Anxiety Sensitivity Uniquely Predicts Exercise Behaviors in Young Adults Seeking to Increase Physical Activity

Abstract

Individuals with elevated levels of anxiety sensitivity (AS) may be motivated to avoid aversive emotional or physical states, and therefore may have greater difficulty achieving healthy behavioral change. This may be particularly true for exercise, which produces many of the somatic sensations within the domain of AS concerns. Cross-sectional studies show a negative association between AS and exercise. However, little is known about how AS may prospectively affect attempts at behavior change in individuals who are motivated to increase their exercise. We recruited 145 young adults who self-identified as having a desire to increase their exercise behavior. Participants completed a web survey assessing AS and additional variables identified as important for behavior change—impulsivity, grit, perceived behavioral control, and action planning—and set a specific goal for exercising in the next week. One week later, a second survey assessed participants’ success in meeting their exercise goals. We hypothesized that individuals with higher AS would choose lower exercise goals and would complete less exercise at the second survey. AS was not significantly associated with exercise goal level, but significantly and negatively predicted exercise at Time 2 and was the only variable to offer significant prediction beyond consideration of baseline exercise levels. These results underscore the importance of considering AS in relation to health behavior intentions. This is particularly apt given the absence of prediction offered by other traditional predictors of behavior change.

Keywords

anxiety sensitivity exercise physical activity grit impulsivity

The physical health benefits of regular exercise have been well documented and include reductions in both the incidence of, and mortality from, a number of chronic health conditions, such as cancer, cardiovascular disease, stroke, and diabetes (Arena, Cahalin, Borghi-Silva, & Myers, 2015; Blair & Morris, 2009; Garrow & Summerbell, 1995; Goldberg & King, 2007; Leitzmann et al., 2007; Nocon et al., 2008; Paffenbarger, Hyde, Wing, & Hsieh, 1986). There are also substantial mental health benefits from regular exercise, with meta-analytic reviews demonstrating reliable reductions in depression and anxiety from programs of moderate-intensity exercise (Asmundson et al., 2013; Cooney, Dwan, & Mead, 2014; DeBoer et al., 2012; Stathopoulou, Powers, Berry, Smits, & Otto, 2006; Wipfli, Rethorst, & Landers, 2008). Despite these clear benefits, less than half (48%) of American adults meet the minimum recommendations for weekly physical activity (Center for Disease Control and Prevention, 2014).

According to a number of behavior change theories (e.g., Ajzen, 1991; Schwarzer, 1992), the key to increasing exercise behavior lies in changes in intentions to exercise. Yet, despite evidence that intentions do indeed have some influence on behavior (Webb & Sheeran, 2006), it is also clear that many individuals fail to follow through on exercise intentions. A recent meta-analysis of 10 studies investigating exercise intentions and subsequent exercise behavior found a schism between intention and behaviors in 46% of the sample (Rhodes & de Bruijn, 2013). In addition, of those who do adopt a regular physical activity program, more than half discontinue within 3 to 6 months (Dishman & Buckworth, 1996; Martin & Dubbert, 1985). Theoretical models such as the Theory of Planned Behavior (TPB) have attempted to account for this schism by considering attitudinal variables, such as perceived behavioral control, which is defined as the extent to which a behavior is perceived as easy or difficult (Ajzen & Madden, 1986). More recently, researchers have also included postintentional processes in predictive models—that is, processes which occur after one has set an intention to engage in a behavior. For instance, action planning refers to the postintentional process of identifying a specific plan (i.e., where, when, and how) to carry out an intended behavior (Gollwitzer, 1999). Although consideration of these factors has been a step forward in the prediction of exercise behavior (e.g., Scholz, Schüz, Ziegelmann, Lippke, & Schwarzer, 2008), there is much room for improvement. Meta-analysis has demonstrated that TPB variables such as intention and perceived behavioral control accounted for only 25% of the variance in behavior (Armitage & Conner, 2001). Similarly, though action planning has been predictive of goal attainment at the level of a moderate effect (d = .65; Gollwitzer & Sheeran, 2006), evidence has been mixed when it is evaluated specifically in the context of exercise (Rhodes & Dickau, 2013).

Given that models of exercise behavior change remain incomplete, it is also important to consider psychological characteristics that may influence an individual’s tendency to set higher exercise goals and to engage in exercise behavior. Recent findings in the health psychology literature suggest that anxiety sensitivity (AS), the fear of anxiety-related sensations, may represent one such factor predicting aversion to and avoidance of exercise. Although AS was initially conceptualized as a vulnerability factor for anxiety disorders (McNally, 2002), it has since been established more broadly as an index of distress intolerance (McHugh & Otto, 2011), which has been defined as the “capacity to experience and withstand negative psychological states” (Simons & Gaher, 2005), and as a predictor of avoidance-based coping across a range of health-related behaviors including maladaptive eating (Hearon, Quatromoni, Mascoop, & Otto, 2014), sleep difficulties (Calkins, Hearon, Capozzoli, & Otto, 2013), smoking (Johnson, Farris, Schmidt, Smits, & Zvolensky, 2013; Zvolensky et al., 2004), and substance use (Buckner et al., 2011; Lejuez et al., 2008). With regard to exercise, a number of correlational investigations have noted an inverse relationship between AS and self-reported exercise behavior (McWilliams & Asmundson, 2001; Moshier et al., 2013; Sabourin, Hilchey, Lefaivre, Watt, & Stewart, 2011; Smits & Zvolensky, 2006).

