Introduction

Type 2 diabetes mellitus (T2DM) is an escalating public health concern, now affecting not only adults but increasingly adolescents – largely due to rising rates of obesity and physical inactivity. The global surge in childhood and adolescent overweight has been a key driver of this earlier onset [1]. Adolescents, in particular, undergo physiological and behavioural changes that increase their susceptibility to insulin resistance and other metabolic disorders, placing them at greater risk.

Modifiable lifestyle factors, including low physical activity levels, central adiposity, and impaired glucose metabolism, are significant contributors to the development of T2DM in adolescents [2]. Physical activity has a protective function through increasing insulin sensitivity, enhancing body composition, and lowering blood glucose levels [3]. Excessive adiposity, particularly in the abdominal region, correlates with impaired glucose tolerance (IGT) and increased cardiometabolic risk [4]. Frequent fast-food consumption has been shown to correlate with higher body mass index (BMI) in adolescents, with studies indicating that adolescents who eat fast food more often have a greater risk of being overweight or obese [5]. The school offers nutritionally balanced meals via a structured catering system. However, students in numerous boarding schools in East Java are permitted to order food from external vendors, including fast food and other less nutritious alternatives.

The relationship among physical activity, adiposity (as indicated by BMI and waist circumference), urinary glucose excretion, frequency of fast-food consumption, and 2-hour plasma glucose (2-hPG) levels in adolescents has not been thoroughly investigated, despite the acknowledged importance of these factors. Boarding schools offer a unique setting in which lifestyle behaviours can be carefully observed and evaluated. This study investigated the correlation among physical activity, adiposity, urinary glucose, 2-hPG, frequency of fast-food consumption and the risk of T2DM in adolescent girls within a private boarding school setting.

Aim of the study

This research examined the association of physical activity, central adiposity, and fast-food consumption with 2-hPG levels, as indicators of T2DM risk, in adolescent girls attending a boarding school in East Java.

Material and methods

A cross-sectional study was conducted at a private boarding school involving adolescent female students as participants. The research involved 127 participants selected via total sampling methodology. The study included adolescent girls aged 12 to 18 years who were enrolled and residing at the private boarding school throughout the study period. Eligibility for inclusion was determined by the participants’ physical ability to complete anthropometric measurements and to undergo a 2-hPG test after an oral glucose load. Students with established chronic medical conditions that may affect glucose metabolism, including type 1 or type 2 diabetes, thyroid disorders, or Cushing syndrome, were excluded from the study. Exclusion criteria were the use of medications that influence glycaemic regulation, such as systemic corticosteroids or metformin, the presence of acute illness or fever during data collection, and the inability to complete the oral glucose tolerance test (OGTT).

Adiposity was assessed using body mass index (BMI) and waist circumference measurements. The calculation of BMI entails dividing weight (kg) by height squared (m²) and categorising the results into underweight, normal, overweight, and obese classifications according to age- and sex-specific percentiles, as specified by the World Health Organization growth references [6].

Waist circumference was measured in centimetres (cm) at the midpoint between the lower edge of the last palpable rib and the top of the iliac crest using a non-elastic measuring tape, following a standardised anthropometric protocol [7]. The Physical Activity Questionnaire for Adolescents (PAQ-A) was used to evaluate physical activity levels, functioning as a validated self-administered tool for assessing general activity over the previous week [8]. The Activity Questionnaire (PAQ) has been translated into Indonesian and assessed for validity and reliability by Andriyani et al. [9], yielding a validity coefficient of 0.881 and a reliability coefficient of 0.745. The Indonesian version of the PAQ exhibited moderate validity and acceptable reliability [10]. Participants were categorised into four levels of physical activity – low, low-moderate, moderate, and moderate-to-vigorous – according to PAQ-A scores and total weekly hours of physical activity.

Blood glucose levels were assessed using a 2-hPG test following a standard 75 g oral glucose load. An oral glucose tolerance test should ideally be performed following an overnight fast. Although participants were encouraged to avoid any caloric intake before testing, strict adherence to a standardized overnight fasting period could not be guaranteed in the boarding school setting. Fasting plasma glucose measurements were therefore excluded from the analysis in order to reduce potential misclassification and bias associated with unclear fasting status.

The oral glucose challenge was successfully completed by every participant, and 2-hPG measurements were collected in all cases. IGT was defined according to the American Diabetes Association criteria as a 2-hPG level of 140–199 mg/dl, normal glucose tolerance as < 140 mg/dl, and diabetes-range hyperglycaemia as ≥ 200 mg/dl [11]. Due to its proven validity and sensitivity in detecting early dysglycaemia in adolescents, especially under uncertain fasting conditions, 2-hPG was chosen as the main glycaemic outcome for this study.

