Elder abuse risk factors: Perceptions among older Chinese, Korean, Punjabi, and Tamil immigrants in Toronto.

OBJECTIVES: Elder abuse is a significant concern worldwide. Several factors are reported to increase the risk for elder abuse, but little is known about which factors are most relevant to immigrant communities. This study explored perceptions of risk factors for elder abuse among older immigrants, which is the first step toward designing effective interventions.
METHODS: This cross-sectional quantitative study was conducted between 2017 and 2019 in the Greater Toronto Area, Ontario, Canada and involved a convenience sample (N = 173) of older women and men from Chinese, Korean, Punjabi, and Tamil immigrant communities. Participants completed a questionnaire about the frequency and importance of risk factors of elder abuse in their respective community. Descriptive statistics were used to analyze the data within each immigrant community and analysis of variance to compare the factor ratings across communities.
RESULTS: The immigrant communities differed (p < .05) in their perception of the risk factors. Factors rated as frequent and important (x̅ > 2.0 - midpoint of the rating scale) were social isolation, financial dependence, and lack of knowledge of English for Korean; financial dependence, physical dependence, and emotional dependence for Chinese; lack of knowledge of English, emotional dependence, and physical dependence for Tamil; and social isolation for Punjabi.
CONCLUSION: The findings highlight the need for collaboration among public health and social services to work with immigrant communities in co-designing interventions to address these key risk factors and thereby reduce the risk of elder abuse.

Introduction

Older adults are the fastest-growing age group in Canada. By 2050, about one in four Canadians is expected to be 65 or older (Statistics Canada 2015). With this change in population demographics, elder abuse has become a growing public health problem. It is defined as “a single or repeated act, or lack of appropriate action, occurring within any relationship where there is an expectation of trust which causes harm or distress to an older person” (WHO 2021). Canadian estimates suggest that 4–10% of older adults experience abuse (Statistics Canada 2015) with significant effects on health (e.g., chronic pain and psychosocial distress) as well as social and economic implications (e.g., placement in nursing homes and increased use of healthcare services) at the individual, family, community, and societal levels (Burnes et al., 2014). The increasing prevalence and the negative consequences of elder abuse make it necessary to design and implement interventions to prevent or to manage this public health problem. To be successful, such interventions should target the specific factors that increase the risk for abuse in particular communities (Day et al., 2017).

Several factors have been identified as increasing the risk for elder abuse, but little is known about the risk factors within immigrant communities. The risk factors for elder abuse can be categorized as those pertaining to the victim/older person, the perpetrator/abuser, or to the relationship between them, the community, or the society (Dong, 2015, Johannesen and LoGiudice, 2013, Pillemer et al., 2006). Various studies have explored risk factors for elder abuse in the general population, but it is not yet clear which factors are common or unique among immigrant communities in Canada.

Previous studies have identified some risk factors that contribute to elder abuse in the Canadian post-migration context. For example, Haukioja (Haukioja, 2011) reported that in British Columbia, older immigrants reported financial abuse, emotional abuse, and neglect as the most frequently experienced types of abuse. Migration is known to affect the structure of immigrant families and their cultural values. Older immigrant adults often expect their children to fulfill their cultural obligations such as filial piety; yet, fulfillment of obligations may be challenging because of their post-migration and settlement stresses (Lai, 2011, Dong X et al., 2014, Sethi et al., 2018, Matsuoka et al., 2013). After migration, older adults are dependent on their children for advice, information, and support due to financial, language, and transportation barriers (Guruge et al., 2015). These become risk factors that lead to vulnerability to abuse. Other studies show that ageism (Harbison, 2016), sexism (Herron and Rosenberg, 2016), and racism (Guruge et al., 2010) experienced by older adults and their adult children in Canada can also be risk factors for elder abuse.

Current Canadian immigration policy makes newcomer older adults dependent on their sponsors for 20 years (Government of Canada 2020), which leaves the older adults financially vulnerable. Financially, newcomer older adults do not qualify for Old Age Security in Canada. Also, they find it difficult to secure employment due to various barriers, including ageism, lack of recognition of credential from home country, and policy barriers (Bélanger et al., 2016, Fulton et al., 2016). Such barriers affect the older adults’ degree of dependence on others financially.

