A groundbreaking new study using artificial intelligence (AI) and machine learning has revealed the most crucial factors that keep people committed to their exercise routines: how much time they spend sitting, their gender, and their education level. Published in the prestigious journal Scientific Reports, this research analyzed health data from nearly 12,000 individuals, offering fresh insights into what helps people meet physical activity guidelines—a finding with important implications for Thailand, where sedentary lifestyles are increasingly common.
The importance of regular exercise can hardly be overstated. In Thailand, where modern life often means long hours at office desks or behind the wheel in Bangkok’s infamous traffic, the risk of inactivity-related diseases like diabetes and heart disease is on the rise. The Ministry of Public Health recommends that Thai adults get at least 150 minutes of moderate activity or 75 minutes of vigorous exercise each week, in line with global recommendations. Yet, as with many countries, a significant proportion of Thai people struggle to meet these benchmarks, raising pressing questions about how to encourage healthier habits.
Led by researchers from the University of Mississippi, this latest study used machine learning techniques to analyze a vast trove of public health data from the US National Health and Nutrition Examination Survey, spanning the years 2009 to 2018. After filtering out incomplete or disease-affected responses, the team examined 11,683 participants, looking for lifestyle, demographic, and health factors that predict who sticks with an exercise program. The top three factors—sedentary time, gender, and educational status—emerged as consistent, powerful predictors in the most accurate machine learning models.
“Physical activity adherence to the guidelines is a public health concern because of its relationship to disease prevention and overall health patterns,” explained Professor Minsoo Kang, a co-author of the study and an expert in sport analytics. By training AI models on such massive data sets, the team was able to identify patterns that would be too subtle or complex for traditional statistical approaches, allowing for more precise predictions and, potentially, more effective interventions (source: Neuroscience News).
Digging deeper, the researchers found that people who spend more time sitting are less likely to meet weekly exercise targets, a pattern seen worldwide, including in Thailand—where the rise of digital devices, remote work, and urban lifestyles have all contributed to more sedentary behavior. Gender also played a significant role, echoing previous studies that have found Thai women, on average, exercise less frequently than men; this could be due to cultural norms, concerns about safety when exercising outdoors, or unequal access to public sports facilities.
Perhaps most intriguing, the study found that education level was a powerful, independent predictor of exercise adherence. “I expected that factors like gender, BMI, race, or age would be important for our prediction model, but I was surprised by how significant educational status was,” commented Ju-Pil Choe, the study’s lead author. Unlike gender or age, which are innate, education is often shaped by social and economic circumstances, suggesting that interventions to promote lifelong learning could have ripple effects on public health (source: Neuroscience News).
For Thai readers, this finding resonates on several levels. Education is highly valued in Thai society—เครื่องแบบนักเรียน (student uniforms) are a symbol of national pride, and education is often regarded as a path to career success and social mobility. This new research suggests it may also be a pathway to better health. Policies that improve access to education, especially for marginalized populations, could have wide-ranging benefits—including increasing overall physical activity rates and reducing the burden of lifestyle-related illnesses.
The researchers also noted some limitations of their study, particularly the reliance on self-reported physical activity, which can lead to overestimates. While technology like smartwatches and step counters is growing in popularity in Thailand, more objective nationwide data could help future research sharpen these predictions. The use of machine learning, however, represents a major leap forward, enabling public health officials and policymakers to craft more targeted, individualized outreach programs.
The potential applications for Thailand are broad. By identifying those at highest risk of inactivity (for example, people in sedentary occupations, women with limited access to sports facilities, or adults with lower educational attainment), Thai policymakers—alongside community leaders, businesses, and educators—can develop programs designed for local realities. This might mean workplace wellness initiatives that incentivize taking breaks to stretch or move, expanding free or low-cost education opportunities, or ensuring safe, well-lit public parks for exercise, especially for women and children.
Historically, Thailand has championed mass-participation sports through events like the “Bike for Dad” and “Bike for Mom” campaigns, as well as grassroots movements such as early morning “aerobic dance” sessions in Bangkok’s Lumpini Park. Yet, the persistent disparities in who actually gets enough exercise point to deeper social determinants—factors now illuminated further by AI.
Looking forward, the authors suggest expanding the research by integrating more objective data sources and exploring factors like dietary supplements or other health behaviors. For Thailand, the rise of big data—combined with AI insights—could usher in a new era of “precision public health,” in which interventions are not just broadcast broadly but tailored to the specific needs of diverse communities.
For everyday Thai readers, the takeaway is clear: pay attention to how much time you spend sitting (หยุดนั่งนานๆ แล้วลุกขึ้นขยับตัวหน่อย!), seek out opportunities for education throughout life, and be aware of how cultural and gender norms may influence daily routines. If you’re struggling to start or maintain a workout habit, consider joining group activities, setting small, achievable goals, and leveraging digital tools to track progress and stay motivated. For policymakers and employers, the call is to reduce sedentary opportunities, promote inclusive access to sports and education, and tackle structural barriers that keep people inactive.
For further reading, see the original article at Neuroscience News, as well as the detailed research published in Scientific Reports by Lee, Choe, and Kang.