Content uploaded by Joanne Belknap
Author content
All content in this area was uploaded by Joanne Belknap
Content may be subject to copyright.
The Impact of Community-Based Outreach on Psychological Distress and
Victim Safety in Women Exposed to Intimate Partner Abuse
Anne P. DePrince
University of Denver
Jennifer Labus
University of California Los Angeles
Joanne Belknap and Susan Buckingham
University of Colorado Boulder
Angela Gover
University of Colorado Denver
Objective: Using a longitudinal, randomized controlled trial, this study assessed the impact of a
community-based outreach versus a more traditional criminal justice system-based referral program on
women’s distress and safety following police-reported intimate partner abuse (IPA). Method: Women
(N ⫽ 236 women) with police-reported IPA were randomly assigned to 1 of 2 interdisciplinary
community-coordinated response program conditions: Outreach (community-based victim advocate
outreach) or Referral (criminal justice system-based victim advocate referrals to community-based
agencies). Participants were interviewed 3 times over a 1-year period: within 26 (median) days of
police-reported IPA, 6 months later, and 12 months later. Primary outcome measures included posttrau-
matic stress disorder and depression symptom severity (Posttraumatic Stress Diagnostic Scale; Beck
Depression Inventory–II), fear appraisals (Trauma Appraisal Questionnaire), IPA revictimization (Re-
vised Conflict Tactics Scale), and readiness to leave the relationship with the abuser. Results: One year
after the initial interview, women in the Outreach condition reported decreased PTSD and depression
symptom severity and fear compared with women in the Referral condition. Although both conditions
were unrelated to revictimization in the follow-up year, women in the Outreach condition reported greater
readiness to leave the abuser and rated services as more helpful than women in the Referral condition.
Conclusions: This is one of the first studies to examine community-based outreach in the context of an
interdisciplinary community coordinated response to police-reported IPA. The findings suggest that
community-based outreach by victim advocates results in decreased distress levels, greater readiness to leave
abusive relationships, and greater perceived helpfulness of services relative to system-based referrals.
Keywords: intimate partner abuse, coordinated community response, PTSD, depression
A considerable amount of research documents the significant
prevalence and scope of intimate partner abuse (IPA) experienced
by women (e.g., Rennison & Welchans, 2000; Tjaden & Thoennes,
2000).
1
IPA is linked to serious forms of psychological distress as
well as ongoing danger in relationships. In terms of psychological
distress, both posttraumatic stress disorder (PTSD) and depression
symptoms have been well documented in IPA victims (e.g., R.
Campbell, Greeson, Bybee, & Raja, 2008; Cattaneo & Goodman,
2010; Coker et al., 2002; Johnson, Zlotnick, & Perez, 2008;
Krause, Kaltman, Goodman, & Dutton, 2006, 2008; Mechanic,
Weaver, & Resick, 2008). In addition, a handful of studies point to
high levels of fear among abused women (Cattaneo & Goodman,
This article was published Online First February 13, 2012.
Anne P. DePrince, Department of Psychology, University of Denver;
Jennifer Labus, Department of Psychiatry and Biobehavioral Sciences,
Semel Institute for Neuroscience and Human Behavior, and Oppenheimer
Family Center for Neurobiology of Stress, University of California Los
Angeles; Joanne Belknap, Department of Sociology, University of Colo-
rado Boulder; Susan Buckingham, Department of Geological Sciences,
University of Colorado Boulder; Angela Gover, School of Public Affairs,
University of Colorado Denver.
This project was funded by National Institute of Justice Office of Justice
Programs, United States Department of Justice Award 2007-WG-BX-0002
to Anne P. DePrince. The opinions, findings, and conclusions or recom-
mendations expressed in this report are those of the authors and do not
necessarily reflect those of the Department of Justice or the National
Institute of Justice.
Thank you to our study partners, including the Denver District Attor-
ney’s Office, City Attorney’s Office, Denver Police Department Victim
Assistance Unite, Denver Domestic Violence Coordinating Council, Safe-
House Denver, Project Safeguard, AMEND, and the Triage Steering Com-
mittee. Thank you to the Traumatic Stress Studies Group, particularly
Melody Combs, Claire Hebenstreit, Ryan Matlow, Courtney Mitchell,
Annarheen Pineda, and Jane Sundermann. Finally, thank you to the women
who trusted us with their stories over the year during which we were
privileged to get to know them.
Correspondence concerning this article should be addressed to Anne P.
DePrince, Department of Psychology, 2155 S. Race Street, Denver, CO
80208. E-mail: adeprinc@du.edu
1
Abuse by intimate partners can include physical and nonphysical abuse
and is increasingly recognized to include physical, sexual, psychological,
and stalking abuse. Given that some of these are not violent per se, we use
the term intimate partner abuse (IPA) instead of intimate partner violence.
Journal of Consulting and Clinical Psychology © 2012 American Psychological Association
2012, Vol. 80, No. 2, 211–221 0022-006X/12/$12.00 DOI: 10.1037/a0027224
211
2010; Fanslow & Robinson, 2010). Because fear is linked to both
PTSD and depression symptoms (e.g., DePrince, Zurbriggen, Chu,
& Smart, 2010), addressing fear as well as PTSD and depression
symptoms following IPA is critical to women’s psychological
health.
Unfortunately, many relationships continue to be dangerous for
women exposed to IPA because of high rates of revictimization
(e.g., Cattaneo & Goodman, 2005; Krause et al., 2006, 2008;
Robinson & Tregidga, 2007), which is associated with worse
PTSD and depression symptoms (Classen et al., 2005; DePrince,
2005; Krause et al., 2006, 2008). Although the question of whether
mental health symptoms contribute to or are consequences of
revictimization remains unclear (Classen et al., 2005), links be-
tween the two make revictimization a critical consideration in
interventions. Importantly, victims cannot control whether their
abusers continue to engage in abusive behaviors; thus, interven-
tions should also affect domains in which women can have more
control, such as victims’ own reasoning regarding staying/leaving.
Applications of the transtheoretical model of change to IPA have
illustrated that women vary in their readiness to leave abusive
partners (Shurman & Rodriguez, 2006); therefore, interventions
that affect readiness to leave may be important to victim safety
considerations.
In light of IPA victims’ psychological and safety needs, re-
searchers and practitioners have pointed to the urgent need for
victim-service providers to make outreach to victims. A handful of
studies have now evaluated victim outreach from battered wom-
en’s shelters (e.g., Bybee & Sullivan, 2002, 2005; Sullivan &
Bybee, 1999; Sullivan, Bybee, & Allen, 2002) or emergency
rooms and health clinics (e.g., Gillum, Sun, & Woods, 2009;
Kendall et al., 2009), demonstrating some positive effects on
women’s well-being (e.g., revictimization, safety behaviors, social
support). Although promising, these evaluations have been limited
to women already using shelter/medical services, many of whom
tend to be disproportionately economically marginalized (Cattaneo
& DeLoveh, 2008).
Challenges for Victims When IPA is Reported to
Law Enforcement
When IPA is reported to law enforcement, victims find them-
selves situated in the criminal justice system, which focuses on the
legal case, rather than psychosocial needs. In the aftermath of the
incident, criminal justice system-based advocates (hereafter re-
ferred to as system-based advocates) are often the first to be in
contact with victims. As criminal justice personnel housed in
police departments and/or prosecuting attorneys’ offices, system-
based advocates are subject to disclosure in legal cases and cannot
offer women confidentiality during their contacts. Therefore,
women may be hesitant to disclose details (e.g., regarding psycho-
logical needs) or ask for support when speaking with system-based
advocates, because that information will be subject to disclosure to
the defense. To connect women with confidential services that
address psychosocial needs, system-based advocates have to make
referrals to community-based agencies.
