Changing Public
& Patient Behavior:
Theory, Evidence, Planning
Training in Clinical Research
Translating Evidence Into Practice
April 29, 2008
Rena J. Pasick




Cancer screening
disparities
Intervention evidence
Behavioral theory and
culture
Planning
Trends in Early-Stage Female
Breast Cancer Incidence by Race/Ethnicity
California, 1988-1999
Percent
60
50
40
30
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
Af. Am.
N-H White
Hispanic
Asian/PI
Source: Cancer in California, 1988-1999. California Department of Health Services, December
2001.
Mammography Status
in a Multiethnic* Sample
Alameda County Women, 40-74
Pathfinders Baseline Survey, 1999-2000 (n=1463)
Language
n
Mammography (%)
Recent**
Not recent
Never
1016
60.2
29.3
10.4
171
50.3
26.3
23.4
Tagalog
56
48.2
33.9
17.9
Spanish
220
58.6
25.9
15.5
English
Cantonese
*African American, Chinese, Filipina, Latina, White women,
** Within 15 months of interview
Cervical Cancer
Average Annual Age-Adjusted Incidence
US, 1988 - 1992
45
Rates per 100,000
40
35
30
25
20
15
10
5
0
White
SEER, 1996
Black
Hisp
Chin
Fil
Viet
Research Background
 Breast & Cervical Cancer
Intervention Study (BACCIS)
NCI
1991 - 1996
Low-income African American, Latina,
Chinese, & White women
N = 3216
 Pathways to Early Cancer Detection
in Four Ethnic Groups
1992 - 1997 NCI
African American, Latina, Vietnamese,
Chinese, & White women
N = 4228
 Early Cancer Detection Among
Filipino American Women
1994 - 1998 DOD
Filipinas - interviewed in English,
Tagalog, Ilocano, & Cebuano
N = 875
 Cancer Screening, Managed Care
and the Underserved
1998 - 2002 NCI
Low-income African American,
Chinese, Filipina, Latina, & White
women
N= 3100 (est.)
Research Background (con’t)
 Behavioral Constructs and Culture in Filipina & Latina Women
Cancer Screening
Qualitative
2001-2007 NCI
 Cross-Cultural Communication for
Colorectal Screening
2002-2006 NCI
African American, Chinese, and Latino
primary care patients at Kaiser &
community clinics
Formative study/pilot
Statewide Communication to Reach
High-Risk Low-SES Women for
Research and Prevention
2007-2012 NCI
African American, Latino, Chinese,
Vietnamese & Korean women calling
statewide Everywoman Counts toll-free
line




Cancer screening
disparities
Intervention evidence
Behavioral theory and
culture
Planning
Interventions
Access Enhancement
Promotion






laws
non-clinical settings
cost reduction
eliminate structural barriers
multi-component
Briss PA et al, Am J Prev Med 2000
Rimer BK et al, Pub Hlth Rep 1993






mass media
small media
small group education
one on one education
incentives
reminders
multi-component
Cancer Screening
Intervention Research






Large body of research
Extensive array of barriers identified
Many effective interventions
Extent to which interventions sustained
“disappointing”
Few interventions found effective are disseminated
Multilevel theoretical and conceptual approaches
are needed to understand the full context of
cancer screening
Meissner H, 2004; Rakowski W, Breslau E. 2004
Lesson 1
Virtually all major types of communitybased cancer screening interventions
have been found to be effective in some
studies. However, no interventions work
for all people, screening tests, or in all
settings.
Pasick, Hiatt, Paskett. 2004
“Focal Point” Framework
Intervention Strategies
Focal Point
Target population
• Setting
• Screening objective
•
•Access
enhancement
•laws
•non-clinical
settings
•cost reduction
•eliminate structural barriers
•multi-component
•Promotion
•mass
Research
•Fundamental
research
•Intervention
research
media
•small media
•small group education
•one on one education
•incentives
•reminders
•multi-component
Fundamental research
(basic behavioral research)
Research
•
Fundamental
research
•
Intervention
research
Develop/refine the mechanisms
that lead to intervention
outcomes
- theory & measurement
- messages
- communication channels
Intervention research
Research
•
Fundamental
research
•
Intervention
research
Optimal intervention for the “focal
point” is identified, implemented
accurately, evaluated rigorously, and
replicated.
- formative evaluation
- process evaluation
- assessment of outcomes
- including cost-effectiveness
Brilliant points of light…..
Research
•
Fundamental
research
•
Intervention
research
…punctuating vast empty spaces
Fundamental
research
Intervention
research




