Transcript MEASURING HEALTH BEHAVIOR CHANGE: PROBLEMS AND PROMISE

MEASURING HEALTH
BEHAVIOR CHANGE:
PROBLEMS AND PROMISE
CARLO C. DICLEMENTE
PROFESSOR & CHAIR
UMBC PSYCHOLOGY
HEALTH PROMOTION &
DISEASE PREVENTION
CANCER PREVENTION
REQUIRE
BEHAVIOR
CHANGE
INITIATION
HEALTH PROMOTION
SAFETY & INJURY
PREVENTION
MODIFICATION
HEALTH PROTECTION
SUBSTANCE ABUSE
CESSATION
The Transtheoretical Model of Intentional Behavior Change
STAGES OF CHANGE
PRECONTEMPLATION  CONTEMPLATION  PREPARATION 
ACTION  MAINTENANCE
PROCESSES OF CHANGE
COGNITIVE/EXPERIENTIAL
BEHAVIORAL
Consciousness Raising
Self-Revaluation
Environmental Reevaluation
Emotional Arousal/Dramatic Relief
Social Liberation
Self-Liberation
Counter-conditioning
Stimulus Control
Reinforcement Management
Helping Relationships
CONTEXT OF CHANGE
1.
2.
3.
4.
5.
Current Life Situation
Beliefs and Attitudes
Interpersonal Relationships
Social Systems
Enduring Personal Characteristics
MARKERS OF CHANGE
Decisional Balance
Self-Efficacy/Temptation
How Do People Change?

People change voluntarily only when they
– Become concerned about the need for change
– Become convinced that the change is in their
best interests or will benefit them more than
cost them
– Organize a plan of action that they are
committed to implementing
– Take the actions that are necessary to make the
change and sustain the change
Model Components (Stages)
1. Precontemplation - Not Ready to Change
2. Contemplation - Thinking About Change
3. Preparation - Getting Ready to Make Change
4. Action - Making the Change
5. Maintenance - Sustaining Behavior Change Until
Integrated into Lifestyle
Relapse and Recycling - Slipping Back to Previous
Behavior and Re-entering the Cycle of Change
Termination - Leaving the cycle of change
Stage of Change Tasks

Precontemplation

Contemplation

Preparation

Action

Maintenance

Awareness,
Concern,Confidence
 Risk-Reward Analysis
& Decision making
 Commitment &
Creating an
Effective/Acceptable
Plan
 Adequate
Implementation of Plan
and Revising as Needed
 Integration into
Lifestyle
Theoretical and practical considerations related to
movement through the Stages of Change
Motivation
Precontemplation
Personal Environmental
Concerns Pressure
Decision-Making
Self-efficacy
Contemplation
Preparation
Action
Maintenance
Decisional
Balance
(Pros & Cons)
Cognitive
Experiential
Processes
Behavioral
Processes
Recycling
Relapse
Prescribed Health Behaviors

Pregnancy and HIV
Prevention
– Condom use
– Abstinence
– Birth control methods





Pills
Patch
Depo injections
Spermicidal agents
Emergency
contraceptives

Cancer Risk Reduction
– Screening (multiple)
– Smoking cessation
– UV Protection
– Environmental exposures
– Dietary changes
 Fat < 30%
 Fiber 20 grams
 Fruits & Vegetables (5)
Prescribed Health Behaviors

Cardiovascular Risk
Reduction
– Physical Activity

Diabetes Prevention
and Treatment
– Obesity Prevention and
– Cholesterol screening
–
–
–
–
and treatment
Weight Reduction
Dietary changes
Aspirin regimen
Alcohol Moderation
–
–
–
–
Reduction
Glucose monitoring
Dietary changes
Regular screening for
associated problems
Alcohol Consumption
Prescribed Health Behaviors

Similar lists of behaviors can be compiled
– Asthma prevention and control
– Obesity prevention
– Chronic Lung Disease
– Preventing and Treatment of Addictions and
Substance Abuse
– Traffic safety
– Occupational Safety
HEALTH BEHAVIORS






MULTIPLE
MULTIDIMENSIONAL
VARY IN FREQUENCY
VARY IN INTENSITY
REQUIRE DIFFERING LEVELS OF
MOTIVATION
CAN BE INTEGRATED INTO DIFFERENT
LIFESTYLES TO VARYING DEGREES
THE FIRST STEP TO
MEASURING HEALTH
BEHAVIORS

