Non-tertiary
Primary	10	19.2	27.0
Secondary	20	38.5	20.5
Tertiary
Diploma	1	1.9	–
University degree	12	23.1	–
Other certificate/degree	9	17.3	–
Tertiary (all)	22	42.3	21.6
Employment
Working
Full-time	8	15.4	38.9
Part-time	11	21.2	18.7
Away from work	–	–	3.8
Employed (total)	19	36.6	61.4
Not in paid employment
Unemployed/Looking for work	8	15.4	3.7
Not in labour force
Retired	12	23.0	–
Homemaker	10	19.2	–
Disability support or pension	2	3.8	–
Student	1	1.9	–
Not in labour force (total)	25	47.9	35.0
Marital status
Never married	2	3.8	34.3
Married/de facto	40	76.9	48.7
Divorced/separated	9	17.3	11.4
Widowed	1	1.9	5.5
Total	52	100.0	100.0

CALD: culturally and linguistically diverse.

There were some socio-demographic characteristic differences between the three sources of sample. There was a higher percentage of CALD carers within the first source (n = 28) who were parents (n = 19, 67.9%), working (n = 15, 53.6%) and had English as their first language (n = 16, 57.1%) compared to the other two sources. As for the second source (n = 13), there was a higher percentage of CALD carers within the source who were male (n = 9, 69.2%) and were spouses/partners (n = 11, 84.6%) compared to the other two sources. A total of 11 carers (6 from Source 1, 4 from Source 2 and 1 from Source 3) dropped out of follow-up assessments due to various reasons such as refusal to participate in follow-up, inability to be contacted and long overseas trip.

Nature of the caregiving relationships

The CALD caregivers were parents (n = 26, 50.0%), partner or spouse (n = 19, 36.5%), siblings (n = 6, 11.5%) and stepmother (n = 1, 1.9%) to the person with mental illness. Most (n = 42, 80.8%) of the carers were living with the person with mental illness. For those carers who were not living with the person with mental illness (n = 10), 7 of them had at least weekly face-to-face contact and phone conversations. Most of the carers only cared for one person with mental illness (n = 47, 90.4%). The median years caring for the person with mental illness were 9.0 years (25th percentile = 4.0 years and 75th percentile = 19.0 years.)

Characteristics of the relatives with psychiatric disorders

The median age of the people with psychiatric disorders was 34.5 years old (range = 19 to 63). Around half of them were female (n = 25, 48.1%). Most (n = 29, 55.8%) of them had schizophrenia, including schizoaffective disorders. Some had bipolar disorders (n = 5), depression (n = 8) and other types of psychotic disorders (n = 5). Some of the CALD carers could not provide the name of their relative’s mental disorder (n = 6) but their relatives were likely experiencing psychotic disorders as they were prescribed antipsychotics and had a case manager supporting them from the psychiatric clinics. In terms of independent living skills, the functioning level of the people with psychiatric illness was similar at baseline and follow-up as measured by LSP-20 (Table 2). There was no statistically significant difference in the baseline and follow-up LSP-20 scores.

Table 2.

Independent functioning skills of relative with psychiatric illness using LSP-20.

	LSP-20 Baseline (n = 52)		LSP-20 Follow-up (n = 41)		Consumers with schizophrenia (n = 200) (Rosen et al., 2001)
	Mean	SD	Mean	SD	Mean
Anti-social (4–16)	14.17	2.25	14.22	2.06	13.04
Compliance (3–12)	10.37	2.43	10.38	1.70	9.54
Self-care (5–20)	14.71	3.89	14.54	3.22	15.54
Bizarre (3–12)	10.15	2.35	9.83	2.37	9.80
Withdrawal (5–20)	13.79	4.51	13.39	4.61	12.37
Total (20–80)	63.19	11.75	61.85	9.94	60.30

LSP-20: Life Skills Profile–20; SD: standard deviation.

Health, quality of life and caregiving consequences of CALD carers

A considerable minority of the CALD carers (13.5%) experienced severe social isolation as measured on the FSS during the baseline period (Table 3). At follow-up, a slightly higher percentage (19.5%) of the carers experienced severe social isolation compared to the baseline period. The baseline and follow-up FSS scores were highly correlated (r_s = .639, p < .001, n = 41), and there was no statistically significant difference between the baseline and follow-up scores (Z = −0.088, p = .930).

Table 3.

Social connectedness and psychological distress of the carers.

