Elder mistreatment in underserved populations: Opportunities and challenges to developing a contemporary program of research

Abstract

This article provides an overview of the status of research on elder mistreatment among underserved populations in the United States, including gaps in our current knowledge base, scientific and structural barriers to growing research on the exploitation, neglect, and abuse of older people from diverse and disadvantaged ethnic/racial, geographic, sexual identity, and socioeconomic groups. High priority areas in need of new elder mistreatment research with underserved populations are identified, and suggestions are given for how this research can be facilitated by researchers, university institutional review boards, and funding agencies.

Introduction

While the majority of empirical research on the mistreatment of older people in the United States focuses on the general population, disadvantaged status due to low income or membership in some ethnic minority groups may in fact put elders at higher risk of mistreatment (Burnes et al., 2015; Laumann, Leitsch, & Waite, 2008; Pillemer, Burnes, Riffin, & Lachs, 2016). Both population aging, as well as the current and projected growth of the U.S. minority population, call for increased attention to the unique circumstances in which these groups find themselves as they get older. This article explores the state of the field with respect to the mistreatment of underserved elders, with the primary goal of identifying scientific and structural barriers that have slowed progress as well as offering recommendations that can advance the field. In doing so, we assess both the general state of knowledge and identify major gaps in knowledge. We then describe approaches that have either been recommended by members of underserved communities themselves or that have been found efficacious in working with these communities. Following this, major structural impediments to developing a program of research on elder mistreatment with underserved populations are outlined. We identify high priority areas that can move the field further, and describe ways in which policymakers can facilitate research. “Underserved” populations are heterogeneously defined within the literature, but the term generally refers to the medically underserved and has been formally defined by the US Department of Health and Human Services as including “Low income populations (defined to include all those with incomes below 200% of the poverty level); Medicaid-eligible populations; linguistically isolated populations; migrant and seasonal farmworkers and their families; homeless populations; residents of public housing; and Native Americans” (Department of Health and Human Services, 2008). Within the larger literature, the term underserved often encompasses other groups believed to be insufficiently served by the health care system, including racial and ethnic minorities, uninsured persons, immigrants, elders, rural populations, persons living in primary care health professional shortage areas, and/or those with various community-wide vulnerabilities (Grumbach, Hart, Mertz, Coffman, & Palazzo, 2003; Silow-Carroll, Alteras, & Stepnick, 2006). The question of exactly who is medically underserved—and how being underserved relates to elder mistreatment—has proven to be an important area of research in and of itself. It could be argued, for instance, that virtually all elders are inadequately served by medical systems with regard to the issue of elder mistreatment. At minimum, the term seems to gesture toward a presumed difference from the standard normative culture bearer (ostensibly a white, native-born, middle-class, heterosexual, able-bodied, urban/suburban, non-elderly male with access to health care). A variation from this “norm” signals the potential for disadvantage or vulnerability.

Methods

Cognizant of the fact that most elder mistreatment research continues to focus on the general population, we conducted an exploratory survey of the extant literature on elder mistreatment among select underserved groups that we predicted would likely be “underrepresented” in empirical research. These groups include the largest US ethnic/racial groups, rural elders, and low SES elders. We also included lesbian, gay, bisexual, and transgender (LGBT) elders, who while not necessarily included within standard definitions of underserved populations, clearly fit the criteria of having unique characteristics that might increase their vulnerability to mistreatment. What follows is a thematic profile of the literature on elder mistreatment among underserved populations (a quantitative meta-analysis of this literature is currently in preparation) (Jervis & Williams, in preparation). We combine these findings with insights derived upon reflection on a Breakout Session at the NIH’s “Multiple Approaches to Understanding and Preventing Elder Abuse and Mistreatment” Workshop in which several of the authors participated (Jervis & Beach, 2015).

The literature review of underserved populations and elder mistreatment within the United States was conducted by searching Google Scholar and PubMed for peer-reviewed articles in English using the following terms: Elder mistreatment, elder abuse, elder neglect; Black; African American; American Indian; Alaska Native; Asian American; Hispanic; Latino; Mexican American; LGBT; LGBTQ; gay; lesbian; low socioeconomic status; low SES; low income; poverty; poor; and rural. Ninety five articles were reviewed that fit these criteria; of these, 53 articles were determined to describe original empirical research. We excluded overview articles and practitioner-oriented reports that were not based on original empirical research.

