Active commuting reduces sociodemographic differences in adherence to recommendations derived from leisure-time physical activity among Brazilian adults

A Data-Driven Approach to Predicting Recreational Activity Participation Using Machine Learning

Purpose: With the popularity of recreational activities, the study aimed to develop prediction models for recreational activity participation and explore the key factors affecting participation in recreational activities. Methods: A total of 12,712 participants, excluding individuals under 20, were selected from the National Health and Nutrition Examination Survey (NHANES) from 2011 to 2018. The mean age of the sample was 46.86 years (±16.97), with a gender distribution of 6,721 males and 5,991 females. The variables included demographic, physical-related variables, and lifestyle variables. This study developed 42 prediction models using six machine learning methods, including logistic regression, Support Vector Machine (SVM), decision tree, random forest, eXtreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM). The relative importance of each variable was evaluated by permutation feature importance. Results: The results illustrated that the LightGBM was the most effective algorithm for predicting recreational activity participation (accuracy: .838, precision: .783, recall: .967, F1-score: .865, AUC: .826). In particular, prediction performance increased when the demographic and lifestyle datasets were used together. Next, as the result of the permutation feature importance based on the top models, education level and moderate-vigorous physical activity (MVPA) were found to be essential variables. Conclusion: These findings demonstrated the potential of a data-driven approach utilizing machine learning in a recreational discipline. Furthermore, this study interpreted the prediction model through feature importance analysis to overcome the limitation of machine learning interpretability.

Preliminary Results of a Bicycle Training Course on Adults' Environmental Perceptions and Their Mode of Commuting

This study was designed to analyze the effects of a bicycle training course on both adults’ environmental perceptions and their mode of commuting. Four bicycle training courses for adults were conducted in Granada, Spain in April 2015 and May 2016. The course program was focused on developing practical skills and attitudes on road. From the initial 65 adults who started the course, only 35 adults met the inclusion criteria and were included in the study. Participants completed twice (i.e., before and after the course) a questionnaire about their perceptions of the environment, usual mode of commuting to daily destinations, and sociodemographic characteristics. After finishing the initial questionnaire, the participants completed a bicycle training course based on the methodology “Bikeability” with a duration of 6 h. The results suggest that participants improved their safety perception in relation to the level of crime in the participants’ neighborhood after the bicycle training course. Cycling training courses should last longer in order to produce changes in the mode of commuting and in the environmental perceptions.

Levels of domain-specific physical activity at work, in the household, for travel and for leisure among 327 789 adults from 104 countries

OBJECTIVE

To compare the country-level absolute and relative contributions of physical activity at work and in the household, for travel, and during leisure-time to total moderate-to-vigorous physical activity (MVPA).

METHODS

We used data collected between 2002 and 2019 from 327 789 participants across 104 countries and territories (n=24 low, n=34 lower-middle, n=30 upper-middle, n=16 high-income) from all six World Health Organization (WHO) regions. We calculated mean min/week of work/household, travel and leisure MVPA and compared their relative contributions to total MVPA using Global Physical Activity Questionnaire data. We compared patterns by country, sex and age group (25-44 and 45-64 years).

RESULTS

Mean MVPA in work/household, travel and leisure domains across the 104 countries was 950 (IQR 618-1198), 327 (190-405) and 104 (51-131) min/week, respectively. Corresponding relative contributions to total MVPA were 52% (IQR 44%-63%), 36% (25%-45%) and 12% (4%-15%), respectively. Work/household was the highest contributor in 80 countries; travel in 23; leisure in just one. In both absolute and relative terms, low-income countries tended to show higher work/household (1233 min/week, 57%) and lower leisure MVPA levels (72 min/week, 4%). Travel MVPA duration was higher in low-income countries but there was no obvious pattern in the relative contributions. Women tended to have relatively less work/household and more travel MVPA; age groups were generally similar.

CONCLUSION

In the largest domain-specific physical activity study to date, we found considerable country-level variation in how MVPA is accumulated. Such information is essential to inform national and global policy and future investments to provide opportunities to be active, accounting for country context.

Active Commuting to University and its Association with Sociodemographic Factors and Physical Activity Levels in Chilean Students