Specify the broad target behavior that
provides the greatest yield in health
outcome for this problem.
 Examine the key component behaviors that
are required to reach this goal target
behavior
 Examples: pregnant drug abusing women;
30% calories from fat; abstinence or
moderation
Defining Action: The First Step

Specifying the behavior or constellation of
behaviors that would characterize the action stage
of change
 Doing a task analysis that would indicate
frequency, intensity, difficulty, and skills needed to
perform the behavior
 Define partial goals and/or associated behaviors
that indicate positive activity but fall short of the
actual target behavior change (harm reduction)
Food for Life Project

Over 2000 women in WIC (Women,
Infants, & Children) programs
 10 sites with each acting as own control and
contributing women to intervention and
control
 Mail and in person intervention that was
intensive
 Significant results: < Fat; > F & V
Dietary behaviors related to
diet of < 30% calories from fat

Drinking 1% or skim milk
 Avoiding fried foods
 Checking labels for fat content
 Buying low fat or fat free products
 Avoiding High fat snacks and sweets
 Avoiding high fat meats
 Eating more fruits & vegetables
Precontemplation for All Low Fat Behaviors (Items 2-8)
No
Eating a Low
Fat Diet
Yes
N
%
N
%
Chi-Square
p-value
Precontemplation
506
29.1%
292
91.8%
448.02
.000
Contemplation
515
29.6%
21
6.6%
Preparation
301
17.3%
3
0.9%
Action
252
14.5%
2
0.6%
Maintenance
165
9.5%
0
0.0%
Totals
1739
Reported
Stage
318
Maintenance for All Low Fat Behaviors (Items 2-8)
No
Eating a Low Fat
Diet
Yes
N
%
N
%
Chi-Square
p-value
Precontemplation
798
39.8%
0
0.0%
321.32
.000
Contemplation
533
26.6%
3
5.9%
Preparation
302
15.1%
2
3.9%
Action
246
12.3%
8
15.7%
Maintenance
127
6.3%
38
74.5%
Totals
2006
Reported
Stage
51
Step 2: Defining Maintenance

What would this behavior look like in terms of
frequency, intensity, and completeness if it were
integrated into the lifestyle of the individual
(mammograms every 2 years; never more that 4-5
drinks of alcohol per occasion)
 What would criteria be for defining a slip
(temporary non adherence) or a relapse (a pattern
that substantively failed to meet criterion)
 Does maintenance make sense for infrequent acts
Proportion of MATCH Outpatients
Avoiding a Heavy Drinking (5 Drinks)
Day as a Function of Time
1.2
1
0.8
0.6
0.4
0.2
0
0
100
# OF DAYS
200
CBT
Days
MET
300
TSF
400
Drinking and Problem Status by
Treatment Condition (Outpatient)
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
cbt
met
tsf
cbt
met
tsf
Abstinent
Heavy Drinking w/minimum consequence
cbt
met
tsf
cbt
met
tsf
Light Drinking
Heavy Drinking w/severe consequence
The Well-Maintained Addiction

Defining action and maintenance is critical
for initiation of health risks, like addiction,
as well as health protection behaviors
 Regular, dependent use of a substance that
creates creates a pattern that eludes selfregulatory control, continues despite
negative feedback, and becomes an integral
part of the individual’s life and coping
The Reality of Relapse

Many individuals who attempt to make a
health behavior change fail to do so
 Non adherence rates for a wide range of
health behaviors range from 20 to 80%
 Adherence is often higher at short-term
follow-up than it is one year after an
intervention
Relapse & Recycling

Relapse is not a problem of substance abuse or
addictions; relapse is part of the process of
behavior change.
 The reality of Relapse requires successive
approximations to instigate successful, sustained
health behavior change.
 Most successful changers make repeated efforts to
get it right that are part of a learning process to
remediate inadequate completion of stage tasks.
Theoretical and practical considerations related to
movement through the Stages of Change
Motivation
Precontemplation
Personal Environmental
Concerns Pressure
Decision-Making
Self-efficacy
Contemplation
Preparation
Action
Maintenance
Decisional
Balance
(Pros & Cons)
Cognitive
Experiential
Processes
Behavioral
Processes
Recycling
Relapse
Stages of Change Model
Precontemplation
Increase awareness of need to change
Contemplation
Motivate and increase confidence
in ability to change
Relapse
Assist in Coping
Maintenance
Encourage active
problem-solving
Termination
Preparation
Negotiate a plan
Action
Reaffirm commitment
and follow-up
Measuring Change:
Behavioral Outcomes

