Effect of Physical Activity on Obesity in Older American Adults over Fourteen Years

Physical activity (PA) is recognized as an effective behavioral intervention for controlling body weight. However, the public’s neglect of importance of PA likely contributes to the prevalence of obesity among older adults. This study analyzed data from the National Health and Nutrition Examination Survey (NHANES) from 2005–2006 to 2017–2018 to examine the association between PA and obesity, and to identify trends and contribution of PA in older obese adults. A total of 5,812 obese participants were selected for association and trend analyses. Logistic regression was applied to estimate the associations between PA and obesity. Trend tests over 14 years were evaluated using orthogonal polynomial coefficients in a regression analysis. Students’ t-test was used to compare the slopes of the obesity trend and the PA trend. In two cycles (2005–2006 and 2017–2018), 1,461 participants were selected for Oaxaca-Blinder regression decomposition analysis to explain the contribution of PA to obesity over 14 years. Results revealed a significant negative correlation between PA and obesity in older adults, suggesting that increased participation in PA can reduce the likelihood of obesity (OR = 0.22, 95% CI=0.15-0.34, p = 0.00). From 2005 to 2018, the trend in PA closely paralleled the obesity trend. Notably, PA was identified as a contributing factor, accounting for 9% of the observed increase in obesity during this period. These findings can guide the development of targeted PA interventions aimed at reducing the widespread prevalence of obesity among older adults in the United States.

Introduction

The prevalence of obesity has been gradually increasing every year in America, affecting not only children and young adults but also older adults, with a prevalence rate exceeding 40% (National Center for Health Statistics, 2018). Recent national surveillance data indicate that obesity prevalence among U.S. adults aged 65 and older has nearly doubled over the past three decades, rising from approximately 22% in 1988–1994 to nearly 40% in 2015–2018 (CDC, 2025). Similar patterns have been observed globally, with increases in overweight and obesity among older adults contributing to higher risks of chronic disease, disability, and loss of independence (World Health Organization, 2025). Obesity in older adults is a serious problem that leads to increased morbidity and mortality and deserves more attention worldwide.

Physical activity (PA) is thought to be the most common method for expending energy and is considered a behavioral intervention for modifying body weight. PA intensity is particularly important because it is strongly related to energy expenditure. A cross-sectional study in Switzerland found that the energy expenditure of high-intensity PA is related to obesity (Bernsteinet al., 2004). Similarly, in obese individuals, body mass index (BMI) is strongly related to PA intensity, including moderate and vigorous activities (Hemmingsson & Ekelund, 2007). Even light intensity PA can be beneficial for older adults. Light-intensity PA has been associated with a lower BMI (Bannet al., 2015) and is closely related to abdominal fat distribution in obese older adults (Pescatello & Murphy, 1998). While the benefits of PA have been widely communicated to the public, older adults engage in significantly fewer minutes of moderate-to-vigorous PA than young adults (Davis & Fox, 2007). More than half (52.5%) of the American adults aged over 60 years reported no participation in leisure-time PA. Only 27% achieved more than 150 weekly minutes of leisure-time PA (Hugheset al., 2008). In the period of 2005–2006, older male adults aged 60–69 years dedicated 2.1% of their day to moderate-to-vigorous PA and 0.1% to vigorous PA. The decline in PA was more pronounced among older females, with 1.3% of their day spent on moderate-to-vigorous PA and 0.02% on vigorous PA. These figures further decrease with advancing age (Chastinet al., 2014). Consequently, most older adults fail to meet the recommended PA levels.

Despite its clear benefits, there is limited evidence on whether specific PA intensities—vigorous, moderate, or a combination of both—have distinct effects on obesity outcomes in older adults. Vigorous activity, which causes heavy sweating or large increases in breathing and heart rate, may provide greater energy expenditure and more pronounced metabolic improvements (Swiftet al., 2018), whereas moderate activity with smaller increases in breathing and heart rate may be more feasible and sustainable for older populations (Paterson & Warburton, 2010). A combined vigorous–moderate approach can balance metabolic efficiency with long-term adherence (Powellet al., 2011). However, most existing studies either aggregate PA intensities or focus on younger populations, leaving uncertainty about the intensity-specific effects in older adults. This study addresses this gap by separately analyzing vigorous, moderate, and combined vigorous–moderate PA in relation to obesity prevalence among U.S. adults.