Cancer screening
disparities
Intervention evidence
Behavioral theory and
culture
Planning
On the one hand…..
the reason for theory is to help guide the
selection of important constructs thought to
exert impact on health behavior
Crosby et al. 2002
On the other hand…
By targeting a specific range of theoretically
important constructs for examination or as a
foundation for designing health promotion
programs…
theory limits the breadth of observations
and investigations and the scope of
intervention
Crosby et al., 2002
Limitations of Theory
for Disparities Populations
 Origins
of theory
 Measurement
 Decontextualization
 Focus on cognition
Measurement
Theory
Interventions &
Messages
Measurement
•
Intention measurement includes action, target, time:
I intend to get a mammogram in the next 12 months
extremely unlikely :___1__:___2__:___3__:___4__:___5__:___6__:___7__: extremely likely
•
Pathfinders translation/adaptation for 5 race/
ethnic and 4 language groups:
Do you plan to have a mammogram in the next 12 mos?
yes •no
•
Perceived Benefits
of Mammography
in a Multiethnic, Multilingual Sample (n=1463)
Perceived Benefits
of Regular Mammography
at baseline
Good for Family
(% agree)
Peace of
Mind
(% agree)
Early
Detection
(% exc./good)
Afr Amer
90
91
95
Chinese
97
98
71
Filipina
95
93
93
Latina
93
95
85
White
86
84
98
Overall
91
92
90
p-value
0.001
0.001
0.001
Perceived Benefits &
Regular Mammography
in a Multiethnic, Multilingual Sample (n=1463)
Perceived Benefits
of Regular Mammography
at baseline
Good for Family
(% agree)
Regular
Mammograms
at baseline
Peace of
Mind
(% agree)
Early
Detection
(% exc./good)
(% had)
Afr Amer
90
91
95
51
Chinese
97
98
71
39
Filipina
95
93
93
43
Latina
93
95
85
39
White
86
84
98
55
Overall
91
92
90
47
p-value
0.001
0.001
0.001
0.001
Summary: 3Cs Quantitative Findings
Pathfinders Baseline & Final Surveys*
 performed as expected
– self-efficacy predicted recent screening longitudinally
across ethnic groups
– subjective norms was associated with mammography
cross-sectionally
 differential predictive strength by ethnicity
– intention
– perceived benefits
 no evidence of association
– perceived susceptibility
* 5 Race/ethnic and 4 language groups
Stewart SL. Behavioral Constructs and Mammography in Five Ethnic Groups (in progress)
Data Quality in Multiethnic Health Surveys
•
A study to assess the validity and comparability of survey
questions across cultures and languages
•
Mixed methods to explain findings and understand more
complex meaning
•
Survey instruments & quantitative data from 3 multilingual breast & cervical screening intervention studies
– Data from 3 studies (6760 interviews)
Pasick, RJ, Stewart SL, Bird JA, D’Onofrio CN. 2001
Data Quality in Multiethnic Health Surveys - Public Health Reports 116: 223-243,
Data Quality in Multiethnic Health Surveys
Quality of multi-ethnic, multi-lingual data should be
regarded as questionable
Measurement is affected by complex factors not readily
observed or understood through standard tests
Objective items present fewer problems than attitudinal
items
All are subject to translation difficulties & lack of crosscultural comparability
Measurement Theory
Communication
Understanding
Culture
Images
Stated
values &
beliefs
Ways of
communicating
Culture
a system of knowledge that allows us to know how to
communicate with others and how to interpret their behaviors
(Gudykunst, 1998)
High Context
Communication
Low Context
Communication
• Associated more with
collectivist cultures*
• Associated more with
individualistic cultures*
• Very little information is in
coded, explicit,
transmitted part of the
message**
• Explicit, direct, precise
messages
* Gudykunst & Ting-Toomey, 1988
** Hall, 1976
Hypothetical Vignette
based on Communicating Effectively with the Chinese*
Scene: Anglo MD recommending CRC
screening to Chinese patient
MD prescribes med for presenting GI
complaint…
* G Gao, S Ting-Toomey 1998 - Sage Publications
Anglo MD & Chinese patient (con’t)
MD: “You’re now 51. At
this age your risk of CRC
cancer goes up. I’d like you
to start doing yearly stool
tests and to have a test called
a sigmoidoscopy every 5
years. Here’s a booklet that
describes this. Stop at the lab
for the test kit and call the
nurse if you have any
questions.”
Patient is silent, thinking:
She thinks I have cancer
Explanation In high-context cultures,
people read between the
lines, expecting to find more
meaning in what is not said
than in what is verbalized
Anglo MD & Chinese patient (con’t)
MD: “Do you have any
questions?”
Patient: “No”
Explanation In some cultures, it is
considered impolite &
disrespectful for a lay person
to question an MD; one
accepts authority without
question.
Understanding
Culture
Images
Stated
values &
beliefs
Ways of
communicating
Socio-Cultural Context
Culture is…