Women face many challenges when trying to find appropriate
confidential services following police-reported IPA. For example,
increases in economic strain (e.g., transitions in housing, economic
problems caused by partner’s arrest) and demands on women’s
time (e.g., additional caregiving and/or job responsibilities caused
by partner’s arrest) may result in practical barriers (e.g., lack of
access to a computer) that hamper women’s efforts to seek out
resources. In addition, cognitive aspects of common posttraumatic
responses (e.g., problems with memory) may make organizing and
initiating efforts to find resources difficult. Further, many IPA
victims have been isolated by perpetrators, leaving them with
impoverished social support systems (Sullivan & Bybee, 1999)
and feelings of alienation (DePrince, Chu, & Pineda, 2011) fol-
lowing IPA that make initiating requests for help difficult. Finally,
women may experience obstacles navigating the range of IPA
services available in many communities to identify the services
most appropriate to her specific case and needs (e.g., counseling
vs. legal advocacy).
Community-Coordinated Response (CCR) Programs
Given the psychosocial needs of and challenges facing IPA
victims (particularly following police-reported IPA), researchers
and practitioners have increasingly called for the development and
evaluation of CCR programs that provide victim advocacy to a
broad base of victims (e.g., Allen, Bybee, & Sullivan, 2004;
Fanslow & Robinson, 2010; Goodman & Epstein, 2005). CCR
programs involve collaboration between community-based (e.g.,
shelters and legal advocacy programs) and criminal justice system-
based (e.g., prosecuting attorney and police) agencies to coordinate
victim advocacy for IPA intervention and prevention (Klevens,
Baker, Shelley, & Ingram, 2008). Despite the need to coordinate
responses to address victim safety and psychological well-being,
most CCR programs have been geared more toward responding to
abusers than victims (Goodman & Epstein, 2005). Given involve-
ment of both criminal justice and community-based agencies, CCR
programs are well positioned to respond to the needs of victims
following police-reported IPA, although research on the impact of
victim advocacy within CCRs on victims is noticeably lacking.
CCR programs that involve outreach directly to victims from
community-based (rather than system-based) advocates may be
particularly well-poised to mitigate deleterious IPA psychological
and safety outcomes. By making outreach directly to victims,
community-based agencies remove barriers that might otherwise
impede women from seeking out community-based services to
address psychosocial needs. Further, the high level of coordination
involved in CCR programs means that community-based agencies
can be selected to outreach to victims based on specific psychos-
ocial needs. Matching services with victim needs at the point of
outreach should lead to better service provision, with resulting
gains in psychological symptoms and safety. Further, the process
of finding resources should be less stressful (e.g., women do not
have to retell their stories as they seek out an agency that meets
their needs), which may also reduce distress. Outreach might be
particularly important to women who are oppressed or marginal-
ized by life circumstances beyond the IPA itself (e.g., socioeco-
nomic status, physical dependence on the offender) that otherwise
add barriers to seeking out services or increase risk of psycholog-
ical symptoms, making examination of moderators important.
Current Study
Addressing the gap in CCR programs that focus on victims, the
current study assesses the impact of two approaches within a CCR
212
DEPRINCE ET AL.
program in a single jurisdiction in a large urban city in the Rocky
Mountain West of the United States. The overarching goal of the
CCR program was to coordinate responses to IPA victims across
an interdisciplinary team of law enforcement, prosecution, and
community-based service agencies to improve victim psycholog-
ical distress and safety (see DePrince, Belknap, Labus, Bucking-
ham, & Gover, in press, for discussion of criminal justice out-
comes from this CCR). Within that overarching goal, two
approaches were used: community-based outreach (Outreach) and
criminal justice system-based referral (Referral).
Commonalities Across Outreach and Referral
Following the initial report to law enforcement, system-based
victim advocates attempted to contact victims to assess victim
needs. Details about the IPA incidents, including the information
gathered by system-based advocates, were then discussed by an
interdisciplinary victim-service team composed of criminal justice
system and community-based victim-service professionals.
Unique Aspects of Outreach
For the Outreach condition, the interdisciplinary victim-service
team identified a specific community-based agency to initiate
phone outreach to each victim based on her unique case and needs.
Outreach by community-based agencies offered women a confi-
dential means of learning about and accessing support and ser-
vices, while not requiring women to initiate their own search for
services in the aftermath of IPA. During the Outreach phone call,
community-based advocates told women that they were calling
because of the police-reported IPA. The advocates then described
the specific services available at that agency (e.g., legal advocacy,
counseling, shelter) and provided a nuanced explanation of how
those specific services might cater to each victim’s particular
needs. Further, because the community-based advocate making the
outreach call already had basic details about the IPA incident from
the interdisciplinary team review, the victim did not have to
describe the incident to learn about services. Because the outreach
phone conversation was confidential, women could freely disclose
information about their reactions or needs related to the IPA.
Unique Aspects of Referral Condition
For the Referral condition, a system-based advocate from the
prosecuting attorney’s office or police department contacted
women to make referrals to community-based agencies based on
recommendations from the interdisciplinary victim-service team.
In these calls, system-based advocates offered victims only basic
information about agencies, and the conversation was not confi-
dential. The onus was then on women to initiate calls to
community-based agencies, at which time women would have to
describe their cases (e.g., to an intake worker) to learn more about
services.
Predictions
The Outreach condition was designed to initiate contact with
women from community-based victim-service agencies that could
provide confidential services specific to each woman’s needs.
Thus, our primary prediction was that Outreach (compared with
Referral) would be associated with greater decreases in common
victim IPA reactions (PTSD, depression, and fear). A secondary
prediction was that Outreach (relative to Referral) would lead to
better victim safety outcomes, including greater readiness to leave
the abusive relationship and decreases in revictimization. To our
knowledge, this is the first CCR evaluation focused on victims and
the only one using a randomized control, longitudinal design.
Participants were interviewed by a separate research team as close
to the IPA incident as possible (Time 1 [T1]) and then 6 (T2) and
12 (T3) months later. Psychological distress (i.e., PTSD, depres-
sion, and fear) and victim safety (i.e., revictimization, readiness to
leave the abusive relationship) outcomes were assessed, while
taking into account potential moderators (socioeconomic status,
living with the offender, physical or economic dependence on the
offender, and ethnic minority status).
Method
All study procedures and measures were reviewed and approved
by a university-based Institutional Review Board.
Participants
The research team obtained publicly accessible police reports to
evaluate inclusion and exclusion criteria. Reports were used for
recruitment if the incident involved nonsexual IPA (incident re-
ports for sex crimes are not publicly available and therefore could
not be accessed for this project); a female, adult victim; and a
male, adult offender. Reports were excluded if women were mono-
lingual non-English speakers and if there was a cross arrest (i.e., if
both male and female partners were arrested). Finally, police
reports had to provide some form of valid contact information for
victims. Based on these criteria, we attempted to contact 827
women; 236 (29%) were enrolled in the study. Additional details
about participant recruitment are described in DePrince et al. (in
press). Figure 1 illustrates the participant flow through the study.
