Response to Reader's Queries.

Sir,

We are thankful to the reader for his interest in our paper and for the concerns raised. This was a noninferiority trial where we wanted to analyze whether the bupropion and iron-folate were as efficacious as ropinirole in the short-term treatment of Willis–Ekbom disease/Restless legs syndrome (WED/RLS). The consort diagram is provided with this letter [Figure 1]. As already mentioned, random allocation was done using a list of random numbers, and this has been reported to be an effective method.[1] Author who was responsible of measuring the improvement was blinded of the group allocation that removes the observer's bias. We agree with the statement mentioned that “both groups were neither superior nor equivalent to the ropinirole,” and we have already mentioned it in the paper “Pair-wise comparison depicted that ropinirole group differed from other two groups in International RLS Severity Rating (IRLS) score (F = 7.06; P = 0.001), which were comparable to each other” [Table 1].[2] As pointed out by the reader, in the absence of placebo control group, it cannot be commented whether bupropion or iron-folate combination was better than placebo. Placebo group could not be included because of restrictions from the regulatory bodies.[3] However, it must be noted that placebo is not an inert substance and this has been found effective in a number of disorders including WED/RLS.[45] In this study, mixed ANOVA was used, and in this statistical method, effect size was depicted using partial eta square. In this study, partial eta square value was 0.145 with 95% power for between-subjects’ effect, and thus, effect size may be considered large.[6] Intention to treat analysis, as mentioned by the reader has its own limitations. One of them is, subjects who do not receive treatment, are assumed to have taken treatment with resultant effect and this may have bearing on the results.[7] Thus, we removed five subjects who did not come for any follow up and the intention to treat analysis was done after including 8 subjects who attended at least one follow-up visit. We used last observation carry forward method of imputation. This method did not change the results as published in the initial paper.

Figure 1

Enrollment of subjects in the study

Table 1

Pair-wise comparison of the three groups

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.