Page 137 - Comprehensive treatment of patients with glucocorticoid-dependent severe asthma

Author’s response:
We thank Dr Konstantinou for his interesting letter [1] in response to our paper on internet-based management of severe asthma [2]. He raises a pertinent question about how pragmatic is a study in which the patients have to give their consent to participate? We agree that asking patients to sign for their participation brings the study a step away from real life settings, but it would be unethical to perform an interventional study without asking the patient’s permission for randomisation and data to be collected. There are alternative approaches, proposed by Zelen in 1979 [3] and mainly used in emergency settings, consisting of post randomization consent. In such designs it is allowed for participants to refuse their allocated treatment or ‘crossover’ to any treatment arm. However this method is ethically very controversial and could bring some serious statistical drawbacks [4].
Another point issued by Dr. Konstantinou was a possible influence by the ‘Hawthorne effect” (i.e., a change in behaviour due to trial participation rather than treatment) on the outcomes of the study [5]. As suggested it would be ideal to verify the patient’s adherence to asthma treatment outside the context of a study. There are reports of objective assessment (blood levels of cortisol) of real life compliance to oral corticosteroid treatment demonstrating that up to 50% of asthma patients were not adherent to the prescribed medicine [6]. Compared to these data we indeed observed rather high levels of adherence, but given the fact that both groups in our study demonstrated similar adherence as well as comparable ratings for satisfaction with the treatment strategy, we do not believe that the “Hawthorne effect” played a major role in influencing the difference between internet-based and conventional management of severe asthma.
How to adjust for the ‘Hawthorne effect’?”
135
8