Older immigrants may prefer to live with their children and grandchildren because of their cultural practices (Bordone and Valk, 2016). But with the changes in familial structure in the post-migration context in Canada, such as nuclearization, new expectation for older adults to care for their adult children and grandchildren, and older immigrants’ willingness to live by themselves (Kim, 2010). Multi-generational co-residence may be a risk factor.

Isolation of older immigrants, often due to language, financial, and transportation barriers, is a significant factor contributing to abuse since it increases older immigrants’ physical and emotional dependency on their families (Gao et al., 2017, Charpentier and Quéniart, 2017, Syed et al., 2017, Stewart et al., 2011, Johnson et al., 2019, Guruge et al., 2015). A lack of linguistically appropriate community resources may constrain their ability to build new social networks (Guruge et al., 2015).

In summary, the previous studies have identified several factors as increasing the risk of elder abuse in the post-migration context: age, gender, racialized status, sponsorship, length of stay in Canada, income, employment, financial dependence, social isolation, lack of knowledge of English, and physical and emotional dependency on others. But immigrant communities are diverse in their cultural beliefs, values, and norms, which influence people's knowledge and understanding of health and social problems (Sun et al., 2012, Hsiao et al., 2016), and hence, their perceived acceptability of interventions to remedy the problems (Sidani et al., 2017, Sidani et al., 2018). It is important to explore the differences in older immigrants’ perceptions of the risk factors most relevant in contributing to elder abuse in their respective communities within the Canadian post-migration context.

The current study is part of a large project that aims to identify the factors that increase the risk of abuse among older immigrant women and men, and to explore the cultural acceptability of evidence-based interventions to address elder abuse in established and recent immigrant communities in the Greater Toronto Area (GTA), Ontario (Guruge et al., 2019). The identification of the most relevant risk factors is the first step toward developing interventions that address elder abuse and are acceptable and responsive to each immigrant community's values (Sidani et al., 2017, Sidani et al., 2018, Araújo-Soares et al., 2018, Srivarathan et al., 2019). The project is informed by the community-based research and the co-production framework that involves a range of stakeholders, including older adults in understanding the problem and identifying solutions (Hawkins et al., 2017).

Material and method

Study design

A cross-sectional, quantitative design was used. Consenting older adults attended a data collection session, at a place and time of convenience to them. At the session, they completed the questionnaire containing items to assess their socio-demographic characteristics as well as their perceptions of the relevance of the risk factors, identified from the extant literature, in contributing to elder abuse in their respective immigrant community.

Ethical considerations

The study protocol was approved by Ryerson University Research Ethics Boards and The Office of Research Ethics at York University, Canada. The facilitator informed participants of the study purpose and the research activities in which they will be involved, the potential benefits and risks, and their rights including the right to refuse to answer any question and to terminate their involvement in the study at any time. The facilitator obtained participants’ written or oral (based on their preference) consent before administering the questionnaire.

Setting and target populations

The study was conducted in the GTA, where almost two in three older adults are immigrants (Um and Lightman, 2017). The target population consisted of older adults from the largest immigrant communities: East Asian and South Asian, comprising 24.0% and 32.3%, respectively, of the total racialized population in the GTA (Statistics Canada 2017). We selected established and relatively recent newcomer immigrant communities to reflect variability in socio-cultural resources (such as availability of community support) that may positively or negatively affect the experience of risk factors for elder abuse. Specifically, we targeted older adults who self-identified as belonging to one of the two East Asian communities (Korean recent and Chinese established) or the two South Asian communities (Sri Lankan Tamil recent and Punjabi established).

Sample

Women and men were eligible to participate in the study if they: (1) were 60 years or older; (2) were non-institutionalized and residing in the GTA; (3) self-identified with one of the four immigrant communities; (4) reported having experienced or knowing others who have experienced elder abuse; and (5) were able to provide informed consent in either English or their own language. Purposive sampling within each immigrant community was used to include women and men of various age groups, lengths of stay in Canada, and sponsorship status.

