A Pilot of an Intervention Delivered to Chinese- and Spanish-Speaking Carers of People With Dementia in Australia

Abstract

There are limited language- and culture-specific support programs for carers of people with dementia living in Australia. A group intervention for use with Chinese and Spanish speakers in the United States was adapted to the Australian context, and a pilot study was undertaken with these 2 communities. The intervention is based on a cognitive behavioral therapy approach and was delivered by bilingual health professionals. The adapted material comprised 7 sessions, spanning 2 hours in duration. All 22 participants completed the Depression Anxiety and Stress Scale—Short form (DASS-21) pre- and postintervention. A significant decrease in depression, anxiety, and stress was observed among Spanish speakers; a significant decrease in depression and anxiety was present among the Chinese speakers. The implications are considered in the context of Australia’s changing aged care service system.

Introduction

Like many countries worldwide, the incidence of dementia in Australia is increasing due to the aging population. A recent report estimated that dementia diagnoses will increase by 30% between 2012 and 2050. ¹ Caring for a relative with dementia often takes a toll on the physical and emotional health of family members. ² The role and work of family carers are recognized and considered in policy and planning of health/welfare services, and various support programs are available for family carers. ³

Australia is a culturally diverse country, and it is estimated that by 2021 more than 30% of the older population will be born outside Australia. ⁴ Older people from culturally and linguistically diverse (CALD) backgrounds and their families struggle with a variety of dementia-related issues. Most research into issues for CALD carers comes from the United States and includes African American, Chinese, and Hispanic/Latino (Spanish speaking) communities. ⁵ Two of these communities, Chinese and Spanish speaking, are the focus of the current study.

In Chinese culture, family goals and achievements are prioritized over individual needs; children must respect, obey, and care for elderly family members, especially their parents. ⁶ Chinese family structures mean residential care is not acceptable, even if the carers have other commitments and limited support. ⁷ In Chinese customs, dementia is sometimes attributed to normal aging and regression to childhood but is also associated with stigma. ⁸ This stigmatization may mean families do not seek support as they fear the shame associated with dementia. ⁹ Chinese carers can experience significant physical and emotional health problems in caring for older relatives without assistance. ¹⁰

Spanish-speaking communities are family orientated. However, although Spanish-speaking carers identify positive aspects of caregiving, such as a sense of purpose ¹¹ they also report feeling stressed and depressed. ^12,13 Specific issues causing carers stress include their own and their relative’s deteriorating health, financial concerns, lack of support from family members, and general worry about the future. ^5,11,14 Spanish speakers may experience guilt and sadness when placing an older relative in residential care, as families derive comfort from living with older family members and feel proud to care for them. ^15,16

Research highlights the need to deliver culturally appropriate programs for CALD carers. ^13,17 In Australia, carers’ support programs for CALD communities are limited. Some of the programs written for Anglo carers have been delivered by bilingual staff or through interpreters, and participants may be provided with translated materials. However, there has been little evaluation as to whether the programs are effective in assisting CALD carers.

Due to the small number of extant Australian CALD carers’ programs, international alternatives were investigated to determine whether they could be used locally. Contact was initiated with colleagues in the United States who developed a cognitive behavioral therapy (CBT)-based program for Chinese- and Spanish-speaking family carers. There is strong evidence for the effectiveness of CBT and specifically for use with carers of people living with dementia. ¹⁸ The efficacy of this specific intervention has been demonstrated in a number of studies. ^{19 –21}

The intervention was adapted to the Australian context and funding sought to run a feasibility study with Chinese and Spanish speakers living in Australia. The foci of this small feasibility study was to trial the material and investigate whether the intervention enabled participants to better cope with caregiving and had a positive impact on their mental health. The intervention was trialed in different locations within Australia; Sydney and Adelaide were selected due to their large multicultural populations. Spanish speakers were recruited from Sydney and Chinese speakers from Adelaide.

Method

Study Design

This study uses a single group pre- and posttest design, where participants completed measures prior to and after the CBT intervention. The intention was to ascertain whether participants experienced an improvement in feelings of depression, anxiety, and stress after receiving the intervention.

