1

British Journal of Health Psychology (2014) © 2014 The British Psychological Society www.wileyonlinelibrary.com

The efficacy of a self-managed Acceptance and Commitment Therapy intervention DVD for physical activity initiation Robyn Moffitt1* and Philip Mohr2,3 1

School of Psychology, Australian Catholic University, Banyo, Queensland, Australia Animal, Food and Health Sciences, Commonwealth Scientific and Industrial Research Organisation, Adelaide, South Australia, Australia 3 School of Psychology, The University of Adelaide, South Australia, Australia 2

Objectives. Initiating and maintaining physical activity presents the individual with challenges of inconvenience, discomfort, and counteractive energy. Addressing these challenges requires an intervention that elicits motivation to engage in this activity, minimizes the direct relationship between unwanted internal experiences and inaction, and is also in itself accessible and convenient. Accordingly, this study investigated the efficacy of a self-managed Acceptance and Commitment Therapy (ACT) intervention delivered via DVD and tailored for physical activity initiation. Design and methods. Fifty-nine minimally active community participants were randomly allocated to receive a 12-week pedometer-based walking programme, or the same walking programme with the additional provision of the ACT DVD. The primary outcome was overall physical activity level (assessed at baseline and post-intervention), and the secondary outcome was pedometer-assessed step count (measured at 4-weekly intervals throughout the intervention period). Results. Participants who received the ACT DVD achieved a significantly greater increase in physical activity levels post-intervention, were more likely to achieve the goals specified in the programme, and reported a higher average step count than participants who received the walking programme in isolation. Conclusions. The ACT intervention, delivered via DVD for the promotion of physical activity, proved a simple, efficient, and accessible method to encourage positive short-term increases in an important health-promoting behaviour.

Statement of contribution What is already known?  ACT interventions can increase physical activity levels through augmenting initiatory self-regulatory control.

*Correspondence should be addressed to Robyn Moffitt, School of Psychology, Australian Catholic University, 1100 Nudgee Road, Banyo, QLD 4014, Australia (email: [email protected]). DOI:10.1111/bjhp.12098

2

Robyn Moffitt and Philip Mohr  Face-to-face delivery presents challenges of accessibility and feasibility for community implementation.  There is a need for effective interventions that maximize impact while minimizing inconvenience. What does this study add?  Supplementing a walking programme with a self-managed ACT DVD produced significant increases in physical activity.  The ACT DVD is a convenient, accessible, and potentially cost-effective approach to physical activity initiation.  ACT lends itself to implementation as a self-managed electronically delivered intervention.

Physical activity decreases the risk of cardiovascular disease, cancer, diabetes, obesity, and all-cause mortality (Bull, 2011). It provides benefits to health independently of changes in body mass and can be valuable for weight-loss maintenance (Warburton, Charlesworth, Ivey, Nettlefold, & Bredin, 2010). However, physical activity can involve inconvenience and counteractive energy, requiring considerable self-control to inhibit thoughts, emotions, and habitual behavioural response patterns that encourage cessation or avoidance (Hall & Fong, 2007; Ulmer, Stetson, & Salmon, 2010). Physical activity is an affectively motivated behaviour, and the discomfort associated with physical activity initiation can be a strong contributor to pre-mature termination (Ekkekakis, Hargreaves, & Parfitt, 2013). Initiating a behaviour that is not habitual or naturally enjoyable can deplete self-regulatory energy and encourage non-adherence, thus contributing to the modest short-term success of behavioural lifestyle interventions (Butryn, Forman, Hoffman, Shaw, & Juarascio, 2011). Increasing research attention has been directed towards the efficacy of Acceptance and Commitment Therapy (ACT) interventions for promoting health behaviours (Hayes, Luoma, Bond, Masuda, & Lillis, 2006). The approach is particularly suitable for physical activity because it encourages the identification of desired life directions, in domains including health and fitness, to motivate committed engagement in behaviours that are consistent with these values (Hayes et al., 2006). ACT teaches cognitive strategies to minimize the relationship between unwanted internal experiences (i.e., obstructive thoughts and feelings) and action (or inaction). It allows the individual to tolerate discomfort, accept uncomfortable experiences, and facilitate progress towards goals (Butryn et al., 2011; Hayes et al., 2006; Ulmer et al., 2010). ACT targets the psychological inflexibility that occurs when the desire for immediate gratification overshadows long-term goals. Patterns of responding to these immediate goals (e.g., feeling good) foster habitual engagement in behaviours inconsistent with long-term values (Hayes et al., 2006). ACT illuminates these maladaptive behavioural patterns by teaching mindfulness, which involves being present and encouraging individuals to refrain from living ‘in their heads’ (Hayes et al., 2006, p. 7). Specifically, the ACT model encompasses acceptance, cognitive defusion, being present, values, and committed action. For physical activity, acceptance involves observing uncomfortable physical sensations and metaphorically making room for these feelings inside one’s body. Defusion involves questioning whether a thought contributes towards or obstructs goal achievement (Hayes et al., 2006). Unworkable thoughts (e.g., ‘exercise is intolerable’) can be defused through labelling such thoughts as reasons that do not dictate behaviour, or by categorizing thoughts into patterns of thinking (e.g., ‘my mind is telling me the

