In the definition of Caspersen et al. [26], physical activity is seen as any voluntary movement produced by the skeletal muscles that result in increased energy expenditure, and exercise is described as a subcategory of physical activity which is planned, structured, and repetitive, with the intent of improving or maintaining one or more patterns of physical fitness or function. Physical activity and exercise can be classified as behaviors, whereas fitness can be classified as performance. When physical activity is understood as a behavior, its longstanding character and cognitive, emotional, and action components are emphasized. The concept of physical activity or sports participation can be seen as voluntary and quite stable physical activity that is measured by type, frequency, intensity, and duration of exercise. It also includes motives and attitudes toward physical activity [27]. The essential element of this definition is consciousness and the attitude of the person himself or herself. For a young person, physical activity participation could be exercising in a sports club for fun or with serious competitive intent, and for an older person, it might also be sports club participation or a walk to the market place or a group-based sport such as water gymnastics for rehabilitation. Often, participation in physical activity is not only based on planned behavior and intention but is performed rather automatically and with less awareness; it could then be defined as a habit. Habit makes a contribution as the automatic component that is strengthened when physically active behavior becomes a routine and is incorporated repeatedly into everyday life [28] such as daily walks, daily home gymnastics, and commuting in a physically active way. Physical activity habits are assumed to develop during childhood through adolescence into adulthood [29]. In addition, the promotion of public health through physical activity interventions and physical education in schools is based on the belief that physical activity is habitual and tracks over time.

Tracking is one of the important concepts of life-course physical activity, while the others are transitions, and determinants and correlates of physical activity in different phases of life. Tracking is usually defined as the tendency of individuals to maintain their rank or position within a group over time [30]. It is expected that physical activity at a young age will have an effect on public health, and a high level of physical activity in youth should predict a high level of physical activity in adulthood. The tracking of inactivity is also an important question from the promotion perspective. The mechanism of tracking and different hypotheses, e.g., the carry-over value hypothesis, ability and readiness hypothesis, and self-selection hypotheses have been presented in the review of tracking studies by Telama [8]. The ability and readiness hypothesis maintains that earlier experiences of physical activities and sports and of the basic skills connected with them make it easy to maintain physical activity or start it again after a possible break, even though the type of activity practiced is different [8]. The concept of carry-over value means that activities in which people participate in adulthood should be learned as early as possible, e.g., in school physical education. The positive experiences and wide range of sports skills acquired in childhood and adolescence could be seen as the preparation for lifelong physical activity. The self-selection hypothesis posits that those individuals who have a hereditary disposition to fitness, early or late maturation in relation to age-mates, or motor performance participate more often in physical activity both at a young age and in adulthood than those who do not have the same disposition [8]. In addition to physical activity, these three models and explanations fit into physical activity motivation.

Self-determination theory (SDT) [31–34], a meta-theory for framing motivational studies, defines that intrinsic and varied extrinsic sources of motivation have an effect on cognitive and social development and on individual differences. Self-determination theory has become a salient motivational theory in recent years and has received a great deal of research attention. The theory propositions focus on how social and cultural factors facilitate or undermine people’s sense of volition and initiative, in addition to their well-being and the quality of their performance. It is argued that conditions supporting the individual’s experience of autonomy, competence, and relatedness foster the most volitional and high quality forms of motivation and engagement for activities, including enhanced performance and persistence in physical activity. Autonomy refers to the belief that one has some level of control over whether to participate in physical activity. Competence refers to the belief that a person possesses sufficient ability to cope in a specific situation, and relatedness is how connected a person feels to those around, such as parents, peers, and colleagues. The stability of perceived competence has been studied to some extent. However, hardly any tracking research exists on perceived autonomy and relatedness. Perceived physical competence is relatively stable over time [30]. However, we do not know how modifiable perceived competence is during the lifespan. When it has been formed, it may crystallize into a stable characteristic or it may remain stable because the environment remains unchanged for many people.

Longitudinal studies provide an opportunity to study prospectively or retrospectively physical activity over time and whether physical activity already in youth leads to a certain pathway with consequences for later activity and well-being. These studies are relatively scarce because they require a large financial effort and the long-term commitment of subjects, and encounter certain problems such as changing measurement techniques, a changing concept of physical activity, and confounding effects that occur in repeated measurements [35]. The concept of physical activity may have varied; those who, nowadays, know that when they walk somewhere (shopping, work), they are taking exercise might not recognize this earlier in life (e.g., 30 years previously). Socioeconomic status may be a confounding effect that occurs in repeated measurements in different ways; those with higher status continue participating more often in later phases of life.