It is hypothesized that AS may influence exercise engagement by increasing distress due to the overlap between feared anxiety sensations and those typically experienced during physical exertion. That is, those who fear bodily sensations such as rapid heart rate, sweating, and chest tightness may be more likely to experience exercise as distressing, leading to subsequent avoidance of the behavior. In an experimental study conducted by Smits, Tart, Presnell, Rosenfield, and Otto (2010), the investigators noted an interaction between body mass index (BMI) and AS such that those elevated on both constructs reported higher levels of fear during a moderate-intensity exercise challenge. This finding highlights the importance of negative affect during exercise as a predictor of avoidance. For example, Williams and colleagues (2008) found that negative affect during moderate-intensity exercise among sedentary participants predicted less physical activity at 6- and 12-month assessments. This may be particularly detrimental to overweight/obese individuals as research shows they are more likely to report negative affect during exercise than those in the normal weight range (Ekkekakis & Lind, 2006). Indeed, in the first prospective study examining AS effects on exercise, Hearon and colleagues (Hearon, Quatromoni, et al., 2014) also noted an AS and BMI interaction whereby obese participants with elevated AS completed less physical activity as measured by actigraphy over a 3-day monitoring period. However, AS effects on exercise behavior may also depend on level of exertion, as elevated AS predicted less self-reported exercise across all weight groups when specifically examining vigorous-intensity activity (Moshier et al., 2013).

Because studies to date have examined AS as a stand-alone predictor of exercise behavior, the relative influence of AS in the context of other trait variables is unknown. Within health behavior research, there is growing focus on traits that may influence one’s ability to initiate and persist with goal-directed activity, such as negative urgency, lack of perseverance, and grit. Each of these characteristics could potentially influence the process of goal selection and implementation of exercise behavior. Negative urgency and lack of perseverance are considered to be personality traits, which may lead to impulsive behavior (Whiteside & Lynam, 2001). Negative urgency refers to a general tendency to engage in impulsive behaviors when experiencing negative affect, whereas lack of perseverance is defined as having difficulty persisting in projects, particularly in the presence of distracting stimuli (Whiteside & Lynam, 2001). Accordingly, individuals high in negative urgency may be more likely to cancel plans to exercise or end exercise in the face of negative emotional experiences, and lack of perseverance could cause individuals to fail to follow through on set goals for exercise when competing and distracting tasks are at hand. Grit, defined as “the tendency to sustain interest in and effort toward very long-term goals” (Duckworth, Peterson, Matthews, & Kelly, 2007), may be predictive of greater success in meeting exercise goals; it may be that individuals who are motivated to stick with a long-term goal of achieving physical fitness or health are better able to maintain a regular exercise routine.

Research has showed initial support for the influence of these characteristics on health behavior change. For instance, Reed (2014) found in a large community sample that individuals who exercised regularly reported higher levels of grit than infrequent exercisers. Although to our knowledge, studies have not previously examined the association between exercise and negative urgency and lack of perseverance, Churchill and colleagues have found across a number of studies that impulsivity predicts health behaviors such as snacking above and beyond TPB variables such as intention and planning (Churchill & Jessop, 2010; Churchill, Jessop, & Sparks, 2008). Furthermore, negative urgency and lack of perseverance have been shown to be associated with related health issues such as obesity (Mobbs, Crepin, Thiery, Golay, & Van der Linden, 2010). Therefore, in the current study, we included examination of grit, negative urgency, and lack of perseverance as traits which, in addition to AS, may be relevant to changing exercise behavior.

Despite emerging evidence that AS is a predictor of exercise avoidance, to our knowledge no studies to date have examined how AS may prospectively affect attempts at behavior change in individuals who are motivated to increase their exercise. That is, does AS predict failure to engage in exercise even among individuals who explicitly express a desire and motivation to increase their time spent exercising? Accordingly, the purpose of the current investigation was to examine the role of AS in a sample of undergraduate students who reported intention to increase their exercise over a 1-week time frame. To determine the unique impact of AS, we also examined other variables relevant to health behavior change models, including past exercise behavior, action planning, and perceived behavioral control. We also evaluated AS as a predictor of exercise behavior relative to impulsivity (negative urgency and lack of perseverance) and grit—trait variables that have been posited to influence self-regulatory behaviors (Duckworth et al., 2007; Whiteside & Lynam, 2001). We hypothesized that AS would predict both goal setting and exercise achievement, such that those with higher AS would set goals that represented a smaller increase in exercise behavior and would report lower exercise rates after the goal-attainment period even when controlling for the aforementioned health behavior change variables. We studied these issues in a young adult sample; this age has been identified as particularly important for determining adulthood exercise patterns (Barnekow-Bergkvist, Hedbert, Janlert, & Jansson, 1996).

Method

Participants

Participants were 145 young adults, mean (±SD) age of 18.8 (±1.3), who completed this study online after identifying “increase exercise” as a goal from among six target behaviors for potential change. All participants were college undergraduates who received partial fulfillment of a course research requirement. Participants were required to be at least 18 years of age and to be a student at Boston University. The sample was 81% female, and the majority of participants self-identified as Caucasian (61%) and Asian (26%) with the rest identifying as African American (4%), American Indian or Alaskan Native (1%), or Other (9%). Eleven percent of the sample identified as of Hispanic or Latino origin.

Procedures

Participants enrolled in the study through an online recruitment system and all surveys were administered online via Qualtrics. Prior to completing any study procedures, participants were presented an informed consent form explaining the voluntary nature of participation and alternate options for receiving course credit. After obtaining consent, participants were instructed to select two health behaviors they wished to change over the course of the following week from six options (i.e., increasing exercise, increasing study time, increasing sleep, decreasing recreational Internet use, decreasing alcohol use, or decreasing marijuana use). Participants then completed several self-report measures, including a demographic questionnaire (i.e., age, sex, race/ethnicity, educational attainment, height, weight), baseline amount of exercise over the past week (i.e., International Physical Activity Questionnaire [IPAQ]), and other variables (e.g., Anxiety Sensitivity Index [ASI], negative urgency, lack of perseverance, grit, perceived behavioral control, and action planning). Finally, participants were asked to set a goal of amount of moderate- and vigorous-intensity exercise to complete for the coming week.

One week later, participants were sent a link for a second survey to complete via Qualtrics. On this survey, participants reported actual amount of exercise completed over the past week via the IPAQ. Participants were then provided a debriefing form explaining study goals and were given credit. All study procedures were approved by the Boston University Institutional Review Board.