Context

The intervention was adapted to the Australian aged care context in terms of services and protocols. Along with these changes to localize the intervention, the project advisory committee, which was made up of members of the research team and community members, discussed the methodology at length and developed strategies about how to optimize delivery of the intervention to Australian communities.

In adapting the intervention, a number of important changes were made to accommodate the Australian context. The number of sessions decreased from 13 to 7. This decision was made for several reasons. First, marked differences exist between the Spanish- and Chinese-speaking communities in Australia and the United States, particularly in regard to Spanish speakers. The largest wave of migration to Australia occurred in the 1970s, with people coming from Uruguay, Chile, and Argentina, many as refugees. The US migrants typically originate from North and Central America, particularly Mexico, Puerto Rico, and Cuba. Spanish is the second language in the United States; Australia has much smaller numbers of Spanish speakers—less than 1% of the total population. ²² There is close proximity between the United States and North/Central America, meaning that US-based Spanish speakers are more transient than Australians. In the United States, participants would often miss sessions of the intervention due to traveling between their country of origin and their current country, visiting friends/family, or collecting supplies such as medication.

Second, CALD Australian families often do not have relatives living nearby to provide care for the person with dementia while a primary carer attends support programs. People typically rely on formal respite services and therefore prefer a smaller number of sessions. Third, since other extant programs in Australia focus on issues like wills and planning for end-of-life, these topics were removed from the intervention. Relatedly, members of the advisory committee felt that some of the terms used in the communication component of the intervention might be culturally inappropriate for the Australian Chinese community (eg, assertive communication). This was discussed with LC, and a senior clinician of Chinese-speaking background provided additional assistance adapting the terminology.

Fourth, it was necessary to adapt the Spanish terminology used in the original intervention as some concepts or words were typical of North and Central American culture as opposed to South American or Spain. Finally, to optimize the uptake of the intervention, we reminded bilingual health professionals that Australian CALD communities are small and carers may be concerned about community reactions to care-related decisions. This meant AM and LC enabled the staff-recruiting participants to anticipate and better manage negative reactions.

Content of the Intervention and the Role of Bilingual Health Professionals

This intervention consisted of 7 sessions, spanning 2 hours in length. All sessions were delivered in a group-based community setting. Session 1 contained an overview of dementia and discussion of how stress affects well-being and behavior. Session 2 focused on understanding what triggers problematic behaviors in the person with dementia and how carers often react to those behaviors—sometimes in ways that increase stress. Session 3 covered methods for changing unhelpful thoughts and related behaviors to more positive thoughts and behaviors. Session 4 expanded on the previous session, teaching carers basic relaxation techniques, including breathing exercises. Session 5 focused on the ways to communicate more effectively with family members. Session 6 focused on improving skills for communicating with the person with dementia. Session 7 involved a review of the intervention’s content and suggestions for how to maintain concepts and techniques learnt. Home practice was provided to complete between all sessions.

The 2 bilingual health professionals (AM and LC)—1 Spanish and 1 Chinese speaker—were recruited based on their clinical qualifications, cultural knowledge, language skills as well as their connections and networks within their respective communities. If community members have some knowledge of the people involved in projects, they are more willing to participate, as they can be more confident the initiative will be language/culture appropriate. ²³

Complementing their extant knowledge base, AM and LC received further training in CBT, dementia, and relevant local services. Specific training on the intervention was provided by teleconference with DGT and LWT in the United States. Drawing on their previous experience in delivering the intervention to Chinese and Spanish speakers, DGT and LWT discussed how to best facilitate sessions and manage issues that typically arise. There was an expectation that AM and LC would be able to provide a culturally safe space to discuss sensitive issues, and this was an additional focus of the training delivered by DGT and LWT. A total of 16 hours of initial training was provided, including feedback on mock facilitation sessions. Additional support and refresher training were provided on an ad hoc basis throughout the 7-week period that the intervention was delivered.