Materials
Symptom measures. PTSD symptom severity was assessed
with the Posttraumatic Stress Diagnostic Scale (PDS; Foa, Cash-
man, Jaycox, & Perry, 1997), a 49-item measure based on Diag-
nostic and Statistical Manual of Mental Disorders (4th ed.; DSM–
IV; American Psychiatric Association, 1994) criteria for PTSD. A
total score that reflects PTSD symptom severity was calculated by
summing all items. At T1, many women did not meet the time
requirement for PTSD; thus, we comment only on symptom se-
verity and not PTSD diagnostic status. Coefficient alphas were as
follows for each time point: T1 ⫽ .82, T2 ⫽ .93, T3 ⫽ .93.
Depression symptoms were assessed with the Beck Depression
Inventory–II (BDI; Beck, Steer, Ball, Ranieri, 1996), one of the
most widely used self-report measures of depression with high
validity and reliability. This 21-item measure assesses depression
symptoms based on DSM–IV criteria. The summed items create a
total depression score. Coefficient alphas were as follows for each
time point: T1 ⫽ .89, T2 ⫽ .91, T3 ⫽ .91. Fear was assessed using
the Fear scale of the Trauma Appraisal Questionnaire (TAQ;
DePrince et al., 2010). The TAQ is a 54-item self-report measure
of posttraumatic appraisals that has demonstrated reliability and
213
IMPACT OF OUTREACH AFTER INTIMATE PARTNER ABUSE
validity. Women were asked to think about current feelings in
relation to thinking about target IPA event while answering the
TAQ. Coefficient alphas for the TAQ fear scale were as follows:
T1 ⫽ .93, T2 ⫽ .91, T3 ⫽ .90.
Victim safety. The occurrence of new IPA incidents during
the study period was assessed using items from the Revised
Conflict Tactics Scale (CTS; Straus, Hamby, Boney-McCoy, &
Sugarman, 1996). The CTS is a widely used and well-validated
instrument for assessing the use of psychological, physical, and
sexual aggression in relationship conflict. Revictimization was
present if women reported one or more tactics within each cate-
gory (psychological, physical, and/or sexual) was used against
them at T2 or T3. Women were asked about aggression perpetrated
by the partner from the initial incident as well as new partners.
Women’s readiness to leave the relationships with the offender
was assessed using an adaptation of the Stages of Change–Short
Form (e.g., DiClemente et al., 1991), a widely used measure based
on the Stages of Change model (e.g., Prochaska & DiClemente,
1983). Consistent with Burke, Mahoney, Gielen, McDonnell, and
O’Campo (2009), the current study used five questions to deter-
mine stage of change: (a) Are you currently in a relationship with
[the offender]? (b) If not, have you been out of the relationship for
over 6 months? (c) Are you thinking about leaving the relationship
sometime in the next 6 months? (d) Are you planning to leave the
relationship in the next 30 days? and (e) Have you left the rela-
tionship or tried to leave sometime in the past year? Women’s
responses were coded as follows: still in a relationship with the
offender and not considering leaving in the next 6 months coded as 1
(precontemplation), still in the relationship with the offender but
considering leaving in the next 6 months (and not planning to leave in
the next 30 days) coded as 2 (contemplation), planning to leave in the
next 30 days and had previously left or tried to leave the relationship
in the past year were coded as 3 (preparation), left the relationship
within the past 6 months coded as 4 (action), left the relationship over
6 months ago coded as 5 (maintenance).
Service utilization and therapy. At each time point, we
asked participants about their experiences with services from
community-based agencies on the CCR team as well as talking to
a therapist. We calculated the percentage of women who reported
contact with these agencies at some point during the study period.
In addition, at Time 2 and Time 3, we developed a question to ask
women to rate the helpfulness of contacts with community-based
agencies as well as the police department victim assistance unit
over the previous 6 months (i.e., from T1 to T2, measured at T2;
Figure 1. Flow of participants through the study. IPA ⫽ intimate partner abuse; T1 ⫽ Time 1; T2 ⫽ Time 2;
T3 ⫽ Time 3.
214
DEPRINCE ET AL.
and from T2 to T3, measured at T3). For each agency, women
responded on a scale of 1 to 5 (1 ⫽ they made things much harder,
3 ⫽ they didn’t really make things easier or harder, and 5 ⫽ make
things a lot easier). The same question about helpfulness was
asked for therapy services. Because we did not have predictions
about time per se, we averaged women’s T2 and T3 responses
across agencies to derive a single score for victim service and
therapy helpfulness.
Moderators. Participant characteristics may influence inter-
vention outcomes (Kraemer, Frank, & Kupfer, 2006). Therefore,
exploratory hypothesis generating moderator analyses have been
highly recommended because the results of these analyses can
have important implications for clinical work, research, and policy
development. We examined the moderating effects of socioeco-
nomic status, perceptions of dependence on the offender, and
ethnicity. Socioeconomic status was assessed with three variables:
highest level of education, occupation, and income. Living with
the offender was assessed by asking women whether they lived
with the offender at the time of the incident that led to the police
report (yes/no). Perceptions of physical and economic dependence
on the offender were assessed with two questions: (a) “How
dependent on [partner’s name] were you for your physical well-
being?” (1 ⫽ not at all dependent to 5 ⫽ entirely dependent) and
(b) “How important to your financial stability was the money
[partner’s name] brings home?” (1 ⫽ not at all important to 5 ⫽
absolutely necessary). Ethnicity was a dichotomous variable coded
as –1 for women who identified only as White and 1 for women
who identified with any racial or ethnic minority group.
CCR Intervention Procedure
Prior to daily interdisciplinary team meetings, the team leader
applied an algorithm to randomly assign all victims identified in
police reports to Outreach or Referral conditions. Also prior to the
team meetings, a system-based advocate attempted initial contact
with victims to assess their needs. Importantly, victims did not
receive the Outreach or Referral interventions if they (a) could not
be reached within 3 business days by the system-based advocate
(as this meant the system-based advocate could not assess their
needs to share with the interdisciplinary team) or (b) told the
system-based advocate that they wanted no further contact.
At the interdisciplinary team meeting, the team conducted em-
pirically guided risk assessments on a case-by-case basis to iden-
tify women who appeared to be at grave risk for imminent serious
harm or fatality. Victims who the team determined to be at grave
safety risk were not eligible for the study and were automatically
referred for community-based outreach. The interdisciplinary team
was blind to the Outreach/Referral condition assignment until after
the risk assessment to assure that risk was not evaluated differently
depending on condition assignment.
Once the risk assessment was completed, the condition assign-
ment was revealed to the interdisciplinary team. Among women
randomly assigned to the Outreach condition, the interdisciplinary
team chose the most appropriate lead community-based agency to
address each victim’s needs. Victim advocates from the lead
community-based agency were then responsible for making phone
outreach to victims to offer services. For those assigned to the
Referral condition, a system-based victim advocate called women
to make referrals to community-based agencies; women in the
Referral condition could then contact community-based agencies
themselves if they so desired.
Research Procedure
The research team retrieved publicly accessible police incident
reports from the Denver, Colorado Police Department several
times per week. Within a day of retrieving these incident reports,
the research team sent lead letters inviting women to participate in
a Women’s Health Study. Approximately 3 days after the lead
letter was mailed, research staff initiated phone calls to invite
potential participants. Recruiting materials (lead letters, phone
scripts) did not mention IPA to decrease safety risks to potential
participants. During the initial phone contact, potential participants
were invited to attend a 3-hr interview. Women who indicated that
they would have to take public transportation to the interview were
offered cab rides.
Participants were greeted by a female interviewer (a graduate-
level interviewer and/or the principal investigator) who was blind
to the intervention condition. During the consent process, women
were told that their names had been identified from police reports.
