Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
‹#›
------=_NextPart_01C91FC5.CF8AD390
Content-Location: file:///C:/1EF64AF4/elliott_files/master14.xml
Content-Transfer-Encoding: quoted-printable
Content-Type: text/xml; charset="utf-8"
The
Future of Violence
Prevention
Research and
Practice
by
Delbert
S. Elliott, Ph.D.
Director,
Center for the Study and Prevention
of
Violence, University of Colorado
Prevention Research-
Agenda for Next = 20 Years
Agenda for Next = 20 Years
•Establish consensus on scientific standard for
certifying effective programs
•Upgrade program evaluation design, methodology
and reporting
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-3.88%'>•<=
/span>The
new research frontier: dissemination
and implementation
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.02%'>•<=
/span>Address
the barriers to dissemination &
programs
<=
span
style=3D'visibility:hidden'>•
Confusion over standard
Defining evidence-based&=
#13;
Federal Program Lists
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-3.86%'>•<=
/span>Center
for Mental Health Services (2000)
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.43%'>•<=
/span>National
Registry (NREPP) (2002)
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.43%'>•<=
/span>Surgeon
General Report (2001)
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.43%'>•<=
/span>Helping
America’s Youth (2007)
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.43%'>•<=
/span>OJJDP
Title V (2007)
<=
span
style=3D'visibility:hidden'>•
Consensus Across 8
Federal Lists
Federal Lists
<=
span
style=3D'font-size:75%'>•No program
appeared on all lists
•Only one program (LST) appeared on 7 of 8 federal =
lists as a model/exemplary/Level 1 program*
<=
/span>
<=
span
style=3D'font-size:75%'>•Two progra=
ms
were on 5 lists: MST & TND
•4 Programs on four lists: ALERT, ATLAS, Early Rise=
rs for Success, & FFT
•11 Programs on 3 lists: BBBS, GBG, TNT, PATHS, MTFC, NFP, Project Northland, Focus on Family, Strengthening Families, Caring School Communities,=
Incredible Years
* Top category on each list
<=
span
style=3D'font-size:50%;visibility:hidden'>•
Federal Working Group Standard for <=
/span>=
Certifying Programs as Effective*
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.43%'>•<=
/span>Experimental
Design/RCT
•Effect
sustained for at least 1 year post-intervention
•At least 1 independent replication with =
RCT
•RCT’s
adequately address threats to internal validity
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.35%'>•<=
/span>No
known health-compromising side
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.43%'>•<=
/span>effects
<=
span
style=3D'visibility:hidden'>•
Hierarchical Program
Classification*<= /b>
Classification*<= /b>
I. Model: Meets all standards
II. Effective: RCT replication(s) not independent&=
#13;
III. Promising: Q-E or RCT, no replication
IV. Inconclusive: Contradictory findings or
non-sustainable effect=
s
V. Ineffective: Meets all standards but with no
statistically signif=
icant
effects
VI. Harmful: Meets all standards but with negative=
main effects or serious side effects
=
VII Insufficient Evidence: All others
=
<=
span
style=3D'font-size:56%;visibility:hidden'>•
Effectiveness, Wo=
rking
Group for the Federal Collaboration on What Works, 2004.
<=
span
style=3D'font-size:56%;visibility:hidden'>•
The new research frontier:
dissemination
and
implementation
…very little is known about the
=
processes required to effectively
implement evidence-based
programs on a national scale.
<=
/div>
Research to support the
implementation activities that
=
are being used is even scarcer.
National Implementation Research Network, 2007=
Blueprints
for Violence
Prevention
Replication:
Factors
for
Implementation
Success
Mihalic et al., 2004. Fund=
ed
by OJJDP
Program Implementation
<=
span
style=3D'font-size:178%'>§Program Dissemination
<=
span
style=3D'font-size:178%'>§ Program Fit
<=
span
style=3D'font-size:178%'>§ Site Preparation
<=
/span>
<=
span
style=3D'font-size:178%'>§ Training
<=
span
style=3D'font-size:178%'>§ Technical Assistance&#=
13;
<=
span
style=3D'font-size:178%'>§ Program
Fidelity/Adaptation
<=
span
style=3D'font-size:178%'>§ Predictors of Program
Quality
<=
span
style=3D'font-size:178%'>§ Program Sustainability=
PROGRAM
DISSEMINATION
CAPACITY
<=
span
style=3D'font-size:156%'>§Published material: handbooks, curriculum,
ma=
nuals,
etc.
<=
span
style=3D'font-size:156%'>§ Certification of
trainers
<=
span
style=3D'font-size:156%'>§ High quality, packaged
T.A.
<=
span
style=3D'font-size:156%'>§ Process evaluation
measures
<=
span
style=3D'font-size:156%'>§ Dissemination organiza=
tion:
dedicated to
marketing and delivery
<=
span
style=3D'font-size:156%'>§ Data management system=
in
place
PROGRAM
FIT
<=
span
style=3D'font-size:156%'>§Does the program address the needs and
existing
barriers to learning at this school?
<=
span
style=3D'font-size:156%'>§ Has it been demonstrat=
ed
effective for the
type
of community/school/students that will
be
involved?
<=
span
style=3D'font-size:156%'>§What level of certification does the program
have?
