Continuing assessment of the significance of the total point count of the answers given in each of these questionnaires led us to believe that it might be appropriate to compare these numbers with those of the non-addictive population, with those who are addictive in some way but who do not have an eating disorder, and with those who do have an eating disorder. Although I still use the figure of a total of twenty points as the rule of thumb indicator of clinical significance, we created the following table to give a more specific range of cut-off points, first for men (in bold print) and then for women according to the percentage of other men or women who do not have that level of problem. For example, men scoring between ten and twenty points, or women scoring between twenty-five and twenty-nine points, on the food bingeing questionnaire will be in the seventieth percentile i.e. only thirty per cent of men or women in the general population score at that leveL

My reason for generally sticking to a cut-off level of twenty points for significant problems is my clinical observation that there are differences between men and women in the addictive outlets that they choose. There maybe the same proportion of men and women who are addicts - probably ten per cent or so - but disproportionately more women than men have eating disorders. Therefore a man who has some anorexic behaviour may be relatively rare among men but still not have a significant eating disorder. Further, the distinctions are not as absolute as the table might indicate. There is no absolute distinction between people who binge and those who starve. Often the same people will binge or starve at different times. People suffering from bulimia, who binge and then vomit or purge, and others whose crash diets end in binges, often score highly in both the bingeing and starving questionnaires. Also, when patients binge on food they commonly also binge on shopping and spending. When they starve they also tend to exercise excessively. Both bingers and starvers may bury themselves in work There is a great deal of overlap so that individual variables get lost from view in a chart of overall averages.