Article Text


Partial syndromes of eating disorders in adolescence are associated with adverse health outcome in adulthood
  1. Phillipa J Hay, FRANZCP, MD, DPhil
  1. School of Medicine University of Western Sydney, Sydney, Australia

Statistics from



What impact do partial syndromes of eating disorder in adolescence have on health and social adjustment in young adulthood?


2032 students selected using a two-stage cluster sample and followed up regularly into young adulthood. In the first wave, one class from each of 44 schools was selected at random. A second class was randomly selected 6 months later (wave 2). The average age of participants at the start of study was 14.5 years. Follow-up was at subsequent 6-month intervals during the remaining school years (waves 3–6) and at two stages in young adulthood (wave 7, aged 20–21 years; wave 8, aged 24–25 years).


Victoria, Australia; August 1992 to March 2003.

Prognostic factors:

Partial syndrome of eating disorder based on DSM-IV criteria for anorexia or bulimia on the Branched Eating Disorders Test. See online notes for further details on how anorexia and bulimia were defined.


Depression or anxiety (assessed on the Revised Clinical Interview Schedule (CIS-R) during waves 1–7 and the 12-item General Health Questionnaire (GHQ) at wave 8), persistent eating disorder (assessed at wave 7 and in females only), substance abuse (nicotine dependence defined as score ⩾4 on the Fagerstrom test; alcohol abuse or dependence assessed using the Composite International Diagnostic Interview), self-reported cannabis or amphetamine use in the past year, BMI, sexual activity, pregnancy, relationships and education.



Prospective cohort study.

Follow-up period:

10 years.


Prevalence of any partial eating disorder among males and females from waves 3–6 (average age 15.9–17.4 years) was 5.5%, 95% CI 4.3% to 6.7% (males 1.4%; females 9.4%). Compared to females without a partial eating disorder in adolescence, those with a partial eating disorder in adolescence had increased risk of: persistence of the disorder at age 20 (prevalence 15%; OR 6.7, 95% CI 2.0 to 22), psychiatric morbidity at age 20 (prevalence 44%; OR 3.3, 95% CI 2.1 to 5.2), psychiatric morbidity at age 24 (prevalence 40%; OR 2.0, 95% CI 1.2 to 3.1), being underweight BMI at age 24 (prevalence 38%; OR 2.1, 95% CI 1.3 to 3.3), having alcohol diagnosis at age 24 (prevalence 19%; OR 1.8, 95% CI 1.0 to 3.4), alcohol dependence at age 24 (prevalence 16%; OR 3.3, 95% CI 1.0 to 4.6), amphetamine use at age 24 (prevalence 17%; OR 2.5, 95% CI 1.2 to 5.0), sexual activity before age 16: prevalence 34%, OR 2.5, 95% CI 1.5 to 3.9; sexual activity by school leaving age: prevalence 57%; OR 2.1, 95% CI 1.3 to 3.5), having left school before year 12 (prevalence 25%; OR 2.2, 95% CI 1.2 to 4.1), having neither a qualification nor be studying (prevalence 34%; OR 1.8, 95% CI 1.1 to 2.9); and not having a degree (prevalence 29%; OR 0.43, 95% CI 0.27 to 0.69).


Partial eating syndromes are common in females during adolescence and are associated with a range of adverse health and social outcomes in young adulthood.


Patton GC, Coffey C, Carlin JB, et al. Prognosis of adolescent partial syndromes of eating disorder. Br J Psychiatry 2008;192:294–9.


Current diagnostic schemes define three groups of eating disorders (EDs): anorexia nervosa (AN), bulimia nervosa (BN) and EDs not otherwise specified (EDNOS) in the DSM-IV.1 The former are well-defined EDs, but are much less common than EDNOS in most clinic samples and in community surveys.2 Longitudinal studies have focused largely on AN, and to a lesser degree BN, and have mostly taken the form of follow-up of clinical case series and other treatment samples in adults.3 It has been unclear how much EDNOS reflects partially recovered ED or EDs in evolution, the latter of note in younger people where early intervention may thus help prevent adult EDs. The findings of a community sample of 55 adolescent girls, aged 14–15 years with partial AN or BN syndromes over a 6-year period, suggested these EDs were self-limiting and, because of the high prevalence of depression and anxiety at follow-up, possibly a “variant” of affective disorder.4 The present paper builds on this with a 10-year follow-up of a community cohort of 1943 adolescents aged around 14–15 years.

This is a complex paper but with two important findings. First, partial AN and BN occurred in nearly 1 in 10 girls aged 15–17 years, and second, these appeared to be a psychologically vulnerable group, both with regards to poorer functional outcomes (poorer education and more frequent teenage pregnancy) and psychiatric morbidity (including increased depression and alcohol dependence). There was little evidence however of progression to full AN or BN syndromes. Young women with ED symptoms may benefit less from specialised eating disorder services and more from broadly-based programmes.

The paper alerts clinicians to an at-risk group of young women with partial AN or BN, but rather than specialist interventions, it supports the development and evaluation of programmes that promote psychological resiliency, improved educational and functional outcomes, and address a range of psychological vulnerabilities likely to be important such as low self-esteem.

View Abstract

Supplementary materials


  • Source of Funding: National Health and Medical Research Council, the Victorian Health Promotion Foundation, and the Victorian Centre for Excellence in Eating Disorders.

  • Additional notes and a reference list are published online only at


  • Competing interests: PH has been sponsored by AstraZeneca to provide a lecture at an educational meeting for psychiatrists.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.