Reply to letter: Narrative approaches to reviews

We thank the authors for their interest in our recently published review article on non-operative therapies in low back pain.¹ As they have rightly pointed out, our study was a narrative review and is subject to limitations of the same. We agree that some of the literature may not have been covered in our review. Also, since our review was not a systematic one, many components of a systematic review such as a systematic search of the available databases, analysis of risks of bias of individual studies, assessment of heterogeneity among studies, etc. were not addressed. However, as mentioned in our review, whenever possible we used data from high-quality systematic reviews. Our review was primarily aimed to summarize the evidence provided in various systematic reviews each of which had adhered to all the guidelines of systemic reviews. The author also mentions about the lack of definition for using Grading of Recommendations Assessment, Development, and Evaluation (GRADE), as well as lack of definition for various outcome measures, comparator agents, and effect sizes. Defining outcome measures, interventions, effect sizes, etc. are relevant while conducting a review of individual trials. However, we had reported the findings as well as GRADE recommendations provided by multiple reviews. For example, for understanding the effect of education on low back pain we used the findings from Engers et al.² and Wood and Hendrick.³ Both these studies were high-quality reviews (not individual trials), which had summarized the findings of multiple studies, and our findings and recommendations were based on these studies. The aim of our review was not to summarize the evidence of different trials, rather summarize the findings of other systematic reviews. Therefore, synthesizing information from individual trials was not necessary when the included reviews have already evaluated in detail the individual topics/interventions. Finally, the authors point out that some of our references were old. We had tried to include the latest references at the time of preparation of the manuscript, and we agree that some studies may have been published after the preparation of the review. The study on pilates was a comprehensive Cochrane review by Yamato et al.,⁴ published in 2015, which is still the latest evidence available on Cochrane on pilates for back pain. However, we do recognize that new studies have been published regarding the topic and are likely to be published in the future. Therefore, periodic future reviews are required to include the updated evidence. Despite multiple limitations of our review, we do believe that our narrative review provides readers a summary of the findings and recommendations of various systematic reviews on interventions in back pain, and is helpful in guiding treatment decisions.