Crucial challenge: Operationally defining
Action, Maintenance, and Relapse for this
particular health behavior
 Creating sensitive and clear evaluations for
each of these three constructs
 Finding ways to validate all of these critical
health behavior change outcomes using both
self-report and more objective measures
Examples of More Objective
Action and Maintenance
Outcome Measures

Steps per day or week measured by
pedometers assessed during a one week
period every three months for a year.
 Self-reported abstinence from illegal drugs
confirmed by random drug screens over one
year with a minimum of 90% clean screens
 Medical record confirmed mammograms
every 2 years (within a 2 to 3 year period)
Step 3: Examining Pre-action

Identifying critical markers of movement
toward action. Various models identify
various indicators: beliefs, intentions,
efficacy, decision making.
 Stage specific tasks: concern and
consideration, decision making,
commitment & planning
 Identifying associated variables
Distinguishing Pre-Action from
Action

It is difficult to evaluate concerns, attitudes,
beliefs, intentions, and plans unless you are
able to distinguish those already engage in
the action and those who do not need to
make changes from those at risk and
needing to change.
 Problem definitions and action/maintenance
criteria are essential to do this.
Food For Life Project

Block Dietary assessment
 Self-reported stage of change for eating a low fat
diet, eating five or more fruits and vegetables per
day, eating a high fiber diet, and for each of the
component dietary behaviors (skim milk, avoiding
high fat)
 How to reconcile objective and self-report
measures and to evaluate what any discrepancies
mean to the individual and for research
Self-Reported Stage of Change for Eating a Low Fat Diet
PC
C
PA
A
M
Ns for Rows
85.5
--
--
--
--
682
Contemplation
--
79.7
--
33.1
21.8
547
Preparation
--
76.6
27.2
33.3
357
Restaged SOC based on FFQ
Precontemplation
14.5
20.3
23.4
39.8
--
397
Maintenance
--
--
--
--
44.8
74
Ns for Columns
798
536
304
254
165
2057
Action
Self-Report and Restaging

For the most part self-report is a very good
approximation of where a person is in the process
of change with significant and substantial
correspondence between objective measures and
reported stage even when there is a vague criterion
like <30%.
 However, eliminating or restaging based on
objective measures can help get rid of problematic
variance
 Identifying discrepant individuals can increase our
understanding of self-evaluations and problems in
measurement
The Importance of Measuring
Pre-Action Status