Background and Objectives: Active commuting to and from university (ACU) could be a strategy to increase physical activity levels (PA) and promote health in young university students. We aimed to a) examine the patterns of commuting to university in Chilean students; b) the association between the mode of commuting to and from university and socio-demographic factors and PA-levels. Materials and Methods: A total of 496 university students (21.6 ± 2.4 years old) from two universities from Valparaíso (central coast of Chile) participated in this study. Personal data, home address, socio-economic status, PA, and the usual mode of commuting to and from the university were self-reported by a questionnaire. The commute distances were objectively measured using Google-Maps-software. Associations were examined using binary logistic regressions. Results: The main mode of commuting was by bus (to university: 55.2% vs. from university: 59.3%; p < 0.001). The least used mode was cycling (1.4% to and from university). Students living >5-km from university were less active commuters than those living in closer distances: (2-5 km, odds ratio (OR): 4.424, 95% and 95% confidence intervals (CI): 2.443-8.011, p < 0.001; 2 km, OR: 143.052, 95% CI: 55.154-371.030, p < 0.001). Students with low PA-levels were less active commuters than those with medium (OR: 1.446; 95% CI: 0.864-2.421; p = 0.160) or higher levels (OR: 1.880; 95% CI: 1.880-1.094; p = 0.022). Students who lived between 2 and 5 km, presented a significant association to be active commuters when they showed medium PA-levels (OR: 5.244, 95% CI: 1.358-20.246; p = 0.016). Conclusions: Chilean university students from Valparaíso are mainly passive commuters using public transport as the main mode of commuting to and from university; longer distances from home to the university are associated with low PA levels. ACU in distances between 2-5 km (mainly walking) could contribute to having medium PA-levels in Chilean university students. Thus, promoting the ACU walking to and from the university in such distances could be an effective strategy to increase the overall PA levels in Chilean university students.

Are Differences in Physical Activity across Socioeconomic Groups Associated with Choice of Physical Activity Variables to Report?

Despite being challenged in recent years, the hypothesis that individuals of higher socioeconomic status (SES) are more physically active than their lower SES counterparts is generally considered a fact. Recent reviews, however, have suggested that differences across groups might be related to which physical activity (PA) domains have been investigated. In the present review, searches for relevant studies were performed in the MEDLINE, ISI Web of Knowledge and SPORTDiscus databases. Search terms included “socioeconomic”, “socio-economic”, “socio economic” and “social class” to meet all variations of the variable “socioeconomic status” in combination with the term “physical activity”. Studies were included when applying the dimensions of intensity, frequency, type/mode, and duration in measuring PA. Fifty-six studies were included and were subsequently split into four PA domains: transport PA (TPA), occupational PA (OPA), housing PA (HPA) and leisure time PA (LTPA). It turned out that the positive relationship held only for LTPA, whereas the relationship was non-existent or even opposite for all other domains. It is concluded that the assumed positive relationship between SES and PA is mainly a relationship between LTPA and SES. It is further suggested that the PA domain should always be considered when studying said relationships.

Factors influencing interest in recreational sports participation and its rural-urban disparity

Objectives

Recreational sports are important leisure activities. However, most studies investigating barrier factors and motivation for participation in recreational sports have been limited to specific areas (e.g., a city or school) or demographic groups (e.g., adolescents). Therefore, this study set out to gain a more comprehensive understanding of the behavioral and socioeconomic factors influencing interest in recreational sports participation in Taiwan, as well as to evaluate the effect of any urban-rural divide.

Methods

This study analyzed data collected by the “Taiwan Social Change Survey” (program five, wave 3) “Leisure Life” questionnaire. We used hierarchical linear modeling to assess respondent interest in recreational sports participation and evaluated the influence of behavioral factors, socioeconomic factors, and residence location (urban/rural).

Results

Of the 2,146 participants in this study, 50.3% were male, and the average age was 43.9 years. Location of residence (urban/rural) accounted for 35.3% of the variation in interest in recreational sports participation, while the remaining 64.7% came from the individual level. Participants who lived in rural settings were less interested in recreational sports than their urban counterparts. Gender, educational attainment, participation frequency, health-motivated interest, and appearance-motivated interest were also associated with interest in recreational sports participation.

Conclusions

Different communication strategies may be needed to effectively reach different demographic groups. We suggest that future public health campaigns aiming to increase recreational sports participation include tailored interventions and messages to effectively encourage leisure physical activities among all, regardless of demographic boundaries.

Sociodemographic, health-related and lifestyle correlates of physical activity in adults with current asthma

OBJECTIVE

Regular physical activity is associated with better asthma control; however, little is known of the determinants of physical activity in a population of adults with asthma. Thus, the purpose of this study was to identify important sociodemographic, health and lifestyle correlates of physical activity among adults with asthma.

METHODS

Data from adults with asthma aged 18-44 years (n = 2740) from the Canadian Community Health Survey (CCHS), 2011-2012 annual component were used for analysis. The main outcome was Daily Energy Expenditure (DEE). This variable was based on frequency and duration of leisure activities in the past 3 months. Self-reported sociodemographic (age, sex, total household income, highest education level, and ethnicity), health-related (body mass index (BMI), self-perceived health, mood disorder, anxiety disorder, and asthma symptoms/attacks, past year) and lifestyle (type of smoker, sedentary time, fruit and vegetable consumption) variables were assessed as correlates for DEE. Data were analysed using linear regressions.

RESULTS

In the final model, sociodemographic correlates accounted for 4.2% of the variation in DEE. Adding health correlates increased the R² to 12.1%. Finally, adding lifestyle correlates increased the R² to 17.6%.

CONCLUSIONS

In adults with current asthma, the main correlates of physical activity are age, total household income, BMI, self-perceived health, anxiety disorder, sedentary time, and fruit and vegetable consumption. These data are important for informing education and exercise interventions for adults with asthma. Future research is needed to determine asthma-specific correlates of physical activity.