The natural 30-year outcome of adolescent-onset anorexia nervosa: a prospective controlled community-based study

Project: Research

Research areas and keywords

UKÄ subject classification

  • Psychiatry


  • child and adolescent psychiatry, eating disorders, anorexia nervosa, neurodevelopmental disorders, epidemiology, clinical outcomes


Anorexia Nervosa is a serious condition with long-lasting influence on central aspects of life – mental and somatic health, as well as social functioning.

The anorexia nervosa outcome study was initiated by Maria Råstam and Christopher Gillberg in 1985, with the aim of conducting an epidemiological study on teenage-onset anorexia nervosa. The individuals were identified through community screening. All 4291 individuals born in 1970 and attending the 8th grade at junior high school 1985 in Gothenburg were screened for anorexia nervosa. 25 individuals constituted the total age cohort. All but one girl were pooled together with 27 anorexia nervosa cases born in adjacent years also identified by the Göteborg school nurses. All 51 individuals (48 women, 3 men) were psychiatrically examined in the original study together with 51 peers with no eating problems, matched for sex, age and school attended.
The epidemiological study showed that the prevalence of anorexia nervosa at 15 years of age was 0.7 %. The accumulated population prevalence for anorexia nervosa up to 18 years of age was 1.08 % for girls and 0.09 % for boys, equivalent to a girl:boy ratio of 11.6:1.

The anorexia nervosa group and the comparison group have been examined on four occasions, initially at 16 years of age and subsequently in follow-up studies at 21, 24 and 32 years of age in a prospective, controlled, very long-term outcome study of anorexia nervosa. Research questions include a population-based estimation of prevalence of teenage anorexia nervosa, a study of background factors relative well-matched healthy peers, the course of the eating problems, environmental and genetic risk factors, and psychosocial functioning, physical and psychiatric health in adult years. The overall anorexia nervosa study has pointed to important biological factors in anorexia nervosa, predictors for long term psychosocial outcome, and the notion of anorexia nervosa as a developmental disorder.

One unexpected finding was that the participants in the anorexia nervosa group had significantly more problems with social cognition and obsessions in early childhood according to their mothers. Such traits were enhanced during their eating disorder and persisted in many cases after recovering from the eating disorder per se. Low age of anorexia nervosa onset, premorbid obsessional traits, and autism-like symptoms in childhood predicted a poor socio-psychological outcome according to the global functioning scales used in this study

Ongoing study

In a collaboration between Lund University and the University of Gothenburg, the research group recently closed our 30-year follow-up study of the same sample where 96 % agreed to participate at age 44. We have assessed mental health, neuronal injury, education, employment, mortality, inpatient care, pharmaceutical consumption, social assistance, health economics, quality of life, and offspring’s mental health. We have used validated instruments, biochemical markers, eye-tracking technique and registers. Collection of data closed in November 2016, and are now being analysed.

This is the only prospective controlled community-based very long-term follow-up study of anorexia nervosa. The healthcare system misses half the anorexia nervosa cases and one in five never seeks psychiatric care. The study will enable us to give a true reflection on the long term outcome of teenage-onset anorexia nervosa with respect to health, brain function, health economics, and offspring status.

Research group

Elisabet Wentz (EW), Maria Råstam (MR), Nouchine Hadjikhani (NH) (visiting professor at the Gillberg Neuropsychiatry Centre and Associate Professor at Harvard Medical School) designed the study. EW, Carina Gillberg (CG), PhD-student Sandra Rydberg Dobrescu (SRD) and PhD-student Lisa Dinkler (LD) examined the participants. Bio-markers will be analysed by Professor Henrik Zetterberg (HZ) at the Department of Psychiatry and Neurochemistry, University of Gothenburg. Professor in Health Economics Kristian Bohlin (KB) (co-advisor for SRD) conducts the register-study. LD, NH and EW are analysing data on eye-tracking and emotion recognition. LD is the research coordinator. MR is co-advisor of SRD and LD. The research group has a long lasting collaboration with Nils-Gunnar Pehrsson, Statistiska konsultgruppen, Göteborg.


Collaboration concerning the anorexia nervosa sample with Professor Michael Lowe, Drexel University, Philadelphia, focus on the importance of weight suppression, premorbid weight, and eating disorders. Child psychiatrist Sören Nielsen, Denmark, has performed earlier analyses based on the Morgan Russell scales, and will analyse outcome data from the 30-year follow-up study.
Effective start/end date1983/09/01 → …