Where do we go from here? Interim analysis to forge ahead in violence prevention.

Abstract

The severity and disparity of interpersonal violent injury is staggering. Fifty-three per 100,000 African Americans (AA) die of homicide yearly, 20 per 100,000 in Latinos, whereas the rate is 3 per 100,000 in Caucasians. With the ultimate goal of reducing injury recidivism, which now stands at 35% to 50%, we have designed and implemented a hospital-based, case-managed violence prevention program uniquely applicable to trauma centers. The Wraparound Project (WP) seizes the "teachable moment" after injury to implement culturally competent case management (CM) and shepherd clients through risk reduction resources with city and community partners. The purpose of this study was to perform a detailed intermediate evaluation of this multi-modal violence prevention program. We hypothesized that this evaluation would demonstrate feasibility and early programmatic efficacy. We looked to identify areas of programmatic weakness that, if corrected, could strengthen the project and enhance its effectiveness.
We performed intermediate evaluation on the 18-month-old program. We selected the Centers for Disease Control and Prevention-recommended instrument used for unintentional injury prevention programs and applied it to the WP. The four sequential stages in this methodology are formative, process, impact, and outcome. To test feasibility of WP, we used process evaluation. To evaluate intermediate goals of risk reduction and early efficacy, we used impact evaluation.
Four hundred thirty-five people met screening criteria. The two case managers were able to make contact and screen 73% of gun shot victims, and 57% of stab wound victims. Of those not seen, 48% were in the hospital for <or=2 days. Fifty-four percent of those screened had identified needs and received CM services. Thirteen percent refused services. Of the high-risk clients receiving full services (N = 45), 60% were AA and 30% were Latino. Sixty percent of the AA had no contact with their fathers. CM "dose": In the first 3 weeks of enrollment, 40% of the time, case managers spent >6 h/wk with the client. Forty-one percent of the time, they spent 3 hours to 6 hours. Seventeen of 18 people who required >6 hours had two to three needs. Attrition rate is only 4%. The table demonstrates percent success thus far in providing risk reduction resources.
WP case managers served high-risk clients by developing trust, credibility, and a risk reduction plan. Cultural competency has been vital. Six of seven major needs were successfully addressed at least 50% of the time. The value of reporting these results has led WP to gain credibility with municipal stakeholders, who have now agreed to fund a third CM position. Intermediate evaluation provided a framework in our effort to achieve the ultimate goal of reducing recidivism through culturally competent CM and risk factor modification.

