Association of Regular Leisure-Time Physical Activity with Body Mass Index and Obesity Risk in Taiwanese Young Adults

Background: The purpose of this study was to determine the association of regular leisure-time physical activity with body mass index and obesity risk in Taiwanese young adults. Methods: A total of 10,802 young adults (18-44 years) were enrolled into this national telephone survey. The questionnaire data of this survey includes socio-demographic characteristics, zip code of residence, leisure-time physical activity (LTPA) behaviors, self-reported health status, and self-evaluations (including height, body weight, and body mass index [BMI]). Results: When using non-regular LTPA as the baseline, participants in regular LTPA group exhibited the lower risks of overweight and underweight (OR, 0.837; 95% CI 0.738–0.948, OR, 0.732; 95% CI 0.611-0.876). But there was no signicant relationship between the regular LTPA and obesity risk when using non-regular LTPA as the baseline after adjusting for potential confounders. Conclusions: The study results revealed that regular LTPA effectively lowered the risks of underweight and overweight. However, for people with obesity, regular LTPA was unable to signicantly decrease their obesity risk. The cross-sectional study data were obtained from the national-wide survey for children and adolescents (13–17 years), adults (18–64 years), and older adults (65 years and older) through the Taiwan National Physical Activity Survey (TNPAS) conducted by the Taiwan. Recruitment was conducted by random-digit-dialing to the proportional-stratied sampling with multiple factors (e.g., age, gender, and geographic districts), and the detailed procedure has been Citizens aged over 13 years and stratied by 22 cities/counties across the were selected to become the The of each city/county was The total was 2020, with sampling errors of 3–5% and condence intervals (CI) of 95%, which allowed for achieving a sucient sample size and statistical power. Subsequently, a two-month telephone interview (CATI) was conducted by a computer-assisted system from August to October 2020. To ensure the quality of data collection, a group of well-trained and experienced interviewers was employed in the CATI. The following data were collected through the telephone survey by socio-demographic characteristics (i.e., age, gender, education, and occupation), physical activity behaviors, self-reported health status, and self-evaluations (e.g., body height and body weight), and zip code of residence. Finally, a total of 10,802 young adults (18–44 years) was selected in this study. The participants were fully informed about the objective, procedure, and content of this study. This study was conducted under the Declaration of Helsinki, and all procedures have been approved by the Institutional Review Board of the Fu Jen Catholic University in Taiwan (FJU- IRB C109085). The oral consent was given before the interview. All relevant information was contained in the de-identied secondary dataset and was released for public research

determine the association of regular leisure-time physical activity with body mass index and obesity risk in Taiwanese young adults.

Study sample and data collection procedures
The cross-sectional study data were obtained from the national-wide survey for children and adolescents (13-17 years), adults (18-64 years), and older adults (65 years and older) through the Taiwan National Physical Activity Survey (TNPAS) conducted by the Sports Administration, Ministry of Education in Taiwan. Recruitment was conducted by random-digit-dialing to the proportional-strati ed sampling with multiple factors (e.g., age, gender, and geographic districts), and the detailed procedure has been described elsewhere [13]. Citizens aged over 13 years and strati ed by 22 cities/counties across the country were selected to become the sampling population. The sample size of each city/county was determined by its proportion to the country population in Taiwan. The total sample size was 25,526 in 2020, with sampling errors of 3-5% and con dence intervals (CI) of 95%, which allowed for achieving a su cient sample size and statistical power. Subsequently, a two-month telephone interview (CATI) was conducted by a computer-assisted system from August to October in 2020. To ensure the quality of data collection, a group of well-trained and experienced interviewers was employed in the CATI. The following data were collected through the telephone survey by socio-demographic characteristics (i.e., age, gender, education, and occupation), physical activity behaviors, self-reported health status, and self-evaluations (e.g., body height and body weight), and zip code of residence. Finally, a total of 10,802 young adults (18-44 years) was selected in this study. The participants were fully informed about the objective, procedure, and content of this study. This study was conducted under the Declaration of Helsinki, and all procedures have been approved by the Institutional Review Board of the Fu Jen Catholic University in Taiwan (FJU-IRB C109085). The oral consent was given before the interview. All relevant information was contained in the de-identi ed secondary dataset and was released for public research purposes.