Act and physical activity

3

exercise-is-boring story again’). Being present, or mindful, facilitates effective in-the-moment behavioural regulation more closely informed by long-term goals and values (Brown, Ryan, & Creswell, 2007). The final behaviour change processes in ACT require the identification of core values and a commitment to engage in behaviours consistent with these values. For physical activity, identifying one’s values involves shifting the emphasis away from goals (e.g., losing weight) and towards ongoing global qualities (e.g., enthusiasm or persistence) that do not cease with goal attainment but permit actions to be coordinated long-term (Hayes et al., 2006). The committed action component of ACT incorporates many techniques common to traditional behaviour therapy (BT) including shaping methods, reinforcement, and setting short- and long-term values-driven goals. Strategies include developing action plans that involve committing to weathering psychological barriers to action and identifying techniques to manage broken commitments (Hayes, Strosahl, & Wilson, 1999; Hayes et al., 2006). The ACT approach promotes resiliency through inevitable lapses, failures in behavioural regulation, and internal psychological distress. The underlying aim of the ACT approach is not to eliminate unwanted thoughts and feelings, but to learn to manage these experiences more effectively, thus creating a less overtly challenging path to behaviour change. Research supports the use of ACT to translate intention into action in the realm of healthy lifestyle behaviours. Two studies have demonstrated the short-term efficacy of ACT approaches specifically for increasing physical activity. One involved female university students wanting to increase the frequency of their visits to an athletic centre. At 5 weeks post-intervention after two 2-hr face-to-face group sessions, women who received an ACT intervention had increased their centre visits significantly more than women who received an educational intervention (Butryn et al., 2011). Another single-group study explored the preliminary effectiveness of an acceptance-based intervention for physical activity and decreased caloric, fat, and sodium intake in a small sample (N = 16) of cardiac patients. Following four 90-min group sessions, participants reported reductions in weight and BMI and increases in physical activity (Goodwin, Forman, Herbert, Butryn, & Ledley, 2012). The efficacy of ACT for self-care behaviour (Gregg, Callaghan, Hayes, & Glenn-Lawson, 2007) and weight loss (Forman, Butryn, Hoffman, & Herbert, 2009; Tapper et al., 2009) has also been demonstrated in overweight and obese individuals. However, notwithstanding the demonstrated efficacy of the ACT approach for lifestyle behaviour and physical activity, face-to-face interventions, as used in these studies, are economically demanding and present challenges for wider community delivery. Their positive effects are also confounded with personalized delivery and contact with researchers. Finally, they potentially add a layer of effort and inconvenience (i.e., attendance at workshops) to the pursuit of behaviours that already present the challenge of inconvenience, thus limiting their utility as a community intervention. These factors, coupled with the less labour-intensive nature of ACT, have encouraged the investigation of technologically delivered ACT self-help interventions (i.e., internet for Tinnitis, Hesser et al., 2012; smartphone for psychological flexibility and life satisfaction, Ly, Dahl, Carlbring, & Andersson, 2012). The purpose of the present study was to determine whether the provision of a 12-week walking programme with the addition of a self-managed ACT intervention DVD would encourage higher levels of physical activity and greater likelihood of achieving the walking goals specified in the programme than the provision of the walking programme in isolation. A walking activity was chosen on the strength of evidence for the effectiveness of pedometer-based walking interventions, and

4

Robyn Moffitt and Philip Mohr

because walking is a convenient form of physical activity (Baker et al., 2008). It was expected that there would be a greater increase in physical activity level in the ACT group than the control group and that a higher proportion of participants in the ACT group would adhere to the walking programme and meet international physical activity recommendations post-intervention. To explore the mediators of change, the psychological variables targeted by ACT (e.g., acceptance, clarity of values, and goal commitment) were also measured (Hayes et al., 2006). It was expected that increases in physical activity in the ACT group would be accompanied by improvements in these psychological variables.

Method Participants A community sample of 95 participants, recruited via media advertisements, was assessed for eligibility. Inclusion and exclusion criteria were dictated by the aim of the intervention, which was not to promote the merits of physical activity but to assist already primed individuals to translate their intentions into action. Readiness was assessed using the Stages of Change Questionnaire (SCQ; Reed, Velicer, Prochaska, Rossi, & Marcus, 1997). Participants indicated which of the five stage descriptions best reflected their current physical activity. Only individuals in the contemplation (intending to begin regular exercise within 6 months) or preparation (intending to begin regular exercise within 1 month) stage of the transtheoretical model were included (Hausenblas, Dannecker, & Downs, 2003). Participants in pre-contemplation, action, or maintenance stages were excluded. Age range was 18–65 years, as the international physical activity guidelines, on which the intervention was based, apply to adults within this age range (Baker et al., 2008). Participants were advised to consult a medical practitioner before commencing the programme and were excluded if they were participating in a structured physical activity programme, suffered from a chronic illness, or were unable to engage in walking-based activity for any health or medical reason. These data were collected via self-report, and participants were screened for eligibility prior to randomization. Only participants assessed as achieving low or moderate physical activity at baseline according to the data-processing guidelines for the 7-day International Physical Activity Questionnaire (IPAQ) were included. Of the 95 participants originally recruited, 15 did not meet the inclusion criteria and 4 declined to participate. The remaining 76 participants were randomly allocated to receive the walking programme in isolation or the walking programme plus the ACT intervention DVD. Randomization was achieved using a permuted-block restricted randomization method with a 1:1 allocation ratio and a block size of 10. Of the 37 control group participants, 5 were non-contactable and 3 withdrew pre-intervention, leaving 29 participants. Of the 39 intervention group participants, 3 did not adhere to experimental protocol, 2 were non-contactable, and 2 withdrew pre-intervention, leaving 32 participants. Two further participants in the control group withdrew from the study prior to follow-up. One of these participants withdrew due to injury; no other adverse events were reported. Figure 1 shows the flow of participants throughout the study. A power calculation was conducted a priori to inform recruitment. To detect a medium-to-large effect (f = .325) at a power of 80% in a two-group comparison of means required a sample of 76. The final sample of 59 gave the study power of 70% at f = .325, 48% to detect a medium effect (f = .25) and 86% to detect a large effect (f = .40) at p < .05 for a simple comparison of means (Cohen, 1988). These represent the minimum power

