Lifetime and 12-month prevalence of eating disorders amongst women in mid-life: a population-based study of diagnoses and risk factors

Eating disorders (EDs) are common amongst women; however, no research has specifically investigated the lifetime/12-month prevalence of eating disorders amongst women in mid-life (i.e., fourth and fifth decade of life) and the relevant longitudinal risk factors. We aimed to investigate the lifetime and 12-month prevalence of EDs and lifetime health service use and to identify childhood, parenting, and personality risk factors. This is a two-phase prevalence study, nested within an existing longitudinal community-based sample of women in mid-life. A total of 5658 women from the UK Avon Longitudinal Study of Parents and Children (ALSPAC; enrolled 20 years earlier) participated. ED diagnoses were obtained using validated structured interviews. Weighted analyses were carried out accounting for the two-phase methodology to obtain prevalence figures and to carry out risk factor regression analyses. By mid-life, 15.3% (95% confidence intervals, 13.5–17.4%) of women had met criteria for a lifetime ED. The 12-month prevalence of EDs was 3.6%. Childhood sexual abuse was prospectively associated with all binge/purge type disorders and an external locus of control was associated with binge-eating disorder. Better maternal care was protective for bulimia nervosa. Childhood life events and interpersonal sensitivity were associated with all EDs. By mid-life a significant proportion of women will experience an ED, and few women accessed healthcare. Active EDs are common in mid-life, both due to new onset and chronic disorders. Increased awareness of the full spectrum of EDs in this stage of life and adequate service provision is important. This is the first study to investigate childhood and personality risk factors for full threshold and sub-threshold EDs and to identify common predictors for full and sub-threshold EDs. Further research should clarify the role of preventable risk factors on both full and sub-threshold EDs.