Social connectedness	CALD caregivers Baseline (n = 52)		CALD caregivers Follow-up (n = 41)		Australian population norm (Hawthorne, 2008)
	n	Percentage	n	Percentage	Percentage
Very isolated	7	13.5	8	19.5	2
Isolated	11	21.2	4	9.8	5
Some isolation	5	9.6	2	4.9	9
Connected	12	23.1	13	31.7	25
Very connected	17	32.7	14	34.1	59

Psychological distress level					Australian Population Norm (Australian Bureau of Statistics, 2012)

Low (10–19)	29	55.8	28	68.3	70.1
Moderate (20–24)	8	15.4	3	7.3	18.4
High (25–29)	8	15.4	1	2.4	7.4
Very high (30–50)	7	13.5	9	22.0	3.4

CALD: culturally and linguistically diverse.

During the baseline period, a significant percentage (28.9%) of the CALD carers experienced high or very high levels of psychological distress. At follow-up, 24.4% of the carers continued to experience high or very high levels of psychological distress. The baseline and follow-up K10 scores were highly correlated (r_s = .622, p < .001, n = 41), and there was no statistically significant difference between follow-up and baseline scores (Z = −0.898, p = .369).

The baseline mean score of the MIVTIG was 50.3 (SD = 17.9, n = 45). The follow-up mean score dropped slightly to 46.7 (SD = 19.8, n = 34). The baseline and follow-up MIVTIG scores were highly correlated (r_s = .726, p < .001, n = 34), and there was no statistically significant difference between the follow-up and baseline scores (Z = −1.663, p = .096).

The perceived quality of life of the CALD carers at baseline was lower than the population norm (Hawthorne, Herrman, & Murphy, 2006) for all the domains (Table 4). The baseline and follow-up WHOQOL-BREF scores were highly correlated (Physical Health: r = .701, p < .001, n = 41; Psychological Health: r = .727, p < .001, n = 41; Social Relationships: r = .516, p = .011, n = 41; Environment: r = .742, p < .001, n = 41). The perceived quality of life was worse after 1 year for all domains, especially the Social Relationships domain. However, there was no statistically significant difference between the follow-up and baseline scores (Physical Health: t = −1.022, df = 40, p = .313; Psychological Health: t = −0.170, df = 40, p = .868; Social Relationships: t = −1.321, df = 40, p = .194; Environment: t = −1.488, df = 40, p = .145).

Table 4.

Perceived quality of life and caregiving consequences of the carers.

Quality of life	CALD caregivers Baseline (n = 52)		CALD caregivers Follow-up (n = 41)		Australian Population Norm (Hawthorne et al., 2006)
	Mean	SD	Mean	SD	Mean	SD
Physical Health	68.8	21.3	67.0	21.6	73.5	18.1
Psychological Health	66.8	20.4	66.1	20.9	70.6	14.0
Social Relationships	70.0	22.4	61.4	25.7	71.5	18.2
Environment	70.8	16.3	68.6	18.2	75.1	13.0

Caregiving consequences					Support Group Carers (Stephens, Farhall, Farnan, & Ratcliff, 2011)

Worrying (0–24)	11.23	6.60	11.56	6.66	13.09	5.63
Tension (0–36)	8.12	6.72	6.32	6.21	9.65	6.27
Urging (0–32)	14.85	8.04	13.49	9.25	10.92	6.94
Supervision (0–24)	4.40	4.62	3.98	4.17	3.32	3.56
Sum score (0–108)	35.96	18.79	33.49	21.02	33.90	16.80

CALD: culturally and linguistically diverse; SD: standard deviation.

The CALD carers experienced moderate caregiving consequences as measured on IEQ. As for the baseline and follow-up IEQ scores, they were highly correlated (Urging: r_s = .653, p < .001, n = 41; Supervision: r_s = .387, p = .013, n = 41; Tension: r_s = .708, p < .001, n = 41; Worrying: r_s = .568, p < .001, n = 41; Sum scores: r_s = .700, p < .001, n = 41). There was no statistically significant difference between the follow-up and baseline scores (Urging: Z = −0.494, p = .621; Supervision: Z = −0.124, p = .902; Tension: Z = −1.349, p = .177; Worrying: Z = −0.712, p = .476; Sum scores: Z = −0.370, p = .712).

