在非随机对照研究质量评价——非随机干预性研究偏倚风险评估工具ROBINS-I (Risk of Bias in Non-randomized Studies-of Intervention) (一)一文中介绍了ROBINS-I的使用方法；在非随机对照研究质量评价——非随机干预性研究偏倚风险评估工具ROBINS-I (Risk of Bias in Non-randomized Studies-of Intervention) (二)一文中介绍了评估清单中的混杂偏倚、研究对象选择的偏倚、干预分类的偏倚、偏离既定干预措施的偏倚。本文主要介绍评估清单中的缺失数据的偏倚、结局测量的偏倚、结果选择性报告的偏倚。
关键词：非随机对照研究质量评价; 非随机干预性研究偏倚风险评估; NRSI; ROBINS-I
二、评估清单
| Table 3 | The Risk Of Bias In Non-randomized Studies——of Interventions (ROBINS-I) assessment tool | |||
| 表3 | 非随机研究的偏倚风险评估工具ROBINS-I | |||
| Bias domain 偏倚域 | Signalling questions 信号问题 | Elaboration 详细解释 | Response options 回答选项 |
| Bias due to missing data 译文：缺失数据的偏倚 | 5.1 Were outcome data available for all, or nearly all, participants? | "Nearly all" should be interpreted as "enough to be confident of the findings", and a suitable proportion depends on the context. In some situations, availability of data from 95% (or possibly 90%) of the participants may be sufficient, providing that events of interest are reasonably common in both intervention groups. One aspect of this is that review authors would ideally try and locate an analysis plan for the study. | Y/PY/PN/N/NI |
| 译文：5.1是否可以获得所有或几乎所有研究对象的结局数据？ | 译文：“几乎所有”应该被解释为“足以对调查结果充满信心”，具体多少比例合适取决于上下文。 在某些情况下，当两个组中的结局事件发生率比较高时，95% (或90%)参与者的数据可用可能就够了。理想的做法是，系统评价员应努力确定一份好的分析计划 | 是/可能是/可能否/否/未知 | |
| 5.2 Were participants excluded due to missing data on intervention status? | Missing intervention status may be a problem. This requires that the intended study sample is clear, which it may not be in practice. | Y/PY/PN/N/NI | |
| 译文：5.2研究对象是否因干预状态数据的缺失而被排除？ | 译文：缺少干预状态可能是一个问题。这就要求所要研究的样本是明确的，但实际过程中可能并不是这样 | 是/可能是/可能否/否/未知 | |
| 5.3 Were participants excluded due to missing data on other variables needed for the analysis? | This question relates particularly to participants excluded from the analysis because of missing information on confounders that were controlled for in the analysis. | Y/PY/PN/N/NI | |
| 译文：5.3 研究对象是否因缺少分析所需的其他变量数据而被排除？ | 译文：该问题特别涉及因分析中控制的混杂因素信息缺失而被排除在分析之外的受试者 | 是/可能是/可能否/否/未知 | |
| 5.4 If PN/N to 5.1, or Y/PY to 5.2 or 5.3: Are the proportion of participants and reasons for missing data similar across interventions? | This aims to elicit whether either (i) differential proportion of missing observations or (ii) differences in reasons for missing observations could substantially impact on our ability to answer the question being addressed. "Similar" includes some minor degree of discrepancy across intervention groups as expected by chance. | NA/Y/PY/PN/N/NI | |
| 译文：5.4 若问题 5.1 回答否/可能否，或问题 5.2/5.3 回答是/可能是，则数据缺失比例和缺失原因在组间是否相似？ | 译文：这旨在确定，(i) 缺少观察结果的比例或(ii) 缺少观察结果的原因是否会严重影响我们回答所解决问题的能力。“相似”是指偶然性所导致的组间微小差异 | 不适用/是/可能是/可能否/否/未知 | |
| 5.5 If PN/N to 5.1, or Y/PY to 5.2 or 5.3: Is there evidence that results were robust to the presence of missing data? | Evidence for robustness may come from how missing data were handled in the analysis and whether sensitivity analyses were performed by the investigators, or occasionally from additional analyses performed by the systematic reviewers. It is important to assess whether assumptions employed in analyses are clear and plausible. Both content knowledge and statistical expertise will often be required for this. For instance, use of a statistical method such as multiple imputation does not guarantee an appropriate answer. Review authors should seek naive (complete-case) analyses for comparison, and clear differences between complete-case and multiple imputation-based findings should lead to careful assessment of the validity of the methods used. | NA/Y/PY/PN/N/NI | |
