CDISC的ADaMIG (V1.2) 中英文对照【4】_第四章（下）实施问题，标准解决方案和示例
本AdaMIG (v1.2)来自CDISC官网以下链接：
https://www.cdisc.org/standards/foundational/adam/adam-implementation-guide-v1-2-release-package
第4章（下）
4.5 Identification of Records Used for Analysis
4.5 识别用于分析的记录
This section addresses how to identify the records of an ADaM dataset that are used for analysis. The four specific issues addressed include: (1) identification of the records used in an LOCF analysis; (2) identification of the record containing the baseline value; (3) identification of post-baseline conceptual timepoint records, such as endpoint, minimum, maximum, or average; and (4) identification of specific records used in an analysis.
本节介绍如何识别用于分析的ADaM数据集的记录。解决的四个具体问题包括：（1）确定LOCF分析中使用的记录；（2）识别包含基线值的记录；（3）识别基线后的概念性时间点记录，例如端点，最小值，最大值或平均值；（4）确定分析中使用的特定记录。
4.5.1 Identification of Records Used in a Timepoint Imputation Analysis
4.5.1识别时间点归因分析中使用的记录
This section considers the issue of how to identify records used in a timepoint-related imputation analysis as well as how to represent data imputed for missing timepoints in an ADaM dataset. LOCF is one of the most commonly used timepoint-related imputation analyses, and is therefore specifically mentioned. However, the methodology is general and is not restricted to LOCF analysis. WOCF analysis is also mentioned to emphasize the generalizability.
本节考虑如何识别与时间点相关的插补分析中使用的记录，以及如何表示ADaM数据集中缺少时间点的插补数据的问题。LOCF是最常用的与时间点相关的归因分析之一，因此专门提及。但是，该方法是通用的，并不限于LOCF分析。还提到了WOCF分析以强调可概括性。
ADaM Methodology
ADaM方法论
When an analysis timepoint is missing, the ADaM methodology is to create a new record in the ADaM dataset to represent the missing timepoint and identify these imputed records by populating the derivation type variable DTYPE.
当缺少分析时间点时，ADaM方法将在ADaM数据集中创建一个新记录来表示缺少的时间点，并通过填充派生类型变量DTYPE来识别这些估算记录。
For example, when an LOCF/WOCF analysis is being performed, create LOCF/WOCF records when the LOCF/WOCF analysis timepoints are missing, and identify these imputed records by populating the derivation type variable DTYPE with values LOCF or WOCF. All of the original records would have null values in DTYPE. It would be very simple to select the appropriate records for analysis by selecting DTYPE=null for Data as Observed (DAO) analysis, DTYPE=null or LOCF for LOCF analysis, and DTYPE=null or WOCF for WOCF analysis. This approach would require understanding and communicating that if the DTYPE flag were not referenced correctly, the analysis would default to using all records, including the DAO records, plus the records derived by LOCF and WOCF. To perform a correct DAO analysis, one would need to explicitly select DTYPE=null.
例如，当执行LOCF / WOCF分析时，在缺少LOCF / WOCF分析时间点时创建LOCF / WOCF记录，并通过使用值LOCF或WOCF填充派生类型变量DTYPE来识别这些估算记录。所有原始记录在DTYPE中都将具有空值。选择适当的记录进行分析非常简单，方法是选择DTYPE = null进行“按观测的数据”（DAO）分析，选择DTYPE = null或LOCF进行LOCF分析，而选择DTYPE = null或WOCF进行WOCF分析。这种方法需要理解和传达，如果未正确引用DTYPE标志，则分析将默认使用所有记录，包括DAO记录以及LOCF和WOCF派生的记录。为了执行正确的DAO分析，需要明确选择DTYPE = null。
Example 1
例子1
Identification of rows used in a LOCF analysis. 标识LOCF分析中使用的行。
In Table 4.5.1.1, some subjects have complete data and others have rows imputed by one method (LOCF). Subjects with no missing data have the same number of rows as in the input dataset, with all DTYPE values blank. Subject 1001 has complete data. DTYPE is blank for all rows indicating they are not imputed. AVISIT matches VISIT (from SDTM) in this example. AVISIT does not always match VISIT from SDTM even in scenarios where there is no missing data. Subject 1002 is missing the Week 2 assessment. Week 2 is imputed using the LOCF method.
在表4.5.1.1中，一些主题具有完整的数据，而另一些主题具有通过一种方法（LOCF）估算的行。没有缺失数据的主题的行数与输入数据集中的行数相同，所有DTYPE值均为空白。主题1001具有完整的数据。对于所有行，DTYPE均为空白，表示未估算。在此示例中，AVISIT与VISIT（来自SDTM）匹配。即使在没有丢失数据的情况下，AVISIT也不总是与SDTM的VISIT匹配。受试者1002缺少第2周评估。第2周使用LOCF方法估算。
AVISIT=Week 2 but VISIT=Week 1 so one can see where the imputed value came from in the original data.
AVISIT =第2周，但VISIT =第1周，因此可以看到估算值在原始数据中的来源。
Subject 1003 is missing Week 2 and 3 data. A Data as Observed (DAO) analysis can be performed by selecting only those rows where DTYPE is null. For a LOCF analysis, all rows (DTYPE=null or DTYPE="LOCF") should be used.
对象1003缺少第2周和第3周数据。可以通过仅选择DTYPE为空的那些行来执行观测数据（DAO）分析。对于LOCF分析，应使用所有行（DTYPE = null或DTYPE =“ LOCF”）。
Table 4.5.1.1 Example 1: ADaM Dataset with Identification of Rows Used in a LOCF Analysis
表4.5.1.1示例1：具有LOCF分析中使用的行标识的ADaM数据集
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
PARAM |
AVAL |
DTYPE |
VSSEQ |
| 1 |
1001 |
Baseline |
Baseline |
-4 |
SUPINE SYSBP (mm Hg) |
145 |
171 |
|
| 2 |
1001 |
Week 1 |
Week 1 |
3 |
SUPINE SYSBP (mm Hg) |
130 |
191 |
|
| 3 |
1001 |
Week 2 |
Week 2 |
9 |
SUPINE SYSBP (mm Hg) |
133 |
201 |
|
| 4 |
1001 |
Week 3 |
Week 3 |
20 |
SUPINE SYSBP (mm Hg) |
125 |
211 |
|
| 5 |
1002 |
Baseline |
Baseline |
-1 |
SUPINE SYSBP (mm Hg) |
145 |
50 |
|
| 6 |
1002 |
Week 1 |
Week 1 |
7 |
SUPINE SYSBP (mm Hg) |
130 |
60 |
|
| 7 |
1002 |
Week 1 |
Week 2 |
7 |
SUPINE SYSBP (mm Hg) |
130 |
LOCF |
60 |
| 8 |
1002 |
Week 3 |
Week 3 |
22 |
SUPINE SYSBP (mm Hg) |
135 |
70 |
|
| 9 |
1003 |
Baseline |
Baseline |
1 |
SUPINE SYSBP (mm Hg) |
150 |
203 |
|
| 10 |
1003 |
Week 1 |
Week 1 |
8 |
SUPINE SYSBP (mm Hg) |
140 |
213 |
|
| 11 |
1003 |
Week 1 |
Week 2 |
8 |
SUPINE SYSBP (mm Hg) |
140 |
LOCF |
213 |
| 12 |
1003 |
Week 1 |
Week 3 |
8 |
SUPINE SYSBP (mm Hg) |
140 |
LOCF |
213 |
Example 2
例子2
Identification of rows used in both LOCF and WOCF analyses.
标识在LOCF和WOCF分析中使用的行。
Table 4.5.1.2 shows a situation where there is more than one imputation method used. In this case, additional rows are generated for each type of imputation. A DAO analysis can be performed by selecting only those rows where DTYPE is null. For LOCF analysis, all rows with DTYPE=null or DTYPE="LOCF" should be used. For WOCF analysis, all rows with DTYPE=null or DTYPE="WOCF" should be used.
表4.5.1.2显示了使用不止一种插补方法的情况。在这种情况下，将为每种插补类型生成其他行。通过仅选择DTYPE为空的那些行可以执行DAO分析。对于LOCF分析，应使用DTYPE = null或DTYPE =“ LOCF”的所有行。对于WOCF分析，应使用DTYPE = null或DTYPE =“ WOCF”的所有行。
Table 4.5.1.2 Example 2: ADaM Dataset with Identification of Rows Used in Both LOCF and WOCF Analyses
表4.5.1.2示例2：具有在LOCF和WOCF分析中使用的行标识的ADaM数据集
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
PARAM |
AVAL |
DTYPE |
VSSEQ |
| 1 |
1002 |
Baseline |
Baseline |
-4 |
SUPINE SYSBP (mm Hg) |
145 |
77 |
|
| 2 |
1002 |
Week 1 |
Week 1 |
3 |
SUPINE SYSBP (mm Hg) |
130 |
78 |
|
| 3 |
1002 |
Week 2 |
Week 2 |
9 |
SUPINE SYSBP (mm Hg) |
138 |
79 |
|
| 4 |
1002 |
Week 3 |
Week 3 |
18 |
SUPINE SYSBP (mm Hg) |
135 |
80 |
|
| 5 |
1002 |
Week 3 |
Week 4 |
18 |
SUPINE SYSBP (mm Hg) |
135 |
LOCF |
80 |
| 6 |
1002 |
Week 2 |
Week 4 |
9 |
SUPINE SYSBP (mm Hg) |
138 |
WOCF |
79 |
| 7 |
1002 |
Week 5 |
Week 5 |
33 |
SUPINE SYSBP (mm Hg) |
130 |
81 |
|
| 8 |
1003 |
Baseline |
Baseline |
-1 |
SUPINE SYSBP (mm Hg) |
145 |
122 |
|
| 9 |
1003 |
Week 1 |
Week 1 |
7 |
SUPINE SYSBP (mm Hg) |
140 |
123 |
|
| 10 |
1003 |
Week 2 |
Week 2 |
15 |
SUPINE SYSBP (mm Hg) |
138 |
124 |
|
| 11 |
1003 |
Week 2 |
Week 3 |
15 |
SUPINE SYSBP (mm Hg) |
138 |
LOCF |
124 |
| 12 |
1003 |
Week 2 |
Week 4 |
15 |
SUPINE SYSBP (mm Hg) |
138 |
LOCF |
124 |
| 13 |
1003 |
Week 2 |
Week 5 |
15 |
SUPINE SYSBP (mm Hg) |
138 |
LOCF |
124 |
| 14 |
1003 |
Week 1 |
Week 3 |
7 |
SUPINE SYSBP (mm Hg) |
140 |
WOCF |
123 |
| 15 |
1003 |
Week 1 |
Week 4 |
7 |
SUPINE SYSBP (mm Hg) |
140 |
WOCF |
123 |
| 16 |
1003 |
Week 1 |
Week 5 |
7 |
SUPINE SYSBP (mm Hg) |
140 |
WOCF |
123 |
Approaches Considered and Not Adopted
考虑和不采用的方法
Another approach considered is to create a complete separate set of records for each analysis type (or a separate dataset), indicating the various analysis types by assigning unique values of the analysis timepoint description AVISIT, for example, "Week 4," "Week 4 (LOCF)," and "Week 4 (WOCF)". This approach would make it more foolproof to perform the DAO, LOCF, and WOCF analysis in one step by referencing only AVISIT. However, because so many records would be duplicated, a very large dataset is one of the major disadvantages for this approach. In addition, this approach might be less tool-friendly, in that one might need to parse AVISIT searching for a key substring such as "(LOCF)." This approach should not be used.
另一种考虑的方法是为每个分析类型(或一个单独的数据集)创建一个完整的单独的记录集，通过分配分析时间点描述AVISIT的唯一值来指示各种分析类型，例如，“第4周”、“第4周(LOCF)”和“第4周(WOCF)”。通过只引用AVISIT，这种方法可以更加简单地一步执行DAO、LOCF和WOCF分析。
但是，由于要复制的记录太多，所以非常大的数据集是这种方法的主要缺点之一。
此外，这种方法可能对工具不太友好，因为可能需要解析AVISIT搜索关键子字符串，如“(LOCF)”。
不应该使用这种方法。
A third approach considered is to create a flag (LOCFFL/LOCFFN) to indicate when a record is created by virtue of last observation carried forward, and similarly for WOCF. This is similar to the specified ADaM methodology, except that a separate flag is created for each derivation type, rather than indicating row derivation type in one column DTYPE. This approach might result in fewer records than the recommended approach (e.g., if the WOCF record is the same as the LOCF record). In other respects, this approach shares the advantages and disadvantages of the recommended approach. This approach of creating separate flags for each derivation type is not recommended.
考虑的第三种方法是创建一个标志（LOCFFL / LOCFFN），以指示何时根据结转的最后观察创建记录，对于WOCF也是类似的。这与指定的ADaM方法相似，除了为每种派生类型创建一个单独的标志，而不是在一个列DTYPE中指示行派生类型。与推荐的方法相比，此方法可能会导致记录减少（例如，如果WOCF记录与LOCF记录相同）。在其他方面，此方法具有推荐方法的优点和缺点。不建议这种为每种派生类型创建单独标志的方法。
4.5.2 Identification of Baseline Records
4.5.2 基准记录的识别
Many statistical analyses require the identification of a baseline value. This section describes how a record used as a baseline is identified.
许多统计分析都需要确定基线值。本节介绍如何识别用作基线的记录。
ADaM Methodology
ADaM方法论
The ADaM methodology is to create a baseline flag column to indicate the record used as baseline (the record whose value of AVAL is used to populate the BASE variable). This method does not require duplication of records in the event that the baseline record is not derived.