the tools that group members use to make sense of the
chaos around them, putting the abstract ideas of
worldview into tangible beliefs and behaviors
Kagawa-Singer M. 1997
 a system of knowledge that allows us to know how to
communicate with others and how to interpret their
behaviors
Gudykunst WB. 1998

a system whose whole pattern (most of which is not
conscious at any particular time) is required to
understand the meanings that people attach to specific
facts and observations
Bourdieu P. 1990
Behavioral Theory:
Like a Compass on Mars
Mars
groups who suffer cancer disparities
Compass behavioral research theory & methods
The Problem

Behavioral theory constructs
– Developed with mainstream populations
– Used as universal

Underlying concepts
– Not tested for cultural appropriateness
– Focus is cognition, not context

Measures
– Not validated for cross-cultural comparability
or validity
The Problem

Behavioral theory constructs
– Developed with mainstream populations
– Used as universal

Underlying concepts
– Not tested for cultural appropriateness
– Focus is cognition, not context

Measures
– Not validated for cross-cultural comparability
or validity
Behavioral Constructs & Culture
in Cancer Screening (3Cs)
funded by the NCI 2001-2006
 Methodological study to
assess the cultural
appropriateness of 5
behavioral theory
constructs
 Mixed methods
 multi-lingual survey
 in-depth interviews
key informant “scholars”
 community gatekeepers
 women in community

Self efficacy Intention Perceived Benefit Perceived Susceptibility Subjective Norms
3Cs Research Team
Investigators
Rena Pasick
Nancy Burke
Susan Stewart
Judith Barker
Joyce Bird
Regina Otero-Sabogal
Noe Tuason
Galen Joseph
Consultants
Bill Rakowski
Melissa Clark
Anthropology * Behavioral Science *
Biostatistics * Health Psychology * Public
Health * Sociology
The Problem

Behavioral theory constructs
• Developed with mainstream populations
• Used as universal

Underlying concepts
• Not tested for cultural appropriateness
• Focus is cognition, not context

Measures
• Not validated for cross-cultural
comparability or validity
Is Intention Comparable
in Meaning & Measurement Across Cultures?
Self efficacy Intention Perceived Benefit Perceived Susceptibility Subjective Norms
intention – the determination to perform certain
activities (e.g., mammography) or to bring about a
certain future state of affairs; the likelihood of doing
something (Theory of Planned Behavior)
mechanisms


cognitive representation of future outcome
goal setting and self-evaluation
assumptions


behavior is volitional
people are rational
Theory of Planned Behavior
OR……
Beliefs + Beliefs + Beliefs > Intention > Behavior
3Cs Qualitative Methods

Key informant (KI) interviews
• Scholars from the ethnic group of interest

Community gatekeeper (GK) interviews
• CBO directors/managers, public health nurses,
religious leaders