Participants were informed about the scope of the study as well as
their rights as participants. Participants completed a “consent quiz”
designed to assess understanding of the consent information. Fol-
lowing consent procedures, participants completed the interview
and questionnaires. At the end of the session, participants were
asked to complete questions to monitor responses to study proce-
dures.
Approximately 3 weeks prior to follow-up interview due dates,
a lead letter was sent letting women know that the researchers
would be calling to schedule a follow-up interview. Approximately
2 weeks prior to T2 and T3 interview dates, we began calling
women to schedule the T2/T3 interview. When women arrived for
the follow-up interviews, we reviewed consent information and
readministered the consent quiz. Participants then completed the
interview and questionnaires after the consent review procedures.
At the end of each interview, women were compensated for their
time ($50 at T1, $55 at T2, and $60 for T3) and debriefed. At the
end of each interview, women were offered a newsletter that
provided referrals to community agencies dealing with health and
violence issues.
Analysis Plan
Primary outcome variables. Intent to treat contrast analyses
were performed to test for hypothesized intervention-induced de-
creases in the primary outcome variables (PSTD, depression, fear)
using a general linear mixed model in SAS (Version 9.2). Partic-
ipants were specified as a random effect, as this designation
produced the best fitting error-covariance structure based on model
fit indices (Akaike information criterion, Bayesian information
criterion). We modeled the primary outcomes using second degree
polynomial functions of time (i.e., linear, quadratic effects) and
their interaction with condition (Outreach or Referral). Specific a
priori contrasts of interests included differences in symptom
change on the primary outcomes between Outreach and Referral
conditions from T1 to T2, T2 to T3, and T1 to T3. In addition,
effect sizes were calculated using Cohen’s d. For each primary
outcome analyses, we graph response trajectory over time by
215
IMPACT OF OUTREACH AFTER INTIMATE PARTNER ABUSE
group, presenting parameter estimates and 95% normal confidence
intervals. To control for measurement variability of the instru-
ments used to measure PTSD, fear, and depression, we used a
completers-only approach to calculate each participant’s reliable
change index (RCI) from T1 to T3 using reliability and standard
deviation estimates from independent samples (Evans, Margison,
& Barkham, 1998). Fisher’s exact test was then applied to test for
group differences in RCI.
Moderator analyses. To test whether trajectories for out-
comes differed between Outreach and Referral groups depending
on the level of the hypothesized moderator variables, we used a
general linear mixed model (GLMM) with sequential sum of
squares (for further details on the method, see Holroyd, Labus, &
Carlson, 2009). Moderator effects are described in Table 1 by
displaying parameter estimates and 95% normal confidence inter-
vals at clinically meaningful levels of the moderator variables.
Given the inherent problems in conducting a moderator analysis in
a moderate size trial (N ⬍ 400 participants; Holroyd et al., 2009;
Kraemer, Wilson, Fairburn, & Agras, 2002), we do not provide p
values. As an alternative, we calculate effect sizes (Cohen’s d;J.
Cohen, 1992) to elucidate the nature of moderator effects. By
convention d ⫽ 0.20, d ⫽ 0.50, and d ⫽ 0.80 are interpreted as
small, moderate, and large effect sizes, respectively. We consider
these moderator analyses descriptive and hypothesis-generating
rather than hypothesis-testing. In the interests of space, we do not
report on nonsignificant moderator analyses.
Secondary outcome variables. GLMMs were also applied to
examine the effects of condition on secondary outcome variables:
service usage, revictimization, and stage of change.
Results
Participant Demographics
Participants (M age ⫽ 33.8 years, SD ⫽ 11.1) described their
ethnic backgrounds as 33% White/Caucasian, 29% Black or Af-
rican American, 2% Asian/Asian American or Pacific Islander,
14% American Indian or Alaskan Native, and 42% Hispanic/
Latina. Across ethnic group categories, 76% of women identified
with at least one ethnic minority group. To capture socioeconomic
status in a single global score, a principal component analysis
using orthogonal rotation was applied to education, occupation,
and income variables in the full sample. A single component
solution emerged (all component loadings above .75); each indi-
vidual’s factor score was used in analyses. Median yearly net
income was $7,644, with a range of $0 to $108,000. Fifty-one
percent of women indicated that they lived with the offender at the
time of the incident. Twenty-two percent of women reported some
degree of physical dependence on the offender, whereas 50% of
women reported some degree of economic dependence. For addi-
tional details about demographics, see DePrince et al. (in press).
We used spatial data to explore the representativeness of the
sample. Using geo-coded addresses, we compared where partici-
pants reported living at the time of the incident with the addresses
of all IPA incidents reported across the county during the recruit-
ment period (see DePrince et al., in press, for additional details
about this comparison). As illustrated in Figure 2, participants
reported addresses that appear to represent the spatial locations of
IPA incident reports and, therefore, spatially relevant demographic
variables (e.g., ethnicity, income).
Primary Outcome Variables: Effects of Condition
(Outreach, Referral) and Moderators
As detailed in DePrince et al. (in press), Outreach and Referral
conditions were equivalent on demographic variables (e.g., ethnic
minority status, age, education, occupation, and income) as well as
IPA-relevant variables (e.g., description of relationship to of-
fender, experiences of IPA in the 6 months prior to the target
incident). Figure 3 provides estimated means and standard errors
for primary outcomes.
Table 1
Within- and Between-Group Effect Sizes and 95% Confidence Intervals Stratified by Moderators
Variable Intervention group T1–T3 T1–T2 T2–T3
Perceptions of physical dependence on offender moderates group effects on fear
Total physical dependence Outreach ⫺0.23 (⫺.54, .08) ⫺.28 (⫺.59, .03) 0.07 (⫺0.24, 0.38)
Referral 0.01 (⫺.38, .40) .03 (⫺.36, .42) ⫺0.01 (⫺0.40, 0.37)
Outreach minus Referral ⫺0.29 (⫺.64, .06) ⫺.32 (⫺.68, .03) 0.08 (⫺0.27, 0.44)
Partial physical dependence Outreach ⫺0.10 (⫺0.41, 0.21) ⫺0.18 (⫺0.49, 0.13) 0.10 (⫺0.21, 0.41)
Referral 0.09 (⫺0.30, 0.48) 0.13 (⫺0.26, 0.52) ⫺0.07 (⫺0.46, 0.32)
Outreach minus Referral ⫺0.01 (⫺0.36, 0.34) ⫺0.20 (⫺0.55, 0.16) 0.15 (⫺0.20, 0.50)
No physical dependence Outreach 0.75 (0.43, 1.07) 0.59 (0.28, 0.91) 0.12 (⫺0.19, 0.43)
Referral 0.40 (0.01, 0.80) 0.56 (0.16, 0.95) ⫺0.16 (⫺0.54, 0.23)
Outreach minus Referral 0.33 (⫺0.03, 0.68) 0 (⫺0.35, 0.35) 0.34 (⫺0.01, 0.70)
Ethnicity moderates group effects on fear
Ethnic minority Outreach 0.74 (0.42, 1.06) 0.62 (0.30, 0.94) 0.12 (⫺0.19, 0.43)
Referral 0.40 (0.01, 0.79) 0.47 (0.07, 0.86) ⫺0.05 (⫺0.44, 0.34)
Outreach minus Referral 0.31 (⫺0.05, 0.66) 0.12 (⫺0.23, 0.48) 0.18 (⫺0.17, 0.53)
Not ethnic minority Outreach 0.25 (⫺0.06, 0.56) 0.08 (⫺0.23, 0.39) 0.17 (⫺
0.14, 0.48)
Referral 0.20 (⫺0.19, 0.59) 0.52 (0.13, 0.92) ⫺0.31 (⫺0.70, 0.08)
Outreach minus Referral 0.07 (⫺0.28, 0.42) ⫺0.01 (⫺0.36, 0.34) 0.47 (0.12, 0.83)
Note. The parameters represent effect size of change, with 95% confidence intervals in parentheses. Effect size was calculated using the parameter
estimates and standard error from the general linear mixed model.