ADaM方法是创建一个基线标志列以指示用作基线的记录（该记录的AVAL值用于填充BASE变量）。如果不导出基线记录，则此方法不需要重复记录。
Although a baseline record flag variable ABLFL is created and used to identify the record that is the baseline record, this does not prohibit also providing a record with a unique value of AVISIT (e.g., "Baseline"), designating the baseline record used for analysis, even if redundant with another record. For more complicated baseline definitions (functions of multiple records), a derived baseline record would have to be created as described in 4.2.1.3, Rule 3: A function of one or more rows within the same parameter for the purpose of creating an analysis timepoint should be added as a new row for the same parameter. This methodology requires that clear metadata be provided for the baseline record variable so that the value can be reproduced accurately.
尽管创建了基线记录标志变量ABLFL并将其用于标识作为基线记录的记录，但这并不禁止还提供具有AVISIT唯一值的记录（例如“ Baseline”），指定用于分析的基线记录，即使与另一个记录重复。对于更复杂的基线定义（多个记录的功能），必须按照4.2.1.3第3条规则中的描述创建派生基线记录： 为了创建分析时间点，在同一参数内具有一个或多个行的功能应该添加为同一参数的新行。此方法要求为基线记录变量提供清晰的元数据，以便可以准确地重现该值。
Example 1
例子1
Identification of baseline rows – using screening visit to impute a baseline row.
识别基线行–使用筛选访问来估算基线行。
This example (Table 4.5.2.1) illustrates the use of a baseline flag variable ABLFL. It also illustrates the inclusion of an additional row for a baseline analysis timepoint (row 6). In this example, a unique value of AVISIT has been defined for the baseline record used for analysis. Subject 1001 had complete data. There was no record that qualified as a baseline value for Subject 1002 in the source data. A derived baseline record (AVISIT="Baseline") is added with DTYPE="LVPD" (Last Value Prior to Dosing) to indicate that the record is imputed to be used as baseline.
此示例（表4.5.2.1）说明了基准标志变量ABLFL的用法。它还说明了基线分析时间点（第6行）的附加行。在此示例中，已为用于分析的基线记录定义了AVISIT的唯一值。受试者1001具有完整的数据。在源数据中没有记录可以作为主题1002的基线值。派生的基线记录（AVISIT =“ Baseline”）与DTYPE =“ LVPD”（加料前的最后值）一起添加，以指示该记录被推算为用作基线。
Table 4.5.2.1 Example 1: ADaM Dataset with Identification of Baseline Rows when Imputation Is Used
表4.5.2.1示例1：使用插补时具有基线行标识的ADaM数据集
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
ABLFL |
PARAM |
AVAL |
BASE |
DTYPE |
VSSEQ |
| 1 |
1001 |
Screening |
Screening |
-12 |
SUPINE SYSBP (mm Hg) |
144 |
1 |
|||
| 2 |
1001 |
Baseline |
Baseline |
1 |
Y |
SUPINE SYSBP (mm Hg) |
145 |
2 |
||
| 3 |
1001 |
Week 1 |
Week 1 |
6 |
SUPINE SYSBP (mm Hg) |
130 |
145 |
3 |
||
| 4 |
1001 |
Week 2 |
Week 2 |
12 |
SUPINE SYSBP (mm Hg) |
133 |
145 |
4 |
||
| 5 |
1002 |
Screening |
Screening |
-14 |
SUPINE SYSBP (mm Hg) |
144 |
1 |
|||
| 6 |
1002 |
Screening |
Baseline |
-14 |
Y |
SUPINE SYSBP (mm Hg) |
144 |
LVPD |
1 |
|
| 7 |
1002 |
Week 1 |
Week 1 |
8 |
SUPINE SYSBP (mm Hg) |
130 |
144 |
2 |
||
| 8 |
1002 |
Week 2 |
Week 2 |
14 |
SUPINE SYSBP (mm Hg) |
133 |
144 |
3 |
Example 2
例子2
Identification of baseline rows – using an average of multiple visits to derive a baseline row.
识别基准行–使用多次访问的平均值得出基准行。
This example (Table 4.5.2.2) illustrates the use of a baseline flag variable ABLFL to identify the record used as baseline for analysis in a scenario where the baseline value is based on the average of the non-missing values collected prior to dosing. Row 3 is a derived "Baseline" record using the average of the values of row 1 and row 2. DTYPE="AVERAGE" to indicate that row 3 is derived. The Baseline flag (ABLFL="Y") indicates that AVAL from row 3 is used to populate the BASE (Baseline) column. VISIT (from SDTM) is left blank on row 3 since AVAL on that record is not merely a copy of AVAL on another record.
此示例（表4.5.2.2）说明了使用基线标志变量ABLFL来识别用作基线的记录的情况，其中基线值基于给药前收集的非缺失值的平均值。第3行是派生的“基线”记录，使用第1行和第2行的值的平均值。DTYPE =“ AVERAGE”表示派生了第3行。基准标志（ABLFL =“ Y”）表示AVAL
第3行中的数据用于填充BASE（基准）列。（来自SDTM的）VISIT在第3行上保留为空白，因为该记录上的AVAL不仅是另一条记录上的AVAL的副本。
Table 4.5.2.2 Example 2: ADaM Dataset with Identification of Baseline Rows when Baseline Is an Average
表4.5.2.2示例2：基线为平均值时具有基线行标识的ADaM数据集
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
ABLFL |
PARAM |
AVAL |
BASE |
DTYPE |
| 1 |
1001 |
Screening |
Screening |
-12 |
SUPINE SYSBP (mm Hg) |
144 |
144.5 |
||
| 2 |
1001 |
Baseline |
Baseline |
1 |
SUPINE SYSBP (mm Hg) |
145 |
144.5 |
||
| 3 |
1001 |
Baseline |
Y |
SUPINE SYSBP (mm Hg) |
144.5 |
144.5 |
AVERAGE |
||
| 4 |
1001 |
Week 1 |
Week 1 |
12 |
SUPINE SYSBP (mm Hg) |
130 |
144.5 |
||
| 5 |
1001 |
Week 2 |
Week 2 |
-14 |
SUPINE SYSBP (mm Hg) |
133 |
144.5 |
Example 3例子3
Identification of baseline rows – using an average of multiple visits to derive a baseline row.
识别基准行–使用多次访问的平均值得出基准行。
This example (Table 4.5.2.3) is the same as Example 2 except that the analysis timepoint description "Screening/Baseline Combination" helps differentiate the derived average baseline record from an existing observed record whose timepoint description is "Baseline." This was helpful in analysis and reporting because it was desired to summarize all scheduled visits in addition to the average baseline visit. The analysis was straightforward using the distinct descriptions of AVISIT. The choice of AVISIT values is up to the producer.
此示例（表4.5.2.3）与示例2相同，不同之处在于分析时间点描述“筛选/基线组合”有助于将导出的平均基线记录与时间点描述为“基线”的现有观察记录区分开。这对分析和报告很有帮助，因为除了平均基线访问外，还希望总结所有计划的访问。使用AVISIT的不同描述可以轻松进行分析。AVISIT值的选择取决于生产者。
Table 4.5.2.3 Example 3: ADaM Dataset with Identification of Baseline Rows, Including Description in Analysis Timepoint Variable
表4.5.2.3示例3：具有基线行标识的ADaM数据集，包括分析时间点变量中的描述
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
ABLFL |
PARAM |
AVAL |
BASE |
DTYPE |
| 1 |
1001 |
Screening |
Screening |
-12 |
SUPINE SYSBP (mm Hg) |
144 |
144.5 |
||
| 2 |
1001 |
Baseline |
Baseline |
1 |
SUPINE SYSBP (mm Hg) |
145 |
144.5 |
||
| 3 |
1001 |
Screening/Baseline Combination |
Y |
SUPINE SYSBP (mm Hg) |
144.5 |
144.5 |
AVERAGE |
||
| 4 |
1001 |
Week 1 |
Week 1 |
12 |
SUPINE SYSBP (mm Hg) |
130 |
144.5 |
||
| 5 |
1001 |
Week 2 |
Week 2 |
-14 |
SUPINE SYSBP (mm Hg) |
133 |
144.5 |
4.5.3 Identification of Post-Baseline Conceptual Timepoint Records
4.5.3 基准后概念时间点记录的标识
When analysis involves cross-timepoint derivations such as endpoint, minimum, maximum, and average post- baseline, questions such as "Should distinct records with unique value of AVISIT always be created even if redundant with an observed value record?", or "Should these records just be flagged?" need to be considered. In this section, Examples 1 and 2 present the approach of adding records, and Example 3 presents the alternate approach of flagging the records.
当分析涉及跨时间点派生（例如端点，最小值，最大值和平均基线后）时，诸如“是否应该创建具有AVISIT唯一值的不同记录，即使有观察值记录是多余的？”或“是否应该这些记录只是被标记？” 需要考虑的。在本节中，示例1和2展示了添加记录的方法，示例3展示了标记记录的替代方法。
ADaM Methodology
ADaM方法论
The ADaM methodology is to create a new record with a unique value of AVISIT in cases where analysis is based on AVISIT. The advantage of this approach is that it is simple and analysis-friendly. It is recognized that such new records might be redundant with observed records for some kinds of conceptual timepoint definitions.
ADaM方法是在分析基于AVISIT的情况下创建具有AVISIT唯一值的新记录。这种方法的优点是它简单且易于分析。人们认识到，对于某些类型的概念性时间点定义，此类新记录可能与观察到的记录无关。
Always creating a record with a unique value of AVISIT designating the record used for analysis (e.g., "Endpoint," "Post-Baseline Minimum," "Post-Baseline Maximum") has the advantage that once the AVISIT values are understood, producers, consumers, and software can rely on these values of AVISIT. This approach represents the general case because any such cross-timepoint derivation can be represented in a new record with a unique AVISIT description. The disadvantage is that the dataset would contain more records, and conventions would have to be communicated and understood.
始终创建具有唯一值AVISIT的记录来指定用于分析的记录（例如，“端点”，“基线后最小值”，“基线后最大值”）具有以下优点：一旦了解了AVISIT值，生产者，消费者，软件可以依靠AVISIT的这些价值。这种方法代表一般情况，因为任何这样的跨时间点派生都可以在具有唯一AVISIT描述的新记录中表示。缺点是数据集将包含更多记录，并且必须传达和理解约定。
In cases where analysis is not based on AVISIT, either solution is valid. It is recognized that in cases where the AVISIT values are not defined in the analysis documentation, adding a flag may be more appropriate. Which methodology is appropriate for situations where an "analysis visit" value is not defined can be driven by how the analysis will be performed. In cases where only a subset of data is analyzed (e.g., only on treatment minimum values), then flagging the values that qualify for analysis might be a better choice than creating an additional record to contain the minimum value. However, where the subset of data is analyzed within the context of a greater pool of data, creating an additional record to contain the minimum value would help facilitate analysis-ready usage and review.
如果分析不是基于AVISIT，则任何一种解决方案都是有效的。已经认识到，在分析文档中未定义AVISIT值的情况下，添加标记可能更合适。哪一个
该方法适用于未定义“分析访问”值的情况，这取决于执行分析的方式。在仅分析数据子集的情况下（例如，仅针对治疗最小值），与创建附加记录以包含最小值相比，标记符合分析条件的值可能是更好的选择。但是，如果在更大的数据池范围内分析数据的子集，则创建包含最小值的附加记录将有助于促进分析就绪的使用和检查。
Example 1 例子1
Identification of endpoint rows. 端点行的标识。
This example (Table 4.5.3.1) shows the creation of an added row with a unique value of AVISIT designating the Endpoint record used for analysis. Subject 1001 discontinued at Week 2, and a derived Endpoint record (AVISIT="Endpoint") is added using the Week 2 visit. DTYPE="LOV" (Last Observed Value) indicates how the AVISIT="Endpoint" record is populated. Subject 1002 did not have any post-baseline visits, and therefore has no Endpoint record.
此示例（表4.5.3.1）显示了添加行的创建，该行具有AVISIT的唯一值，该值指定用于分析的端点记录。受试者1001在第2周停产，并使用第2周访问添加了派生的端点记录（AVISIT =“ Endpoint”）。DTYPE =“ LOV”（最后观察值）指示如何填充AVISIT =“ Endpoint”记录。受试者1002没有任何基线后访问，因此没有端点记录。
Table 4.5.3.1 Example 1: ADaM Dataset with Identification of Endpoint Rows
表4.5.3.1示例1：具有端点行标识的ADaM数据集
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
PARAM |
AVAL |
DTYPE |
| 1 |
1001 |
Screening |
Screening |
-12 |
SUPINE SYSBP (mm Hg) |
144 |
|
| 2 |
1001 |
Baseline |
Baseline |
1 |
SUPINE SYSBP (mm Hg) |
145 |
|
| 3 |
1001 |
Week 1 |
Week 1 |
6 |
SUPINE SYSBP (mm Hg) |
130 |
|
| 4 |
1001 |
Week 2 |
Week 2 |
12 |
SUPINE SYSBP (mm Hg) |
133 |
|
| 5 |
1001 |
Week 2 |
Endpoint |
12 |
SUPINE SYSBP (mm Hg) |
133 |
LOV |
| 6 |
1002 |
Screening |
Screening |
-14 |
SUPINE SYSBP (mm Hg) |
144 |
|
| 7 |
1002 |
Baseline |
Baseline |
-1 |
SUPINE SYSBP (mm Hg) |
144 |
Example 2
例子2
Identification of endpoint and post-baseline minimum, maximum, and average rows.