However, much of the process of change happens
prior to action being initiated
 Subdividing pre-action status into stages helps to
understand challenges of individuals and
populations of interest prior to action
 Enables fine tuning of intervention efforts
including targeting feedback and adapting
interventions
 Provides a more sensitive and fine-grained
assessment of movement and intervention impact
over time
Stage Based Epidemiology
PC
PC
C
M
PA A
M
C
A
PA
Smoking Cessation Stages of Change: Ever Smokers in the State of Maryland
4000
3767
3500
Numbers of Ever Smokers
3000
2500
2000
1664
1500
988
1000
691
621
500
267
0
N
Precontemplation
Contemplation
Preparation
Action
Maintenance (6 mos - 5 years)
Long Term Maintenance (5+ years)
Table 3. Stage of Change by County of Residence (Weighted)
Stage of Change
% of Ever
County
Smokers
a
Precontemplation
Contemplation
Preparation
Action
Maintenance
Allegany
44.1%
52.9%
10.9%
10.2%
8.6%
17.4%
Anne Arundel
49.1%
44.2%
17.2%
11.6%
8.5%
18.6%
Baltimore
49.1%
40.1%
17.1%
11.6%
9.8%
21.4%
Calvert
51.9%
42.9%
15.7%
16.2%
4.2%
21.0%
Caroline
51.9%
40.3%
18.5%
12.0%
5.4%
23.8%
Carroll
45.7%
48.1%
12.1%
15.3%
5.0%
19.5%
Cecil
50.5%
44.2%
22.4%
14.0%
4.5%
14.9%
Charles
45.4%
45.8%
11.4%
15.1%
3.6%
24.1%
Dorchester
54.9%
42.2%
23.3%
12.0%
2.7%
19.7%
Frederick
46.5%
43.3%
18.0%
17.2%
4.5%
17.0%
Garrett
48.1%
46.4%
12.0%
20.7%
2.9%
18.0%
Harford
49.0%
37.7%
15.3%
17.4%
8.3%
21.3%
Howard
39.3%
41.9%
12.2%
16.1%
4.6%
25.3%
Kent
53.5%
38.3%
11.3%
13.9%
5.3%
31.1%
Montgomery
38.9%
35.1%
8.3%
17.4%
5.9%
33.3%
Prince George's
39.6%
34.3%
12.5%
20.2%
8.3%
24.7%
Queen Anne's
50.9%
36.6%
21.1%
18.9%
2.2%
21.2%
St. Mary
49.9%
39.5%
17.4%
18.8%
7.9%
16.4%
Somerset
51.7%
32.5%
19.1%
16.2%
7.0%
25.1%
Talbot
43.5%
38.1%
18.6%
14.5%
5.1%
23.7%
Washington
49.6%
50.4%
22.4%
12.5%
1.4%
13.3%
Wicomico
50.5%
43.8%
16.9%
12.4%
4.1%
22.7%
Worcester
49.3%
49.6%
14.3%
16.3%
4.0%
15.8%
Baltimore City
53.6%
37.6%
25.9%
19.1%
2.8%
14.7%
Table 4. Stage of Change (Current Smokers) by County of Residence (Weighted)
Current Smokers
County
% Current Smokers
Precontemplation
Contemplation
Preparation
Allegany
18.9%
71.5%
14.7%
13.7%
Anne Arundel
19.2%
60.6%
23.6%
15.8%
Baltimore
16.8%
58.3%
24.9%
16.8%
Calvert
21.3%
57.3%
21.0%
21.7%
Caroline
23.1%
56.9%
26.2%
16.9%
Carroll
18.0%
63.7%
16.0%
20.2%
Cecil
23.3%
54.8%
27.8%
17.4%
Charles
18.7%
63.4%
15.8%
20.9%
Dorchester
25.8%
54.5%
30.0%
15.5%
Frederick
17.9%
55.2%
22.9%
21.9%
Garrett
21.1%
58.6%
15.2%
26.2%
Harford
17.8%
53.5%
21.7%
24.8%
Howard
11.8%
59.7%
17.3%
23.0%
Kent
18.9%
60.3%
17.8%
21.9%
Montgomery
9.2%
57.7%
13.6%
28.7%
Prince George's
14.1%
51.2%
18.7%
30.1%
Queen Anne's
23.8%
47.7%
27.6%
24.7%
St. Mary
20.8%
52.1%
23.1%
24.8%
Somerset
19.6%
47.9%
28.2%
23.9%
Talbot
14.2%
53.6%
26.1%
20.4%
Washington
22.0%
59.1%
26.2%
14.7%
Wicomico
22.0%
59.9%
23.1%
17.0%
Worcester
21.4%
61.9%
17.8%
20.3%
Baltimore City
29.9%
45.5%
31.4%
23.1%
Measuring Pre-Action






Can approximate how far or close individuals are
to being committed and planning action using
many different methods
Measures of attitudes and self-statements
(URICA, Readiness to Change; pros & cons)
Stage classification algorithms
Simpler ruler or ladder types of assessments
Interview evaluations
Self or peer nominations
Stage of Change by RUNG (Q56)
10.00
9.00
8.00
7.00
6.38
6.00
5.15
5.00
4.00
3.00
2.88
2.00
1.00
0.00
RUNG
Precontemplation
Contemplation
Preparation
Measuring Associated
Markers of Change