Elder Mistreatment and Underserved Groups: What is known

Using these search parameters, it became apparent that a predominant thematic emphasis in the extant literature was ascertaining how various ethnic/racial groups understood mistreatment. The bulk of research attention on this topicwas devoted to two groups, Asian American elders (Dong, 2014; Le, 1998; Lee & Shin, 2010; Lee et al., 2011; Lee, Kaplan, & Perez-Stable, 2014; Moon et al., 1998; Moon, Tomita, & Jung-Kamei, 2002; Pablo & Braun, 1998; Yan & Tang, 2001) and African Americans (Benton, 1999; Hudson & Carlson, 1999; Shellman, 2004; Tauriac & Scruggs, 2006). Only a few empirically based research articles on the conceptualization of elder mistreatment had a primary emphasis on Latino (Franco, Gray, Gregware, & Meyer, 2000; Mitchell, Festa, Francis, Juarez, & Lamb, 1999; Paulino, 1998) or American Indian/Alaska Native elders (Hudson, Armachain, Beasley, & Carlson, 1998; Jervis, 2014). Methodologically, articles on this topic leaned toward quantitative methods, although qualitative research was also well-represented.

A number of articles investigated how differing support systems influenced elder mistreatment risk factors among ethnic/racial groups, again with a primary focus on Asian American and African American elders (Benton, 1999; Brown, 1989; Dimah & Dimah, 2002; Dong, Chang, Wong, & Simon, 2013; Dong, Chen, Fulmer, & Simon, 2014; Griffin & Williams, 1992; Le, 1998; Moon & Benton, 2000; Moon & Evans-Campbell, 2000; Paranjape, Sprauve-Holmes, Gaughan, & Kaslow, 2009; Paulino, 1998; Shellman, 2004; Strumpf, Glicksman, Goldberg-Glen, Fox, & Logue, 2001). A relatively small literature examined prevalence of mistreatment among ethnic/racial groups, either as part of a general population prevalence study (Acierno et al., 2010; Burnes et al., 2015; Laumann et al., 2008) or in ethnic/racially specific samples (Beach, Schulz, Castle, & Rosen, 2010; Buchwald, Tomita, Ashton, Furman, & Manson, 2000; DeLiema, Gassoumis, Homeier, & Wilber, 2012; Dong et al., 2014). A very small number of articles focused on elder mistreatment measurement (Jervis, Fickenscher, & Beals, 2014) or interventions in American Indian communities (Holkup, Salois, Tripp-Reimer, & Weinert, 2007).

While a hallmark feature of underserved populations is poverty (Braveman & Gruskin, 2003; U.S. Department of Health and Human Services, 2008), research-based articles with a targeted focus on low SES and elder mistreatment were essentially non-existent. Only a few empirically based articles had a primary emphasis on elder mistreatment among rural (Begle et al., 2010; Dimah & Dimah, 2004) or LGBT elders (D’Augelli & Grossman, 2001). While rural and low-SES groups were seldom the sole focus of a study, rural or low income issues were sometimes discussed as part of studies of specific ethnic/racial groups.

Culture, Race/Ethnicity, and Elder Mistreatment

Based on their review of international prevalence studies on elder abuse, Pillemer and colleagues conclude that “Culturally specific forms of elder abuse and cultural attitudes toward prevention and treatment (including barriers) remain virtually unexplored” (Pillemer et al., 2016, p. S202). Our literature search, as described above, found that cultural understandings of abuse, on the contrary, received a fair amount of attention within the literature on some US underserved groups, although few of these studies examined implications for prevention and treatment nor were they prevalence studies per se. One population-representative prevalence study found virtually no race or ethnicity-based differences in types of mistreatment (Hernandez-Tejada, Amstadter, Muzzy, & Acierno, 2013). This study did not explore, however, whether some groups might be vulnerable to unique forms of abuse related to their group identity (e.g., spiritual abuse among Native Americans (Jervis, 2014) or the “outing” of an older LGBT elder against their will). Further, the study utilized self-report via interviews on home telephones, which could be particularly problematic for some underserved groups without home telephones or who would not be comfortable disclosing this information to an outsider.

