Article Text

Download PDFPDF

Developing a comprehensive understanding of elder abuse prevention in immigrant communities: a comparative mixed methods study protocol
  1. Sepali Guruge1,
  2. Souraya Sidani1,
  3. Atsuko Matsuoka2,
  4. Guida Man3,
  5. Diane Pirner1
  1. 1 Daphne Cockwell School of Nursing, Ryerson University, Toronto, Ontario, Canada
  2. 2 School of Social Work, York University, Toronto, Ontario, Canada
  3. 3 Department of Sociology, York University, Toronto, Ontario, Canada
  1. Correspondence to Dr Sepali Guruge; sguruge{at}


Introduction Older adults are the fastest growing age group in Canada. Elder abuse has significant individual and societal implications, so it is critical to address. While interest in this topic is increasing, little is known about the risk factors for elder abuse in immigrant communities in Canada, or about culturally relevant strategies to address these risk factors.

Methods and analysis This mixed-methods study is guided by the intersectionality and ecological frameworks. We will include two long-term (ie, established) and two recent immigrant communities from East Asian and South Asian communities in the Greater Toronto Area: Chinese, Korean, Punjabi and Tamil. Through structured group interviews, we will first identify factors that contribute to elder abuse within and across each of the immigrant communities and then explore culturally relevant strategies to address those risk factors. Group interviews will be conducted separately with five stakeholder groups in each of the four languages: older women, older men, family members, community leaders and service providers. Quantitative and qualitative data will be analysed at the level of the particular interview groups, subgroups and communities, and will be integrated across communities to identify common and unique risk factors and strategies to address elder abuse.

Ethics and dissemination The study protocol has received ethics approval from the two universities associated with the research team. Given the comprehensive approach to incorporate local knowledge and expert contributions from multi-level stakeholders, the empirical and theoretical findings will facilitate practice change and improve the well-being of older men and women in immigrant communities.

  • public health
  • social medicine
  • primary care

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Strengths and limitations of this study

  • The total sample across four communities will include more than 350 participants who will be diverse in terms of gender, age, immigration experience, time spent in Canada and stakeholder group (older adults, children/children in-law, grandchildren, community leaders and non-community service providers).

  • The design includes both recent and long-term immigrant communities, allowing for comparison of risk factors and prevention strategies with regard to the quantity, diversity, accessibility and quality of resources available to different immigrant communities.

  • The design includes two South Asian and two East Asian communities, enabling comparison between groups that historically share some cultural values and how these may be shaped by similar or different post-migration contexts.

  • Some of the limitations include the potential for self-selection and social desirability biases in responses.


Elder abuse is a growing problem with significant societal implications. It can be 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’.1 Abuse may include physical, psychological, material and sexual violence; mistreatment; isolation and abandonment; violation of legal and medical rights; and deprivation of choices, decisions, status, finances and respect.2 Canadian estimates suggest that 4%–10% of all older adults (65 years or older) experience abuse.3 4 However, according to the WHO,1 such numbers likely represent under-reporting by as much as 80% in the general population, within and across countries.

Canada is the home for immigrants of different ethnic backgrounds. The province of Ontario has the large share of people born outside Canada, with the majority settling in the Greater Toronto Area (GTA). The GTA is a metropolitan city in which 45.7% of the population represent immigrants with the highest percentages of immigrants including East Asian (ie, Chinese, Korean) and South Asian (ie, Pakistani, Indian, Sri Lankan).5 Estimating the incidence of elder abuse within these immigrant communities is complicated by factors such as language barriers and differences in what is thought to constitute abuse across cultures.6 7 No national prevalence rates are available for older immigrants in Canada, but the service providers and community leaders who work with us are aware of the existence of elder abuse.