| 译文：5.5 若问题 5.1、5.2或5.3 回答为否/可能否，则回答是否应用了合适的统计学方法处理缺失数据？ | 译文：可靠的证据来自分析中处理缺失数据的方式以及研究者是否进行了敏感性分析，或者偶尔来自系统评价者进行的其他分析。重要的是要评估分析中使用的假设是否明确和合理。为此往往需要丰富的专业知识和统计知识。例如，使用多重插补方法并不能保证得到适当的答案。评价者应寻求原始(完整病例)分析进行比较，如果完整病例分析和多重插补分析之间存在明显差异，则需要仔细评估所使用方法的有效性 | 不适用/是/可能是/可能否/否/未知 | |
| Optional: Risk of bias judgement | See Table 2 | Low / Moderate / Serious / Critical / NI | |
| 译文：可选项：缺失数据的偏倚方向 | 见表2 | 低/中/高/极高/未知 | |
| Bias in measurement of outcome 译文：结局测量的偏倚 | 6.1 Could the outcome measure have been influenced by knowledge of the intervention received? | Some outcome measures involve negligible assessor judgment, e.g. all-cause mortality or non-repeatable automated laboratory assessments. Risk of bias due to measurement of these outcomes would be expected to be low. | Y/PY/PN/N/NI |
| 译文：6.1结局测量是否受到已分配的干预措施相关知识的影响？ | 译文：一些结局测量可以忽略评价者的判断，例如全因死亡率或不可重复的自动化实验室检测结果。测量这些结局所导致的偏倚风险一般较低 | 是/可能是/可能否/否/未知 | |
| 6.2 Were outcome assessors aware of the intervention received by study participants? | lf outcome assessors were blinded to intervention status, the answer to this question would be 'No'. In other situations, outcome assessors may be unaware of the interventions being received by participants despite there being no active blinding by the study investigators; the answer this question would then also be 'No'. In studies where participants report their outcomes themselves, for example in a questionnaire, the outcome assessor is the study participant. In an observational study, the answer to this question will usually be 'Yes' when the participants report their outcomes themselves. | Y/PY/PN/N/NI | |
| 译文：6.2结局评价者是否知道研究对象接受的干预？ | 译文：如果结局评价者对干预状态不知情，则该问题的答案为“否”。在其他情况下，结局评价者可能不知道受试者正在接受的干预措施，尽管研究者没有主动设盲。那么这个问题的答案也是“否”。在患者报告结局(Patient Reported Outcomes, PRO)的研究中(如问卷调查)，结果评价者为患者本人。在观察性研究中，当为患者报告结局研究时，该问题答案通常为“是” | 是/可能是/可能否/否/未知 | |
| 6.3 Were the methods of outcome assessment comparable across intervention groups? | Comparable assessment methods (i.e. data collection) would involve the same outcome detection methods and thresholds, same time point, same definition, and same measurements. | Y/PY/PN/N/NI | |
| 译文：6.3各组间结局评价方法是否具有可比性？ | 译文：可比的评价方法(即数据收集)涉及相同的结局检测方法和阈值、相同的时间点、相同的定义和相同的度量 | 是/可能是/可能否/否/未知 | |
| 6.4 Were any systematic errors in measurement of the outcome related to intervention received? | This question refers to differential misclassification of outcomes. Systematic errors in measuring the outcome, if present, could cause bias if they are related to intervention or to a confounder of the intervention-outcome relationship. This will usually be due either to outcome assessors being aware of the intervention received or to non-comparability of outcome assessment methods, but there are examples of differential misclassification arising despite these controls being in place. | Y/PY/PN/N/NI | |
| 译文：6.4是否存在与接受干预措施相关的反映在结局测量上的系统性误差？ | 译文：这个问题涉及结果的差异性错误分类。如果存在测量结局的系统性误差(与干预或干预-结果关系的混杂因素有关)，则可能导致偏倚。这通常是由于结果评价者知道所接受的干预措施或结果评估方法不具可比性，尽管有控制措施，但仍存在差异性错误分类的例子 | 是/可能是/可能否/否/未知 | |