Notions regarding what constitutes the proper and improper treatment of older adults may differ culturally (Jervis, 2014). Additionally, different ethnic/racial groups have been found to rate the severity of abuse scenarios differently (Hudson & Carlson, 1999). Understanding how groups conceptualize abuse is imperative for developing culturally relevant and valid measures, as well as for operationalizing culturally competent adult protection, policing, and criminal justice systems. While proper attention to culture is crucial, it is equally important to avoid an overreliance on culture as the default explanation for behavioral differences, as more sophisticated analyses may reveal that other factors, such as low SES, are responsible for actions that are too easily attributed to culture (Kleinman & Benson, 2006). However, while race and SES are frequently conflated, it is important to be aware that they are not equivalent (Hamby, 2015; Steinberg & Fletcher, 1998).

Other factors may also contribute to increased vulnerability to abuse. While less studied in the area of elder abuse, ample research indicates that criminal justice, health care, and social services responses to violence and abuse are not distributed equally across the population (Binswanger, Redmond, Steiner, & Hicks, 2012; Donnelly, Cook, & Wilson, 1999; Hamby, 2008, 2009; Hamby, 2015; Walker, Spohn, & DeLone, 2011). The burden of discrimination and microaggressions, which are subtle comments or actions that, intentionally or not, reinforce stereotypes, might leave certain categories of elders (e.g., the impoverished and some ethnic minorities) more vulnerable to victimization. Theoretically, the lingering effects of racism and poverty related to the legacy of slavery (Cross, 1998) and/or historical trauma in communities subjected to the disruptions of colonialism/post-colonialism (Braveheart-Jordan & DeBruyn, 1995; Jervis & AI-SUPERPFP Team, 2009; Whitbeck, Adams, Hoyt, & Chen, 2004) could also compound the challenges of dealing with elder mistreatment. And a poverty-related lack of social capital certainly imposes challenges with regard to navigating health care and social service systems (Bourdieu & Wacquant, 1992).

A closer examination of the characteristics of different groups can yield insights not only into challenges but also protective factors (Hamby, 2015). These may include extended family networks, multigenerational households, cultural identity and values, and spirituality which can alternatively be sources of strength or challenges (Roman, Jervis, & Manson, 2012). In addition to SES and rurality, other demographic characteristics can vary across groups and can impact access to services, (e.g., language barriers and cultural norms regarding help seeking). A true understanding of group differences in elder mistreatment will require going beyond demographic characteristics as mere “social addresses” (Bronfenbrenner, 1986) and understanding the factors and processes that produce these differences.

Elder Mistreatment and Underserved Groups: Knowledge Gaps

Numerous knowledge gaps exist for underserved persons with respect to elder mistreatment. In addition to the aforementioned issue of determining the state of elder abuse services for all elders, other areas await investigation. While general population prevalence studies have been conducted that include major US ethnic minorities (Acierno et al. 2010), these should not necessarily be taken as the final word due to the many methodological difficulties incumbent in conducting culturally valid research with and across ethnically diverse populations (Greenfield, 1997; Peng, Nisbett, & Wong, 1997; Solano-Flores & Nelson-Barber, 2001). Other studies that employ different methods and/or that utilized culture members as part of the research team have obtained considerably different results (DeLiema et al. 2012; Buchwald et al. 2000), raising questions about how best to understand the findings of the respective studies. Further, general population studies on elder mistreatment tend to treat SES and rural residence as variables, rather than viewing low-income and rural groups as target populations that experience unique issues deserving of attention in and of themselves.