Participants

Participants were recruited through local ethnic media, community organizations, and aged care networks in Adelaide and Sydney. The 2 bilingual health professionals were employed by ethno-specific, nongovernment organizations that were well known to the Spanish- and Chinese-speaking communities. Recruitment, enrollment, pretesting, and the intervention took place over a 6-month period. In total, 29 participants were recruited (16 Spanish speaking and 13 Chinese speaking); 5 participants withdrew during the course of the intervention and 2 participants had incomplete data, yielding a sample of 22 participants for analyses. The majority (85%) of Chinese-speaking participants were female and aged between 30 and 80 years. They were born in Hong Kong, Malaysia, Taiwan, or the People’s Republic of China. With 1 exception, most carers had been living in Australia for at least 10 years. Similar to the Chinese participants, 87% of the Spanish-speaking sample was female, and participants were aged between 45 and 82 years. These carers migrated from Chile, Spain, Uruguay, Argentina, Venezuela, and Ecuador. Most participants had been living in Australia for more than 20 years.

Eligibility Requirements

Participants were screened initially for entry into the study via telephone by the bilingual health professionals. To be eligible for the study, the respondent had to be engaged in 8 or more hours of care per week (physical or emotional support) to a member of their immediate or extended family who has a diagnosis of dementia. Priority entry into the study was given to first-generation family carers, as they were more likely to be living with the care recipient and have higher needs. If participants met these eligibility requirements they were invited to attend an appointment where they were provided with additional information about the study.

Participants were excluded if they did not speak Cantonese, Mandarin, or Spanish at a sufficiently fluent level to enable them to fully participate in the study. Further, participants were excluded if they had a more immediate (physical or mental health) problem requiring referral to appropriate services. Individuals with a manageable, current health condition that would not interfere with their ability to participate in the study were included.

Procedures

Detailed information about the study’s background and rationale was provided to the participants. Informed consent was given by all participants prior to commencement of the study. All 22 participants were administered the Depression Anxiety and Stress Scale—Short form (DASS-21) ²⁴ through face-to-face interviews a week prior to receiving the intervention and within a week following the last session. Since this was the first use of a CALD-specific intervention program of this type, the bilingual health professionals also collected feedback from participants about the program’s utility and to help refine the intervention for future use with Australian CALD communities.

Instrument

Depression Anxiety and Stress Scale—Short form

The DASS-21 is an abbreviated version of the original DASS-42 scale. The instrument includes 3 subscales, depression (low mood, low self-esteem, and sense of hopelessness); anxiety (autonomic arousal and fearfulness); and stress (persistent tension, irritability, easily upset, or frustrated). Each subscale comprises 7 items. Participants were asked to indicate the extent to which the items applied to them in the past week using a 4-point scale, ranging from 0 (did not apply to me at all) to 3 (applied to me much of the time). Items in each subscale were summed and multiplied by 2 (to facilitate comparison with norms generated under the DASS-42). Consistent with the extant literature, severity ratings were subsequently assigned to each subscale (see Table 1).

Table 1.

DASS-21 Severity Ratings.

	Depression	Anxiety	Stress
Normal	0-9	0-7	0-14
Mild	10-13	8-9	15-18
Moderate	14-20	10-14	19-25
Severe	21-27	15-19	26-33
Extremely severe	28+	20+	34+

Abbreviation: DASS-21, Depression Anxiety and Stress Scale—Short form.

Clinical significance was evaluated following guidelines in the extant literature, ²⁵ which provide cutoff scores and changes in scores over the depression, anxiety, and stress subscales to indicate reliable change and improvement (see Table 2).

Table 2.

Cutoff Scores and Changes in Depression, Anxiety, and Stress Subscale Scores Indicating Change.^a

	DASS-21
	Depression	Anxiety	Stress
Cutoff scores separating patient range
Normal and outpatient ranges	9.03	6.27	12.27
Outpatient and inpatient ranges	22.53	15.26	22.55
Minimum movement required to indicate reliable change
Within normal range	3.86	3.85	4.90
Between normal and outpatient  ranges	5.01	5.38	5.55
Within outpatient range	6.15	6.92	6.20
Between outpatient and inpatient  ranges	6.28	8.08	6.36
Within inpatient range	6.41	9.23	6.52

Abbreviation: DASS-21, Depression Anxiety and Stress Scale—Short form.

^a Movement from the inpatient range to the normal range is always classified as reliable and thus not presented here (Ronk et al ²⁵ ).

Analyses

Internal consistency of the DASS-21 subscales pre- and postintervention was examined using Cronbach’s α. Paired t tests were conducted to evaluate whether any differences occurred in participants’ perceptions of caregiver burden and perceived emotional health pre- and postintervention.

