The field of translational exercise science faces the challenge of bridging mechanistic research and evidence-based medicine (Berg et al., 2025). However, the field wrestles with the inconsistent use and definition of key terms, such as ‘physical activity’, ‘exercise’ or ‘intensity’, creating barriers to translation from mechanistic insight to clinical and public health application. This challenge highlights the importance of ongoing efforts to standardize definitions and terms in this field (Merrell et al., 2024). Forty years ago, Carl J. Caspersen and colleagues from the Centers for Disease Control and Prevention provided definitions for ‘physical activity’, ‘exercise’ and ‘physical fitness’ (Caspersen et al., 1985), which provided an interpretative framework that empowered research to relate these concepts to health. This work has had an enduring influence and accumulated >5000 citations (Web of Science; 1 December 2025). More recently, attempts have been made to standardize definitions of exercise intensity (Bishop et al., 2025; Norton et al., 2010) and (re)define terms such as durability, fatigability, repeatability and physiological resilience (Eldadah, 2010; Jones, 2024; Meixner et al., 2025). Such efforts are crucial for the aim of providing clinically relevant and evidence-based exercise recommendations and prescriptions. With the continued need to revisit and reconsider key definitions, and in acknowledgement of the legacy of their seminal work, here we discuss the definitions of Caspersen et al. (1985) and offer complementary perspective for related terms that are frequently misunderstood or used interchangeably, such as ‘physical inactivity’ and ‘sedentary behaviour’ (Box 1). Acute exercise bout: A single session of exercise, at a given duration and intensity, and that may include resistance or endurance exercise. Balance: The maintenance of equilibrium while stationary or moving (Caspersen et al., 1985). Body composition: The relative amounts of muscle, fat, bone and other vital parts of the body (Caspersen et al., 1985). Cardiorespiratory endurance: Synonymous with terms such as cardiorespiratory fitness or aerobic capacity, cardiorespiratory endurance is defined by the body's maximal rate of oxygen transport and utilization. Chronic exercise: Structured and repetitive acute exercise bouts, performed a specified number of times and at a defined frequency, leading to health benefits associated with physical fitness. Exercise: A subset of physical activity that is planned and structured to achieve a final or intermediate objective of gaining health benefits associated with improving or maintaining physical fitness. Exercise training: A subset of chronic exercise that is planned for the purpose of improving a specific outcome and that results in health-related improvements in physical fitness. Flexibility: The range of motion available at a joint (Caspersen et al., 1985). Muscle strength: The amount of external force that a muscle can exert (Caspersen et al., 1985). Muscular endurance: The ability of the muscle groups to exert external force for many repetitions or successive exertions in dynamic or static conditions that are not limited by the cardiovascular or pulmonary systems. Physical activity: A quantifiable term referring to any bodily movement produced by the voluntary contraction of skeletal muscle that increases energy expenditure. This can encompass exercise in addition to occupational, domestic and all other activities. Physical fitness: The anatomical and physiological qualities that provide an individual with the ability to meet the demands of a particular task in a steady state. Physical inactivity: Physical activity levels below those required for optimal health and the prevention of premature death. Power: The maximum rate at which one can perform work for no longer than a few seconds. Sedentarism: Any waking behaviour characterized by an energy expenditure of 1.5 metabolic equivalents or lower. Caspersen et al. (1985) defined physical fitness as ‘a set of attributes that are either health- or skill-related’. ‘Health-related’ attributes were defined as: (1) cardiorespiratory endurance; (2) muscular endurance; (3) muscular strength; (4) body composition; and (5) flexibility. This definition is useful because the degree to which an individual possesses a particular attribute is quantifiable and can be measured using specific laboratory assessments (Table 1). These attributes have consistently shown strong associations with mortality, disease-free survival and independent living (Araújo et al., 2024; Jochem et al., 2019; Kokkinos et al., 2023; Li et al., 2023; Roshanravan et al., 2017). In contrast, skill-related domains of physical fitness were defined as: (1) agility; (2) balance; (3) coordination; (4) speed; (5) power; and (6) reaction time (Caspersen et al., 1985). As noted by Caspersen et al., health-related components are more strongly related to health outcomes than are the skill-related components. However, the definition by Caspersen et al. (1985) is also a limited one, because it constrains fitness attributes to five health-related domains. With our ever-increasing ability to assess specific physical fitness attributes and the relationship