Risk factors are of special concern with respect to elder mistreatment. Many risk and protective factors are unclear for US underserved populations. A recent review article (Pillemer et al., 2016) suggested that race/ethnicity may be a potential risk factor for elder abuse, citing increased risk for financial and psychological abuse for African American elders (Beach et al., 2010) and decreased risk of emotional and financial abuse and well as neglect among Latino elders (Burnes et al., 2015; Laumann et al., 2008). Low income/SES emerged as a strong risk factor for neglect (Acierno et al., 2010; Pillemer et al., 2016). It is interesting that several population-based studies find significant race/ethnicity differences even after statistically controlling for common sociodemographic, social, physical, and mental health correlates in multivariate models (Beach et al., 2010; Burnes et al., 2015; Laumann et al., 2008). Thus, the challenge is explaining why these race/ethnicity differences in risk for elder abuse remain – what are the causal mechanisms and mediators? The question is “What are we not measuring or capturing about the experiences of underserved populations that relates to increased/decreased risk for abuse?” We suggest here that race/ethnicity per se is not what puts an older adult at risk. Rather, race and ethnicity are markers for the effects of discrimination, cultural factors, and other factors that might operate as contributors. Tracing these mechanisms out would be an important area for future research to investigate. To more directly investigate these mechanisms, actual measures of cultural values and racial, ethnic and cultural identity could be included. Research would also benefit from measures of exposure to discrimination and micro-aggressions, including considering those as part of the lifetime burden of victimization. Research is needed on health disparities and the ways in which poor medical care and poor access to medical care can increase the vulnerabilities of elders and potentially contribute to a cycle of violence. More research is also urgently needed on whether there is differential effectiveness of prevention and intervention programs across racial and ethnic groups. For this type of research, sufficient numbers of people from diverse communities should be enrolled to offer enough statistical power to test for group differences. Research attention also needs to examine factors other than SES that might vary across racial and ethnic groups and contribute to vulnerability to elder abuse or the effectiveness of programming. For example, some types of prevention might be more effective for people living in multi-generational households versus elders living alone, since racial and ethnic minorities are more likely to live in such households (Fry & Passel, 2014).

There is clearly room for more in-depth examination of elder mistreatment risk and protective factors in underserved populations. Important research questions include whether unique mistreatment risk and protective factors exist among the various underserved populations, whether existing findings regarding risk and protective factors in other populations apply to underserved groups, or whether new theories need to be developed to better account for the unique dynamics that occur in some of these environments/communities. In some cases, a widening of the existing net of research to encompass more underserved communities without the development of completely new theories is likely sufficient. For instance, some risk factors have been consistently found to put elders at greater risk of abuse (e.g., functional dependence/disability or cognitive impairment among victims; substance abuse or mental illness among perpetrators) (Pillemer et al., 2016) likely have widespread applicability among underserved populations as well as the general population. In other cases, further investigation is needed to understand the unique cultural and social dynamics that put particular groups at risk.

In addition, we lack sufficient data for most—if not all—underserved populations on how best to tailor prevention and intervention programs to best fit their needs. This is a crucial area, in that there is no reason to assume that a program that works for the general population will work unaltered in an underserved population. Baseline data is needed to develop prevention and intervention programs for target communities in order to increase the likelihood of success.

Approaches to Research on Elder Mistreatment with Underserved Groups

Addressing research gaps in underserved populations requires different approaches with respect to establishing working relationships with communities, involving target groups as well as stakeholders at multiple levels (e.g. health care providers, protective services, community or faith-based organizations or leaders, family members and so forth) in the development of research design and protocols, securing study approvals, recruiting participants, project management, and dissemination of findings. These are essentially the principles advocated by Community Based Participatory Research (CBPR) (Minkler, Blackwell, Thompson, & Tamir, 2003; Wallerstein & Duran, 2006), an approach that has been quite fruitful within the health disparities field, and that has been utilized in elder mistreatment research with American Indians (Jervis et al., 2010; Jervis & the Shielding American Indian Elders Project Team, 2009). CBPR promotes respectful relationships with research communities, advocating community involvement from the inception of a project through the dissemination phase (Jervis & the Shielding American Indian Elders Project Team, 2009). Some have gone a step further and suggested a rigorous self-critical examination of the power and privilege dynamics within the research team itself, as well as the consideration of adopting a social justice orientation (Muhammad et al., 2015). Some authors have even compared the researcher’s relationship to participating communities in a CBPR context to a spiritual covenant (Salois, Holkup, Tripp-Reimer, & Weinert, 2006).