Numerous risk factors for elder abuse have been identified. In their scoping review, Pillemer et al 8 categorised the factors at different levels including the individual victim (eg, functional dependence, poor physical and mental health), the individual perpetrator (eg, substance misuse), the victim-perpetrator relationship (eg, type of relationship), the community (eg, geographic location) and the society (eg, social and cultural norms). Researchers, however, have not systematically explored issues such as the relevance of gender, culture, age, time since immigration and size of immigrant community to the risk of abuse; how individual or micro-level, community or meso-level and society or macro-level factors combine to affect risk; or what strategies can reduce risk in various immigrant communities. Our research, along with that of others (eg, Baars9; George and Chaze10; Guruge et al 11; Lai and Chau12; Tam and Neysmith13), suggests that various risk factors (eg, knowledge of English, social isolation, financial dependency) in the post-migration context can combine to affect elder abuse in complex ways. However, to date most research has been based on small samples representing one specific immigrant community, and lacks a systematic analysis of multi-level factors within and across cultural and immigrant communities.

Nevertheless, past research can provide important points of departure in identifying potential risk factors and strategies to address these factors. Our recent scoping review by Guruge titled Identification, mitigation, and prevention of elder abuse: a scoping review of interventions, of 30 studies (as well as 13 previous reviews including the one by Pillemer et al 8) from Canada and other countries revealed that elder abuse identification and mitigation interventions have primarily involved abused persons, and to a lesser degree, perpetrators, caregivers, and health, social, and settlement service providers. The mitigating interventions included psycho-educational support, community-based case management, legal interventions, social services, home visits by a domestic violence counsellor and police, education about abuse and volunteers providing support and advocacy in the use of the criminal justice system. Measures of client outcomes have included recurrence of abuse, case resolution, relocation, psycho-social outcomes, and knowledge of abuse and awareness of services. Based on our review and those of others,14–20 it is still unclear which elder abuse interventions are most acceptable and effective, in what circumstances, and for which immigrant communities.

Clarifying the factors that contribute to elder abuse and developing strategies to address these requires focusing on: (1) diversity within and between immigrant communities; (2) shared experiences within and across communities and (3) contextual factors beyond individual actions/actors that influence the construction of vulnerabilities and resiliencies. Our study will achieve this by integrating elements of intersectionality21 and an ecological model.22 Use of an intersectionality perspective helps clarify the complexity surrounding multiple elements of social identity (eg, gender, race, class, immigration status), and how these interrelate at various levels of society to contribute to abuse of older immigrant women and men, and thereby inform strategies to address risk factors.21 An ecological model helps clarify how individuals are situated within and influenced by micro-level, meso-level, and macro-level systems, and how victimisation is affected by the dynamic interplay of multi-level influences.21 Together, these two approaches can help identify common and unique factors for elder abuse in immigrant communities. They can also help determine which strategies work for which factors, for which individuals, and in/across which contexts by exploring multiple elements of risk and incorporating multi-level stakeholders’ perspectives.

The study objectives are therefore: (1) to identify key micro-level, meso-level and macro-level factors that contribute to the abuse of older immigrant women and men; (2) to explore culturally relevant strategies to address these risk factors within and across groups of immigrants.

Methods and analysis


This study will be conducted in the GTA, where 63% of the older adult population are immigrants.22 We will target the two largest immigrant communities in the GTA: East Asians and South Asians, who make up 27% and 22%, respectively, of the older adult population in Canada. Within the East Asian community, we will focus on Korean (a relatively recent) community and Chinese (a more established) community. Within the South Asian community, we will focus on the Sri Lankan Tamil (a relatively recent) community and Punjabi (a more established) community. Chinese and Punjabi are the two largest racialised communities in the GTA.22 Korean and Tamil communities are among the fastest growing East and South Asian communities in Canada, and the majority of Korean and Tamil immigrants to Canada have settled in the GTA. An advisory committee will be formed with representatives from each of the four communities to provide overall guidance for the project.