Data collection
Multiple demographic characteristics of participants were required and recorded in this study, and they are age, gender, education, occupation, and self-reported health status. The age of participants was divided into 18-24 25-29 30-34 35-39, and 40-44 years, and education was categorized by the elementary school or lower, junior or senior high school, and college or higher. Detailed categories of occupation included white collar, government servant, blue-collar, owner/manager, specialists, student, housewife, retired, free-lancer, jobless, and others. The self-reported health status contained excellent or good, fair, and very bad or poor.

LTPA Assessment
A series of questions through a CATI was utilized to determine either regular or non-regular LTPA in this study. The process of a CATI with the step-by-step questions as described in the following: First, the participant's current LTPA participation was asked by the following question, "Have you taken part in any LTPA in the past month?" Second, both frequency and duration of LTPA participation were asks by the following questions when the respondent provided a positive response, "How many times do you participate in LTPA per week? "; "How many minutes do you usually spend at one time?" Third, the LTPA intensity was assessed by the description of breathing and sweating status from the participant, and those questions included "When you are doing LTPA, you usually feel….". And then, the following structural answers were selected and described by the respondent, "No changes in my breath and sweating," "I breathe faster but do not sweat," "I breathe normally but sweat," "I breathe quickly and sweat." When the respondent indicated they usually breathed quickly and sweated that was considered to engage in moderate-intensity LTPA. Finally, regular and non-regular LTPA groups were de ned by the following Statistical analysis SAS 9.4 (SAS Institute, Cary, NC, USA) was utilized to perform multiple statistical analyses in this study including student's t-test, chi-square test, and multiple linear regression analysis. Continuous variables were analyzed through the student's t-test and categorical variables were analyzed by using the chisquare test. BMI was a dependent variable and used to examine the relationship between regular LTPA and BMI after the adjustment of potential confounders by using multiple linear regression analysis.
Adjusted odds ratios (ORs) with 95% con dence intervals (CIs) were applied on obese, overweight, or underweight were calculated from unconditional logistic regression models based on regular LTPA. In this study, test values were presented as means ± standard deviation (SD) or frequency percentages, and test results were evaluated with two-tailed and statistically signi cant at p < 0.05. Table 1 shown the result of the demographic characteristics. There were 10,802 data were dichotomized into groups regarding their Leisure Time Physical Activity (LTPA) status and most of participants (76%) were classi ed into the non-regular LTPA group. Except BMI, the signi cant differences were shown between groups (p < 0.05) on all relevant variables include age, gender, height, body weight, obese status, education, occupation and self-reported health status. Participants who were in non-regular LTPA group, the rates fo overweight (19.50%) and obese (15.00%) are higher than regular LTPA group and 18%   higher than non-regular LTPA group (06.40%). In addition, both men (12.00%; 08.80%) and women (06.20%; 03.80%) the obesity prevalence of non-regular LTPA group were signi cant higher than regular LTPA in age 25-29. On the contrary, the overweight prevalence of non-regular LTPA group were signi cant lower than regular LTPA in women (08.90%; 12.70%). The overweight (11.00%; 09.50%) and obesity (11.30%; 00.00%) prevalence of non-regular LTPA were signi cant higher than regular LTPA group in age 30-34 group in women, in men the signi cant difference were found in age 35-39 group too (overwight:31.30%; 29.10%, obesity:21.70%; 20.90%).  CI 0.611-0.876). But there was no signi cant relationship between the regular LTPA and obesity risk when using non-regular LTPA as the baseline in Model 2.  According to the study results, regular LTPA reduced the odds of being underweight by 26.8%. The causes of being underweight include anorexia nervosa, cognitive behavior [15], and being picky eaters [16], which result in insu cient calorie intake for extended periods of time. Studies have indicated that 120-min aerobic exercise per week can effectively increase the weight and tness of patients who are underweight [17]. For patients with anorexia nervosa, exercise intervention can effectively improve their nutritional status [18] and quality of life [19]. Additionally, exercise intervention can improve their weight and eating disorder [20] considerably more than that achieved with traditional cognitive-behavioral therapy.