Recruitment

Several strategies were used to recruit older immigrants. Staff at partnering community agencies assisted in informing potentially eligible participants about the study. Older immigrants who participated in the study also helped with the recruitment (i.e., word of mouth). Additional recruitment was done via flyers posted at ethnic grocery stores, religious/community centres, health clinics, and libraries; outreach to older adult day programs; and announcements at community events.

Development of questionnaire

Literature search was done using combination of key words including: “risk/factors/elder/abuse/(X),” “risk/factors/senior/abuse/(X),” and “risk/factors/older/adult/abuse/(X).” The (X) represents each of the immigrant community. By, adding each of the community as a keyword in the search strategy, we were able to generate articles that report on risk factors that are applicable to each of the target immigrant community in this study. The search was limited to the past 5 years in order to maximize the relevance of the risk factors to the current socio-political context of immigration. The review of the literature identified recurring risk factors within each of the immigrant community and across the four communities. To measure older immigrants’ perceptions of the risk factors for elder abuse, we generated a list of factors identified through our integrative review of previous quantitative and qualitative studies (N = 4454). The quantitative studies described the frequency and the associations between risk factors and elder abuse, whereas the qualitative studies elicited participants’ perceptions of elder abuse and contributing factors, of older adults in diverse immigrant communities. A total of 13 factors (full list available in Table B and C) were commonly found (across quantitative and qualitative studies, and across immigrant communities) to be related to elder abuse.

Table B

Comparison of frequency ratings (mean (SD)) of risk factors among immigrant communities.

Factor	Korean	Chinese	Sri Lankan Tamil	Punjabi	F test, p (df = 3, 168)
Advanced age	1.63 (1.13)	2.02 (1.13)	2.57 (1.13)	3.11 (0.88)	F(3, 165) = 14.9, p < .001
Gender	1.56 (1.13)	2.17 (1.14)	2.40 (1.24)	2.18 (1.09)	F(3, 166) = 4.49, p = .005
Length of time in Canada	1.71 (1.19)	1.98 (0.99)	2.58 (1.14)	2.25 (1.13)	F(3, 165) = 4.98, p = .002
Sponsorship status	1.31 (1.10)	2.07 (1.01)	2.65 (1.19)	2.19 (1.49)	F(3, 166) = 10.0, p < .001
Lack of knowledge of English	2.28 (1.23)	2.66 (1.09)	3.44 (0.83)	2.66 (1.42)	F(3, 165) = 7.95, p < .001
Income	2.08 (1.18)	2.39 (1.02)	3.00 (1.18)	2.79 (1.42)	F(3, 164) = 5.05, p = .002
Employment	1.55 (1.02)	2.41 (0.92)	2.63 (1.18)	2.79 (1.30)	F(3, 164) = 11.4, p < .001
Physical dependence on others	2.00 (1.10)	2.98 (1.06)	3.05 (0.95)	2.97 (1.10)	F(3, 167) = 10.51, p < .001
Emotional dependence on others	2.00 (1.08)	2.75 (1.06)	3.15 (0.99)	2.30 (1.27)	F(3, 163) = 9.21, p < .001
Financial dependence on others	2.29 (1.16)	3.02 (1.00)	2.85 (1.22)	2.89 (1.20)	F(3, 164) = 3.75, p = .01
Multi-generational co-residence	1.73 (1.04)	2.45 (0.99)	2.67 (1.07)	2.82 (1.21)	F(3, 165) = 9.03 p < .001
Social isolation	2.41 (1.18)	2.68 (0.94)	2.52 (1.22)	3.11 (0.83)	F(3, 165) = 3.48, p = .02
Racialized, cultural, or ethnic group status	2.14 (1.23)	2.02 (0.91)	2.86 (1.14)	3.00 (1.11)	F(3, 165) = 8.18, p < .001

Table C

Comparison of importance ratings (mean (SD)) of risk factors among immigrant communities.

