An Evaluation of AoA's Program to Prevent Elder Abuse: Final Report. EXECUTIVE SUMMARY


A. Background

The Elder Abuse Prevention Interventions demonstration, authorized by the Elder Justice Act and funded by the Administration on Aging (AoA), U.S. Department of Health and Human Services (HHS) in FY 2013, provided funding to test interventions designed to prevent elder abuse, neglect, and exploitation. The Elder Abuse Prevention Interventions program provided $5.5 million to five states and three Tribes.

The HHS Office of the Assistant Secretary for Planning and Evaluation (ASPE) contracted with NORC at the University of Chicago to design and conduct an evaluation of the interventions being tested through this demonstration. The purpose of the evaluation was to study the development and implementation of the state grantees' elder abuse interventions and report findings on the characteristics of victims and perpetrators of elder abuse or those at-risk, the use of prevention services, and outcomes. Awards for the five states ranged from $625,000-$1,020,000 for a three-year period. The five grantees funded by AoA were:

  • Alaska Division of Senior and Disabilities Services (AK DSDS)--Using a Critical Time Intervention Approach for Elder Services Case Management

  • New York State Office for the Aging (NYSOFA)--Using Enhanced Multi-Disciplinary Teams to Address Financial Exploitation

  • Texas Department of Family and Protective Services and the WellMed Charitable Foundation (TX/WellMed)--Implementing Elder Abuse Screening and Embedding APS Specialists in Clinical Settings

  • University of Southern California (USC)--Take AIM against Elder Abuse: The Abuse Intervention Model

  • University of Texas Health Science Center (UTHSC)--The Self-management of Medication of Independent Living Elders who Self-Neglect (SMILES) Study

All pilot projects shared common goals and requirements, including: (1) the design of a selective and/or indicated preventive intervention; (2) targeting of 1-3 categories of people at high risk of elder abuse; (3) the establishment of key stakeholder partnerships; (4) provision of local and state-level Adult Protective Services (APS) administrative data; and (5) agreement to collect a core set of data elements. Beyond these five objectives, grantees were afforded broad discretion in developing prevention interventions tailored to their specific communities and contexts.

Collectively, the interventions included the development and/or use of various screening and assessment tools, time-limited case management, tailored health promotion, enhanced multi-disciplinary teams (E-MDTs), improved coordination of referral and care, projects supported by multiple and diverse partnerships, and provision of education and training to a variety of target audiences (e.g., clients, clinicians, professionals, communities of interest). All projects were directly responsible for developing and customizing care plans. However, some projects directly administered those services to clients, whereas others either coordinated existing services or provided a combination of the two. Pilot projects were also characterized by their heterogeneity, including a focus on one type of abuse or potentially all forms, implementation in a variety of settings (primary care, APS, multi-disciplinary teams, etc.) and geographic areas (urban and rural), as well as assorted recruitment strategies or points of entry.

B. Objectives (Research Questions)

The evaluation of the five state cooperative agreements awarded by AoA's Elder Abuse Prevention Interventions program is an important element in building the evidence base on effective approaches to prevent elder abuse and enhancing existing data collection systems. The research questions of interest to ASPE and AoA were:

  1. What is the infrastructure within which the interventions rest and the structure of elder abuse prevention interventions?

  2. What are the facilitators of and barriers to implementation of the interventions and how are barriers addressed?

  3. What are the characteristics of victims and perpetrators of elder abuse in the grantees' communities?

  4. What are the characteristics of the interventions and how do victims and perpetrators of elder abuse participate in the grantees' intervention?

  5. What data are available at the state, local, and national levels to measure the outcomes associated with those interventions?

To address these questions, the evaluation assessed the implementation and outcomes of individual grantee prevention interventions.

C. Evaluation Design

We used a mixed-methods approach to conduct the process evaluation of the elder abuse prevention interventions. To address the first two research questions that focus on examining the implementation and infrastructure of the prevention interventions, we conducted site visits with each of the five grantees and met with grantee staff, partners, and providers that implemented the elder abuse prevention interventions in late 2014 and early 2015. Topics addressed during the site visits included the theoretical or clinical basis of the prevention intervention; implementation of the core components of the intervention; partnerships; context; facilitators and barriers; service utilization; state and local data collection systems; and project replicability and lessons learned.

