There is evidence to support that low fitness and physical activity, excess body weight, and excess abdominal fatness are associated with increased health risk. Moreover, evidence exists to support that sufficient levels of physical activity can significantly improve body weight, fatness, and health risk. Physical activity has been shown to have a modest effect on body weight that is typically <3% of initial body weight, but has an additive effect when combined with dietary restriction. Moreover, physical activity has been shown to be an important behavioral factor for enhancing long-term weight loss and minimizing weight regain; however, this may require relatively high doses of physical activity that approach 300min/week. Physical activity may concurrently reduce abdominal adiposity, and which may serve as a pathway by which there is also an improvement in health-related risk factors for various chronic diseases. There are important areas of research that require further investigation, with particular need to further examine the dose of physical activity that significantly affects these health outcomes. Moreover, there is a need for improved interventions to promote the adoption and long-term maintenance of physical activity, which can lead to improved weight control, abdominal adiposity, and chronic disease risk factors. Future research is also needed to understand the physiological/metabolic pathways and mechanisms that explain the influence of physical activity on long-term regulation of body weight.
Excess body weight has been associated with an increase in health-related diseases such as heart disease, diabetes, certain forms of cancer, musculoskeletal disorders, and other related problems (1). The morbidity and mortality of obesity-related conditions have resulted in an increase in the associated health-care cost of these conditions. For example, in the United States, the estimated cost associated with obesity exceeds $117 billion annually (2). Thus, it is important to consider intervention opportunities to lower the health risk associated with obesity, which in turn may reduce the health-care costs associated with these conditions. One of the lifestyle-intervention opportunities that has been shown to be effective is physical activity due to its importance in body-weight regulation and its independent effect on related health outcomes.
There have been numerous studies that have reported on the importance of physical activity for weight control. These studies have focused on prevention of weight gain, successful weight loss, and the prevention of weight regain. The interest in physical activity as a lifestyle strategy to combat the increasing prevalence of overweight and obesity stems from the fact that it is the one method that can be consistently used to increase energy expenditure. As highlighted by Ravussin and Bogardus (3) physical activity is the most variable component of total daily energy expenditure. Therefore, it is important to understand the contribution of an increase in energy expenditure resulting from physical activity on weight loss, long-term weight-loss maintenance, and the prevention of weight gain. Optimal weight control through physical activity may also contribute to reductions in abdominal adiposity and reductions in metabolic risk.
Does Physical Activity Affect Health Risk Independent of Body Weight?
There is a growing body of scientific literature to support the effect of physical activity on health-related outcomes independent of body weight. For example, data from the Aerobics Center Longitudinal Study has consistently shown an inverse relationship between fitness and mortality in adults regardless of level of weight status (normal weight, overweight, obese) (4,5,6,7). The pattern of an inverse association between health risk and physical activity, independent of weight status, is also supported by a number of other investigations (8,9,10). Lee et al. (11) have also reported an association between improved risk factor profiles and fitness at varying levels of abdominal adiposity measured by either computed tomography or magnetic resonance imaging.
Despite these findings, there is evidence to support the position that physical activity does not completely alleviate the negative health outcomes associated with obesity (8,9,10). This conclusion is consistent with the findings of Wing et al. (12) in a study of over 5,000 individuals with type 2 diabetes, which demonstrated that selected risk factors were significantly influenced by weight status (e.g., systolic blood pressure) whereas others were significantly influenced by fitness (e.g., HbA1c, ankle brachial index, Framingham Risk Score). Moreover, both physical activity and a smaller waist circumference have recently been shown to be associated with classification of enhanced metabolic health (presence of 0 or 1 cardiometabolic abnormality) in obese individuals (13). These results may suggest that it is important to intervene on both physical inactivity and excess body weight to have the greatest improvement in health outcomes.