Results
Although this study did not explore the speci c causes of becoming underweight, the multivariate logistic regression analysis (adjusted for variables such as age, gender, self-reported health status, occupation, and education) found that exercise effectively decreased the odds of being underweight, suggesting that developing regular exercise habits may be key to lowering the odds of being underweight.
The current study indicated that LTPA can effectively diminish the odds of being overweight. LTPA can successfully reduce abdominal and visceral fat [9,10], and the amount of physical activity and fat reduction are signi cantly and positively correlated [11]. Sedentary lifestyle and minimal physical activity are the main causes of increasing obesity prevalence [21]. In the United Kingdom, decreased physical activity level has been found to be the primary cause of high obesity prevalence [22]. Consistently, the present study discovered that regular LTPA can effectively lower the odds of being overweight.
Although LTPA could reduce the odds of being overweight considerably, it was unable to lower the odds of obesity; this may be because of the de nition of LTPA. The ACSM asserted that 150 min of moderateintensity exercise per week can effectively maintain people's weight and health status. However, 150-200 min of moderate-intensity exercise per week can only achieve minimal reduction in body weight. Without dietary intervention, > 250 min of moderate-intensity exercise per week is required to show any pronounced effect [12]. Other review-based studies have also indicated that 150 min of moderateintensity exercise is ineffective in reducing weight [23], and recommended that 150-300 min of moderate-intensity or 75-150 min of high-intensity exercise per week [24] for weight loss. Therefore, for patients with obesity, exercise intervention may be insu cient, and dietary intervention or cognitivebehavioral therapy may be necessary.
This study had a su cient sample size similar to those of previous studies on physical activity and BMI [25][26][27], and used questionnaires with favorable reliability and validity which have also been used in other studies [26]; thus, the sample size, study reliability, and study validity were representative of the overall population.
Although this study incorporated various intervening variables, it still had the following limitations: First, the data were limited to Taiwan. Future studies should investigate other ethnic groups, countries, and people with different nancial statuses. Second, factors other than those covered in the questionnaires may have caused the participants to be underweight or overweight/obese, which were not examined in this study. Future studies should include other factors such as dietary habits, energy intake, and nutritional intake, particularly for people who are underweight.

Conclusions
In summary, this study found that 150 min of LTPA per week effectively lowered the odds of being underweight and overweight but was signi cantly reducing the risk of obesity (BMI ≥ 27 kg/m 2 ). Future studies may investigate the effects of a combined approach of physical activity and dietary control on obesity.
Abbreviations ACSM: American College of Sports Medicine; BMI: Body mass index; CATI: two-month telephone interview; CI: con dence interval; LTPA: Leisure-time physical activity; OR: odds ratio; SD: standard deviation; TNPAS: Taiwan National Physical Activity Survey.

Declarations
Ethics approval and consent to participate: The design and protocol of this study was approved by the Institutional Review Board of Fu Jen Catholic University in Taiwan (FJU-IRB C109085). The informed consent was obtained from all subjects involved in this study via CATI based survey. The present study was conducted according to the guidelines of the Declaration of Helsinki.
Consent for publication: Not applicable.
Availability of data and material: The raw data are not publicly available due to restrictions ethical.
Competing interests: The authors declare that they have no competing interests Funding: No funding was obtained for this study.
Author Contributions: C.L. participated in the design, conducted the statistical analyses, interpreted the data, and drafted the manuscript. S.-W. L. and P.-F. L. helped to manage and analyze the data, assisted in data interpretation. C.-C, H. supervised the study and critically reviewed and revised the manuscript. All authors have read and agreed to the published version of the manuscript.