标识端点和基线后的最小，最大和平均行。
This example (Table 4.5.3.2) shows the creation of rows with unique values of AVISIT designating the Endpoint record, and the Post-Baseline Minimum, Maximum, and Average rows. Subject 1001 had minimum post-baseline result at Week 1, maximum post-baseline result at Week 2, and the average post-baseline result was based on the average of Week 1 and Week 2. This subject discontinued at Week 2. A derived Endpoint record (AVISIT="Endpoint") is added using the Week 2 visit. DTYPE="LOV" (last observed value) indicates that the AVISIT="Endpoint" record is a derived record. Subject 1002 did not have any post-baseline visit. Therefore, the Post-Baseline Minimum, Post-Baseline Maximum, Post-Baseline Average, and Endpoint rows could not be derived for that subject.
这个例子(表4.5.3.2)展示了使用AVISIT的唯一值创建端点记录的行，以及基线后的最小、最大和平均行。
1001名受试者基线后第1周的结果是最小的，基线后第2周的结果是最大的，基线后的平均结果是基于第1和第2周的平均值。这个题目在第2周停止了。使用第2周访问添加派生端点记录(AVISIT="Endpoint")。DTYPE="LOV"(最后观察到的值)表示AVISIT="Endpoint"记录是一个派生记录。受试者1002没有任何基线后访问。因此，无法获得该受试者基线后的最小值、基线后的最大值、基线后的平均值和端点行。
Table 4.5.3.2 Example 2: ADaM Dataset with Identification of Endpoint and Post-Baseline Minimum, Maximum, and Average Rows
表4.5.3.2示例2：具有端点和基线后最小，最大和平均行标识的ADaM数据集
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
PARAM |
AVAL |
DTYPE |
VSSEQ |
| 1 |
1001 |
Screening |
Screening |
-12 |
SUPINE SYSBP (mm Hg) |
144 |
1 |
|
| 2 |
1001 |
Baseline |
Baseline |
1 |
SUPINE SYSBP (mm Hg) |
145 |
2 |
|
| 3 |
1001 |
Week 1 |
Week 1 |
6 |
SUPINE SYSBP (mm Hg) |
130 |
3 |
|
| 4 |
1001 |
Week 2 |
Week 2 |
12 |
SUPINE SYSBP (mm Hg) |
133 |
4 |
|
| 5 |
1001 |
Week 1 |
Post-Baseline Minimum |
6 |
SUPINE SYSBP (mm Hg) |
130 |
MINIMUM |
3 |
| 6 |
1001 |
Week 2 |
Post-Baseline Maximum |
12 |
SUPINE SYSBP (mm Hg) |
133 |
MAXIMUM |
4 |
| 7 |
1001 |
Post-Baseline Average |
SUPINE SYSBP (mm Hg) |
131.5 |
AVERAGE |
|||
| 8 |
1001 |
Week 2 |
Endpoint |
12 |
SUPINE SYSBP (mm Hg) |
133 |
LOV |
4 |
| 9 |
1002 |
Screening |
Screening |
-14 |
SUPINE SYSBP (mm Hg) |
144 |
22 |
|
| 10 |
1002 |
Baseline |
Baseline |
-1 |
SUPINE SYSBP (mm Hg) |
144 |
23 |
Example 3
例子3
Identification of post-baseline minimum and maximum rows.
标识基线后的最小和最大行。
This example (Table 4.5.3.3) shows the identification of the Post-Baseline Minimum and Maximum rows. Subject 1001 had minimum post-baseline result at Week 1 (identified with ANL01FL=Y) and maximum post-baseline result at Week 2 (identified with ANL02FL=Y). Subject 1002 did not have any post-baseline visit. Therefore, the Post- Baseline Minimum and Post-Baseline Maximum could not be identified for that subject.
此示例（表4.5.3.3）显示了基准后最小值和最大值行的标识。受试者1001在第1周的基线后结果最低（以ANL01FL = Y标识），在第2周的基线后最大结果（以ANL02FL = Y标识）。受试者1002没有进行任何基线后访问。因此，无法为该主题确定基线后最小值和基线后最大值。
Table 4.5.3.3 Example 3: ADaM Dataset with Identification of Post-Baseline Minimum and Maximum Rows
表4.5.3.3示例3：具有基准后最小和最大行标识的ADaM数据集
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
PARAM |
AVAL |
ANL01FL |
ANL02FL |
| 1 |
1001 |
Screening |
Screening |
-12 |
SUPINE SYSBP (mm Hg) |
144 |
||
| 2 |
1001 |
Baseline |
Baseline |
1 |
SUPINE SYSBP (mm Hg) |
145 |
||
| 3 |
1001 |
Week 1 |
Week 1 |
6 |
SUPINE SYSBP (mm Hg) |
130 |
Y |
|
| 4 |
1001 |
Week 2 |
Week 2 |
12 |
SUPINE SYSBP (mm Hg) |
133 |
Y |
|
| 5 |
1002 |
Screening |
Screening |
-14 |
SUPINE SYSBP (mm Hg) |
144 |
||
| 6 |
1002 |
Baseline |
Baseline |
-1 |
SUPINE SYSBP (mm Hg) |
144 |
4.5.4 Identification of Records Used for Analysis – General Case
4.5.4确定用于分析的记录-一般情况
It is important to identify the records used in or excluded from analysis. Should records used in the analysis be identified via flags or by unique values of analysis timepoint window description AVISIT?
确定用于分析或从分析中排除的记录很重要。分析中使用的记录是否应通过标记或通过分析时间点窗口描述AVISIT的唯一值来标识？
ADaM Methodology
ADaM方法论
The ADaM methodology is to use an analysis flag (ANLzzFL) to indicate the records that fulfill specific requirements for one or more analyses. For example, ANLzzFL=Y indicates records meeting the requirements for analysis and is blank (null) in other records, such as a duplicate record that was not the one selected for analysis, or pre-specified post-study timepoints not included in the analysis. This allows multiple records within a parameter with the same value of AVISIT. However, it also requires flags to be added to the dataset to be used in selecting appropriate records for analysis. Understanding of the flags is required for correct analysis results to be generated. In addition to ANLzzFL, additional flags might also be required, such as record-based population flags (e.g., ITTRFL, PPROTRFL).
ADaM方法是使用分析标记（ANLzzFL）来指示满足一项或多项分析特定要求的记录。例如，ANLzzFL = Y表示满足分析要求的记录，在其他记录中为空白（空），例如不是为分析选择的重复记录，或者分析中未包含的预先指定的研究后时间点。这允许一个参数中具有相同AVISIT值的多个记录。但是，它还要求将标志添加到数据集，以用于选择适当的记录进行分析。要生成正确的分析结果，需要了解标记。除ANLzzFL外，还可能需要其他标志，例如基于记录的填充标志（例如ITTRFL，PPROTRFL）。
Please note that there can be multiple ANLzzFL variables. In this case, it will be imperative to have clear and robust metadata to indicate the basis for the creation and population of each ANLzzFL variable.
请注意，可以有多个ANLzzFL变量。在这种情况下，必须具有清晰且健壮的元数据来指示每个ANLzzFL变量的创建和填充的基础。
Example 1
例子1
Identification of rows used for analysis – multiple visits that fall within a visit window.
标识用于分析的行–属于访问窗口的多次访问。
This example (Table 4.5.4.1) illustrates the use of the analysis flag variable ANLzzFL to indicate the rows that were chosen for analysis from among the multiple visits that fall within the analysis timepoint windows of "Baseline" and "Week 2". Subject 1001 had two observed Baseline and Week 2 analysis timepoints according to analysis window definitions. The one that is used in analysis is flagged with ANL01FL=Y. This approach is used because all original visits (rows) are included in the dataset, and those selected for analysis must be identified. For traceability reasons, the AWTARGET and AWTDIFF columns are included in order to indicate more clearly how the analyzed rows were selected from among the candidate rows within each analysis window. In this example, the record that falls closest to the scheduled visit day is the one that will be analyzed.
本例(表4.5.4.1)说明了如何使用分析标志变量ANLzzFL来指示在“基线”和“第2周”的分析时间点窗口内的多次访问中选择进行分析的行。根据分析窗口定义，受试者1001有两个观察基线和第2周分析时间点。在分析中使用的标记为ANL01FL=Y。之所以使用此方法，是因为所有原始访问(行)都包含在数据集中，并且必须识别那些选择用于分析的访问。由于可跟踪性的原因，包含AWTARGET和AWTDIFF列是为了更清楚地表明如何从每个分析窗口的候选行中选择所分析的行。在本例中，最接近预定访问日的记录将被分析。
Table 4.5.4.1 Example 1: ADaM Dataset with Identification of Rows Used for Analysis when Multiple Visits Fall Within a Visit Window
表4.5.4.1示例1：ADaM数据集，其中包含多次访问属于访问窗口时用于分析的行的标识
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
PARAM |
AVAL |
DTYPE |
ANL01FL |
AWTARGET |
AWTDIFF |
| 1 |
1001 |
Screening |
Baseline |
-5 |
SUPINE SYSBP (mm Hg) |
144 |
1 |
5 |
||
| 2 |
1001 |
Baseline |
Baseline |
1 |
SUPINE SYSBP (mm Hg) |
145 |
Y |
1 |
0 |
|
| 3 |
1001 |
Week 1 |
Week 1 |
7 |
SUPINE SYSBP (mm Hg) |
130 |
Y |
7 |
0 |
|
| 4 |
1001 |
Week 2 |
Week 2 |
12 |
SUPINE SYSBP (mm Hg) |
133 |
Y |
14 |
2 |
|
| 5 |
1001 |
Week 3 |
Week 2 |
17 |
SUPINE SYSBP (mm Hg) |
125 |
14 |
3 |
||
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
PARAM |
AVAL |
DTYPE |
ANL01FL |
AWTARGET |
AWTDIFF |
| 6 |
1001 |
Week 4 |
Week 4 |
30 |
SUPINE SYSBP (mm Hg) |
128 |
Y |
28 |
2 |
Example 2
例子2
Identification of rows used for analysis – visit falls outside of a target window.
标识用于分析的行–访问不在目标窗口之内。
In this example (Table 4.5.4.2), the Week 3 visit for Subject 1001 was outside the day window of analysis Week 3, so "Post-Study" was assigned to AVISIT. This visit as well as the first baseline visit were excluded from the analysis per the SAP. The "Worst Post-Baseline" analysis timepoint (Row 6) was imputed by worst observed case (DTYPE=WOC). The "Endpoint" row was derived using the "Week 2" visit, since it was the last available eligible observation based on the SAP. Both of the derived rows are flagged with ANL01FL=Y since they were rows selected for analysis.
在此示例中（表4.5.4.2），对象1001的第3周访问不在分析第3周的日窗口之内，因此将“研究后”分配给AVISIT。根据SAP，此访问以及首次基准访问均从分析中排除。“最差的基准后”分析时间点（第6行）是由观察到的最差情况（DTYPE = WOC）估算得出的。“端点”行是使用“第2周”访问得出的，因为它是基于SAP的最后一个可用的合格观察。这两个派生的行都被标记为ANL01FL = Y，因为它们是选择进行分析的行。
Table 4.5.4.2 Example 2: ADaM Dataset with Identification of Rows Used for Analysis when Visit Falls Outside of a Target Window
表4.5.4.2示例2：ADaM数据集，其中包含用于访问落在目标窗口之外的分析行的标识
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
VISITDY |
PARAM |
AVAL |
DTYPE |
ANL01FL |
| 1 |
1001 |
Screening |
Baseline |
-5 |
1 |
SUPINE SYSBP (mm Hg) |
144 |
||
| 2 |
1001 |
Baseline |
Baseline |
1 |
1 |
SUPINE SYSBP (mm Hg) |
145 |
Y |
|
| 3 |
1001 |
Week 1 |
Week 1 |
7 |
7 |
SUPINE SYSBP (mm Hg) |
150 |
Y |
|
| 4 |
1001 |
Week 2 |
Week 2 |
12 |
14 |
SUPINE SYSBP (mm Hg) |
133 |
Y |
|
| 5 |
1001 |
Week 3 |
Post-Study |
40 |
21 |
SUPINE SYSBP (mm Hg) |
140 |
||
| 6 |
1001 |
Week 1 |
Worst Post-Baseline |
7 |
7 |
SUPINE SYSBP (mm Hg) |
150 |
WOC |
Y |
| 7 |
1001 |
Week 2 |
Endpoint |
12 |
14 |
SUPINE SYSBP (mm Hg) |
133 |
LOV |
Y |
Example 3
例子3
Identification of rows used for analysis – a visit not flagged for the analysis is used to create imputed LOCF rows.
标识用于分析的行–未标记为分析的访问用于创建估算的LOCF行。
This example (Table 4.5.4.3) illustrates a scenario where two visits occur within a window (Week 2). The first record (on Row 4) is analyzed as is (it is the record chosen to represent analysis timepoint Week 2 based on an algorithm defined in the SAP and referred to in the metadata of ANL01FL). The second Week 2 timepoint record (on Row 5) is the basis for the LOCF derivation of analysis timepoints Week 3, 4 and 5 (Rows 6, 7, and 8). In the LOCF analysis, Week 2 is based on the observed data on Row 4, and Weeks 3, 4, and 5 are imputed using the last available observation on Row 5.
