Discussion 是你这项研究真正赢得学术贡献的地方。它要说明你的结果意味着什么、与已有知识的关系如何、边界在哪里、又为什么重要。一个扎实的 Results 后面接一个薄弱的 Discussion，审稿人读完会产生这样的判断：“数据本身没问题，但作者似乎并不理解自己发现了什么。”
下面这五类错误，出现在医学与生命科学稿件的 Discussion 中最为频繁。每一类都会削弱这一节的解读价值，并给审稿人留下要求大修的理由。
一、复述结果，却没有真正解读
Discussion 最常见的错误，是开头就把 Results 重讲了一遍。作者换几个词把自己的发现总结一下，却没有加入任何新的解读，也没有与已有文献对比，更没有机制层面的说明。审稿人一眼就能看出来：如果 Discussion 的第一段读起来像 Results 的浓缩版，那作者其实还没有开始”讨论”。
常见错误写法：
In this study, we found that patients in the treatment group had significantly lower HbA1c levels compared with the control group (6.8% vs. 7.9%, p < 0.01). The treatment group also showed improved fasting glucose levels and reduced insulin resistance.
修改后：
The reduction in HbA1c levels observed in the treatment group (6.8% vs. 7.9%, p < 0.01) is consistent with prior reports of SGLT2 inhibitor efficacy in type 2 diabetes management. Notably, the magnitude of this reduction (1.1 percentage points) exceeds the 0.5-point change commonly accepted as clinically meaningful in glycemic control, suggesting that this intervention may offer benefits beyond those achieved by existing first-line therapies.
修改后的版本从同一个结果出发，但立刻把它放进了语境中：与既有文献对比，量化其临床相关性，并提示它对治疗决策可能的意义。这才是解读。
二、回避与自己结论相矛盾的研究
当已有研究报告了与你结论相冲突的结果时，回避这些文献并不会让冲突消失。审稿人熟悉这个领域的文献，如果一项高被引研究给出了相反的结论，而你的 Discussion 只字未提，审稿人要么认为你根本不知道这项研究，要么认为你在刻意回避一处不利的对比。这两种判断都会削弱稿件的可信度。
直面相悖证据不是软肋，而是科研成熟度的体现。
常见错误写法：
Our results are consistent with previous studies showing that high vitamin D levels are associated with reduced fracture risk in postmenopausal women (Smith et al., 2020; Lee et al., 2021).
（完全没有提及 VITAL 试验或其他未发现显著关联的大型研究。）
修改后：
Our results are consistent with observational studies reporting an inverse association between vitamin D levels and fracture risk in postmenopausal women (Smith et al., 2020; Lee et al., 2021). However, our findings contrast with those of the VITAL randomized trial, which found no significant reduction in fracture incidence with vitamin D supplementation in a general population cohort (LeBoff et al., 2022). This discrepancy may reflect differences in study population (our cohort included only women with documented vitamin D deficiency) and supplementation dosing (our protocol used 4,000 IU daily versus 2,000 IU in VITAL). These differences suggest that the protective effect of vitamin D supplementation may be concentration-dependent and limited to deficient populations.
修改后的版本承认了相反结论的研究，具体说明可能造成差异的设计因素，并借助这种对比把当前研究的解读进一步收紧。
三、罗列局限性，却没说明它会带来什么影响
大多数作者都知道 Discussion 需要有一段讲局限性。问题在于，很多人把这一段当成例行公事：列出若干潜在不足，却不解释每一条局限对结果解读具体会产生什么影响。一段读起来像清单的局限性段落，会让审稿人觉得作者只是”被要求写”才写的。
常见错误写法：
This study has several limitations. First, the sample size was relatively small. Second, it was a single-center study. Third, the follow-up period was short. Fourth, we did not include a placebo control.
修改后：
This study has several limitations that affect the generalizability of our findings. The sample size (n = 86) may have been insufficient to detect small but clinically meaningful differences in secondary endpoints, particularly the subgroup analysis of patients over 65 years. As a single-center study conducted at a tertiary referral hospital, the patient population may not represent community-level disease severity, and the intervention’s effectiveness in primary care settings remains to be established. The 6-month follow-up period captures short-term treatment response but cannot address long-term durability of benefit or late-emerging adverse effects.
修改后的版本解释了每条局限的后果。样本量不仅仅是”小”，它可能让某个亚组的效应没被检测到。单中心不仅仅是一条局限，它影响的是结果向特定临床场景外推的能力。每一条局限都与一个具体的解读后果相连。
四、在有限数据之上做出过大外推
单中心、120 例患者的回顾性研究，无法支持关于”所有该病患者”的断言。细胞实验无法直接预测临床结局。观察性研究无法建立因果关系。然而 Discussion 中常常把结论推到研究设计本身无法支撑的程度。
这类问题与 Results 中的过度声明不同（Results 里的过度声明我们在先前一篇文章里讨论过）。在 Discussion 里，过度外推指的是作者把解读延伸到自己研究并未直接检验的人群、场景或机制上。
常见错误写法：
Based on our findings, this biomarker should be incorporated into routine clinical screening for early detection of pancreatic cancer.
修改后：
Our findings suggest that this biomarker warrants further investigation as a potential screening tool for pancreatic cancer. Validation in a larger, multicenter prospective cohort is needed to establish its diagnostic accuracy across different disease stages and patient demographics before clinical implementation can be considered.
修改后的版本保留了结果的积极意义，同时明确指出要把结论推进到临床应用之前，还需要哪些验证步骤。措辞从 “should be incorporated”（明确的临床建议）转为 “warrants further investigation”（与当前证据匹配的下一步方向）。
五、结尾没有落到明确的意义或未来方向
Discussion 的最后一段，是审稿人进入整体评估前留下的最后一印象。如果这段以一句笼统的复述收尾（“In conclusion, our study found that…”），或者在列完局限性之后戛然而止，整篇稿件就会显得没写完。
一个有效的收尾段应当做两件事：说明这项研究在实践或理论上的意义（这对领域意味着什么？），并指出具体、可操作的未来研究方向（下一步应该做什么？）。
常见错误写法：
In conclusion, our study found that treatment X was associated with improved outcomes in patients with condition Y. Further studies are needed to confirm these findings.
修改后：
These findings position treatment X as a candidate adjunct therapy for patients with condition Y who do not respond adequately to standard first-line management. Two specific questions merit investigation in future trials: whether the treatment effect persists beyond the 12-month observation window examined here, and whether the benefit extends to patients with comorbid condition Z, a population excluded from the current study.
修改后的版本明确了获益人群（对一线治疗反应不佳者），指出了两个具体的研究问题，并把它们与当前研究的具体边界对应起来。“Further studies are needed” 被一条清晰的研究议程所替代。
提交前自查清单
- Discussion 的开头段是在解读结果，还是只是在复述？如果它能被 Results 中的一段直接替换，那就需要重写。
- 是否讨论了那些得出不同结论的已发表研究？是否解释了差异为何可能存在？
- 每一条列出的局限，是否都说明了它对结果或对结果解读的具体影响？
- 结论是否保持在研究设计所能支持的边界内？单中心研究可以支持假设，但不能确立普适性的临床建议。
- 最后一段是否说清了这项工作的实际意义，并给出了具体、可检验的未来研究问题？
一段写得到位的 Discussion，能把数据转化为认识。这一节往往是区分”可以发表的稿件”和”需要大修的稿件”的关键。如果你希望有专业人员帮你判断 Discussion 是否真正为你的数据争取到了应有的分量，ScholarMemory 为医学与生命科学研究者提供专业学术润色服务。如需联系，请发送邮件至 contact@scholarmemory.com。