Data will be collected in two phases, through structured group interviews with 6–8 participants in each. These interviews will be run separately with: (1) older women; (2) older men; (3) family members; (4) community leaders and (5) service providers who work with (but are not from) the four communities to maximise comfort and encourage dialogue. Each group interview will follow a semi-structured format. Interviews with older immigrant women and men, as well as with family members of older immigrants, will be conducted in their own language and facilitated by a bilingual, bicultural moderator (Research Assistant) who is intensively trained in group interviewing techniques. Interviews with community leaders and service providers will be conducted in English. Unlike focus groups, group interviews are structured sessions that allow participants to respond to each other’s comments; to question, clarify, and elaborate on ideas; and to reach consensus about collective knowledge.23–25 All group interview sessions will be audio-recorded with informed consent.

Participant recruitment

We will recruit older women, men and family members from the selected communities primarily via referral through our community connections and word of mouth. We will recruit community leaders and service providers using our existing networks and contacts.

The inclusion criteria for older women and men are: community-dwelling (ie, non-institutionalised) older (60+ years) adults (the age criterion is chosen to be consistent with that used in their countries of origin); self-identification as having experienced, or knowing someone who has experienced, elder abuse; immigration to Canada within the last 20 years and current residence in the GTA; self-identification with one of the four selected immigrant communities; and ability to understand and consent to participation (orally or in writing). Purposive sampling will be used to recruit a range of participants in terms of age (‘young-old,’ ‘middle-old’ and ‘old-old’), length of stay in Canada (<5, 6–10, 11–20 years), gender and sponsorship status (self-sponsored; sponsored by spouse or children; refugee).

The inclusion criteria for family members are: current residence in the GTA; self-identification as first-generation (for adult children and their spouses) or first-generation or second-generation (for grandchildren); provision (past or current) of care and support to an older adult who does or does not live in the same household; and ability to understand and consent to participation. Purposive sampling will be used to recruit a diverse group in terms of length of stay in Canada, fluency in English, income, employment, having (or not) children who live at home, having (or not) an older adult living with them and extended family coresidence.

Community leaders include individuals from the selected immigrant communities who take leadership roles in working with, providing various supports for, and/or advocating on behalf of older women and men in their community. Examples of community leaders are: faith leaders, media figures, community advocates, social or settlement workers and healthcare providers. Community leaders will be included if they are from one of the four selected immigrant communities who are identified by our connections as community leaders.

Service providers include social, settlement and health workers who provide services to older immigrant women and men from any of the four communities but do not belong to the four selected immigrant communities. Inclusion criteria are: health, social or settlement service providers who work regularly with, and can speak about, elder abuse within any of the selected communities.

Data collection

Quantitative and qualitative data will be collected in each study phase.

Phase 1

Data collection will focus on identifying factors, occurring at multiple levels, perceived as contributing to elder abuse in the selected immigrant communities. Data will be gathered during group interviews, with a mix of quantitative and qualitative approaches. The quantitative approach involves rating risk factors found to contribute to elder abuse. The list of factors was generated from a synthesis of previous study findings (including those reported by Pillemer et al 8; Baars9; George and Chaze10; Guruge et al 11; Lai and Chau12; Tam and Neysmith13 and our recent scoping review by Guruge titled Identification, mitigation, and prevention of elder abuse: a scoping review of interventions) that is, the risk factors were commonly reported across these scoping reviews and studies. The factors included those at the micro-levels (ie, knowledge of English, financial dependency, physical dependency/disability, emotional dependency), meso-levels (ie, multigenerational coresidence, social isolation) and macro-levels (ie, racism). To help participants rate the factors, we generated a list that clearly labelled each factor and described how the factor contributes to abuse, and we translated the list into the primary language spoken by each immigrant community. We followed phases 3–5 of the integrated method for the cultural adaptation and translation of measures.25 The translation was done by bilingual and bicultural health care professional and community experts, and the translated version of the list was pre-tested for comprehension and linguistic appropriateness with five adults from each immigrant community.

The qualitative approach consisted of engaging participants in a group discussion to further explore the immigrant community’s perspective on the factors. The open-ended questions (eg, how much does the factor align with your knowledge of elder abuse in your community; how do you think it leads to elder abuse in your community) were translated into the respective community’s primary language, reviewed and approved by the study advisory committee.