Correlations of caregiving variables with demographic characteristics of carers

The gender of the carers was not statistically correlated to any baseline caregiving outcome variables. The age of the carers was statistically correlated to IEQ Urging (r_s = −.343, p = .013, n = 52), Tension (r_s = −.316, p = .022, n = 52) and Sum scores (r_s = −.295, p = .33, n = 52), indicating that older carers experienced less negative caregiving consequences. Scatterplot examinations showed that spouses/partners were generally younger than other caregivers. One-way ANOVA was conducted for caregiving relationships with IEQ which showed that partners/spouses had higher mean scores for IEQ, and thus greater negative caregiving consequences, than other caregiving relationships, and these differences were statistically significant: Tension subscale (F(3,48) = 3.959, p = .013), Urging subscale (F(3,48) = 4.992, p = .004), Supervision (F(3,48) = 5.533, p = .002), Worry subscale (F(3,48) = 3.002, p = .040) and Sum IEQ (F(3,48) = 6.808, p = .001).

In addition, there were frequent correlations found between the caregiving outcome variables indicating that the carers who were psychologically distressed were also socially isolated and experienced more grief, greater negative caregiving consequences and poorer quality of life (Table 5).

Table 5.

Correlations between baseline caregiving outcome variables.

		FSS (n = 52)	K10 (n = 52)	MIVTIG (n = 45)	Physical health (n = 52)	Psychological health (n = 52)	Social relationships (n = 52)	Environment (n = 52)	IEQ urging (n = 52)	IEQ supervision (n = 52)	IEQ tension (n = 52)	IEQ worry (n = 52)	IEQ sum (n = 52)
FSS	r_s		−.733^**	−.482^**	.617^**	.671^**	.673^**	.608^**	−.351^*	−.418^**	−.646^**	−.562^**	−.604^**
p			<.001	.001	<.001	<.001	<.001	<.001	.011	.002	<.001	<.001	<.001
K10	r_s	−.733^**		.636^**	−.713^**	−.757^**	−.668^**	−.625^**	.349^*	.336^*	.660^**	.569^**	.604^**
p	<.001			<.001	<.001	<.001	<.001	<.001	.011	.015	<.001	<.001	<.001
MIVTIG	r_s	−.482^**	.636^**		−.549^**	−.441^**	−.385^**	−.375^*	.375^*	.183	.334^*	.476^**	.481^**
p	.001	<.001			<.001	.002	.009	.011	.011	.228	.025	.001	.001
Physical health	r_s	.617^**	−.713^**	−.549^**		.708^**	.594^**	.781^**	−.275*	−.256	−.576^**	−.321^*	−.461^**
p	<.001	<.001	<.001			<.001	<.001	<.001	.049	.067	<.001	.021	.001
Psychological health	r_s	.671^**	−.757^**	−.441^**	.708^**		.694^**	.738^**	−.296^*	−.245	−.572^**	−.417^**	−.458^**
p	<.001	<.001	.002	<.001			<.001	<.001	.033	.080	<.001	.002	.001
Social relationships	r_s	.673^**	−.668^**	−.385^**	.594^**	.694^**		.596^**	−.102	−.210	−.562^**	−.514^**	−.440^**
p	<.001	<.001	.009	<.001	<.001			<.001	.473	.135	<.001	<.001	.001
Environment	r_s	.608^**	−.625^**	−.375^*	.781^**	.738^**	.596^**		−.279*	−.381^**	−.564^**	−.330^*	−.456^**
p	<.001	<.001	.011	<.001	<.001	<.001			.045	.005	<.001	.017	.001
IEQ urging	r_s	−.351^*	.349^*	.375^*	−.275*	−.296^*	−.102	−.279*		.606^**	.382^**	.392^**	.803^**
p	.011	.011	.011	.049	.033	.473	.045			<.001	.005	.004	<.001
IEQ supervision	r_s	−.418^**	.336^*	.183	−.256	−.245	−.210	−.381^**	.606^**		.417^**	.435^**	.723^**
p	.002	.015	.228	.067	.080	.135	.005	<.001			.002	.001	<.001
IEQ tension	r_s	−.646^**	.660^**	.334^*	−.576^**	−.572^**	−.562^**	−.564^**	.382^**	.417^**		.582^**	.694^**
p	<.001	<.001	.025	<.001	<.001	<.001	<.001	.005	.002			<.001	<.001
IEQ worry	r_s	−.562^**	.569^**	.476^**	−.321^*	−.417^**	−.514^**	−.330^*	.392^**	.435^**	.582^**		.769^**
p	<.001	<.001	.001	.021	.002	<.001	.017	.004	.001	<.001			<.001
IEQ sum	r_s	−.604^**	.604^**	.481^**	−.461^**	−.458^**	−.440^**	−.456^**	.803^**	.723^**	.694^**	.769^**
p	<.001	<.001	.001	.001	.001	.001	.001	<.001	<.001	<.001	<.001

FSS: Friendship Scale; MIVTIG: Mental Illness Version of the Texas Inventory of Grief; IEQ: Involvement Evaluation Questionnaire.