Most commonly, physical activity is studied across two or three time points, for example, using Spearman’s rank order correlations in tracking studies; across childhood into adolescence, from adolescence into young adulthood, and during adulthood. Other methods for considering activity stability or change in activity include using baseline characteristics to predict physical activity and fitness several years later and relating current activity with retrospective recall of earlier activity levels.

Lifespan is characterized by various important, temporally arranged transitions. These include the transition from early childhood to formal schooling, the transition from childhood to puberty, the transition from school to the work, marriage, children with associated responsibilities, and retirement. In addition to these transitions are factors or life events that may influence the transitions or create new ones, for example, the occurrence of a disease, job changes, ecological and social and political events (changes in political power, war, etc.). Earlier events in the life span may condition later events and in turn influence patterns of stability (tracking) or instability (change, low tracking). Earlier events may thus predict the present levels of activity, but they do not predict the future. Changes in patterns of physical activity related to specific events in the lifespan help us to better understand the stability or instability of habitual physical activity in individuals and groups [30].

It is important to study transitions and life changes because individuals or groups may be more receptive to health-promoting interventions around the times of these transitions [44]. However, it is currently somewhat unclear which transitions are most important and what factors are associated with changes in physical activity at these times. Physical activity behavior changes over the life course, and individuals add and delete behaviors in their life, but sometimes, changes may be gradual or changes in physical activity behavior may occur during periods of life transition when individuals’ roles, relationships, and ecological contexts are altered. The transition from one school to the next, going to college or university, marriage, becoming a parent, and retirement have been found to be periods for change in physical activity behavior [45, 46]. In a review by Allender et al. [47], which consists of 19 publications, five life change categories were found: change in employment status, change in residence, change in physical status, change in relationships, and change in family structure. The review suggests that life change events affect participation in physical activity, generally decreasing participation after a change in employment or a change in residence status or having children, but having no effect during pregnancy, or even producing some increase in participation. After serious diseases, the results have varied; in two studies of women with cancer, Pinto et al. [48] found no change in exercise participation, but Satia et al. [49] reported an increase in physical activity participation over a 2-year period of follow-up. Brown et al. [50] studied young (22–27 years), middle-aged (51–56 years), and older (73–78 years) Australian women on two occasions, 3 years apart. Decreasing physical activity was associated with marriage and childbirth in young women and with declining health in older women. Increasing physical activity was associated with retirement and the death of a spouse in the middle-aged women. Stressful events such as divorce, harassment at work, and violence were also associated with changing physical activity.

Lunn et al. [39] presented some interesting recall data from a representative sample of 3,080 adults with a median age of 41 in Ireland. The study identified key transitions in the sports and exercise life course. Among the various significant associations, the importance of the period around the end of secondary schooling and the change in the relationship between the sexes was emphasized. The likelihood of making the transition from secondary schooling and hence continuing to participate in later adulthood is strongly linked to background characteristics, especially socioeconomic status. Income and educational attainment seemed to be a strong determinant of participation, especially of taking up individual activities. The findings revealed that participation in sports and exercise is higher in more recent cohorts.

Retirement allows people ample opportunity to be physically active in their leisure time. There are also social factors and particular times of life among adults and older adults in retirement, which may have an adverse effect on participation in physical activity, such as becoming disabled and the death of a spouse. Thus, also the results in different studies have varied. Slingerland et al. [6] examined the effect of retirement on changes in physical activity in the GLOBEStudy, in the Dutch 13-year prospective follow-up into retirement of employees aged 40–65 years at baseline after which they were still employed or had retired. Physical activity included work-related transportation, sports participation, and other leisure-time physical activity. They found that physical activity associated with work-related transportation greatly decreased and that the reduction was not compensated by increased sports participation or other leisure-time physical activity. Touvier et al. [50] investigated the relationships between retirement and 3-year changes in leisure-time physical activity patterns and sedentary behavior (watching television) in middle-aged 45–64-year-old French adults. Mean leisure-time physical activity increased by about 2 h/week in men and women who had retired, whereas no change was observed in employed persons. The positive change in leisure-time physical activity following retirement was mainly related to an increase in activities of moderate intensity, such as walking. Retirement did not modify the ranking of the most frequently performed leisure time physical activities, but the number of participants and the duration increased through retirement. Retirement was also associated with time spent watching TV. The study by Chung et al. [51] found that physical activity decreased with retirement from a physically demanding job but increased with retirement from a sedentary job in the Health and Retirement Study (1996–2002) conducted in the USA.