216
DEPRINCE ET AL.
PTSD total symptoms. PTSD symptom severity scores de-
creased significantly T1 to T2 for Outreach, t(206) ⫽ 3.33, p ⫽
.001, Cohen’s d ⫽ 0.34 (0.03, 0.66), and Referral, t(211) ⫽ 4.48,
p ⬍ .0001, d ⫽ 0.60 (0.20, 1.00), conditions (see Figure 3A).
Differences in symptom change from T1 to T2 in Referral relative
to Outreach conditions did not achieve statistical significance,
t(207) ⫽ –1.59, p ⫽ .11, d ⫽ –0.29 (– 0.68, 0.11). From T2 to T3,
reduction in symptom severity was maintained among women in
the Outreach condition, d ⫽ 0.08 (– 0.02, 0.39), but symptom
severity increased for the Referral condition, t(204) ⫽ –2.11, p ⫽
.04, d ⫽ – 0.28(–.0.67, 0.11). As such, we observed significant
differences between Outreach and Referral conditions in symptom
change T2 to T3, t(203) ⫽ 2.18, p ⫽ .03, d ⫽ 0.40 (0.04, 0.75). For
both interventions, within-group effect size change from T1 to T3
in PTSD symptom severity was moderate: Outreach condition d ⫽
0.43 (0.11, 0.74), and Referral condition d ⫽ 0.31 (– 0.08, 0.70).
RCI of 8.77 was calculated based on a reliability sample, where
Cronbach’s alpha for PTSD ⫽ 0.93 (SD ⫽ 11.96; DePrince et al.,
2011). For the Outreach condition, 22 (34%) of 64 individuals
demonstrated reliable change at Time 3. For the Referral condi-
tion, only 11 (29%) of 38 individuals showed reliable change.
Fisher’s exact test indicated no significant differences between
groups on RCI for PTSD (
2
⫽ 2.32, p ⫽ .14). Relative risk
computations indicated that reliable change is twice as likely for
Outreach compared with Referral.
Depression symptoms. Depressive symptoms significantly
decreased from T1 to T2 in the Outreach, t(204) ⫽ 2.16, p ⫽
.0319, d ⫽ 0.22 (–0.09, 0.53), and Referral, t(212) ⫽ 3.84, p ⫽
.0002, d ⫽ 0.52 (– 0.12, 0.91), conditions. As illustrated in Figure
3B, symptom change from T1 to T2 showed a trend to be greater
in the Referral condition compared with the Outreach condition,
t(209) ⫽ –1.81, p ⫽ .07, d ⫽ – 0.33 (– 0.069, 0.02). From T2 to T3,
women in the Outreach condition showed a trend for continued
improvement, t(199) ⫽ 1.51, p ⫽ .13, d ⫽ 0.15 (– 0.16, 0.46),
whereas women in the Referral condition appeared to worsen,
t(365) ⫽ –1.54, p ⫽ .12, d ⫽ – 0.21 (– 0.60, 0.18). Because of
relapse of depressive symptoms in the Referral condition, compar-
ison of symptom change T2 to T3 was significantly different,
t(202) ⫽ 2.13, p ⫽ .03, d ⫽ 0.39 (0.04, 0.75). Overall, from T1 to
T3, both Outreach, t(206) ⫽ 3.61, p ⫽ .0004, d ⫽ 0.37 (0.06,
0.69), and Referral, t(210) ⫽ 2.26, p ⫽ .02, d ⫽ 0.30 (– 0.09, 0.69),
conditions showed moderate decreases in depressive symptoms.
RCI of 8.50 was calculated based on a reliability sample where
Cronbach’s alpha for depression ⫽ 0.90 (SD ⫽
9.76; DePrince et
al., 2011). For the Outreach condition, 17 (28%) of 61 individuals
demonstrated reliable change at Time 3, whereas for the Referral
condition, 5(14%) of 35 individuals showed reliable change. Fish-
er’s exact test indicated no significant differences between the
groups on RCI for depression (
2
⫽ 2.32, p ⫽ .14).
Fear. Similar decreases from T1 to T2 in fear symptoms,
t(206) ⫽ .55, p ⫽ .58, were observed for both the Outreach,
t(205) ⫽ 5.26, p ⬍ .0001, d ⫽ 0.59 (0.28, 0.91), and Referral,
t(206) ⫽ 4.69, p ⬍ .0001, d ⫽ 0.67 (0.27, 1.07), conditions (see
Figure 3C). From T2 to T3, there was a trend for continued
reduction in symptoms for the Outreach condition, t(199) ⫽ 1.68,
p ⫽ .09, d ⫽ 0.18 (– 0.13, 0.49), concurrent with a trend for relapse
of fear symptoms in the Referral condition, t(201) ⫽ –1.48, p ⫽
.14, d ⫽ – 0.21(– 0.61, 0.18), resulting in significant differences
between Outreach and Referral for fear symptom change T2 to T3,
t(200) ⫽ 2.19, p ⫽ .03, d ⫽ 0.40 (0.05, 0.76). Because of the
reinstatement of fear symptoms for the Referral condition from T2
to T3, there was a trend from T1 to T3, t(207) ⫽ 1.70, p ⫽ .09, d ⫽
– 0.31 (– 0.024, 0.66), for the Outreach condition, t(205) ⫽ 6.92,
p ⬍ .0001, d ⫽ 0.77 (0.45, 1.09), to report greater reductions in
fear symptoms than the Referral condition, t(208) ⫽ 3.18, p ⫽
.0017, d ⫽ 0.45 (0.06, 0.85). RCI of .89 was calculated based on
a reliability sample, where Cronbach’s alpha for fear ⫽
0.89
(SD ⫽ 0.97; DePrince et al., 2011). For the Outreach condition, 31
Figure 2. Participants’ addresses relative to location of overall intimate partner abuse (IPA) reports during the
recruitment period. Copyright 2011 by Susan Buckingham. Reprinted with permission.
217
IMPACT OF OUTREACH AFTER INTIMATE PARTNER ABUSE
(51%) of 61 individuals demonstrated reliable change at Time 3,
whereas for the Referral condition, only 10 (28%) of 36 individ-
uals showed reliable change. Fisher’s exact test indicated signifi-
cant differences between the groups on RCI for fear (
2
⫽ 4.93,
p ⫽ .03). Relative risk computations indicated that reliable change
is twice as likely for Outreach compared with Referral.
Ethnicity significantly moderated the quadratic, F(1, 199) ⫽
5.81, p ⫽ .02, and main effects, F(1, 119) ⫽ 3.46, p ⫽ .07, of
condition on TAQ fear scores (see Figure 4). For individuals
identifying as ethnic minorities, both interventions yielded mod-
erate to large reductions in symptoms—Outreach T1–T3 d ⫽
0.074 (0.42, 1.06), T1–T2 d ⫽ 0.62 (0.30, 0.94); Referral T1–T3
d ⫽ 0.40 (0.01, 0.79), T1–T2 d ⫽ 0.47 (0.07, 0.86)—with trends
suggesting Outreach had a greater effect than Referral: Outreach ⫺
Referral, T1–T3 d ⫽ 0.31 (– 0.05, 0.66). For ethnic majority
participants, only the Referral condition showed significant de-
crease in PTSD symptoms T1–T2, d ⫽ 0.53 (0.13, 0.92); however,
an apparent relapse in symptoms in these individuals, d ⫽ – 0.31
(– 0.70, 0.08), resulted in small overall effects for Referral, d ⫽
0.20 (–0.19, 0.59), compared with Outreach, d ⫽ 0.25 (– 0.06,
0.56).
