TPB health behavior intervention
1
An Extended Theory of Planned Behavior Intervention to Promote Physical Activity and
2
Healthy Eating Among Older Adults Diagnosed with Type 2 Diabetes and Cardiovascular
3
Disease
4
1
Running Head: TPB health behavior intervention
5
6
Katherine M. Whitea*(PhD), Deborah J. Terryb (PhD), Carolyn Troupb (PhD), Lynn A.
7
Rempelc (PhD), Paul Normand (PhD), Kerry Mummerye (PhD), Malcolm Rileyf (PhD),
8
Natasha Posnerg (PhD), and Justin Kenardyb (PhD)
9
10
a
11
4001 Australia
12
b
13
c
14
d
15
e
16
Australia
17
f
18
g
School of Psychology and Counselling, Queensland University of Technology, Brisbane,
School of Psychology, The University of Queensland, St Lucia, Brisbane, 4072 Australia
Department of Nursing, Brock University, St. Catharines, ON L2S 3A1 Canada
Department of Psychology, University of Sheffield, Sheffield, S10 2TP UK
College of Health and Human Services, Central Queensland University, Rockhampton, 4702
Dairy Australia, Level 5 IBM Centre, 60 City Road Southbank Victoria 3006 Australia
University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK
19
20
*
21
km.white@qut.edu.au
Corresponding author. Phone: +61 7 3138 4689; Fax: + 61 7 3138 0486; Email:
22
23
Acknowledgment of financial and/or other support: This study was conducted as part of a
24
grant to the second author from Queensland Health, Diabetes Australia – Queensland and The
25
Heart Foundation – Queensland.
TPB health behavior intervention
2
Acknowledgements
1
2
The authors would like to thank Kylie Burton and Barbara Sponza, at Diabetes Australia –
3
Queensland for their assistance in obtaining volunteers for this study. Thanks also to: Theresa
4
Collison, Jan Coad, Ruth Dukes, Ann Dyne, Helen Elliott, Claire Hyde, Rene Hinton, Debbie
5
McGrath, Betty Mulder, Susan Mylne, Cheryl Pearson, Gaylene Weir and Cindy Wood for
6
their assistance in data collection.
7
8
TPB health behavior intervention
3
Abstract
1
2
A randomized controlled trial evaluated the effectiveness of a 4-week extended Theory of
3
Planned Behavior (TPB) intervention to promote regular physical activity and healthy eating
4
among older adults diagnosed with Type 2 diabetes and/or cardiovascular disease (N = 183).
5
Participants completed TPB measures of attitude, subjective norm, perceived behavioral
6
control, and intention, as well as planning and behavior, at pre-intervention, 1-week, and 6-
7
weeks post-intervention for each behavior. No significant time by condition effects emerged
8
for healthy eating. For physical activity, significant time by condition effects were found for
9
behavior, intention, planning, perceived behavioral control, and subjective norm. In particular,
10
compared to control participants, the intervention group showed short-term improvements in
11
physical activity and planning with further analyses indicating that the effect of the
12
intervention on behavior was mediated by planning. The results indicate that TPB-based
13
interventions including planning strategies may encourage physical activity among older
14
people with these serious conditions.
15
16
TPB health behavior intervention
4
1
Over the past decade, the incidence of chronic diseases such as diabetes and
2
cardiovascular disease (CVD) has risen substantially in developed countries. Between 2004
3
and 2005, approximately 18 percent of the Australian population was diagnosed with CVD,
4
placing enormous costs on the health care system (Australian Institute of Health and Welfare,
5
2006). Similarly, based on findings from the first national diabetes lifestyle study, AusDiab, 1
6
in 4 individuals has diabetes or is at a high risk of developing the disease within the next 5 to
7
10 years (Diabetes Australia, 2004). As well as being an independent risk factor for CVD,
8
Type 2 diabetes shares similar risk factors to CVD and many individuals suffer from
9
symptoms of both diseases. In further evidence for the similarities between the two conditions
10
of CVD and Type 2 diabetes, the American Heart Association states that "from the point of
11
view of cardiovascular medicine, it may be appropriate to say, 'diabetes is a cardiovascular
12
disease’” (Grundy et al., 1999, p. 1134). Consequently, the need for effective interventions
13
and improved management of these conditions is essential.
14
It is generally accepted that adherence to healthy eating patterns is central to the
15
prevention and optimal management of these two chronic conditions (e.g., Vessby, 2000).
16
Dietary guidelines for individuals with diabetes and CVD are usually the same as those
17
recommended for the general population. The Australian Guide to Healthy Eating published
18
by the Commonwealth Department of Health and Ageing (2001) recommends daily intake of
19
foods based on three general principles; namely, (1) reducing the dietary consumption of fat,
20
with an emphasis on decreasing saturated fat, (2) increasing the consumption of
21
carbohydrates, particularly those that are more slowly digested and, (3) increasing the intake
22
of vegetables and fruit. There are a plethora of examples of interventions to encourage healthy
23
eating among the general population (e.g., Anderson et al., 1998), including some focusing on
24
healthy eating specifically among older people (e.g., Kelley & Abraham, 2004), with varying
25
degrees of success in meeting their intervention goals.
TPB health behavior intervention
1
5
In addition to healthy eating, improved physical activity is also important for the
2
prevention and optimal management of these two chronic conditions (Kavookjian, Elswick, &
3
Whetsel, 2007; Vessby, 2000; Warburton, Nicol, & Brenin, 2006). As with the general
4
population, individuals with diabetes and CVD are recommended to undertake 30 or more
5
minutes of moderate physical activity during their leisure time on 5 or more days of the week
6
(Australian Government Department of Health and Ageing, 2005). Numerous interventions to
7
improve physical activity levels (Dishman & Buckworth, 1996; Hillsdon, Foster, &
8
Thorogood, 2005; Kahn et al., 2002; Muller-Riemenschneider, Reinhold, Nocon, & Willich,
9
2008), including some specifically aimed at people diagnosed with diabetes or CVD (Furber
10
et al., 2008; Gleeson-Kreig, 2006; Graham-Clarke & Oldenburg, 1994; Richardson et al.,
11
2007; Steptoe, Doherty, Rink, Kerry, Kendrick, & Hendrick, 1999; Wing, Vendetti, Jakicic,
12
Polley, & Lang, 1998), have been developed using a number of different approaches (e.g.,
13
pedometer wearing/diary recording, behavior therapy, self-monitoring) to encourage positive
14
changes in people’s physical activity levels. However, while some improvements to people’s
15
physical activity levels have been observed (Page, Harnden, Cook, Turner, 1992; Simons-
16
Morton, Calfas, Oldenburg, & Burton, 1998; Steptoe et al., 1999; Wing et al., 1998) they have
17
typically been, at best, only modest and often not maintained. Furthermore, the randomized
18
controlled trials that specifically target physical activity in individuals at risk of, or with
19
diabetes or CVD, often lack a theoretical basis that specifies the underlying process of
20
decision-making for the desired behavior change.
21
Previous researchers have drawn on a variety of social cognitive theories, such as the
22
theory of planned behavior (TPB; Ajzen, 1991), to engender health-related behavior change
23
(see Hardeman, Johnston, Johnston, Bonetti, Wareham, & Kimmonth, 2002). According to
24
the TPB (Ajzen, 1991), the immediate antecedent of behavior is intention. Intentions are
25
determined by attitude, subjective norm, and perceived behavioral control. Attitudes are an
TPB health behavior intervention
6
1
individual’s overall positive or negative evaluation of performing the behavior. Subjective
2
norms represent perceived (dis)approval from significant others for behavioral performance.