Act and physical activity

5

Figure 1. Flow of participants throughout the study.

levels available, with repeated measures and adjustment for a covariate in analyses of variance both contributing increased power (Rosenthal & Rosnow, 1991). The expected effect sizes correspond to those reported in recent ACT interventions assessing outcomes on the IPAQ. The control group (n = 27) comprised 22 women and 5 men, aged 27– 65 years (M = 43.93, SD = 10.33) with mean BMI of 31.71 (SD = 5.98) based on self-reported height and weight. The intervention group (n = 32) comprised 27 women and 5 men, aged 19–65 years (M = 43.47, SD = 12.21) with mean BMI of 32.03 (SD = 7.10). Both mean BMIs fell in the Obese Class 1 category (BMI 30.00–34.99). BMI was calculated at baseline only to ensure equivalence across experimental groups. Post-intervention weight was excluded from the study protocol to maintain the emphasis of the intervention on physical activity engagement and values rather than weight-loss goals. Measures International Physical Activity Questionnaire At baseline and 12 weeks post-intervention, participants completed the IPAQ (long form) to measure physical activity in the preceding 7 days (Hagstr€ omer, Oja, & Sj€ ostr€ om, 2005). Total IPAQ activity was computed in METs (multiples of the resting metabolic rates) and

6

Robyn Moffitt and Philip Mohr

converted to MET-minutes by multiplying the MET score of an activity by minutes performed. According to IPAQ classification guidelines, to be categorized as high in physical activity, participants had to have engaged in vigorous-intensity activity on at least 3 days and accumulated at least 1,500 MET-minutes, or have completed 7 or more days of any combination of walking, moderate-intensity, or vigorous-intensity activities achieving a minimum of 3,000 MET-minutes. Moderate physical activity required 3 or more days of vigorous activity of at least 20 min, 5 or more days of moderate-intensity activity or walking of at least 30 min, or 5 or more days of any combination of walking, moderate-intensity, or vigorous-intensity activities achieving a minimum of 600 MET-minutes. Participants low in physical activity did not meet the criteria for a categorization as moderate or high. Step count During the 12-week intervention period, participants were asked to record the duration and number of steps for every walking session undertaken. As this information was specific to the pedometer-based walking programme introduced during the intervention period, no baseline step count data were collected. Goal achievement For each 4-week period, participants were categorized as achieved goals (i.e., met programme specifications) or did not achieve goals (i.e., did not meet programme specifications) in relation to step count data. Acceptance and Action Questionnaire The 16-item Acceptance and Action Questionnaire (AAQ) measures desire for emotional and cognitive control, avoidance of negative private events, and susceptibility to cognitive barriers to action (Hayes et al., 2004). Participants rated the truth of each statement (e.g., ‘I try hard to avoid feeling depressed or anxious’) on a scale from 1 (never true) to 7 (always true). Higher scores indicate greater experiential acceptance. Valued Living Questionnaire Participants rated the importance of physical health and fitness on a scale from 1 (not at all important) to 10 (extremely important), and how consistent their actions had been with their health and fitness values during the past week on a scale from 1 (not at all consistent) to 10 (extremely consistent; Wilson, Sandoz, Kitchens, & Roberts, 2010). Physical activity goals At post-intervention, participants rated their commitment to, successful attainment of, and progress towards their exercise goals on scales from 0% to 100% (Nigg, Borrelli, Maddock, & Dishman, 2008).

Procedure Institutional ethics approval and informed consent were obtained prior to study commencement. Participants completed the IPAQ prior to attending an information