Women’s perceptions of physical dependence on the offender
moderated the linear, F(1, 166) ⫽ 4.43, p ⫽ .04, effects of
Condition on TAQ fear scores. For individuals who report not
being at all dependent on the offender, both Outreach, d ⫽ 0.75
(0.43, 1.07), and Referral, d ⫽ 0.40 (0.01, 0.80), conditions led to
moderate-to-large reductions in fear over time; a trend suggested
the greatest reduction for the Outreach condition: Outreach ⫺
Referral, T1–T3 d ⫽ 0.33 (– 0.03,0.68). However, for individuals
who reported partial or total dependence on the offender, neither
Outreach nor Referral conditions had significant effects on fear.
Secondary Outcome Variables
Victim safety. In terms of continued aggression by target
offenders, 75% and 81% of women in the Outreach and Referral
conditions, respectively, reported at least one incident that in-
volved psychological aggression; 30% and 29%, respectively, re-
ported physical aggression; and 32% and 26%, respectively, re-
ported sexual aggression. In terms of aggression by new offenders,
24% of women in the Outreach condition and 24% of women in
the Referral condition reported psychological aggression; 15% and
10% reported physical aggression, respectively; and 8% and 7%
reported sexual aggression, respectively. Chi-square test of links
between the presence/absence of revictimization by Outreach/
Referral conditions for the target offender and new partners were
nonsignificant.
To characterize women’s plans for staying/leaving relationships
with the offender from the target incident, we examined Stages of
Change scores at T3. Compared with women in the Referral
condition (M ⫽ 3.43, SD ⫽ 1.86), women assigned to the Outreach
condition (M ⫽ 4.15, SD ⫽ 1.26) had higher Stages of Change
scores at T3, equal variances not assumed, t(56.60) ⫽ 2.07, p ⫽
.04. Higher scores indicate greater likelihood of having left the
relationship or having a plan to do, with a medium effect size,
Cohen’s d ⫽ 0.47 (0.06, 0.89).
Figure 3. Estimated means (with standard error bars) for primary out-
come variables by time and condition. PTSD ⫽ posttraumatic stress
disorder; O ⫽ community-based outreach; R ⫽ criminal justice system-
based referral.
Figure 4. Estimated means (with standard error bars) for moderator
analyses for fear by time and condition. O ⫽ community-based outreach;
R ⫽ criminal justice system-based referral.
218
DEPRINCE ET AL.
Service utilization. We examined contact with any of the
community-based agencies on the interdisciplinary team over time
but found no differences between the Outreach and Referral con-
ditions using chi-square analyses. At T1, 48% of women in the
Outreach condition and 38% of women in the Referral condition
reported contact with at least one community-based agency. Con-
tact with community agencies at T2 was made by 32% and 37% of
women in the Outreach and Referral conditions, respectively; and
at T3 contact was made by 19% and 13% of women in the
Outreach and Referral conditions, respectively. Likewise, groups
did not differ in the likelihood that they spoke with a therapist over
the follow-up period: At T2, 41% and 33% women in the Outreach
and Referral conditions, respectively, reported contact with a ther-
apist; those rates were 30% and 37%, respectively, at T3. The
groups did differ, however, in how helpful they found their con-
tacts with system- and community-based CCR team members,
t(76) ⫽ 2.28, p ⫽ .03. According to average ratings of helpfulness
of contacts with community-based agencies and the Denver Police
Department Victim Assistance Unit at T2 and T3, women in the
Outreach condition reported greater perceptions of the helpfulness
of contacts (n ⫽ 47; M ⫽ 3.67, SD ⫽ 0.79) than women in the
Referral condition (n ⫽ 31; M ⫽ 3.21, SD ⫽ 0.98). The effect size
was medium, Cohen’s d ⫽ 0.53 (0.07, 0.98). In contrast, groups
did not differ in how helpful they rated contacts with therapists
across the follow-up period (Outreach n ⫽ 30, M ⫽ 3.98, SD ⫽
0.89; Referral n ⫽ 18, M ⫽ 3.92, SD ⫽ 1.15).
Discussion
Using a randomized control, longitudinal design, this research
demonstrates that community-based Outreach from victim advo-
cates for women exposed to police-reported IPA results in de-
creases in distress 1 year later. Women in the community-based
Outreach condition reported greater decreases in both PTSD and
depressive symptom severity and fear 1 year later, compared with
women in the Referral condition. In contrast, women in the Re-
ferral condition reported increases in symptoms from T2 to T3,
suggesting that community outreach by system victim advocates
may have important prevention influences long after the abuse
occurs. These findings point to the importance of future research
on the role that community-based victim advocates can play in
addressing distress following IPA in addition to other more fre-
quently studied responders, such as therapists and medical provid-
ers.
Moderator analyses revealed that community-based outreach
was almost three times more effective for ethnic minority women
in decreasing fear, compared with White women. Although mod-
erate to large decreases in fear were observed among ethnic
minority women, only small decreases were observed among
White women. This finding may have particularly important im-
plications for the intersection of psychological well-being and
criminal justice participation. Fear is important not only to psy-
chological outcomes (DePrince et al., 2010) but also criminal
justice participation (Belknap & Sullivan, 2003; Fischer & Rose,
1995; Goodman, Bennett, & Dutton, 1999). Given the overrepre-
sentation of ethnic minority women in the criminal justice system
as victims (Gabbidon & Green, 2008), outreach may be valuable
for decreasing victim fear and increasing women’s engagement in
the criminal justice system.
Community-based outreach was less effective for women who
perceived greater physical dependence on the offender, relative to
their peers; however, economic dependence on the offender did not
moderate the impact of outreach. The absence of an effect for
economic dependence is interesting because survivors who are
economically dependent on their abusers have been found to be
less likely to engage the criminal legal system in the future (e.g.,
Erez & Belknap, 1998; Fleury-Steiner, Bybee, Sullivan, Belknap,
& Melton, 2006). Further, economic disadvantage has been found
to increase women’s risk of violence among intimate partners
(Benson & Fox, 2004). However, we did not see evidence of the
impact of economic dependence on psychological distress here.
Instead, we found that physical dependence moderated the impact
of the intervention. To our knowledge, physical dependence on the
offender has not received attention in the empirical literature.
Thus, future work should identify the mechanisms by which phys-
ical dependence on the offender exerts an influence on women’s
well-being and interventions that can effectively support these
women.
Women faced significant ongoing danger following police-
reported IPA. Even though this sample was recruited from incident
reports indicating nonsexual IPA, nearly one third of women in the
sample reported that the target offender engaged in at least one
sexually aggressive tactic in the year following the T1 interview.
For almost one third of the women, conflicts with the offender
continued to be physically aggressive. Roughly one quarter of
women reported that a new partner engaged in at least one psy-
chologically aggressive tactic. In addition, 7–15% of women re-
ported physically aggressive or sexually aggressive behaviors by
new partners. Thus, women continued to face significant safety
risks over the follow-up year.