Correlation is significant at the .05 level (two-tailed).

Correlation is significant at the .01 level (two-tailed).

Discussion

Our key findings are that a sizeable minority of CALD carers experienced social isolation as measured by FSS and psychological distress as measured by K10 and, even though they had been caring on average for some years, they were still experiencing moderate grief as measured by MIVTIG. The percentage of CALD carers who were socially isolated was high compared to only 2% of the general population who experienced severe social isolation (Hawthorne, 2008). Similarly for psychological health, more than a quarter of CALD carers were experiencing psychological distress compared to 17.7% of carers experiencing high or very high levels of psychological distress in Australia (Pirkis et al., 2010). Their quality of life as measured by WHOQOL-BREF was also poorer than the general population. Furthermore, the impact of caregiving as measured by IEQ was comparable with carers attending support groups (Stephens et al., 2011), with younger CALD carers and partners/spouses experiencing greater negative caregiving consequences than other carers. More importantly, the health and wellbeing of these carers did not improve over 1-year follow-up, in the context of routine service delivery in a relatively well-resourced health system with some family and CALD-inclusive services.

Strengths and limitations

To the authors’ knowledge, this is likely the first Australian study to partially use a prevalence study of psychosis to investigate the experiences of diverse CALD carers of people with psychiatric illness. Other than this innovative method in recruiting CALD carers, there were other strengths in this study. The study included CALD carers who did not speak English. There was diversity in the languages used in the study, namely English, Cantonese, Mandarin, Arabic, Assyrian, Greek, Italian, Turkish and Vietnamese. By adding two other sources of sample, the study was able to recruit a higher percentage of CALD carers who had limited English-speaking ability. In most caregiver studies, carers who did not speak English were usually excluded (Minas et al., 2013). The diverse languages used also reflected the multi-cultural background of Australia and specifically, the multi-cultural composition of the LGAs. To the authors’ knowledge, this is the first such study that included such a wide diversity of language groups in Australia. The other strength was that this study provided longitudinal findings which were not examined in the limited CALD caregiving studies conducted in Australia (Endrawes et al., 2007; Kokanovic et al., 2006; Morse & Messimeri-Kianidis, 2001; Poon et al., 2013). The longitudinal follow-up provided some insight into the experiences of CALD carers of people with psychiatric illness receiving routine mental health care over time.

There were some limitations in the study. First, although the study attempted to recruit a cohort of CALD carers within the SHIP prevalence framework, the study had to recruit more CALD carers from the relevant mental health clinics in order to achieve a reasonable sample size. Therefore, although we attempted to use a variety of sources to achieve a representative sample of CALD carers from among all such carers in the two Victorian sites, we cannot be confident that this was the case. Furthermore, the extent to which these findings apply to other settings is unclear, although it is plausible that the situation would be comparable in similar urban health settings in culturally diverse environments in Australia and elsewhere. The other limitation was that the final sample size of this study was relatively small, and the results must be interpreted with care. In addition, there were limited published population-based studies on carers which limited the comparison between the findings of this study regarding CALD carers with other carers who were born in Australia.

All people recruited into this study were caregivers of people with evidence of psychotic disorders with the diagnosis of psychosis being made through SHIP using the Diagnostic Interview for Psychosis (Castle et al., 2006) or the validated Psychosis Screener (Morgan et al., 2012) or based on prescription of antipsychotic medication. We have, therefore, considered that it is highly likely that they were experiencing a type of psychotic disorder. Although, we cannot assert that all people with mental illness who were cared for by CALD carers in this study had psychotic disorders, their independent functioning was similar to a cohort of consumers with schizophrenia (Table 2) (Rosen et al., 2001).

Profile of CALD carers

First, the findings show that CALD carers were predominantly older people, consistent with the profile of carers in population-based studies conducted in Australia (Pirkis et al., 2010) and elsewhere (van Wijngaarden et al., 2003). Around half of them were parents of the person with mental illness, also consistent with local findings concerning all carers, regardless of CALD status (Pirkis et al., 2010). In addition, all the carers were relatives of the people with mental illness which was consistent with the cultural phenomenon of family caregiving in CALD communities (Endrawes et al., 2007; Poon et al., 2013). However, around 50% of the CALD carers were male which is contrary to other studies with two-thirds of carers of people with mental illness being female (Pirkis et al., 2010) and 68.3% of carers of people with schizophrenia being female (van Wijngaarden et al., 2003). In addition, there was a higher percentage of male spousal carers with limited English-speaking proficiency who were recruited from the second source, indicating that male spousal carers might be common in CALD communities but not commonly seen in other caregiving studies, possibly due to language barriers. These findings indicate the need for further study regarding gender issues in caregiving relationships within CALD communities.