此示例（表4.5.4.3）说明了一个窗口中发生两次访问的情况（第2周）。第一条记录（第4行）按原样进行分析（根据SAP中定义的算法，该记录被选择代表第2周的分析时间点，并在ANL01FL的元数据中引用该记录）。第二个第2周时间点记录（第5行）是LOCF得出第3、4和5周（行6、7和8）分析时间点的基础。在LOCF分析中，第2周基于第4行的观测数据，第3、4和5周是使用第5行的最后可用观测值估算的。
Table 4.5.4.3 Example 3: ADaM Dataset with a Value that is Carried Forward but Not Included in the Analysis
表4.5.4.3示例3：ADaM数据集的值被转发但未包括在分析中
| Row |
USUBJID |
VISIT |
AVISIT |
ADY |
PARAM |
AVAL |
DTYPE |
ANL01FL |
| 1 |
1001 |
Screening |
Baseline |
-5 |
SUPINE SYSBP (mm Hg) |
144 |
||
| 2 |
1001 |
Baseline |
Baseline |
1 |
SUPINE SYSBP (mm Hg) |
145 |
Y |
|
| 3 |
1001 |
Week 1 |
Week 1 |
7 |
SUPINE SYSBP (mm Hg) |
130 |
Y |
|
| 4 |
1001 |
Week 2 |
Week 2 |
12 |
SUPINE SYSBP (mm Hg) |
133 |
Y |
|
| 5 |
1001 |
Week 3 |
Week 2 |
17 |
SUPINE SYSBP (mm Hg) |
125 |
||
| 6 |
1001 |
Week 3 |
Week 3 |
17 |
SUPINE SYSBP (mm Hg) |
125 |
LOCF |
Y |
| 7 |
1001 |
Week 3 |
Week 4 |
17 |
SUPINE SYSBP (mm Hg) |
125 |
LOCF |
Y |
| 8 |
1001 |
Week 3 |
Week 5 |
17 |
SUPINE SYSBP (mm Hg) |
125 |
LOCF |
Y |
Approaches Considered and Not Adopted
考虑和不采用的方法
Another option considered was to create unique values of the timepoint window description AVISIT. For example, add an asterisk to the end of AVISIT such as "Week 2*" if not analyzed. This approach might be less confusing because one would not need to be aware of a flag. The disadvantage is that one would need to have a convention for AVISIT values, and tools would need to parse values of AVISIT for correct results to be generated. For these reasons, this approach was not chosen.
考虑的另一种选择是创建时间点窗口描述AVISIT的唯一值。例如，如果未分析，请在AVISIT的末尾添加一个星号，例如“ Week 2 *”。这种方法可能会减少混乱，因为不需要知道标志。缺点是需要对AVISIT值有一个约定，并且工具需要解析AVISIT的值才能生成正确的结果。由于这些原因，未选择此方法。
4.6 Identification of Population-Specific Analyzed Records
4.6 确定特定人群的分析记录
It is not uncommon in the statistical analysis of clinical trials to conduct analyses based on multiple populations of interest. The population of interest can be defined either at the subject level, the record (measurement) level, or both. For example, when defining an analysis population, a subject may be included in one analysis population such as Intent-to-Treat but may be excluded from another analysis population such as Per-Protocol. Analysis populations may also be defined using characteristics of individual measurements. For example, a measurement that was assessed outside of a pre-specified time window for a particular visit may not be included in a per-protocol visit- level population. In this section, it is assumed that the definition of a record-level analysis population is dependent on the definition of the subject-level population. In other words, if a subject is excluded from the subject-level Per- Protocol population, then none of that subject's records would be candidates for inclusion within the record-level Per-Protocol population. Given the variety of possible population definitions, the same record in an analysis dataset could be included in one analysis and excluded from another, depending on characteristics of the subject as a whole and the characteristics of the individual measurement. Therefore, the issue becomes how best to select records for each analysis.
在临床试验的统计分析中，基于多个目标人群进行分析并不少见。可以在主题级别，记录（度量）级别或两者上定义感兴趣的人群。例如，当定义分析人群时，受试者可以被包括在一个分析人群中，例如意图治疗，但是可以从另一个分析人群中被排除，例如Per-Protocol。还可以使用单个测量的特征来定义分析总体。例如，针对特定访问的在预定义时间窗口之外评估的度量可能不包括在按协议访问级别的总体中。在本节中，假定记录级别分析总体的定义取决于主题级别总体的定义。换句话说，如果某个受试者被排除在受试者级别的按协议人群中，则该受试者的记录都不会成为包含在该记录级别的按协议人群中的候选对象。给定各种可能的人口定义，一个分析数据集中的同一条记录可以包含在一项分析中，而从另一项分析中排除，这取决于对象的整体特征和单个测量的特征。因此，问题就变成了如何最好地为每个分析选择记录。分析数据集中的同一记录可以包含在一个分析中，而可以从另一个分析中排除，这取决于对象的整体特征和单个测量的特征。因此，问题就变成了如何最好地为每个分析选择记录。分析数据集中的同一记录可以包含在一个分析中，而可以从另一个分析中排除，这取决于对象的整体特征和单个测量的特征。因此，问题就变成了如何最好地为每个分析选择记录。
ADaM Methodology
ADaM方法论
The ADaM methodology for this analysis issue is to create a single ADaM dataset that can be used to perform multiple analyses using population flag variables to identify records that are used for each type of analysis. An advantage of this approach is that this single ADaM dataset can be used for multiple analyses. Flag variables obviate the need to replicate records for each type of analysis. This approach promotes efficiency in the operational aspects of electronic submissions, clarity of analyses, and ease in comparing selected values for each population.
This approach does, however, require that clear metadata be provided for the flag variables so that each specific analysis can be reproduced accurately. Below are several examples of the use of population flag variables to identify records used for different analyses.
针对此分析问题的ADaM方法论是创建一个单个ADaM数据集，该数据集可用于使用总体标记变量来执行多种分析，以识别用于每种分析类型的记录。这种方法的优势在于，该单个ADaM数据集可用于多个分析。标志变量消除了为每种类型的分析复制记录的需要。这种方法提高了电子提交的操作方面的效率，分析的清晰性以及易于比较每个人群的选定值。但是，此方法确实需要为标记变量提供清晰的元数据，以便可以准确地复制每个特定的分析。以下是使用种群标记变量来标识用于不同分析的记录的几个示例。
Example 1
例子1
Use of subject-level flag variables (ITTFL and PPROTFL) and record-level flag variables (ANL01FL and PPROTRFL).
使用主题级别的标志变量（ITTFL和PPROTFL）和记录级别的标志变量（ANL01FL和PPROTRFL）。
In some statistical analyses, even if a subject is included in the Per-Protocol population, some or all data for that subject in a particular dataset may not be appropriate for a per-protocol analysis. Consider a situation in HIV studies where a Per-Protocol analysis excludes all data after permanent discontinuation of study medication or addition of other antiretroviral therapy. An example of an ADaM dataset to support this type of analysis is illustrated in Table
在某些统计分析中，即使一个受试者包含在按方案人群中，特定数据集中该受试者的某些或全部数据可能也不适合按方案进行分析。考虑一下HIV研究中的一种情况，根据方案分析将永久终止研究药物或添加其他抗逆转录病毒疗法后的所有数据排除在外。表中说明了支持这种分析的ADaM数据集的示例。
This ADaM dataset (Table 4.6.1) can be used to repeat analyses based on multiple populations of interest either at the subject level or at the record (measurement) level.
此ADaM数据集（表4.6.1）可用于在主题级别或记录（度量）级别基于多个感兴趣的人群重复分析。
ITTFL and PPROTFL are subject-level analysis population flags. If a subject is in the Intent-to-Treat population, then the column ITTFL will have the value of "Y" ("N" if not). In Table 4.6.1, subjects 1001, 1002, and 1003 are in the Intent-to-Treat population. Similarly, if a subject is in the Per-Protocol population, the column PPROTFL will have the value of "Y" ("N" if not). Subjects 1001 and 1003 in Table 4.6.1 are in the Per-Protocol population while Subject 1002 with PPROTFL=N is excluded from any Per-Protocol analysis. These indicator variables are used to identify individual subjects that belong to each subject-level population.
ITTFL和PPROTFL是受试者级别的分析人群标志。如果受试者在意向治疗人群中，则ITTFL列的值为“ Y”（如果不是，则为“ N”）。在表4.6.1中，受试者1001、1002和1003在“意向治疗”人群中。类似地，如果受试者属于按协议人群，则PPROTFL列的值将为“ Y”（否则为“ N”）。表4.6.1中的受试者1001和1003在按方案人群中，而具有PPROTFL = N的受试者1002被排除在任何按方案分析之外。这些指标变量用于识别属于每个学科水平人群的个体学科。
In contrast to the subject-level population flags, the column PPROTRFL is the per-protocol analysis flag at the record level. As illustrated in Table 4.6.1, if a record is a candidate for the Per-Protocol analysis, the variable PPROTRFL is set to "Y"; it is null if the record does not fulfill the criteria for this analysis. In the example, Subjects 1001 and 1002 continue with study medication after Week 2; the last dose of study medication for Subject 1003 is at Week 1. In Table 4.6.1, all three records for Subject 1002 and two of four records for Subject 1003 are not record- level Per-Protocol data and would not be selected for a Per-Protocol analysis when we apply the subset condition: PPROTRFL="Y". PPROTRFL is null on the last two records for Subject 1003 and will be excluded from any record-level Per-Protocol data analysis as they occur after the subject discontinued study medication.
与主题级别的填充标志相反，列PPROTRFL是记录级别的按协议分析标志。如表4.6.1所示，如果记录是按协议分析的候选者，则变量PPROTRFL设置为“ Y”；如果记录不满足此分析的条件，则为null。在该示例中，受试者1001和1002在第2周后继续接受研究药物治疗；受试者1003的研究药物的最后剂量是在第1周。在表4.6.1中，受试者1002的所有三个记录和受试者1003的四个记录中的两个都不是记录级别的按协议数据，因此不会选择当我们应用子集条件：PPROTRFL =“ Y”时，按协议进行分析。
Not all records in Table 4.6.1 are included for analysis purposes. In this example, the analysis flag ANL01FL is null for one record (USUBJID=1003, VISIT=Week 1, AVISIT=Week 1, AVAL=999) because its value was replaced by the retest result in the next record (USUBJID=1003, VISIT=Retest, AVISIT=Week 1, AVAL=49). The analysis flag for the Retest record is Y.
并非出于分析目的就包含了表4.6.1中的所有记录。在此示例中，对于一个记录（USUBJID = 1003，VISIT =第1周，VISIT =第1周，AVAL = 999），分析标志ANL01FL为空，因为它的值已替换为下一个记录（USUBJID = 1003， VISIT =重新测试，AVISIT =第1周，AVAL = 49）。重新测试记录的分析标记为Y。
Table 4.6.1 Example 1: ADaM Dataset with Subject-Level and Record-Level Indicator Variables
表4.6.1示例1：具有主题级和记录级指示符变量的ADaM数据集
| Row |
USUBJID |
ITTFL |
PPROTFL |
VISIT |
AVISIT |
PARAMCD |
AVAL |
ANL01FL |
PPROTRFL |
| 1 |
1001 |
Y |
Y |
Week 0 |
Week 0 |
TEST1 |
500 |
Y |
Y |
| 2 |
1001 |
Y |
Y |
Week 1 |
Week 1 |
TEST1 |
400 |
Y |
Y |
| 3 |
1001 |
Y |
Y |
Week 2 |
Week 2 |
TEST1 |
600 |
Y |
Y |
| 4 |
1002 |
Y |
N |
Week 0 |
Week 0 |
TEST1 |
500 |
Y |
|
| 5 |
1002 |
Y |
N |
Week 2 |
Week 1 |
TEST1 |
48 |
Y |
|
| 6 |
1002 |
Y |
N |
Week 2 |
Week 2 |
TEST1 |
46 |
Y |
|
| 7 |
1003 |
Y |
Y |
Week 0 |
Week 0 |
TEST1 |
999 |
Y |
Y |
| 8 |
1003 |
Y |
Y |
Week 1 |
Week 1 |
TEST1 |
999 |
Y |
|
| 9 |
1003 |
Y |
Y |
Retest |
Week 1 |
TEST1 |
49 |
Y |
|
| 10 |
1003 |
Y |
Y |
Week 2 |
Week 2 |
TEST1 |
499 |
Y |
To identify records used for an Intent-to-Treat analysis for parameter code "TEST1" at Week 1 requires the following selection specification:
要确定在第1周用于参数代码“ TEST1”的意图分析的记录，需要以下选择规范：
| AVISIT="Week 1" & PARAMCD="TEST1" & ANL01FL="Y" & ITTFL="Y" |
Similarly, to identify records used for a Per-Protocol analysis of values of for parameter code "TEST1" at Week 1 requires the following selection specification:
同样，要识别用于在第1周对参数代码“ TEST1”的Per-Protocol分析值的记录，需要以下选择规范：
| AVISIT="Week 1" & PARAMCD="TEST1" & ANL01FL="Y" & PPROTRFL="Y" |
Because an error in the specification of the selection for either of the above conditions will yield incorrect results, it is important that the metadata be clear for each indicator variable. In addition, ADaM analysis results metadata will specify the selection criteria to provide clear documentation of how the indicator variables were used to select analyzed records for identified analyses.
由于以上两种情况之一的选择说明中的错误都会产生错误的结果，因此对于每个指标变量而言，清除元数据非常重要。此外，ADaM分析结果元数据将指定选择标准，以提供有关如何使用指标变量选择已确定分析的记录的清晰文档。
Example 2 例子2
Use of subject-level indicator variables and parameter-level indicator variables. 使用主题级别的指标变量和参数级别的指标变量。
For the purposes of this example, it is assumed that the producer's SAP included a definition of an efficacy analysis population, defining it as consisting of subjects with a baseline efficacy assessment and at least one post-baseline efficacy assessment, without restriction to a specific assessment. In this example, there are two efficacy parameters (Test 1 and Test 2), and three visits (Week -1, Baseline, and Week 2). The subjects whose data is presented in Table 4.6.3 have results for the assessments as noted in Table 4.6.2.