Factor	Korean	Chinese	Sri Lankan Tamil	Punjabi	F test, p (df = 3, 168)
Advanced age	2.22 (1.16)	2.61 (0.95)	2.86 (0.72)	2.78 (1.05)	F(3, 164) = 3.71, p = .01
Gender	1.90 (1.78)	2.34 (1.13)	2.33 (1.16)	2.21 (1.07)	F(3, 163) = 1.53, p = .21
Length of time in Canada	1.78 (1.01)	2.12 (1.12)	2.57 (1.02)	2.19 (1.19)	F(3, 164) = 4.14, p = .007
Sponsorship status	1.59 (1.06)	2.39 (1.00)	2.56 (1.08)	2.08 (1.44)	F(3, 166) = 6.36, p < .001
Lack of knowledge of English	2.44 (1.05)	2.78 (0.99)	3.28 (0.70)	2.42 (1.42)	F(3, 164) = 6.21, p = .001
Income	2.38 (1.02)	2.80 (0.95)	3.00 (1.15)	2.41 (1.42)	F(3, 164) = 3.08, p = .03
Employment	1.97 (1.01)	2.59 (0.97)	2.88 (0.94)	2.49 (1.30)	F(3, 165) = 6.38, p < .001
Physical dependence on others	2.31 (1.07)	3.20 (0.78)	3.05 (0.96)	2.76 (1.19)	F(3, 165) = 6.89, p < .001
Emotional dependence on others	2.39 (0.95)	2.88 (0.94)	3.00 (0.88)	2.51 (1.31)	F(3, 163) = 3.44, p = .02
Financial dependence on others	2.49 (0.89)	3.15 (0.77)	2.95 (0.85)	3.03 (1.10)	F(3, 165) = 4.61, p = .004
Multi-generational co-residence	1.98 (0.92)	2.78 (0.66)	2.90 (0.96)	2.61 (1.22)	F(3, 165) = 8.52 p < .001
Social isolation	2.71 (0.94)	2.83 (0.75)	2.83 (0.74)	3.03 (0.89)	F(3, 164) = 1.01, p = .39
Racialized, cultural, or ethnic group status	2.33 (1.13)	2.29 (0.93)	2.69 (0.90)	2.62 (1.16)	F(3, 165) = 1.61, p = .19

Data collection

Quantitative data related to participants’ demographic profile and their ratings of the risk factors for elder abuse were obtained with a questionnaire developed for the study. In addition to reporting on their socio-demographic characteristics, participants responded to items pertaining to the risk factors for elder abuse identified from the studies (reviewed previously). They were instructed to read each item that identified and describe a risk factor, and to rate the respective factor in terms of its frequency of occurrence in the participants’ immigrant community and its importance in contributing to elder abuse. The rationale for asking participants to rate the frequency and importance is that some risk factors may not be prevalent in a community, yet they could significantly lead to elder abuse. This information is useful in developing interventions so that the intervention does not only address the commonly reported risk factors, but also those that are important but hard for service providers to locate.

Quantitative data related to participants’ demographic profile and their ratings of the risk factors for elder abuse were obtained with a questionnaire developed for the study. The questionnaire was translated into the four languages of the selected immigrant communities by bilingual and bicultural research staff. The translated versions were content validated by community experts and then pilot-tested for comprehension and linguistic appropriateness with 5 older adults from each community.

Socio-demographic profile

Standard items, derived from Statistics Canada's Canadian Community Health Survey – Healthy Aging, were used to assess participants’ age, gender, marital status, number of children, country of birth, educational achievement, streams of immigration, and proficiency in English.

Perceptions of risk factors

The questionnaire items included a clear label of each factors (e.g., lack of English knowledge) and a description of how each factor may contribute to abuse (e.g., older people who cannot understand or speak English well are at risk of abuse). The description was based on relevant empirical evidence (derived from the integrative review) and written in simple language. Each factor is rated in terms of perceived frequency of occurrence (i.e., extent to which participants encounter the factor) in the respective community, on a five-point scale ranging from never (0) to very often (4), and of perceived importance in increasing the risk for elder abuse in the respective community, on a five-point scale ranging from not at all important (0) to extremely important (4). The content validity of the list was maintained by deriving the factors and their descriptions from pertinent literature and by having service providers working with older adults review it. In addition, the list was pilot tested with 5 older adults as recommended (Streiner et al., 2015).

Data analysis

Descriptive statistics (frequency, measures of central tendency and dispersion) were used to characterize the demographic profile of participants within each immigrant community. One-way analysis of variance (ANOVA), followed by the post-hoc Tukey test, were conducted to compare the ratings of the risk factors for elder abuse between the four communities.

