**Background:** Prediabetes has more than doubled in prevalence among adolescents in the past 2 decades, now affecting approximately 40% of youth with obesity. Youth-onset type 2 diabetes leads to microvascular complications within the first 2 decades after diagnosis in approximately 80% of individuals. Because no medications have demonstrated reduction in progression from prediabetes to type 2 diabetes in youth, the primary treatment remains intensive lifestyle changes. According to the Health Belief Model, an individual's engagement in health-promoting behavior is driven by disease risk perception, barriers to and benefits of change, and cues to act such as awareness of a diagnosis. However, due to developmental differences, adolescents perceive and respond to health risks differently than adults. This study investigated whether diabetes risk perception, risk awareness, and potential barriers to behavior change were associated with diabetes risk-reducing health behaviors among adolescents at higher risk for diabetes based on elevated BMI.
**Methods:** This cross-sectional study included youths aged 12–17 years with BMI in the 85th percentile or higher for age and sex and available HbA1c measurement in NHANES waves 2011–2018. Patients with known diabetes and those who did not answer diabetes risk-related survey questions were excluded. The final cohort comprised 1,341 individuals representing 8,716,794 US youth. Risk perception was defined by the question "Do you feel you could be at risk for diabetes or prediabetes?" Risk awareness was defined as an affirmative response to having been told by a doctor about prediabetes or diabetes risk. Potential barriers included household food security, health insurance type, health care access, income, and household size. Outcomes included consumption of fast food, pizza, or non-home-prepared meals; physical activity and sedentary time; and frequency of attempted weight loss. Linear and logistic regression were used to evaluate associations, with separate models without and with adjustment for confounders (age, sex, race/ethnicity, BMI z-score, HbA1c).
**Key Results:** Mean age was 15.0 years (95% CI, 14.9–15.2) and mean BMI z-score was 1.76 (95% CI, 1.73–1.79). Elevated HbA1c (≥5.7%) was present in approximately 9% of the cohort (prediabetes range: 8.3% [95% CI, 6.5%–10.5%]; diabetes range: 0.3% [95% CI, 0.1%–0.7%]). Only 30.1% (95% CI, 23.1%–38.1%) of youth with elevated HbA1c reported feeling at risk for diabetes, and 26.5% (95% CI, 20.0%–34.2%) reported awareness of prediabetes or diabetes risk. In multivariable models adjusted for confounders, risk perception was associated with more TV watching (β=0.3 hours/day; 95% CI, 0.02–0.5) and 1.2 fewer days per week with at least 60 minutes of physical activity (β=−1.2; 95% CI, −2.0 to −0.4). Risk awareness was not associated with any physical activity, sedentary, or nutrition behaviors. Several barriers were associated with adverse outcomes: public insurance (vs private) was associated with approximately 20 fewer minutes/day of physical activity (β=−20.7; 95% CI, −35.5 to −5.8), and routine health care obtained through the emergency department was associated with 2.7 fewer days per week physically active (β=−2.7; 95% CI, −4.0 to −1.4). However, larger household size (≥5 vs 1–2 members) was associated with lower screen time (β=−1.1 hours/day; 95% CI, −2.0 to −0.3) and lower consumption of non-home-prepared meals (OR, 0.4; 95% CI, 0.2–0.7). Risk perception was associated with double the odds of attempted weight loss in unadjusted analysis (OR, 2.3; 95% CI, 1.0–5.1), but this association was no longer significant after adjustment for BMI z-score (aOR, 1.4; 95% CI, 0.6–3.5).
**Clinical Implications:** This study demonstrates that among US adolescents at higher risk for type 2 diabetes due to elevated BMI, diabetes risk perception and awareness are not associated with greater engagement in risk-reducing health behaviors. The finding that nearly 75% of youth with elevated HbA1c were unaware of their diabetes risk represents a substantial missed opportunity for early intervention. However, the paradoxical association between greater risk perception and less healthy behaviors (lower physical activity, more screen time) suggests that knowledge of risk alone is insufficient to motivate behavior change. The association between risk perception and weight loss attempts was explained by higher BMI rather than risk perception itself. These findings highlight the critical need to address structural barriers—including food insecurity, limited health care access, and economic disadvantage—rather than focusing solely on raising risk awareness. The study's limitations include its cross-sectional design (preventing causal inference), reliance on self-reported physical activity and nutrition, and inability to assess additional risk factors such as family history of diabetes, polycystic ovary syndrome, or hypertension. Future prospective studies are needed to evaluate causal relationships between risk perception and health behaviors, and to develop evidence-based interventions that identify and reduce barriers to effective lifestyle change in adolescents.