就本示例而言，假设生产者的SAP包括功效分析总体的定义，将其定义为包括具有基线功效评估和至少一项基准后功效评估的受试者，而不受特定评估的限制。在此示例中，有两个功效参数（测试1和测试2），以及三个访问（第-1周，基线和第2周）。表4.6.3中提供了数据的受试者的评估结果列于表4.6.2中。
Table 4.6.2 Example 2: Data Available for Each Subject in Illustration
表4.6.2示例2：插图中每个主题可用的数据
| Subject |
Does Subject have a Baseline TEST1 Assessment? |
Does Subject have a Post- Baseline TEST1 Assessment? |
EFFPFL for TEST1 |
Does Subject have a Baseline TEST2 Assessment? |
Does Subject have a Post- Baseline TEST2 Assessment? |
EFFPFL for TEST2 |
EFFFL |
| 1001 |
Y |
Y |
Y |
Y |
Y |
Y |
Y |
| 1002 |
Y |
Y |
Y |
N |
N |
Y |
© 2019 Clinical Data Interchange Standards Consortium, Inc. All rights reserved Page 93
2019-10-03
CDISC Analysis Data Model Implementation Guide (Version 1.2 Final)
| Subject |
Does Subject have a Baseline TEST1 Assessment? |
Does Subject have a Post- Baseline TEST1 Assessment? |
EFFPFL for TEST1 |
Does Subject have a Baseline TEST2 Assessment? |
Does Subject have a Post- Baseline TEST2 Assessment? |
EFFPFL for TEST2 |
EFFFL |
| 1003 |
Y |
N |
Y |
Y |
Y |
Y |
|
| 1004 |
Y |
N |
Y |
N |
N |
In contrast to subject-level population flags, the column EFFPFL is a parameter-level population flag. A subject is included in the efficacy analysis population for a specific parameter if the subject has a baseline efficacy assessment and at least one post-baseline efficacy assessment for that parameter. If a subject is eligible for the efficacy analysis for the specific parameter, the variable EFFPFL is set to "Y" for the subject's records within the parameter; in this example, it is null if the subject is not a candidate for the analysis of the parameter. In Table 4.6.3, the efficacy analysis population for TEST1 includes Subjects 1001 and 1002; the efficacy analysis population for TEST2 includes Subjects 1001 and 1003.
与主题级别的总体标志相反，EFFPFL列是参数级别的总体标志。如果受试者具有基线功效评估和针对该参数的至少一项基线后功效评估，则该受试者包括在特定参数的功效分析人群中。如果某个受试者符合特定参数的功效分析的条件，则对于该参数内的受试者记录，变量EFFPFL设置为“ Y”；在此示例中，如果主题不是参数分析的候选者，则为null。在表4.6.3中，TEST1的功效分析人群包括受试者1001和1002；受试者1001和1002。TEST2的功效分析人群包括受试者1001和1003。
Table 4.6.3 Example 2: ADaM Dataset with Subject-Level and Parameter-Level Indicator Variables
表4.6.3示例2：具有主题级别和参数级别指示符变量的ADaM数据集
| Row |
USUBJID |
EFFFL |
AVISIT |
PARAMCD |
AVAL |
EFFPFL |
| 1 |
1001 |
Y |
Wk -1 |
TEST1 |
500 |
Y |
| 2 |
1001 |
Y |
Bsln |
TEST1 |
500 |
Y |
| 3 |
1001 |
Y |
Wk 2 |
TEST1 |
600 |
Y |
| 4 |
1001 |
Y |
Wk -1 |
TEST2 |
10 |
Y |
| 5 |
1001 |
Y |
Bsln |
TEST2 |
10 |
Y |
| 6 |
1001 |
Y |
Wk 2 |
TEST2 |
12 |
Y |
| 7 |
1002 |
Y |
Wk -1 |
TEST1 |
500 |
Y |
| 8 |
1002 |
Y |
Bsln |
TEST1 |
500 |
Y |
| 9 |
1002 |
Y |
Wk 2 |
TEST1 |
46 |
Y |
| 10 |
1002 |
Y |
Wk -1 |
TEST2 |
11 |
|
| 11 |
1003 |
Y |
Wk -1 |
TEST1 |
780 |
|
| 12 |
1003 |
Y |
Bsln |
TEST1 |
799 |
|
| 13 |
1003 |
Y |
Wk -1 |
TEST2 |
28 |
Y |
| 14 |
1003 |
Y |
Bsln |
TEST2 |
30 |
Y |
| 15 |
1003 |
Y |
Wk 2 |
TEST2 |
32 |
Y |
| 16 |
1004 |
N |
Wk -1 |
TEST1 |
250 |
|
| 17 |
1004 |
N |
Bsln |
TEST1 |
300 |
|
| 18 |
1004 |
N |
Wk -1 |
TEST2 |
15 |
|
| 19 |
1004 |
N |
Bsln |
TEST2 |
15 |
CDISC Analysis Data Model Implementation Guide (Version 1.2 Final)
4.7 确定满足用于分析目的的预定标准的记录
For analysis purposes, criteria are often defined to group results based on the collected value's relationship to one or more algorithmic conditions. For example, subjects who had a result greater than 5 times the upper limit of the normal range or subjects who had a systolic blood pressure value >160 mmHg with at least a 25-point increase from the BASE value. In addition to creating subgroups of subjects, the categorization of the presence or absence of a criterion is often used in listings, tabular displays, or statistical modeling (as a covariate or a response variable).
出于分析目的，通常定义标准以根据收集的值与一种或多种算法条件的关系对结果进行分组。例如，结果大于正常范围上限的5倍的受试者，或收缩压值> 160 mmHg且比BASE值至少升高25点的受试者。除了创建主题的子组外，列表，表格显示或统计建模（作为协变量或响应变量）中经常使用对标准存在或不存在的分类。
ADaM Methodology
ADaM方法论
ADaM methodology provides an analysis criterion variable, CRITy, paired with a criterion evaluation result flag, CRITyFL, to identify whether a criterion is met. These variables are defined in Section 3.3.4, Analysis Parameter Variables for BDS Datasets. The variables MCRITy and MCRITyML are defined in Section 3.3.4, Analysis Parameter Variables for BDS Datasets, for use in situations where the criterion can have multiple responses (as opposed to CRITy, which has binary responses).
ADaM方法提供了一个分析标准变量CRITy，与标准评估结果标志CRITyFL配对，以识别是否满足标准。这些变量在部分3.3.4，BDS数据集的分析参数变量中定义。变量MCRITy和MCRITyML在第3.3.4节“ BDS数据集的分析参数变量”中定义，用于在条件可能具有多个响应（而不是具有二进制响应的CRITy）的情况下使用。
CRITy is populated with a text description defining the conditions necessary to satisfy the presence of the criterion. The definition of CRITy can use any variable(s) located on the row, and the definition must stay constant across all rows within the same value of PARAM. A complex criterion which draws from multiple rows (different parameters or multiple rows for a single parameter) will require a new PARAM be created (see Example 3).
CRITy中填充了文本描述，该文本描述定义了满足标准的条件所需的条件。CRITy的定义可以使用该行上的任何变量，并且该定义必须在同一PARAM值内的所有行中保持不变。从多行（不同的参数或单个参数的多行）得出的复杂标准将需要创建一个新的PARAM（请参见示例3）。
CRITyFL, "Criterion Evaluation Result Flag," is the character indicator of whether the criterion described in CRITy was met. Variable CRITyFL must be present on the dataset if variable CRITy is present. CRITyFN is permitted if a numeric result flag is needed.
CRITyFL，“标准评估结果标志”，是指示是否满足CRITy中描述的标准的字符指示器。如果存在变量CRITy，则变量CRITyFL必须存在于数据集上。如果需要数字结果标志，则允许CRITyFN。
ADaM methodology allows the option of only populating CRITy on a row if the CRITy criterion is met for that row (see Example 1). In that case, CRITyFL is set to "Y" only if CRITy is populated and is null otherwise. If this option is not used and CRITy is populated on all rows within the parameter (see Example 2), then CRITyFL is set to "Y" or "N" or null. The choice of populating CRITy on only the rows where the criteria is met versus on all rows is dependent on the analysis need, as shown in the examples that follow.
如果满足该行的CRITy标准，则ADaM方法仅允许在该行中填充CRITy（请参见示例1）。在这种情况下，仅当填充了CRITy时，CRITyFL才设置为“ Y”，否则为null。如果不使用此选项，并且CRITy填充在参数中的所有行上（请参见示例2），则CRITyFL设置为“ Y”或“ N”或为null。仅在满足条件的行上填充CRITy，而不是在所有行上填充CRITy取决于分析需求，如以下示例所示。
CRITy and CRITyFL facilitate subgroup analyses. The ADaM methodology does not preclude the addition of rows (in contrast to the addition of multiple CRITy and CRITyFL columns) to the BDS for the criterion CRITy. However, CRITy must be kept constant (if populated) across all rows within the same value of PARAM.
CRITy和CRITyFL促进了亚组分析。ADaM方法并不排除将行CRITy的行添加到BDS中（与添加多个CRITy和CRITyFL列相反）。但是，必须在同一PARAM值内的所有行中将CRITy保持恒定（如果已填充）。
CRITy, CRITyFL, and CRITyFN are not parameter-invariant in that CRITy can vary across parameters within a dataset, as can the Controlled Terminology used for the corresponding CRITyFL and CRITyFN. In other words, CRITy for one parameter can be different than CRITy for a different parameter in the same dataset. (See Example 8: Categorical Analysis of Subjects Meeting Hy's Law Criteria in the document "ADaM Examples in Commonly Used Statistical Analysis Methods", available at https://www.cdisc.org/standards/foundational/analysis-data-model- adam/adam-examples-commonly-used-statistical-analysis.)
CRITy，CRITyFL和CRITyFN并不是参数不变的，因为CRITy可以在数据集中的各个参数之间变化，用于相应CRITyFL和CRITyFN的受控术语也可以。换句话说，一个参数的CRITy可以不同于同一数据集中不同参数的CRITy。（见例8：主题会议的海氏法则标准在文件“中常用的统计分析方法ADAM实例”，可用的范畴分析在https://www.cdisc.org/standards/foundational/analysis-data-model- 亚当/ adam-examples-常用统计分析。）
Example
CRITy populated only when criterion met.
例
仅在满足条件时填充CRITy。
Using this approach, when a criterion is defined for a PARAM but conditions are not met on a specific row, both CRITy and CRITyFL are set to null. CRITy and CRITyFL are also set to null if one or more missing data inputs to a criterion result in an unevaluable criterion (unevaluability is producer-defined, and is not necessarily triggered by missing data inputs).
使用这种方法，当为PARAM定义了标准但在特定行上不满足条件时，CRITy和CRITyFL都设置为null。如果一个标准的一个或多个缺失数据输入导致无法评估的标准，则CRITy和CRITyFL也设置为null（不可评估性是生产者定义的，不一定由缺失的数据输入触发）。
One purpose of this option is to facilitate subsetting within a parameter when the interest is in the subgroup of subjects who fulfilled the criterion. It is also relevant when simple counts of criteria are desired. The following conditions must be true when this option is used:
此选项的一个目的是当感兴趣的对象符合条件的子组时，便于在参数内进行子集。当需要简单计数标准时，它也很重要。使用此选项时，必须满足以下条件：
-
-
-
- Variables CRITy and CRITyFL are present on the dataset.
-
-
变量CRITy和CRITyFL存在于数据集中。
-
-
-
- Analysis Variable Metadata defines CRITy relative to the specific parameter.
-
-
分析变量元数据相对于特定参数定义了CRITy。
-
-
-
- CRITy and CRITyFL are set to null for rows within the parameter where the criterion is not met or is unevaluable.
-
-
对于参数中不符合条件或无法评估的行，将CRITy和CRITyFL设置为null。
Table 4.7.1.1 illustrates ADaM methodology option "CRITy populated only when criterion met". The presence of a value in CRIT1 indicates Subject 1001 satisfied the criterion. With this option, CRIT1 facilitates subsetting when the interest is in the subgroup of subjects who fulfilled the criterion. The null value in CRIT1 is because Subject 1002 did not satisfy the criterion. The null value in CRIT1 is because the criterion is unevaluable due to missing inputs for Subject 1003.
表4.7.1.1说明了ADaM方法选项“仅在满足条件时才填充”。CRIT1中存在值表示受试者1001满足标准。使用此选项，当兴趣位于满足条件的主题子组中时，CRIT1有助于子集。CRIT1中的空值是因为主题1002不满足该条件。CRIT1中的空值是因为缺少主题1003的输入，因此该标准无法评估。
Table 4.7.1.1 Example 1: ADaM Dataset with CRITy Populated Only when Criterion Met
表4.7.1.1示例1：仅在满足标准时才填充具有CRITy的ADaM数据集
| Row |
USUBJID |
PARAM |
AVAL |
BASE |
CHG |
CRIT1 |
CRIT1FL |
| 1 |
1001 |
Systolic Blood Pressure (mm Hg) |
163 |
148 |
15 |
Systolic Pressure >160 |
Y |
| 2 |
1002 |
Systolic Blood Pressure (mm Hg) |
140 |
148 |
-8 |
||
| 3 |
1003 |
Systolic Blood Pressure (mm Hg) |
120 |
Example 2
例子2
CRITy populated on all rows within a parameter.
CRITy填充在参数内的所有行上。
Using this approach, CRITy is populated on all rows within the parameter and CRITyFL is set to "Y" or "N" or null. The purpose of this option is to facilitate analyses where the criterion is used in tabular displays and/or statistical modeling for the parameter.
使用这种方法，将CRITy填充到参数中的所有行上，并将CRITyFL设置为“ Y”或“ N”或为null。此选项的目的是促进分析在表格显示和/或参数的统计建模中使用标准的情况。
Table 4.7.1.2 illustrates the ADaM methodology option "CRITy populated on all rows within a parameter". Since this criterion is used for modeling or analysis in this example, it is necessary to populate the rows which fail to satisfy the criterion. CRIT1FL indicates whether or not the subject meets the criterion. CRIT1FL is set to null for Subject 1005 because the criterion is unevaluable due to missing input(s).
