A major UK cohort study uncovers widespread eating disorder symptoms among diverse adolescents, exposing urgent gaps in recognition, cultural care, and long-held assumptions about who is truly at risk.
Study: The Prevalence and Sociodemographic Correlates of Eating Disorder Symptoms in a Diverse School-Based Population Study. Image Credit: Jack_the_sparow / Shutterstock
*Important notice: Preprints with The Lancet / SSRN publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
In a recent study posted to the SSRN Preprints with The Lancet server*, researchers investigated associations between sociodemographic characteristics and eating disorder (ED) symptoms in adolescents.
EDs are psychiatric disorders linked to extensive psychiatric and physical comorbidity and high mortality. ED incidence in primary care in the UK has increased over the past two decades, especially since the COVID-19 pandemic, and in young adolescents aged 11–16, a period of peak ED incidence.
A UK survey found that 13% of people aged 11–16 met the probable ED threshold in 2022. However, these estimates were from the Development and Well-being Assessment’s ED module, which is susceptible to false positives and shows moderate agreement with clinical diagnoses. Other UK estimates have focused on clinical samples, limiting generalizability.
About the study
In the present cross-sectional study, researchers examined the prevalence of ED symptoms and their associations with sociodemographic characteristics in UK adolescents. Data from the “Born in Bradford Age of Wonder” cohort of school pupils aged 12–15 were used, which has a high proportion of South Asian participants. Data were collected from September 2023 to July 2024 using two sequential questionnaires, with all ED measures drawn from the first survey. The EDE-QS collected data on ED symptoms.
EDE-QS captured the frequency of key symptoms, including binge eating, restrictive eating, and weight/shape concerns. Probable ED was defined as an EDE-QS score ≥ 15, a screening cut-off validated in adult community samples, and sensitivity analyses were performed using a slightly lower cut-off to define probable ED (≥ 13). Because item 10 depends on reporting loss of control at item 9, missing responses on item 10 were set to zero when item 9 indicated no loss of control. The authors noted that the psychometric performance of EDE-QS in ethnically diverse early adolescents has not been fully validated.
Probable ED was defined using an EDE-QS score ≥ 15. ED symptom distribution was described by sociodemographic characteristics: sex, school year group, ethnicity, area-level deprivation, and gender identity. Area-level deprivation was analyzed using the Index of Multiple Deprivation 2019.
Participants self-reported sex as male or female as assigned at birth and gender identity as female, male, or non-binary; there was also a free-text option to further describe gender identity. The researchers created a gender identity variable based on these free-text descriptions and responses: gender diverse (gender identity differed from sex assigned at birth) or cisgender (gender identity aligned with sex assigned at birth). Probable ED prevalence by gender identity was not stratified by sex due to small numbers.
The team estimated the proportion of participants meeting the probable ED threshold by sociodemographic variables. Sex-stratified probable ED prevalence was reported by area deprivation and ethnicity. Linear regression was performed to explore associations between sociodemographic characteristics and EDE-QS scores.
Findings
Overall, 9,498 individuals completed at least one survey. Among these, 7,434 had complete data and were included in analyses. Most participants were female (51.9%), Asian (45.9%), and lived in the most deprived areas (60.9%). The median EDE-QS score was four, and more than 17% of participants had an EDE-QS score ≥ 15. Probable ED prevalence was higher in females (23.5%) than in males (10.7%) and among gender diverse participants (37.9%) than in cisgender individuals (17.1%).
Moreover, gender diverse individuals and females had higher EDE-QS scores than cisgender participants and males, respectively. Asian participants had lower EDE-QS scores than White individuals. Among females, probable ED prevalence was the highest among White subjects, while among males, it was the highest in those of other ethnicities. Probable ED prevalence was lower in the least deprived areas.
Participants from the most deprived areas had higher EDE-QS scores than those in the least deprived areas. Probable ED prevalence was higher among females and males in the most deprived areas than in the least deprived areas. Moreover, males and females from the most deprived areas had higher EDE-QS scores compared to those from the least deprived areas. The most frequently reported symptoms were dissatisfaction with shape or weight and a strong desire to lose weight.
Compulsive exercise was the most frequently reported ED symptom among males, whereas dissatisfaction with shape or weight was the most common in females. Laxative use or self-induced vomiting were the least common symptoms in males and females. Gender diverse participants had higher endorsement of all symptoms than their cisgender peers. All ED symptoms were less prevalent in Asian females compared to White females.
Asian, Black, or other ethnic males had a higher prevalence of fasting and dietary restriction than White males. Females from the most deprived areas showed a higher prevalence of binge eating, fasting, purging, and dietary restriction than those from the least deprived areas. Males from the most deprived areas had a higher prevalence of nearly all symptoms. In sensitivity analyses, associations were consistent with those in the primary analysis.
Conclusions
Together, there was a high prevalence of ED symptoms in UK adolescents. Shape and weight dissatisfaction were common across sociodemographic groups. Ethnic minority males had more ED symptoms than White males, whereas White females had more symptoms than their Black and Asian peers. The findings challenge the misconception that EDs predominantly affect affluent White populations, highlighting a need to address inequalities in recognition and care. Overall, the results suggest the need for culturally sensitive strategies to prevent and treat EDs, accounting for the interplay between ethnicity and sex. The authors also note that EDE-QS psychometrics are not fully established in ethnically diverse young adolescents and that the cohort’s composition may affect generalizability.
*Important notice: Preprints with The Lancet / SSRN publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.