Act and physical activity

7

session for their group (intervention or control). Here, they completed the remaining baseline questionnaires (AAQ and Valued Living Questionnaire [VLQ]), provided demographic information, and were issued with a G-Sensor 2026 pedometer with 98% step accuracy and 7-day memory. The G-Sensor’s motion sensor technology provides superior accuracy and reliability to the well-validated Yamax Digiwalker (SW700 and SW200) models used in previous pedometer-based research studies (Bravata et al., 2007). Participants also received an information booklet detailing a 12-week walking programme based on a successful pedometer-based intervention (Baker et al., 2008). The booklet also contained information regarding the benefits of physical activity, recommended guidelines, and motivational tips adapted from publicly available health promotional materials. The programme consisted of graduated walking goals, including step-based guidelines regarding appropriate walking intensity, to build the inactive adult up to meet international activity recommendations by week 11 (3,000 steps in 30 min on 5 days each week) and maintain this level of activity through to the end of week 12 (Marshall et al., 2009). It differed from the Baker et al. (2008) intervention in that participants were asked to only wear their pedometer during planned continuous walks of no less than 10-min duration (Baker et al., 2008; Marshall et al., 2009). These guidelines were designed both to follow international regulations and, given the IPAQ’s insensitivity to episodes shorter than 10 min, to ensure meaningful assessment of compliance to a physical activity regime. It is acknowledged, however, that health benefits are possible following shorter, more intense, bouts of physical activity (Powell, Paluch, & Blair, 2011). Participants allocated to the ACT group also received the ACT intervention DVD, comprising an introduction and five modules based on the ACT core processes (outlined in Table 1). Each module was between 15 and 25 min, with the entire contents lasting 113 min. All participants commenced the programme and began recording walking activity approximately 1 week following the information session. During the intervening time, participants practised using their pedometer; those in the ACT group were asked to complete 1 DVD module per day independently in their home environment and to confirm having viewed the DVD before commencing the walking programme using a yes– no response format. The intervention was prepared by the primary author (a registered psychologist trained in the delivery of ACT) using a variety of materials (Harris, 2009; Hayes et al., 2006; Moffitt, Brinkworth, Noakes, & Mohr, 2012). The DVD was recorded in a studio using a professional narrator and audio and visual aids. The content was tailored for physical activity initiation and a general-public audience. Every 4 weeks, participants reported the duration and number of steps for each programme-based walking session undertaken (as recorded by pedometer). Non-responders received a follow-up phone call. At the end of the study, participants completed the IPAQ, AAQ, and VLQ a second time. They also reported their progress towards physical activity goals. The ACT DVD was posted to control group participants at the conclusion of the study.

Data analysis The primary outcome variable was IPAQ score, which was collected at baseline and post-intervention and provided an indication of between-groups physical activity changes. The secondary outcome variable was step count. These data were collected at 4-weekly intervals during the intervention period and reflected uptake of the walking

8

Robyn Moffitt and Philip Mohr

Table 1. Description of Acceptance and Commitment Therapy (ACT) DVD intervention protocol Introduction: Welcome and Program Outline Purpose and Content Theoretical rationale and goals of ACT, creative hopelessness, control agenda, illusion of control, your mind is not your friend Exercises and Techniques Two-scales metaphor Module 1: Why Should I Bother With This? Purpose and Content Uncovering health-related values, differences between values and goals, the life compass Exercises and Techniques Compass metaphor, What matters to you exercise, Value Compass worksheet, Valued Direction worksheet Module 2: What Exactly Should I Be Aiming For? Purpose and Content Goal-setting, SMART goals, emotional goals, dead person’s goals Exercises and Techniques Values-Based Goals worksheet Module 3: How Do I Keep Going When The Going Gets Tough? Purpose and Content Acceptance and willingness, clean versus dirty discomfort, physicalizing discomfort Exercises and Techniques Physicalization exercise, swamp metaphor, soap bubble metaphor Module 4: How Do I Overcome My Mental Barriers? Purpose and Content Cognitive defusion, getting to know your mind, labelling the contents of your mind Exercises and Techniques Reasons versus facts, ‘buts’ versus ‘ands’ Module 5: How Do I Stay Committed For The Long-Term? Purpose and Content Committed action, FEAR and DARE acronyms, public commitments, commitment patterns, broken commitments Exercises and Techniques Willingness and action plan worksheet, expanding balloon metaphor, Self-compassion exercise

programme across experimental groups. Analyses were by means of mixed model repeated-measures ANOVA. The independent variables were experimental condition (ACT and control) and, in the case of repeated-measures analyses, time (baseline and 12 weeks post-intervention) for the IPAQ, AAQ, and VLQ scores, and week (4, 8, and 12) for the step count scores. On the basis of post-intervention IPAQ scores, participants were categorized as low-, medium-, or high-activity according to IPAQ guidelines. In addition, step count data were used to categorize participants according to goal achievement. Chi-Square tests of independence were used to compare the proportions of participants in each group who met international physical activity recommendations and achieved the goals of the walking programme. Three independent-samples t-tests were used to assess differences in goal commitment, attainment, and satisfaction between groups. Preliminary assumption testing was conducted to check for violations prior to all analyses. Because of the small number of males (n = 10), all analyses conducted were subsequently repeated with males omitted from the sample; no changes were observed except for two instances noted below where omission of males resulted in marginal effects reaching significance.

Results Descriptive statistics There were no baseline differences across experimental conditions for age (control: M = 43.93, SD = 10.33; ACT: M = 43.47, SD = 12.21), BMI (control: M = 31.71,

Act and physical activity

9

SD = 5.98; ACT: M = 32.03, SD = 7.10), or IPAQ, AAQ, and VLQ scores (reported in Table 2), suggesting that randomization was successful. Over half the participants in both experimental groups reported low physical activity levels at baseline using IPAQ classification guidelines (control: Low 51.9%, moderate 48.1%; ACT: Low 53.1%, moderate 46.9%).