表4.7.1.2说明了ADaM方法选项“在参数内的所有行上填充的CRITy”。由于在此示例中将此准则用于建模或分析，因此必须填充不满足该准则的行。CRIT1FL指示受试者是否符合标准。对于主题1005，CRIT1FL设置为null，因为由于缺少输入，因此该标准无法评估。
Table 4.7.1.2 Example 2: ADaM Dataset with CRITy Populated on All Rows Within a Parameter
表4.7.1.2示例2：在参数内的所有行上填充有CRITy的ADaM数据集
| Row |
USUBJID |
PARAM |
AVAL |
BASE |
CHG |
CRIT1 |
CRIT1FL |
| 1 |
1001 |
Systolic Blood Pressure (mm Hg) |
163 |
148 |
15 |
Systolic Pressure >160 and Change from Baseline in Systolic Pressure>10 |
Y |
| 2 |
1002 |
Systolic Blood Pressure (mm Hg) |
140 |
148 |
-8 |
Systolic Pressure >160 and Change from Baseline in Systolic Pressure>10 |
N |
| 3 |
1005 |
Systolic Blood Pressure (mm Hg) |
120 |
Systolic Pressure >160 and Change from Baseline in Systolic Pressure>10 |
Example 3
例3：
Compound Criteria化合物标准
If the definition of a criterion uses values located on multiple rows (different parameters or multiple rows for a single parameter), then a new row must be added with the value of PARAM being the textual description of the criterion (see Section 4.2.1, Rule 4: A function of multiple rows within a parameter should be added as a new parameter. and Rule 5: A function of more than one parameter should be added as a new parameter.). The text of PARAM is producer-defined and can be as long or as short as needed to be meaningful, within the 200-character limitation for the column.
如果标准的定义使用位于多行上的值（不同的参数或单个参数的多行），则必须添加新行，并以PARAM的值作为标准的文字描述（请参见第4.2.1节，规则4：一个参数中有多个行的函数应作为新参数添加；规则5：一个以上参数的函数应作为新参数添加。PARAM的文本是生产者定义的，可以有意义，可以在该列的200个字符的范围内，根据需要长或短。
For compound criterion rows, AVALC must always be populated with Y/N/null. If an analysis also requires a numeric indicator variable, either of the following two options may be chosen:
对于复合标准行，AVALC必须始终用Y / N / null填充。如果分析还需要数字指标变量，则可以选择以下两个选项之一：
-
-
-
- CRITy may be set to the same criterion text as PARAM, CRITyFL set to the same Y/N/null value as AVALC, and CRITyFN set to 1/0/null.
-
-
1可以将CRITy设置为与PARAM相同的标准文本，将CRITyFL设置为与AVALC相同的Y / N / null值，并将CRITyFN设置为1/0 / null
-
-
-
- AVAL may be set to a numeric 1/0/null indicator value.
-
-
2 AVAL可以设置为数字1/0 /空指示符值。
If an analysis requires only simple subsetting of the "hits" on a particular compound criterion, it is acceptable to add only the "compound criterion met" (AVALC="Y") rows to the dataset. If this option is chosen, rows are not added where the assessment of a compound criterion in PARAM would result in AVALC="N" or null.
如果分析仅需要对特定复合标准上的“命中”进行简单的子集设置，则可以仅将“满足复合标准”（AVALC =“ Y”）行添加到数据集中。如果选择此选项，那么在PARAM中评估复合标准将导致AVALC =“ N”或null的情况下，不添加行。
Note that if a compound criterion is defined, then its components do not have to exist on their own in the dataset unless these components are themselves used for subsetting, display, or modeling purposes, or are needed for traceability.
请注意，如果定义了复合条件，则除非数据成分本身用于子集，显示或建模目的或可追溯性所需，否则其成分不必在数据集中单独存在。
Table 4.7.1.3 illustrates a compound criterion (Row 3) included in the same dataset with non-compound criteria (Rows 1 and 2). For the compound criterion row, AVALC is populated with Y to indicate the compound criterion has been met. Note it is not necessary to populate CRITy, CRITyFL, or CRITyFN on the compound criterion row since the values would be redundant with what is already stored in PARAM, AVALC, and AVAL, respectively.
表4.7.1.3说明了包含在具有非复合标准（行1和2）的同一数据集中的复合标准（行3）。对于复合标准行，将使用Y填充AVALC以指示已满足复合标准。请注意，不必在复合标准行上填充CRITy，CRITyFL或CRITyFN，因为这些值对于分别存储在PARAM，AVALC和AVAL中的值将是多余的。
Table 4.7.1.3 Example 3: ADaM Dataset with Both Compound and Non-compound Criteria
表4.7.1.3示例3：同时具有复合标准和非复合标准的ADaM数据集
| Row |
USUBJID |
PARAM |
AVAL |
AVALC |
BASE |
CHG |
CRIT1 |
CRIT1FL |
CRIT1FN |
CRIT2 |
CRIT2FL |
CRIT2FN |
| 1 |
1001 |
Systolic Blood Pressure (mm Hg) |
163 |
148 |
15 |
Systolic Pressure > 160 |
Y |
1 |
Change from Baseline in Systolic Pressure > 10 |
Y |
1 |
|
| 2 |
1001 |
Diastolic Blood Pressure (mm Hg) |
96 |
87 |
9 |
Diastolic Pressure > 95 |
Y |
1 |
||||
| 3 |
1001 |
Systolic Pressure >160 and Diastolic Pressure > 95 |
Y |
Note that criterion "Diastolic Pressure >95" (Row 2) can coexist in the same CRIT1 column with "Systolic Pressure >160" (Row 1). Each of these criteria is specific to its own subset of PARAM rows.
请注意，标准“舒张压> 95”（行2）可以与“收缩压> 160”（行1）共存于同一CRIT1列中。这些标准中的每一个都特定于其自己的PARAM行子集。
ADaM Methodology
ADaM方法论
ADaM methodology provides an analysis criterion variable, MCRITy, paired with a criterion evaluation result flag, MCRITyML ("Multi-Response Criterion y Evaluation"), to identify which level of a multiple response criterion is met. These variables are defined in Section 3.3.4, Analysis Parameter Variables for BDS Datasets.
ADaM方法提供了分析标准变量MCRITy，并与标准评估结果标志MCRITyML（“多响应标准y评估”）配对，以识别满足多响应标准的哪个级别。这些变量在部分3.3.4， BDS数据集的分析参数变量中定义。
MCRITy is populated with a text description identifying the criterion being evaluated. The definition of MCRITy can use any variable(s) located on the row and the definition must stay constant across all rows within the same value of PARAM. A complex criterion which draws from multiple rows (different parameters or multiple rows for a single parameter) will require a new PARAM be created.
MCRITy填充有文本描述，用于标识要评估的标准。MCRITy的定义可以使用该行上的任何变量，并且该定义必须在同一PARAM值内的所有行中保持不变。从多个行（不同的参数或单个参数的多个行）得出的复杂标准将需要创建一个新的PARAM。
MCRITyML is the character flag variable that indicates which level of the criterion defined in MCRITy was met. Variable MCRITyML must be present on the dataset if variable MCRITy is present. MCRITyMN is permitted if a numeric result flag is needed.
MCRITyML是字符标志变量，它指示满足了MCRITy中定义的标准的哪个级别。如果存在变量MCRITy，则数据集上必须存在变量MCRITyML。如果需要数字结果标志，则允许MCRITyMN。
MCRITy and MCRITyML facilitate subgroup analyses. The ADaM methodology does not preclude the addition of rows (in contrast to the addition of multiple MCRITy and MCRITyML columns) to the BDS for the criterion MCRITy. However, MCRITy must be kept constant (if populated) across all rows within the same value of PARAM.
MCRITy和MCRITyML促进了亚组分析。ADaM方法并不排除将行MCRITy的行添加到BDS中（与添加多个MCRITy和MCRITyML列相反）。但是，必须在同一PARAM值内的所有行中将MCRITy保持恒定（如果已填充）。
MCRITy, MCRITyML, and MCRITyMN are not parameter-invariant in that MCRITy can vary across parameters within a dataset, as can the Controlled Terminology used for the corresponding MCRITyML and MCRITyMN. In other words, MCRITy for one parameter can be different than MCRITy for a different parameter in the same dataset.
MCRITy，MCRITyML和MCRITyMN不是参数不变的，因为MCRITy可以在数据集中的各个参数之间变化，用于相应MCRITyML和MCRITyMN的受控术语也可以。换句话说，一个参数的MCRITy可以与同一数据集中不同参数的MCRITy不同。
Example 1 例子1
Table 4.7.2.1 illustrates partial laboratory data for Alanine Aminotransferase (IU/L). As with other examples, some of the necessary columns for analysis and traceability (e.g., CHG, PCHG, AWTARGET, AWTDIFF, LBSEQ) have been excluded from the illustration. In this example, ALT values are evaluated for changes in toxicity grade. (Laboratory Grading in this example is based on CTCAE Version 4 so Grade 1 is >ULN - 3.0 x ULN, Grade 2 is >3.0 - 5.0 x ULN, Grade 3 is >5.0 - 20.0 x ULN, and Grade 4 >20.0 x ULN.) In a typical analysis situation, it is of interest to know the shift in the number of toxicity grades from baseline. Generally, an increase from a Grade 1 at baseline to a Grade 3 at a post baseline visit is treated the same as an increase from Grade 2 to Grade 4; that is, both of these are considered an increase in 2 Grades.
表4.7.2.1说明了丙氨酸转氨酶（IU / L）的部分实验室数据。与其他示例一样，用于分析和可追溯性的某些必要列（例如CHG，PCHG，AWTARGET，AWTDIFF，LBSEQ）已从插图中排除。在此示例中，评估ALT值的毒性等级变化。（此示例中的实验室评分基于CTCAE版本4，因此1级为> ULN-3.0 x ULN，2级为> 3.0-5.0 x ULN，3级为> 5.0-20.0 x ULN，而4级为> 20.0 x ULN 。）在典型的分析情况下，了解毒性等级数与基线之间的变化是很有意思的。一般而言，基线访视后从基线的1级增加到基线后的3级与从2级增加到4级相同。也就是说，这两个等级均提高了2个等级。
Note that for some laboratory analytes, only increases in toxicity grades are of interest. For other analytes, the interest is only in decreases in toxicity grades. Finally, for a few analytes, the change in toxicity in both directions (increases and decreases) is of interest.
请注意，对于某些实验室分析物，仅需要提高毒性等级。对于其他分析物，仅关注降低毒性等级。最后，对于一些分析物，两个方向的毒性变化（增加和减少）是有意义的。
In this example, it is increases in toxicity grades that are of interest. Within the analysis dataset, MCRIT1 identifies the criterion being evaluated and MCRIT1ML contains the level of criterion met, and MCRIT1MN contains a numeric version of the response. In contrast, CRIT1 assesses whether or not the value of ALT exceeded 8 times ULN.
在这个例子中，令人关注的是毒性等级的增加。在分析数据集中，MCRIT1标识要评估的标准，MCRIT1ML包含满足的标准级别，而MCRIT1MN包含响应的数字版本。相反，CRIT1评估ALT值是否超过ULN的8倍。
Note that in this example, the producer has elected to not populate the values of MCRIT1ML on the screening and baseline records. Values of MCRIT1ML represent the number of grade increases from baseline. Note that for visits where there is either no increase in toxicity grade OR where the grade decreases, MCRIT1ML is given a value of "No Grade Increase". This approach is suitable for analytes where only grade increases is of interest. Should decreases also be of interest, then a second set of MCRIT variables would be added to contain the observed number of toxicity grade decreases.
请注意，在此示例中，生产者选择不填充筛选记录和基线记录上的MCRIT1ML值。MCRIT1ML的值表示从基线开始的等级增加的数量。请注意，对于毒性等级没有增加或等级降低的访问，MCRIT1ML的值为“没有等级增加”。该方法适用于仅关注等级增加的分析物。如果降低也很有意义，那么将添加第二组MCRIT变量以包含观察到的毒性等级降低数量。
This example also illustrates other BDS variables, notably SHIFT1 and SHIFT2. SHIFT1 is defined as the shift between BNRIND and ANRIND while SHIFT2 is defined as the shift between BTOXGR and ATOXGR. These shifts of normal ranges and CTCAE toxicity grades are often of interest for analysis.