A potential area for additional research related to the independent effect of physical activity on health risks in obese individuals may lie in the study of individuals with Class II (BMI: 35.0–39.9 kg/m2) or III obesity (BMI ≥40 kg/m2). Church et al. (5) acknowledged that there are limited data on the relationship between fitness and weight status among individuals with Class II or III obesity when examining data from the Aerobic Center Longitudinal Study due to few individuals with this weight status being classified as fit. This observation is confirmed by Wing et al. (12) in analysis of data from the Look AHEAD Study in which only 10% of individuals with Class III obesity were in the highest 40% for fitness classification, with only 1.8% of individuals in the highest tertile for waist circumference being in the highest fitness category. These findings may suggest that it may be difficult and uncommon for individuals in the highest categories of obesity and/or waist circumference to be able to achieve sufficiently high levels of fitness to improve health outcomes independent of concomitant reductions in body weight. Again, this may suggest that for some individuals, it is important to intervene on both physical inactivity and excess body weight to have the greatest improvement in health outcomes.
Effect of Physical Activity on Weight Loss
To effectively reduce body weight it is widely accepted that a negative energy balance needs to be present; whereas, to maintain weight loss and prevent weight gain, energy balance needs to be present. Thus, physical activity can be an important intervention to achieve desired body-weight regulation. However, the effect of physical activity alone on reductions in body weight may be somewhat modest. For example, in a study of overweight and obese adults at risk for the onset of type 2 diabetes, Wing et al. (14) compared interventions consisting of diet alone, physical activity alone, and the combination of diet plus physical activity. Results following the initial 6-month intervention period revealed weight losses of 9.1, 2.1, and 10.4% in these intervention conditions, respectively, with weight loss in the physical activity alone conditioning being significantly less than the weight loss in the diet alone and diet plus physical activity conditions. In a study of men, Wood et al. (15) reported that weight loss in response to physical activity was 3.0 and 4.0 kg at 7 and 12 months, which was significantly less than the 7.6 and 7.2 kg weight loss observed in response to a dietary intervention. In response to a 12-week intervention, Hagan et al. (16) report significantly less weight loss for men and women (0.6%) in response to physical activity alone compared to 8.4 and 5.5% in response to dietary restriction alone, and compared to 11.4 and 7.5% in response to the combined effects of diet plus physical activity. These data appear to indicate that physical activity has a modest impact on short-term weight loss in overweight and obese adults.
Donnelly et al. (17) examined supervised physical activity of 45min/day on 5 days/week over a 16-month period in overweight and Class I obese adults. The results of this study demonstrated that when dietary intake is unchanged, for men physical activity resulted in weight decrease of 5.2 kg vs. a 0.5 decrease observed in non-physical-activity control men. In contrast, physical activity for women blunted weight gain with weight increasing by 0.6 kg in exercising women vs. a 2.9 kg increase in non-physical-activity control women. The physical activity also resulted in significant reductions in abdominal adiposity as measured by computed tomography. For males, nonsignificant differences at 16-month that favored the physical activity vs. the control conditions were observed for visceral fat (difference of 9.9 cm2), subcutaneous fat (difference of 23.2 cm2), and total fat (difference of 33.2 cm2). For females, differences favoring the physical activity vs. the control group at 16-months were observed for visceral fat (difference of 8.6 cm2; P = 0.88), subcutaneous fat (difference of 53.3 cm2; P = 0.02), and total fat (difference of 61.9 cm2; P = 0.04) with the pattern of change in abdominal adiposity mirroring the observed changes in body weight between genders. These gender differences may be a result of the supervised physical activity resulting in different energy expenditures, with the average energy expenditure for each physical activity session 667 kcal (6.7 kcal/kg of fat-free mass) in men vs. 438 kcal (5.4 kcal/kg fat-free mass) in women (18). Thus, the observed gender differences may be a result of differences in energy expenditure resulting from physical activity. Thus, it will be important for future studies to examine whether clamping energy expenditure will result in similar changes in body weight in men and women.
The scientific evidence presented in the Physical Activity Guidelines Committee Report in the United States (http:www.health.govPAguidelines) confirms that physical alone typically results in weight change of <3% of initial body weight, with absolute weight loss ranging from ∼0.5 to 3.0 kg. This is consistent with the magnitude of 2.4 kg reported by the National Heart, Lung and Blood Institute's Expert Panel (1), and the 1–2 kg reduction in body weight reported in a review conducted by Wing (19).