Theory

This cross-sectional study is a part of a larger study that is guided by an ecosystemic framework and an intersectionality lens. This current cross-sectional study reports only on the quantitative data from Phase one of the study. Following Guruge and Khanlou's work (Guruge and Khanlou, 2004), which provides guidance on how to apply this framework in understanding health issues related to immigrants, this study reports on the factors at the micro (individual), meso‑ (community) and macro (societal) levels. This approach is particularly useful in informing future work on developing interventions that target each of these levels.

Results

Demographic profile

A total of 173 older adults participated: 49 Korean, 41 Chinese, 43 Tamil, and 40 Punjabi. These group sizes were adequate to detect moderate-to-large differences between communities in the risk factor ratings, setting power at 0.80 and p at 0.05 (Cohen, 1992). Participants within the four immigrant communities were comparable in terms of age and gender distribution (Table A).

Table A

Demographic profiles of participants by immigrant community.

Variable	Korean (n = 49)	Chinese (n = 41)	Sri Lankan Tamil (n = 43)	Punjabi (n = 40)
Age (mean ± SD)	69.1 ± 6.1	68.2 ± 5.5	68.5 ± 6.6	70.0 ± 5.1
Gender – women (%)	53.1	51.2	51.2	52.5
Marital status – married (%)	71.4	78.0	48.8	67.5
Number of children
1 – 2 children (%)	75.5	65.9	46.5	50.0
≥ 3 children (%)	10.2	9.8	39.5	47.5
Education – highest degree obtained (%)
Primary school	0	7.3	16.3	12.5
Some high school	12.2	17.1	39.5	17.5
High school graduate	24.5	24.4	4.7	2.5
College diploma	30.6	29.3	14.0	17.5
Baccalaureate degree	14.3	7.3	4.7	15.0
Master or doctoral degree	16.3	12.2	2.3	22.5
Streams of immigration (%)
Skilled worker	28.6	26.8	4.7	7.5
Dependent on skilled worker	14.3	9.8	0	7.5
Sponsored parent/grandparent	18.4	17.1	48.8	65.0
Refugee claimant	0	0	39.5	5.0
Business immigration	20.4	17.1	0	2.5
Temporary worker	0	0	2.3	0
International student	2.0	0	0	0
Family sponsorship	8.2	12.2	0	0
English language proficiency (mean ± SD) [scale: 1 to 4]	1.49 ± 0.85	2.68 ± 0.82	2.07 ± 0.89	2.38 ± 0.96

On average, participants were in their late 60 s (range: 60–85 years) and slightly more than half were women. However, they varied in the reported country of birth, first language, and other demographic characteristics.

Most Korean participants were married with 1–2 children, and had graduated from high school. About half arrived in Canada from 1981 to 1999, primarily as skilled workers or business immigrants. Participants reported a low level of English proficiency.

Most Chinese participants were married with 1–2 children. They graduated from high school and arrived in Canada before 2000, primarily as skilled workers or business immigrants. Almost two-thirds reported good proficiency in English.

Less than half of Tamil participants were married with at least 3 children. Most had graduated from high school and arrived in Canada after 2000, primarily sponsored by their children or as refugees. All reported average proficiency in English.

Most Punjabi participants were married, with 1–2 children, and having graduated from high school. Most arrived in Canada before 2010, primarily sponsored by their children. They reported an average level of proficiency in English.

Perceptions of risk factors

Participants rated the perceived frequency of occurrence (Table B) and importance (Table C) of the 13 risk factors for elder abuse.

Consistent with the respective rating scales, factors with a mean rating score ≥ 3 were considered to be perceived as very frequent and very important in increasing the risk for elder abuse.

Perceived frequency of risk factors. Korean participants perceived the 13 risk factors as somewhat frequently to frequently occurring in their community (Table B). Chinese participants rated financial dependence as a very frequent risk factor. Tamils considered four factors (lack of English knowledge, income, physical dependence, and emotional dependence) to occur frequently, and Punjabis perceived three factors (advanced age, social isolation, and racialized, cultural, or ethnic group status) to occur frequently in their community.