3
Perceived behavioral control comprises the perceived extent to which the behavior is under
4
the person’s control and influences both intentions and behavior (when estimates of actual
5
control are accurate). Each of the TPB constructs is determined by underlying belief sets
6
focusing on the perceived outcomes of the behavior, beliefs of specific referents, and
7
facilitating and inhibiting factors. Much support has been provided for the TPB across a range
8
of behaviors (Armitage & Conner, 2001), and for health-related behaviors specifically
9
(McEachan, Conner, Taylor, & Lawton, in press), including healthy eating (e.g., Conner,
10
Norman, & Bell, 2002; Payne, Jones, & Harris, 2004) and physical activity (Hagger,
11
Chatzisarantis, & Biddle, 2002; Hausenblas, Carron, & Mack, 1997). In addition, the TPB has
12
been used to predict physical activity (Boudreau & Godin, 2009; Plotnikoff, Lippke,
13
Courneya, Birkett, & Sigal, 2010) and healthy eating (White, Terry, Troup, Rempel, &
14
Norman, 2010) for people diagnosed with Type 1 and Type 2 diabetes and/or CVD.
15
Despite this support, some researchers (Gollwitzer, 1993, 1999; Sheeran, 2002) have
16
argued that the TPB (and other social cognition models) should be expanded to include
17
volitional variables (i.e., self-regulatory strategies that encourage enactment of one’s
18
intention) to aid in the prediction of behavior. In this respect, Norman and Conner (2005)
19
suggest that planning is a key volitional variable that can aid the transition from intention to
20
behavior. Thus, specifying where, when, and how a behavior is to be performed may ensure
21
that strong intentions are translated in behavior. In their study examining undergraduate
22
students’ physical activity behavior, Norman and Conner found evidence to suggest that the
23
impact of intentions on behavior may be mediated by the extent to which one undertakes
24
planning activities (see also Gutiérrez-Doña, Lippke, Renner, Kwon, & Schwarzer, 2009;
25
Jones, Abraham, Harris, Schulz, & Chrispin, 2001; Luszczynska & Schwarzer, 2003; van
TPB health behavior intervention
1
Osch, Beenackers, Reubsaet, Lechner, Candel, & de Vries, 2009). Similarly, experimental
2
work has shown that forming implementation intentions (i.e., specific if-then plans that link
3
an appropriate behavioral response to a situational cue) helps people to translate their goal
4
intentions into behavior (Gollwitzer & Sheeran, 2006). Adhering to healthy eating choices
5
and performing physical activity often requires a series of preparatory steps. For healthy
6
eating, these steps include selecting appropriate products, purchasing the items, choosing
7
recipes, and preparing and cooking meals and, for physical activity, there are considerations
8
such as what activity to perform, acquiring the relevant clothing and/or equipment, and
9
making plans with others. Therefore, the role of planning as a potential mediating variable
10
11
7
between intentions and behavior in an extended TPB framework was examined.
Given that the TPB has been shown to strongly predict healthy eating choices and
12
physical activity in a range of contexts and for a variety of populations, and planning appears
13
to play an important role in the intention-behavior relationship, an extended TPB
14
(incorporating planning) should provide a good basis for the development of interventions to
15
improve levels of physical activity. In their review of 24 TPB-based behavioral interventions,
16
Hardeman et al. (2005) concluded that the TPB was mainly used to predict intentions and
17
behaviors and less commonly to develop intervention programs. Importantly, Hardeman et al.
18
reported that approximately half of the interventions were effective in changing intentions and
19
two-thirds in changing behavior, albeit with small effect sizes (where calculable). There are
20
several TPB-informed healthy eating (Gratton, Povey & Clark, 2007; Jemmott et al., 2011)
21
and physical activity (Chatzisarantis & Hagger, 2005; Parrott, Tennant, Olejnik, &
22
Pouderijne, 2008) interventions for non-clinical populations; however, only a few TPB
23
interventions have been reported for clinical populations (Jones, Courneya, Fairey, Mackey,
24
2005; Kelley & Abraham, 2004).
25
The Current Study
TPB health behavior intervention
1
8
The aim of this research was to design, deliver, and evaluate an extended TPB-based
2
intervention. The intervention aimed to specifically target people diagnosed with Type 2
3
diabetes and/or CVD. This study aimed to determine the effect of an intervention targeting
4
people diagnosed with Type 2 diabetes and/or CVD on healthy eating and physical activity
5
and to compare the effect of the intervention with that of no intervention on outcomes at 1 and
6
6 weeks post-intervention.
7
The present study used an extended TPB for both the development of the content of
8
the intervention as well as for the purposes of measurement and intervention evaluation. The
9
healthy eating goal of the intervention was based on recommendations contained in the
10
Dietary Guidelines for Australians and the Australian Guide to Healthy Eating
11
(Commonwealth Department of Health and Ageing, 2001). These recommendations include:
12
(1) using only low-fat dairy products, (2) using a monounsaturated or polyunsaturated
13
cooking oil and (3) trimming all visible fat from meats. The physical activity goal of the
14
intervention was based on the current Australian physical activity recommendations for adults
15
(Australian Government Department of Health and Ageing, 2005) as well as
16
recommendations of The Heart Foundation - Australia and Diabetes Australia which
17
recommend regular moderate physical activity be performed for individuals with diabetes
18
and/or CVD. The broad physical activity intervention goal for participants, then, was to
19
undertake 30 or more minutes of moderate physical activity during their leisure time on 5 or
20
more days of the week (with the 30 minutes of activity able to be done in bouts of at least 10-
21
minute blocks). However, it is also suggested that sedentary individuals should increase their
22
activity levels slowly, and it is recommended that elderly individuals increase their moderate
23
level activity by no more than 5% per week (Christmas & Anderson, 2000). Thus, a revised
24
target behavior based on previous recommendations (see Terry & Hogg, 1996) was defined as
25
engaging in moderate physical activity on a regular basis (i.e., ‘moderate physical activity’
TPB health behavior intervention
9
1
was defined as any movement that causes a slight but noticeable increase in breathing and
2
heart rate and may cause light sweating in some people and ‘a regular basis’ was defined as at
3
least three 3 occasions per week).
4
It was hypothesized that, compared to a waitlist control group, participants in the
5
intervention group would show an improvement in their pre-intervention to 1 week post-
6
intervention ratings for (1) the extended TPB constructs of attitudes, subjective norm,
7
perceived behavioral control, intentions, and planning in relation to healthy eating and
8
engaging in regular physical activity and (2) self-report assessments of healthy eating and
9
physical activity, and that this improvement from pre-intervention to 1 week post-intervention
10
would be maintained at 6 weeks post-intervention. It was also predicted that the effects of the
11
extended TPB intervention on behavior would be mediated by planning.
Methods
12
13
A randomized controlled trial was conducted to evaluate the effectiveness of a 4-week
14
extended Theory of Planned Behavior (TPB) intervention. The university ethics committee
15
and relevant government health district authorities approved the study.
16
Participants
17
Time 1 (pre-intervention) participants were 183 predominantly older adults diagnosed
18
with Type 2 diabetes and/or CVD recruited from 7 community health center sites in
19
Queensland, Australia. Participants were recruited voluntarily at an advertised information
20
session and provided their informed consent to participate in a 4-week intervention trial
21
designed to encourage healthy behaviors. Of the Time 1 participants, 116 provided data at all
22
three collection points (pre-intervention, 1 week post-intervention, and 6 weeks post-
23
intervention) on healthy eating, and 111 on physical activity. The participants comprised
24
primarily of (note there is a small amount of missing data for each demographic variable):
25
older (M = 61.17 years; SD = 8.81), Caucasian (n = 107, 99%), married (n = 83, 76%),
TPB health behavior intervention
10
1
females (n = 67, 61%). The most commonly reported occupations were retired (n = 40, 39%)
2
and homemaker (n = 32, 31%). Half of the participants (n = 54, 49%) reported being
3
diagnosed with diabetes only, a further 45% (n = 49) had been diagnosed with both Type 2
4
diabetes and CVD, and 6% (n = 7) had a CVD only diagnosis. The average length of time
5
since diagnosis was 5.55 years (SD = 6.60) for people with a Type 2 diabetes diagnosis and
6
11.69 years (SD = 12.08) for people with a CVD diagnosis.