We originally predicted that outreach would increase victim
safety by preventing revictimization by the abuser; however, we
came to think differently about this prediction with hindsight.
Although other research suggests that contacts with advocates
and/or outreach are associated with positive effects in terms of
victim engagement with the criminal legal system (DePrince et al.,
in press) and improved victim well-being (reported here), these
things are, to some degree, within women’s capacity to influence.
For example, women can exert control over decisions to engage
with the legal system and to seek services and/or other support that
may mitigate symptoms. Women, however, cannot control their
partners’ behaviors. Thus, predicting that outreach to victims
would increase safety failed to take into account specific domains
in which women may or may not be able to effect change in their
lives. Although outreach did not affect victim safety, women
assigned to the Outreach (relative to Referral) condition had higher
stage of change scores, indicating they were more likely to artic-
ulate plans to leave or to have already left the offender at T3. This
suggests that outreach is associated with changes that are more
directly in women’s control (e.g., plans to leave) than in offenders’
control (e.g., whether he engages in aggressive behaviors).
Across the two conditions, women did not differ in terms of
contact with community-based agencies or therapists during the
study period. Community-based Outreach did not improve on
women’s overall access to victim advocacy or therapy services;
however, the groups did differ in their perceptions of how helpful
advocacy services were (but not therapy). In particular, women
assigned to the Outreach condition reported that their victim-
219
IMPACT OF OUTREACH AFTER INTIMATE PARTNER ABUSE
service contacts were more helpful than did women in the Referral
condition. Community-based outreach was implemented by an
agency selected based on each victim’s needs; thus, women may
have connected more quickly or more successfully with relevant
services, which helped their coping and well-being overall (as
reflected in reduced symptoms of distress, for example). Women
in the Referral condition may have failed to find the fit required to
address their specific needs as they worked their way through the
complex world of victim services.
Study Limitations
We did not have data on women’s contact with system- or
community-based personnel prior to T1; therefore, we do not know
what (if any) effect the initial contact with system-based advocates
had on women prior to T1. We had no feasible way of interviewing
women prior to their first contact with system-based advocates
(e.g., police department advocates sometimes go to the scene of
incidents). However, because both groups were contacted by a
system-based advocate, this initial contact was held constant
across groups. We had no way of testing the randomization to
groups or controlling for preexisting conditions; however, groups
were equivalent on key variables, including reports of exposure to
violence in the 6 months prior to the target incident (DePrince et
al., in press).
Conclusions
The current study provides evidence of positive effects of early,
coordinated, victim-focused outreach on women’s well-being:
Community-based outreach from victim advocates through a CCR
was linked to decreases in IPA-related mental health problems,
including PTSD and depression symptom severity as well as fear.
Strikingly, community-based outreach by victim advocates was
almost three times more effective for ethnic minority women in
decreasing fear compared with White women, whereas physical
dependence on the offender mitigated the impact of outreach.
Although outreach was linked to greater readiness to leave the
offender, women in both conditions reported that ongoing aggres-
sion by the original offender was all too common. The findings
from this study address historical gaps between victim advocates
and mental health providers, demonstrating that research and pol-
icy attention should be paid to understanding the role that
community-based outreach and interdisciplinary CCRs can play in
addressing IPA.
References
Allen, N. E., Bybee, D. I., & Sullivan, C. M. (2004). Battered women’s
multitude of needs: Evidence supporting the need for comprehensive
advocacy. Violence Against Women, 10, 1015–1035. doi:10.1177/
1077801204267658
American Psychiatric Association. (1994). Diagnostic and statistical man-
ual of mental disorders (4th ed.). Washington, DC: Author.
Beck, A. T., Steer, R. A., Ball, R., & Ranieri, W. (1996). Comparison of
Beck Depression Inventories–IA and –II in psychiatric outpatients.
Journal Personality Assessment, 67, 588 –597. doi:10.1207/
s15327752jpa6703_13
Belknap, J., & Sullivan, C. M. (2003). Longitudinal study of battered
women in the system: The victims’ and decision-makers’ perceptions:
Final report (Document No. 202946). Washington, DC: National Insti-
tute of Justice. Retrieved from http://www.ncjrs.gov/pdffiles1/nij/grants/
202946.pdf
Benson, M. L., & Fox, G. L. (2004). Concentrated disadvantage, economic
distress, and violence against women in intimate relationships. In B. S.
Fisher (Ed.), Violence against women and family violence: Develop-
ments in research, practice, and policy (NIJ100709). Washington, DC:
National Criminal Justice Resource Center.
Burke, J. G., Mahoney, P., Gielen, A., McDonnell, K. A., & O’Campo, P.
(2009). Defining appropriate stages of change for intimate partner vio-
lence survivors. Violence and Victims, 24, 36 –51. doi:10.1891/0886-
6708.24.1.36
Bybee, D. I., & Sullivan, C. M. (2002). The process through which an
advocacy intervention resulted in positive change for battered women
over time. American Journal of Community Psychology, 30, 103–132.
doi:10.1023/A:1014376202459
Bybee, D. I., & Sullivan, C. M. (2005). Predicting re-victimization of
battered women three years after exiting a shelter program. American
Journal of Community Psychology, 36, 85–96. doi:10.1007/s10464-005-
6234-5
Campbell, R., Greeson, M. R., Bybee, D., & Raja, S. (2008). The co-
occurrence of childhood sexual abuse, adult sexual assault, intimate
partner violence, and sexual harassment: A mediational model of post-
traumatic stress disorder and physical health outcomes. Journal of Con-
sulting and Clinical Psychology, 76, 194 –207. doi:10.1037/0022-
006X.76.2.194
Cattaneo, L. B., & DeLoveh, H. L. M. (2008). The role of socioeconomic
status in helpseeking from hotlines, shelters, and police among a national
sample of women experiencing intimate partner violence. American
Journal of Orthopsychiatry, 78, 413–422. doi:10.1037/a0014558
Cattaneo, L. B., & Goodman, L. A. (2005). Risk factors for reabuse in
intimate partner violence: A cross-disciplinary critical review. Trauma,
Violence, & Abuse, 6, 141–175. doi:10.1177/1524838005275088
Cattaneo, L. B., & Goodman, L. A. (2010). Through the lens of therapeutic
jurisprudence the relationship between empowerment in the court sys-
tem and well-being for intimate partner violence victims. Journal of
Interpersonal Violence, 25, 481–502. doi:10.1177/0886260509334282
Classen, C. C., Palesh, O. G., & Aggarwal, R. (2005). Sexual revictimiza-
tion: A review of the empirical literature. Trauma, Violence, & Abuse, 6,
103–129. doi:10.1177/1524838005275087
Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155–159.
doi:10.1037/0033-2909.112.1.155
Coker, A. L., Smith, P. H., Thompson, M. P., McKeown, R. E., Bethea, L.,
& Davis, K. E. (2002). Social support protects against the negative
effects of partner violence on mental health. Journal of Women’s
Health & Gender-Based Medicine, 11, 465– 476. doi:10.1089/
15246090260137644
DePrince, A. P. (2005). Social cognition and revictimization risk. Journal
of Trauma and Dissociation, 6, 125–141. doi:10.1300/J229v06n01_08
DePrince, A. P., Belknap, J., Labus, J., Buckingham, S. E., & Gover, A. R.
(in press). The impact of victim-focused outreach on criminal legal
system outcomes following police-reported intimate partner abuse. Vi-
olence Against Women.