As shown in Table B, the perceived frequency of occurrence of all risk factors for elder abuse differed significantly among the four communities. Post-hoc comparisons revealed that the mean ratings by Korean participants were the lowest for almost all risk factors. Chinese participants had mean rating scores comparable to those of Tamil and Punjabi participants on most risk factors. Compared to others, Tamil participants perceived certain risk factors as frequently occurring in their community: advanced age, gender, length of time in Canada, lack of English knowledge, income, and emotional dependence. Punjabi participants rated advanced age and social isolation as frequently occurring risk factors in their community. Chinese and Korean participants rated racialized, cultural, or ethnic group status as a less frequent risk factor than Tamils and Punjabis.

Perceived importance of risk factors. None of the 13 factors were perceived by Korean participants as very important in increasing the risk of elder abuse in their community (Table C). Chinese participants rated physical and financial dependence as very important. Tamil participants viewed lack of English knowledge, income, physical dependence, and emotional dependence as very important risk factors. Punjabi participants considered financial dependence and social isolation to be very important.

The four immigrant communities differed in the perceived importance of 10 risk factors for elder abuse (Table C). On average, they rated gender as moderately low, social isolation as moderately high, and racialized, cultural, or ethnic group status as moderately important, risk factors. The post-hoc comparisons revealed that the mean rating scores were lowest for the Korean community on most factors. The mean rating scores for the Punjabi community did not differ significantly from those of the Chinese and Tamil communities. Compared to others, Chinese participants perceived financial dependence as a highly important risk factor in their community. Tamil participants reported that advanced age, length of time in Canada, lack of English knowledge, and income were highly important risk factors in their community.

Overall ranking of risk factors. To determine the most relevant risk factors within each community, we examined the concordance in the factors perceived by participants as most frequent and most important (based on mean rating scores). Overall, there was concordance in the perception of the most frequent and most important risk factors for elder abuse within each of the Korean, Chinese, and Tamil, but not the Punjabi, communities. The three highly rated risk factors varied across communities; in descending order, these were: (1) social isolation, financial dependence, and lack of knowledge of English in the Korean community; (2) financial dependence, physical dependence, and emotional dependence in the Chinese community; and (3) lack of knowledge of English, emotional dependence, and physical dependence in the Tamil community. Punjabi participants viewed social isolation as the most frequent and most important risk factor; they considered advanced age and racialized, cultural, or ethnic group status as the next two frequent factors.

Discussion

This study explored how older Korean, Chinese, Tamil, and Punjabi immigrants perceive factors that increase the risk of elder abuse in the Canadian post-migration context. Overall, the findings indicated that they differed in their perceptions of the extent to which 13 factors occur frequently and are important in increasing the risk of elder abuse in their respective community.

In general, the mean frequency and importance rating scores hovered around the midpoint (i.e., 2.0) of the rating scale. Responses suggested that participants perceived most factors as relevant, and a few factors as very relevant, in increasing the risk of elder abuse in their respective community. This pattern of responses may be associated with high levels of acquiescence and a tendency to select midpoints on rating scales among Asian immigrants (He and Van de Vijver, 2013, Wang et al., 2008). The perceived relevance of the risk factors differed across the four communities, with the mean rating scores of Korean and Chinese participants slightly lower than those of Tamil and Punjabi participants. It is unclear whether these differences were related to specific beliefs, values, or resources available to immigrant communities. Qualitative research involving various stakeholder groups (e.g., older immigrants, community leaders) is needed to explore explanations of differences in the perceived relevance of risk factors for elder abuse.