Results

Internal Consistency of the DASS-21

Overall, the α coefficients for the DASS-21 subscales were high at preintervention (depression: α = .86, anxiety: α = .84, and stress: α = .91) and postintervention (depression: α = .82, anxiety: α = .70, and stress α = .79). The α coefficient for the overall DASS-21 scale was high at preintervention (α = .95) and postintervention (α = .90).

Depression Anxiety and Stress Scale—Short form

In the entire sample, significant improvement in the DASS-21 was observed from pre- to postintervention: t ₂₁ = 4.38, P < .001. This was evident across the depression (t ₂₁ = 3.53, P < .01), anxiety (t ₂₁ = 3.74, P < .01), and stress subscales (t ₂₁ = 3.84, P < .01). On closer inspection, DASS-21 severity scores indicated that the majority of participants reported a normal range of emotional disturbance on the depression (40.9%), anxiety (40.9%), and stress subscales (50%) at preintervention. At postintervention, a greater number of participants reported normative functioning on the depression (63.6%), anxiety (63.6%), and stress subscales (86.4%). This is also reflected in the mean DASS-21 pre- and postintervention scores presented in Table 3.

Table 3.

Individual Group Analyses of DASS-21 Subscale Scores Pre- and Postintervention.

Outcome Measure	Chinese (n = 10), Mean (SD)	Spanish (n = 12), Mean (SD)	All participants (n = 22), Mean (SD)
DASS-21: total score
Preintervention	31.00 (19.47)	53.17 (32.84)	43.09 (29.24)
Postintervention	17.80 (12.56)	26.17 (17.78)	22.36 (15.85)
DASS-21: depression subscale
Preintervention	10.00 (6.04)	15.67 (12.47)	13.09 (10.27)
Postintervention	5.60 (4.05)	8.33 (7.48)	7.09 (6.32)
DASS-21: anxiety subscale
Preintervention	7.00 (5.19)	14.50 (10.59)	11.09 (9.21)
Postintervention	3.60 (3.24)	6.67 (6.23)	5.27 (5.22)
DASS-21: stress subscale
Preintervention	14.00 (9.71)	23.0 (11.65)	18.91 (11.51)
Postintervention	8.60 (5.74)	11.17 (7.26)	10.00 (6.59)

Abbreviations: DASS-21, Depression Anxiety and Stress Scale—Short form; SD, standard deviation.

More specifically, with regard to the Spanish-speaking group (n = 12), significant improvement was observed in the DASS-21 overall (t ₁₁ = 3.74, P < .01), including significant differences in the depression (t ₁₁ = 2.60, P < .05), anxiety (t ₁₁ = 3.06, P < .05), and stress subscales (t ₁₁ = 3.54, P < .01). In terms of the Chinese-speaking group (n = 10), significant improvement in the DASS-21 was observed from pre- to postintervention (t ₉ = 2.54, P < .05), including significant differences in the DASS-21 depression (t ₉= 2.70, P < .05) and anxiety (t ₉ = 2.68, P < .05) subscales. No significant pre- and postintervention differences were observed in the stress subscale (t ₉ = 1.82, P >.05).

The clinical significance of changes in DASS-21 scores across the depression, anxiety, and stress subscales are presented in Table 4. The classifications indicate that approximately half of all the participants exhibited a clinically significant change (either recovered or recovering) during the intervention. Specifically, the depression (45.5%) and stress (45.5%) subscales displayed the highest rate of recovery, followed by the anxiety subscale (36.4%).

Table 4.

Rates of Recovery, Recovering, Improvement, No Change, and Deterioration Across the DASS-21 Scales (N = 22).

Classification	DASS 21
	Depression, %	Anxiety, %	Stress, %
Recovered	31.8	27.3	31.8
Recovering	13.6	9.1	13.6
Improved	13.6	18.2	13.6
Unchanged	18.2	36.4	27.3
Deteriorated	22.7	9.1	13.6

Abbreviation: DASS-21, Depression Anxiety and Stress Scale—Short form.