Urine dipstick analysis was conducted to identify glucose and protein in urine, which may suggest glycosuria or early renal involvement, and functions as a screening tool for metabolic disturbances [12]. Urine dipstick testing was conducted as a component of the standard health screening protocol in the boarding school clinic. Participants received instructions on mid-stream urine collection from trained nurses to reduce contamination. Proteinuria was incorporated solely as a screening parameter, given that elevated urine protein may correlate with metabolic risk in adolescents. Nonetheless, it was not a primary outcome variable in this research. Participants reported their frequency of fast-food consumption through the Youth/ Adolescent Food Frequency Questionnaire – a semi-quantitative tool developed for individuals aged 9–18 years, demonstrating acceptable validity and reliability in evaluating dietary intake among adolescents. The test–retest reliability coefficients varied between r = 0.58 and r = 0.68, indicating its applicability in adolescent populations [13, 14]. This variable was subsequently analysed in conjunction with BMI to evaluate associations.

All statistical analysis was carried out using IBM SPSS Statistics 25.0 (IBM Corp., Armonk, NY, USA). Before conducting any comparisons, the distribution of each continuous variable was checked with the Shapiro-Wilk test, as this approach is well suited for samples of the size used in the study. Variables that followed a normal distribution were summarised as mean and standard deviation, while skewed data were expressed as median values together with their range. Depending on how the data were distributed, differences between groups were examined with either the independent samples t-test or the Mann-Whitney U test. Categorical data were evaluated with the χ² test. Associations among 2-hPG, physical activity, waist circumference, and BMI were explored through correlation analysis, using Pearson or Spearman coefficients, as appropriate. In addition, average waist circumference was compared between participants with normal glucose tolerance and those with IGT using an independent samples t-test. Statistical significance was set at p < 0.05.

Results

Among the 127 adolescent girls enrolled, 1.6% were classified as obese and 10.2% as overweight according to the BMI classification. Sixty-three percent of the participants exhibited a normal BMI. The average 2-hPG value among participants was 110.46 mg/dl. The characteristics of the included participants are shown in Table I.

Urinalysis results indicated that 1% of participants exhibited positive findings for glucose and protein, whereas 1.6% demonstrated proteinuria classified as +1. No association was identified between proteinuria and anthropometric or glycaemic parameters; detailed analyses are not provided. Based on PAQ-A scores, 10.2% of participants had low physical activity, 18.1% had low-moderate activity, 46.5% had moderate activity, and 12.6% engaged in moderate-vigorous activity. Correlation analyses indicated a significant relationship between physical activity and 2-hPG levels (r = 0.205; p = 0.021). No significant correlation was observed between waist circumference and 2-hPG (r = 0.041, p = 0.643) or between BMI and 2-hPG (r = 0.032; p = 0.723), as shown in Table II. The analysis revealed a significant correlation between BMI and the frequency of fast-food consumption, indicated by a weak negative correlation (r = –0.208, p = 0.020). Correlation analyses demonstrated no significant association between the frequency of fast-food consumption and 2-hPG levels (r = 0.038; p = 0.674). A total of 16 students (12.6%) had 2-hPG levels in the range 140–199 mg/dl (indicative of IGT). These students had significantly larger waist sizes (p = 0.03). However, there was no significant relationship between BMI categories and 2-hPG levels (p = 0.445). All participants were found to have at least one risk factor for T2DM.

Discussion

This study investigated the association between physical activity, adiposity (measured by BMI and waist circumference), urine glucose, and blood glucose levels in adolescent girls within a boarding school environment. The results indicate the existence of early risk factors for T2DM within this population, even among younger individuals. While the majority of participants exhibited a normal BMI, around 11.8% were categorised as overweight or obese. This aligns with global trends indicating a rising prevalence of adolescent obesity, particularly in structured settings such as boarding schools [15]. Eating habits in boarding school adolescents are shaped by structured meal provision, peer influence, nutritional knowledge, and access to unhealthy foods. In developing countries such as Indonesia, rapid dietary shifts toward fast food and low physical activity amplify obesity and diabetes risks. The burden is particularly pronounced compared to other developing nations due to urbanization [5]. Developed countries benefit from better resources but face similar challenges with fast food and eating disorders [16]. Interventions focusing on education, healthy meal planning, and physical activity are critical to improving eating habits and reducing the associated health risks, particularly in adolescents vulnerable to metabolic disorders [17, 18].

Central adiposity is acknowledged as a risk factor for impaired glucose regulation. However, this study did not identify a statistically significant correlation between waist circumference and 2-hPG. Adolescents with IGT demonstrated a slightly larger waist circumference, although these differences are relatively small and require careful interpretation. Prior studies have shown that visceral adiposity has a stronger correlation with insulin resistance compared to BMI alone [19, 20]. However, these patterns were not clearly evidenced in this analysis. Skinfold thickness measurements, such as subscapular or triceps skinfold, primarily indicate subcutaneous fat and have demonstrated greater accuracy in predicting total body fat compared to BMI in adolescents. However, they are not reliable indicators of visceral adipose tissue [21, 22].