7
The flow of participants throughout the study is shown in Figure 1. Considering
8
responses to the healthy eating questionnaires, 27% of participants (intervention n = 33, 25%;
9
control n = 11, 32%) were lost to follow-up at 1-week post-intervention. This number
10
increased to 37% (intervention n = 46, 35%; control n = 21, 40%) at 6-week follow-up. For
11
the physical activity questionnaires, 27% of participants (intervention n = 31, 24%; control n
12
= 18, 34%) were lost to follow-up at 1-week post-intervention and this number increased to
13
39% (intervention n = 47, 37%; control n = 23, 43%) at 6-week follow-up. No significant
14
differences were found between attrition rates for the intervention and control conditions at
15
both follow-up time points. Moreover, no significant differences were also found between
16
completers and non-completers at both follow-up time points on the demographic factors and
17
pre-intervention extended TPB variables.
<Insert Figure 1 about here>
18
19
20
Design and Procedure
The design of the study was a 2 (Condition: intervention vs. waitlist control) by 3
21
(Time: pre-intervention vs. 1-week post-intervention vs. 6-weeks post-intervention) mixed
22
measures design, with Time as the repeated measures factor. Participants were randomly
23
assigned via a color-coded draw into 2 groups; two thirds were assigned to an intervention
24
group and one third to the waitlist control group. The uneven design of the 2 groups was
25
based on the collaborating community health center sites requiring their patients/clients to
TPB health behavior intervention
11
1
benefit from the intervention as early as possible and to avoid a large wait-list group. A pre-
2
intervention questionnaire assessing the extended TPB constructs (including planning and
3
past behavior), for healthy eating and physical activity was completed at the information
4
session held 1 week prior to the intervention. Follow-up questionnaires assessing the extended
5
TPB constructs, healthy eating, and physical activity were mailed to all participants at 1 and 6
6
weeks after completion of the 4-week intervention. Most of the data collection period
7
occurred across an annual holiday period (Australian Summer encompassing Christmas and
8
New Year).
9
Control Group. Participants in the control group received no intervention during the
10
data collection phase. These participants completed the extended TPB questionnaire at the
11
three data collection points (pre-intervention, 1-week post-intervention, and 6-weeks post-
12
intervention). People in the control group were offered the opportunity to attend intervention
13
sessions once all follow-up questionnaires were returned.
14
Intervention Group. The intervention group received an extended TPB intervention
15
consisting of weekly 2-hour sessions held over a 4-week period. The intensity of the
16
intervention was based on the ability to be able to engage both participants and facilitators in
17
a brief, cost-effective program that may serve to complement other offerings available to
18
those individuals diagnosed with these chronic conditions The sessions were facilitated by
19
health professionals (e.g., diabetes educators, physiotherapists) trained in the program’s
20
delivery by research team members during a 1 or 2 day program, with intervention sessions
21
conducted in a primarily interactive mode with participants. The sessions were group-based;
22
approximately 5 to 12 participants attended each session across the multiple test sites. The
23
comprehension of the learning in each session was checked by the facilitators at the end of
24
each of the four sessions to ensure that participants understood the issues that were raised in
25
the workshop sessions. An important component of the extended TPB intervention was the
TPB health behavior intervention
12
1
use of participants’ pre-intervention attitudes and beliefs about healthy eating and physical
2
activity. These beliefs were used as a basis for discussion and to assist in the development of
3
intentions to engage in healthy behaviors.
4
Intervention protocol. The weekly 2-hour intervention sessions held over a 4 week
5
period covered a series of TPB-related topics. Session 1 explored participants’ attitudes and
6
beliefs about healthy eating and physical activity. Pre-intervention information of the group’s
7
overall attitudes and beliefs was used as a basis to consider the perceived advantages (e.g.,
8
feeling healthy, losing weight) and disadvantages (e.g., reducing the taste of food, feeling
9
tired) of adhering to healthy eating choices and engaging in physical activity. In Session 2,
10
participants considered the barriers (e.g., cost, lack of time) that prevent them from making
11
healthy eating choices and engaging in regular physical activity and common triggers to
12
unhealthy behaviors and how unhealthy habits develop. Discussion also focused on the
13
group’s overall perceptions of social support from significant others (subjective norms)
14
adhering to healthy eating choices and performing physical activity. Participants were
15
encouraged to identify people in their lives (e.g., spouse or partner) who impact upon their
16
eating choices and level of physical activity and to develop strategies for dealing with
17
unsupportive individuals or groups. The focus of Session 3 was the role of planning to enact
18
behavior change. Participants learnt and practiced the steps of effective planning (including
19
goal-setting) to bring about achievable behavior change. Session 4 completed the intervention
20
with a focus on fostering a sense of control over behavior change. Participants generated
21
strategies to deal with barriers preventing them from meeting their healthy eating and physical
22
activity goals.
23
Measures
24
In addition to demographic information, the pre-intervention questionnaire and follow-
25
up questionnaires obtained measures of the extended TPB constructs in relation to adhering to
TPB health behavior intervention
13
1
healthy eating choices (i.e., consumption of foods low in saturated fats) and engaging in
2
regular physical activity. Participants completed all measures, as part of a larger
3
questionnaire, assessed at pre-intervention, 1- and 6-weeks post-intervention. All of the TPB
4
items were constructed in line with guidelines specified by Ajzen (1991). Some of these
5
measures included negatively-worded items to minimize response bias. Items were assessed
6
on 7-point Likert scales: 1 (strongly disagree) to 7 (strongly agree) unless otherwise stated.
7
The item examples below are for physical activity where the target behavior was termed
8
“engaging in moderate physical activity on a regular basis during the next month”; for healthy
9
eating, participants completed identical items where the target behavior was termed “eating
10
11
12
13
foods low in saturated fats during the next month”.
Intention. Participants indicated the extent to which they agreed that: “It is likely that
I will engage in moderate physical activity on a regular basis during the next month”.
Attitude. Attitude was assessed on four 7-point evaluative semantic differential items:
14
unpleasant/pleasant, good/bad, negative/positive and favorable/unfavorable, the average of
15
which served as a reliable measure of attitude across the three time points (healthy eating, s
16
= .82, .88 and .89; physical activity, s = .81, .82, and .81).
17
Subjective norm. Subjective norm was assessed using two items: “Most people who
18
are important to me would approve of my engaging in moderate physical activity on a regular
19
basis during the next month” and “Those people who are important to me would want me to
20
engage in moderate physical activity on a regular basis during the next month”. The two
21
items were averaged to create a subjective norm measure which were correlated significantly
22
(p < .001) across the three time points (healthy eating, rs = .72, .79, and .72; physical activity,
23
rs = .59, .87, and .87).
TPB health behavior intervention
1
14
Perceived behavioral control. One item assessed perceived behavioral control: “I am
2
confident that I could engage in moderate physical activity on a regular basis during the next
3
month”.
4
Planning. A planning index was formed based on items by Norman and Conner
5
(2005). For healthy eating, participants completed four items asking the extent to which they
6
had planned “what foods to buy to ensure that you eat foods low in saturated fats during the
7
next month”, “how to prepare meals to ensure that you eat foods low in saturated fats”,
8
“where to purchase food to ensure you eat foods low in saturated fats”, and “how you would
9
handle the situation if you don’t feel like eating foods low in saturated fats”. For physical
10
activity, participants completed six items asking the extent to which they had planned: “how
11
you will engage in moderate physical activity on a regular basis during the next month”,
12
“what physical activities you will engage in”, “when you will be physically active”, “where
13
you will be physically active”, “who you will be physically active with” and “how you will
14
handle the situation if you don’t feel like engaging in physical activity”. Items were
15
completed on 7-point Likert scales from 1 (completely) to 7 (not at all), but reversed scored
16
so that high values indicated high levels of planning. The items were averaged to form a scale
17
which was reliable across the three time points (healthy eating, s = .88, .85, and .86;
18
physical activity, s = .91, .90, and .91).