此示例还说明了其他BDS变量，尤其是SHIFT1和SHIFT2。SHIFT1定义为BNRIND和ANRIND之间的转换，而SHIFT2定义为BTOXGR和ATOXGR之间的转换。正常范围和CTCAE毒性等级的这些变化通常是分析所关注的。
Table 4.7.2.1 Example 1: ADaM Dataset with a Criterion that Has Multiple Responses
表4.7.2.1示例1：具有多个响应条件的ADaM数据集
| Row |
USUBJID |
PARAMCD |
AVISIT |
VISIT |
ADY |
AVAL |
ANRLO |
ANRHI |
ANRIND |
ATOXGR |
ABLFL |
ANL01FL |
BASE |
BNRIND |
BTOXGR |
SHIFT1 |
SHIFT2 |
MCRIT1 |
MCRIT1ML |
MCRIT1MN |
CRIT1 |
CRIT1FL |
| 1 |
ABC- 0001 |
ALT |
Baseline |
SCREENING |
-14 |
30 |
0 |
35 |
Normal |
0 |
Y |
Y |
30 |
Normal |
0 |
ALT Grade Increase |
ALT > 8*ULN |
N |
||||
| 2 |
ABC- 0001 |
ALT |
Week 1 |
WEEK 1 |
2 |
31 |
0 |
35 |
Normal |
0 |
Y |
30 |
Normal |
0 |
Normal to Normal |
0 to 0 |
ALT Grade Increase |
No Grade Increase |
0 |
ALT > 8*ULN |
N |
|
| 3 |
ABC- 0001 |
ALT |
Week 3 |
WEEK 3 |
22 |
45 |
0 |
35 |
High |
1 |
Y |
30 |
Normal |
0 |
Normal to High |
0 to 1 |
ALT Grade Increase |
Increase of 1 Grade |
1 |
ALT > 8*ULN |
N |
|
| 4 |
ABC- 0001 |
ALT |
Week 5 |
WEEK 5 |
34 |
81 |
0 |
35 |
High |
1 |
Y |
30 |
Normal |
0 |
Normal to High |
0 to 1 |
ALT Grade Increase |
Increase of 1 Grade |
1 |
ALT > 8*ULN |
N |
|
| 5 |
ABC- 0001 |
ALT |
Week 7 |
WEEK 7 |
51 |
110 |
0 |
35 |
High |
2 |
Y |
30 |
Normal |
0 |
Normal to High |
0 to 2 |
ALT Grade Increase |
Increase of 2 Grades |
2 |
ALT > 8*ULN |
N |
|
| 6 |
ABC- 0001 |
ALT |
Week 9 |
WEEK 9 |
65 |
554 |
0 |
35 |
High |
3 |
Y |
30 |
Normal |
0 |
Normal to High |
0 to 3 |
ALT Grade Increase |
Increase of 3 Grades |
3 |
ALT > 8*ULN |
Y |
|
| 7 |
ABC- 0001 |
ALT |
Month 3 |
MONTH 3 |
92 |
1077 |
0 |
35 |
High |
4 |
Y |
30 |
Normal |
0 |
Normal to High |
0 to 4 |
ALT Grade Increase |
Increase of 4 Grades |
4 |
ALT > 8*ULN |
Y |
|
| 8 |
ABC- 0002 |
ALT |
Screening |
SCREENING |
-14 |
30 |
0 |
31 |
Normal |
0 |
32 |
High |
1 |
ALT Grade Increase |
ALT > 8*ULN |
N |
||||||
| 9 |
ABC- 0002 |
ALT |
Baseline |
WEEK 1 |
1 |
32 |
0 |
31 |
High |
1 |
Y |
Y |
32 |
High |
1 |
ALT Grade Increase |
ALT > 8*ULN |
N |
||||
| 10 |
ABC- 0002 |
ALT |
Week 3 |
WEEK 3 |
21 |
23 |
0 |
31 |
Normal |
0 |
Y |
32 |
High |
1 |
High to Normal |
1 to 0 |
ALT Grade Increase |
No Grade Increase |
0 |
ALT > 8*ULN |
N |
|
| 11 |
ABC- 0002 |
ALT |
Week 3 |
UNSCHEDULED |
25 |
25 |
0 |
31 |
Normal |
0 |
32 |
High |
1 |
High to Normal |
1 to 0 |
ALT Grade Increase |
No Grade Increase |
0 |
ALT > 8*ULN |
N |
||
| 12 |
ABC- 0002 |
ALT |
Week 5 |
WEEK 5 |
39 |
33 |
0 |
31 |
High |
1 |
Y |
32 |
High |
1 |
High to High |
1 to 1 |
ALT Grade Increase |
No Grade Increase |
0 |
ALT > 8*ULN |
N |
|
| 13 |
ABC- 0002 |
ALT |
Week 7 |
WEEK 7 |
53 |
100 |
0 |
31 |
High |
2 |
Y |
32 |
High |
1 |
High to High |
1 to 2 |
ALT Grade Increase |
Increase of 1 Grade |
1 |
ALT > 8*ULN |
N |
|
| 14 |
ABC- 0002 |
ALT |
Week 9 |
WEEK 9 |
64 |
27 |
0 |
31 |
Normal |
0 |
Y |
32 |
High |
1 |
High to Normal |
1 to 0 |
ALT Grade Increase |
No Grade Increase |
0 |
ALT > 8*ULN |
N |
|
| 15 |
ABC- 0002 |
ALT |
Month 3 |
MONTH 3 |
89 |
22 |
0 |
31 |
Normal |
0 |
Y |
32 |
High |
1 |
High to Normal |
1 to 0 |
ALT Grade Increase |
No Grade Increase |
0 |
ALT > 8*ULN |
N |
|
| 16 |
ABC- 0002 |
ALT |
Month 6 |
MONTH 6 |
181 |
20 |
0 |
31 |
Normal |
0 |
Y |
32 |
High |
1 |
High to Normal |
1 to 0 |
ALT Grade Increase |
No Grade Increase |
0 |
ALT > 8*ULN |
N |
-
- Examples of Timing Variables
Table 4.8.1 provides a schematic example of the use of the BDS variables for phase, period, and subperiod. The example is of a study in which there are three analysis phases: Screening, Treatment, and Follow-up. The treatment phase consists of a two-period crossover design. In each period of the treatment phase, there are distinct subperiods in which the dose of the corresponding therapy is escalated, then maintained, and then de-escalated.
表4.8.1提供了BDS变量在相位，周期和子周期中使用的示意性示例。该示例是一项研究，其中包含三个分析阶段：筛选，治疗和随访。处理阶段包括两阶段交叉设计。在治疗阶段的每个阶段，都有不同的子期间，在该期间中，相应的治疗剂量应先递增，然后维持并再递减。
Table 4.8.1 Example of Phase, Period, and Subperiod Variables表4.8.1相位，周期和子周期变量的示例
| Variable |
Variable Values |
|||||||
| APHASE |
Screening |
Treatment |
Follow-up |
|||||
| APHASEN |
1 |
2 |
3 |
|||||
| APERIOD |
1 |
2 |
||||||
| APERIODC |
Crossover Period 1 |
Crossover Period 2 |
||||||
| ASPER |
1 |
2 |
3 |
1 |
2 |
3 |
||
| ASPERC |
Escalation |
Maintenance |
De-escalation |
Escalation |
Maintenance |
De-escalation |
||
Note that, in general, there is no requirement to use all three of APHASE, APERIOD, and ASPER when only one or two suffice. Also note that, in general, there is no requirement that the number and nature of subperiods, if any, be the same in each period. If ASPER is used, APERIOD must also be present.
注意，通常，当仅一个或两个就足够时，不需要使用APHASE，APERIOD和ASPER的全部三个。还应注意，一般而言，没有要求每个时期的子时期的数量和性质（如果有）相同。如果使用ASPER，则还必须存在APERIOD。
-
- Examples of Bi-Directional Lab Toxicity Variables
Table 4.9.1 contains an example of variables that support the analysis of parameters for which toxicity is defined in two directions.
In this example, the sponsor decided to include the word “Grade” in ATOXGRL, BTOXGRL, ATOXGRH, and BTOXGRH; however, the inclusion of the word “Grade” is not part of the ADaM standard and is neither encouraged or discouraged. ATOXDSCL and ATOXDSCH are descriptions of the toxicity being assessed; in this example, they are based on the LBTOX values in SDTM, but in other cases could be sponsor- defined.
表4.9.1包含变量示例，这些变量支持对在两个方向上定义毒性的参数进行分析。
在此示例中，发起人决定在ATOXGRL，BTOXGRL，ATOXGRH和BTOXGRH中加入“等级”一词；但是，“等级”一词不是ADaM标准的一部分，也不鼓励或不鼓励。ATOXDSCL和ATOXDSCH是评估毒性的描述；在此示例中，它们基于SDTM中的LBTOX值，但在其他情况下，可以由发起者定义。
Table 4.9.1 Example of Bi-Directional Lab Toxicity Variables
表4.9.1双向实验室毒性变量示例
| USUBJID |
PARAMCD |
AVISITN |
AVAL |
BASE |
ABLFL |
ANRLO |
ANRHI |
ATOXDSCL |
ATOXGRL |
BTOXGRL |
ATOXDSCH |
ATOXGRH |
BTOXGRH |
|
| 1 |
001-0001 |
HGB |
1 |
7.4 |
7.4 |
Y |
11 |
16.1 |
Anemia |
Grade 3 |
Grade 3 |
Hemoglobin increased |
Grade 0 |
Grade 0 |
| 2 |
001-0001 |
HGB |
2 |
20.5 |
7.4 |
11 |
16.1 |
Anemia |
Grade 0 |
Grade 3 |
Hemoglobin increased |
Grade 3 |
Grade 0 |
|
| 3 |
001-0001 |
AST |
1 |
33 |
33 |
Y |
5 |
25 |
Aspartate aminotransferase increased |
Grade 1 |
Grade 1 |
|||
| 4 |
001-0001 |
AST |
2 |
55 |
33 |
5 |
25 |
Aspartate aminotransferase increased |
Grade 1 |
Grade 1 |
||||
| 5 |
001-0001 |
AST |
3 |
60 |
33 |
5 |
25 |
Aspartate aminotransferase increased |
Grade 1 |
Grade 1 |
© 2019 Clinical Data Interchange Standards Consortium, Inc. All rights reserved Page 100
2019-10-03
CDISC Analysis Data Model Implementation Guide (Version 1.2 Final)
| USUBJID |
PARAMCD |
AVISITN |
AVAL |
BASE |
ABLFL |
ANRLO |
ANRHI |
ATOXDSCL |
ATOXGRL |
BTOXGRL |
ATOXDSCH |
ATOXGRH |
BTOXGRH |
|
| 6 |
001-0001 |
AST |
4 |
77 |
33 |
5 |
25 |
Aspartate aminotransferase increased |
Grade 2 |
Grade 1 |
||||
| 7 |
001-0001 |
PLAT |
1 |
250 |
250 |
Y |
150 |
450 |
Platelet count decreased |
Grade 0 |
Grade 0 |
|||
| 8 |
001-0001 |
PLAT |
2 |
100 |
250 |
150 |
450 |
Platelet count decreased |
Grade 1 |
Grade 0 |
||||
| 9 |
001-0001 |
PLAT |
3 |
99 |
250 |
150 |
450 |
Platelet count decreased |
Grade 1 |
Grade 0 |
||||
| 10 |
001-0001 |
PLAT |
4 |
75 |
250 |
150 |
450 |
Platelet count decreased |
Grade 1 |
Grade 0 |
||||
| 11 |
001-0001 |
PLAT |
5 |
49 |
250 |
150 |
450 |
Platelet count decreased |
Grade 3 |
Grade 0 |
||||
| 12 |
001-0002 |
HGB |
1 |
21.1 |
21.1 |
Y |
11 |
16.1 |
Anemia |
Grade 0 |
Grade 0 |
Hemoglobin increased |
Grade 3 |
Grade 3 |
Because PARAMCD PLAT has toxicity grading only in the low direction, only BTOXGRL, ATOXGRL, and other toxicity variables in the low direction are populated. None of the high-direction toxicity variables for PARAMCD PLAT are ever populated, even if the value is out of range in the high direction (ANRIND=HIGH).
由于PARAMCD PLAT仅在低方向具有毒性等级，因此仅填充BTOXGRL，ATOXGRL和其他在低方向的毒性变量。即使该值在高方向上超出范围（ANRIND = HIGH），也不会填充PARAMCD PLAT的高方向毒性变量。
In this example, ATOXDSCL is populated whenever AVAL is not null and grading is in the LOW direction, even if ATOXGRL is null. Similarly, ATOXDSCH is populated whenever AVAL is not null and grading is in the HIGH direction, even if ATOXGRH is null.
在此示例中，即使ATVALGRL为空，只要AVAL不为空且渐变为LOW，就填充ATOXDSCL。同样，只要AVAL不为空且分级为HIGH方向，即使ATOXGRH为空，也会填充ATOXDSCH。
Note that the analysis toxicity grade variables (i.e., ATOXGRH and ATOXGRL) and the corresponding baseline toxicity grade variables (i.e., BTOXGRH and BTOXGRL) can have numeric counterparts (i.e., ATOXGRLN, ATOXGRHN, BTOXGRLN, and BTOXGRHN), which are not shown in this example. The population of ATOXGRL and ATOXGRH is sponsor-defined. In this example, ATOXGRL and ATOXGRH were determined from SDTM LBTOXGR with the addition of "Grade 0" for assessments that did not meet any of the grading criteria in a given direction. ATOXGRL, ATOXGRH, and the corresponding baseline toxicity variables were used to derive shifts from baseline in toxicity grade, as is illustrated in the example in Table 4.9.2. In this case, SHIFT1 represents the shift from baseline in the low direction and is derived from BTOXGRL and ATOXGRL, whereas SHIFT2 represents the shift from baseline in the high direction and is derived from BTOXGRH and ATOXGRH.