Following the visits, we prepared summaries that were shared with the grantees, ASPE, and AoA. We used this information as the basis for a series of Research Briefs for each grantee that were disseminated during the White House Conference on the Aging in 2015. We also periodically reviewed grantee progress reports provided by AoA.

To address research questions 3-5 that focus on describing the characteristics of participants, the interventions themselves and available data, we developed a Cross-Grantee Data Analysis Plan that called for the collection of core data elements on client characteristics, program activities and outcome measures across grantees. Given the heterogeneity in scope and program features of the grantee initiatives, this unified approach allowed for comparison of client and service utilization characteristics and outcomes from the diverse interventions. The systematic collection of core data elements enabled the preparation of risk factor profiles on victims/care recipients and perpetrators/caregivers that are served by each intervention.

The core set of data elements describe demographic, psychological/physical health and social well-being indicators that are risk factors for elder abuse. Other elements pertain to referral source, type(s) of abuse, service utilization and outcomes. Identification of the common data elements for inclusion in the cross-site framework was guided by a balance between any additional burden placed on the grantees and the increased scientific rigor achieved from collecting identical information that could be compared across sites. We note that given the heterogeneity and some gaps in the data we found that they could not be reliably harmonized and pooled across the grantees.

We executed Data Use Agreements with each grantee and their partners, as appropriate, and specified the variables needed for analysis. Data transfers between the grantees and NORC's Data Enclave, a secure, protected environment, were conducted through securely encrypted transfer of incoming confidential data via National Institute of Standards and Technology-certified secure file transfer protocol applications. The grantees provided data dictionaries and assisted the team by reviewing analyses.

D. Key Findings

Given the complex and multi-dimensional nature of elder abuse, as well as different underlying theories guiding elder abuse subtypes, each of the five grantees developed a variety of multi-component and/or multi-disciplinary prevention interventions that addressed victims and elders at-risk, as well as care recipients and caregivers. Key findings are presented below.

  • AK DSDS through the APS Unit and in partnership with the Anchorage Police Department and other community partners implemented, tested and measured the performance of the Critical Time Intervention case management model to prevent elder abuse, neglect and exploitation.

  • NYSOFA, in conjunction with multiple partners, implemented an E-MDT incorporating forensic accountants and geriatric psychiatrists to investigate and intervene in complex cases of elder financial exploitation and elder abuse.

  • The USC Keck School of Medicine in partnership with the California Department of Aging, California Department of Social Services, Legal Aid Society of Orange County, and the Orange County Elder Abuse Forensic Center piloted a multi-dimensional intervention called Abuse Intervention Model (AIM) that designed and piloted a multi-component model for primary and secondary prevention of abuse of elders with dementia.

  • The UTHSC at Houston, in partnership with APS, the Texas Department of Aging and Disability Services, and the Houston area justice system piloted an intervention to increase medication adherence in older adults who have chronic health conditions and who neglect themselves.

  • TX/WellMed developed and tested clinical screening protocols within WellMed Clinics, including use of the Elder Abuse Suspicion Index© (EASI) screening tool to identify at-risk elders and prevent elder abuse. TX/WellMed also embedded two APS Specialists within WellMed Medical Management, a primary care physician group, to provide technical assistance, communication facilitation, and education supporting increased screening to prevent elder abuse.


An essential component of the evaluation was to examine the infrastructure and structure of the prevention intervention. As required by the grant, each of the prevention interventions had the support and active involvement of APS, whether serving as the lead entity (AK DSDS), a key implementation partner (NYSOFA, UTHSC, TX/WellMed), or a referral source (USC). Across the grantees, there was broad representation of community partners in implementing the core components of the prevention intervention as well as providing services to address elders' needs, such as protection and safety, medical care, food security, housing or legal and financial assistance. As some grantees served local communities, Area Agencies on Aging were enlisted to support the interventions. Partnerships with the justice system were rooted in long-standing organizational affiliations to address elder abuse in the grantees' communities, and included law enforcement and legal services, to varying degrees, depending on the intervention. Three grantees involved the justice system as part of the operating structure of the prevention intervention (NYSOFA, USC, AK DSDS). With two grantees, the justice system played a more peripheral role in the prevention intervention but was actively involved with elder abuse prevention and APS activities (UTHSC, TX/WellMed). Partnerships formed to implement the prevention intervention benefitted from active and sustained participation of its members.