[many pushed prematurely with only
efficacy
trial]
SITE
PREPARATION
<=
span
style=3D'font-size:156%'>§Most failures due to limited site
capacity
<=
span
style=3D'font-size:156%'>§Critical elements: Local champion,
<=
span
style=3D'font-size:156%'>§Develop clear expectations and
contracts
TRAINING
•Hire all staff before
training
•Hold line on requisite
training
•<=
/span>Review
program plans with staff before training
•Have administrators at=
tend
training
•Plan and budget for st=
aff
turnover
•Implement immediately
after training
•<=
/span>
TECHNICAL ASSISTANCE
•Quality declined over
time
•Lack of proactive
delivery
•<=
/span>T.A.
providers hard to reach, slow response
•School-based programs
delivered best
•Family-based most
consistent & proactive
•<=
/span>Variation
in perceived need by program type
•<=
/span>
FIDELITY COMPONENTS
•<=
/span>Adherence:
delivered as designed and evaluated
•
BP 86%-100%; LST- 81%-86%
•Exposure/Dosage
•
School BP-33%-50%; LST- 56%-78%
•Quality of program
delivery
•Participant
responsiveness
•<=
/span>
Fidelity vs Adaptation
<=
span
style=3D'font-size:88%'>•Adaptation=
s must
fit with program rationale
justified
–
Little evidence for race/ethnicity, gender, or
class
differences in school program effects
<=
span
style=3D'font-size:88%'>•Most frequ=
ent
threats to fidelity:
–
Frontline implementers
–
Disseminating Agency
<=
span
style=3D'font-size:88%;visibility:hidden'>•
Fidelity vs Adaptation
•Adaptation is as likely to reduce effects as enhance
them
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.08%'>•<=
/span>Local
adaptation may increase “buy in”
but also creates uncertainty
about
program effects
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.33%'>•<=
/span>Program
success must be judged by
real changes in behavior, not
number
of adoptions or survival
=
;
<=
span
style=3D'visibility:hidden'>•
Overcoming
barriers to
widespread
dissemination
Why Are We Not
Implementing Evidence-Based Violence
Prevention Progr= ams?
Implementing Evidence-Based Violence
Prevention Progr= ams?
<=
span
style=3D'font-size:75%'>•It’s=
hard
to sell prevention- the focus typically is on
improving
responses to violence
<=
span
style=3D'font-size:75%'>•Programs n=
ot
addressing strongest risk/protective
factors
or clusters
<=
span
style=3D'font-size:75%'>•Confusion =
about
standard for EB certification
<=
span
style=3D'font-size:75%'>•Politics and
parochial judgment often trump research
•Increasing professional resistance to EB programs =
and practices
<=
span
style=3D'font-size:75%'>•Failure to
implement with fidelity
<=
span
style=3D'font-size:75%;visibility:hidden'>•
Professional Resistance
“I particularly enjoyed your most recent art=
icle
warning about the potenti=
al
tyranny of evidence based practices …
I think you underplayed the possibility that an emphasis on such programs can inadvertently undermine rather than enhance school-wide reform <=
/span>efforts. …there is virtually no evidence that
evidence-based practices
contribute to overall school effectiveness,
as data on such an issue are never gathered.”
Unidentified “well-respected scientist”=
;,
Enews, August, 2007 (=
Vol
11, #11)
<=
span
style=3D'font-size:75%;visibility:hidden'>•
Impact of Unsafe Schools on
Health and Aca= demic Performance
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.43%'>•<=
/span>Poorer
Student Health
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.43%'>•<=
/span>Higher
Rates of Dropout
<=
span
style=3D'mso-special-format:bullet;position:absolute;left:-4.43%'>•<=
/span>Lower
Test Scores
•Smaller Gains in Academic Performance over
time
Controlling for grade in school, race/ethnic
composition, % subsid=
ized
meals, average parent education, %ESL students
<=
span
style=3D'font-size:75%;visibility:hidden'>•
National Survey = of School-
Based Prevention Programs
•Over two-thirds of schools reported use of at leas=
t one substance abuse program; almost half reported usin=
g 3 or more programs.
• Only=
26.8%
of schools were implementing an effective (research based) substance abuse prevention
program.
•In general, the quality of school-based prevention=
(delinquency, substance abuse, violence) practices=
is low.
Sources: Ringwalt et al., 2002. The Prevalence of
Effective Substance Use Prevention
Curricula in U.S. Middle Schools. =
Prevention
Science 3:257-272.
Gottfredson & Gottfredson, 2002. Quality of
School-Based Prevention
Programs:
Results from a National Survey. Jo=
urnal
of Research in Crime and
Delinquency 39:3-35
<=
span
style=3D'font-size:56%;visibility:hidden'>•
Feasibility Example
•Cost to provide every
student in U.S. a
model drug prevention progra=
m like
LST is $550 million per year
•<=
/span>Current
national drug control spending is approximately $40 billion=
per
year
•<=
/span>This
represents 1.5% of the current drug control spending
•<=
/span>
Conclusions
•We Need A Uniform Scientific Standard For Certifyi=
ng “Evidence-Based” Programs
<=
/span>
•Existing Federal Lists Provide Some Guidance, But =
Programs Other Than Those In The Top Category Are =
Often Problematic
•EB Program Should Be Selected For Its Known Effect=
On Particular Risks & Protective Factors For
Specific Groups
•If You Decide To Use A Program Not Certified as EB=
, You Must Commit To Evaluating It
•Do Not Use Any Program Found to Be Ineffective or =
Harmful
<=
span
style=3D'font-size:75%;visibility:hidden'>•
THANK
YOU
Center for the Study and Prevention of Violence
www.colorado.edu/cspv
9/26/2008
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
‹#›
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
' + ENDSHOW_MESG + '
![](3D"notes_flag.gif")
No=
tes![](3D"buttons.gif")