We need to understand associated behaviors
and activities that coincide with stage status
 These markers can provide additional
targets of intervention or assessment
 For interventions that do not produce gross
behavior change, stage tasks and markers
represent the only way to evaluate if they
have had any effect on the process
Stage of Change by Average Number of Cigarettes Smoked per Day in the Past 30
Days (Q10)
20.0
18.0
17.4
15.3
16.0
13.3
14.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0
Past 30 days, avergage cigarettes smoked/day
Precontemplation
Contemplation
Preparation
Stage of Change by Number of Times Stopped Smoking Cigarettes 1+ Days
because Trying to Quit (Q46A)
9.00
8.22
8.00
6.59
7.00
6.00
5.00
5.38
4.85
4.45
4.00
3.00
2.00
1.00
0.00
# of Times Stopped Smoking 1+ Days
Precontemplation
Contemplation
Preparation
Action
Maintenance
Stage of Change by Generally How Purchase Cigarettes (Q18)
90.0%
80.2%
80.0%
69.3%
70.0%
60.0%
50.0%
54.3%
45.7%
40.0%
30.7%
30.0%
19.8%
20.0%
10.0%
0.0%
Precontemplation
Contemplation
Carton
Preparation
Pack
Stage of Change by Percentage Self/Others who Smoked in Their Home (Q76)
80.0%
70.3%
70.0%
66.0%
60.4%
60.0%
50.0%
40.0%
29.3%
30.0%
19.6%
20.0%
10.0%
0.0%
Percentage of Self/Others who Smoked in Home During Past Week
Precontemplation
Contemplation
Preparation
Action
Maintenance
Stage of Change by Perceived Cost of Last Pack of Cigarettes (Q19)
$3.35
$3.33
$3.30
$3.24
$3.25
$3.20
$3.17
$3.15
$3.10
$3.05
Cost of Last Pack Purchased
Precontemplation
Contemplation
Preparation
Stage of Change by Number of 4 Closest Friends who Use Tobacco Products
(Q88)
4.00
3.50
3.00
2.63
2.50
2.27
2.21
1.92
2.00
1.38
1.50
1.00
0.50
0.00
Number of 4 Closest Friends Who Use Tobacco Products
Precontemplation
Contemplation
Preparation
Action
Maintenance
Stage of Change by Percentage who Asked Someone Else around them Not to
Smoke in the Past Year (Q72)
35.0%
32.1%
30.0%
26.4%
26.2%
25.0%
20.0%
20.0%
15.0%
11.7%
10.0%
5.0%
0.0%
% Asked Someone Not to Smoke
Precontemplation
Contemplation
Preparation
Action
Maintenance
TTM Profile: Outpatient PDA Baseline
0.8
Standard Scores
0.6
0.4
Abstinent
Moderate
Heavier
0.2
0
-0.2
-0.4
-0.6
-0.8
Pre
Con
Act
Main
TTM Variables
Conf
Temp
TTM Profile: Outpatient PDA Post Treatment
0.8
0.6
Standard Scores
0.4
Abstinent
Moderate
Heavier
0.2
0
-0.2
-0.4
-0.6
-0.8
Pre
Con
Act
Main Conf Temp
TTM Variables
Exp
Beh
Cautions in Assessing PreAction

Pre-action stage status is volatile and
changeable (even during the course of an
interview)
 Individuals move both forward and
backward in considering and planning for
change
 Even for those planning change priorities
change and competing problems interfere
Cautions continued

Assessment of readiness needed for overall
goal behavior does not necessarily indicate
readiness for all component behaviors.
 Are importance and efficacy the only
ingredients needed for readiness?
 Prior attempts (recycling) and success or
failure with similar changes are important to
consider and evaluate
Pros and Cons of Various
Types of Measures

Simple Continuous Measures (rulers)
 Multi-component attitudinal measures
 Algorithms (a series of dichotomous
response questions)
 Related assessments (pros and cons; selfefficacy; intention, beliefs)
 Self-reported stage status
Conclusions about measuring
Health Behavior Change

There are significant differences in attitudes and
activities of individuals in different pre-action
stages no matter how these are assessed (not every
study but every type of measure)
 It is complicated evaluating pre-action
assessments once individuals have made behavior
changes
 Patterns of change vary greatly over time: more
stability than change; rapid change; recycling
Conclusions II

What is needed are multiple assessment
over short and long periods of time. Longterm follow-ups will not help us understand
the process of change. Short-term followups emphasize momentary changes and
action but underestimate the long haul.
 Successful health behavior changes must be
viewed incrementally not dichotomously
Challenges I

We must sharpen our thinking and
conceptualizing of health behaviors. Broad,
general conceptualizations do as much
damage to health promotion research as
simply looking at regions of the brain and
not neurotransmitters would do for brain
research. Specificity and sophistication
must be the hallmarks of the future.
Challenges II

Basic research to understand, define and
assess health behaviors must precede largescale efforts to change these behaviors
 New technology should be incorporated into
the assessment of actual behavior change
(pedometers, MEMS Caps, body fat
composition, computerized assessments)
but cannot supplant self-reported behavior.
Challenges III

We must continue to develop more sophisticated
assessments of critical attitudes, intentions and
plans related to the specific health behavior
change
 We must look for benchmarks or additional
markers related to movement toward change
 We must develop a better understanding of how
cultural and ethnic influences impact our
outcomes and our assessments
The Promise of Accurate
Assessment
More sophisticated understanding of health
behaviors and health behavior change
More sensitive analyses of mechanisms,
contextual influences, and change
Increased accuracy of goals and target
behaviors
Better targeted interventions
Better evaluation of interventions