Over the past two decades, domestic violence
issues have been the focus of growing attention. Numerous studies have
been undertaken to study violence in the home and to test ways of
reducing it. Although a significant amount of research has been
conducted on abused wives and abused children, very little research
had been done on the domestic abuse of elderly persons. Similarly, a
large number of evaluations have been conducted on the effectiveness
of domestic violence programs in general, but little research had been
undertaken on the effectiveness of elder abuse programs and of various
types of interventions for elder abuse. This project consisted of an
evaluation of an elder abuse program run by the New York Police
Department and Victim Services Research. The focus of the study was
domestic elder abuse, which generally refers to any of several forms
of maltreatment, physical abuse, sexual abuse, psychological abuse,
neglect, and/or financial exploitation of an older person. The
program, conducted in New York City public housing, had two
complementary parts. The community-level component provided education
through leaflets and posters to all elderly residents of housing
projects that were targeted to receive educational materials about
elder abuse. In the individual-level component, crisis response teams
(each consisting of a police officer and a social worker) were
dispatched to randomly-selected residents to follow up on the initial
patrol response to elder abuse reports and to link elder abuse victims
and abusers to long-term social services. The intent of the two
interventions was to improve the victim's awareness and use of
resources, the victim's knowledge of elder abuse, and the willingness
of victims to call the police. The theory behind the interventions
posited that, through the provision of information and the impression
of increased police surveillance, a reduction in repeat victimization
would be observed. This reduction of repeat victimization and the
provision of supportive services were expected to have an effect on
the psychological health and well-being of the victims.
This study employed community-level and
individual-level interventions combined in a nested randomized
experimental design. First, public housing projects in Manhattan were
assigned to one of two levels of public education (i.e., to receive or
not to receive educational materials about elder abuse). The public
education intervention consisted of several components, including
distribution of leaflets, hanging of posters, and community
presentations. Brochures and posters describing elder abuse and giving
phone numbers for assistance were distributed to building managers at
the intervention sites. The managers placed the posters in lobbies,
laundry rooms, mail areas, rental offices, and other common
areas. Brochures were delivered to the apartments of elderly
residents. For targeted housing projects, presentations on elder abuse
were made at public housing tenant association meetings. The remaining
housing projects were assigned to the control group. Once the public
education treatment had been implemented, 403 older adult residents of
the housing projects who reported elder abuse to the police during the
next ten months were assigned to one of two levels of follow-up to the
initial police response (i.e., to receive or not to receive a home
visit) as the second part of the project. The home visit intervention
consisted of a strong law enforcement response designed to prevent
repeat incidents of elder abuse. A team from the Domestic Violence
Intervention and Education Program (DVIEP), consisting of a police
officer and a social worker, followed up on domestic violence
complaints with a home visit within a few days of the initial patrol
response. Cases assigned home visits received a letter and phone
calls, which provided referral numbers and notified of an upcoming
home visit. During the home visit, the DVIEP team discussed current
and past abuse with victims, and provided information about elder
abuse. The team emphasized to victims that elder abuse is a crime, and
the team explained procedures for filing an order of protection and
the difference between family court and criminal court. If the
offender was present, the police officer informed the abuser that the
household was being monitored. Four weeks after the initial home
visit, the victim received a follow-up telephone call from the
counselor. Each victim was given the name and telephone number of
their local DVIEP counselor and other local Victim Services counselors
specializing in elder abuse. The team assessed the household for
service needs and made referrals as needed (e.g., to respite care,
counseling programs for abusers, home care services for
seniors). Cases not assigned to the home visit group were assigned to
the control group. Following the usual patrol response to the elder
abuse complaint, these cases received a generic DVIEP letter that was
similar to the intervention group's letter but omitted information on
elder abuse and information about the home visit. Victims were
interviewed about new victimizations following the intervention on
three occasions, six weeks after the trigger incident, six months
after the trigger incident, and twelve months after the trigger
incident. The DVIEP databases maintained on households at each of the
three participating PSAs were searched to identify new police reports
of elder abuse for households in the sample within 12 months following
the trigger incident.
Nested randomized experimental design using dual sampling
frames.
Elderly residents of housing projects in Manhattan.
Individuals
telephone and personal interviews
survey data
experimental data
Interviews at the three time points were identical
except for the omission of background information on the second and
third interviews. Demographic data collected during the first
interview only included age, gender, ethnicity, education, employment,
income, legal relationship with abuser, living situation, number of
people in the household, and health. For each time point, data provide
measures of physical, psychological, and financial abuse, knowledge of
elder abuse, knowledge and use of social services, satisfaction with
the police, assessment of service delivery, and self-esteem and
well-being. The DVIEP databases maintained on households at each of
the three participating Police Service Areas (PSAs) were searched to
identify new police reports of elder abuse for households in the
sample within 12 months following the trigger incident. Variables from
the DVIEP databases include age, race, ethnicity, and sex of the
victim and the perpetrator, relationship of perpetrator to victim,
type of abuse reported, charge, whether an arrest was made, if an
order of protection had been obtained, if the order of protection was
violated, use of weapons, if the victim had been injured, and if the
victim was taken to the hospital. Several time lapse variables between
different time points are also provided.
The completion rate with victims was 67 percent
for the first interview, 69 percent for the second interview, and 67
percent for the third interview. Between one-third and one-half of
researchers' attempts to interview the victims failed due to victim
refusals. The refusal rate for victims was 11 percent for the first
interview, 14 percent for the second interview, and 14 percent for the
third interview. Three percent of the sample was lost because of
out-of-town moves and 2 percent due to death and illness. The
remainder of those not interviewed could not be contacted by telephone
or visits to their homes.
To assess history of violence and frequency and severity
of violence, this project used a variant of the Conflict Tactics Scale
(Straus, 1979) adapted by Pillemer and Finkelhor (1988) in their elder
abuse research. To measure knowledge and the use of social services,
the researchers adapted a scale used in their earlier research (Davis
and Taylor, 1997). The project used the Rosenberg Self-Esteem Scale
(Rosenberg, 1979) to gauge self-perceptions of victims. The Bradburn
Affect-Balance Scale was administered to tap the affective status of
the victims in the sample. A Likert-type scale was also used for the
questions on public education.