Full text

Page 1

ORIGINAL ARTICLE
Where Do We Go From Here? Interim Analysis to Forge Ahead in
Violence Prevention
Rochelle A. Dicker, MD, Sebastian Jaeger, MS, Mary M. Knudson, MD, Robert C. Mackersie, MD,
Diane J. Morabito, RN, MPH, Javier Antezana, and Michael Texada
Background: The severity and disparity of interpersonal violent injury is
staggering. Fifty-three per 100,000 African Americans (AA) die of homicide
yearly, 20 per 100,000 in Latinos, whereas the rate is 3 per 100,000 in
Caucasians. With the ultimate goal of reducing injury recidivism, which now
stands at 35% to 50%, we have designed and implemented a hospital-based,
case-managed violence prevention program uniquely applicable to trauma
centers. The Wraparound Project (WP) seizes the “teachable moment” after
injury to implement culturally competent case management (CM) and shep-
herd clients through risk reduction resources with city and community
partners. The purpose of this study was to perform a detailed intermediate
evaluation of this multi-modal violence prevention program. We hypothe-
sized that this evaluation would demonstrate feasibility and early program-
matic efficacy. We looked to identify areas of programmatic weakness that,
if corrected, could strengthen the project and enhance its effectiveness.
Methods: We performed intermediate evaluation on the 18-month-old
program. We selected the Centers for Disease Control and Prevention-
recommended instrument used for unintentional injury prevention pro-
grams and applied it to the WP. The four sequential stages in this
methodology are formative, process, impact, and outcome. To test fea-
sibility of WP, we used process evaluation. To evaluate intermediate
goals of risk reduction and early efficacy, we used impact evaluation.
Results: Four hundred thirty-five people met screening criteria. The two case
managers were able to make contact and screen 73% of gun shot victims, and
57% of stab wound victims. Of those not seen, 48% were in the hospital for
?2 days. Fifty-four percent of those screened had identified needs and
received CM services. Thirteen percent refused services. Of the high-risk
clients receiving full services (N ? 45), 60% were AA and 30% were Latino.
Sixty percent of the AA had no contact with their fathers. CM “dose”: In the
first 3 weeks of enrollment, 40% of the time, case managers spent ?6 h/wk
with the client. Forty-one percent of the time, they spent 3 hours to 6 hours.
Seventeen of 18 people who required ?6 hours had two to three needs.
Attrition rate is only 4%. The table demonstrates percent success thus far in
providing risk reduction resources.
Conclusions: WP case managers served high-risk clients by developing
trust, credibility, and a risk reduction plan. Cultural competency has been
vital. Six of seven major needs were successfully addressed at least 50% of
the time. The value of reporting these results has led WP to gain credibility
with municipal stakeholders, who have now agreed to fund a third CM
position. Intermediate evaluation provided a framework in our effort to
achieve the ultimate goal of reducing recidivism through culturally compe-
tent CM and risk factor modification.
Key Words: Violence prevention, Health disparities, Hospital-based vio-
lence prevention, Public health model.
(J Trauma. 2009;67: 1169–1175)
I
According to the Centers for Disease Control, homicide was
responsible for 18,124 deaths in 2005. This represents nearly
600,000 potential life years lost, giving credence to the
concern that interpersonal violence disproportionately affects
our young people.1Homicide is the second leading cause of
death in people 15 years to 24 years of age and third in those
25 years to 34 years of age. Disadvantaged minority popula-
tions are disproportionately represented. Homicide is the
main cause of death in African Americans (AA) aged 10
years to 24 years old and second most among Hispanics.
Fatalities from assault represent the tip of the iceberg; non-
fatal injuries are believed to outnumber fatal injuries on the
order of 100 to 1.2As a result of the tremendous societal
effects of violent injury, violence prevention is considered a
fundamental goal of “Healthy People 2010.”
The rate of injury recidivism from interpersonal vio-
lence is 35% to 50% nationally.3With the ultimate purpose of
reducing injury and criminal recidivism, we designed and
implemented a hospital-based and community driven vio-
lence prevention program (VPP), the Wraparound Project
(WP) for Comprehensive Rehabilitation. The conceptual
model of Wraparound is based on three critical components:
(1) The Public Health Model for injury prevention succeeds
based on evidence that addressing root causes and risk factors
of violence can prevent future injury and incarceration.4(2)
Health communication in this country is marred by a lack of
cultural competency.5Our case managers function to undue
this nationally identified health care flaw by providing solid
understanding, cultural competency, and the capacity to es-
tablish trust between client and the case manager. (3) We
believe that when a major event like trauma occurs, it pro-
vides a golden window, a teachable moment in which some-
one is more likely to participate in changes to secure health
and welfare for the future.6Many violently injured youth and
young adults are not provided “reentry” opportunities in a
like manner to the criminal justice system. Physical rehabil-
nterpersonal violent injury is pervasive in the United States,
and trauma centers stand on the front lines of the epidemic.
Submitted for publication March 9, 2009.
Accepted for publication August 17, 2009.
Copyright © 2009 by Lippincott Williams & Wilkins
From the Department of Surgery, University of California-San Francisco
and Division of Acute Care Surgery, San Francisco General Hospital, San
Francisco, California.
This article was scheduled to be presented at the 38th Annual Meeting of the
Western Trauma Association, February 24–March 1, 2008, Squaw Valley,
California, but due to unforeseen circumstances the paper was not presented.
Address for reprints: Rochelle A. Dicker, MD, Department of Surgery, 1001
Potrero Avenue, Ward 3A, San Francisco, CA 94110; email: dickerr@sfghsurg.
ucsf.edu; radicker@hotmail.com.
DOI: 10.1097/TA.0b013e3181bdb78a
The Journal of TRAUMA®Injury, Infection, and Critical Care • Volume 67, Number 6, December 2009
1169