There was overall concordance in the mean rating scores for the perceived frequency and importance of the risk factors in all communities except the Punjabi community, a finding that requires further exploration. Concordance enhances the validity of conclusions regarding the perceived relevance of risk factors. However, variability was observed among the four communities in the rank ordering of the factors most relevant in increasing the risk of elder abuse. For Koreans, the most relevant risk factors were social isolation, financial dependence, and lack of English knowledge, whereas for Chinese participants, the three most relevant factors were financial dependence, physical dependence, and emotional dependence. Tamils rated lack of English knowledge, emotional dependence, and physical dependence as the most relevant factors, and Punjabis considered social isolation as most relevant, followed by financial dependence and multi-generational co-residence as most important. The reasons for this variability in ranking of risk factors are unclear. The variability could potentially be accounted for by the variation in health status and social capital among the four groups. It is interesting to note that the Hong Kong Chinese and the Punjabi group did not rate lack of English knowledge as a risk factor in contributing to elder abuse. In addition to the colonial legacy of English-language education, these two communities are more established in the GTA. Hence. They also have a better-established language-specific service network in Toronto as compared to the Korean and Tamil groups. The Chinese and the Punjabi communities have also established themselves longer compared to the Korean and the Tamil communities. Therefore, they might have had a higher extent of integration into the English-speaking Canadian society.

The variability in ranking of risk factors could potentially be explained by the immigration experience and the culture of each group. It is possible that Tamils may have experienced the atrocities and consequences of the civil war, which may have interfered with their educational achievements, personal capital, and physical and mental health. They may not have had the opportunity to acquire the proficiency in English to access the health and social services they require to address their physical and emotional health needs.

Previous studies on Korean older adults have indicated that the use of honorific and specific suffixes to refer to seniors in Korean language is related to their willingness to access services (Kang et al., 2013). In the post-migration context where the majority of formal services are provided in English, a language that does not include a polite form to address the older adults, Korean older adults might feel a sense of disrespect when they try to access services. Without the access to formal services, these Korean older adults are confined in the area with which they are familiar with. They might end up having a more difficult time building new social capitals, which results in social isolation.

The identification of social isolation as the most relevant risk factor in Punjabi group resonates with the finding in Tyyskä et al.’s study (Tyyskä et al., 2012). According to these researchers, the majority of children sponsors their parents/grandparents to Canada to become babysitters (Tyyskä et al., 2012). In our study, Punjabi group data also point to systemic barriers in leading to elder abuse. The majority of the Punjabi group participants came to Canada as sponsored parents. Under the current Canadian immigration policy, sponsors need to provide financial care for the sponsored parents/grandparents for 20 years. The majority of the Punjabi group participants were within the 20-year range of being in Canada. Therefore, their rating reflects the impact of this systemic risk factor.

In the Chinese community, Dong identified that most Chinese older adults perceive psychological abuse, physical abuse, financial exploration, and neglect as dimensions of elder abuse (Dong, 2015). The ratings of the Chinese participants in our study also reflect this perception. The ranking of the risk factors indicates that Chinese older adults might perceive financial exploitation as the most common form of elder abuse in Canada.

There are reports that co-residence among Korean families is declining and Korean older adults are more individualistic and embrace egalitarianism in Canada (Kim, 2010). With the increasing emphasis on older adults’ independence in society, they might perceive social isolation as an important risk factor in leading to abuse or neglect that they might experience. Linguistically, all participants in the Korean group indicate that only Korean is their first language. Their rating of English proficiency is also the lowest. This could potentially explain their rating of lack of English knowledge as one of the top risk factors. However, more research is needed to explore how other stakeholders (e.g., community leaders and service providers) perceive the most relevant risk factors; convergence of the findings would strengthen the conclusions about most relevant risk factors within each immigrant community.

The generally lower rating by the Korean group could potentially be explained by the low reporting rate and hence the relatively low awareness of incidents of elder abuse in the community. Prior studies found Korean older adults were more tolerant to elder abuse in comparison with other cultural groups (Joseph and Gonzalez, 2018, Lee and Eaton, 2009, Lee et al., 2014). As a result, the reporting rate of elder abuse is generally lower in the Korean group. As this study attempts to understand the perceived relevance of these risk factors, the rating of these risk factors ultimately relies on participants’ understanding and awareness of elder abuse in their community. Because of the lower reporting rate of elder abuse, the Korean participants might not be aware of the occurrence of elder abuse in the community. In addition, the Korean group recorded the lowest proficiency in English language. Since language barrier hinders access to social services, Korean older adults are connected to others in their community only, which may limit their knowledge of what should be counted as elder abuse is (Guruge et al., 2015).

