The lifetime prevalence of AN in our sample was somewhat higher than the prevalence reported among female Australian (Wade et al., 2006) and Finnish (Keski-Rahkonen et al., 2007) twins (3.6% versus 1.9% and 2.2%, respectively), but well in line with estimates that AN affects between 1%–4% of the population (Hudson et al., 2007; Smink, van Hoeken, & Hoek, 2012). This increased prevalence could be due to the use of a broader AN definition in our study. The prevalence of SA was lower than that reported in a large nationally representative US sample [4.6%; (Kessler, Borges, & Walters, 1999)]; however, it should be noted that SA in the present study only reflected attempts resulting in clinical care and thus less serious attempts that did not warrant medical attention would not have been detected, yielding a conservative estimate of SA. Although the lifetime prevalence of MDD in our sample was higher (26.6%) than in the US sample [16.6%; (Kessler et al., 2005)], which could reflect true differences or variable estimates due to different assessment strategies (i.e., self-report versus interview), it was similar (25.1% in women) to that reported in another Swedish twin sample (Kendler et al., 2006).