Page 2

itation at our trauma centers is provided in the aftermath of
injury, however, providing services to reduce or eliminate
risk factors associated with violent injury are not traditionally
offered on hospital discharge. The WP serves as a vital point
of entry, provides mentorship, and shepherds clients to es-
sential risk-reduction resources with our city and community
partners. Finally, Wraparound provides long-term follow-up
and crisis intervention. As stated in the Committee on Trauma’s
Resources for the Optimal Care of the Injured Patient 2006,
“Institutions caring for injured patients can and should estab-
lish and aggressively pursue a leadership role in injury
prevention; they are in the best position to do so.”
The field of violence prevention and the programs
dedicated to this cause have suffered from the absence of a
template to measure feasibility, progress, shortfalls, and both
short- and long-term results. Programmatic evaluation is
helpful from inception to completion to create a systematic
way to critique a program and make necessary changes
during its evolution. Early and intermediate data are funda-
mental in maintaining a feasible and sustainable program that
appropriately serves the target population and meets the
expectations of all stakeholders.7,8Graduated evaluation can
also serve as a foundation by which a successful program can
be interpreted and recreated at another trauma center. Finally,
early and intermediate evaluation aids in efforts to secure fund-
ing and create a platform when addressing policy makers.
The evolution of Wraparound can be broken into three
phases. Phase 1 consisted of the early drafting and formatting
of the program and initiation of the pilot project, which was
funded with starter scholarship monies and fees generated
through professional income. A formative evaluation was
part of phase 1. Phase 2 included the launch of the full
program using funds provided through city government. Ad-
justments to programmatic activities based on the formative
evaluation from phase 1 were also part of this phase. An
intermediate evaluation, including both a process phase and
an impact phase were essential elements of phase 2 (see
Methods). Phase 3 (ongoing) includes efforts to expand the
program, and to finesse the elements necessary for exporta-
tion nationally. The final evaluation during phase 3 will be
essential in our efforts to secure stable funding for our
program and similar hospital-based VPP through regional and
national advocacy efforts. This article will focus on the
results of the intermediate evaluation conducted during phase
2 of the WP. To the best of our knowledge, this is the first
study to perform a detailed intermediate evaluation of a VPP.
We hope that it will serve as a template for other, like
programs focusing on the epidemic of interpersonal violence.
METHODS
Before implementation of the WP, we initiated a pilot
program. As a result, our violence prevention efforts became
very well regarded in the community. Due to this notoriety,
we could not conduct a randomized study with eligible
victims of interpersonal violence because we did not have
equipoise in the community the program serves. Instead, we
offer the WP to all eligible individuals treated at our level I
trauma center and implemented an evaluation process de-
signed to critique the feasibility, benchmarks, and long-term
outcomes of the program, as described below.
Eligibility criteria include individuals injured from in-
terpersonal and youth violence between the ages of 12 years
and 30 years. Individuals assessed to be victims of child
abuse or domestic violence were excluded from this study
and referred to appropriate providers. Two case managers
from Wraparound received referrals based simply on eligi-
bility criteria from Emergency Department and Ward Social
Workers, Trauma Nurse Practitioners, Resident, and Attend-
ing Physicians and Ward Nurses. These providers contacted
case managers by cellular phone. In addition, case managers
received a daily trauma log by which they recruited eligible
subjects by going directly to their bedsides to introduce the
program. Based on the case managers’ initial assessments of
risk for reinjury, individuals were placed in either a “high-
risk” category and therefore offered full Wraparound ser-
vices, or “low risk” and offered basic education on resources
such as the Victim of Crime Office of the District Attorney.
We recorded the latter group as our “one-time advocacy”
group. Evaluation of the WP was approved by the University
of California Institutional Review Board.
The structure of our evaluation process follows the
guidelines of two key publications7,9: Demonstrating Your
Program’s Worth, March 2000, Centers for Disease Control
and Prevention and “Framework for Program Evaluation in
Public Health,” MMWR; 17 September, 1999/Volume 48/No.
RR-11. Although the former is geared toward evaluating
programs for unintentional injury prevention, the basic ten-
ants of program analysis were implemented for VPP given
the similarities in public health principles. The evaluation
strategy breaks the process into four parts, allowing programs
to scrutinize activities at various junctures along the evolution
of the program, instead of waiting to find successes and
failures during a final outcome analysis. To date, there does
not exist a validated measure for evaluation specific for
violence prevention. The process included:
a. Formative evaluation: Utility is an early evaluation mea-
sure that we began to assess during phase 1 of the Wrap-
around Pilot Project. Utility refers to serving the needs of
a particular community identified in surveillance data. It is
through formative evaluation of our current program that
we identified the specific need for culturally sensitive case
managers for the Latino or a population in San Francisco,
and for the AA community. The specific Feasibility In-
strument that we have adapted is from work done through
the US Department of Health and Human Services in
Substance Abuse Prevention.
b. Process evaluation: Indicators are critical as short-term ele-
ments that are associated with programmatic interventions
and overall feasibility of the program. The following indica-
tors were assessed beginning at 6 months from initiation of
the program: Initial identification rate of eligible patients,
participation rate, early attrition rate, capacity to deliver
based on case manager’s availability, accountability, and
delivery of services by our city and community partners.
c. Impact evaluation: Intermediate indicators in violence pre-
vention such as employment status and changes in social
Dicker et al.
The Journal of TRAUMA®Injury, Infection, and Critical Care • Volume 67, Number 6, December 2009
© 2009 Lippincott Williams & Wilkins
1170

