Comment and reply on: A randomized crossover trial investigating the ease of use and preference of two dry powder inhalers in patients with asthma of chronic obstructive pulmonary disease

Click to increase image sizeClick to decrease image size

Declaration of interest

The author states no conflict of interest and has received no payment in preparation of this manuscript.

Author's response

Affiliation

Job van der Palen^†1, Paul van der Valk², Martijn Goosens³, Karin Groothuis-Oudshoorn⁴, Marjolein Brusse-Keizer¹

^† Author for correspondence

¹ Medical School Twente, Medisch Spectrum Twente, Haaksbergerstraat 55, Enschede, 7513 ER, The Netherlands [email protected]

² Department of Pulmonary Medicine, Medisch Spectrum Twente, Enschede, The Netherlands

³ Gelre Ziekenhuizen, Zutphen, The Netherlands

⁴ University Twente, Health Technology and Services Research, Drienerlolaan 5, Enschede, The Netherlands

We would like to respond to this letter that according to the writer has the purpose ‘to demonstrate that this [our] conclusion is misleading since it is based on serious oversights of the design of the study’. The author is referring to our basic conclusion: ‘With DK fewer errors were made, more patients preferred DK over EH and patients were more satisfied with DK’.

The author is correct in assuming that the ‘instructions of usage’ were translated from English to Dutch. We would like to thank the writer of this letter for pointing out our error in not including ‘visual verification of dose uptake’ in our error checklist. What is not clear from the manuscript is the exact content of the instruction sheet. This was translated from English and did include the statement on ‘visual verification of dose uptake’. Therefore, the patients knew that they could verify whether or not the dose had been successfully inhaled.

Below, I will summarize the consequences of our omission of adding ‘visual verification of dose uptake’.

Our primary outcome was the percentage of patients making one or more critical error after reading the insert, while the secondary outcomes were: the percentage of patients making one or more critical errors after the first instruction by the trainer; the number of instructions needed to demonstrate a perfect inhalation technique; overall patient satisfaction; and inhaler preference. If one would consider visual verification of dose uptake as a critical error, then the percentage of patients making a critical error with the Elpenhaler could only go up. In the unlikely event that no patient would make this error, the difference in critical error rate would remain 17% for Diskus versus 35% for Elpenhaler (p = 0.001). Any extra patient making an error with the item ‘visual verification of dose uptake’ would further enlarge the difference in favor of the Diskus. The same is true for critical error rate following instruction by the trainer and the number of instructions needed.

The effect of adding the visual inspection item to the checklist will not have influenced our third secondary outcome, which is overall patient satisfaction, because this was included in the instruction sheet. The same probably holds true for our last secondary outcome, which is inhaler preference, in which the observed difference is very large, with 73.2% of patients preferring Diskus versus 26.8% preferring Elpenhaler (p < 0.001).

Furthermore, the author of the letter states: ‘…however, it can be questioned whether patients were really naive to DK, since—as stated in Table 2—disease history ranged between 3 and 13 years. In addition, 72.6% of patients suffered from COPD and fluticasone–salmeterol combination is the drug of choice for this disease’. On this topic we can be clear. All the included patients were truly naive to the Diskus. Contrary to the belief of the author, there are many patients that do not use the Diskus for the fluticasone–salmeterol combination in the Diskus.

In a further comment, the role of familiarity of the instructors with the Diskus and not the Elpenhaler was addressed: ‘Doctors and nurses were also naive to EH and not naive to DK; the instructions of medical doctors and nurses to the patients for the correct use of EH relied on the content of the insert and not on their real-life clinical experience with the EH device, compared to DK’. This is undoubtedly true, but would not have made any difference to our primary outcome: ‘the percentage of patients making one or more critical errors after reading the insert’.

Finally, the author questions the methodology of the study, ‘… since current medical practice for asthma and COPD patients requires the training of the patients before regular use of the inhalation device’. We do not agree that the methodology of the study is incorrect. One can only criticize us for the choice of primary outcome parameter. We chose assessment of critical errors after reading the instructions only – not because we feel this is how it should be done in practice but because we often encounter referred patients who have never been instructed. Our paper shows that this is an unacceptable practice, but more so for some inhalers than for others.