Definitions of multi-impulsivity and sample characteristics between studies of impulsivity vary widely leading to inconsistent results. Outcome data was examined to assess the comparative response of multi-impulsive and non multiimpulsive bulimic patients to a supervised self-help CBT programme. Multiimpulsivity was defined as the presence of two or more impulsive behaviours in the last 2 months in addition to bulimia nervosa as reported with the MultiImpulsivity Scale. Multi-impulsive patients reported similar levels of bulimic behaviours but higher levels of depression at pre-treatment. They achieved similar levels of change by the end of treatment. However, despite improvements, their levels of depressive symptoms remain high and bulimic symptoms subclinical and may be at higher risk of relapse.

Low self-esteem is widely recognized as a predisposing, precipitating and maintaining factor in the aetiology of eating disorders. This paper examines the effectiveness of a self-esteem group for women with eating disorders, delivered repeatedly over a 20-month period. Fortyone patients began and 33 completed the programme and both preand post-measures. Eight were excluded from the final analysis as they were in concurrent psychological treatment. Outcome was assessed using standardized measures and records of symptom levels and drop-out rates. Significant improvements in self-esteem, depression, and eating attitudes were observed. It is concluded that a treatment focusing solely on self-esteem can be beneficial in reducing eating disordered attitudes.

The study investigated factors influencing failure to engage among a group of bulimic and binge eating disordered clients referred to a supervised self-help programme. A total of 125 patients referred to a supervised self-help programme for people with bulimia and binge eating disorder were grouped into those who were seen in treatment ( n ¼ 85), and those who failed to engage i.e. did not attend their initial appointment ( n ¼ 40). Diagnosis, age, gender and waiting times of the two groups were compared. Diagnosis, age and gender were not significantly different between the two groups. The only significant factor measured was the length of time that they had waited between being referred to the service and their appointment date. Clients who waited less than 4 weeks were 2.4 times more likely to attend their appointment. For every week that lapsed between referral and appointment date, the odds of attending were reduced by 15% ( p ¼ 0.002). The results are discussed in light of the importance of maximizing engagement in treatment.

A detailed description of internet research methodology, an exploration of the demographics of an internet recruited sample and reflections upon what cyber-research may mean for traditional, pencil-and-paper sampling.

No abstract