The profile of CALD carers was ageing, well-educated, having lived in Australia for a long time and with close contact with the person with mental illness. In addition, most of them were unemployed or otherwise not in the labour force compared with the general population (Australian Bureau of Statistics, 2013). The age range appears to indicate that those who were unemployed were younger (range = 33.4 to 65.3) than those who were employed (range = 46.0 to 81.8). While noting the small sample of CALD carers who were unemployed and older age of the total sample, there were some possible explanations for the low employment rate. First, the limited English language proficiency of CALD carers might hinder them in finding employment reflecting the difficulties of migrants in seeking employment in Australia (Kostenko, Harris, & Zhao, 2012). Second, CALD carers had poor social support as assessed by FSS, which implied that most of the caregiving responsibilities were borne by them. This finding of relative exclusion from the labour force might represent one of the economic costs of caregiving (Schofield, Herrman, Bloch, Howe, & Singh, 1997) especially since these CALD carers appeared to be relatively well-educated.

Health and wellbeing of CALD carers

The findings also showed that a notable proportion of the CALD carers were highly isolated, experienced psychological distress and poor quality of life and they had caregiving experiences which were similar to other convenience samples of CALD caregivers in Australia (Endrawes et al., 2007; Kokanovic et al., 2006; Morse & Messimeri-Kianidis, 2001; Poon et al., 2013). Furthermore, despite their lengthy experience of caring for most, many of these CALD carers still experienced grief which was also reported in other caregiver studies (Bland, 1998; Young, Bailey, & Rycroft, 2004). Their access to support groups with potential to address some of these adverse experiences is likely to be restricted by language and other cultural barriers. Furthermore, if carers’ health and wellbeing continues to be compromised, the caregiving burden will eventually be felt by their relatives with psychosis (Froggat, Fadden, Johnson, Leggatt, & Shankar, 2007).

In addition, from the positive correlations between baseline and follow-up scores, the findings showed that the carers continued to experience similar levels of social isolation, psychological distress, grief, poor quality of life and negative caregiving consequences after 1 year, in the context of current mental health services in a reasonably well-resourced and organised community mental health service system with a policy and service delivery emphasis on cultural responsiveness (Rural and Regional Health and Aged Care Services, 2009). In similar well-funded mental health services in the United Kingdom, carers’ satisfaction remained similar over time whether their family members with psychosis were receiving standard services or intensive community mental health services (Szmukler, Wykes, & Parkman, 1998). This longitudinal study on CALD carers adds to the literature that CALD carers’ health and wellbeing are unlikely to improve unless there are adequate and culturally appropriate support services for carers.

CALD carers who were spouses or partners were likely experiencing more adverse caregiving consequences than other carers as seen from the correlations between IEQ scores with caregiving relationships. This is consistent with findings on the impact faced by CALD spouses or partners who were caring for people with mental illness in Australia (Endrawes et al., 2007; Poon et al., 2013). This specific finding could be due to the combination of significant social isolation, carers’ age and poor marital satisfaction as found in other studies (Jungbauer & Angermeyer, 2002; Jungbauer, Wittmund, Dietrich, & Angermeyer, 2004). This suggests that this subgroup may require specific attention to alleviate their caregiving burden, since current services and programmes for carers, such as support groups and more structured family psycho-education programmes, tend to be geared towards parental caregivers as well as being harder for spouses or partners to access, especially if they are employed (Jungbauer & Angermeyer, 2002; Maurin & Boyd, 1990; Stephens et al., 2011). In contrast, the older carers in our study experienced less negative caregiving consequences, perhaps indicating adaptation to the challenges of caregiving over lengthy periods, or, more likely, less severe consequences for parents in this respect.

The frequent correlations found between the caregiving outcome variables showed that the likely impact of the relative’s psychiatric illness was on many domains of the carers’ health and wellbeing. Thus, those carers who experienced negative impacts in one area of their lives were likely to also experience other negative impacts within other areas of their lives. This is consistent with many previous caregiving studies, although none have specifically studied CALD carers in this respect (Fadden, Bebbington, & Kuipers, 1987; Schene, 1990).

