Copied from the New Zealand thread This is the paper she is referring to: The pathway from glandular fever to chronic fatigue syndrome: can the cognitive behavioural model provide the map? I think the paper is worth a discussion because it is one of the few papers that actually tried to test the CBT-model using a prospective design. My first impression is that they didn't correct for multiple comparisons, so many of the 'significant' results in the individual logistic regression analysis such as perfectionism should probably not have been reported as such. When all these factors were tested in a Multivariate logistic regression analysis along with gender, age and symptoms, only 'all-or-nothing behavior' remained a significant predictor. With correction for multiple comparisons, this would also be questioned. So I don't think these results are robust. The sample was also a bit small with only 17 EBV-patients who developed CFS after 6 months.