While CBPR has been extremely important in conducting more community oriented research with health disparities populations, there is not a one-size-fits-all approach that can be plugged into every research setting. Communities, even those that share the same demographic “label,” are complex entities. Learning how to work with particular communities in their uniqueness in order to ensure that research aligns with local needs and values will prove infinitely more useful than relying on stereotypes about groups based on their cultural, ethnic/racial, geographic, or sexual identity. It is important to ascertain the degree of community involvement that makes sense for the group in question and the project at hand. Some communities and projects call for a very high degree of collaboration and others for lesser. Regardless, when working with underserved communities, there is always a need for researchers to establish and earn trust with participating communities. CBPR, at whatever level employed, must be meaningful, not just a token effort.

Conducting research with underserved communities often requires a reorientation in not only how researchers work with participating communities but also in how they carry out the practice of research itself. It often becomes clear early in the process that the dearth of knowledge regarding elder mistreatment in underserved contexts means that exploratory, developmental, and conceptual work is needed in order to understand the unique dynamics at work in these communities. An exclusively quantitative methodology employed right from the start may be doomed to failure from a cultural validity standpoint, and in some settings, from a feasibility perspective as well. In underserved communities, qualitative research is often a crucial first step in terms of even knowing what questions to ask as well as figuring out how to ask them. Likewise, in the later stages of a research agenda, when one is developing prevention programs and interventions, it is crucial to have a thorough understanding of the complexities of how elder mistreatment is embedded in the social fabric and cultural norms of a communities (Jervis, 2014), as well as what messaging makes sense for the local context. This is a fundamentally qualitative enterprise.

Qualitative research is a vital part of a mixed methods project in terms of hypothesis generation, measurement development, improved cultural validity of quantitative research, and triangulating findings (Creswell & Plano Clark, 2010; DeWalt, Rothrock, Yount, & Stone, 2007; Tashakkori & Teddlie, 2010). Qualitative research, however, should not be understood as merely a handmaiden to quantitative research, but as a diverse set of research traditions, including varied methods such as ethnography, grounded theory, and focus groups (Agar, 1996; Bernard, 2012; Bryman, 2012; Creswell, 1998; Denzin & Lincoln, 2000; Glaser & Strauss, 1967; Guba & Lincoln, 1994; Guest & MacQueen, 2008; Krueger & Casey, 2008; Seale, Giampietro, Gubrium, & Silverman, 2004; Spradley, 1979; Strauss & Corbin, 1990) that have merit as stand-alone research. Moreover, with an orientation toward inductive reasoning and hypothesis generation, alternative criteria for evaluating the quality of qualitative research have been proposed (Guba & Lincoln, 1994; Sandelowski, 1986; Sandelowski, 1993).

Scientific and Structural Barriers to Developing an Evidence-Base for Elder Mistreatment with Underserved Populations

One of the greatest scientific barriers to empirical research on mistreatment with underserved elders has been a lack of measures with demonstrated cultural validity either in specific underserved populations or with proven reliability and validity across groups. This barrier to valid measurement stifles important efforts such as prevalence studies and intervention research. Although the lack of valid measures primarily impacts quantitative research, ironically it results from a neglect of qualitative research by both the larger scientific community and major funders. As previously detailed, many unanswered questions regarding elder mistreatment among underserved populations are best answered via qualitative inquiry—such as the language and concepts used to describe mistreatment in specific locales; the meaning of forms of abuse in various underserved populations; how mistreatment functions in dynamic community contexts; and how messaging around elder abuse is understood and responded to on the local level. Valid measurement requires valid constructs for the communities in question. These are fundamentally questions of meaning that call for qualitative methods. However, as these constructs are integral to valid measurement, the fact that qualitative research has historically been undervalued and underfunded in biomedical research limits opportunities to develop culturally valid measures for underserved populations by harnessing insights derived from these studies.