It is probable that many transitions and life events are important and individually, socially, and economically determined, and thus, there are also multiple possibilities why people become physically active. It seems that already from very a young age, physical activity and sports participation are a key factor in building a future physically active lifestyle and thus a long-lasting habit. In some studies, retirement appears to be associated with a relative reduction in physical activity [6], although leisure time physical activity may increase [6, 51, 52], while the objective opportunities for being more physically active are better after retirement. Being physically active in older age has great health benefits, and it is important that physical activity promotion is targeted toward people who are preparing for retirement and who have recently retired.

The factors that influence physical activity during the life course are many, and most studies of tracking do not include such correlates or covariates. However, programs and policies will benefit from more information on the reasons why and when people start a physically active way of life or discovering whether critical times for stopping participation exist. Why do some people maintain their physical activity over long time periods while others remain inactive? What determinants are related to changing to a more active or more inactive lifestyle? Li et al. [1] found in their review article that relatively little is known about decline in physically active lifestyle, including whether it occurs in all domains of physical activity. Of special interest from a population health and prevention perspective is identifying the characteristics of those who maintain their level of capability and function at a higher level than would be expected from their lifetime exposure, or are able to recover after adverse health events [2]. The role of physical activity may be very important. There is presumably a wide continuum of physical activity levels among people with and without health problems, and it is known that despite the overall greater risk, people with even severe health problems may be able to prevent further disability and mortality by means of physical exercise [53].

Implications for how physical activity policy is applied over the life course suggests targeting lower socioeconomic groups in early adulthood, offering special support from teachers and a broader range of activities to young women. A wide range of extracurricular physical activities should be provided by schools and more opportunities made available for self-organized activities close to residential areas to decrease social inequality. In addition, a child needs physical competence, parent support, direct helps from parents, and support from significant others, e.g., peers [7].

In adulthood, when physical activity is seen as sociocultural by nature, the need for repeated social reinforcement especially in transition periods or life change periods, such as a change in employment status, change of residence, and change in family structure, is emphasized [47]. At the individual level, being obese, a smoker, and less well-educated may increase the probability of remaining sedentary. Starting and keeping up physical activities during one’s adult years require determination. The challenge seems to be simultaneously personal and political. One of the challenges facing politicians and professionals is to understand difficulties at the individual level and the structural aspects behind people’s behavior. Information on the negative side of tracking is important in the planning of intervention. It may be a kind of chain reaction—older habits (especially inactivity) seem to remain unchanged in different stages of life. However, earlier experiences could be transformed again by later positive experiences, e.g., interventions that increase levels of physical activity in the short term could be a start to increasing long-term physical activity and also health and longevity. Intervention conditions supporting the individual’s experience of autonomy, competence, and relatedness may generate fruitful consequences [33]. There is also a need for ecological and cultural support such as the World Health Organization’s project Healthy Cities [54], which identified a wide range of aspects and determinants that exert an influence at the community level. These include neighborhoods and safe routes to work that encourage walking and cycling, providing opportunities for physical activity at the workplace (such as fitness programs, showers, and changing rooms) and creating opportunities for work colleagues to be active together during breaks and after work hours.

All in all, the ongoing improvements in the educational level, health status, and functional capacity of the elderly population probably enable a larger proportion of older people than previously to exercise. Furthermore, seeing old age as an active phase of life is becoming the predominant view in present-day society. Culture and policy in communities and society as a whole are key determinants of physical activity. The European Network for Action on Aging and Physical Activity proposed that national policy approaches should be sought to increase physical activity for older adults [55]. Effective policy approaches are enhanced access to physical activity places (e.g., groups, transportation to venues for physical activity), urban planning approaches, and infrastructure changes. In addition, life transitions such as retirement may be a new opportunity to start a more active lifestyle, especially in terms of leisure-time physical activity. For older people, regular physical activity is pictured as an important lifestyle behavior, which increases active life expectancy through its influence on chronic disease risk factors, and is associated with better mental and cognitive health [56, 57] and social integration [58]. A recent American College of Sport Medicine Position Stand provides a summary of the most important issues connected with exercise and physical activity in older adults [59]. Several evidence-based conclusions were drawn from exercise and physical activity programs on health and functional capacity in previously sedentary individuals. One of these was the conclusion that physical activity self-efficacy is an important predictor of sustained behavioral change in sedentary populations [60]. The elderly population requires special attention, and supporting the individual’s experience of autonomy, competence, and relatedness is essential to form motivation and persistence in physical activity [31, 32, 34]. The satisfaction of these basic needs through physical activity provides older people with enjoyment and intrinsic motivation. Extending and applying recent advances grounded in self-determination theory over the life course would potentially help in understanding older people’s physical activity behaviors [61].