This modest magnitude of weight loss resulting from physical activity alone is consistent with the definition of “weight stability” as defined by Stevens et al. (20), suggesting that a greater dose of physical activity or physical activity combined with reductions in energy intake may be required to elicit greater reductions in body weight. However, it is important to note that even modest reductions in body weight may be beneficial for reducing cardiometabolic risk. For example, Wing et al. (14) reported that a 2.3 kg reduction in body weight over a period of 24 months from physical activity or the combination of diet plus physical activity decreased the relative risk of developing diabetes by ∼30% in individuals initially identified with normal or impaired glucose tolerance.
Effect of Physical Activity on Long-Term Weight Loss and Weight-Loss Maintenance
Although physical activity may have modest effects on initial weight loss, physical activity may be a critically important strategy for the maintenance of weight loss. Jakicic et al. (21) have recently reported that relatively high doses of physical activity may be important for the long-term maintenance of weight loss. In a secondary analysis of data from a study of adult women, it was reported that individuals who achieved a weight loss of 10% of initial body weight that was maintained at 24 months were engaging in ∼275min/week of physical activity above baseline levels. Achievement of a 10% weight loss is consistent with the weight-loss target recommended by the National Heart, Lung and Blood Institute for overweight and obese adults (1). The importance of improved long-term weight loss in this study appears to also affect sustained reduction in abdominal adiposity as measured by waist circumference, sagittal diameter, waist-to-hip ratio (Figure 1).
These results confirm the importance of relatively high doses of physical activity to maintain weight loss as reported by others (22,23,24,25,26), and may confirm recommendations for weight-loss maintenance that have been reported (27,28,29,30). This is of importance because the maintenance of initial weight loss has been challenging. For example, it has been reported that following weight loss, ∼33–50% of initial weight loss is regained in the ensuing 12–18 months (31). However, physical activity appears to play a key role in the long-term maintenance of weight loss, which may ultimately result in sustained improvements in metabolic risk.
The National Weight Control Registry (NWCR) is a registry of individuals who have self-reported to have lost at least 30 pounds and have maintained this weight loss for at least 1 year (ref. 25). This provides interesting data related to the behaviors of individuals who have successfully lost and maintained significant weight loss. Of interest is the pattern of physical activity in this cohort. Klem et al. (25) report that individuals in the NWCR self-report participating in ∼2,800 kcal/week of leisure-time physical activity to maintain their body weight. McGuire et al. (32) further highlight the importance of physical activity for weight-loss maintenance by reporting that individuals in the National Weight Control Registry who regained weight after 1 year in this registry reported greater decreases in energy expenditure when compare to those individuals who continued to maintain their significant weight loss.
Evidence of the importance of physical activity for maintenance of weight loss is not limited to the National Weight Control Registry. For example, Kayman et al. (33) reported that 76% of individuals who had maintained significant weight loss reported using physical activity as a weight-loss strategy vs. 36% of individuals who relapsed from initial weight loss. When comparing individuals who had maintained a weight loss of 15 pounds vs. those who had relapsed, Gormally and Rardin (34) found that 54% of maintainers reported regular physical activity vs. 36% of those who had relapsed. Moreover, Leser et al. (35) reported that self-reported physical activity was inversely correlated (r = −0.53) with weight regain at a 3-year follow-up to a very-low-calorie diet intervention for weight loss. Physical activity was also shown to be predictive of maintenance of weight loss following an intervention involving pharmacotherapy (36). Despite these findings, it is important to note that physical activity is just one component of successful weight-loss maintenance, with Jakicic et al. (37) reporting that long-term weight-loss success is achieved through a constellation of behaviors, one of which is physical activity.
Effect of Physical Activity on Prevention of Weight Gain
As reported by Jakicic et al. (21), ∼27% of individuals engaged in a behavioral weight-loss intervention achieved a 10% weight loss at 24 months, with 49% of individuals achieving at least 5% weight loss. This confirms the difficulty of achieving long-term weight loss as reported by others (31). Thus, it may be important to focus intervention efforts on prevention of weight gain and obesity, and physical activity may be an important component of intervention efforts to prevent weight gain.
