A common claim in weight stigma research is that stigma sets off a vicious cycle: stigma-related stress and distress are thought to lead to unhelpful eating and avoidance behaviours, those to weight gain, and weight gain to more stigma. A study published on 19 January 2026 followed 345 young adults in Hong Kong for a year to see whether that cycle shows up in the data. It did not. The more interesting part is why the result is smaller than it sounds.
The paper, “Exploring how weight stigma relates to psychological distress, physical activity, and eating behaviors over time: a longitudinal study among young adults in Hong Kong”, is by Xavier C. C. Fung, Benson W. M. Lau and colleagues. It appeared in the Journal of Eating Disorders (volume 14, article 50, open access). The authors say empirical tests of the cycle are scarce, especially among East Asian populations.
Who took part
Participants were recruited by convenience sampling at a Hong Kong university. They were 18 to 30 years old, read Traditional Chinese and gave consent. People with self-reported neurological disorders, functional impairments or psychotic conditions were excluded. Data came from four online surveys, four months apart, between March 2021 and June 2022. The study was not pre-registered.
345 valid participants entered the first wave. Follow-up rates for the later waves were 73.3, 67.5 and 67.8 percent. The mean age was 22.9, and 63.2 percent were women. The mean BMI was 21.15, and 21.2 percent met the cutoff for overweight or obesity that the authors use for Asian adults (BMI of 23 or higher). By my arithmetic that is roughly 70 of the 345 people.
What was measured
Perceived weight stigma was a 10-item yes/no scale of stigmatising situations, such as being treated less politely because of weight. Weight-related self-stigma, meaning internalised negative beliefs about one’s own weight, was an 11-item scale. The survey also covered psychological distress (anxiety and depression, 14 items), self-reported physical activity, three eating-behaviour scales, perceived control over exercising and over avoiding “inappropriate eating behaviours”, and self-reported height and weight for BMI.
Two statistical models were used. A parallel-process growth model asks whether the trajectories of the variables move together. A cross-lagged model asks whether earlier self-stigma predicts later BMI, and earlier BMI predicts later self-stigma.
The findings
Stigma and self-stigma moved together. People with higher perceived stigma at the start also had higher self-stigma (standardised coefficient 0.58), and those whose perceived stigma rose over the year also saw self-stigma rise (0.25). Self-stigma was linked to lower perceived control over avoiding “inappropriate eating behaviours”, both in level and in change over time (change: -0.26).
Distress. Perceived stigma was linked to distress at the starting level (0.55). Changes in the variables over the year were not otherwise linked to changes in distress, physical activity, eating behaviour or BMI. Of the seven hypotheses, the authors report one supported, one partly supported and five not supported.
The cycle test. Higher self-stigma was followed by slightly lower BMI at the next wave (coefficients between -0.11 and -0.14, each p = 0.032). Earlier BMI did not predict later self-stigma (coefficients around zero). The authors conclude that the vicious cycle was not identified, and that this challenges the assumption that more stigma leads to more weight.
Reading it carefully
The negative link between self-stigma and later BMI should not be read as stigma lowering weight. The effect is small, the authors name possible unexamined factors, and the mean BMI barely moved across the four waves (21.15, 21.13, 21.32, 21.16). A cycle that depends on weight gain is hard to detect in a group whose weight hardly changes.
The sample is the larger limit. A mean BMI near 21 and about one in five people above the overweight cutoff means the study says little about people in larger bodies, which is who the cycle hypothesis is usually about. That is my reading of the numbers, not a claim the authors make. They list their own limits: self-reported measures, a small sample of mostly university students, one year of follow-up, and data collected during the COVID-19 pandemic in Hong Kong, when restrictions changed exercise and eating habits.
Self-stigma was also very stable. The authors report correlations of about 0.8 between waves, so there was little change for the models to explain. A three-month Taiwanese study the paper cites found the opposite direction, higher self-stigma going with higher BMI; the authors point to that study’s short duration as a limitation. The data are available only on request.
This article reports the design and findings of one study. It does not evaluate any treatment or weight-related intervention.
Why it belongs here
The stigma-and-weight cycle is often quoted as settled. This paper tested part of it, found no support in this sample, and still found the part that does not depend on weight: perceived stigma goes together with self-stigma, and self-stigma goes together with feeling less in control. Those links are associations, not proof of cause, but they do not depend on what the scale says.
Germany and the United Kingdom are not covered by this data.
Source: Fung XCC, Siu AMH, Ye J, Chen JH, Chen JS, Bevan N, Griffiths MD, Lin CY, Lau BWM. “Exploring how weight stigma relates to psychological distress, physical activity, and eating behaviors over time: a longitudinal study among young adults in Hong Kong.” Journal of Eating Disorders 2026;14:50, doi.org/10.1186/s40337-026-01525-w. Full text read on 2 October 2026.