19
Behavior. Participants were provided with the relevant definitions for the target
20
behaviors. Eating foods low in saturated fats was defined as eating low fat dairy products, fat-
21
trimmed meat, and using mono- and poly-unsaturated oils. Engaging in moderate physical
22
activity on a regular basis was defined as any movement that causes a slight but noticeable
23
increase in breathing and heart rate and may cause light sweating in some people. A regular
24
basis was defined as at least three 3 occasions per week. For both target behaviors,
25
respondents indicated the extent to which they had performed the behaviors during the (1)
TPB health behavior intervention
15
1
past month (baseline and 6-week follow-up) or (2) past week (1-week follow-up). Responses
2
were rated on a 7-point scale, from 1 (a small extent) to 7 (a large extent). In an effort to
3
improve the accuracy of the behavior measures, a memory prompt required participants to
4
complete a checklist indicating their food consumption (including eating foods low in
5
saturated fats) and recreational activities they commonly performed (e.g., walking) during the
6
previous week/month.
Results
7
8
9
Overview of Data Analysis
To examine the utility of the extended TPB intervention, 2 (intervention/control) by 3
10
(pre-intervention/1-week follow-up/6-week follow-up), mixed measures MANOVAs were
11
performed on the extended TPB predictor variables, intentions, and self-report behavior for
12
both healthy eating and physical activity. In addition, on the basis of the MANOVA results
13
for physical activity, correlational and regression analyses were conducted to examine the
14
extent to which the extended TPB predictor variables including planning mediated any
15
extended TPB intervention effects on physical activity. Due to the preliminary nature of the
16
study and the relatively small sample size, alpha was adjusted to .10 for the main analyses.
17
Accordingly, all effects at p < .10 will be interpreted for the main analyses.
18
Prior to the main analyses, two MANOVAs were performed to determine if there were
19
any baseline differences on the extended TPB measures of (i) healthy eating and (ii) physical
20
activity between intervention and control participants, between males and females, or based
21
on diagnosis type (i.e., diabetes only, CVD only, or diabetes and CVD). For healthy eating,
22
these analyses revealed no significant differences as a function of group (intervention or
23
control), F(6, 147) = 0.52, p = .80, gender, F(6, 147) = 0.50, p = .80, or diagnosis, F (12, 294)
24
= 1.31, p = .21. For physical activity, these analyses also revealed no significant differences as
25
a function of group (intervention or control), F(6, 133) = 0.76, p = .60, gender, F(6, 133) =
TPB health behavior intervention
16
1
0.89, p = .51, or diagnosis, F(12, 266) = 1.58, p = .10. Therefore, a combined sample of men
2
and women and a combined diagnosis type sample (i.e., a combined sample of those people
3
diagnosed with diabetes only, CVD only or diagnosed with both conditions) was used in the
4
study’s analyses.
<Insert Table 1 about here>
5
6
The Utility of the TPB as an Intervention to Promote Healthy Eating and Regular
7
Physical Activity
8
9
For the 2 x 3 MANOVAs, the intervention Condition (intervention vs. waitlist control
group) was the between subjects factor and the within-subjects factor was Time (pre-
10
intervention, follow-up at 1 week, and 6 weeks). The extended TPB variables and behavior
11
served as the dependent variables. Table 1 displays the Ms and SDs for the extended TPB
12
measures at pre-intervention, 1-, and 6-week follow-up, for healthy eating and physical
13
activity.
14
15
16
Healthy Eating: Main and interaction effects
For healthy eating, there were no significant Time or Time by Condition effects. There
17
was a significant main effect for Condition, F(6, 72) = 2.78, p =.017, partial η2 = .188, with
18
significant univariate effects for behavior, F(1, 77) = 3.33, p =.072, partial η2 = .041,
19
intention, F(1, 77) = 4.14, p = .045, partial η2 = .05, and perceived behavioral control, F(1,
20
77) = 9.81, p = .002, partial η2 = .113. These results showed that participants in the
21
intervention condition reported higher scores on average across all time points for behavior
22
(M = 6.03, SE = .14), intention (M = 6.25, SE = .09), and perceived behavioral control (M =
23
6.40, SE = .09) than participants in the control condition (M = 5.53, SE = .23 for behavior, M
24
= 5.88, SE = .16 for intention, and M = 5.85, SE = .15 for perceived behavioral control,
TPB health behavior intervention
1
respectively). Given that there was no evidence for the impact of the intervention over time
2
for healthy eating, no further analyses were conducted.
17
3
4
5
Physical Activity: Main effects
For physical activity, there were no significant effects for Condition. A significant
6
effect of Time was found, F(12,63) = 3.10, p = .002, partial η2 = .371. Univariate analyses
7
revealed significant Time effects on intention, F(2, 74) = 6.51, p = .002, partial η2 = .081;
8
perceived behavioral control, F(2, 74) = 3.41, p = .036, partial η2 = .044; and subjective
9
norm, F(2, 74) = 8.02, p < .001, partial η2 = .098, although these effects were qualified by
10
significant Time x Condition effects.
11
12
13
Physical Activity: Interaction effects
Overall, there was a significant multivariate Time x Condition effect, F(12, 63) =
14
1.70, p = .089, partial η2 = .244. Univariate tests were conducted to examine these effects in
15
more detail which revealed significant Time x Condition effects on behavior, F(2, 74) = 2.81,
16
p = .067, partial η2 = .037; planning, F(2, 74) = 6.12, p = .004, partial η2 = .076; intention,
17
F(2, 74) = 8.36, p < .001, partial η2 = .101; perceived behavioral control, F(2, 74) = 5.79, p =
18
.004, partial η2 = .073; and subjective norm, F(2, 74) = 2.64, p = .077, partial η2 = .034.
19
There was no significant Time x Condition effect on attitude.
20
These Time x Condition interactions were examined further, with corrections to
21
control for the Type 1 error rate (alpha set at .05). For self-reported behavior, there was a
22
significant difference across time in the intervention, F(2, 88) = 5.35, p = .006, partial η2 =
23
.108, but not in the control condition, F(2, 88) = .75, p = .475, partial η2 = .017 . Pairwise
24
comparisons (utilizing a Bonferroni adjustment with alpha set at .008) for the simple effects
25
of Time within the intervention condition were then conducted. These results showed that the
26
pre-intervention levels of self-reported physical activity increased significantly at the 1 week
TPB health behavior intervention
18
1
follow-up. There were no other significant differences. For the planning of physical activities,
2
there was a significant difference across time in the intervention, F(2, 91) = 3.67, p = .029,
3
partial η2 = .075, but not the control condition, F(2, 91) = 2.43, p = .094, partial η2 = .051.
4
Pairwise comparisons for the simple effects of Time within the intervention condition showed
5
that the pre-intervention levels of planning for engaging in physical activities increased
6
significantly at the 1-week follow-up. There were no other significant differences.
7
For intention to engage in physical activity, there was a significant difference across
8
time in the control, F(2, 92) = 13.05, p < .001, partial η2 = .221, but not the intervention
9
condition, F(2, 92) = .71, p = .50, partial η2 = .015. Pairwise comparisons for the simple
10
effects of Time within the control condition revealed that pre-intervention physical activity
11
intention levels reduced significantly between baseline and the 1-week follow-up and between
12
baseline and the 6-week follow-up. There were no other significant differences. For perceived
13
behavioral control, there was a significant difference across time in the control, F(2, 91) =
14
8.80, p < .001, partial η2 = .162, but not the intervention condition, F(2, 91) = .56, p = .57,
15
partial η2 = .012. Pairwise comparisons for the simple effects of Time within the control
16
condition showed a significant reduction in perceived behavioral control levels for physical
17
activity between pre-intervention and 1-week follow-up. No other significant differences
18
emerged. For subjective norm, there was a significant effect of Time in the control, F(2, 92) =
19
6.56, p = .002, partial η2 = .125, but not the intervention condition, F(2, 92) = 1.87, p = .16,
20
partial η2 = .039. Pairwise comparisons for the simple effects of Time within the control
21
condition showed a significant reduction in levels of perceived pressure to engage in physical
22
activity (i.e., subjective norm) between pre-intervention and 1-week follow-up. There were no
23
other significant differences.