of these attributes to health, we are beginning to understand that some of the so-called ‘skill-related’ domains of physical fitness, such as power and balance, also play important roles in individual health. Therefore, a contemporary view of ‘exercise as medicine’ might benefit from a broader definition of physical fitness. In particular, both power and balance are strongly associated with risk of mortality (Araújo et al., 2025; Cao et al., 2021). In addition, these attributes are quantifiable (Table 1); thus, we propose that power and balance be included as health-related attributes as we consider a broader definition of physical fitness (Figure 1). The term ‘fitness’ has been used from at least the 16th century. Before Darwin used it to describe the ‘quality of fulfilling the requirements of a particular environment for survival and reproduction’, the biological definition of physical fitness was the ‘ability to fulfil a particular physical task’ (see Box 2 regarding task specificity). As such, fitness encompasses a broad definition of anatomical and physiological attributes. Furthermore, outside of athletic training and competitions, humans rarely perform tasks in our daily lives that challenge the maximum capacity for these attributes. Therefore, a more contemporary definition of physical fitness, particularly in the context of ‘exercise as medicine’, might be ‘the ability to meet the demands of a particular physical task in a steady state’ (Sietsema et al., 2021). This recognizes that the relationship between maximum capacity of the neuromuscular–cardiorespiratory systems to power exercise and the ability to sustain exercise at high fractions of that maximum power are both malleable traits. Given that the mechanism(s) that link physical fitness to health remain elusive, the lexicon of attributes that contribute to physical fitness should be broad enough to facilitate their discovery. Soon after the British rower Sir Steven Redgrave won his fifth consecutive Olympic gold medal in 2000, he ran the London Marathon. With 16 years as the world's top rower, few would deny that Sir Redgrave was among the fittest endurance athletes in history. However, even with a maximum oxygen uptake approaching ∼7 L/min, he was unable to lug his 108 kg and 195 cm frame over 42.2 km in a time better than 4 h 55 min; far off times achieved by elite marathon runners (Maffetone et al., 2017). In contrast, elite runners, who tend to be shorter and weigh less than Sir Redgrave, may be capable of sustaining an effort equivalent to 90% of their maximum oxygen uptake during an entire race (Jones et al., 2021), contributing to their elite times. This example emphasizes the task specificity of physical fitness. With the above in mind, we will briefly revisit the definition of each health-related component of physical fitness. The definition of cardiorespiratory endurance offered by Caspersen et al. (1985) has some shortcomings. The term cardiorespiratory endurance was envisioned as synonymous with more contemporary terms, such as ‘cardiorespiratory fitness’ or ‘aerobic capacity’, and is a prime mediator of the ability to sustain endurance exercise. We propose that cardiorespiratory endurance should be defined by the body's ‘maximum rate of oxygen transport and utilization’, because it depends on the effective functioning of the pulmonary, circulatory, metabolic and neuromuscular systems to supply substrates (particularly oxygen) required for repeated muscle contractions and eliminate products of metabolism that contribute to fatigue, dyspnoea or pain. Assessment techniques for cardiorespiratory endurance therefore require sufficient challenge to all the physiological systems involved, necessitating sustained activation of a large muscle mass, high rates of convective and diffusive oxygen transport and high rates of ventilation that are needed to achieve maximum oxygen uptake (Table 1). Importantly, cardiorespiratory endurance is very strongly related to all-cause mortality (Kokkinos et al., 2023; Mandsager et al., 2018), supporting its status as a pillar of good health. Caspersen et al. (1985) defined muscular endurance as ‘the ability of the muscle groups to exert external force for many repetitions or successive exertions’. However, a challenge associated with this definition is that repetitive muscle contractions also tend to occur in activities that test cardiorespiratory endurance. Therefore, it seems appropriate to update the definition of muscular endurance to ‘the ability of the muscle groups to exert external force for many repetitions or successive exertions in dynamic or static conditions that are not limited by the cardiovascular or pulmonary systems’. This definition more closely aligns with the small muscle mass protocols used to assess muscular endurance (Table 1). This update will also help to differentiate cardiorespiratory endurance from muscular endurance. Overall, muscular endurance is an important health-related component of physical fitness, because it is associated with mortality and physical function (Roshanravan et al., 2017). Muscle strength is appropriately defined by Caspersen et al. (1985) as ‘the amount of external force that a muscle can exert’. Muscle strength is