During the group interviews, the moderator will read the information about the factors included in the list, and then ask participants to individually rate each factor on paper using a 10-point scale, in terms of its importance in contributing to abuse in their own community. Then, the moderator will engage participants in a semi-structured discussion to further explore how each factor contributes to elder abuse in their community. Next, the moderator will ask participants to identify and discuss any additional factors that may contribute significantly to abuse in their community. The group interviews will be audio-recorded with informed consent. For each immigrant community, we will conduct: three group interview sessions with older women; three sessions with older men; one session with daughters and daughters-in-law; one session with sons and sons-in-law; one session with grandchildren and one session each with community leaders and service providers.

Data from phase 1 will be analysed by session, by subgroup (defined in terms of category of participants and gender) and by community. The quantitative ratings obtained from participants in each group interview will be analysed descriptively, providing measures of central tendency and dispersion. Factors with mean ratings >5 (out of 10) will be considered important. The qualitative discussion will be translated, transcribed verbatim and content-analysed26 to reflect the group’s agreement on the cultural relevance of factors. Quantitative and qualitative findings will be integrated across all sessions held with subgroups and immigrant community. The quantitative ratings will be compared across group interview sessions (using appropriate parametric or non-parametric statistics, based on normality of distribution) prior to pooling the data, and mean ratings will be estimated for each subgroup and community. A constant comparison method will be used to determine convergence of coded qualitative data across group interview sessions, for each subgroup. Comparison across communities will be done using analysis of variance for quantitative ratings and a data matrix for qualitative ratings to delineate common and unique factors. Any factors that emerge as important from both quantitative and qualitative data analyses will be identified and integrated into phase 2.

Phase 2

Data collection will use the same approach and steps to generate a list of strategies or interventions that could be applied to address the factors contributing to elder abuse within and across the four immigrant communities. Specifically, we will synthesise the findings of the scoping reviews done by Pillemer et al 8 and our team (Guruge, Identification, mitigation, and prevention of elder abuse: a scoping review of interventions) to determine strategies found promising in preventing and/or managing elder abuse. The strategies include those aimed at identifying abuse (eg, healthcare professionals’ screening for abuse), mitigating abuse (eg, psychological therapies to reduce distress among victims, emergency shelter) and preventing abuse (eg, education and anger management targeting perpetrators, empowerment of older adults). We will follow the process for cultural adaptation of interventions, described by our team (Sidani et al 25) to involve participants in a mapping exercise to determine the desirability, acceptability, and cultural relevance27 of each strategy within and across immigrant communities, as it pertains to each factor identified in phase 1. The same five stakeholder groups in each immigrant community will be included in the mapping exercise.

Participants in phase 2 will consist of those who participated in phase 1 and signed a ‘willingness to be re-contacted’ form. We expect only about a 8–9-month gap between the two phases of data collection. Because this population is not highly mobile, we do not anticipate much difficulty in recruiting participants who were involved in phase 1. However, we are aware that there may be some attrition due to illness/death, changes in housing and so on, which may require us to recruit additional participants using the eligibility criteria and recruitment strategies specified for phase 1.

Procedures for the group interviews will be similar to those used in phase 1, with minor modifications. Prior to the group interviews, we will generate the list of strategies reported as promising in addressing each factor confirmed/revealed in phase 1 as important in contributing to elder abuse within and across specific subgroups. We will develop a description for each strategy to clarify its goal, components, mode and dose of delivery.24 This list of strategies and their descriptions will be translated for and pilot-tested in each immigrant community as explained previously. During group interviews, the moderator will read the description of each strategy and invite participants to individually rate it (using a 10-point scale) to determine the extent to which they perceived it as acceptable. Next, the moderator will involve the group in a discussion that will focus on reaching consensus or majority agreement (>75%) about which strategies are most desirable in addressing each factor contributing to elder abuse, which strategies are consistent with the immigrant community’s beliefs and values, and which aspects of the strategy require modification (and how they should be modified) to improve their cultural relevance. Finally, the moderator will ask participants about strategies that could be important to their community but are missing from the list.