| Optional: Risk of bias judgement | See Table 2 | Low / Moderate / Serious / Critical / NI | |
| 译文：可选项：结局测量的偏倚方向 | 见表2 | 低/中/高/极高/未知 | |
| Bias in selection of the reported result 译文：结果选择性报告的偏倚 | Is the reported effect estimate likely to be selected, on the basis of the results,from... | ||
| 译文：所报告的效应量是否基于以下结果进行的选择 | |||
| 7.1. ... multiple outcome measurements within the outcome domain? | For a specified outcome domain, it is possible to generate multiple effect estimates for different measurements. If multiple measurements were made, but only one or a subset is reported, there is a risk of selective reporting on the basis of results. | Y/PY/PN/N/NI | |
| 译文：7.1某个结局的多重测量 | 译文：对于某个特定的结局，可以是不同的测量生成的多个效应量。如果进行了多次测量，但仅报告了一次或一个子集，则存在基于结果进行选择性报告的风险 | 是/可能是/可能否/否/未知 | |
| 7.2 ... multiple analyses of the intervention-outcome relationship? | Because of the limitations of using data from non-randomized studies for analyses of effectiveness (need to control confounding, substantial missing data, etc), analysts may implement different analytic methods to address these limitations. Examples include unadjusted and adjusted models; use of final value vs change from baseline vs analysis of covariance; different transformations of variables; a continuously scaled outcome converted to categorical data with different cut-points; different sets of covariates used for adjustment; and different analytic strategies for dealing with missing data. Application of such methods generates multiple estimates of the effect of the intervention versus the comparator on the outcome. lf the analyst does not pre-specify the methods to be applied, and multiple estimates are generated but only one or a subset is reported, there is a risk of selective reporting on the basis of results. | Y/PY/PN/N/NI | |
| 译文：7.2干预-结局关系的多重数据分析 | 译文：由于使用非随机化研究数据进行有效性分析的局限性(需要控制混杂，大量缺失数据等)，分析者可能会采用不同的分析方法来解决这些局限性。比如，采用未调整和调整模型；使用最终值 vs 相对于基线变化值 vs 协方差分析；变量的不同变换；连续性变量转换为分类变量；用于调整的不同协变量集；以及处理缺失数据的不同分析策略。应用这些方法可以对干预措施与对照措施对结果的影响进行多重估计。如果分析人员未预先指定分析方法，并且生成了多个估计值，但仅报告了一个或一个子集，则存在基于结果进行选择性报告的风险 | 是/可能是/可能否/否/未知 | |
| 7.3 ... different subgroups? | Particularly with large cohorts often available from routine data sources, it is possible to generate multiple effect estimates for different subgroups or simply to omit varying proportions of the original cohort. If multiple estimates are generated but only one or a subset is reported, there is a risk of selective reporting on the basis of results. | Y/PY/PN/N/NI | |
| 译文：7.3不同亚组分析 | 译文：特别是，对于从日常监测数据系统中获得的大型队列，有可能不同的亚组可生成多个效应估计值。如果生成了多个估计，但仅报告了一个或一个子集，则存在基于结果进行选择性报告的风险 | 是/可能是/可能否/否/未知 | |
| Optional: Risk of bias judgement | See Table 2 | Low / Moderate / Serious / Critical / NI | |
| 译文：可选项：偏倚风险评估 | 见表2 | 低/中/高/极高/未知 | |
| Overall bias | Risk of bias judgement | See Table 2 | Low / Moderate / Serious / Critical / NI |
| 偏倚风险评估 | 见表2 | 低/中/高/极高/未知 | |
注：本文内容是参考相关文献后对非随机干预性研究偏倚风险评估工具ROBINS-I (Risk of Bias in Non-randomized Studies-of Intervention)的概述，仅代表本网站观点。关于ROBINS-I 的更多内容详见网站(https://methods.cochrane.org/bias/risk-bias-non-randomized-studies-interventions)或论文ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions 。
End