24
25
Physical Activity: Mediation Analyses
TPB health behavior intervention
1
19
Given that all but one of the significant Time x Condition interactions for physical
2
activity were due to changes observed between baseline and 1-week follow-up, further
3
analyses were conducted focusing on the effect of the intervention from pre-intervention to
4
the 1-week follow-up only. These results revealed a significant Time x Condition effect, F(6,
5
101) = 3.93, p = .001, partial η2 = .189. This Time x Condition interaction was examined
6
further, with Bonferroni corrections. Importantly, for self-reported behavior, there was a
7
significant difference across time in the intervention condition, F(1, 126) = 9.44, p = .003,
8
partial η2 = .070, but not in the control condition, F(1, 126) = 2.17, p = .14, partial η2 = .017.
9
Participants in the intervention condition showed a significant increase in self-reported
10
physical activity at the 1 week follow-up whereas control participants did not. The next
11
analysis assessed the extent to which the extended TPB variables at 1-week post-intervention,
12
especially planning, mediated the impact of the intervention on the target behavior at 1-week
13
follow-up where an improvement in self-reported behavior from pre-intervention ratings was
14
observed for experimental participants. As shown in Table 2 (showing variables at 1-week
15
post-intervention), all extended TPB predictors were significantly correlated with behavior,
16
with planning emerging as the strongest correlate, followed by intention. Procedures
17
developed by Preacher and Hayes (2008) were used to examine the mediational hypotheses.
18
Condition was entered along with all potential mediators (i.e., the extended TPB variables)
19
simultaneously and the pre-intervention behavior scores as a covariate. The analyses revealed
20
that Condition (IV) had a significant effect on behavior (DV), B = .80, SE = .35, p = .02.
21
However, this effect became non-significant, B = .26, SE = .33, p = .43, when the extended
22
TPB variables (i.e., the potential mediators) were controlled for, suggesting mediation (Baron
23
& Kenny, 1986). Using bootstrapping procedures, the total mediated effect was found to be
24
significant, B = .53, SE = .25, CI = .121 to 1.098. Inspection of the individual mediator
TPB health behavior intervention
1
variables revealed that only planning significantly mediated the effect of Condition on
2
behavior, B = .25, SE = .15, CI = .041 to .681.
3
<Insert Table 2 about here>
4
Discussion
5
20
The present study tested the efficacy of an extended TPB-based intervention including
6
planning in relation to healthy eating and physical activity in predominantly older adults
7
diagnosed with Type 2 diabetes and/or CVD. The results of this study provided only some
8
support for the efficacy of the intervention to the extent that, for the intervention participants,
9
there was evidence of significant short-term improvement in their self-reported physical
10
activity behavior and degree of planning to engage in such activities following the conclusion
11
of the intervention sessions. In contrast, participants in the control condition maintained only
12
moderate levels of planning and activity during this time. However, intervention participants
13
did not report any significant improvement (or maintenance) in their level of planning and
14
self-reported behavior at the 6-week post-intervention follow-up. As a result, these somewhat
15
encouraging findings for physical activity can only be described as temporary as the positive
16
shifts in planning and behavior are potentially reactive to participation in an intervention
17
targeting behavior change.
18
Effectiveness of the Intervention on Social Cognitions and Intention
19
There was no evidence that the intervention facilitated change for participants’ healthy
20
eating cognitions or intentions. It is possible that, as diabetes in particular is recognized as a
21
food-related condition, cognitions and intentions were already at or close to desired levels.
22
Inspection of the means for the extended TPB variables for healthy eating revealed high levels
23
of reported endorsement for the constructs for both conditions across the time-points.
24
25
In relation to physical activity, while not originally the source of change expected,
some support for the efficacy of the extended TPB intervention was evidenced by stability in
TPB health behavior intervention
21
1
people’s perceptions of normative pressure from others, control, and behavioral intentions
2
across time for participants in the intervention condition compared to participants in the
3
control condition. The latter participants reported a decrease in scores on these constructs over
4
time (with this decrease significant only in the short term for subjective norm and perceived
5
behavioral control but significant between baseline and both 1- and 6- week follow-up time
6
points for intention). Participants in the control condition perceived less normative pressure,
7
less control, and had weaker intentions in relation to physical activity than participants
8
assigned to the intervention condition across time, whereas participants in the intervention
9
condition maintained their high levels of perceived normative pressure, a sense of control, and
10
strong intentions in relation to engaging in regular physical activity. This finding may be
11
explained by the timing of the data collection period across the Australian Summer
12
(encompassing Christmas and New Year) holiday period which traditionally is a time when
13
healthy practices, such as physical activity, can be compromised by an increase in social
14
activities and changes from one’s usual daily routine while on leave or in non-familiar
15
settings if holidaying. Thus, the intervention may have helped participants maintain their
16
positive cognitions. Contrary to expectation, there was no significant improvement across
17
time on the measure of attitude for intervention, as opposed to control, participants. Across
18
the course of the study, intervention and control participants reported similarly very positive
19
attitudes towards engaging in physical activity, with these very favorable opinions leaving
20
little room for improvement on this construct for participants in either condition.
21
Effectiveness of the Intervention on Behavior
22
As was the case for social cognitions and intention, there was no evidence that the
23
intervention served to improve participants’ healthy eating behavior over time. As stated
24
above for cognitions and intention, reports of adherence to healthy eating choices for
TPB health behavior intervention
22
1
consumption of foods low in saturated fats for all participants were high, indicating a strong
2
endorsement of recommended dietary guidelines for participants in the present study.
In a similar vein to the findings of Hardeman et al.’s (2002) review of TPB-based
3
4
interventions, the present study found support for an improvement in reported physical
5
activity behavior, albeit with a similarly small effect size. As hypothesized, whereas there was
6
no evidence of change for participants in the control condition, there was a significant
7
increase in reported levels of physical activity for intervention participants at the 1 week
8
follow-up. This encouraging finding suggest some efficacy of the intervention in promoting
9
adherence to regular, moderate physical activity, albeit over a short period of time.
10
Mediation Effects of Planning
The present study also examined the extent to which the effects of the extended TPB
11
12
intervention on behavior would be mediated by planning. According to Norman and Conner
13
(2005), there is some evidence to suggest that the impact of cognitions on behavior is
14
mediated by the extent to which one undertakes sufficient planning for engagement in the
15
behavior (Jones et al., 2001; Luszczynska & Schwarzer, 2003; van Osch et al., 2009). The
16
results of the present study revealed that planning mediated the impact of the TPB-based
17
intervention on reported levels of physical activity at one-week follow-up, indicating that the
18
planning strategies people engage in serve as the means by which the positive impact of the
19
intervention translates into behavioral change. This role for planning highlights the
20
importance of examining variables comprising the factors proposed to serve as proximal
21
facilitators of behavior change (see also Gollwitzer, 1993, 1999, regarding implementation
22
intentions) and points to the value of considering theories that focus on self-regulation (e.g.,
23
including planning and goal-setting) as a basis for intervention (e.g., the action and coping
24
planning components of the Health Action Process Approach (HAPA) model; Schwarzer,
25
1992).
TPB health behavior intervention
1
2
23
Study Strengths and Limitations
The study has a number of strengths as it addressed a number of criticisms noted by
3
Hardeman et al.’s (2002) review of TPB interventions by (i) utilizing the model to inform the
4
intervention, (ii) stating explicitly which TPB components were targeted, (iii) using a
5
randomized controlled design, and (iv) employing standardized measures of TPB constructs
6
as process and outcome measures. Nonetheless, some limitations of the study should be noted.