Physical activity (IPAQ) As shown in Table 2, there was a large significant main effect of time on IPAQ data, with activity levels increasing from baseline to post-intervention across both experimental groups. A moderate significant condition-by-time interaction indicated that the increase was greater in the ACT group. Parallel results were obtained using an intention-to-treat (ITT) analysis (substituting baseline IPAQ scores for 13 participants with missing data post-intervention). The main effect for time remained, but the moderate condition-by-time interaction was no longer significant (p = .05). Post-intervention activity classifications were significantly lower in the control group (low 28.6%, moderate 66.7%, high 4.8%) than in the ACT group (low 8%, moderate 52%, high 40%), v² (2, n = 46) = 9.12, p = .01, Cramer’s V = .45. The effect size was moderate to large. Overall, a higher percentage of participants met recommended physical activity guidelines at post-intervention in the ACT (92%) than the control group (71.5%); the odds of a participant in the ACT group meeting activity guidelines at post-intervention were 4.6 times those of the control group. Moderate significant group differences remained using ITT analysis where baseline activity classifications were substituted for missing data at post-intervention (ACT: Low 21.9%, moderate 46.9%, high 31.2%; control: Low 29.6%, moderate 66.7%, high 3.7%), v² (2, n = 59) = 7.33, p = .026, Cramer’s V = .35.

Step count Because the programme-specific nature of the step count data precluded their collection at baseline, baseline IPAQ score was entered as a covariate to control for baseline physical activity. There was a moderate significant main effect for condition, F(1, 42) = 5.70, p = .022, partial g2 = .12, with intervention group participants reporting a higher average step count across the 12-week period than controls. The condition-by-week interaction (Table 2) was also significant and large. Figure 2 reflects the divergent trends in step count scores between weeks 8 and 12. Contrasts on the repeated-measures variable indicated a small to moderate non-significant difference in step count changes between groups from week 4 to 8 (p = .32, d = .31), but a large and significant increase in step count for participants in the ACT group relative to the control group from week 8 to 12 (p = .007, d = .87). An ITT analysis was conducted with last-reported step data carried forward for six participants with missing data at weeks 8 or 12. The moderate significant main effect for condition was retained, F(1, 47) = 5.99, p = .018, partial g2 = .11, and the interaction was not significant, p = .06.1

The condition-by-time interaction for IPAQ scores (p = .038) and the condition-by-week interaction for step count scores (p = .027) both reached significance when males were removed from the ITT analyses.

1

.20

.15

.03

.01

.07

.12

.09

11.05**

10.23**

.73

.20

2.92

5.50*

3.96

g2p

2.35

1.00

2.35

2.99

3.81*

4.02

4.34*

F

.06

.03

.06

.11

.16

.07

.09

g2p

Interaction (cond 9 time)

3.20 (1.54) 3.68 (2.06)

8.10 (1.62) 7.95 (2.27)

68.25 (12.19) 71.91 (12.56)

– –

– –

466.13 (419.68) 540.09 (608.82)

507.88 (454.91) 658.18 (634.38)

Baseline

M (SD)

– –

– –

– –

23.16 (14.28) 35.48 (21.47)

23.11 (14.28) 34.17 (22.34)

– –

– –

Week 4

– –

– –

– –

27.41 (25.63) 34.04 (24.44)

27.74 (26.44) 34.50 (25.73)

– –

– –

Week 8

3.40 (2.28) 5.23 (2.52)

6.80 (2.53) 7.43 (2.13)

67.85 (12.87) 64.55 (19.96)

20.04 (17.23) 42.68 (33.28)

19.70 (17.45) 46.22 (34.50)

776.72 (731.63) 1892.72 (2840.93)

907.21 (775.64) 2397.84 (3035.22)

Week 12

Notes. IPAQ, International Physical Activity Questionnaire; AAQ, Acceptance and Action Questionnaire; VLQ (Imp), Valued Living Questionnaire Importance; VLQ (Cons), Valued Living Questionnaire Consistency; ITT, Intention-to-Treat analysis; Step count data are reported in thousands. *p < .05; **p < .01.

IPAQ Control ACT IPAQ (ITT) Control ACT Step count Control ACT Step count (ITT) Control ACT AAQ Control ACT VLQ (Imp) Control ACT VLQ (Cons) Control ACT

F

Main effect (time)

Table 2. Physical activity and psychological outcomes across time and experimental condition

10 Robyn Moffitt and Philip Mohr

Act and physical activity

11

Figure 2. Mean step count at 4-weekly intervals during the 12-week walking programme across experimental conditions. Error bars depict the standard error of the mean.

Goal achievement The respective proportions of ACT and control participants to achieve the specified walking goals were 34.6% and 8.3% during weeks 1–4, v2(1, n = 50) = 3.61, p = .057, φ = .32, 17.4% and 21.7% during weeks 5–8, v2(1, n = 46) = 0.00, p = 1.00, φ = .06, and 30.4% and 4.3% during weeks 9–12, v2(1, n = 46) = 3.78, p = .052, φ = .34. All differences were non-significant, although effects for weeks 1–4 and 9–12 were moderate in size.