One possible reason for the increase in obesity may be the insufficient attention given by the public and researchers to the impact of PA on obesity. If the reduction in PA plays a significant role in the rising rates of obesity among the elderly in recent decades, it warrants greater scrutiny. Conversely, analyzing the trends in PA and their relationship to obesity could enhance our understanding of health among older American adults and facilitate the development of effective public health policies and evidence-based interventions to combat obesity in the elderly. However, there is a notable gap in trend analysis concerning PA and obesity in the elderly, particularly in studies spanning over ten years. Accordingly, this study seeks to answer whether different intensities of physical activity (vigorous, moderate, and combined vigorous and moderate) are associated with obesity prevalence in older U.S. adults, how trends in physical activity parallel changes in obesity over time, and to what extent reductions in physical activity contribute to rising obesity rates; we hypothesize that higher participation in all forms of physical activity is inversely related to obesity, that declines in activity mirror increases in obesity, and that reductions in moderate and combined vigorous and moderate activity significantly explain the growth of obesity among older adults. Therefore, due to the limited research in this area and the lack of comprehensive trend analysis, the present study aimed to investigate the association, trend, and impact of PA on obesity among older American adults.

Methods

Considering the consistency of physical activity questionnaires and the impact of the pandemic on obesity, this study used data from the National Health and Nutrition Examination Survey (NHANES) spanning multiple waves: 2005–2006, 2007–2008, 2009–2010, 2011–2012, 2013–2014, 2015–2016, and 2017–2018. The NHANES received approval from the National Center for Health Statistics Research Ethics Review Board and all participants provided written informed consent.

Participant Recruitment and Sample Selection

The NHANES uses a multistage sampling method to represent the civilian, non-institutionalized U.S. population. This process selects the counties, households, and individuals. Some groups, such as older adults, non-Hispanic Black people, and Hispanic people, are sampled more often to improve data accuracy. Trained interviewers visited participants’ homes for interviews, and then completed physical examinations and lab tests at Mobile Examination Centers (MECs).

Older adults were defined as individuals aged 65 years or older. Obesity, based on BMI (kg/m², body weight and body height were measured by CDC specialists.), was represented as a dichotomous variable: 1 for BMI ≥ 30 kg/m² and 0 otherwise. Individuals with missing responses for any key study variable or who answered “don’t know” or refused to answer physical activity questions were excluded.

During the analysis, logistic regression and trend examination were conducted by pooling data from seven waves of data spanning 2005–2006 to 2017–2018. This aggregation yielded a sample size of 5,812 for obesity analysis. The sample sizes for each wave were as follows: cycle 2005–2006 (n = 702), cycle 2007–2008 (n = 995), cycle 2009–2010 (n = 976), cycle 2011–2012 (n = 772), cycle 2013–2014 (n = 806), cycle 2015–2016 (n = 811), and cycle 2017–2018 (n = 800). In the Oaxaca-Blinder decomposition analysis, the obesity group comprised 1,502 participants, encompassing data from only two cycles (2005–2006 and 2017–2018).

Intensities of physical activity

The Physical Activity file of the NHANES questionnaire data included inquiries regarding participants’ engagement in moderate and vigorous physical activity during their leisure time. Vigorous activity was defined as “sports, fitness, or recreational activities that cause heavy sweating or large increases in breathing or heart rate.” Examples in the 2005–2006 cycle included running, lap swimming, and aerobics class or fast bicycling, while in 2017–2018, examples were running or basketball. Moderate activity was characterized as “sports, fitness, or recreational activities that cause a small increase in breathing or heart rate.” Examples provided in the 2005–2006 cycle included brisk walking, bicycling for pleasure, golf, and dancing, whereas in the 2017–2018 cycle, examples were brisk walking, bicycling, swimming, or volleyball. The questions were phrased as follows: “Over the past 30 days, did you do [. . . ] activities for at least 10 minutes [. . . ]?” (2005–2006) and “In a typical week, do you do [. . . ] activities for at least 10 minutes [. . . ]?” (2017–2018). Respondents provided “yes” or “no” answers to each question regarding moderate- and vigorous-intensity activities.

In this analysis, four intensity categories were established based on responses to two questions: moderate intensity if respondents answered “Yes” to the moderate-intensity PA question but “No” to the vigorous-intensity PA question; vigorous intensity if respondents answered “Yes” to the vigorous-intensity PA question but “No” to the moderate-intensity PA question; moderate combined with vigorous intensity if respondents answered “Yes” to both questions; and low-intensity or none if respondents answered “No” to both questions. In the regression analysis, the group with low intensity or no PA was designated as the reference group, compared with the vigorous-intensity PA, moderate-intensity PA, and moderate combined with vigorous-intensity PA groups.