7
Importantly, due to the preliminary nature of the study and the relatively small sample size,
8
marginal effects (adopting a p <.10 significance cutoff) were interpreted for the main analyses
9
due to their theoretical significance but should be considered cautiously, pending replication
10
in future studies. In relation to methodological issues, the majority of the sample was
11
Caucasian and married, bringing into question the generalizability of the findings. In addition,
12
due to time constraints, single-item measures for two of the study’s components (intention
13
and perceived behavioral control) were used. However, multi-item measures of these
14
constructs typically have very high levels of internal reliability, suggesting that the use of
15
single items in this context may not have unduly impacted on issues of reliability. Further, the
16
reliance on self-report data may have inflated people’s assessment of their performance of the
17
two health behaviors. Baseline levels on behavior were high which may indicate social
18
desirability effects or the selectiveness of the sample in terms of motivation to change. In the
19
case of physical activity, the use of objective measures may provide a more accurate level of
20
people’s accumulated physical activity, although self-report measures of physical activity
21
correlate significantly with maximum oxygen consumption (VO2max) (Godin & Shephard,
22
1997). Nonetheless, the present findings should be replicated using objective measures of
23
physical activity behavior (McAuley & Jacobson, 1991; Westerterp, 2009). It is possible also
24
that the mere measurement effect, whereby merely asking questions about intentions to
25
engage in healthy behaviors increases the performance of healthy behavior (e.g., French &
TPB health behavior intervention
1
Sutton, 2010; Godin et al., 2010; Godin, Sheeran, Conner, & Germain, 2008; Sandberg &
2
Conner, 2009), may have reduced the size of the intervention effects.
3
24
An additional limitation relates to the finding that the reported means on the TPB
4
measures were all high (Ms ≈ 6.00 on a 7 point scale) pre-intervention. As a result, there was
5
little room for improvement which meant that the intervention aimed to prevent reductions
6
over time on these variables. However, for planning, the mean pre-intervention score was
7
closer to the mid-point of the scale, providing greater scope for the intervention to increase
8
levels of planning. It should be noted also that stronger effects may have occurred had all data
9
collection taken place in a non-holiday season when sedentary and unhealthy eating behavior
10
may be less common. The timing of the data collection may have provided a more
11
conservative test of the intervention and may also explain the reduced intervention effects and
12
the direction of the findings, although this suggestion remains to be tested. It is unclear, then,
13
whether the intervention would have had similar effects at other times of the calendar year. A
14
further limitation when considering the results of the study is the possibility that the effects of
15
the intervention may relate to the impact of others members within any given intervention
16
session given the group-based nature of the program.
17
A final limitation is that there was a substantial amount of attrition for participants in
18
the study which may have impacted on the power of the analyses. In addition, previous
19
research has indicated that program non-completers in a physical activity intervention for
20
older adults may be more likely to be of a lower socioeconomic class, overweight, and less
21
physically active (Jancey, Lee, Howat, Clarke, Wang, & Shilton, 2007), although no attrition
22
biases were found in the current study. Future research should continue to assess the utility of
23
the intervention in larger samples comprising other clinical and non-clinical groups of
24
participants to examine the extent to which the findings are generalizable across broader
25
populations. Further, reasons for participant dropout should be collected as a matter of course.
TPB health behavior intervention
25
1
Finally, studies should also seek to examine cognitive and behavior change in data collection
2
periods not coinciding with holiday periods, although these times are noted for their difficulty
3
in maintaining healthy behaviors so are useful to examine in their own right.
4
Conclusions
5
Overall, the results of the study provide some limited, short-term, evidence for the
6
efficacy of an extended TPB-based intervention to increase physical activity among older
7
adults diagnosed with Type 2 diabetes and/or CVD; however, the intervention had no effect
8
on healthy eating. Nevertheless, there are some important implications for the context of
9
diabetes and chronic condition education. First, the findings suggest than an emphasis on
10
planning strategies (including goal-setting) would be most beneficial in encouraging healthy
11
lifestyle adherence in regular moderate physical activity among this cohort (i.e., developing
12
detailed plans of how to engage in regular moderate physical activity such as when, where
13
and with whom). Second, for physical activity behavior maintenance during often difficult
14
periods for adherence such as holidays, efforts focusing on the perceived approval from others
15
(e.g., partners, families) to be physically active and bolstering one’s self-confidence to be able
16
to maintain regular physical activity may prove helpful. It is likely to be less useful to target
17
people’s attitudes (i.e., highlighting the benefits and minimizing the costs of regular physical
18
activity) as positive attitudes towards behavioral performance appear to be already
19
established, at least in the present sample. Finally, given that the current intervention only had
20
a short term effect on physical activity behavior, it may be useful to supplement interventions
21
with “booster” reminders after the completion of the program in order to maintain early
22
changes in physical activity behavior.
23
TPB health behavior intervention
26
1
References
2
Ajzen, I. (1991). Theory of planned behavior. Organizational Behavior and Human Decision
3
Processes, 50, 179–211. doi:10.1016/0749-5978(91)90020-T.
4
Anderson, A. S., Cox, D. N., McKellar, S., Reynolds, J., Lean, M. E. J., & Mela, D. J. (1998).
5
Take five, a nutrition education intervention to increase fruit and vegetable intakes:
6
Impact on attitudes towards dietary change. British Journal of Nutrition, 80, 133-140.
7
Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behaviour: A meta-
8
analytic review. British Journal of Social Psychology, 40, 471–499.
9
doi:10.1348/014466601164939.
10
Australian Government Department of Health and Ageing. (2005). An active way to better
11
health: national physical activity guidelines. Canberra: Commonwealth of Australia.
12
Australian Institute of Health and Welfare (2006). Cardiovascular Disease FAQs. Retrieved
13
14
December 18, 2007 from http://www.aihw.gov.au/cvd/faqs.cfm.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
15
psychological research: Conceptual, strategic, and statistical considerations. Journal of
16
Personality and Social Psychology, 51, 1173-1182. doi:10.1037/0022-3514.51.6.1173.
17
Boudreau, F., & Godin, G. (2009). Understanding physical activity intentions among French
18
Canadians with type 2 diabetes: an extension of Ajzen's theory of planned behavior.
19
International Journal of Behavioral Nutrition and Physical Activity, 6, 35-45.
20
doi:10.1186/1479-5868-6-35.
21
Chatzisarantis, N. L. D, & Hagger, M. S. (2005). Effects of a brief intervention based on the
22
theory of planned behaviour on leisure-time physical activity participation. Journal of
23
Sport and Exercise Psychology, 27, 470-487.
24
25
Christmas, K., & Anderson, R. A. (2000). Exercise and older patients: Guidelines for the
clinician. Journal of the American Geriatric Society, 48, 318-324.
TPB health behavior intervention
1
2
3
4
5
6
Commonwealth Department of Health and Ageing (2001). The Australian guide to healthy
eating. Retrieved May 30, 2002, from www.health.gov.au/pubhlth/strateg/food/guide/.
Conner, M., Norman, P., & Bell, R. (2002). The theory of planned behavior and healthy
eating. Health Psychology, 21, 194-201. doi: 10.1037//0278-6133.21.2.194
Diabetes Australia (2004). AusDiab Preliminary Results (media release). Retrieved
November 17, 2004 from http://www.diabetes.com.au/research/AusDiab.htm#media.
7
Dishman, R. K., & Buckworth, J. (1996). Increasing physical activity: A quantitative
8
synthesis. Medicine and Science in Sports and Exercise, 28, 706-719.
9
French, D. P., & Sutton, S. (2010). Reactivity of measurement in health psychology: How
10
much of a problem is it? What can be done about it? British Journal of Health
11
Psychology, 15, 453-468. doi: 10.1348/135910710X492341.
12
27
Furber, S., Monger, C., Franco, L., Mayne, D., Jones, L. A., Laws, R., & Waters, L. (2008).
13
The effectiveness of a brief intervention using a pedometer and step-recording diary in
14
promoting physical activity in people diagnosed with type 2 diabetes or impaired
15
glucose tolerance. Health Promotion Journal of Australia, 19, 189-195.
16
Gleeson-Kreig, J. M. (2006). Self-monitoring of physical activity. Effects on self-efficacy and
17
behavior in people with type 2 diabetes The Diabetes Educator, 32, 69-77. doi:
18
10.1177/0145721705284285.