Feedback From Participants

Since this was a feasibility study, the bilingual health professionals collected verbal feedback from participants to help refine the intervention for use with these and other Australian CALD communities. Participants were asked about the relevance of the specific content and their ability to put the concepts learnt into practice. The majority of participants reported that the content was relevant to their individual situations. Participants reported a reduction in stress levels and feeling better able to relax and cope with caring for their relative with dementia. They attributed their increased coping ability to improved communication skills and a greater understanding of dementia, which increased their level of empathy for the care recipient.

Although some participants reported struggling to complete the specific homework tasks associated with the sessions, the majority felt they were able to apply the skills they learnt and some reported having already implemented their new-found learning in practice. Participants reported that it was helpful to discuss and develop strategies to manage their specific caregiving experiences with people who had an understanding of their situation.

Discussion

The results of this feasibility study examining the application of an adapted intervention program to support dementia carers in Australia are consistent with studies conducted in the United States. In a large study of Spanish-speaking carers, ¹⁹ participants receiving the intervention displayed a significant improvement in depressive symptoms and stress postintervention compared to carers in the control group. Similarly, in a separate study of Chinese-speaking carers ²⁰ significant improvement in symptoms of depression was found postintervention among those receiving the supportive program in comparison to a minimal contact, control condition. In the current study, the results of the DASS-21 suggest the intervention had a significant positive effect on both the CALD groups. Examination of individual subscales suggests a statistical and clinical reduction in depression, anxiety, and stress levels among carers in the Spanish-speaking community and levels of depression among the Chinese-speaking community.

Participants’ verbal reports about the intervention complimented the DASS-21 results. Carers reported feeling less stressed following the intervention. They reported having a better understanding of dementia and improved communication skills, making it easier to cope with caregiving. Some carers also derived benefit from sharing experiences with other members of the group.

This small-scale feasibility study highlights the utility of a supportive intervention for CALD carers of people with dementia in Australia. This program holds much promise for addressing the needs of our aging population. Future work should be targeted toward implementing the intervention among larger groups of and carers from other Australian CALD communities.

There are a number of important issues arising from this work which warrant consideration. With increasing rates of worldwide migration, practitioners should consider using supportive programs for carers that have been demonstrated to be effective for CALD communities living in other Western countries. Although this occurs to some degree, it is important to have genuine collaboration with those who developed such interventions and to modify the interventions for local populations and systems.

Person-centered care is the “treatment and care provided by health services [that] places the person at the center of their own care and considers the needs of the older person's carers.” ²⁶ Person-centered approaches are part of the aged care reform currently taking place in Australia. Aged care has largely become the responsibility of the Commonwealth government and central tenets of the new system, named Living Longer, Living Better, ⁴ include a centralized system for intake/assessment and the consumer having more control and choice about the type of care they receive. Although this system may offer benefits for some consumers, there are concerns about CALD clients’ ability to participate in the service system, as there may be language and cultural barriers. ^27,28

However, if appropriate education about the new service systems is provided, person-centered approaches to care will be of benefit to CALD communities. There is a general consensus that quality care is language, religion, and culture appropriate ²⁹ while remembering that “diversity within diversity” ⁴ exists among CALD communities. This intervention was person centered in that it accommodated different learning styles and considered the social and political circumstances participants faced in migration, while acknowledging that not all group members may have had similar experiences. In reflecting on what occurred in this feasibility study, some carers were more interested in completing specific homework tasks, while others preferred to grasp concepts more generally. Another consideration was that, although not a specific component of the intervention, there was an awareness that some carers in the study had come from refugee-like backgrounds and may need to address this issue. The bilingual health professionals were provided with details of mental health and other appropriate services to refer participants to if necessary.

Clear communication between service provider and client is key to person-centered care. ²⁸ In working with CALD communities, it is essential to have bilingual/bicultural health professionals involved in delivery. Engaging health professionals who are also community members is essential for ensuring that carers are comfortable participating in the group and practicing new skills. Bilingual health professionals can draw upon shared knowledge and reference points to deliver information in a way that is culturally acceptable.

In closing, the findings of this study should be tempered by the small sample size. However, this was a feasibility study to demonstrate the feasibility of implementing a tailored CALD intervention program for use in Australia. The findings hold much promise for wider application of the intervention to these and other CALD communities.