Ethnographic interviews with women
• Latina and Filipina women in the community



Transcription (and translation)
Collaborative coding
Identification of themes
Qualitative Methods in Health Research vs
the Anthropological Approach
Qualitative Health Research
Anthropologic Approach
Focus groups/semi-structured
interviews
Multiple qualitative methods
(Ethnography)
Often used to deconstruct complex
phenomena into simpler concepts
Constructs multifaceted
Understanding/multiple influences
Explores beliefs and practices
related to predefined biomedical
phenomena
Explores broader context for
alternative meaning and/or
multifaceted influences
Focuses on lay understanding to
change individual behavior
Focuses on both lay and expert
knowledge as objects of inquiry
Focuses on cognition; what is
accessible to individual
knowledge/memory
Explores influences beyond individual
consciousness
(e.g., social context)
Pasick R, Burke N (2007) A Critical Review of Theory in Breast Cancer Screening
Promotion across Cultures. Ann Rev Pub Hlth
Themes
with relevance to intention
CONFLICT WITH
BARRIERS: women may have
positive attitudes toward
screening, but are required to
make choices within a stressful
environment; thus, despite
stated intentions to get
screened, they do not go
… there's no opposition to
getting screened… they
know they should… they
want to, etcetera… it's
just a question of finding
a time to do that.
Latino key informant
KI-09-04142003.txt - 9:20
ABSENCE OF
PREDICTABILITY:
underlying the association
between intention and behavior
is the implicit assumption of
predictability; a favorable
attitude & intention cannot
always lead to behavior when
predictability is absent
..what you're dealing with in a
highly chaotic sort of fluid
community where you're
very susceptible to layoffs… poverty is chaos
because there's no
predictability.
Latina key informant
KI07-021203.txt - 7:32
.. We don't get screened because of
decidia; decidia means we put it off -we're negligent"…
Latino key informant, KI-004142003.txt - 9:6
ACCEPTANCE OF
PUTTING THINGS OFF:
When they told me about this, I thought
well I should have it done because I
Many women may be
have never got it done. But no. I think
accustomed to being unable to
that we are “desidioso.” I think that
do all they intend; they cope
most people are “desidioso”….. We
don’t take or make the decision that we
by accepting that some
have to do this. We leave it for later.
important things must be put
Indecisive. I say it like that. I don’t
off.
know how you say it.
Latina woman, L-12-121304
INTENTION IN THE
CONTEXT OF THE
UNFAMILIAR:
Immigrant women may form
positive attitudes & intentions
but so much of their world is
unfamiliar they don’t want to
attempt navigating a system
that makes them feel
vulnerable
.. Sometimes it takes people a
period of time to become
adjusted to a concept,
especially if we're talking
about a concept that's totally
alien to them…. sometimes
it's comfortable and
sometimes it's not. So that's
where I find more professional
settings around people saying
yes but meaning no.
Filipina gatekeeper, P22
INTENTION BASED ON
RELATIONSHIP: in
relational cultures, it is very
likely to find women getting
screened because they
perceive that the person
asking them truly cares; this
is not based on any perceived
benefit of the test. Indeed,
they may not go in the
absence of this relationship
If you don't have that
relationship with them,
they're less likely to do
it….I think they do it mostly
for us…Yeah… because
we're recommending that
they get this mammogram
… they themselves don't
understand why they need
to…
Filipina gatekeeper, P15
.. they don't want to see the
nutritionist, but it's really impolite
to say, "No, I don't want to see
INTENTION AS
your nutritionist." So they'll just
say, "Yes, I'll go see the
DESIRABILITY/POLITENESS:
nutritionist" or "Yes, I'll go to the
part of respect is being agreeable
doctor and I'll do this." And then
and not saying no
they never do it….They say yes
because they don't want to be
impolite. Saying no is impolite.
Filipina gatekeeper, P22
MEANING OF STATED
INTENTION: for many
reasons, including the
subtleties of indirect
communication, people will
say yes but mean no
… it's like they say, "Well, yeah, I
made the appointment", or "Yeah,
I've been there", but they are
actually saying just to say, "I'll be
there"… but they are not really
planning to come.
Latina gatekeeper, P12
You have to tell them, "It's important
that you come… if not, you just tell
me… "It's OK for you to tell me
that you can't come", you have to
tell them that… that's how I deal
with it. "It's OK for you to tell me
that you can't come, you know, I
won't get mad".
Filipina gatekeeper, P13
Intention Across Cultures
The construct of intention should not be used as
universal due to:
•
•
•
•
Differences in meaning of stated intention
Differences in relationship of intention to behavior
Missing predictors
Entirely different issues and relationships
Conclusions



Gaps in understanding of diverse cultures
can best be addressed through mixed
method research on theoretical concepts
& constructs, and thus to inform
interventions and explain outcomes
Cognition may be only a minor influence
on behavior
Pursue the “unpacking” of social context
for more complete understanding of
behavior
Avoid Pitfalls
in Multicultural Research
- Review theoretical premises & hypotheses
-
-
-
-
Can cultural insights provide refinements?
Field test instruments with diverse groups
Expect & plan for differential refusal rates
Assess constructs for group-specific patterns
Share results with community to draw implications
collaboratively
Community members/judges to carefully review
“off-the-shelf” interventions
Marin & Marin, 1991; Vega, 1992
Using What Works (UWW)
National Cancer Institute
Tips For Trainers Module
1- Introduction: What Do We Mean by Evidence-Based?
2- Needs Assessment: Getting To Know Your Audiences
Better
3- Finding an Evidence-Based Program
4- Making the Evidence-Based Program Fit Your Needs:
Adaptation and Your Program Summary
5- Does It Work? Evaluating Your Program
http://cancercontrol.cancer.gov/use_what_works/start.htm
UWW Adaptation Guidelines
1. Determine the needs of your audience/does this
program address those needs?
2. Review the program and its materials with intended
audience
3. Define the extent of adaptation/how to implement
4. Develop “mock-up” of the adapted products.
5. Consult expert advisors on maintaining fidelity to the
original
6. Pilot test
8. Implement
9. Evaluate
http://cancercontrol.cancer.gov/use_what_works/start.htm
Descargar

Document