19
Godin, G., Sheeran, P., Conner, M., Delage., G., Germain, M., Belanger-Gravel, A., &
20
Naccache, H. (2010). Which survey questions changes behavior? Randomized
21
controlled trial of mere measurement interventions. Health Psychology, 29, 636-644.
22
doi: 10.1037/a0021131.
23
Godin, G., Sheeran, P., Conner, M., & Germain, M. (2008). Asking questions changes
24
behavior: Mere measurement effects on frequency of blood donation. Health
25
Psychology, 27, 179-184. doi: 10.1037/0278-6133.27.2.179.
TPB health behavior intervention
1
2
Godin, G., & Shephard, R. J. (1997). Godin leisure-time exercise questionnaire. Medicine &
Science in Sports & Exercise, 29, S36-8.
3
Gollwitzer, P. M. (1993). Goal achievement: The role of intentions. In W. Stroebe & M.
4
Hewstone (Eds.), European Review of Social Psychology (Vol. 4, pp. 141–185).
5
Chichester: Wiley.
6
7
28
Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans. American
Psychologist, 54, 493–503. doi:10.1037/0003-066X.54.7.493.
8
Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A
9
meta-analysis of effects and processes. Advances in Experimental Social Psychology,
10
11
38, 69-119.
Graham-Clarke, P., & Oldenburg, B. (1994). The effectiveness of a general practice-based
12
physical activity intervention on patient physical activity status. Behaviour Change,
13
11, 132-144.
14
Gratton, L., Povey, R., & Clark-Carter, D. (2007). Promoting children's fruit and vegetable
15
consumption: Interventions using the theory of planned behaviour as a framework.
16
British Journal of Health Psychology, 12, 639-650. doi: 10.1348/135910706X171504.
17
Grundy, S. M., Benjamin, I. J., Burke, G. L., Chait, A., Eckel, R. H., Howard, B. V.,
18
… & Sowers, J. R. (1999). Diabetes and cardiovascular disease: A statement for
19
healthcare professionals from the American Heart Association. Circulation, 100,
20
1134-1146.
21
Gutiérrez-Doña, B., Lippke, S., Renner, B., Kwon, S., & Schwarzer, R. (2009). Self-
22
efficacy and planning predict dietary behaviors in Costa Rican and South
23
Korean women: Two moderated mediation analyses. Applied Psychology:
24
Health and Well-being, 1, 91-104.
TPB health behavior intervention
1
29
Hagger, M. S., Chatzisarantis, N. L. D., & Biddle, S. J. H. (2002). A meta-analytic review of
2
the theories of reasoned action and planned behavior in physical activity: Predictive
3
validity and the contribution of additional variables. Journal of Sport and Exercise
4
Psychology, 24, 3-32.
5
Hardeman, W., Johnston, M., Johnston, D. W., Bonetti, D., Wareham, N. J., & Kimmonth, A.
6
L. (2002). Application of the theory of planned behaviour in behaviour change
7
interventions: A systematic review. Psychology &Health, 17, 123-158. doi:
8
10.1080/08870440290013644.
9
Hausenblas, H. A., Carron, A. V., & Mack, D. E. (1997). Application of the theories of
10
reasoned action and planned behavior to exercise behavior: A meta-analysis. Journal
11
of Sport and Exercise Psychology, 19, 36-51.
12
Hillsdon, M., Foster, C., & Thorogood, M. (2005, January 24). Interventions for promoting
13
physical activity. Cochrane Database of Systematic Reviews 2005(1), Article
14
CD003180. Retrieved December 10, 2008, from The Cochrane Library Database.
15
Jancey, J., Lee, A., Howat, P., Clarke, A., Wang., & Shilton, T. (2007). Reducing attrition in
16
physical activity programs for older adults. Journal of Aging and Physical Activity, 15,
17
152-165.
18
Jemmott, J. B., Jemmott, L. S., O’Leary, A., Ngwane, Z., Icard, L., Bellamy, S., … &
19
Makiwane, M. B. (2011). Cognitive-behavioural health-promotion intervention
20
increases fruit and vegetable consumption and physical activity among South African
21
adolescents: A cluster-randomised controlled trial. Psychology & Health, 26, 167-185.
22
doi: 10.1080/08870446.2011.531573.
23
24
Jones, F., Abraham, C., Harris, P., Schulz, J., & Chrispin, C. (2001). From knowledge to
action regulation: Modeling the cognitive prerequisites of sun screen use in Australian
TPB health behavior intervention
1
and UK samples. Psychology & Health, 16, 191-206. doi:
2
10.1080/08870440108405499.
3
30
Jones, L. W., Courneya, K. S., Fairey, A. S., & Mackey, J. R. (2005). Does the theory of
4
planned behaviour mediate the effects of an Oncologist’s recommendation to exercise
5
in newly diagnosed breast cancer survivors? Results from a randomised trial. Health
6
Psychology, 24, 189-197. doi: 10.1037/0278-6133.24.2.189.
7
Kahn, E. B., Ramsey, L. T., Brownson, R. C., Heath, G. W., Howze, E. H., Powell, K. E.,
8
…the Task Force on Community Preventive Services. (2002). The effectiveness of
9
interventions to increase physical activity: A systematic review. American Journal of
10
11
Preventive Medicine, 22, S73-S107. doi:10.1016/S0749-3797(02)00434-8.
Kavookjian, J., Elswick, B. M., & Whetsel, T. (2007). Interventions for being active among
12
individuals with diabetes. The Diabetes Educator, 33, 962-988. doi:
13
10.1177/0145721707308411.
14
Kelley, K., & Abraham, C. (2004). RCT of a theory-based intervention promoting healthy
15
eating and physical activity amongst out-patients older than 65 years. Social Science
16
& Medicine, 59, 787-797. doi:10.1016/j.socscimed.2003.11.036.
17
Luszczynska, A., & Schwarzer, R. (2003). Planning and self-efficacy in the adoption and
18
maintenance of breast self-examination: A longitudinal study on self-regulatory
19
cognitions. Psychology & Health, 18, 93-108. doi: 10.1080/0887044021000019358.
20
21
22
McAuley, E., & Jacobson, L. (1991). Self-efficacy and exercise participation in sedentary
adult females. American Journal of Health Promotion, 5, 185-191.
McEachan, R. R. C., Conner, M., Taylor, N. J., & Lawton, R. J. (in press). Prospective
23
prediction of health-related behaviours with the theory of planned behaviour: A meta-
24
analysis. Health Psychology Review. doi: 10.1080/17437199.2010.521684.
TPB health behavior intervention
1
31
Müller-Riemenschneider, F., Reinhold, T., Nocon, M., & Willich, S. N. (2008). Long-term
2
effectiveness of interventions promoting physical activity: A systematic review.
3
Preventive Medicine, 47, 354-368. doi:10.1016/j.ypmed.2008.07.006.
4
Norman, P., & Conner, M. (2005). The theory of planned behavior and exercise: Evidence for
5
the mediating and moderating roles of planning on intention-behavior relations.
6
Journal of Sport and Exercise Psychology, 27, 488-505.
7
Page, R. C., Harnden, K. E., Cook, J. T., & Turner, R. C. (1992). Can lifestyles of subjects
8
with impaired glucose tolerance be changed? A feasibility study. Diabetic Medicine,
9
9, 562-566. doi: 10.1111/j.1464-5491.1992.tb01839.x.
10
Payne, N., Jones, F., & Harris, P. R. (2004). The role of perceived need within the
11
theory of planned behaviour: A comparison of exercise and healthy eating.
12
British Journal of Health Psychology, 9, 489-504.
13
Parrott, M. W., Tennant, L. K., Olejnik, S., & Pouderijne, M. S. (2008). Theory of planned
14
behavior: Implications for an email-based physical activity intervention. Psychology of
15
Sport and Exercise, 9, 511-526. doi:10.1016/j.psychsport.2007.07.002.