Demographic Characteristics

The demographic characteristics of the participants were extracted from the Demographic Variables & Sample Weights files in the NHANES database. In the gender regressions, women were set as the reference group. Age was analyzed using “Age in years at screening”, ranging from 65 to 80 years, as the NHANES top-codes age at 80. Dichotomous variables were created in four age categories: 65–69 years old, 70–74 years old, 75–79 years old, and 80 years old and over, with the reference group set as 65–69 years old in the regression models.

Ethnicity was represented by dichotomous variables for Non-Hispanic White, Mexican Americans, other Hispanic, Non-Hispanic Black, and other races, with non-Hispanic White as the reference category in the regressions. Education level, obtained from the NHANES demographic questionnaire, included a question “What is the highest grade or level of school completed or the highest degree received”. Five dichotomous variables were categorized as follows: less than 9th grade, 9−11th grade, high school, some college or associate (AA) degree, and college graduate or above. In the regressions, less than 9th grade the reference category.

Marital status was grouped into four categories: never married, married or living with a partner, widowed, divorced or separated, with married or living with a partner as the reference group in the regression models. Annual family income variables reflected six categories: under $20,000, $20,000–34,999, $35,000–44,999, $45,000–54,999, $55,000–64,999, and $65,000 and over, with the under $20,000 group set as the reference group in the regression models.

Statistical Analyses

All analyses were conducted using the Stata software (Version 15.1, StataCorp, LP, USA) with multi-year survey weights. Following the NHANES guidelines, we calculated these weights by dividing the 2-year sample weights by the number of combined survey cycles to ensure nationally representative results that accounted for the complex survey design. We conducted a multivariable logistic regression analysis with the outcome being obese (vs. non-obese) defined by BMI, and the independent variables were participations in different PA intensities. Covariates in this logistic regression included age, gender, ethnicity, education level, marital status, and annual family income.

Tests for trends in different intensities of PA participation across seven cycles were assessed using linear, quadratic, cubic, quartic, quintic, and sextic orthogonal polynomial coefficients in regression analysis. The Student’s t-test was utilized to compare the slope of the obesity trend with the slopes of the PA trends. The calculation of this t-value was outlined by Andrade and Estévez-Pérez (Andrade & Estévez-Pérez, 2014). In brief, we initially established two regression lines for the trends in obesity and PA participation. Next, we calculated the residual variances (squared standard errors of the two regressions) to estimate the variances of the two regressions. We then compared the F value of the experimental t-test with the F value of the tabulated one. If the F value of experimental t-test was less than the F value of the tabulated one (p > 0.05), the null hypothesis could not be rejected. If, however, the F value of experimental t-test was greater than the F value of the tabulated one (p < 0.05), the null was rejected, indicating that the two slopes were considered to be statistically significantly different.

The Oaxaca-Blinder (OB) decomposition (Blinder, 1973; Oaxaca, 1973) was originally designed to analyze wage differentials between male-female and/or white-black, decomposing these differences into percentages and directions of contributions. This method has since been expanded to explain health disparities, including obesity research (Sen, 2014). In addition to analyzing racial or gender differences, period differentials can also be decomposed using this approach (Nieet al., 2018). This study applied OB decomposition to evaluate the role of PA in explaining the change in obesity between the 2005–206 cycle and 2017–2018 cycle. An alpha level of 0.05 was set to determine statistical significance.

Results

Table I presents the characteristics of the sample comprising obese older American adults. The associations between PA and obesity among older adults from the NHANES 2005–2018 are listed in Table II, with demographic variables included in the logistic regression as covariates. As indicated, among older adults, vigorous, moderate, and moderate combined vigorous PA intensities were significantly negatively correlated with obesity (p < 0.05, OR < 1). In simpler terms, for every 1% increase in participation in vigorous, moderate, and moderate combined vigorous-intensity PA, the odds of obesity were reduced by 48%, 47%, and 78%, respectively (OR = 0.52, 0.53, 0.22; 95% CI = 0.31, 0.89; 0.44, 0.64; 0.15, 0.34). Thus, regardless of PA intensity, all forms of PA can lower the risk of obesity in the elderly population. The trends in PA intensities (vigorous, moderate, and moderate combined vigorous) over seven two-year cycles (2005–2018) are depicted in Figs. 1–3. Specifically, the trend for vigorous PA was quadratic (Fig. 1), whereas both trends for moderate and moderate combined vigorous intensities PA were cubic (Fig. 2).

