In their letter,Reference Twisk and Corsius¹ Twisk & Corsius raise several issues with respect to our study testing the efficacy of internet-based cognitive–behavioural therapy (iCBT) for chronic fatigue syndrome (CFS).Reference Janse, Worm-Smeitink, Bleijenberg, Donders and Knoop² We will respond point by point to the most important issues raised with respect to efficacy and safety of iCBT.

First, Twisk & Corsius suggest that many patients who improved following iCBT had depression. All patients met Centers for Disease Control and Prevention criteria for CFS. Psychiatric comorbidity that could explain the presence of fatigue, including a depressive disorder, was ruled out using the Mini International Neuropsychiatric Interview (MINI). A subgroup reported clinically significant levels of depressive symptoms (Beck Depression Inventory for Primary Care score of ≥4) but only about 10% met the criteria of any depressive disorder (MINI). Far more patients, about 40%, reported fatigue scores within the normal range following iCBT. Furthermore, 44% of patients without significant levels of depressive symptoms reported fatigue levels in the normal range. These data show that patients with CFS without depressive symptoms also improve following iCBT and that it is not true that many patients who improved had depression.

Second, Twisk & Corsius state that iCBT is insufficient to achieve normal levels of fatigue. We used a cut-off score of 35 or higher on the Checklist Individual Strength fatigue subscale for severe fatigue. A recent study showed that this score discerns between severe fatigue and ‘fatigue levels in the normal range’.Reference Worm-Smeitink, Gielissen, Bloot, van Laarhoven, van Engelen and van Riel³ A substantial subgroup of patients score below this cut-off after iCBT. For these patients iCBT is sufficient to achieve normal levels of fatigue.

Third, Twisk & Corsius state the effects of iCBT on objective measures are not reported.

CFS is diagnosed on the basis of symptoms and functional impairments reported by patients and the fact that there is no known medical condition that can explain the presence of symptoms. To determine if an intervention for CFS is effective one needs to determine if symptoms and functional impairments have decreased. For this, self-report measures were used. To assess symptoms one has to rely on self-report.

Finally, Twisk & Corsius state iCBT contains graded exercise and is unsafe. Graded activity is an element of CBTReference Knoop and Bleijenberg⁴ and is aimed at increasing levels of activity. Graded exercise is also aimed at increasing the fitness of a patient, this is not the case in graded activity. The safety of CBT for CFS has been studied, analysing data from seven randomised controlled trials.Reference Janse, Worm-Smeitink, Bleijenberg, Donders and Knoop^2, Reference Heins, Knoop, Prins, Stulemeijer, van der Meer and Bleijenberg^5–Reference Nijhof, Bleijenberg, Uiterwaal, Kimpen and van de Putte⁷ There was no evidence suggesting that CBT for CFS is unsafe.

We thank Twisk & Corsius for their feedback and for providing us with the opportunity to clarify our findings that show that iCBT for CFS is an efficacious and safe intervention.