Despite across-community differences in ranking, lack of English knowledge is considered more relevant within the two more recent communities (Korean and Tamil) and financial dependence is considered more relevant in the two more established communities (Chinese and Punjabi). Physical dependence, emotional dependence, and social isolation were also rated as highly relevant by more than one community. Public health and social service providers should collaborate with and engage older immigrants in co-designing interventions to address these risk factors and thereby prevent elder abuse. It is also important to ensure that recent older immigrants have the means and opportunities to attend English classes. It would be useful to revise the content and methods for delivering these classes to embrace the principles of adult learning so that the content is tailored to the learning needs of older adults (e.g., the focus could be on conversation and terms used in everyday life) and ensure that the methods are flexible and engaging, not intimidating. Although most participants qualified for Canadian Old Age Security (OAS), they still considered financial dependence – either older immigrants depending on children or vice versa – as a relevant risk factor for elder abuse. It is possible that they perceive access to the full amount of pension as insecure, or they may find it demeaning, i.e., “losing face” (Lai, 2011), to ask their children for financial assistance. Older immigrants may benefit from discussion with financial advisors and from community engagement, so it may be helpful to create part-time paid job opportunities suitable to the situation and skills of older immigrants. Several interventions have been designed to address social isolation (e.g., the FAMILY project), (Shen et al., 2017) as well as physical and emotional dependence (e.g., educational sessions to empower older adults for healthy aging) (Estebsari et al., 2018). Future research should focus on the perceived acceptability of such interventions to older immigrants within the Korean, Chinese, Tamil, and Punjabi communities.

Policy recommendations

Although the factors perceived to increase the risk of elder abuse varied among immigrant communities, the most relevant factors across the four communities were lack of English knowledge, financial, physical, and emotional dependence, and social isolation. These findings suggest the need for changes to public health policies to address financial and physical dependence among older immigrants in order to reduce their dependence on their families and caregivers, who are often the abusers. The findings also underscore the importance of offering language-specific services and designing strategies to address social isolation to prevent elder abuse among immigrants.

Strengths and limitations

This is the first study in Canada to compare and contrast how applicable the risk factors identified in previous research are to various immigrant communities. Compared to Canadian-born older adults, older immigrants in Canada face unique barriers that impact their agency in mitigating the risk of experiencing elder abuse. For example, results from our Punjabi group suggested that the Canadian immigration policy forces older immigrants to depend on their sponsors financially for 20 years. As a consequence, older immigrants might be less eager to seek services or report to the authority about elder abuse for fear of losing their family members’ and their own citizenship status in Canada. Our study was able to capture some of these unique risk factors that concerns specific the immigrant populations in Canada.

In terms of limitations, our study focused only on older immigrants living in the GTA. Compared to many other areas in Ontario, Canada, the GTA has a large number of immigrant-serving agencies and a relatively more developed transit system. Our participants may have accessed social services to address their concerns over elder abuse and may have different perceptions of these risk factors. Therefore, the findings may not reflect the perceptions of older immigrants who (re)settled in suburban and rural areas.

Our participants were primarily recruited via word-of-mouth and snowball sampling. Therefore, they were, to some extent, socially connected in their community and able to receive informal support from their friends and/or neighbours. For those living in isolation and have limited access to informal supports, their perceptions of these risk factors may be different.

Lastly, our study focused only on the quantitative data. While this approach allows us to identify risk factors and compare differences between the four communities, it does not enable the exploration of reasons underlying our participants’ perceptions of the factors, of possible pathways linking the factors to elder abuse, and of additional structural or contextual factors that increase the risk of elder abuse within and across immigrant communities. A comprehensive understanding of the risk factors is foundational for the design of interventions that are acceptable and tailored to specific immigrant communities’ perspectives.

Conclusion

Elder abuse in a growing public health concern worldwide that must be prevented in order to improve older persons’ health and well being. With the growing populations of immigrants in Canada, it is important to provide adequate support for immigrants as they age. Our study is the first step towards better understanding how Canada can prevent elder abuse among immigrant communities, by identifying the most influential risk factors and designing relevant intervention. Our results highlight that immigrant communities differ in the factors perceived as relevant in increasing the risk of elder abuse, reflecting possible variations in cultural beliefs and values, and highlighting the need to tailor public health interventions and social services to the risk factors most relevant to each community.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.