请注意，分析毒性等级变量(即ATOXGRH和ATOXGRL)和相应的基准毒性等级变量(即BTOXGRH和BTOXGRL)可以有对应的数值(即ATOXGRLN、ATOXGRHN、BTOXGRLN和BTOXGRHN)，这些数值在本例中没有显示。
ATOXGRL和ATOXGRH的人口由主办方定义。
在这个例子中，ATOXGRL和ATOXGRH是由SDTM LBTOXGR确定的，并添加了“0级”，用于在给定方向上不符合任何评分标准的评估。
使用ATOXGRL、ATOXGRH和相应的基准毒性变量来推导从基准毒性等级的变化，如表4.9.2中的示例所示。
在本例中，SHIFT1表示从基线往低方向的位移，由BTOXGRL和ATOXGRL衍生而来;SHIFT2表示从基线往高方向的位移，由BTOXGRH和ATOXGRH衍生而来。
Table 4.9.2 Example Use of Shift Variables
表4.9.2使用Shift变量的示例
| USUBJID |
AVAL |
BASE |
ABLFL |
ATOXGRL |
BTOXGRL |
ATOXGRH |
BTOXGRH |
SHIFT1 |
SHIFT2 |
|
| 1 |
001-0001 |
7.4 |
7.4 |
Y |
Grade 3 |
Grade 3 |
Grade 0 |
Grade 0 |
||
| 2 |
001-0001 |
20.5 |
7.4 |
Grade 0 |
Grade 3 |
Grade 3 |
Grade 0 |
Grade 3 to Grade 0 |
Grade 0 to Grade 3 |
|
| 3 |
001-0001 |
33 |
33 |
Y |
Grade 1 |
Grade 1 |
||||
| 4 |
001-0001 |
55 |
33 |
Grade 1 |
Grade 1 |
Grade 1 to Grade 1 |
||||
| 5 |
001-0001 |
60 |
33 |
Grade 1 |
Grade 1 |
Grade 1 to Grade 1 |
||||
| 6 |
001-0001 |
77 |
33 |
Grade 2 |
Grade 1 |
Grade 1 to Grade 2 |
||||
| 7 |
001-0001 |
250 |
250 |
Y |
Grade 0 |
Grade 0 |
||||
| 8 |
001-0001 |
100 |
250 |
Grade 1 |
Grade 0 |
Grade 0 to Grade 1 |
||||
| 9 |
001-0001 |
99 |
250 |
Grade 1 |
Grade 0 |
Grade 0 to Grade 1 |
||||
| 10 |
001-0001 |
75 |
250 |
Grade 1 |
Grade 0 |
Grade 0 to Grade 1 |
||||
| 11 |
001-0001 |
49 |
250 |
Grade 3 |
Grade 0 |
Grade 0 to Grade 3 |
||||
| 12 |
001-0002 |
21.1 |
21.1 |
Y |
Grade 0 |
Grade 0 |
Grade 3 |
Grade 3 |
4.10 Other Issues to Consider 4.10其他要考虑的问题
The issues presented in the previous sections represent analysis decisions that commonly occur when creating ADaM datasets. However, the ADaM Team recognizes that those are not an exhaustive list. This section provides comment on some additional issues that may arise.
上一节中介绍的问题代表创建ADaM数据集时通常发生的分析决策。但是，ADaM团队认识到这些列表并不详尽。本节提供对可能出现的其他一些问题的评论。
4.10.1 Adding Records to Create a Full Complement of Analysis Timepoints for Every Subject
4.10.1添加记录以为每个主题创建完整的分析时间点
It is not unusual for a given subject to have missing data for a specified analysis timepoint. For example, suppose an analysis is to be performed for the data obtained at each of four visits and that no imputation is to be performed. For subjects who did not attend all four visits, it would be possible to create records in the ADaM dataset for these missed assessments, with AVAL and AVALC missing (null) and appropriate variable(s) set to indicate these added records. For example, DTYPE could contain a producer-defined value such as "PHANTOM." There are some advantages of having an ADaM dataset contain the same number of observations for each subject. For example, programming is facilitated by having the same data dimensions for all subjects, and by explicitly representing missing data rather than implicitly representing it by the absence of a record. This also allows ADaM datasets to support listing creation, especially for data that is not present in the source SDTM dataset (e.g., added analysis parameters). For some categorical analyses, the denominators can be obtained directly from the ADaM dataset rather than from another input such as ADSL. The disadvantage of this approach is that it may require additional metadata to explain the use of these derived blank records and would require in some cases that subsetting statements be used to exclude the rows on which AVAL is missing. In general, the ADaM Team neither advocates nor discourages this practice for BDS datasets.
对于给定的主题，在指定的分析时间点缺少数据并不罕见。例如，假设将对在四个访问中的每个访问中获得的数据进行分析，并且不进行插补。对于未参加全部四次访问的受试者，可以在ADaM数据集中为这些错过的评估创建记录，其中AVAL和AVALC缺失（空），并且设置了适当的变量来指示这些添加的记录。例如，DTYPE可以包含生产者定义的值，例如“ PHANTOM”。ADaM数据集包含每个受试者相同数量的观察值有一些优点。例如，通过对所有主题使用相同的数据维度来简化编程，通过显式表示丢失的数据，而不是通过缺少记录来隐式表示。这也允许ADaM数据集支持列表创建，尤其是对于源SDTM数据集中不存在的数据（例如，添加的分析参数）。对于某些分类分析，可以直接从ADaM数据集而不是从其他输入（例如ADSL）获得分母。这种方法的缺点是，可能需要其他元数据来解释这些派生的空白记录的使用，并且在某些情况下，需要使用子集语句来排除缺少AVAL的行。通常，ADaM团队既不主张也不反对BDS数据集的这种做法。特别是对于源SDTM数据集中不存在的数据（例如，添加的分析参数）。对于某些分类分析，可以直接从ADaM数据集而不是从其他输入（例如ADSL）获得分母。这种方法的缺点是，可能需要其他元数据来解释这些派生的空白记录的使用，并且在某些情况下，需要使用子集语句来排除缺少AVAL的行。通常，ADaM团队既不主张也不反对BDS数据集的这种做法。特别是对于源SDTM数据集中不存在的数据（例如，添加的分析参数）。对于某些分类分析，可以直接从ADaM数据集而不是从其他输入（例如ADSL）获得分母。这种方法的缺点是，可能需要其他元数据来解释这些派生的空白记录的使用，并且在某些情况下，需要使用子集语句来排除缺少AVAL的行。通常，ADaM团队既不主张也不反对BDS数据集的这种做法。这种方法的缺点是，可能需要其他元数据来解释这些派生的空白记录的使用，并且在某些情况下，需要使用子集语句来排除缺少AVAL的行。通常，ADaM团队既不主张也不反对BDS数据集的这种做法。这种方法的缺点是，可能需要其他元数据来解释这些派生的空白记录的使用，并且在某些情况下，需要使用子集语句来排除缺少AVAL的行。通常，ADaM团队既不主张也不反对BDS数据集的这种做法。
4.10.2 Creating Multiple Datasets to Support Analysis of the Same Type of Data
4.10.2 创建多个数据集以支持对相同类型数据的分析
The SAP often specifies that an analysis will be performed using slightly different methodologies. For example, the primary efficacy analysis may be performed using two different imputation algorithms for missing values. The producer must decide whether to include both sets of the imputed observations in one ADaM dataset or create two ADaM datasets, each representing just one of the imputation algorithms. The ADaM provides variables that can be used to identify records that are used for different purposes. However, this does not imply that the producer should not or cannot submit multiple ADaM datasets of similar content, each designed for a specific analysis.
SAP通常会指定将使用略有不同的方法进行分析。例如，可以使用两种不同的插补算法对缺失值进行主要功效分析。生产者必须决定是将两组估算观测值都包括在一个ADaM数据集中还是创建两个ADaM数据集，每个数据集仅代表一种估算算法。ADaM提供了可用于标识用于不同目的的记录的变量。但是，这并不意味着生产者不应该或不能提交相似内容的多个ADaM数据集，每个数据集都是为特定分析而设计的。
4.10.3 Size of ADaM Datasets
4.10.3 ADaM数据集的大小
It is important to consider the size of ADaM datasets, because large datasets can pose problems for transferring between parties, loading into data warehouses, or software processing. The maximum size of a dataset and how to handle large datasets should be discussed with the recipient of the data and clearly documented. Refer to the FDA website (see Section 1.2, Background) for recommendations concerning sizes of submitted datasets.
重要的是要考虑ADaM数据集的大小，因为大型数据集可能会在各方之间转移，加载到数据仓库或软件处理中造成问题。数据集的最大大小以及如何处理大型数据集应与数据的接收者讨论并明确记录在案。有关提交的数据集大小的建议，请参阅FDA网站（请参阅第1.2节，背景）。
4.10.4 Traceability when the Multiple Imputation Method Is Used
4.10.4 使用多重插补方法时的可追溯性
There has been increased attention in the analysis of clinical trial data to address problems associated with missing data, and with this increased attention has come new ways to deal with this problem. In the past, simple methods such as "last observation carried forward" or "baseline observation carried forward" were routinely used to replace missing values. Such "single-point imputation" methods have been shown to underestimate the standard error of the estimates of various statistics computed from the data. A more sophisticated method, termed "multiple imputation," was introduced in 1987 and is now fully supported by frequently used software packages. In brief, this methodology deals with the uncertainty of the missing data by employing a three-step process. The first step is the creation of multiple datasets in which plausible values for each missing data value are imputed. The second step is to analyze each of these datasets with the desired statistical procedure and capture the resulting statistical estimates. The third and final step is to use these estimates to generate a combined (pooled) estimate. It is these estimates which are based on the pooling of the estimates from the multiple imputation datasets that are used to evaluate statistical significance.
在临床试验数据分析中，人们越来越关注解决与缺失数据相关的问题，随着关注的增加，已经出现了解决这一问题的新方法。过去，通常使用简单的方法（例如“结转最近的观察”或“结转基线观察”）来替换缺失值。这种“单点插补”方法已显示出低估了根据数据计算出的各种统计数据的估计的标准误差。1987年引入了一种更复杂的方法，称为“多重插补”，现在已得到常用软件包的完全支持。简而言之，这种方法通过采用三步过程来处理丢失数据的不确定性。第一步是创建多个数据集，其中为每个缺失的数据值推定合理的值。第二步是使用所需的统计程序分析这些数据集中的每个数据集，并捕获生成的统计估计值。第三步（也是最后一步）是使用这些估算值来生成组合（合并）估算值。这些估计是基于来自多个插补数据集的估计的汇总来评估统计显着性的。
Using the SAS software as an example, the above three-step process is achieved using PROC MI to create the multiple datasets (Step 1). Step 2 would utilize the procedure associated with the desired statistical model, such as PROC LOGISTIC. Step 3 would use PROC MIANALYZE to create the combined pooled estimates.
以SAS软件为例，使用PROC MI创建多个数据集即可完成上述三步过程（步骤1）。步骤2将利用与所需统计模型（例如PROC LOGISTIC）相关的过程。第3步将使用PROC MIANALYZE创建合并的合并估计。
In the ADaM, the documentation of derived variables via variable level metadata, and statistical results via analysis results metadata, is paramount to achieving the concept of traceability. However, documenting the traceability of estimates created via multiple imputation cannot be achieved with these current metadata methods. In addition, it would not be practical to include all datasets that are created from the PROC MI process as part of a submission. To address traceability, the ADaM recommendation is to provide the program statements from the three procedures mentioned above as a part of the analysis results metadata. This would allow the reviewer to re-create the analysis as desired. Of primary importance is to ensure that the options used in PROC MI, specifically the value of the seed, the number of iterations, and the method used for imputation, are clearly denoted.
在ADaM中，通过变量级元数据对派生变量进行文档化，通过分析结果元数据对统计结果进行文档化，这对于实现可追溯性的概念至关重要。然而，用这些当前的元数据方法无法记录通过多重估算产生的估算的可追溯性。此外，将所有从PROC MI进程创建的数据集作为提交的一部分是不实际的。为了解决可追溯性问题，ADaM建议提供来自上述三个过程的程序语句，作为分析结果元数据的一部分。这将允许审阅者按照需要重新创建分析。最重要的是要确保在PROC MI中使用的选项，特别是种子的值、迭代次数和用于imputation的方法被清楚地标注出来。
Here is an example taken from a study where multiple imputations need to be performed in order to impute missing covariate values. The output from PROC MI is fed into PROC PHREG, and the hazard ratios are obtained. The output is then routed back through PROC MIANALYZE to get the point estimates and confidence intervals. This example shows six covariates in the model (represented by generic variables COVAR1-COVAR6). For PROC MIANALYZE, the variables should be listed in the order in which they were identified by the forward selection model. This information would all be included directly in the analysis results metadata, with the generic variables COVAR1-COVAR6 replaced by the names of the actual variables selected by the model.
这是一个来自研究的示例，在该研究中，需要执行多个插补才能插补缺失的协变量值。来自PROC MI的输出被馈入PROC PHREG，并获得危险比。然后将输出路由回PROC MIANALYZE，以获得点估计值和置信区间。此示例显示了模型中的六个协变量（由通用变量COVAR1-COVAR6表示）。对于PROC MIANALYZE，变量应按前向选择模型识别变量的顺序列出。此信息将全部直接包含在分析结果元数据中，而通用变量COVAR1-COVAR6将替换为模型选择的实际变量的名称。
4.10.5 Copying Values onto a New Record
4.10.5 将值复制到新记录
As a general rule, when a new record is derived from a single record in the dataset, retain on the derived record any variable values from the original record that do not change and which make sense in the context of the new record (e.g., --SEQ, VISIT, VISITNUM, --TPT, covariates). When a record is derived from multiple records, retain on the derived record all variable values that are constant across the original records, do not change, and which make sense in the context of the new record. Note that there are situations when retention of values from an original record or records would make no sense on the derived record; in such cases, do not retain those values. Refer to Table 4.2.1.3 and Table 4.2.1.4 for examples.
通常，从数据集中的单个记录派生新记录时，请在派生记录上保留原始记录中任何不变且在新记录的上下文中有意义的变量值（例如- -SEQ，VISIT，VISITNUM，-TPT，协变量）。当从多个记录派生一条记录时，请在派生记录上保留所有在原始记录中不变，不变的变量值，这些变量值在新记录的上下文中是有意义的。请注意，在某些情况下，保留原始记录或多个记录中的值在派生记录上毫无意义；在这种情况下，请勿保留这些值。有关示例，请参见表4.2.1.3 和表4.2.1.4 。
第四章（下） 完