Measures

IPAQ

Baseline exercise as well as exercise at Time 2 were assessed via the IPAQ. The IPAQ is a seven-item measure of physical activity completed over the course of the previous week. On this measure, participants report the amount of time engaged in vigorous- and moderate-intensity activity as well as time spent walking and sitting over the past week (e.g., “During the last 7 days, how much time did you spend sitting on a week day?”). The IPAQ has demonstrated strong test–retest reliability and validity in multiple studies including an extensive reliability and validity survey conducted across 12 countries (Booth et al., 2003; Brown, Trost, Bauman, Mummery, & Owen, 2004). As recommended by IPAQ scoring guidelines (Guidelines for data processing and analysis of the International Physical Activity Questionnaire [IPAQ], 2005), scores were calculated separately for walking, moderate, vigorous, and total physical activity by multiplying the estimated energy expenditure for each category by the total minutes per week engaged in each activity category. This calculation yields a total score expressed in Metabolic equivalents (MET-minutes) per week. At Time 1, participants were also asked to set a goal for the amount of moderate- and vigorous-intensity exercise to complete over the following week, and this was converted into MET-minutes.

ASI

The ASI (R. A. Peterson & Reiss, 1993) is a 16-item self-report measure assessing the tendency to respond fearfully to anxiety-related symptoms. Participants rate responses on a Likert-type scale ranging from very little (0) to very much (4). The ASI total score is calculated by summing the responses to the 16 items. In addition, the ASI can be divided into three subscales representing Physical Concerns (e.g., “It scares me when I feel faint”), Mental Concerns (e.g., “When I cannot keep my mind on a task, I worry that I might be going crazy”), and Social Concerns (e.g., “It embarrasses me when my stomach growls”). Both the higher order general factor and the subscales demonstrate strong internal consistency and favorable reliability and validity (Reiss, Peterson, Gursky, & McNally, 1986; Zinbarg & Barlow, 1996; Zinbarg, Barlow, & Brown, 1997).

Perceived behavioral control

Perceived behavioral control has been defined as the extent to which exercise is perceived as being easy or difficult (Ajzen, 1991). Perceived behavioral control was assessed via three questions developed by Ajzen and Madden (1986), consistent with other studies examining the intention–behavior relationship (de Bruijn, Rhodes, & van Osch, 2012; Rhodes, de Bruijn, & Matheson, 2010). These questions assessed participants’ beliefs about their own capacity to determine whether they will meet their behavioral intentions. The questions were as follows: “For me, meeting my exercise goal in the next week will be” (with participants choosing a rating from 1 = extremely difficult to 7 = extremely easy), “How much control do you feel you have over meeting your exercise goal in the next week?” (rating from 1 = very little control to 7 = complete control), and “How much I exercise over the next week is completely up to me” (rating from 1 = strongly disagree to 7 = completely agree).

Action planning

Action planning was measured using four items assessing the extent to which participants have a plan for carrying out their specific behavioral goal. Participants were asked to rate their agreement on 7-point scale ranging from strongly disagree to strongly agree. The items were as follows: “I have made specific plans for the coming week regarding where I will exercise,” “I have made specific plans for the coming week regarding when I will exercise,” “I have made specific plans for the coming week regarding how often I will exercise,” and “I have made specific plans for the coming week regarding with whom I will exercise.” These items have been utilized across multiple studies of the TPB (Sniehotta, Schwarzer, Scholz, & Schüz, 2005; Wiedemann, Schüz, Sniehotta, Scholz, & Schwarzer, 2009).

UPPS-P Impulsive Behavior Scale

The UPPS-P (Cyders et al., 2007) is a 59-item self-report measure assessing five dimensions of personality, which may lead to impulsive behavior: negative and positive urgency, (lack of) premeditation, (lack of) perseverance, and sensation seeking. The UPPS-P is a modified version of the UPPS impulsive behavior scale developed by Whiteside and Lynam (2001). The current study utilized mean scores on the negative urgency and (lack of) perseverance subscales. Negative urgency refers to a general tendency to engage in impulsive behaviors when experiencing negative affect (e.g., “When I feel rejected, I will often say things that I later regret”; Whiteside & Lynam, 2001). Lack of perseverance is defined as having difficulty persisting in projects, particularly in the presence of distracting stimuli (e.g., “ I tend to give up easily”; Whiteside & Lynam, 2001). Participants responded on a scale from 1 (agree strongly) to 4 (disagree strongly).

Short Grit Scale

Grit, the “tendency to sustain interest in and effort toward very long-term goals” (Duckworth et al., 2007), was assessed via the Short Grit Scale (Duckworth & Quinn, 2009), which consists of eight items. Items are rated on a 5-point Likert-type scale from not like me at all to very much like me, and includes items such as “Setbacks don’t discourage me” or “I often set a goal but later choose to pursue a different one.” This scale demonstrates strong psychometric properties and predictive validity of a number of performance variables (Duckworth & Quinn, 2009) and has been assessed in several relevant populations, including students and military trainees (Eskreis-Winkler, Shulman, Beal, & Duckworth, 2014).

Data Analysis

Two primary outcomes were considered: exercise goal and MET-minutes at Time 2. Exercise goal was expressed as the rank order of the percent increase in goal exercise level over baseline level. To place findings in context, we first examined the intercorrelations between potential predictors at baseline, that is, baseline MET-minutes, ASI, perceived behavioral control, action planning, UPPS negative urgency, UPPS (lack of) perseverance, and grit. We next assessed the relationship between the predictor variables and the two outcomes using bivariate correlations. We then conducted stepwise forward regression analyses for each outcome (exercise goal and Time 2 METs) to identify which predictors were nonredundant. These two regression analyses included all variables that had demonstrated significant bivariate correlations with the outcome measure of interest. If ASI subscale scores were significantly correlated with an outcome, we conducted follow-up regression analyses that included the ASI subscale of interest rather than ASI total scores. Statistical significance was set at p < .05 for all analyses, with control of inflation of alpha from the bivariate correlations addressed in the context of the stepwise multiple regression analysis for goal selection and exercise attainment. All analyses were conducted with SPSS.

Results

Sample Characteristics

Exercise behavior, and mean scores for AS, perceived behavioral control, action planning, negative urgency, (lack of) perseverance, and grit are reported in Table 1. The sample was highly active at baseline, with 75% of participants meeting the Centers for Disease Control and Prevention (CDC) recommendations for physical activity. Participants were on average in the healthy weight range, and only 21% of participants had a BMI in the overweight or obese range (i.e., BMI ≥ 25).

