Erratum to ’Comment on “The effect of timing of remotely supervised exercise on glucose control in people with type 1 diabetes during Ramadan: A randomised crossover study” by Ozairi E et al.’ [Diabetes Metabol Syndr: Clin Res Rev 18 (2) (February 2024) 102967] (Diabetes & Metabolic Syndrome: Clinical Research & Reviews (2024) 18(2), (S1871402124000286), (10.1016/j.dsx.2024.102967))

Aima Asim, [email protected]. Ziauddin Medical College, Karachi, Pakistan. Author links open overlay panelSatesh Kumar. Shaheed Mohtarma Benazir Bhutto Medical College Lyari Karachi, Pakistan. Author links open overlay panelMahima Khatri. Dow University of Health Sciences, Karachi, Pakistan. The publisher regrets that the authors’ response was missed in the letter. Authors’ Response to Letter by Aima Asim et al. We thank the authors for their letter regarding our study [1] and for the opportunity to respond to their insightful comments. The authors suggest that there may be attrition bias within our study due to differences in withdrawal from respective groups. The current study was a crossover study and so there are no groups per se, rather there was a randomisation to the order in which they performed afternoon and evening exercise. In the analysis there is, therefore, balance in the comparisons between afternoon and evening exercise. There were no dropouts in the first period of the sequence and 8 in the second, with 5 doing evening and 3 afternoon exercise, which we contend is a small difference. The reasons for withdrawal were due to other commitments, such as work, family, and religious duties during Ramadan, which meant that people struggled to attend the exercise sessions. Overall we suggest the risk of attrition bias is, therefore, small. The authors suggest we should broaden our exclusion criteria to exclude participants, for example, taking glucocorticoids. We respectfully disagree with this suggestion. The strategy in the current study was to maximise external validity of the study by having broad inclusion criteria whilst ensuring participant safety. The authors also suggest that a longer wash out period may be required, or that we should have directly tested this prior to the study. In our opinion we do not think this is a major issue as there is a host of data, in a variety of populations, which shows that the effects of acute exercise on glucose control generally last up to 72 hours post exercise [2,3]. We are happy the authors note our limitations section particularly around the challenges in measuring intensity in this pragmatic study. Our choice of RPE was based on balancing logistics, participant burden and validity – the latter having been shown for RPE in a variety of populations e.g. Ref. [4]. We appreciate the suggestion of using the category ratio-10 scales but there is no data, to our knowledge, which has shown it can enhance the validity of RPE to estimate exercise intensity in a similar population as we have studied. Whilst this may have been suggested in the reference provided by the authors, no data was given to support the suggestion, and we prefer to base choices on robust evidence. Nonetheless we acknowledge again that RPE to quantify exercise intensity is not perfect and has limitations, but this represents a pragmatic choice in our study. We agree on the importance of carbohydrates and insulin use. As presented in the paper there were no differences in overall insulin dosage during afternoon and evening exercise weeks. We had some issues in the current study in accurately quantifying carbohydrate intake, although participants self-reported no overall difference, and we acknowledge this as a limitation and is something we aim to address in future work. The publisher would like to apologise for any inconvenience caused.