Table I. Characteristics of the Sample in Older Adults (n = 5,812)

Characteristics	Descriptive statistics
Age (%)
65–69	33.54% ± 0.01
70–74	26.35% ± 0.01
75–79	17.59% ± 0.01
80 and over	22.52% ± 0.01
Gender (%)
Female	56.46% ± 0.01
Male	43.54% ± 0.01
Ethnicity (%)
Non-Hispanic White	82.69% ± 0.01
Mexican American	3.18% ± 0.00
Other Hispanic	2.99% ± 0.00
Other races	4.19% ± 0.00
Non-Hispanic Black	6.95% ± 0.01
Education level (%)
Less than 9th	8.30% ± 0.01
College graduate above	26.88% ± 0.01
9–11 grade	11.51% ± 0.01
High school	25.93% ± 0.01
Some college or AA degree	27.38% ± 0.01
Marital status (%)
Married & living with partner	62.44% ± 0.01
Divorced & separated	12.10% ± 0.01
Widowed	22.68% ± 0.01
Never married	2.78% ± 0.00
Annual family income (%)
Under 20,000	18.68% ± 0.01
20,000–34,999	22.05% ± 0.01
35,000–44,999	11.85% ± 0.01
45,000–54,999	9.50% ± 0.01
55,000–64,999	5.40% ± 0.00
65,000 and over	29.00% ± 0.02
20,000 and over	3.52% ± 0.00
PA (%)
Low/none	54.29% ± 0.01
Vigorous	2.59% ± 0.00
Moderate	35.44% ± 0.01
Vigorous & moderate	7.68% ± 0.01

Table II. Associations Between Physical Activity and Obesity in Older Adults (n = 5,812)

Physical activity intensity	OR	SE	t	P > |t|	95% CI
Low/none (reference)
Vigorous	0.52	0.14	−2.42	0.02*	0.31	0.89
Moderate	0.53	0.05	−6.71	0.00*	0.44	0.64
Moderate & vigorous	0.22	0.05	−6.96	0.00*	0.15	0.34

Fig. 1. Quadratic trend in vigorous physical activity over time.

Fig. 2. Cubic trend in moderate physical activity over time.

Fig. 3. Quartic trend in combined moderate and vigorous physical activity over time.

In this study, Oaxaca-Blinder regression decomposition was applied to assess the extent to which changes in obesity could be attributed to PA from 2005–2006 to 2017–2018. A sample of 1,461 subjects was selected to investigate the impact of PA on obesity in the elderly during these two cycles. Table III presents the findings, indicating a significant increase in both BMI and obesity percentage among the older adults from 2005 to 2018 (p < 0.001). Specifically, BMI rose by 5% (from 27.70 in 2005–2006 to 29.04 in 2017–2018) and the percentage of obesity surged by 38% (from 27.54% in 2005–2006 to 38.01% in 2017–2018) over fourteen-year period. Moreover, more American older adults opted for low-intensity PA or no PA in 2017–2018 compared than in 2005–2006. However, participation in moderate-intensity PA and moderate combined vigorous-intensity PA decreased significantly during this timeframe (p < 0.05). In the OB decomposition analysis, PA totally explained 9% of the increase in obesity from 2005 to 2018. Moderate-intensity PA and moderate combined vigorous-intensity PA accounted for 4.1% and 4.83%, respectively (Table IV). Therefore, it is recommended that older American adults engage in more moderate and/or combined vigorous-intensity PA to counteract the rising prevalence of obesity. Dietary factors were not included in this analysis, as the primary objective was to examine the impact of physical activity on obesity. Influences of other factors on obesity have been extensively addressed in other studies.