Psychological outcomes As shown in Table 2, there were no significant effects on AAQ scores for time or the condition-by-time interaction. The main effect for condition was also not significant, F < 1. A main effect for time on rated importance of health and physical well-being revealed a decline from baseline to post-intervention in both groups. The main effect for condition, F < 1, and the condition-by-time interaction were not significant. Participants were not significantly more likely to report living consistently with their health and physical well-being values at post-intervention than at baseline (p = .054). Although mean scores were higher in the ACT group, three independent-samples t-tests revealed no significant difference between groups for post-intervention levels of self-reported goal commitment (ACT: M = 6.32, SD = 2.55; control: M = 5.20, SD = 2.82; t(40) = 1.35, p = .19, d = .43), goal attainment (ACT: M = 4.64, SD = 2.85; control: M = 3.50, SD = 3.22; t(40) = 1.21, p = .23, d = .39), or goal satisfaction (ACT: M = 4.18, SD = 3.50; control: M = 3.45, SD = 3.09; t(40) = 0.72, p = .48, d = .23).

Discussion This study investigated the efficacy of a self-managed ACT intervention for initiating a physical activity programme. IPAQ data indicate that ACT group participants were more physically active at post-intervention than controls. This difference was most evident in the proportions achieving high levels of activity: 40% of ACT and 4.8% of control participants. As expected, participants in the ACT group also reported a higher step count on planned walks prescribed in the walking programme. The between-groups difference in step count

12

Robyn Moffitt and Philip Mohr

was large and the most pronounced from week 8 to week 12, suggesting that the benefits of the intervention were retained over the intervention period. According to step count data, the ACT group were also more likely to meet the walking specifications of the programme, particularly during the initial and final phases of data collection. At the final data collection point, adherent participants had reached a physical activity level consistent with current public health recommendations (3,000 steps in 30 min on 5 days each week); 30.4% of participants in the ACT group had successfully achieved these goals compared with 4.3% of participants in the control group. These findings are encouraging for the efficacy of the intervention for helping individuals achieve a level of physical activity of sufficient magnitude to produce health benefit if maintained long-term. Moreover, as baseline BMI for both groups fell in the Obese Class 1 category, the findings supplement previous research demonstrating the value of ACT for lifestyle change with overweight or obese populations (Forman et al., 2009; Gregg et al., 2007; Tapper et al., 2009). The physical activity outcomes observed indicate that ACT, with its encouragement of committed engagement in health behaviours in the face of physical and mental deterrents, was effective for encouraging initiation of, and adherence to, a walking-based programme (Butryn et al., 2011). Much empirical research investigating interventions for health-related behavioural activation has emphasized the efficacy of cognitive and emotional control strategies to modify or eliminate these problematic experiences (Hall & Fong, 2007). However, suppressing unwanted emotional responses (e.g., dread) or challenging de-motivating cognitions (e.g., ‘this is too hard’) may only further deplete self-regulatory energy and encourage avoidance and cessation (Butryn et al., 2011). There were no significant differences between groups in goal commitment, goal attainment, and satisfaction of progress towards goals, although effect sizes in the predicted direction were moderate. However, a consideration regarding the underlying mechanisms of change is the lack of relative improvement observed for the ACT group on the remaining psychological variables (AAQ and VLQ). This may be due to insensitivity of the measures. A measure of acceptance designed specifically for use in the physical activity context, such as the Physical Activity Acceptance and Action Questionnaire (PAAAQ), may have been more appropriate (Butryn et al., 2011; Goodwin et al., 2012). Although the results reported here are consistent with those observed in other ACT interventions for physical activity (Butryn et al., 2011; Goodwin et al., 2012), the findings are of particular interest for two main reasons. First, both experimental groups received an intensive pedometer-based walking programme that included a substantial self-monitoring component known to increase activity levels (Foreyt & Goodrick, 1991; Perri, 1998). Indeed, physical activity increases were observed in both experimental groups. However, although some participants in the control group obtained gains consistent with public health recommendations, the behavioural changes observed in the ACT group were moderate to large and exceeded those accompanying provision of the walking programme alone. In addition, the retained effect at the 12-week time point suggests that the impact of the DVD for intervention group participants may have been cumulative. This finding is encouraging and indicates that the ACT approach may be efficacious for long-term commitment to a physical activity regime. Of further relevance is the efficacy of the intervention for specifically promoting walking-based physical activity; walking is an accessible, safe, and cost-effective form of physical activity that is conducive to adoption at the community level (Baker et al., 2008). The second point of significance is that the current findings were obtained using an ACT intervention that was delivered via DVD. This intervention was brief and entirely self-managed and thus demonstrates the accessibility of ACT at a wider community level