Results from cross-sectional and prospective observational studies appear to support the need for adequate levels of physical activity to prevention weight gain. Lee et al. (8) demonstrated from cross-sectional data that individuals who report levels of physical activity that are consistent with the consensus public health guidelines have a lower body weight than those not reporting this level of physical activity (8). A potential limitation of this study is the self-reported assessment of physical activity. Ball et al. (38) followed a cohort of 8,726 young women (age = 18–23 years) for a period of 4 years to examine predictors of weight maintenance. Results from this study demonstrated that 41% of these young women gained weight over this period of time, which was defined as gaining ≥5% of their initial body weight. Individuals who reported spending ≥33h/week sitting were 17–20% more likely to gain weight compare to those reporting sitting <33h/week. These results are consistent with other studies that have demonstrated sedentary behavior such as television viewing to be associated with increasing levels of body weight and obesity (39). A potential mechanism for this relationship is that a higher level of television viewing is associated with a lower resting metabolic rate in children (40). However, an alternative hypothesis would be that sedentary activities reduce the opportunities for physical activity, which results in an overall reduction in energy expenditure. Therefore, additional studies are required to conclusively determine how sedentary behaviors such as sitting and television viewing contribute to weight gain and obesity.
In a study that used an objective measurement of cardiorespiratory fitness, it was reported that weight gain was attenuated over a 4-year period as cardiorespiratory fitness increased over this same observation period (41). Cardiorespiratory fitness has been shown to improve with increased levels of physical activity. Therefore, these findings may reflect an increase in physical activity, with this contributing to the prevention of weight gain. In a study to examine predictors of weight gain in the Healthy Worker Project, French et al. (42) reported that over a 2-year period the average weight gain was 0.6 kg for women and 0.4 kg for men. Results of this study demonstrated that an increase of one walking session per week was predictive of a decrease in body weight over the 2-year period of 0.8 kg and 0.4 kg in women and men, respectively. However, these levels of physical activity combined with dietary behaviors and weight-loss history accounted for ∼9–10% of the weight loss observed.
There is also evidence from intervention studies supporting the importance of physical activity in the prevention of weight gain. Sherwood et al. (43) report that physical activity was predictive of weight change in the Pound of Prevention Study. Results of multivariate cross-sectional analysis revealed that moderate-intensity physical activity was predictive of less weight gain in this study; whereas multivariate prospective analysis revealed that high-intensity physical activity was predictive of weight change over the 3-year period. When examining data from intervention studies, the effect of physical activity on weight gain may result from the non-physical-activity control group gaining a modest amount of weight with those individuals participating in physical activity losing a modest amount of weight. For example, Slentz et al. (44) reported that physical activity resulted in a modest decrease in body weight and measures of body fatness, whereas control subjects not participating in physical activity had a modest increase in body weight in the STRIDDE Study.
In summary, based on the scientific evidence, it appears that physical activity can be an important component of intervention efforts for body-weight regulation. These findings are important as it appears that there is a concurrent reduction in abdominal adiposity with weight loss resulting from physical activity. However, the weight loss achieved with physical activity alone appears to be modest and is typically <3% of initial body weight. However, the addition of physical activity to a dietary intervention can result in improved weight loss compared to what is achieved through dietary modification alone. Moreover, it appears that physical activity is a key lifestyle behavior in longer term weight loss and the maintenance of weight loss. However, as summarized in the Physical Activity Guidelines Advisory Committee Report, additional research is needed to fully understand the contribution of physical activity to body-weight regulation (45). In general, these areas of research include the following:
1. Appropriately designed and powered studies are needed to understand the effect of different modes, intensities, duration, and frequencies of physical activity on weight loss, weight stability, and weight regain.
2. There may be a need for further research to understand how demographic characteristics such as age, gender, race/ethnicity influence the effect of physical activity on weight loss, weight stability, and weight regain.
3. There is a need for further research to be conducted that focus specifically on behavioral strategies that improve the adoption and long-term maintenance of physical activity in overweight and obese adults.
4. There is a need to examine whether the dose of physical activity that promotes improved long-term outcomes of weight control is different from the dose of physical activity that results in improvements in chronic disease risk factors and the onset of chronic disease across the entire range of body weight including those with severe obesity (e.g., Class II and Class III obesity).
J.M.J. has received consulting fees and grant support from BodyMedia, Proctor & Gamble, Free & Clear, UPMC Health Plan.