A major structural barrier encountered by the field are university Institutional Review Boards (IRBs). Some IRBs have exhibited extreme caution when considering studies on abuse in what they perceive to be a “protected class” (i.e., elderly). When compounded with ethnic minority status and/or low SES, this extreme caution has sometimes escalated to the point of completely stopping research. In an attempt to protect participants deemed vulnerable, IRBs have been known to demand numerous protocol revisions to the extent that projects are unduly delayed or that scientific validity may be compromised. Further, a prospective researcher’s fear that a study may not make it through an IRB could be enough to keep one from proposing a study with an underserved focus. In addition, IRB-related obstacles to research have a career impact on investigators required to show research productivity for tenure and promotion in academia. Given the primary barrier to entry the general field of elder abuse is already facing—ageism—undue barriers posed by IRBs is an obstacle that the field cannot afford. It is unclear how proposed changes to subpart A of 45 CFR part 46, known as the “Common Rule” for the Protection of Human Subjects, will impact studies on elder mistreatment with the underserved (Federal Register Tuesday, September 8, 2015). While changes to federal regulations could well lead to increased scrutiny by IRBs, it is also possible that IRBs could use these changes as an opportunity to develop more sophisticated practices around evaluating risk that could benefit the elder mistreatment field. Studies of elder mistreatment are not equally risky, and individuals should not automatically be assumed to need increased human subjects protections simply due to being an older adult.

Another major structural barrier to elder mistreatment research is lack of policy attention and funding devoted to this area. Indeed, there are large federal funding disparities in types of family violence research, with elder mistreatment receiving substantially less annually (e.g., child abuse - $6 Billion; intimate partner violence - $249 Million; and elder abuse -$19 million (Grantmakers in Aging, 2016). While historically, there has been a relative lack of policy attention to elder abuse, attention to elder justice issues during the 2015 White Conference on Aging (https://whitehouseconferenceonaging.gov/, the subsequent NIH workshop, “Multiple Approaches to Understanding and Preventing Elder Abuse and Mistreatment” (Saylor, 2016), and the Elder Justice Roadmap initiative (Connolly, Brandl, & Breckman, 2014) have raised awareness and attention from policy makers.

High Priority Elder Mistreatment Research Areas for Underserved Populations

While some leaders in the general elder mistreatment field are urging the field to move toward interventions (https://videocast.nih.gov/summary.asp?Live=17290&bhcp=1), it is clear that many underserved groups continue to need baseline research. Interventions that aren’t grounded in foundational and/or developmental research are premature and stand a much lower chance of success. Sound measures that are in accord with a group’s linguistic and cultural norms (Greenfield, 1997) are crucial in order to conduct the epidemiological research that is still needed for many underserved groups. For instance, it is unlikely that we have culturally valid and population-representative prevalence figures for a number of groups (e.g., American Indian/Alaska Natives) given the well-known issues related to sampling, recruitment, and cultural validity in this population (Beals, Manson, Mitchell, Spicer, & Team, 2003; Jervis, 2014). We also do not have an empirically based picture of risk factors for many underserved communities. One area that would benefit the field as a whole would be examining how elder mistreatment is intertwined in the social fabric of families and communities, how this embeddedness functions to perpetuate or inhibit mistreatment, and how this knowledge can help one to design interventions that are more likely to meet with success.

As research with underserved populations proceeds, it would do well to incorporate a longitudinal perspective in order to more effectively understand risk factors, causal mechanisms, and the “natural history” or “trajectory” of elder abuse in various underserved populations. What are the implications for potential customization of interventions (Dong, 2014)? One possibility in this regard is to build on existing longitudinal samples that include ethnic minorities or other underserved populations. Another high priority area over the longer term is the acquisition of knowledge on how underserved groups interface with the social service, legal, and justice systems, and the determination of which populations—if any—are sufficiently served. Which groups are in need of better, more, or different responses to elder mistreatment, and how can they be better served?

As this effort proceeds, it is important that those who fund research understand the ways in which those who want to do this work are facilitated and impeded. First and foremost, funding agencies—and those who review grant applications—must recognize the value of qualitative research both as a stand-alone method and as part of mixed methods projects. Not judging qualitative methods by the standards of quantitative research, but by its own standards (Guba & Lincoln, 1994) will be crucial to this effort. This, however, involves ensuring that a sufficient number of reviewers are included on panels who understand qualitative research and are aware of the existence of alternative criteria for its evaluation. Questions arise as to how to build qualitative research capacity within the scientific workforce and within funding agencies, as well as ensuring that reviewers are equipped to fairly review qualitative proposals on elder mistreatment. One approach to this problem is for new and early stage investigators with qualitative research training and experience to become involved with the Early Career Reviewer Program sponsored by the Center for Scientific Review at the NIH and for experienced qualitative researchers to submit Curriculum Vitae to Program and Scientific Review Officers for consideration as reviewers. Selection of reviewers depends on the scientific expertise demanded by the applications to be reviewed; however, the experience of being a reviewer provides benefits for both the scientist and the NIH. For the NIH, experienced qualitative researchers who serve as scientific reviewers help to disseminate information and understanding of the value of qualitative methods to deepening understanding and evaluation of complex phenomena. For the scientist/reviewer, the act of reviewing the work of others facilitates an educational process in the desired information on research methods needed for a successful application. Both are essential to support the development of a robust appreciation of elder mistreatment and to address any data gaps. These processes are likely similar at other funding agencies as well, including but not limited to the National Institute of Justice (NIJ), the Centers for Disease Control (CDC), the Health Resources and Services Administration (HRSA), the Agency for Healthcare Research and Quality (AHRQ).