associated with all-cause mortality in both healthy and patient populations (García-Hermoso et al., 2018; Jochem et al., 2019). It is important to acknowledge that muscular strength is associated with muscle mass and, more specifically, physiological cross-sectional area (Riviati Li et al., 2023). As stated above, we propose that balance be reclassified from a skill-related to a health-related component of physical fitness. This is because balance disorders are associated with increased all-cause mortality, in addition to increased risk of death from cardiovascular diseases and cancer (Cao et al., 2021). Beyond this reclassification, the definition of balance as ‘the maintenance of equilibrium while stationary or moving’ (Caspersen et al., 1985) remains appropriate. Power is the other component of physical fitness that we propose should fall under health-related rather than skill-related attributes. It is now known that power, defined as ‘the rate at which one can perform work’, is strongly related to mortality (Alcazar et al., 2021; Araújo et al., 2025) and is therefore very important for maximizing individual health. This definition is somewhat ambiguous because the rate at which one can perform work depends strongly on the duration for which that power is measured. For example, aerobic power might be defined as the power output that can be supported at maximum oxygen uptake. However, the definition here relates to peak power output, which should be assessed over short intervals of time. We therefore propose that power be defined as ‘the maximum rate at which one can perform work for no longer than a few seconds’. Overall, understanding the health-related components of physical fitness provides important insight into how to use exercise as medicine to improve health and well-being. Caspersen et al. (1985) define ‘physical activity’ as any bodily movement produced by skeletal muscle contraction that results in energy expenditure. They categorize physical activity in daily life into occupational, sports, conditioning, household or other activities. Like physical fitness, physical activity is a quantifiable entity. As specified by Caspersen et al., the increase in energy expenditure attributable to physical activity can be assessed over a specified duration. However, accurate assessment of daily expenditure is complex. Indirect calorimetry requires equipment to measure oxygen uptake and carbon dioxide output and is typically applied during individual physical tasks rather than for extended durations (Macfarlane, 2017; Overstreet et al., 2017). The doubly labelled water method is the gold standard for measuring total daily energy expenditure in free-living humans (Buchowski, 2014) but is expensive to apply in large cohorts. Historically, researchers have relied on questionnaires and, more recently, on accelerometery to estimate energy expenditure. These estimates, which suffer from variability and inaccuracy (Plasqui Prince et al., 2008), are based on the concept of metabolic equivalents (METs). METs are the ratio of energy expenditure, relative to body mass, during daily living compared with a reference, which is, by convention, fixed for all people at 3.5 mL of oxygen consumption per kilogram of body mass per minute (this value represents the mean resting energy expenditure during quiet sitting of an average person). One MET is also equal to 1 kcal/kg body mass/h. The use of counting ‘steps’ with pedometers as a measure of physical activity is not recommended, because steps do not capture energy expenditure directly (Kumahara et al., 2009; Nielson et al., 2011). Energy expenditure measured via doubly labelled water, accelerometers and questionnaires is associated with important health outcomes (Andersen et al., 2014; Dempsey et al., 2022; Manini et al., 2006). Current physical activity guidelines are based on these observational studies that associate mortality risk with physical activity and call for accumulating ≥150 min of moderate-intensity (defined as activity that requires 3–6 METs) or 75 min of vigorous-intensity (> 6 METs) physical activity (or a combination) per week (World Health Organization, 2021). Unfortunately, we are lacking randomized controlled trials that are of long enough duration and adequately powered to assess the effects of physical activity on survival. The ‘Generation 100’ study (Stensvold et al., 2015) is an example of a randomized controlled trial to examine the effect of a 5 year exercise training programme on survival in >1500 people >70 years of age. Although there was no difference in mortality in the control, moderate- or high-intensity training groups, the overall mortality rate in the study was low, limiting the conclusions of the study. As yet, a causal link remains elusive. Additionally, randomized controlled trials on the effect of high rates of physical activity during early life or midlife on survival have yet to be performed. An important caveat to this story is whether it is physical activity (energy expenditure) or physical fitness (the ability to meet a particular task in a steady state) that imparts health and survival benefit (e.g., Myers et al., 2021). High rates of energy expenditure confer beneficial physiological adaptations