Treatment of PTSD

host of clinical studies which he mixes up (like apples and pears) with the handful of community based studies that exists.One of the few methodo logically acceptable community based studies of anorexia nervosa (see Treasure, 1990; Patton & King, 1991) â€"¿ the one performed in Gà ¶teborg, Sweden in the l980s (Râstamet a!, 1989) â€"¿ was excluded from Fombonne's analysis for â€oe¿ obvious reasons―.It seems these â€oe¿ obvious reasons― were (1) that the material of the Goteborg study was presented in sufficient detail to allow specific analysis of whether DSMâ€"III or DSMâ€"IIIâ€"R criteria applied, (2) that partial syndromes â€"¿ later meeting full DSMâ€"III--R criteria (Gillberg et al, 1994) â€"¿ were included as a separate group in the original study, and (3) that the birth-cohort was followed up for a few years leading to the appear ance of new cases.The findings were presented in a way which has made it possible for Fombonne to calculate all sorts of rates needed for a thorough review.He himself complains that several studies have not provided enough information about the diagnostic criteria used, and that few authors have looked at cohorts in a longitudinal fashion, so we had some difficulty understanding what was so â€oe¿ obvious― about the reasons for excluding this study.The prevalence rate of anorexia nervosa in the community based studies was considerably higher than the median rate calculated by Fombonne.Again, this should not be taken as evidence that there has been an increase in prevalence rate over the years.However, it is essential that conclusions be based on the most reasonable data sets rather than those that, according to the standards set out by the author of a review/meta-analysis, are less than adequate.