Implications for service delivery for CALD carers

The results of this study indicate that CALD carers require access to effective support services to achieve an improvement in their health and wellbeing over time. However, Australian studies showed that CALD carers faced many barriers in accessing services (Endrawes et al., 2007; Kokanovic et al., 2006; Poon et al., 2013) which likely further contributed to their poor health and wellbeing. Some of the ways to overcome these barriers include employing and matching clinicians from diverse ethnic backgrounds with corresponding consumers, providing cultural competency training for staff, creating awareness and respect for cultural beliefs in mental health services and encouraging sensitivity in communication with CALD clients and their families among clinicians (Cross & Bloomer, 2010; Poon et al., 2013; Ziguras, Klimidis, Lewis, & Stuart, 2003).

In addition, the lack of improvement in CALD carers’ health and wellbeing may be due to the neglect of health services in providing suitable interventions to address the needs of CALD carers. Previous research has shown that evidence-based family approaches must be modified to be effective within migrant populations (Harvey & O’Hanlon, 2013). Modified family psycho-education programmes are likely to be beneficial to support CALD carers, especially if modified to better address the needs of spouses or partners (Bradley et al., 2006; Jungbauer & Angermeyer, 2002). In Australia, some modified family psycho-education programmes have shown promising results (Bradley et al., 2006; Hayes et al., 2013). However, in general, such ethnic-sensitive interventions are lacking in multi-cultural Australia and other similar countries.

Conclusion

The study suggests that CALD carers were experiencing considerable negative caregiving impact within the current level of service provision in Australia, which may reflect the situation in other countries with a large influx of migrants and similar health delivery systems. The profile and caregiving impact experienced by the CALD carers suggests that there is a need to review existing policies and services to better support CALD families. There is also a need to modify evidence-based practices to be ethnic-sensitive. Finally, there is a need for further research with a larger sample size and multi-cultural representation for generalisable results.

Footnotes

Acknowledgements

This publication is based on data collected (specifically, the diagnosis variable) in the framework of the 2010 Australian National Survey of High Impact Psychosis. The members of the Survey of High Impact Psychosis Study Group are V Morgan (National Project Director), A Jablensky (Chief Scientific Advisor), A Waterreus (National Project Coordinator), R Bush, V Carr, D Castle, M Cohen, C Galletly, C Harvey, B Hocking, A Mackinnon, P McGorry, J McGrath, A Neil, S Saw, and H Stain. Ethics approvals for the study were obtained from relevant institutional human research ethics committees. The SHIP study was funded by the Australian Government Department of Health and Ageing. This report acknowledges all the effort and hard work by the local site coordinators. We would also like to thank the hundreds of mental health professionals who participated in the preparation and conduct of the survey and the many Australians with psychotic disorders and their relatives who gave their time and whose responses form the basis of this publication. The authors are especially grateful to SHIP investigator, Professor David Castle from St Vincent’s Mental Health Service (Melbourne) for his support on this carer study. The authors are also grateful to Nga Tran from St Vincent’s Mental Health Service (Melbourne) who helped to contact and introduce the Vietnamese-speaking clients to the researchers.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical standards

The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the William Buckland Foundation, which is managed by ANZ Trustees (Reference Number: CT 12530).

References

Australian Bureau of Statistics. (2012). Australian Health Survey: First results. Retrieved from http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/4338.0main+features42011-13

Australian Bureau of Statistics. (2013). 2011 Census QuickStats. Retrieved from http://www.censusdata.abs.gov.au

Australian Bureau of Statistics. (2014). Community profile. Retrieved from http://profile.id.com.au

Barrio

Yamada

A.-M.

(2010). Culturally based intervention development: The case of Latino families dealing with schizophrenia. Research on Social Work Practice, 20, 483–492.

Bland

(1998). Understanding grief and guilt as common themes in family response to mental illness: Implications for social work practice. Australian Social Work, 51, 27–34.

Bradley

Couchman

Perlesz

Nguyen

Singh

Riess

(2006). Multiple-family group treatment for English-and Vietnamese-speaking families living with schizophrenia. Psychiatric Services, 57, 521–530.

Cantor-Graae

Jean-Paul

(2005). Schizophrenia and migration: A meta-analysis and review. The American Journal of Psychiatry, 162, 12–24.

Castle

D. J.

Jablensky

McGrath

J. J.

Carr

Morgan

Waterreus

. . . Farmer

(2006). The diagnostic interview for psychoses (DIP): Development, reliability and applications. Psychological Medicine, 36, 69–80.