Target Population

The target populations for four prevention interventions were elders at risk of abuse, neglect or exploitation (USC, AK DSDS, NYSOFA, TX/WellMed). One prevention intervention focused exclusively on substantiated victims of self-neglect (UTHSC). The minimum age for eligibility in the intervention was 60 years for two prevention interventions (UTHSC, NYSOFA) and 65 for the other three (AK DSDS, TX/WellMed, USC). Three prevention interventions targeted elders with cognitive impairment or dementia (AK DSDS, NYSOFA, USC) and four targeted elders with a physical impairment or health problem(s) (AK DSDS, NYSOFA, UTHSC, TX/WellMed). One prevention intervention targeted elders with detectable signs of possible financial exploitation present (NYSOFA).

Certain prevention interventions emphasized the connection between a vulnerable elder and a trusted person in his/her social network and the potential for abuse (i.e., the focal subject and responsible actor). This focus on relationships varied across the prevention interventions, as did the clinical or service delivery effort. USC explicitly focused on older adults with dementia at risk for abuse and their primary caregivers. In certain cases served by AK DSDS, there was a known abuser who was dependent on the victim. NYSOFA identified social isolation and inadequate social support as risk factors (and eligibility criteria), along with identification of perpetrators of financial exploitation, for cases served by the E-MDTs in the Finger Lakes region and Manhattan.

The prevention interventions varied in the number of elders served over the course of the three-year grant period. Original expectations were tempered by the ebb and flow of referrals from partners or the willingness of elders to participate. Three of the prevention interventions had rolling enrollment but with defined periods for participation and completion. AK DSDS received 170 referrals and had 87 elders participate in Elder Services Case Management. UTHSC recruited and enrolled 34 elders in the medication adherence prevention intervention. USC recruited a cohort of 76 dyads. Two prevention interventions had a more fluid referral stream. The NYSOFA E-MDTs served more than 220 elders, which included new cases and follow-up cases. TX/WellMed screened 11,426 elders using the EASI tool. Of these, 35 elders were referred to APS. Additionally, 588 WellMed patients were served through the APS Specialists and 474 were referred to APS.

Collectively, the prevention interventions targeted and addressed multiple forms of abuse, neglect, and exploitation and its co-occurrence. While the eligibility criteria for each prevention intervention focused on defined risks, co-morbid problems were addressed through the intervention. Those that emerged through assessments or over the course of the intervention were addressed through referrals to service partners.

Core Components

The core components of the five prevention interventions were implemented as intended, with some minor adjustments. They were conducted within the time period designated by the protocol. To varying degrees, each of the prevention interventions were standardized (or manualized) by creating manuals and protocols for staff implementation. As to be expected given the heterogeneity of the five prevention interventions, their delivery methods and service duration varied, and depended on the population targeted and the nature of abuse or risk. One common delivery element across all of the prevention interventions was the use of home visits as a primary method to reach at-risk elders, although the degree of contact varied. The intensity or dose of services varied with each prevention intervention, depending on the identified needs, the treatment protocol or case plan, the resource capacity of providers, and uptake by the elder. The duration of the prevention interventions varied, as well. Three were time-limited, with the duration ranging from three-months (USC), six-months (UTHSC), or nine-months (AK DSDS). Two were open-ended and depended on case resolution by the E-MDT (NYSOFA) or APS intervention as a result of screening or care coordination efforts (TX/WellMed).

The role and scope of service providers' involvement varied--from limited to extensive--across the prevention interventions. One had limited contact with external service providers in the community, but could turn to APS or a primary care physician in the event a problem or urgent need was identified (UTHSC). Prevention interventions that used a case management model or targeted elders' service needs coordinated referrals and service linkage with a range of community service providers (AK DSDS, NYSOFA, TX/WellMed, USC).