Table 1.

Mean Scores on Study Measures.

Measure	M (SD)
BMI	22.51 (3.49)
Baseline METs	1,818.32 (1,910.67)
Exercise goal in METs	2,658.29 (2,104.41)
Time 2 METs	2,080.56 (1,895.23)
ASI	25.03 (12.21)
Action planning	18.96 (5.09)
Perceived behavioral control	13.43 (3.13)
UPPS Negative Urgency subscale	2.06 (0.66)
UPPS Lack of Perseverance subscale	1.81 (0.42)
Short Grit Scale	3.37 (0.64)

Note. BMI = body mass index; ASI = Anxiety Sensitivity Index; MET = metabolic equivalent.

Relations Among Variables at Baseline

Correlations between all study measures are presented in Table 2. Baseline exercise was significantly and positively associated with action planning and perceived behavioral control, but was not significantly associated with ASI, UPPS negative urgency, UPPS (lack of) perseverance, or grit. AS was positively correlated with UPPS negative urgency and negatively correlated with grit, and was marginally positively associated with UPPS (lack of) perseverance. Grit was significantly and negatively related to UPPS (lack of) perseverance and UPPS negative urgency. Perceived behavioral control and action planning were significantly correlated, but neither variable was related to the four trait variables assessed.

Table 2.

Correlations Between Study Measures.

Measure	1	2	3	4	5	6	7	8	9	10	11	12
Baseline METs	1
METs goal (ranked)	−.61**	1
Time 2 METs	.50**	−.36**	1
ASI total	−.07	.01	−.21*	1
ASI physical	−.10	.04	−.22**	.94**	1
ASI mental	−.06	.02	−.22**	.82**	.64**	1
ASI social	.07	−.10	−.00	.71**	.53**	.49**	1
Grit	.05	−.13	.08	−.28**	−.23**	−.33**	−.16	1
UPPS negative urgency	.02	.05	−.05	.40**	.34**	.47**	.22*	−.46**	1
UPPS lack of perseverance	−.11	.17*	−.08	.16	.10	.28**	.05	−.74**	.41**	1
Action planning	.25**	−.17*	.26**	−.09	−.14	−.06	.06	.07	−.09	−.11	1
Perceived behavioral control	.27**	−.15	.24**	−.06	−.08	−.07	.03	−.15	.03	.15	.50**	1

Note. ASI = Anxiety Sensitivity Index.

Correlation is significant at the .05 level (two-tailed). ** Correlation is significant at the .01 level (two-tailed).

Prediction of Goal Setting

Participants set an average goal of 2,658 (±2,104) MET-minutes of exercise for Time 2, an average increase of 803 (±1,752) METs over their baseline totals (i.e., an increase of approximately 100 min of vigorous activity or 200 min of moderate activity). In predictive models, exercise goals were examined as a ranked variable reflecting the percent increase in physical activity from baseline (expressed in terms of moderate- and vigorous-intensity METs). Zero-order correlations of goal levels indicated that ASI total and subscale scores were not associated with exercise goal levels. Exercise goal levels were significantly negatively associated with action planning and positively associated with UPPS lack of perseverance. In addition, those with higher METs at baseline selected more conservative exercise goals relative to current levels. In multiple regression analysis (see Table 3), which included all significant predictors of exercise goal (baseline METs, action planning, and UPPS lack of perseverance), only baseline METs contributed significantly to prediction of exercise goal levels (β = −.61, t = −9.08, p < .001).

Table 3.

Stepwise Regression Analyses for Variables Predicting Exercise Goal and Time 2 METs.

Variable	B	SE (B)	β	t
Dependent variable: Exercise goal
Model 1
Time 1 METs	−0.01	0.00	−.61	−9.08**
Excluded variables
Action planning				−0.29
Lack of perseverance				1.50
Dependent variable: Time 2 METs
Model 1
Time 1 METs	0.60	0.05	.68	10.96**
Model 2
Time 1 METs	0.59	0.05	.67	10.99**
ASI total score	−23.29	8.40	−.17	−2.77**
Excluded variables
Exercise goal				0.98
Action planning				0.91
Perceived behavioral control				0.15

p < .01.

Prediction of Exercise Attainment at Time 2

Bivariate correlations indicated that baseline exercise was strongly related to Time 2 exercise. ASI total score, and the ASI Physical Symptoms and Mental Concerns subscales were significantly negatively associated with Time 2 exercise. Furthermore, significant positive associations were found between Time 2 exercise and both action planning and perceived behavioral control. Negative urgency, lack of perseverance, and grit were not significantly associated with Time 2 exercise.

We next conducted a forward stepwise regression including the variables that had predicted Time 2 METs in bivariate analyses (baseline METs, ASI total, perceived behavioral control, and action planning). Results are presented in Table 3. In this analysis, we first entered baseline METs (β = .67, p < .001) and found that ASI total score offered subsequent significant prediction (β = −.17, p = .006), whereas action planning (p = .36) and perceived behavioral control (p = .89) did not.

To examine the predictive value of specific ASI subscales, the forward stepwise regression was repeated twice to include the ASI Physical Symptoms and ASI Mental Concerns subscale scores rather than ASI total score. This yielded similar results to when ASI total score was examined; both subscales offered significant prediction of exercise at Time 2 (ASI Physical Symptoms: β = −.15, p = .02; ASI Mental Concerns: β = −.19, p = .002), and action planning and perceived behavioral control did not enter into the model significantly in either regression.

Discussion

This study was designed to evaluate the importance of AS in predicting future exercise in a sample of individuals with high intentions for change. One hundred forty-five college students who expressed desire to increase their exercise behavior were asked to set a specific goal for exercise within the next week. One week later, exercise behavior was assessed to evaluate goal achievement. Consistent with the research literature documenting a large intention–behavior gap for exercise behavior, only 37% of the sample was able to meet their goal for exercise in the next week. Knowledge of baseline levels of exercise was important for understanding both goal setting and Week 2 exercise levels. Those with higher levels of exercise at baseline tended to set a lower relative increase in exercise goals and also achieved higher levels of subsequent exercise. Contrary to our hypothesis, AS was not significantly associated with exercise goal. However, AS was significantly and negatively predictive of exercise behavior 1 week later. Importantly, AS was incrementally predictive of exercise behavior above and beyond the influence of baseline exercise levels, when other variables associated with Time 2 exercise were not.

This distinctive prediction offered by AS and the lack of prediction offered by other measures related to self-regulation—including action planning, perceived behavioral control, grit, and impulsivity—speak to the potential value of considering distress intolerance in health behavior outcomes. Specifically for exercise, AS may identify those individuals who are more likely to become distressed in response to symptoms of exertion, and thereby have increased motivation to avoid exercise. That AS predicts exercise outcomes among a sample of young adults specifically identifying the desire to increase exercise is noteworthy and underscores the importance of distress intolerance in derailing goal-directed behavior. This finding for exercise is consistent with the role of AS in predicting the failure of stated goal attainment in other areas, such as session attendance for drug use treatment (Lejuez et al., 2008), reductions in Internet use (Yamada, Moshier, & Otto, 2014), and smoking cessation success (Zvolensky, Stewart, Vujanovic, Gavric, & Steeves, 2009). In each case, AS may identify individuals for whom somatic or emotional distress may be amplified, thereby hastening avoidance behavior and derailing goal persistence. Interestingly, both the Physical Concerns and Mental Concerns subscales of the ASI significantly predicted exercise behavior at Time 2, suggesting that the relationship between AS and exercise is not driven solely by distress related to physical sensations of exercise, but may reflect the role of a more general difficulty tolerating distress.

Baseline levels of exercise were related to both action planning (r = .25) and perceived behavioral control (r = .27), with effects approaching moderate effect sizes. Individuals with a specific plan for exercise behavior, and individuals who felt a stronger sense of control over their exercise behavior tended to achieve more exercise. Accordingly, these variables were also associated with the degree of increase in exercise that participants targeted in goal setting. Nonetheless, action planning and perceived behavioral control did not offer useful prediction of goal setting or future exercise behavior beyond that offered by knowledge of current exercise levels. This pattern is consistent with previous work finding that such variables offer less predictive value when considered in models that also include past behavior (Conner & Armitage, 1998).

Variables thought to be associated with self-regulatory ability were not predictive of baseline, goal target, or Time 2 exercise in our sample. For example, grit was not a useful predictor of existing exercise levels, goal setting for exercise, or subsequent exercise levels. Grit has shown incremental predictive validity of important educational outcomes (e.g., grade point average, class retention at the United States Military Academy; Duckworth et al., 2007) as well as self-report of goal attainment over time (Sheldon, Jose, Kashdan, & Jarden, 2015). It has been less studied with regard to exercise behavior. However, one study found that grit was a significant positive predictor of stages of change for exercise behavior, showing an association on the order of a small effect size (rs = .15 and .16 for moderate and vigorous exercise, respectively; Reed, Pritschet, & Cutton, 2013). In the current study, the effect size estimate between grit and exercise behavior was roughly half that reported by Reed et al. (2013). It may be that grit is more closely related to exercise behavior when considering change over longer periods of time, whereas our study examined behavior change within an acute time frame. Consistent with this idea, the stage of change measure used by Reed and colleagues asked participants to consider their attitudes over the past 6-month period, raising the possibility that grit may better reflect attributions about past rather than future exercise behavior.

In addition, lack of perseverance and negative urgency were not associated with goal setting or baseline or future exercise behavior. These UPPS trait variables (negative urgency in particular) have been shown to be associated with a range of maladaptive behavioral processes, including problematic alcohol and substance use (e.g., Jones, Chryssanthakis, & Groom, 2014; Verdejo-García, Bechara, Recknor, & Pérez-García, 2007), nonsuicidal self-injury (C. M. Peterson & Fischer 2012), eating pathology (C. M. Peterson & Fischer, 2012), and compulsive hoarding (Timpano et al., 2013). However, they have been relatively unstudied in the context of achieving healthy, goal-directed behaviors. The lack of relationship between the UPPS variables and exercise behavior found here may suggest that these traits are more strongly predictive of engagement in impulsive behaviors rather than the derailment of persistence toward a goal. That is, individuals with high levels of urgency or low levels of perseverance may have difficulty inhibiting a prepotent response, but may not have difficulty acting purposefully toward a specific goal.

This study is the first to our knowledge to examine the prospective relationship between AS and exercise behavior in individuals reporting a desire to increase their exercise behavior. A particular strength of this study is that it considers the role of AS in the context of past behavior and relative to a number of other trait and self-regulatory variables. However, our results need to be considered in light of several limitations. The sample consisted of a group of physically active college students who were predominantly female and Caucasian, which may limit generalizability of the results. It is also noteworthy that the mean ASI score in the sample was 25, reflecting a higher level than what has been previously found in samples of young adults (for instance, M = 18 on the ASI in Schmidt, Lerew, & Jackson, 1997). In the future, diagnostic assessment of psychiatric illness would help to better characterize study samples and add to the understanding of how depression, anxiety, and other mental health problems may interact with the relationships examined in the current study. In addition, it is important to note that this study focused on a brief duration, assessing exercise behavior within a single week of the participant’s initial setting of a goal. Therefore, it remains unclear whether AS is related to exercise behavior when individuals are attempting to make longer term change. In addition, exercise behavior was measured via self-report, and more recent versions of the ASI, such as the ASI-3, may have improved subscale validity (Taylor et al., 2007). Further research would benefit from examination of objective exercise data and from the longer term study of well-characterized clinical populations (e.g., individuals with psychiatric disorders, obesity, diabetes, heart disease) for whom exercise is of particular importance.

Footnotes

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr. Smits and Dr. Otto receive royalties for books on exercise for mood and anxiety disorders for Oxford University Press. The authors have no additional disclosures.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Samantha J. Moshier, MA, is a doctoral student at Boston University and is currently completing her predoctoral psychology internship at the Institute of Living in Hartford, Connecticut. She conducts research investigating psychological processes involved in health behavior change and the mechanisms and treatment of mood and anxiety disorders.

Kristin L. Szuhany, MA, is a doctoral student in the clinical psychology program at Boston University. Her research focuses on novel interventions for mood and anxiety disorders, including exercise interventions, and on the impact of distress intolerance and neuropsychological functioning on persistence and behavior change.

Bridget A. Hearon, PhD, is a postdoctoral fellow in Boston University’s Department of Psychology and Brain Sciences, where she studies health behaviors, including eating, exercise, and substance use.

Jasper A. J. Smits, PhD, is professor at the University of Texas at Austin, where he studies the treatment of anxiety and related problems (e.g., nicotine dependence, overeating, physical inactivity, and depression).

Michael W. Otto, PhD, is professor in the Department of Psychological and Brain Sciences at Boston University. His clinical and translational research includes investigations of novel treatments, brain–behavior relationships in therapeutic learning, principles underlying behavior change failures, and health behavior promotion, including investigations of addictive behaviors, medication adherence, sleep, smoking, and exercise.

References

Ajzen

(1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179-211.

Ajzen

Madden

T. J.

(1986). Prediction of goal-directed behavior: Attitudes, intentions, and perceived behavioral control. Journal of Experimental Social Psychology, 22, 453-474.

Arena

Cahalin

L. P.

Borghi-Silva

Myers

(2015). The effect of exercise training on the pulmonary arterial system in patients with pulmonary hypertension. Progress in Cardiovascular Diseases, 57, 480-488. doi:10.1016/j.pcad.2014.03.008

Armitage

C. J.

Conner

(2001). Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40, 471-499.

Asmundson

J. G.

Fetzner

M. G.

DeBoer

L. B.

Powers

M. B.

Otto

M. W.

Smits

J. A. J.

(2013). Let’s get physical: A contemporary review of the anxiolytic effects of physical exercise for anxiety and its disorders. Depression and Anxiety, 30, 362-373. doi:10.1002/da.22043

Barnekow-Bergkvist

Hedbert

Janlert

Jansson

(1996). Physical activity pattern in men and women at the ages of 16 and 34 and development of physical activity from adolescence to adulthood. Scandinavian Journal of Medicine & Science in Sports, 6, 359-370.

Blair

S. N.

Morris

J. N.

(2009). Healthy hearts—And the universal benefits of being physically active: Physical activity and health. Annals of Epidemiology, 19, 253-256. doi:10.1016/j.annepidem.2009.01.019

Booth

M. L.

Ainsworth

B. E.

Pratt

Ekelund

Yngve

Sallis

J. F.

. . .Oja

(2003). International physical activity questionnaire: 12-country reliability and validity. Medicine & Science in Sports & Exercise, 35, 1381-1395.

Brown

Trost

Bauman

Mummery

Owen

(2004). Test-retest reliability of four physical activity measures used in population surveys. Journal of Science and Medicine in Sport, 7, 205-215.

10.

Buckner

J. D.

Zvolensky

M. J.

Smits

J. A. J.

Norton

P. J.

Crosby

R. D.

Wonderlich

S. A.

Schmidt

N. B.

(2011). Anxiety sensitivity and marijuana use: An analysis from ecological momentary assessment. Depression and Anxiety, 28, 420-426. doi:10.1002/da.20816

11.

Calkins

A. W.

Hearon

B. A.

Capozzoli

M. C.

Otto

M. W.

(2013). Psychosocial predictors of sleep dysfunction: The role of anxiety sensitivity, dysfunctional beliefs and neuroticism. Behavioral Sleep Medicine, 11, 133-143. doi:10.1080/15402002.2011.643968

12.

Center for Disease Control and Prevention. (2014). Facts about physical activity. Atlanta, GA: Center for Disease Control and Prevention. Retrieved from: www.cdc.gov/physicalactivity/data/facts.html

13.

Churchill

Jessop

(2010). Spontaneous implementation intentions and impulsivity: Can impulsivity moderate the effectiveness of planning strategies? British Journal of Health Psychology, 15, 529-541.

14.

Churchill

Jessop

Sparks

(2008). Impulsive and/or planned behaviour: Can impulsivity contribute to the predictive utility of the theory of planned behaviour? British Journal of Social Psychology, 47, 631-646.

15.

Conner

Armitage

C. J.

(1998). Extending the theory of planned behavior: A review and avenues for further research. Journal of Applied Social Psychology, 28, 1429-1464.

16.

Cooney

Dwan

Mead

(2014). Exercise for depression. Journal of the American Medical Association, 311, 2432-2433. doi:10.1001/jama.2014.4930

17.

Cyders

M. A.

Smith

G. T.

Spillane

N. S.

Fischer

Annus

A. M.

Peterson

(2007). Integration of impulsivity and positive mood to predict risky behavior: Development and validation of a measure of positive urgency. Psychological Assessment, 19(1), 107-118.

18.

DeBoer

L. B.

Powers

M. B.

Utschig

A. C.

Otto

M. W.

Smits

J. A. J.

(2012). Exploring exercise as an avenue for the treatment of anxiety disorders. Expert Review of Neurotherapeutics, 12, 1011-1022. doi:10.11586/ERN.12.73

19.

de Bruijn

G. J.

Rhodes

R. E.

van Osch

(2012). Does action planning moderate the intention-habit interaction in the exercise domain? A three-way interaction analysis investigation. Journal of Behavioral Medicine, 35, 509-519.

20.

Dishman

R. K.

Buckworth

(1996). Increasing physical activity: A quantitative synthesis. Medicine & Science in Sports & Exercise, 28, 706-719.

21.

Duckworth

A. L.

Peterson

Matthews

M. D.

Kelly

D. R.

(2007). Grit: Perseverance and passion for long-term goals. Journal of Personality and Social Psychology, 92, 1087-1101.

22.

Duckworth

A. L.

Quinn

P. D.

(2009). Development and validation of the Short Grit Scale (GRIT-S). Journal of Personality Assessment, 91, 166-174.

23.

Ekkekakis

Lind

(2006). Exercise does not feel the same when you are overweight: The impact of self-selected and imposed intensity on affect and exertion. International Journal of Obesity, 30, 652-660.

24.

Eskreis-Winkler

Shulman

E. P.

Beal

S. A.

Duckworth

A. L.

(2014). The grit effect: Predicting retention in the military, the workplace, school and marriage. Frontiers in psychology, 5, 1-12.

25.

Garrow

J. S.

Summerbell

C. D.

(1995). Meta-analysis: Effect of exercise, with or without dieting, on the body composition of overweight subjects. European Journal of Clinical Nutrition, 49, 1-10.

26.

Goldberg

J. H.

King

A. C.

(2007). Physical activity and weight management across the lifespan. Annual Review of Public Health, 28, 145-170.

27.

Gollwitzer

P. M.

(1999). Implementation intentions: Strong effects of simple plans. American Psychologist, 54, 493-503. doi:10.1037/0003-066X.54.7.493

28.

Gollwitzer

P. M.

Sheeran

(2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69-119. doi:10.1016/S0065-2601(06)38002-1

29.

Hearon

B. A.

Quatromoni

P. A.

Mascoop

J. L.

Otto

M. W.

(2014).The role of anxiety sensitivity in daily physical activity and eating behavior. Eating Behaviors, 15, 255-258. doi:10.1016/j.eatbeh.2014.03.007

30.

IPAQ. (2005). Guidelines for data processing and analysis of the “International Physical Activity Questionnaire”. Retrieved from: https://sites.google.com/site/theipaq/scoring-protocol

31.

Johnson

K. A.

Farris

S. G.

Schmidt

N. B.

Smits

J. A. J.

Zvolensky

M. J.

(2013). Panic attack history and anxiety sensitivity in relation to cognitive-based smoking processes among treatment-seeking daily smokers. Nicotine & Tobacco Research, 15, 1-10. doi:10.1093/ntr/ntr332

32.

Jones

K. A.

Chryssanthakis

Groom

M. J.

(2014). Impulsivity and drinking motives predict problem behaviours relating to alcohol use in university students. Addictive Behaviors, 39, 289-296. doi:10.1016/j.addbeh.2013.10.024

33.

Leitzmann

M. F.

Park

Blair

Ballard-Barbash

Mouw

Hollenbeck

A. R.

Schatzkin

(2007). Physical activity recommendations and decreased risk of mortality. Archives of Internal Medicine, 167, 2453-2460.

34.

Lejuez

C. W.

Zvolensky

M. J.

Daughters

S. B.

Bornovalova

M. A.

Paulson

Tull

M. T.

. . .Otto

M. W.

(2008). Anxiety sensitivity: A unique predictor of dropout among inner-city heroin and crack/cocaine users in residential substance use treatment. Behavior Research and Therapy, 46, 811-818. doi:S0005-7967(08)00072-7

35.

Martin

J. E.

Dubbert

P. M.

(1985). Adherence to exercise. Exercise and Sport Sciences Reviews, 13, 137-167.

36.

McHugh

R. K.

Otto

M. W.

(2011). Domain-general and domain-specific strategies for the assessment of distress intolerance. Psychology of Addictive Behaviors, 25, 745-749. doi:10.1037/a0025094

37.

McNally

R. J.

(2002). Anxiety sensitivity and panic disorder. Biological Psychiatry, 52, 938-946. doi:S0006322302014750

38.

McWilliams

L. A.

Asmundson

G. J.

(2001). Is there a negative association between anxiety sensitivity and arousal-increasing substances and activities? Journal of Anxiety Disorders, 15, 161-170. doi:S0887-6185(01)00056-1

39.

Mobbs

Crepin

Thiery

Golay

Van der Linden

(2010). Obesity and the four facets of impulsivity. Patient Education and Counseling, 79, 372-377.

40.

Moshier

S. J.

Hearon

B. A.

Calkins

A. C.

Szuhany

K. L.

Utschig

A. C.

Smits

J. A. J.

Otto

M. W.

(2013). Clarifying the link between distress intolerance and exercise: Elevated anxiety sensitivity predicts less vigorous exercise. Cognitive Therapy and Research, 37, 476-482.

41.

Nocon

Hiemann

Müller-Riemenschneider

Thalau

Roll

Willich

S. N.

(2008). Association of physical activity with all-cause and cardiovascular mortality: A systematic review and meta-analysis. European Journal of Cardiovascular Prevention and Rehabilitation, 15, 239-246. doi:10.1097/HJR.0b013e3282f55e09

42.

Paffenbarger

R. S.

Jr. Hyde

R. T.

Wing

A. L.

Hsieh

C. C.

(1986). Physical activity, all-cause mortality, and longevity of college alumni. New England Journal of Medicine, 314, 605-613.

43.

Peterson

C. M.

Fischer

(2012). A prospective study of the influence of the UPPS model of impulsivity on the co-occurrence of bulimic symptoms and non-suicidal self-injury. Eating Behaviors, 13, 335-341. doi:10.1016/j.eatbeh.2012.05.007

44.

Peterson

R. A.

Reiss

(1993). Anxiety Sensitivity Index revised test manual. Worthington, OH: International Diagnostic Systems Publishing Corporation.

45.

Reed

(2014). A survey of grit and exercise behavior. Journal of Sport Behavior, 37(4), 390-406.

46.

Reed

Pritschet

B. L.

Cutton

D. M.

(2013). Grit, conscientiousness, and the transtheoretical model of change for exercise behavior. Journal of Health Psychology, 18, 612-619.

47.

Reiss

Peterson

R. A.

Gursky

D. M.

McNally

R. J.

(1986). Anxiety sensitivity, anxiety frequency and the prediction of fearfulness. Behaviour Research and Therapy, 24(1), 1-8.

48.

Rhodes

R. E.

de Bruijn

G. J.

(2013). How big is the physical activity intention-behaviour gap? A meta-analysis using the action control framework. British Journal of Health Psychology, 18, 296-309.

49.

Rhodes

R. E.

de Bruijn

G. J.

Matheson

D. H.

(2010). Habit in the physical activity domain: Integration with intention temporal stability and action control. Journal of Sport & Exercise Psychology, 32(1), 84-98.

50.

Rhodes

R. E.

Dickau

(2013). Moderators of the intention-behaviour relationship in the physical activity domain: A systematic review. British Journal of Sports Medicine, 47, 215-225. doi:10.1136/bjsports-2011-090411

51.

Sabourin

B. C.

Hilchey

C. A.

Lefaivre

M.-J.

Watt

M. C.

Stewart

S. H.

(2011). Why do they exercise less? Barriers to exercise in high-anxiety-sensitive women. Cognitive Behavioral Therapy, 40, 206-215. doi:10.1080/16506073.2011.573572

52.

Schmidt

N. B.

Lerew

D. R.

Jackson

R. J.

(1997). The role of anxiety sensitivity in the pathogenesis of panic: Prospective evaluation of spontaneous panic attacks during acute stress. Journal of Abnormal Psychology, 108, 532-537.

53.

Scholz

Schüz

Ziegelmann

J. P.

Lippke

Schwarzer

(2008). Beyond behavioural intentions: Planning mediates between intentions and physical activity. British Journal of Health Psychology, 13, 479-494. doi:10.1348/135910707X216062

54.

Schwarzer

(1992). Self-efficacy in the adoption and maintenance of health behaviors: Theoretical approaches and a new model. In Schwarzer

(Ed.), Self-efficacy: Thought control of action (pp. 217-243). Washington, DC: Hemisphere Publishing.

55.

Sheldon

K. M.

Jose

P. E.

Kashdan

T. B.

Jarden

(2015). Personality, effective goal- striving, and enhanced well-being: Comparing 10 candidate personality strengths. Personality and Social Psychology Bulletin, 41, 575-585.

56.

Simons

J. S.

Gaher

R. M.

(2005). The Distress Tolerance Scale: Development and validation of a self-report measure. Motivation and Emotion, 29, 83-102.

57.

Smits

J. A.

Tart

C. D.

Presnell

Rosenfield

Otto

M. W.

(2010). Identifying potential barriers to physical activity adherence: Anxiety sensitivity and body mass as predictors of fear during exercise. Cognitive Behaviour Therapy, 39(1), 28-36. doi:10.1080/16506070902915261

58.

Smits

J. A.

Zvolensky

M. J.

(2006). Emotional vulnerability as a function of physical activity among individuals with panic disorder. Depression & Anxiety, 23(2), 102-106. doi:10.1002/da.20146

59.

Sniehotta

F. F.

Schwarzer

Scholz

Schüz

(2005). Action planning and coping planning for long-term lifestyle change: Theory and assessment. European Journal of Social Psychology, 35, 565-576.

60.

Stathopoulou

Powers

M. B.

Berry

A. C.

Smits

J. A.

Otto

M. W.

(2006). Exercise interventions for mental health: A quantitative and qualitative review. Clinical Psychology: Science and Practice, 13(2), 179-193.

61.

Taylor

Zvolensky

M. J.

Cox

B. J.

Deacon

Heimberg

R. G.

Ledley

D. R.

. . .Cardenas

S. J.

(2007). Robust dimensions of anxiety sensitivity: Development and initial validation of the Anxiety Sensitivity Index–3. Psychological Assessment, 19, 176-188.

62.

Timpano

K. R.

Rasmussen

Exner

Rief

Schmidt

N. B.

Wilhelm

(2013). Hoarding and the multi-faceted construct of impulsivity: A cross-cultural investigation. Journal of Psychiatric Research, 47, 363-370. doi:10.1016/j.jpsychires.2012.10.017

63.

Verdejo-García

Bechara

Recknor

E. C.

Pérez-García

(2007). Negative emotion-driven impulsivity predicts substance dependence problems. Drug and Alcohol Dependence, 91, 213-219. doi:10.1016/j.drugalcdep.2007.05.025

64.

Webb

T. L.

Sheeran

(2006). Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychological Bulletin, 132, 249-268.

65.

Whiteside

S. P.

Lynam

D. R.

(2001). The five factor model and impulsivity: Using a structural model of personality to understand impulsivity. Personality and Individual Differences, 30, 669-689.

66.

Wiedemann

A. U.

Schüz

Sniehotta

Scholz

Schwarzer

(2009). Disentangling the relation between intentions, planning, and behaviour: A moderated mediation analysis. Psychology & Health, 24, 67-79.

67.

Williams

D. M.

Dunsiger

Ciccolo

J. T.

Lewis

B. A.

Albrecht

A. E.

Marcus

B. H.

(2008). Acute affective response to a moderate-intensity exercise stimulus predicts physical activity participation 6 and 12 months later. Psychology of Sport and Exercise, 9, 231-245. doi:10.1016/j.psychsport.2007.04.002

68.

Wipfli

B. M.

Rethorst

C. D.

Landers

D. M.

(2008). The anxiolytic effects of exercise: A meta-analysis of randomized trials and dose-response analysis. Journal of Sport & Exercise Psychology, 30, 392-410.

69.

Yamada

Moshier

S. J.

Otto

M. W.

(2014, November). Distress intolerance predicts the intention-behavior gap for the intention of reducing recreational internet use in young adults. Poster presented at the 48th annual meeting of the Association for Behavioral and Cognitive Therapies, Philadelphia, PA.

70.

Zinbarg

R. E.

Barlow

D. H.

(1996). Structure of anxiety and the anxiety disorders: A hierarchical model. Journal of Abnormal Psychology, 105, 181-193.

71.

Zinbarg

R. E.

Barlow

D. H.

Brown

T. A.

(1997). Hierarchical structure and general factor saturation of the Anxiety Sensitivity Index: Evidence and implications. Psychological Assessment, 9, 277-284.

72.

Zvolensky

M. J.

Baker

K. M.

Leen-Feldner

Bonn-Miller

M. O.

Feldner

M. T.

Brown

R. A.

(2004). Anxiety sensitivity: Association with intensity of retrospectively-rated smoking-related withdrawal symptoms and motivation to quit. Cognitive Behavior Therapy, 33, 114-125.

73.

Zvolensky

M. J.

Stewart

S. H.

Vujanovic

Gavric

Steeves

(2009). Anxiety sensitivity and anxiety and depressive symptoms in the prediction of early smoking lapse and relapse during smoking cessation treatment. Nicotine & Tobacco Research, 11, 323-331.