London School of Hygiene & Tropical Medicine

The school a girl attends can affect her chance of being diagnosed with an eating disorder, according to new research carried out by a joint UK-Swedish team and published in the International Journal of Epidemiology.

Researchers from Oxford University, UCL, the University of Bristol, the London School of Hygiene & Tropical Medicine and the Karolinska Institute in Stockholm used routinely collected data from Sweden to take account of individual factors that would make someone more likely to develop an eating disorder. Even after accounting for these factors, there were still differences in the rates of eating disorder according to the school attended.

Girls attending schools with higher proportions of female students, and high proportions of university-educated parents were more likely to be diagnosed with an eating disorder than girls at schools with lower proportions of female students and fewer university-educated parents.

Eating disorders, including anorexia nervosa, bulimia nervosa, binge eating disorder, and other eating disorders which are not as easily classified, affect 5.7% of adolescent girls. They are serious illnesses, and someone with bulimia nervosa is around twice as likely to die young as someone without it, while someone with anorexia nervosa is about six times more likely to die young.

Dr Helen Bould, Child and Adolescent Psychiatrist at the University of Oxford, Department of Psychiatry, led the research. She said: "Eating disorders have an enormous effect on the lives of young people who suffer from them - it is important to understand the risk factors so that we can address them. 

"For a long time clinicians in the field have noted that they seem to see more young people with eating disorders from some schools than others, but this is the first empirical evidence that this is the case."

The research team accounted for factors as diverse as parental income, whether parents had a history of mental ill health, parental education, the number of siblings and birth weight among others. Even allowing for all these characteristics, there were still variations between schools.

Bianca DeStavola, Professor of Biostatistics at the London School of Hygiene & Tropical Medicine, who contributed to the study, said: "This study is an example of the research potentials of linking large administrative and health record data in a country like Sweden, which has a very long tradition in using e-Health records in research.

"It allowed us to investigate why eating disorders seem to be more frequent among girls attending certain schools rather than others. We found that there were characteristics of the schools, in particular the percentage of girls and of highly educated parents, that contributed to the observed clustering of cases, even after accounting for the girls family circumstances and own characteristics.

"This study cannot tell us the reasons for this - it could be due to the pressurised environment of some schools, or it could also be a greater awareness of eating disorders in these schools and hence detection of cases.

Sweden does not have any single sex schools due to its strict laws on gender equality, so it is difficult to apply the findings to the UK education system, where there are selective all-girls schools that are likely to have a high proportion of highly educated parents.

Prof DeStavola said: "Extrapolation of these findings to the UK should be taken very cautiously given the differences in education system, and we are repeating this investigation with UK-based data."

Publication:

Helen Bould, Bianca De Stavola, Cecilia Magnusson, Nadia Micali, Henrik Dal, Jonathan Evans, Christina Dalman, Glyn Lewis, The influence of school on whether girls develop eating disorders, International Journal of Epidemiology. DOI: 10.1093/ije/dyw037