Cross

W. M.

Bloomer

M. J.

(2010). Extending boundaries: Clinical communication with culturally and linguistically diverse mental health clients and carers. International Journal of Mental Health Nursing, 19, 268–277.

10.

Department of Health. (2014). Kessler 10 measure. Retrieved from http://www.healthtranslations.vic.gov.au/bhcv2/bhcht.nsf/PresentDetail?Open&s=Kessler_10_measure

11.

Department of Health and Ageing. (2011). Fact sheet mental health services for CALD communities. Canberra, Australian Capital Territory, Australia: Author.

12.

Endrawes

O’Brien

Wilkes

(2007). Egyptian families caring for a relative with mental illness: A hermeneutic study. International Journal of Mental Health Nursing, 16, 431–440.

13.

Fadden

Bebbington

Kuipers

(1987). The burden of care: The impact of functional psychiatric illness on the patient’s family. The British Journal of Psychiatry, 150, 285–292.

14.

Froggat

Fadden

Johnson

D. L.

Leggatt

Shankar

(2007). Families as partners in mental health care: A guidebook for implementing family work. Toronto, Ontario, Canada: World Fellowship for Schizophrenia and Allied Disorders.

15.

Guada

Hoe

Floyd

Barbour

Brekke

(2012). How family factors impact psychosocial functioning for African American consumers with schizophrenia. Community Mental Health Journal, 48, 45–55.

16.

Harvey

O’Hanlon

(2013). Family psycho-education for people with schizophrenia and other psychotic disorders and their families. Australian and New Zealand Journal of Psychiatry, 47, 516–520.

17.

Hawthorne

(2006). Measuring social isolation in older adults: Development and initial validation of the friendship scale. Social Indicators Research, 77, 521–548.

18.

Hawthorne

(2008). Perceived social isolation in a community sample: Its prevalence and correlates with aspects of peoples’ lives. Social Psychiatry and Psychiatric Epidemiology, 43, 140–150.

19.

Hawthorne

Herrman

Murphy

(2006). Interpreting the WHOQOL-Bref: Preliminary population norms and effect sizes. Social Indicators Research, 77, 37–59.

20.

Hayes

Harvey

Farhall

(2013). Family psychoeducation for the treatment of psychosis. InPsych, 35, 16–17.

21.

Horwitz

A. V.

Reinhard

S. C.

(1995). Ethnic differences in caregiving duties and burdens among parents and siblings of persons with severe mental illnesses. Journal of Health and Social Behavior, 36, 138–150.

22.

IBM Corp. (2012). IBM SPSS Statistics for Windows (Version 21.0) [Software]. Armonk, NK: Author.

23.

Jungbauer

Angermeyer

M. C.

(2002). Living with a schizophrenic patient: A comparative study of burden as it affects parents and spouses. Psychiatry, 65, 110–123.

24.

Jungbauer

Wittmund

Dietrich

Angermeyer

M. C.

(2004). The disregarded caregivers: Subjective burden in spouses of schizophrenia patients. Schizophrenia Bulletin, 30, 665–675.

25.

Kessler

R. C.

Andrews

Colpe

L. J.

Hiripi

Mroczek

D. K.

Normand

S. L. T.

. . . Zaslavsky

A. M.

(2002). Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychological Medicine, 32, 959–976.

26.

Kokanovic

Petersen

Klimidis

(2006). ‘Nobody can help me … I am living through it alone’: Experiences of caring for people diagnosed with mental illness in ethno-cultural and linguistic minority communities. Journal of Immigrant and Minority Health, 8, 125–135.

27.

Kostenko

Harris

Zhao

(2012). Occupational transition and country-of-origin effects in the early stage occupational assimilation of immigrants: Some evidence from Australia. Applied Economics, 44, 4019–4035.

28.

Lloyd

Singh

Merritt

Shetty

Yiend

Singh

Burns

(2011). A comparison of levels of burden in Indian and white parents with a son or daughter with schizophrenia. International Journal of Social Psychiatry, 57, 300–311.

29.

Magaña

S. M.

García

J. I. R.

Hernández

M. G.

Cortez

(2007). Psychological distress among Latino family caregivers of adults with schizophrenia: The roles of burden and stigma. Psychiatric Services, 58, 378–384.

30.

Maurin

J. T.

Boyd

C. B.

(1990). Burden of mental illness on the family: A critical review. Archives of Psychiatric Nursing, 4, 99–107.

31.