Act and physical activity

13

without the requirement for intensive therapist contact. Although experimental protocol required that participants in the ACT group viewed the DVD once, the DVD format enables repeated viewing and is available as an ongoing resource for individuals to revisit, as required, during periods of low motivation. Moreover, it is one of a number of available electronic media (e.g., smartphones) that may lend themselves to the delivery of such a programme, some of which offer advantages in flexibility, accessibility, and convenience over the DVD format. The physical activity benefits demonstrated in the ACT group thus have applied significance in the context of community lifestyle intervention. A strength of this study was the use of objective accelerometer technology to supplement the self-reported step count measurement. This is diminished somewhat by the fact that no baseline assessment of step count was obtained (i.e., using sealed accelerometer technology). However, this method is more suited to the measurement of incidental walking than to situations where the aim of the intervention is to encourage adherence to a structured programme with defined minimum requirements for intensity and duration. A related limitation was the reliance on self-report data for the remaining outcome measures. This is a common characteristic of the evaluation of behaviour change interventions that is not confined to circumstances, such as the present, where the integrity of the intervention demands minimal experimenter involvement. Other limitations were a lack of power to detect effects of moderate size, evident in non-significant changes on some of the psychological variables and a number of effects using IPAQ scores that lost statistical significance when ITT provisions were applied. The small sample size for men (n = 10), while characteristic of many lifestyle intervention studies also limits the generalizability of the findings to males. Importantly, the inclusion of males in the analyses attenuated some of the findings for females. Although a recent meta-analysis found that gender did not moderate ACT treatment outcomes for depression and quality of life (Ruiz, 2012), future research could explore whether ACT may be differentially relevant as a treatment approach across gender. Finally, given the null results on the AAQ, it is conceivable that the physical activity findings reflect a non-specific treatment effect. However, this seems unlikely given the sustained nature of the ACT group’s advantage over the control group. Future research is warranted to ascertain the longer-term effectiveness of such a brief and self-managed intervention. Furthermore, it would be useful to determine which aspects of the ACT theoretical model are most able to produce physical activity behaviour change using sensitive and well-validated measures. In addition, further research could explore the efficacy of ACT for high intensity, strength-based, or anaerobic forms of physical activity as these activities may introduce greater challenges for adherence than walking (Ekkekakis et al., 2013). Although individuals in the pre-contemplative stage of change according to the transtheoretical model were excluded here, future research should explore whether ACT is equally efficacious for non-motivated individuals. The theoretical premise of the ACT model suggests that it may be appropriate for use with this population.

Conclusions Research investigating psychological interventions targeting modifiable lifestyle behaviours, such as physical inactivity, is imperative (Goodwin et al., 2012). The present findings, using an ACT intervention DVD, offer encouragement for proponents of low-intensity, brief, and self-managed interventions employing electronic media. Given the potential efficacy of an intervention such as this for chronic disease prevention and

14

Robyn Moffitt and Philip Mohr

management and reduction of the high morbidity and mortality rates associated with obesity and physical inactivity, future research should investigate the efficacy of a self-administered ACT-based intervention beyond the 12-week period across a range of target populations.

References Baker, G., Gray, S. R., Wright, A., Fitzsimons, C., Nimmo, M., Lowry, R., & Mutrie, N. (2008). The effect of a pedometer-based community walking intervention ‘Walking for Wellbeing in the West’ on physical activity levels and health outcomes: A 12-week randomised controlled trial. International Journal of Behavioral Nutrition and Physical Activity, 5, 44. doi:10.1186/ 1479-5868-5-44 Bravata, D. M., Smith-Spangler, C., Sundaram, V., Gienger, A. L., Lin, N., Lewis, R., . . . Sirard, J. R. (2007). Using pedometers to increase physical activity and improve health: A systematic review. Journal of the American Medical Association, 208, 2296–3204. doi:10.1001/jama.298.19.2296 Brown, K. K., Ryan, R. M., & Creswell, J. D. (2007). Mindfulness: Theoretical foundations and evidence for its salutary effects. Psychological Inquiry, 18, 211–237. doi:10.1080/ 10478400701598298 Bull, F. C. (2011). Global advocacy for physical activity – Development and progress of the Toronto charter for physical activity: A global call for action. Research in Exercise Epidemiology, 13, 1–10. Butryn, M. L., Forman, E., Hoffman, K., Shaw, J., & Juarascio, A. (2011). A pilot study of Acceptance and Commitment Therapy for promotion of physical activity. Journal of Physical Activity and Health, 8, 516–522. Cohen, J. (1988). Statistical power analysis for the behavioural sciences. Hillsdale, NJ: Lawrence Erlbaum Associates. Ekkekakis, P., Hargreaves, E. A., & Parfitt, G. (2013). Envisioning the next fifty years of research on the exercise-affect relationship. Psychology of Sport and Exercise, 14, 751–758. doi:10.1016/j. psychsport.2013.04.007 Foreyt, J. P., & Goodrick, G. K. (1991). Factors common to the successful therapy for the obese patient. Medicine and Science in Sport and Exercise, 23, 292–297. doi:10.1249/ 00005768-19910300-00005 Forman, E. M., Butryn, M. L., Hoffman, K. L., & Herbert, J. D. (2009). An open trial of an acceptance-based behavioural intervention for weight loss. Cognitive and Behavioral Practice, 16, 223–235. doi:10.1016/j.cbprq.2008.09.005 Goodwin, C. L., Forman, E. M., Herbert, J. D., Butryn, M. L., & Ledley, G. S. (2012). A pilot study examining the initial effectiveness of a brief acceptance-based behaviour therapy for modifying diet and physical activity among cardiac patients. Behavior Modification, 36, 199–217. doi:10. 1177/0145445511427770 Gregg, J. A., Callaghan, G. M., Hayes, S. C., & Glenn-Lawson, J. L. (2007). Improving diabetes self-management through acceptance, mindfulness, and values: A randomised controlled trial. Journal of Consulting and Clinical Psychology, 75, 336–343. doi:10.1037/0022-006X.75.2.336 Hagstr€ omer, M., Oja, P., & Sj€ ostr€ om, M. (2005). The International Physical Activity Questionnaire (IPAQ): A study of concurrent and construct validity. Public Health Nutrition, 9, 755–762. doi:10.1079/PHN2005898 Hall, P. A., & Fong, G. T. (2007). Temporal self-regulation theory: A model for individual health behaviour. Health Psychology Review, 1, 6–52. doi:10.1080/17437190701492437 Harris, R. (2009). ACT made simple: An easy-to-read primer on Acceptance and Commitment Therapy. Oakland, CA: New Harbinger Publications. Hausenblas, H. A., Dannecker, E. A., & Downs, D. S. (2003). Examination of the validity of a stage of exercise change algorithm. Journal of Applied Social Psychology, 33, 1179–1189. doi:10.1111/ j.1559-1816.2003.tb01944.x