With respect to quantitative research, rigorous methodological studies are needed that test elder abuse assessment and screening tools for measurement invariance or similar item functioning across ethnic/cultural groups. As noted above, these quantitative measurement studies should build on prior qualitative work that informs conceptual development and item wording. The major challenge is developing measures of elder abuse that are both universally culturally/ethnically valid but also standardized to allow cross-cultural quantitative comparisons. Given the importance of CBPR in research with underserved communities, it is important that community engagement is built into grant applications as well as the review processes. NIH has paved the way with a number of collaborative efforts to support CBPR, including funding opportunities which include a special CBPR study section (National Institutes of Health, 2016). NIMHD, more specifically, has over the past years issued a series of funding announcements within a Community Based Participatory Research initiative to provide support for community engagement in the research process (3 years), implementation of the research project (5 years) and dissemination of research findings (3 years). Another recommendation is to establish an interagency working group that funds capacity building efforts. The Native American Research Centers for Health (NARCH) includes elements that could serve as an example for such a working group. This funding initiative developed by the Indian Health Service (IHS) in collaboration with NIH Institutes and Centers is now managed by the National Institute of General Medical Sciences in conjunction with the other Institutes/Centers of the National Institutes of Health (NIH). The purpose of the Native American Research Centers for Health (NARCH) initiative is to reduce health disparities, support health research projects prioritized by tribal communities, enhance health research partnerships, and reduce distrust of research by American Indian/Alaska Native (AI/AN) communities while developing a cadre of AI/AN scientists and health research professionals. By empowering tribes to set their own priorities for research as well as develop capacity to carry it out, NARCH instantiates many of the best principles of a CPBR approach.

Conclusions

Underserved populations have historically been underrepresented in elder mistreatment research. Much of the extant research has focused on examining conceptions of mistreatment among Asian American and African American elders, with a smaller body of literature focusing on risk factors in these same groups. It is possible that including overview, policy-based, and practitioner-oriented articles in our literature review would have yielded a wider range of topics and target groups. Further, with its broad scope, our literature review may not have identified every article relevant to elder mistreatment among underserved groups. Conceptual work continues to be crucial for underserved populations, as culturally valid measurement, identification, and interventions rest upon an accurate understanding of a group’s perception of a phenomenon. It is clear, however, that a) this conceptual work is not yet completed and/or has not been conducted for all underserved groups, and, even more importantly, that b) this is not the only kind of developmental work that needs to be undertaken. Moreover, it is not clear that the conceptual work that has been undertaken to date has been fully incorporated into other research in the field (e.g., epidemiological studies, risk factors, etc.). In the elder mistreatment field, as in many others, quantitative and qualitative researchers tend to operate in silos. Further, the existing research has not sufficiently explored the role of low SES, rurality, or sexual identity in relation to elder mistreatment. This article has described a need to widen the net beyond cultural perceptions of abuse, to examine other areas, such as low-SES, rurality, inadequate access to health care, and the burden of discrimination. Further, sexual orientations other than heterosexuality can be expected to increasingly come into focus in terms of the older demographic—including abused, neglected, and exploited persons. As this research moves forward, it needs to be mindful of how to employ community-engaged approaches that make sense for the setting in question, and that ensure meaningful (rather than tokenized) inclusion of underserved research populations. In order to ensure success, we need to find solutions to the bureaucratic obstacles imposed by university IRBs as well as scientific obstacles imposed by paradigmatic biases within the research and funding communities.