Three of the five prevention interventions identified areas that may be important to change in future expansion or replication efforts, based on their implementation experience. This included: Allowing for greater flexibility in the case management timeframe for an evidence-based intervention (AK DSDS); Providing scripts and language to clinics to better communicate with family members about the need for mandatory reporting and adapting elder abuse screening processes to better fit within an organization's existing protocols (TX/WellMed); and Using a less intensive staffing model or a more triaged assessment with a tiered intervention for a home-based intervention (UTHSC).

Facilitators and Barriers

The evaluation also addressed implementation facilitators and barriers. A number of common factors were identified across the five prevention interventions. To various degrees, all were grounded in strong partnerships with APS and community partners that assisted with intervention planning and/or implementation (AK DSDS, NYSOFA, TX/WellMed, UTHSC, USC). Although there was some turnover, continuity in staffing and leadership across the prevention interventions was critical in providing consistency in implementation and maintaining relationships developed between case managers and clients (AK DSDS), research staff and elders (UTHSC, USC), APS specialists and clinic staff (TX/WellMed), and E-MDT coordinators and community partners (NYSOFA).

Four of the prevention interventions had established referral partners that contributed resources in various capacities: to recruit and enroll elders in the intervention protocol (UTHSC, USC); take up a case with the E-MDT (NYSOFA); or provide community-based services once needs were assessed (AK DSDS, NYSOFA, UTHSC, USC). Use of a client-driven or patient-driven approach in the social service or clinical settings of the prevention interventions was extended by the involvement of partners, community agencies, advocacy organizations, and other entities in monthly standing meetings to address elder's needs stemming from abuse or risk of harm. Such forums helped expedite service delivery by specialists (NYSOFA, USC), provide complementary services and reduce fragmentation (TX/WellMed), and build awareness of available resources for referrals (AK DSDS, UTHSC).

Most challenges tended to be site-specific; a few were common to the prevention interventions, such as lower than expected recruitment, limited uptake of referrals, and retention of elders in the intervention. Limited services and lack of access to services affected two of the prevention interventions (AK DSDS, UTHSC).


A key task of this study is to describe the characteristics of victims, at-risk elders, care recipients, perpetrators and caregivers who participated in the five interventions. While we report broad patterns that emerge in selected characteristics of participants, these findings need to be understood within the context of each intervention's goals and eligibility criteria. An intervention's focus on a particular type of abuse (i.e., self-neglect or financial exploitation or all forms) and selection factors for inclusion (i.e., physical and cognitive impairment and social isolation as well as age minimums) not only shape the pool of elders for participation from the outset of the study but are in part determined by risk factors of abuse themselves. The differences--and similarities--then, that we observe across interventions are in part due to the intervention's focus and recruitment process. A risk factor for one type of abuse, furthermore, may not be a risk factor for another form.

With these caveats, we describe herein the characteristics of the five grantee interventions and their participants and where possible, draw on prior research on specific forms of abuse and risk factors in order to place the findings in context. In terms of age, elders served by the prevention interventions ranged from 74 years to 81 years. The majority of victims and at-risk elders was female, spoke English as their primary language, had low income levels and lived alone and in a private home. Greater variation was observed across grantees with respect to the race and ethnicity of elders served, education levels, and marital status. The high number of female victims and at-risk elders in the interventions is consistent with elderly women's greater representation in APS caseloads (Wolf 1997). At the same time, Pillemer and Finkelhor (1988) have noted that this may be due to elderly women's greater numbers in the senior population. Their study found that the victimization rate was higher for men (5.1 percent) than women (2.5 percent). In terms of living arrangement, living alone was found to be a protective factor against elder mistreatment (Lachs et al. 1997). Shared residence increases opportunities for contact and has been linked to violence, particularly when Alzheimer's patients live with immediate family members (Paveza et al. 1992). It should be noted, however, that living arrangement is likely to play a differential role depending on the type of abuse being examined. For example, a shared living arrangement may not be as relevant in cases of self-neglect compared to other forms of mistreatment such as physical abuse or financial exploitation.

