The study, published in Discover Psychology, was led by Leo C. M. Cheung of the Sports Nutrition & Monitoring Centre at the Hong Kong Sports Institute, working alongside colleagues from the institute and the Department of Sports Science and Physical Education at The Chinese University of Hong Kong. The team administered the Eating Attitudes Test-26, or EAT-26, a twenty-six item self-report instrument that has served as the workhorse of eating-attitude screening since the 1980s. Respondents rate statements about food preoccupation, dieting behavior, and eating-related guilt on a six-point scale, and a total score of twenty or above conventionally flags a need for clinical follow-up. The instrument does not diagnose anorexia nervosa or bulimia, but it reliably identifies individuals whose attitudes toward food have drifted into territory associated with those conditions.

Statistically, the headline number is the overall at-risk proportion of 4.0 percent. Beneath that aggregate, however, the researchers uncovered a series of gradients that tell a more interesting story than any single prevalence figure could. Female athletes registered a risk proportion of 7.1 percent, roughly four times the 1.7 percent observed in male athletes, a sex difference consistent with the broader epidemiological literature in which women carry a substantially higher burden of eating pathology. Yet even the female figure remained modest, suggesting that the structured environment of Hong Kong’s elite sport system, with its professional nutrition support, may buffer some of the pressures that push other populations toward restriction.

The more striking patterns emerged when the researchers stratified their sample by dieting history, age, and self-perceived body image. Athletes who reported previous dieting experience scored significantly higher on the EAT-26 than those who had never dieted, with mean scores of 8.08 plus or minus 9.03 compared with 4.92 plus or minus 5.73, a difference that reached statistical significance at the five percent level. This finding matters because dieting attempts are widely understood as a gateway behavior; in athletes, a single episode of deliberate caloric restriction can spiral into chronic under-fueling, and under-fueling in turn drives the syndrome known as Relative Energy Deficiency in Sport, which degrades bone density, endocrine function, and performance itself.

Age produced an equally compelling signal. Athletes under nineteen years old scored significantly higher than their older counterparts, with mean EAT-26 values of 9.24 plus or minus 11.21 versus 5.33 plus or minus 5.71. Pearson correlation analysis confirmed a weak but significant negative association between age and EAT-26 score, with a correlation coefficient of -0.234 and a p-value below 0.05. In plain terms, the younger the athlete, the more disordered the eating attitudes tended to be. The authors point to this developmental window as a critical target for intervention, and the logic is sound: adolescence is precisely the period when body image consolidates, when sport-specific pressures about weight class and aesthetics intensify, and when nutritional habits harden into lifelong patterns.

Body image proved to be the strongest single correlate in the dataset. Participants rated their self-perceived body image on a nine-point scale, and those scoring below the midpoint of five showed markedly elevated EAT-26 scores, averaging 11.08 plus or minus 7.88, compared with 5.41 plus or minus 6.99 among athletes who rated their body image at five or above. The difference was significant at the one percent level, and Pearson analysis revealed a moderate negative correlation of -0.344 between body image and EAT-26 score with a p-value below 0.001. Athletes who felt worse about their bodies, in other words, consistently harbored more disturbed attitudes toward eating, independent of the sport they played.

To disentangle the overlapping effects of age and body image, the team ran a multiple linear regression, and both variables survived as independent negative predictors of EAT-26 scores. Age carried a standardized coefficient of -0.228 with a p-value below 0.05, while self-perceived body image carried a standardized coefficient of -0.344 with a p-value below 0.001. Because both predictors point in the same direction, the regression supports a coherent risk profile: the athlete most likely to exhibit problematic eating attitudes in this sample is young, has experimented with dieting, and holds a negative view of her own body. That profile gives sports medicine staff something actionable, because every one of those characteristics can be screened for during routine athlete monitoring.

Methodologically, the study is a descriptive, cross-sectional survey, and the authors classify it as Level V evidence, the lowest rung on the clinical evidence ladder. Cross-sectional designs capture a single moment in time, which means they can identify associations but cannot establish whether poor body image causes disordered eating or vice versa, nor can they track how risk evolves across an athlete’s career. The sample size of 100, while respectable for a single-institute study, is small enough that subgroup estimates, particularly the 1.7 percent figure for male athletes, rest on a handful of individuals. The researchers themselves frame the work as a foundation for tailored health education programs and differentiated screening rather than as a definitive prevalence estimate.

Even with those caveats, the study adds a valuable data point to a literature that the authors describe as frequently contradictory. Prevalence estimates for disordered eating among athletes vary wildly across studies, driven by differences in sport type, assessment tools, cultural context, and how risk is defined. High-risk sports such as gymnastics, distance running, and weight-class combat disciplines consistently show elevated rates, but pooled figures are notoriously unstable. By anchoring the Hong Kong data to a standardized instrument and a defined at-risk threshold, the study gives regional sport bodies a benchmark against which future surveillance can be measured, and it suggests that the often-cited narrative of an eating-pathology epidemic in elite sport may require regional calibration rather than blanket assumptions.

The practical implications ripple outward from the institute itself. Screening programs that incorporate the EAT-26, a simple body-image rating, and a question about dieting history could be embedded into existing athlete health checks at negligible cost, flagging young athletes with poor body image for early nutrition counseling before attitudes calcify into behavior. The research, which received no external funding and was approved by the Research Ethics Committee of the University of Hong Kong School of Professional and Continuing Education, emerged from an MSc project in Sport and Exercise Nutrition, a reminder that some of the most immediately useful findings in sports science come from practitioners working directly with athletes rather than from large-scale trials. For coaches and parents watching a teenage athlete quietly shrink portions or fixate on body shape, the study’s message is direct: age, dieting history, and body dissatisfaction together form the earliest warning system, and acting on that warning early is far easier than treating an eating disorder once it takes hold.

Subject of Research: Disordered eating risk factors among Hong Kong Chinese athletes

Article Title: A cross-sectional study of eating attitudes in Hong Kong athletes

Article References: Cheung, L. C. M., Sze, J. W. H., Siu, F. P. L., Li, S. L., Pun, G. T. K., Lau, J. T. Y., O’Reilly, J., & Kuo, M. J. (2026). A cross-sectional study of eating attitudes in Hong Kong athletes. Discover Psychology. https://doi.org/10.1007/s44202-026-00848-8

Image Credits: AI Generated

DOI: 10.1007/s44202-026-00848-8

Keywords: disordered eating, EAT-26, Hong Kong athletes, body image, sports nutrition, eating disorder risk, cross-sectional study, adolescent athletes, dieting behavior, sport psychology, Relative Energy Deficiency in Sport, athlete screening