A statistically significant but weak correlation exists between 2-hPG and physical activity levels. Recent evidence indicates that physical activity is more closely associated with glycaemic regulation in adolescents than BMI. A two-year longitudinal study of children and adolescents with diabetes indicated that increased levels of moderate-to-vigorous physical activity and decreased sedentary time correlated with lower interstitial glucose and glycated hemoglobin levels. This suggests that movement behaviours significantly impact glycaemic stability, independent of body weight [23]. The magnitude of the association in the present study was small, and self-reported PAQ-A scores may be influenced by recall or social desirability bias. Teenagers most commonly participated in physical activities at school by climbing stairs to their classes and walking between buildings for religious activities.

The assessment that 12.6% of participants exhibited IGT, defined as a 2-hPG level of 140–199 mg/dl, is important to note. Adolescents with IGT are at increased risk of developing T2DM in the absence of intervention. Recent longitudinal studies indicate that early detection and lifestyle interventions, particularly in high-risk youth, can prevent or postpone the onset of diabetes [19, 24].

Urine dipstick testing identified glucosuria and proteinuria in a limited number of participants. The presence of glucosuria, while not diagnostic, may indicate transient hyperglycaemia or initial renal involvement. Dipstick screening is a straightforward, non-invasive, and economical approach that could facilitate the early detection of metabolic disturbances in educational environments [25]. These findings should be interpreted with caution, as traces of protein may reflect hydration status, exercise, or menstrual contamination.

A weak negative correlation exists between BMI and the frequency of fast-food consumption, suggesting that individuals who consumed fast food more frequently tended to have slightly lower BMI. This finding aligns with the study by Poudel et al. [26], which indicates that fast-food consumption is associated with lower BMI in adolescents. It suggests that the timing of fast-food consumption is critical, as eating fast food after school is more commonly linked to underweight status. However, given the weak strength of the correlation, this finding should be interpreted with caution and may be influenced by confounding variables.

Kirana and Wirjatmadi [5] observed that regular intake of fast food, which is characterised by high levels of calories, fat, and sugar, correlates with an increased risk of overweight in young individuals. The discrepancy in our findings may be due to underreporting or social desirability bias caused by social stigma or inaccurate recall. Individuals with higher BMIs may underreport their fast-food consumption, leading to potential reverse causality. Adolescents aware of their overweight status may mitigate this condition by reducing fast-food consumption [26]. Van Erpecum et al. [27] found that the availability of healthy food options and neighbourhood socioeconomic status significantly influenced BMI more than the frequency of fast-food access. Adolescents with elevated BMI may engage in efforts to reduce their food intake, whereas those with lower BMI may exhibit less dietary regulation and consume more fast food in a boarding school environment [28].

Fast-food consumption is associated with impaired glucose metabolism, attributed to its high glycaemic load and poor nutritional quality. This intake correlates with elevated fasting glucose levels and an increased risk of metabolic syndrome [29]. This study identified no significant correlation between fast-food consumption frequency and 2-hPG levels (r = 0.038; p = 0.674). This discrepancy may indicate factors including inconsistent portion sizes, variations in food composition, individual insulin sensitivity, compensatory dietary behaviours during other meals, or underreporting influenced by social desirability bias. Additionally, fast-food consumption is recognised as a significant factor in the development of obesity and cardiometabolic risk among adolescents [28, 30]. The influence on glucose tolerance within a controlled setting, such as a boarding school, may be moderated by the overall dietary composition and levels of physical activity. This finding highlights the complex relationship among diet, lifestyle, and metabolic health, indicating that the frequency of fast-food consumption may not be a reliable predictor of increased 2-hPG in this particular population.

This study’s notable strength is its use of objective anthropometric measurements and a standardized 2-hour OGTT, which is infrequently performed in adolescent populations. The boarding school environment provided a regulated context for the observation of dietary and activity patterns. The selection of 2-hPG as the principal glycaemic outcome requires further explanation. Fasting plasma glucose is frequently used in metabolic screening. However, post-challenge glucose measurements have demonstrated reliability and sensitivity in identifying IGT in children and adolescents, particularly in the early stages of dysglycaemia [11]. This is particularly important in practical contexts where maintaining strict overnight fasting proves challenging, such as in boarding school settings.

Several limitations should be acknowledged. The cross-sectional design does not allow for causal inference. The sample was sourced from a single educational institution, which restricts generalizability. Data regarding pubertal stage, total energy intake, and family history were not collected, which may impact glucose metabolism. Self-reported physical activity can be influenced by recall and reporting biases. The lack of complete fasting glucose data restricted comparisons with 2-hPG and reduced the capacity to analyse glycaemic trends.

Conclusions

This research identified early indicators of metabolic risk in adolescent girls, such as central adiposity, IGT, and decreased physical activity levels. Students with IGT exhibited significantly larger waist circumference compared with those with normal glucose tolerance (p = 0.03). However, the associations observed were relatively small and warrant cautious interpretation, given that most correlations lacked statistical significance. No correlation was identified between BMI and the frequency of fast-food consumption, indicating that alternative lifestyle or environmental factors may have a more significant influence on glucose regulation. The findings underscore the necessity for early school-based initiatives that promote regular physical activity and healthy nutritional practices to enhance metabolic health in adolescents.