Turning to physical and psychosocial characteristics of victims, at-risk elders and care recipients, the physical functioning of elders served by the interventions tended to be fairly low-to-moderate. Levels of cognitive impairment, on the other hand, varied. Whereas self-neglecting elders were cognitively intact, most care recipients were cognitively impaired. There was also variation with respect to levels of anxiety and stress experienced by participants and limited evidence for depression among the elders served. Elders served by three prevention interventions reported low-to-moderate levels of social support but elders experiencing financial exploitation tended to be socially isolated. Past research has found that low levels of social support increases the risk of elder mistreatment (Lachs et al. 1994) and is associated with caregivers' verbal and physical abuse (Compton et al. 1997).

Limited information about perpetrators was available for two interventions (NYSOFA and TX/WellMed). Alleged perpetrators tended to be middle-aged or elderly, and included both males and females. Race and ethnicity was known for only a subset, but perpetrators were predominantly Caucasian. The educational background of the alleged perpetrators of financial exploitation ranged from those with limited education to the highly-educated. Most alleged perpetrators were family members or relatives. These findings are consistent with previous research indicating that victims' family members (adult children and spouses) tend to be perpetrators (Acierno et al. 2009). Alleged perpetrators also tended to have issues with substance abuse. Previous studies have also known that alcohol or drug abuse problems as well as a history of mental illness are relatively common among perpetrators (Greenberg et al. 1990; Wolf & Pillemer 1989).

Findings about caregivers are drawn exclusively from one intervention (USC). Caregivers were mostly female, Caucasian, married to the care recipient, college-educated, and had fairly high incomes. Many caregivers were adult children. Although exhibiting low levels of anxiety and burden, and with moderate levels of support, caregivers showed signs of depression. A study by Paveza et al. (1992) found that depression among Alzheimer's caregivers predicted physical abuse.

In terms of types of abuse experienced, a finding across the prevention interventions was that self-neglect was the most common type of abuse experienced and co-occurred with all forms of abuse, reinforcing that elder self-neglect is a serious public health problem and a prevalent concern for APS (Naik et al. 2008). Financial exploitation co-occurred with other forms of abuse. Many elders served by the preventions interventions experienced more than one type of abuse. Thus, elders participating in the prevention intervention had multiple service needs.

With respect to outcomes, we examined whether cases had been referred to APS once the intervention had been completed (i.e., recidivism for those with prior APS histories) for a subset of elders served by AK DSDS and UTHSC. For NYSOFA, we examined outcomes achieved regarding financial exploitation. For TX/WellMed, we examined APS data collected on reasons for case closure. We found that most elders served by UTHSC did not have a re-referral to APS, but about one-third did. For the elders served by AK DSDS, 90 percent did not have a subsequent referral to APS. Intervention by NYSOFA's E-MDTs stopped financial exploitation of elder assets. TX/WellMed's use of the EASI screening tool identified few patients as at-risk for elder abuse. At the same time, at least 588 WellMed patients were brought to the attention of APS Specialists and served by WellMed's Complex Care services and/or APS whose needs may not have been addressed otherwise.

Changes measured in elders' state of vulnerability, characteristics or circumstances varied across the prevention interventions. As the type of change was intervention-specific, the measures and quantity of data available also varied.

In terms of social support and risk, findings varied across the prevention interventions. The moderate level of social support and physical functioning reported for elders served by UTHSC at baseline remained stable following the intervention. While elders served by AK DSDS reported less vulnerability following the intervention, they may have been susceptible to harm by others. High risk of financial exploitation decreased for a subset of elders served by one of NYSOFA's E-MDTs. Care recipients' sense of vulnerability and coercion fluctuated over the course of the dyadic intervention. Yet the perceived degree of social support remained constant for care recipients and caregivers served by USC.

In closing, this evaluation provides information about the development and implementation of the five Elder Abuse prevention interventions, focused on the characteristics of victims and at-risk elders, care recipients and caregivers, along with perpetrators of elder abuse; service utilization; and outcomes. Despite the limitations noted, collectively, the implementation and outcomes findings point to field-initiated approaches that merit further investigation and effectiveness testing using rigorous scientific designs, in an effort to build the knowledge base and prevent and reduce elder abuse, neglect, and exploitation.