DePrince, A. P., Chu, A. T., & Pineda, A. S. (2011). Links between
specific posttrauma appraisals and three forms of trauma-related distress.
Psychological Trauma: Theory, Research, Practice, and Policy. Ad-
vance online publication. doi:10.1037/a0021576
DePrince, A. P., Zurbriggen, E. L., Chu, A. T., & Smart, L. (2010).
Development of the Trauma Appraisal Questionnaire. Journal of
Aggression, Maltreatment & Trauma, 19, 275–299. doi:10.1080/
10926771003705072
DiClemente, C. C., Prochaska, J. O., Fairhurst, S., Velicer, W. F., Rossi,
J. S., & Velasquez, M. (1991). The process of smoking cessation: An
analysis of precontemplation, contemplation and contemplation/action.
220
DEPRINCE ET AL.
Journal of Consulting and Clinical Psychology, 59, 295–304. doi:
10.1037/0022-006X.59.2.295
Erez, E., & Belknap, J. (1998). In their own words: Battered women’s
assessment of the criminal processing system’s responses. Violence and
Victims, 13, 251–268.
Evans, C., Margison, F., & Barkham, M. (1998). The contribution of
reliable and clinically significant change methods to evidence-based
mental health, Evidence-Based Mental Health, 1, 70 –72. doi:10.1136/
ebmh.1.3.70
Fanslow, J. L., & Robinson, E. M. (2010). Help-seeking behaviors and
reasons for help seeking reported by a representative sample of women
victims of intimate partner violence in New Zealand. Journal of Inter-
personal Violence, 25, 929 –951. doi:10.1177/0886260509336963
Fischer, K., & Rose, M. (1995). When “enough is enough”: Battered
women’s decision-making around court orders of protection. Crime &
Delinquency, 41, 414 – 429. doi:10.1177/0011128795041004003
Fleury-Steiner, R., Bybee, D. I., Sullivan, C. M., Belknap, J., & Melton,
H. C. (2006). Contextual factors impacting battered women’s intentions
to re-use the criminal legal system. Journal of Community Psychology,
34, 327–342. doi:10.1002/jcop.20102
Foa, E. B., Cashman, L., Jaycox, L., & Perry, K. (1997). The validation of
a self-report measure of posttraumatic stress disorder: The posttraumatic
diagnostic scale. Psychological Assessment, 9, 445– 451. doi:10.1037/
1040-3590.9.4.445
Gabbidon, S. L., & Green, H. T. (2008). Race and crime (2nd ed.).
Thousand Oaks, CA: Sage.
Gillum, T. L., Sun, C. J., & Woods, A. B. (2009). Can a health clinic-based
intervention increase safety in abused women? Results from a pilot
study. Journal of Women’s Health, 18, 1259 –1264. doi:10.1089/
jwh.2008.1099
Goodman, L., Bennett, L., & Dutton, M. A. (1999). Obstacles to victim’s
cooperation with the criminal prosecution of their abusers: The role of
social support. Violence and Victims, 14, 427–444.
Goodman, L., & Epstein, D. (2005). Refocusing on women: A new
direction for policy and research on intimate partner violence. Journal of
Interpersonal Violence, 20, 479 – 487. doi:10.1177/0886260504267838
Holroyd K. A., Labus J. S., & Carlson, B. (2009). Moderation and medi-
ation in the psychological and drug treatment of chronic tension-type
headache: The role of disorder severity and psychiatric comorbidity.
Pain, 143, 213–222. doi:10.1016/j.pain.2009.02.019
Johnson, D. M., Zlotnick, C., & Perez, S. (2008). The relative contribution
of abuse severity and PTSD severity on the psychiatric and social
morbidity of battered women in shelters. Behavior Therapy, 39, 232–
241. doi:10.1016/j.beth.2007.08.003
Kendall, J., Pelucio, M. T., Casaletto, J., Thompson, K. P., Barnes, S.,
Petit, E., & Aldrich, M. (2009). Impact of emergency department inti-
mate partner violence intervention. Journal of Interpersonal Violence,
24, 280 –306. doi:10.1177/0886260508316480
Klevens, J., Baker, C. K., Shelley, A., & Ingram, E. M. (2008). Exploring
the links between components of coordinated community responses and
their impact on contact with intimate partner violence services. Violence
Against Women, 14, 346 –358. doi:10.1177/1077801207313968
Kraemer, H. C., Frank, E., & Kupfer, D. J. (2006). Moderators of treatment
outcomes: Clinical, research, and policy importance. Journal of the
American Medical Association, 296, 1286 –1289. doi:10.1001/
jama.296.10.1286
Kraemer, H., Wilson, G., Fairburn, C., & Agras, W. (2002). Mediators and
moderators of treatment effects in randomized clinical trials. Archives of
General Psychiatry, 59, 877– 883.
Krause, E. D., Kaltman, S., Goodman, L., & Dutton, M. A. (2006). Role of
distinct PTSD symptoms in intimate partner reabuse: A prospective
study. Journal of Traumatic Stress, 19, 507–516. doi:10.1002/jts.20136
Krause, E. D., Kaltman, S., Goodman, L. A., & Dutton, M. A. (2008).
Avoidant coping and PTSD symptoms related to domestic violence
exposure: A longitudinal study. Journal of Traumatic Stress, 21, 83–90.
doi:10.1002/jts.20288
Mechanic, M. B., Weaver, T. L., & Resick, P. A. (2008). Mental health
consequences of intimate partner abuse: A multidimensional assessment
of four different forms of abuse. Violence Against Women, 14, 634 –654.
doi:10.1177/1077801208319283
Prochaska, J. O., & DiClemente, C. C. (1983). Stages and processes of
self-change for smoking: Toward an integrative model of change. Jour-
nal of Consulting and Clinical Psychology, 51, 390 –395. doi:10.1037/
0022-006X.51.3.390
Rennison, C. M., & Welchans, S. (2000).
Intimate partner violence.
Washington, DC: Bureau of Justice Statistics, United States Department
of Justice.
Robinson, A. L., & Tregidga, J. (2007). The perceptions of high-risk
victims of domestic violence to a coordinated community response in
Cardiff, Wales. Violence Against Women, 13, 1130 –1148. doi:10.1177/
1077801207307797
Shurman, L. A., & Rodriguez, C. M. (2006). Cognitive-affective predictors
of women’s readiness to end domestic violence. Journal of Interpersonal
Violence, 21, 1417–1439. doi:10.1177/0886260506292993
Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. (1996).
The revised Conflict Tactics Scale (CTS2). Journal of Family Issues, 17,
283–316. doi:10.1177/019251396017003001
Sullivan, C. M., & Bybee, D. I. (1999). Reducing violence using
community-based advocacy for women with abusive partners. Journal of
Consulting and Clinical Psychology, 67, 43–53. doi:10.1037/0022-
006X.67.1.43
Sullivan, C. M., Bybee, D. I., & Allen, N. E. (2002). Findings from a
community-based program for battered women and their children. Jour-
nal of Interpersonal Violence, 17, 915–936. doi:10.1177/
0886260502017009001
Tjaden, P., & Thoennes, N. (2000). Prevalence and consequences of
male-to-female and female-to-male partner violence as measured by the
National Violence Against Women Survey. Violence Against Women, 6,
142–161. doi:10.1177/10778010022181769
Received January 10, 2011
Revision received December 15, 2011
Accepted December 22, 2011 䡲
221
IMPACT OF OUTREACH AFTER INTIMATE PARTNER ABUSE
A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Consulting and Clinical Psychology
This content is subject to copyright. Terms and conditions apply.
