We were delighted that our systematic review aroused people's interest and consideration regarding the association between sarcopenia and depression.¹ Zhang et al. raised several points in their letter that we respond to below.²

First, we appreciate Zhang et al.'s careful reading and noting the article reported by Chen.³ Of course, we would not ignore it when screening the literature. However, the data of Chen and Wang's⁴ studies came from the same cohort, our team's Chengdu Yulin cohort in 2014. In Table 1, we compared the differences between the two studies. The prevalence of sarcopenia was different due to the use of different diagnostic criteria. We selected Wang's study in our analysis because it was based on the association between sarcopenia and depression, which was more relevant to our topic.

A cohort study conducted by Chen et al.⁵ aimed to examine risk factors for new-onset depressive symptoms over the course of a 1-year follow-up in community-dwelling older adults and found that sarcopenia was one of the risk factors for depression. However, patients with depression were excluded at baseline, and the prevalence of depression in sarcopenia could not be calculated and was excluded from our analysis. Although OR values could be obtained from both cross-sectional and cohort studies, there are essential differences between the two types of research. Cohort studies are prospective and confirmatory, confirming the relationship between cause and outcome, from cause to outcome. Cross-sectional studies explore the relationship between cause and outcome and are unable to distinguish cause and outcome. Due to the different natures of the studies, cohort studies cannot be combined with cross-sectional studies when systematic reviews are conducted.

Second, regarding heterogeneity, we have already discussed it in the original paper. Although Zhang et al. found that excluding two studies reduced the heterogeneity to 54.19%, which still indicates a large heterogeneity, excluding these two studies did not solve the problem. More rigorous large-sample real-world studies are needed to confirm these questions in the future.

Third, we thank you for pointing out the lack of assessment of publication bias. We have performed a funnel plot in our latest letter.⁶

我们很高兴我们的系统评价引起了人们对肌肉减少症和抑郁症之间关联的兴趣和思考。¹张等人。在他们的信中提出了几点，我们在下面作出回应。²

首先，我们感谢张等人的仔细阅读并注意到陈报道的文章。³当然，我们在筛选文献时也不会忽视它。然而，陈和王的⁴项研究的数据来自同一队列，即我们团队 2014 年的成都榆林队列。在表 1 中，我们比较了两项研究之间的差异。由于使用不同的诊断标准，肌肉减少症的患病率也不同。我们在分析中选择了王的研究，因为它基于肌肉减少症和抑郁症之间的关联，这与我们的主题更相关。

Chen 等人进行的一项队列研究。⁵旨在检查社区居住的老年人在 1 年随访过程中新发抑郁症状的危险因素，发现肌肉减少症是抑郁症的危险因素之一。然而，抑郁症患者在基线时被排除在外，肌肉减少症中抑郁症的患病率无法计算，因此被排除在我们的分析之外。虽然 OR 值可以从横断面研究和队列研究中获得，但两种类型的研究之间存在本质差异。队列研究是前瞻性和确认性的，确认原因和结果之间的关系，从原因到结果。横断面研究探讨了原因和结果之间的关系，无法区分原因和结果。由于研究性质不同，

其次，关于异质性，我们已经在原论文中讨论过。虽然张等人。发现排除两项研究将异质性降低到54.19%，这仍然表明异质性很大，排除这两项研究并没有解决问题。未来需要更严格的大样本真实世界研究来证实这些问题。

第三，我们感谢您指出缺乏对发表偏倚的评估。我们在最近的一封信中做了一个漏斗图。⁶