Study design and participants
The cross-sectional study design was applied in order to analyze the association between various physical fitness measurements and overweight/obesity risk. A secondary dataset from 2014-15 Taiwan’s National Physical Fitness Survey (TNPFS) was utilized for further analysis. 46 examination stations within 20 major cities or counties in Taiwan were responsible for the TNPFS data collection from October 2014 to March 2015. The TNPFS as well as these examination stations were supervised by the Sports Administration, Ministry of Education, Taiwan. The examiners of the TNPFS were qualified from the official training courses and capable for different types of physical fitness test (i.e., physical fitness test for adults aged 23 to 64, and for elder adults aged 65 years and above). All relevant data are contained as a secondary database and has been released for public research purposes.
At first, 25,271 data from Taiwanese elders aged 65 years and above were included in 2014-15 TNPFS. However, in order to focus on the overweight/obese and normal weight (as the reference group) population, the data exclusion has applied according to their BMI classification. Finally, data from 21,630 healthy Taiwanese elderly aged 65 to 96 years were reviewed for the present study.
Data collection
The TNPFS conducted the convenience sampling at each examination station. The data collection contained three different approaches. First, qualified examiners and medical specialists (usually nurse or doctor) were preliminary checked participants’ blood and resting heart rate, as well as assessed the potential safety risks by a structured questionnaire. All participants were required to pass the preliminary safety assessment then allowed to proceed next step. Second, participants were requested to fill (or verbally answer, if unavailable) the demographic questionnaire, as well as to complete the anthropometric measurements. After completed the second step, the participants were instructed for a light warming up (dynamic and static muscle stretching) around 10 minutes. Then, a series of physical fitness measurements were performed by the participants. Interval breaks are permitted for 2-4 minutes.
Measurements
Demographic characteristics
The demographic questionnaire included the questions such as age, gender education, and monthly income, marital status, and health-related life habits such as smoking, betel-nut chewing, and perceived health status, were recorded. Education level was divided into three categories: elementary or lower, junior or senior school, and college or higher. Monthly income level was divided into three categories: 20,000 NTD (New Taiwan Dollar) or under, between 20,001–40,000 NTD, and 40,001 NTD or above. Marital status was divided into three categories: married, never married, and divorced/separated/widowed. Self-reported health status was divided into three categories: excellent or good, fair, and very bad or poor. Smoking and chewing betel status were both divided into three categories: never, current, and former. This questionnaire was developed by Taiwan's Sports Administration, Ministry of Education, and has implemented in the annual nationwide survey for years. The same questionnaire has also reported and published previously [13].
Anthropometric assessment
In this study, anthropometric assessments, including body weight (kg), height (m), waist, and hip circumference (WC & HC). All participants were required to remove their shoes and heavy clothes during measuring. Body mass index (BMI) of the participants then calculated (kg/m2). The WC and HC were measured to the nearest 0.1 cm by a soft measuring tape at the natural waist and greater trochanter level. Thus, it allowed the present study to calculate the waist-to-hip ratio (WHR). The cut-off values for BMI were suggested by the Taiwanese Ministry of Health and Welfare. The cut-offs for underweight, normal weight, overweight, and obesity are 18.5, 24, and 27 kg/m2 [14]. However, as mentioned above, only normal weight, overweight, and obesity population were included for present study.
Physical fitness measurements
2-minute step test (reps), 30-second arm curl (reps), 30-second chair stand (reps), back scratch (cm), chair sit-and-reach (cm), one-leg stance with eye open (seconds), and 8-foot up-and-go (seconds) were measured to access the functional capacity among the participants. These measurements were representing individual aerobic endurance, muscle strength, muscle endurance, flexibility, and balance. The protocol of these measurements was conducted by the qualified examiners. Besides, the protocol of these measurements was mainly performed by the Senior Fitness Test manual [15], except the one-leg stance with eye open was performed according to the previous approach [16]. The results of these measurements were classified respectively into four quartiles for further analysis.
Statistical analyses
The Statistical Analysis System (SAS) software package (Version 9.4, SAS Institute Inc., Cary, NC) was used for data analysis. Chi-square tests and one-way analysis of variance (ANOVA) were used to analyze demographic characteristics and health-related physical fitness measurements among groups of participants. Tukey's post hoc tests were applied for those significant differences has found (p < 0.05) in order to determine the sequences among the groups. Significant differences between groups are considered as the potential confounders for the logistic regression model adjustment. Further, the normal weight population was appointed as the reference group for logistic regression analyses for general physical fitness test results as well as the quartiles of the results. Thus, the overweight and obesity risks (ORs) of each physical fitness performance were estimated, respectively. All values were expressed as means ± standard deviation, or the percentage (frequency). The significant level within each analysis was p < 0.05 to reject the null hypothesis and with a confidence interval (CI) of 95%.