Phase 2 data analysis will also follow a process similar that used in phase 1: quantitative and qualitative data will be analysed separately at the level of group interviews and subgroups to identify strategies that are acceptable and applicable to all or certain subgroups (eg, gender/immigrant community). The results will be incorporated into a matrix linking factors and respective strategies, which will serve as the foundation for developing and tailoring programmes for preventing abuse at the immigrant community and individual levels. Analysis of quantitative ratings will include: (1) descriptive statistics at the particular group interview level, providing measures of central tendency and of dispersion pointing to strategies rated as acceptable (mean rating >5 out of 10); (2) descriptive statistics to examine mean ratings of the strategies for the following subgroups (estimated across sessions with the same subgroup): older men and older women in each immigrant community, family members, informal and formal community leaders, and service providers, as well as parametric statistics (independent sample t-test or analysis of variance, based on the level of between-subject factor) to explore differences in ratings by gender, community, and subgroup and (3) mixed linear models to compare ratings while accounting for intra-group correlation. The latter analysis will be applied to: (a) subgroups of older men and older women of the same community and (b) the target population, where variability in ratings of cultural relevance and acceptability of strategies will be examined by subgroup, gender and community. Qualitative data will be content-analysed28 29 to reflect subgroup agreement about the cultural relevance and acceptability of the strategies, which will be summarised in a matrix linking factors and strategies. For each factor that may be addressed by a strategy, the groups that have judged the strategy as acceptable and relevant to that factor will be listed in the intersecting cell. The results of the mixed linear models will be integrated into the matrix to delineate linkages among factors and strategies reported as relevant (as evidenced by the convergence of quantitative and qualitative findings) to various and/or all subgroups.

Patient and public involvement

Patients were not involved in the development of this protocol. Community members were invited to join the research team and informed the development of the questions that will be used to collect data from participants in group interviews.


Participants will be informed that their participation is voluntary, that they may choose to withdraw from the study at any time, and that their personal identity and information will remain confidential. Participants will review and sign informed consent, in English or their own language (based on their preference), prior to participating. During the recruitment and consent process, they will be advised that they will be asked about their personal experience with elder abuse only in the written (individual) survey, and that they are not required to respond to any question. They will also be advised that during discussion groups, they will not be asked to share any personal experiences with elder abuse. Additionally, if any participant appears troubled, the moderator will provide supportive listening and information on how to access suitable agencies/services as needed. Participants will have the option to leave the group discussion at any time for any reason.


We will work closely with community leaders and service providers to package evidence in a format that meets their unique needs. We will prepare plain-language summaries for ethnic newspapers, brochures to be distributed through local libraries and community services, and key messages for radio call-in shows and TV channels. The project website will make information (such as plain language summaries, Powerpoint presentations) available to all stakeholders. Alternative formats such as webinars, podcasts, ‘lunch and learn’ forums at research and community centres, and university-based forums will also be pursued. Publications will include papers related to methods, results and emerging theory. We will also share the findings through papers, posters and workshops at refereed conferences. Policy dialogues will be planned to bring together other key stakeholders for a guided, interactive discussion of the findings, which will be summarised in policy briefs.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.


  • Contributors SG: Principal investigator of the project: leading the project upon which this protocol paper is based and drafted the paper. SS: Methodological expertise to the project; drafted the methods section of the project application upon which this paper is based; provided critical revisions on an earlier version of this paper. AM: Co-investigator, provides expertise in immigrants and elder abuse and provided feedback on the project application, revisions and finalising the survey tools. GM: Co-investigator, provided feedback on the project application upon which this paper is based; reviewed and provided critical feedback on this paper. DP: Expertise in the area of gerontological nursing; reviewed and provided critical feedback on an earlier version of this paper. All authors revised and edited the manuscript critically for important intellectual content of the material. All authors approved the final version of the manuscript.

  • Funding This project was reviewed and awarded funding by the Social Sciences and Humanities Research Council for the years 2016–2020.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval This study has received approval from the Research Ethics Boards of Ryerson and York Universities in Toronto, Ontario.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Author note Actual project work can be accessed at: