Regular exercise reduces the risk of developing many chronic illnesses including cardiovascular disease, diabetes, depression, osteoporosis, etc. [1]. Women are less physically active than men, and women over 50 constitute one of the most sedentary populations in the United States [1, 2]. In addition, as women age their physical activity participation decreases [3]. Physical activity, however, could benefit women in midlife in many ways. Midlife women who are physically active during menopause gain less weight and experience less stress and negative affect [4]. Unfortunately, sustaining physically active lives is not easy. While a number of interventions can help individuals successfully initiate an exercise program, most interventions have failed to show that the new lifestyle is maintained [5, 6]. To date, the most commonly used public health theories have not been adequate for producing sustainable changes [7]. Moreover, most theories used in exercise research do not address the influential role that goals play in participation despite goals being central to motivation and self-regulation processes [8, 9].

Goals Have Multiple Levels

Goals differ in level of abstraction, and are connected in a hierarchical manner [14, 21]. According to Carver and Scheier's (1990) theory of self-regulation there is a three-level hierarchy of goals (Figure 1) [19]. In this model, the focal goal represents the concrete goal intention, or what the individual is striving to achieve with their behavior - in this case exercise (e.g., decreased cholesterol, weight loss). Below the focal level is the subordinate-level goal. This is the lowest tier in the goal structure. It represents the specific action for how individuals will achieve their focal goal (e.g., walking 30 minutes 5 days/week). Above the focal level is the superordinate-level goal. This goal is more abstract and represents the reason(s) why individuals strive for their focal goal (e.g., longevity, popularity). Investigating the different levels of exercise goals within the goal hierarchy might help us better understand how individuals have been socialized to pursue exercising.

The target of this study is the superordinate-level exercise goal. Superordinate goals, which have also been referred to as "be goals" [14], connect to the greater life values and principles that individuals hold [22]. Because they reflect individuals' idealized selves, superordinate-level goals are considered to be important self-regulatory guides for behavior, which has been modeled and tested empirically [19, 21, 23]. We propose that by understanding more about how exercise is connected to the self via superordinate goals we will be able to develop improved communications and methods to make exercise participation more deeply compelling to the individual; something that might improve sustainability [24, 25]. Research using this hierarchical framework has had predictive validity in many different areas of research. One study, for example, reported that superordinate goals influenced hypertensive patients' beliefs, feelings and self-regulation decisions [19]; in another study they predicted volunteering for the Italian Army [20]. Research on branding has shown that consumers regulated their behavior and considered purchasing different brands of cars based on what they reported at different goal levels within their goal hierarchies [26].

Virtually all of the research on exercise goals has investigated the "focal-goal" level [9, 61]. Yet, superordinate-level goals are thought to contribute to a more profound and lasting motivational experience than focal-level goals [14]. Because superordinate goals reflect the principles that individuals value [14], researching these higher-level goals may illuminate how exercise fits into individual's greater life objectives and their personal goal structures [10]. This is the first study to qualitatively assess the content of midlife women's superordinate exercise goals and investigate quantitatively which superordinate goals are most valued and most predictive of greater exercise participation over time. The majority of participants reported superordinate exercise goals related to their health in some way, but less than 25% of participants mentioned goals related to enhancing quality of life.

That such a small proportion reported quality-of-life superordinate exercise goals is concerning given that participants with Quality of Life goals exercised between 15% and 34% more than those with other types of goals. In general, as individuals age, they are more interested in obtaining subjective well-being experiences from physical activity [62]. This lower prevalence of quality-of-life goals may simply reflect that women have not been socialized to consider exercise as an effective way to enhance the quality of their daily lives. In contrast, that the majority of participants listed health or healthy aging superordinate exercise goals probably represents their socialization to exercising [17], given that these goals reflect the typical way exercise has been promoted within culture.

It is easy to recognize that the dominant messaging about exercise and physical activity, for both women and men, has promoted physical activity primarily for the health and/or weight control benefits [33, 35, 36]. Furthermore, exercise is typically prescribed to patients for its medical and health value rather than as a good way to enhance mood or quality of life [38]. When physicians recommend exercise to their patients it is usually discussed within the specific context of the need to diet and lose weight [39]. This makes losing weight the purpose for exercise.

In recent years, leading organizations like the American Heart Association (AHA) have developed health communications that promote quality of life alongside the health and longevity benefits of exercise: "You'll feel better and your life depends on it." [45]. Yet, the dominant messaging in their communications still emphasize disease prevention and life expectancy. In addition, the AHA's women-specific promotions have maintained their primary focus on heart health as the reason for participating in health behaviors like exercise: "Go Red BetterU is a FREE 12-week online nutrition and fitness program that can makeover your heart" [36]. In addition, a recent 2010 American Cancer Society (ACS) campaign, "Choose You," encourages women to put their own health first in the fight against cancer [63, 64], also clearly touting disease prevention as the main reason women should adopt a health behavior like exercise.

The role of leading organizations like the ACS and the AHA is to improve the health of individuals. Yet, we suggest that a health-related organization's primary goals may be very different than, and possibly incompatible with, the specific messaging that is most engaging and persuasive to the end user. While other research has called for shifting the focus and promotion of exercise from body weight to health [65], these and other data suggest that promoting "health" as the main motivation to engage in exercise may also not be the most strategic message to facilitate optimal engagement and participation among individuals [9].

The dominant messaging about exercise seems to have created a "behavioral branding" problem. Branding is a process that purposefully aims to influence how individuals perceive, think about, and expect from a particular product, service, organization, and even a country or a person [66]. In other words, branding refers to creating an imprint of specific associations and expectations in someone's mind regarding an object or concept. Branding is a marketing concept and not one frequently discussed in the behavioral medicine, public health, and exercise literatures. Yet, the end result of branding is simply a socialization process that creates particular schemas for and expectations about something.

We suggest that the specific socialization to exercise that individuals have had through the media, health care, and society in general has explicitly branded exercise primarily as a vehicle that promotes "weight loss," "health benefits," and "disease prevention." These desired outcomes from exercise are clearly not negative! Yet, promoting exercise primarily within health care and society as a method to "improve health" or to "be thinner" might inherently foster a feeling of compliance instead of autonomy toward exercising because cultural expectations and pressures undergird these specific goals [27, 31, 67].

Many consider "health" to be an autonomous outcome to strive for and an exercise goal specifically [61, 68], as we had thought it would be before our previous research [18]. However, we now argue that exercising to achieve health benefits medicalizes exercise and reflects normative pressures for what is idealized in our culture, making exercise a moral imperative, something else that we "should" be doing [31, 69]. There is an important distinction between what values a culture fosters in its members and whether these values are congruent with human psychological needs and optimally motivate individuals [70]. Thus, while the societal branding of exercise has successfully been internalized by most, it may have inadvertently created a compliance-oriented brand of exercising.

Feeling controlled toward a specific behavior (e. g, feeling that one "should" do it), instead of feeling autonomous towards it (e.g., what is personally important and/or satisfying), leads individuals to feel pressure to "comply," things that are known to undermine goal pursuit and behavioral sustainability [71, 72]. If the societal branding of exercise results in individuals feeling a controlled or extrinsic regulation toward exercising (instead of autonomy) than we can consider this to be non-optimal for improving population-level physical activity participation [18].

Extrinsic motives, in general, are thought to lead to poorer psychological well-being compared to intrinsic ones [25]. In addition, avoidance goals, those that focused on avoiding a negative state, have even been associated with negative physical symptoms [73]. Socialization to exercising in our culture and especially within health care has emphasized the use of exercise specifically to avoid poor health and chronic illness [40]. Thus, while counterintuitive, exercising with health goals, especially those that aim to avoid a negative state such as illness, may not be quite as healthy as one would hope.

Moreover, the relatively recent campaign devised by the prestigious American College of Sports Medicine and supported by many leading international organizations (the American Medical Association, Exercise and Sports Science Australia, the President's Council on Physical Fitness and Sports, etc.) promotes and explicitly brands "exercise is medicine" [40], something that exacerbates this problematic branding of exercise. If clinicians analogize exercise to "taking a pill" or "medicine" when speaking to their patients it may further attenuate participation, given the well-documented low adherence rates to prescription medication [74, 75].

These data also suggest that what an individual espouses as important does not necessarily translate into behavior. It is logical and commonly thought that placing a high value on health will motivate individuals to practice health behaviors [76]. Moreover, other research suggests that health is highly endorsed as a reason for exercising [62, 77]. Yet, despite all three groups equally valuing their goals, participants with exercise goals related to Current Health and Healthy Aging participated in significantly less exercise over time than those who had Quality of Life goals. This discrepancy is important to explore.

One explanation for the discrepancy between what one says they value and what they do could be that when women exercise "for health," "healthy aging," or "weight loss" they do not receive quick, if any, concrete feedback that they are achieving their main goal for exercising. Research shows that individuals disengage from pursuing goals when they do not receive sufficient feedback that they are making progress [14]. Furthermore, individuals have a tendency to choose smaller, immediate rewards over larger ones that occur later in time, especially when self-control is involved [78–80]. Thus, larger delayed rewards for exercising, like staying healthy or preventing illness, may not be as motivating or provide as good of feedback as smaller, immediate rewards, like improving mood or decreasing stress [81, 82].

By shifting our paradigm from medicine to marketing, we can glean insights into what we might be missing in our traditional promotion of exercise. Increasing participation among individuals in sustainable ways might be a question of improving how we market and "sell" exercise through principles such as branding [66, 83]. Instead of promoting the end points that clinicians, business, and government care about achieving from having individuals exercise (e.g., "improved health" in service of health care savings), health communications might become more meaningful and persuasive if they were based on the exercise benefits that will be most compelling to individuals [20, 84, 85].

Reading the language participants used to describe their superordinate goals offers insight into why exercising to enhance quality of life may trump health-related motives. Quality of Life participants wrote, "Being centered," "being balanced and relaxed," "feeling good," and "happiness" as some superordinate exercise goals. Given women's constant juggling of roles and responsibilities, it is no surprise that they want their limited leisure time to represent "relaxation," "personal freedom," "lack of constraints," and "self-determination" [86, 87].

We propose that it would be strategic to rebrand exercise as a primary method to enhance aspects of daily quality of life (e.g. through social marketing, advertising, programming, and prescribing practices). Rebranding exercise with this new, in-the-moment purpose emphasizes the immediate benefits, such as stress reduction and increased vitality, and may also trigger individuals to appreciate the downstream benefits that enrich daily living (e.g., being a patient parent, enjoying life, creativity and focus at work, etc.). Striving to attain these personally meaningful and self-determined benefits might better promote well-being, engagement, and on-going participation [25, 72, 88, 89].

Exercise that specifically aims to enhance aspects of daily living might optimize the value of exercising and make it more compelling for women to fit into their busy schedules and stressful lives [90, 91]. In support of this idea, we previously reported that midlife women who exercised with focal-level goals aiming to improve the quality of their lives through reducing stress and enhancing well-being planned physical activity into their lives more frequently and reported higher participation levels over one year compared to those with focal-level health or weight-loss exercise goals [9]. Another study using a different design, sample, and methods also found that exercising for more autonomous goals predicted greater exercise participation and that this relationship was fully mediated by greater self-regulation strategies like planning [8]. These study findings suggest that exercise might most effectively compete against other daily goals and responsibilities if its primary purpose aims to enhance individuals' daily living experience in noticeable, pertinent, and significant ways [9, 92].

Our rebranding recommendation could be considered a form of "reward substitution," a strategy from the field of behavioral economics to improve adherence by switching the motive for a behavior away from distant rewards like disease prevention to immediately-experienced incentives like increased energy [82, 93, 94]. Statistical modeling shows that motivation for a behavior is partially channeled through the desire an individual feels toward their behavioral motive [12]. Thus, we need to pay much more attention to our population-level messaging about the reasons why individuals should take time out of their busy days to exercise.

One important limitation of this study is that the findings cannot be generalized outside of this specific demographic of mostly white women who worked full time in the United States. Which exercise goals will most effectively motivate participation should vary across the lifespan with changes in roles, responsibilities, and priorities [27, 77, 95]. Thus, as midlife women age and reach retirement, see aging loved ones develop chronic conditions, and have more time to spend in leisure activities, exercise to produce good health and healthy aging might become more compelling to fit into their lives.

In addition, "health" might mean different things and be differently valued in distinct cultures, especially ones that have a health care system different than the United States. Thus, exercising for health might be experienced as more intrinsic and autonomous in different countries, as others have reported [96, 97]. Studying which types of exercise goals are associated with more controlled or autonomous motivation and participation is an exciting new area of study [18, 98]. The findings from this emerging literature should help inform how, as a society, we can better promote physical activity and exercise participation to better engage individuals and make it more compelling to sustain.

It should also be noted that our method for eliciting participant's superordinate goals asked them to identify only one instead of an exhaustive list of goals. While this inductive measurement method is a study strength because individuals can have multiple goals for exercising (including both autonomous and controlled), this strategy is also a weakness because it may underestimate a more complex relationship between numerous higher order goals and participation.

Using self-reported exercise data is an important limitation of this study because it is often over-reported [99]. Given that the aim of this research was to assess relationships within the data, however, there is no reason to think that those who over-reported were not evenly distributed between the superordinate goal categories. Over-reporting should also not affect the associations between the Superordinate Exercise Goal variable and the outcomes variables.

This study focused on which types of goals and messaging may most effectively persuade individuals that physical activity creates concrete and discernable value so they will feel compelled to fit it into their busy days. Type of goal may be one of many important factors that influences women's participation. Others facilitating factors include increasing women's comfort with and skills for making their own self-care a high priority, self-regulation techniques, social support, and family-friendly facilitates [59, 90, 100, 101].

This study has many strengths. It utilized a longitudinal design over one year. It also randomly selected participants from the sample population and had excellent baseline response rate and retention of participants across the study. It was a person-centered, idiographic approach to understanding differences between individuals with similar types of goals on producing exercise behavior over time. By implementing a person-centered strategy for investigating sustained behavior the research question can go beyond making generalizations from the mean response of variables to the mean response of individuals [102]. The benefit of such a methodological strategy is that the findings have direct translation into application and improved external validity, something that has been lacking in the field of behavioral medicine [103]. Moreover, using quantitative and qualitative methods is another strength because mixed-method designs produce a more comprehensive understanding, especially of a new topic [104].

Behavior can only be understood by identifying the goals to which it is attached [14]. This research adds to the emerging literature on how superordinate goals influence behavior. Our data suggest that superordinate exercise goals related to health and healthy aging are associated with less exercise than those related to enhancing daily quality of life, despite being equally valued. Individuals have been socialized to perceive and value exercise primarily as a vehicle to promote health, prevent disease, and lose weight [18, 105]. While important, these types of benefits might not make exercise compelling enough to successfully compete against other daily responsibilities and priorities [43, 100]. Because immediate payoffs motivate behavior better than distant goals [81, 82], a more effective "hook" for promoting higher participation levels might be to rebrand exercise as a primary way individuals can enhance the quality of their daily lives [90, 106]. These findings have important implications for how we as a culture, especially those in fitness-related businesses, health promotion, health care, and public health, prescribe and market exercise on individual and population levels.