Miller

F. E.

Dworkin

Ward

Barone

(1990). A preliminary study of unresolved grief in families of seriously mentally ill patients. Hospital & Community Psychiatry, 41, 1321–1325.

32.

Minas

Kakuma

Too

L. S.

Vayani

Orapeleng

Prasad-Ildes

. . . Oehm

(2013). Mental health research and evaluation in multicultural Australia: Developing a culture of inclusion. International Journal of Mental Health Systems, 7, 23.

33.

Morgan

V. A.

Waterreus

Jablensky

Mackinnon

McGrath

J. J.

Carr

. . . Saw

(2012). People living with psychotic illness in 2010: The second Australian national survey of psychosis. Australian and New Zealand Journal of Psychiatry, 46, 735–752.

34.

Morse

C. A.

Messimeri-Kianidis

(2001). Keeping it in the family: Caregiving in Australian-Greek families. Social Work in Health Care, 34, 299–314.

35.

Office of Multicultural Affairs and Citizenship. (2013). Victoria’s diverse population 2011 census. Melbourne, Victoria, Australia: Department of Premier and Cabinet.

36.

Pearce

McGovern

Barrowclough

(2006). Assessment of need for psychosocial interventions in an Asian population of carers of patients with schizophrenia. Issues and Innovations in Nursing Practice, 54, 284–292.

37.

Pirkis

Burgess

Hardy

Harris

Slade

Johnston

(2010). Who cares? A profile of people who care for relatives with a mental disorder. Australian and New Zealand Journal of Psychiatry, 44, 929–937.

38.

Poon

W. C.

Joubert

Harvey

(2013). Experiences of Chinese migrants caring for family members with schizophrenia in Australia. Social Work in Health Care, 52, 144–165.

39.

Rooney

Wright

O’Neil

(2006). Issues faced by carers of people with a mental illness from culturally and linguistically diverse backgrounds: Carers’ and practitioners’ perceptions. Australian e-Journal for the Advancement of Mental Health, 5, 1–13.

40.

Rosen

Trauer

Hadzi-Pavlovic

Parker

(2001). Development of a brief form of the Life Skills Profile: The LSP-20. Australian and New Zealand Journal of Psychiatry, 35, 677–683.

41.

Rural and Regional Health and Aged Care Services. (2009). Cultural responsiveness framework – Guidelines for Victorian health services. Melbourne, Victoria, Australia: Rural and Regional Health and Aged Care Services, Victorian Government, Department of Health.

42.

Schene

(1990). Objective and subjective dimensions of family burden. Social Psychiatry and Psychiatric Epidemiology, 25, 289–297.

43.

Schofield

H. L.

Herrman

H. E.

Bloch

Howe

Singh

(1997). A profile of Australian family caregivers: Diversity of roles and circumstances. Australian and New Zealand Journal of Public Health, 21, 59–66.

44.

Stephens

J. R.

Farhall

Farnan

Ratcliff

K. M.

(2011). An evaluation of Well Ways, a family education programme for carers of people with a mental illness. Australian and New Zealand Journal of Psychiatry, 45, 45–53.

45.

Szmukler

G. I.

Wykes

Parkman

(1998). Caregiving and the impact on carers of a community mental health service. The British Journal of Psychiatry, 173, 399–403.

46.

van Wijngaarden

Schene

A. H.

Koeter

Becker

Knapp

Knudsen

H. C.

, … EPSILON Study Group. (2003). People with schizophrenia in five countries: Conceptual similarities and intercultural differences in family caregiving. Schizophrenia Bulletin, 29, 573–586.

47.

van Wijngaarden

Schene

A. H.

Koeter

Vázquez-Barquero

J. L.

Knudsen

H. C.

Lasalvia

McCrone

(2000). Caregiving in schizophrenia: Development, internal consistency and reliability of the Involvement Evaluation Questionnaire-European Version. The British Journal of Psychiatry, 177(39), s21–s27.

48.

WHOQOL Group. (1998). Development of the World Health Organization WHOQOL-BREF quality of life assessment. Psychological Medicine, 28, 551–558.

49.

World Health Organization. (2004). The World Health Organization Quality of Life (WHOQOL)-BREF. Geneva, Switzerland: Author.

50.

Young

Bailey

Rycroft

(2004). Family grief and mental health: A systemic, contextual and compassionate analysis. Australian and New Zealand Journal of Family Therapy, 25, 188–197.

51.

Ziguras

Klimidis

Lewis

Stuart

(2003). Ethnic matching of clients and clinicians and use of mental health services by ethnic minority clients. Psychiatric Services, 54, 535–541.