Act and physical activity

15

Hayes, S. C., Strosahl, K., & Wilson, K. G. (1999). Acceptance and Commitment Therapy: An experiential approach to behavior change. New York: The Guilford Press. Hayes, S. C., Luoma, J. B., Bond, F. W., Masuda, A., & Lillis, J. (2006). Acceptance and Commitment Therapy: Model, processes and outcomes. Behaviour Research and Therapy, 44, 1025. doi:10. 1016/j.brat.2005.06.006 Hayes, S. C., Strosahl, K. D., Wilson, K. G., Bissett, R. T., Pistorello, J., Toarmino, D., . . . McCurry, S. M. (2004). Measuring experiential avoidance: A preliminary test of a working model. The Psychological Record, 54, 553–578. Hesser, H., Gustafsson, T., Lunden, C., Henrikson, O., Fattahi, K., Johnsson, E., . . . Kaldo, V. (2012). A randomised controlled trial of internet-delivered Cognitive Behaviour Therapy and Acceptance and Commitment Therapy in the treatment of Tinnitus. Journal of Consulting and Clinical Psychology, 80, 649–661. doi:10.1037/a0027021 Ly, K. H., Dahl, J., Carlbring, P., & Andersson, G. (2012). Development and initial evaluation of a smartphone application based on Acceptance and Commitment Therapy. SpringerPlus, 1, 11. Marshall, S. J., Levy, S. S., Tudor-Locke, C. E., Kolkhorst, F. W., Wooten, K. M., Ming Ji, M. A., . . . Ainsworth, B. E. (2009). Translating physical activity recommendations into a pedometer-based step goal: 3000 steps in 30 minutes. American Journal of Preventive Medicine, 36, 410–415. doi:10.1016/j.amepre.2009.01.021 Moffitt, R., Brinkworth, G., Noakes, M., & Mohr, P. (2012). A comparison of cognitive restructuring and cognitive defusion as strategies for resisting a craved food. Psychology and Health, 27, 74– 90. doi:10.1080/08870446.2012.694436 Nigg, C. R., Borrelli, B., Maddock, J., & Dishman, R. K. (2008). A theory of physical activity maintenance. Applied Psychology: An International Review, 57, 544–560. doi:10.1111/j. 1464-0597.2008.00343.x Perri, M. G. (1998). The maintenance of treatment effects in the long-term management of obesity. Clinical Psychology: Science and Practice, 5, 526–543. doi:10.1111/j.1468-2850.1998. tb00172.x Powell, K. E., Paluch, A. E., & Blair, S. N. (2011). Physical activity for health: What kind? How much? How intense? On top of what? Annual Review of Public Health, 32, 349–365. doi:10.1146/ annurev-publhealth-031210-101151 Reed, G. R., Velicer, W. F., Prochaska, J. O., Rossi, J. S., & Marcus, B. H. (1997). What makes a good algorithm: Examples from regular exercise. American Journal of Health Promotion, 12, 57–66. doi:10.4278/0890-1171-12.1.57 Rosenthal, R., & Rosnow, R. L. (1991). Essentials of behavioural research: Methods and data analysis (2nd ed.). Singapore: McGraw-Hill. Ruiz, F. J. (2012). Acceptance and Commitment Therapy versus Traditional Cognitive Behavioural Therapy: A systematic review and meta-analysis of current empirical evidence. International Journal of Psychology & Psychological Therapy, 12, 333–357. Tapper, K., Shaw, C., Ilsley, J., Hill, A. J., Bond, F. W., & Moore, L. (2009). Exploratory randomised controlled trial of a mindfulness-based weight loss intervention for women. Appetite, 52, 396– 404. doi:10.1016/j.appet.2008.11.012 Ulmer, C. S., Stetson, B. A., & Salmon, P. G. (2010). Mindfulness and acceptance are associated with exercise maintenance in YMCA exercisers. Behaviour Research and Therapy, 48, 805–809. doi:10.1016/j.brat.2010.04.009 Warburton, D. E. R., Charlesworth, S., Ivey, A., Nettlefold, L., & Bredin, S. S. D. (2010). A systematic review of the evidence for Canada’s physical activity guidelines for adults. International Journal of Behavioral Nutrition and Physical Activity, 7, 39. doi:10.1186/1479-5868-7-39 Wilson, K. G., Sandoz, E. K., Kitchens, J., & Roberts, M. (2010). The Valued Living Questionnaire: Defining and measuring valued action within a behavioural framework. The Psychological Record, 60, 249–272. Received 22 May 2013; revised version received 18 February 2014