Page 3

circle are direct indicators of ongoing presence or absence
of risk factors associated with violent injury. The follow-
ing indicators were evaluated at 6 months after each
client-specific enrollment and were culled from the
follow-up screening session or exit interview of partici-
pants. The measures included assessment of the following:
Y Education status
Y School fighting
Y Employment status
Y Attrition rate from program
Y Social circle
Y Attrition from mental health services
Y Substance abuse
Y Fulfillment of other needs (driver license, relocation,
tattoo removal, etc.)
d. Outcome evaluation: This component will look at the
degree to which our intervention meets the ultimate goal:
reduction in reinjury based on reversal of risk factors.
Associated activities, such as incarceration history since
program enrollment, will also be recorded. The original
questionnaire used to establish a needs assessment for
each client will be the basis of the outcome evaluation.
This will allow for direct analysis of risk factor modifica-
tion in comparison with the intake interview. Outcome
evaluation will be the next step in analysis of the WP,
anticipated to be implemented when the program has been
running for 3 years to 4 years.
The two intermediate evaluations, process and impact,
were conducted during this study: to test intermediate feasi-
bility of the WP, we used process evaluation. To evaluate
intermediate goals of risk reduction and early programmatic
efficacy, we used impact evaluation. To aid in assessment of
risk, we utilized the Centers for Disease Control and Prevention-
validated Youth Risk Behavior Surveillance System, targeting
questions particularly relevant to risk for violent injury.10
This survey tracks health-risk behaviours, which contrib-
ute to the leading causes of morbidity and mortality in
youth and young adults. The behaviors are thought to be
interrelated and potentially preventable. The survey mon-
itors six categories of health risks including risks that con-
tribute to intentional injury. In our Wraparound survey, we
Figure 1. Wraparound program algorithm.
The Journal of TRAUMA®Injury, Infection, and Critical Care • Volume 67, Number 6, December 2009
Interim Analysis to Forge Ahead in Violence
Prevention
© 2009 Lippincott Williams & Wilkins
1171