Table III. Weighted Means or Percentage of Obesity and Physical Activity in Cycles of 2005–2006 and 2017–2018 (n = 1,461)

Variables	2005–2006 (n = 690)	2017–2018 (n = 771)	t	p
Obesity
BMI (kg/m2)	27.70	29.04	−3.71	0.00*
Obesity % (BMI)	27.54	38.01	−3.81	0.00*
Physical Activity Intensity
Low/none	42.11	50.32	−5.99	0.00*
Vigorous	3.24	3.13	−0.52	0.61
Moderate	41.04	36.97	3.93	0.00*
Vigorous & moderate	13.62	9.58	4.18	0.00*

Table IV. Contributions of Physical Activity in Explaining the Difference of Elderly Obesity Over 14 Years (n = 1,461)

	OB decomposition
	Coef.	Contribution % (explained %)	95% CI
Overall
2017–2018	29.04		28.48	29.59
2005–2006	27.70		27.23	28.17
Difference	1.34		0.61	2.07
Explained
PA*	0.12	9.00	0.00	0.24
Low/none (reference)
Vigorous	0.00	0.07	−0.02	0.02
Moderate	0.05	4.10	−0.04	0.15
Moderate & vigorous	0.06	4.83	−0.03	0.16

Discussion

This study demonstrated a significant negative association between PA and obesity among American older adults. Obesity prevalence increased from 27.54% in 2005–2006 to 38.01% in 2017–2018, while participation in moderate and combined vigorous-moderate PA significantly declined. Logistic regression confirmed that all intensities of PA-vigorous, moderate, and combined-were inversely associated with obesity risk. Trend analyses revealed that declines in PA participation closely paralleled increases in obesity. Oaxaca-Blinder decomposition further showed that reduced PA explained 9% of the observed increase in obesity over the 14-year period, with combined vigorous–moderate PA contributing the largest share. These findings highlight the critical role of sustained PA in mitigating obesity among older adults.

In the current study, the results revealed a significant increase in the obesity percentage among older American adults (Table III). Notably, this obesity percentage for 2017–2018 was slightly lower than the that reported by the CDC, which indicated an obesity percentage of over 40% among older adults in 2013–2016 (National Center for Health Statistics, 2018). Comparing the periods 2005–2006 and 2017–2018, there was a 38% increase in the obesity percentage. Over the course of 14 years (seven two-year cycles), the average annual change in obesity percentage was 2.71%. These results align with previous research indicating a recent increase in obesity (National Center for Health Statistics, 2018).

Furthermore, obesity in older adults is associated with numerous fatal diseases and significantly affects the their quality of life (Samper-Ternent & Al Snih, 2012). PA has long been regarded as the most effective means of controlling obesity (Dubnovet al., 2003). However, participation in PA among older adults, including vigorous, moderate, and moderate combined vigorous intensities of PA, decreased from 2005 to 2018, coinciding with an increase in low or no PA over the 14 years (Table III). Specifically, the number of participants with moderate PA decreased significantly by 10%. The largest decrease among the three PA intensities was observed in the moderate combined vigorous PA group, which decreased by 30% over 14 years. Hence, it is crucial to identify the association and impact of PA on obesity to prevent and intervene in the increasing trend of obesity in the elderly as soon as possible.

Undoubtedly, the findings of the present study underscore the significant impact of PA on obesity. Among older adults, participation in vigorous, moderate, and moderate combined vigorous-intensity PA was significantly and inversely associated with the odds of obesity (p < 0.05, OR < 1). Remarkably, individuals who engaged in moderate combined vigorous PA had lower odds of being obese than those who engaged in either moderate or only vigorous-intensity PA alone. This is in line with previous research indicating that high-intensity PA was correlated with lower rates of obesity rather than moderate PA alone (Bernsteinet al., 2004), and that moderate-to-vigorous PA was strongly linked to obesity (Maheret al., 2013). Consequently, older adults are encouraged to incorporate both vigorous and moderate PA into their routines, rather than solely on one format of PA.

A national estimate of 5,589 older adults aged 60 years and over, based on the NHANES 1999-2004, found that 52.5% of older American adults had no leisure-time PA (Hugheset al., 2008). Our study continued this research on older adults and found that 54.29% of participants had low or no PA (data not shown), indicating an increase in the proportion of older adults engaging in low or no PA after 2004. Additionally, more than one-third of individuals participated in moderate PA, while only 3% participated in vigorous PA. The percentage of participants who engaged in both moderate and vigorous PA was less than 10%. These data highlight a low participation rate in PA among older adults, particularly in high-intensity PA. This could be attributed to movement limitations in older adults, such as sarcopenia (Hairiet al., 2010). Furthermore, the trends of all three PA intensities mirrored the obesity trend. Therefore, all PA participation levels contributed to the prevalence of obesity among older American adults from 2005 to 2018.