16
Plotnikoff, R. C., Lippke, S., Courneya, K., Birkett, N., & Sigal, R. (2010). Physical activity
17
and diabetes: An application of the theory of planned behaviour to explain physical
18
activity for Type 1 and Type 2 diabetes in an adult population sample. Psychology &
19
Health, 25, 7-23. doi: 10.1080/08870440802160984.
20
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and
21
comparing indirect effects in multiple mediator models. Behavior Research Methods,
22
40, 879-891. doi: 10.3758/BRM.40.3.879.
23
Richardson, C. R., Mehari, K. S., McIntyre, L. G., Janney, A. W., Fortlage, L. A., Sen, A., …
24
Piette, J. D. (2007). A randomized trial comparing structured and lifestyle goals in an
25
internet-mediated walking program for people with type 2 diabetes. International
TPB health behavior intervention
32
1
Journal of Behavioral Nutrition and Physical Activity, 4, 59. doi:10.1186/1479-5868-4-
2
59.
3
Sandberg, T., & Conner, M. (2009). A mere measurement effect for anticipated regret:
4
Impacts on cervical screening attendance. British Journal of Social Psychology, 48,
5
221-236. doi: 10.1348/014466608X347001.
6
Schwarzer, R. (1992). Self-efficacy in the adoption and maintenance of health behaviours:
7
Theoretical approaches and a new model. In: R. Schwarzer (ed.), Self-efficacy: Thought
8
control of action (pp. 217-243). London: Hemisphere.
9
10
11
Sheeran, P. (2002). Intention-behavior relations: A conceptual and empirical review.
European Review of Social Psychology, 12, 1-36. doi: 10.1080/14792772143000003.
Simons-Morton, D. G., Calfas, K. J., Oldenburg, B., & Burton, N. W. (1998). Effects of
12
interventions in health care settings on physical activity or cardiorespiratory fitness.
13
American Journal of Preventive Medicine, 15, 413-430.
14
Steptoe, A., Doherty, S., Rink, E., Kerry, S., Kendrick, T., & Hilton, S. (1999). Behavioural
15
counselling in general practice for the promotion of healthy behaviour among adults at
16
increased risk of coronary heart disease: Randomised trial. British Medical Journal,
17
319, 943-948.
18
Terry, D. L., & Hogg, M. A. (1996). Group norms and the attitude–behavior relationship: A
19
role for group identification. Personality and Social Psychology Bulletin, 22, 776–793.
20
doi:10.1177/0146167296228002.
21
van Osch, L., Beenackers, M., Reubsaet, A., Lechner, L., Candel, M., de Vries, H. (2009).
22
Action planning as predictor of health protective and health risk behavior: an
23
investigation of fruit and snack consumption. International Journal of Behavioral
24
Nutrition and Physical Activity, 6, 69-79. doi: 10.1186/1479-5868-6-69.
25
Vessby, B. (2000). Dietary fat and insulin action in humans. British Journal of Nutrition, 83,
TPB health behavior intervention
1
2
S91-S96. doi:10.1017/S000711450000101X.
Warburton, D. E. R., Nicol, C. W., & Bredin, S. S. D. (2006). Health benefits of physical
3
activity: The evidence. Canadian Medical Association Journal, 174, 801-809.
4
doi:10.1503/cmaj.051351.
5
Westerterp, K. R. (2009). Assessment of physical activity: a critical appraisal. European
6
Journal of Applied Physiology, 105, 823-828. doi: 10.1007/s00421-009-1000-2.
7
White, K. M., Terry, D. J., Troup, C., Rempel, L., & Norman, P. (2010). Predicting the
8
consumption of food low in saturated fats among people diagnosed with Type 2
9
diabetes and cardiovascular disease: The role of planning in the theory of planned
10
11
behaviour. Appetite, 55, 348-354. doi: 10.1016/j.appet.2010.07.011.
Wing, R. R., Vendetti, E., Jakicic, J. M., Polley, B. A., & Lang, W. (1998). Lifestyle
12
intervention in overweight individuals with a family history of diabetes. Diabetes
13
Care, 21, 350-359. doi: 10.2337/diacare.21.3.350.
14
33
TPB health behavior intervention
1
Eligible to participate
(N = 183)
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
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Randomized
(N = 183)
Control Group
(n = 53)
Intervention
(n = 130)
Baseline assessment
(n = 130/128)
Baseline assessment
(n = 53/53)
Lost to follow-up
(n = 46/47)
Lost to follow-up
(n = 21/23)
1 week (n = 33/31)
1 week (n = 17/18)
6 weeks (n = 13/16)
6 weeks (n = 4/5)
Included in analysis at
6-weeks
(n = 84/81)
Included in analysis at
6-weeks
(n = 32/30)
46
47
48
Note. Time 1 = pre-intervention. Time 2 = 1-week post-intervention. Time 3 =
6-weeks post-intervention. Ns refer to healthy eating/physical activity.
49
50
Figure 1. Flow of recruitment and participants through study.
34
TPB health behavior intervention
35
1
Table 1
2
Descriptive Data (Means and Standard Deviations) Examining Time by Condition Effects for
3
TPB Measures of Healthy Eating and Physical Activity at Baseline, 1- and 6-Week Follow-up
Healthy Eating
Behavior
Intention
Planning
Perceived behavioral control
Subjective Norm
Attitude
Intention
Planning
1-week
6-week
intervention
follow-up
follow-up
(n = 183)
(n = 133)
(n = 116)
Experimental
5.78 (1.43)
6.20 (1.20)
6.10 (1.08)
Control
5.55 (1.70)
5.50 (1.50)
5.55 (1.43)
Experimental
6.10 (0.92)
6.25 (0.99)
6.39 (0.70)
Control
5.95 (0.94)
5.85 (0.93)
5.85 (0.75)
Experimental
5.35 (1.38)
5.46 (1.30)
5.49 (1.22)
Control
5.25 (1.61)
5.36 (1.30)
5.09 (1.49)
Experimental
6.39 (0.64)
6.44 (0.70)
6.36 (0.76)
Control
5.75 (1.41)
5.95 (1.00)
5.85 (0.93)
Experimental
6.26 (0.85)
6.19 (1.21)
6.15 (0.98)
Control
6.00 (0.86)
6.00 (0.99)
5.60 (0.99)
Experimental
6.19 (1.11)
6.42 (.77)
6.05 (1.29)
Control
6.31 (1.03)
6.09 (1.03)
6.10 (1.02)
(n = 181)
(n =132)
(n = 111)
Experimental
4.27 (1.95)
4.95 (1.63)
4.60 (1.84)
Control
4.43 (1.69)
3.95 (1.69)
4.19 (1.81)
Experimental
5.87 (1.25)
5.98 (1.16)
5.84 (1.17)
Control
6.33 (.58)
5.00 (1.41)
5.43 (1.12)
Experimental
4.61 (1.62)
5.22 (1.39)
5.05 (1.43)
Control
4.92 (.96)
4.32 (1.26)
4.41 (1.64)
Physical Activity
Behavior
Pre-
TPB health behavior intervention
Perceived behavioral control
Subjective Norm
Attitude
1
36
Experimental
5.65 (1.36)
5.80 (1.24)
5.73 (1.31)
Control
6.00 (0.84)
4.90 (1.41)
5.38 (1.12)
Experimental
6.26 (0.77)
6.08 (1.04)
5.99 (0.97)
Control
5.98 (0.78)
5.19 (1.37)
5.40 (1.04)
Experimental
6.02 (1.21)
6.25 (0.92)
5.94 (1.28)
Control
5.90 (0.99)
5.77 (0.95)
5.60 (1.11)
TPB health behavior intervention
1
Table 2
2
Bivariate Correlations Between the Extended TPB Variables and Self-Report Physical
3
Activity at 1-Week Follow-up (N = 132)
1
1. Attitude
2. Subjective norm
3. Perceived behavioral control
4. Intention
5. Planning
6. Self-report behavior
*p < .05, **p < .01, ***p < .001.
4
37
2
3
4
5
6
.35***
.43***
.42***
.38***
.27**
.61***
.57***
.28**
.18*
.81***
.54***
.43***
.60***
.54***
.56***