Page 4

focused on violent injury-related behaviors when compiling
our needs assessments.
The structure of the WP is shown in Figure 1. Intensive
case management (CM) begins at the bedside. Once the client
is deemed “high risk” and the patient agrees to be in the
program, enrollment for “full services” ensues. The case
managers then follow clients beyond hospital discharge for 6
months to 1 year depending on the needs and progress of the
client. The partnership between the case manager and the
client is vital. The time case managers spend with clients, is
a valuable, and of course limited resource. We refer to this as
case manager “dose.” We measured time spent during the
most intense portion of program enrollment per client: the
first 3 weeks from initial contact. It includes personal and
phone contact, as well as advocacy.
In addition, it is very common for the case manager to
incorporate the support of family and friends on the client’s
behalf. CM included crisis management, home visits, phone
contacts, escorts to risk reduction resources, and recommen-
dations for family therapy if necessary. Based on the needs
assessment, the mentorship provided is coupled with other
risk reduction resources as outlined in this figure.
RESULTS
In conducting process evaluation, we found the follow-
ing results: 435 people met eligibility criteria at San Fran-
cisco General Hospital, the only City and County Level I
Trauma Center, in the first 12 months of the program. The
two case managers were able to make contact and assess 73%
of the gun shot victims and 57% of the stabbing victims.
Fifty-four percent of the individuals assessed had needs
identified and received one of two forms of CM services: for
lower-risk individuals, assistance with victim of crime ser-
vices within the city and some basic CM was provided. Risk
assessment was based on the criteria listed in the methods
section and a more subtle assessment by the case managers:
presence of certain tattoos, the mention of names of their
friends, where they spend their time and the result of frank
discussions. If deemed high-risk for injury recidivism, indi-
viduals received full programmatic services. Forty-five peo-
ple were assessed as high risk. Of those eligible but not
assessed by the case managers, 48% were in the hospital for
?2 days. Typically, by the time, the case managers were able
to come to the bedside to see the patient, he or she was
already discharged. This was particularly true for the week-
end admissions. Figure 2 profiles the outcomes of individuals
who were able to be assessed by the case managers.
The demographics of the 45 individuals who received
full CM services are seen in Table 1. Of note, nearly 60% of
the violently injured victims in this group are AA. According
to the San Francisco Violent Injury Surveillance System,11
60% of overall victims of violence in our city are also AA,
although this ethnic group makes up only 6% of the city’s
population.
Lack of education and unemployment are two character-
istics closely associated with violent injury risk. Of our high-risk
Figure 2. Subjects assessed by case managers.
Dicker et al.
The Journal of TRAUMA®Injury, Infection, and Critical Care • Volume 67, Number 6, December 2009
© 2009 Lippincott Williams & Wilkins
1172

Page 5

clients, 90% were unemployed at the time of injury. Figure 3
demonstrates the lack of education in our population.
In an effort to understand time expenditure per client,
we measured CM “dose,” as described above. We found that
40% of the time, case managers spent ?6 h/wk with an
individual client. Forty-one percent of the time, they spent 3 h
to 6 h/wk, and 19% of the time, they spent ?3 h/wk.
Seventeen of 18 people who required ?6 hours had two to
three service needs. A greater number of identified needs did
not correlate with greater CM time.
For impact evaluation, we assessed how well Wrap-
around was able to address risk factors specifically associated
with violent injury. Table 2 illustrates our early success rate
at placing our highest-risk clients in risk-reduction programs
and jobs. Fifty-five percent of our clients were on probation
when they entered the WP. Our success in Court Advocacy
was measured by our capacity to work with Probation Offic-
ers, Judges, and Public Defenders who entrusted Wraparound
case managers with supervisory roles. Success in obtaining
Driver Licenses was limited. This can be attributed to clients’
failure to address prior moving violations and illiteracy.
Successfully placing clients in General Education Develop-
ment programs or in schools where they could finish formal
secondary education was in large part due to case managers
working with School Counsellors and having schools willing
to take our clients in an alternative setting. Specific sites such
as Goodwill Industries and a local trucking company were
vital in helping with placement of our clients in solid jobs.
Finding safe and affordable housing in San Francisco has
been very labor intensive and has taken tremendous time by
the case managers. Finally, through a Memorandum of Un-
derstanding with a local mental health outpatient program,
The Trauma Recovery Center, we have been able to provide
clients with treatment for substance abuse and post-traumatic
stress disorder. The biggest hurdle is the cultural stigma
associated with seeking mental health care. In addition to the
services listed, we also have sent several clients to a city-
funded tattoo removal service. This is a critical step for many
TABLE 1. Demographics Data for Enrolled Clients
Age (yr)
Gender
Female
Male
Race/ethnicity
African American
Latino
Pacific Islander
White
Length of stay
ICU days
Hospital days
19 ? 3.86
6 (14)
38 (86)
26 (59)
13 (29)
2 (4)
3 (7)
3 ? 9.51
10 ? 15.33
Values are N ? SD or N (%).
ICU, intensive care unit.
Figure 3. Education level in population of high-risk subjects.
The Journal of TRAUMA®Injury, Infection, and Critical Care • Volume 67, Number 6, December 2009
Interim Analysis to Forge Ahead in Violence
Prevention
© 2009 Lippincott Williams & Wilkins
1173