A few factors contribute to the low participation of all-intensity PA among older adults. Sociocultural shifts, such as increased sedentary leisure activities and reduced engagement in community-based exercise, have limited opportunities for regular movement (Kinget al., 1998). Environmental barriers, including inadequate pedestrian infrastructure, limited access to safe recreational spaces, and extreme weather conditions, can further discourage outdoor activity (Satarianoet al., 2010). Policy changes, such as reduced funding for senior fitness programs or a lack of supportive workplace and transportation policies, also play a role in lowering physical activity participation (Heathet al., 2012). Collectively, these factors can create an obesogenic environment, reinforcing the link between declining physical activity and increasing obesity prevalence in this population.

OB decomposition was used to determine how much variation in obesity could be attributed to PA from 2005–2006 to 2017–2018. Overall, our findings suggested that a 9% increase in obesity over 14 years can be explained by PA participation among older American adults (Table IV). The decline in participation in vigorous PA, moderate PA, and moderate combined vigorous PA contributed to the rise in obesity rates between 2005 and 2018. Among the three intensities of PA, moderate combined vigorous-intensity PA made the largest contribution to obesity in the elderly (4.83% out of 9%), followed by moderate-intensity PA (4.1% out of 9%). Several studies have focused on the low participation in PA in older adults and the prevalence of obesity (Davis & Fox, 2007; Tudor-Lockeet al., 2010). However, this study further quantified the exact contribution of PA participation to obesity, considering different intensities of PA. Thus, PA, being an important contributor to obesity in the elderly, should be given greater attention in efforts to control weight gain among older American adults.

In summary, this study highlighted the negative association between PA participation and obesity prevalence, the parallel trends between them, and the important contribution of PA participation to obesity among older American adults. The participants of this study represent a nationally representative sample of older adults in the United States, allowing for the generalization of the results to the broader American older adult population. To our knowledge, this is the first study to use OB decomposition regression to assess the impact of PA participation on the difference in elderly obesity over a 14-year period. These associations between PA and obesity lay the foundation for promoting healthy ageing. Therefore, there is an urgent need for tailored public health strategies aimed at sustaining or increasing physical activity among older adults, such as community-based exercise programs, age-friendly urban designs, and social support initiatives. Such targeted interventions can help mitigate obesity risk while improving the overall health and quality of life in this growing population.

Limitation

This study is a secondary analysis using the NHANES dataset and relies exclusively on self-reported leisure-time physical activity data collected through questionnaires without objective measurements such as accelerometers. This reliance introduces a significant risk of measurement bias, as responses may be affected by memory errors, misinterpretation of questions, or social desirability bias, particularly among older adults. Moreover, non-leisure activities may represent the predominant form of physical activity in this population but were not captured in the dataset. It is also noteworthy that only 2.6% of participants reported engaging in vigorous physical activity. This small subgroup size substantially reduces the statistical power for analyses involving this group, may yield unstable estimates, and limits the generalizability of vigorous-PA related findings to the broader older adult population. In addition, only variables related to participation in different PA intensities were consistent from 2005–2006 to 2017–2018, resulting in the exclusion of other potentially relevant PA variables. Future studies should examine the associations between additional PA characteristics (e.g., duration and frequency) using objective measurements. Furthermore, differences in question wording between survey years and specifically, the change from asking about activities “over the past 30 days” in 2005–2006 to “in a typical week” in 2017–2018—may have affected how participants reported physical activity, potentially limiting direct comparability of data across cycles. Additionally, since NHANES is a cross-sectional study rather than a longitudinal follow-up of the same individuals, establishing causal relationships between PA and obesity is not feasible. Further research is needed to delve deeper into these relationships and understand the underlying mechanisms involved.

Conclusion

This study presents compelling evidence of a significantly negative association between PA and obesity among older adults in the United States. The analysis of PA trends revealed similarities to obesity trends. Moreover, the decomposition regression showed that PA accounted for a 9% increase in obesity from 2005 to 2018. These findings have important implications for obesity prevention and intervention and offer recommendations for weight management. Effective strategies aimed at improving PA participation could potentially reduce the prevalence of obesity in older adults.

Acknowledgment

This study was supported by the Department of Nutrition and Hospitality Management at University of Mississippi, U.S.

Conflict of Interest

The authors declare that they do not have any conflict of interest.