when they are frequent, intense and progressive, and subsequent adaptations are specific to the tissues and energy systems involved. Hence, the ‘one-size-fits-all’ MET framework might obfuscate the causal relationships that studies are attempting to elucidate. Nonetheless, physical activity is a central tenet of the ‘exercise as medicine’ framework. In the paper by Caspersen et al. (1985), ‘exercise’ is defined as a subset of physical activity that is planned, structured and repetitive, with a final or intermediate objective of improving or maintaining physical fitness. Although valuable, we believe that this definition requires some qualification. In our experience, this definition does not adequately define a single acute bout of exercise, which might be spontaneous, might not be repeated and might not have the final or immediate objective of improving or maintaining physical fitness. An ‘acute exercise bout’ refers to a single session of exercise, defined by a given duration and intensity, and that might include resistance or endurance exercise. Chronic exercise, in contrast, is ‘structured and repetitive’. ‘Structured’ means that the exercise has a defined duration, intensity and type, whereas ‘repetitive’ denotes the frequency of acute structured exercise bouts. Chronic exercise encompasses ‘exercise training’, which includes chronic exercise performed with the objective of achieving a specific goal. In the spirit of this ‘exercise as medicine’ article collection, we propose that the concept of exercise training should include gaining health benefits associated with physical fitness. This is not to exclude improved physical performance (e.g., winning in a sports competition or improving a personal best); however, from a public health standpoint we believe it is important to identify the link between exercise and health; a link that is mediated (in still emerging ways) by physical fitness. We therefore advocate for the use of the terms ‘acute exercise’ and ‘chronic exercise’ (or ‘exercise training’ when appropriate) to distinguish between a ‘single bout’ of exercise and exercise that is ‘structured and repetitive’, where exercise itself is defined as ‘a subset of physical activity that is planned and structured to achieve the final or intermediate objective of gaining health benefits associated with improving or maintaining physical fitness’. Caspersen et al. (1985) do not provide definitions for the terms ‘physical inactivity’ and ‘sedentary behaviour’. Both terms should be distinguished from ‘rest’, which can be expressed as either basal metabolic rate or resting metabolic rate. Basal metabolic rate has classically been defined as a condition where all skeletal muscles, apart from those involved in normal tidal breathing, are inactive, with no motor impulses from the CNS, and is the minimum rate of energy expenditure needed for life metabolic in contrast, is measured in somewhat less conditions (e.g., after a min of et al., 2014) and therefore will be than basal metabolic rate Although the Centers for Disease Control and Prevention definitions for exercise do not encompass ‘physical the Health the term to describe the to meet physical activity recommendations (World Health Organization, 2021). Although this definition is particularly useful in public health research and it might be when applied to mechanistic studies and clinical trials in exercise It might also how We propose the of a more such as that by and colleagues et al., who defined physical as ‘physical activity levels below those required for optimal health and the prevention of premature This definition should remain relevant an physical and environment and a more Although we its that it might be in terms and to in it provides a framework that to Importantly, is from physical to this it has been defined as any waking behaviour characterized by an energy expenditure of METs while or of include work, and The guidelines this definition to include sitting activities and total sitting such as (World Health Organization, 2021). Although the 1.5 MET seems some sitting such as can this whereas still can be METs in et al., As such, this definition should be in the context of the specific research and study there is a at which be as or inactive, is more to be as a measured such as accelerometers and in addition to questionnaires and h & are to a amount of time but even those who are elite of time As in any other of definitions of central terms are ‘exercise as medicine’ and related remain relevant and some definitions and as the field The definitions in Box 1 a to achieve were involved in the and of the work, and all to the work or it for important the final of the and to be for all of the work in that related to the or of any of the work are appropriately and as for and all those who for are from the and is involved in clinical research with and is a at the of and the of a by The to and is of of the a of the and to the and a of the to the and is also a of the for and from and The for Physical is supported by and The had no in study and to or of the is supported by from of Health and Disease and of is supported by ongoing from the the the and and the Health and is a
Craighead et al. (Tue,) studied this question.