ADaM Datasets Example: What You Need To Know About ADaM Standards

Health | November 14, 2022, Monday // 07:22|  views

Adobe stock

The Clinical Data Interchange Standards Consortium, or CDISC, creates medical research standards for the healthcare and pharmaceutical industry, especially for medical research such as clinical trials. And one of the most important CDISC standards is the Analysis Data Model (ADaM) for clinical trial submissions. 

What Is ADaM?  

The CDISC has different standards for clinical submissions, including content and data exchange standards. The content standards include the Clinical Data Acquisition Standards Harmonization (CDASH), Protocol Representation Model (PRM), Study Data Tabulation Model (SDTM), and ADaM, defining what datasets and variables are allowed in clinical trial submissions. 

SDTM is the original source for ADaM data. From there, analysis datasets enable the scientific and statistical analysis of the clinical study results. Moreover, ADaM specifies the standards and principles to ensure a clear sequence from clinical data collection to analysis.  

While SDTM is the standard for organizing data collected in animal and human clinical trials, ADaM involves creating data ready for analysis. You can discover here other essential things you should know about ADaM standards.  

ADaM Dataset Examples 

SDTM datasets are categorized to ease visual analysis. This way, a reviewer can filter the datasets across the entire clinical trial and detect patterns and anomalies. On the other hand, ADaM data sets are grouped for an easier calculation to derive specific results in the Table, Listings, and Figure (TLF). This process requires variables from SDTM sources and then rearranging them into a different grouping to obtain the result. 

Below are some ADaM datasets examples from the original Pilot 01 study of the CDISC: 

  • vital signs (‘advs’)  

  • subject-level (‘adsl’) 

  • laboratory chemistry data('adlbc') 

  • concomitant medications ('adcm')  

  • adverse event (‘adae’)  

  • pharmacokinetic parameters (‘adpp’) 

This ADaM dataset follows the Version 2.0 standard, which contains the augmented and modified dataset of the CDISC’s original Pilot 01 study. 

 

ADaM Implementation Guide 

Because the Food and Drug Administration (FDA) now requires clinical trial data submissions to use CDISC standards, many companies make their analysis datasets ADaM-compliant. To do this, clinical trial submissions must follow the ADaM implementation guide, which specifies the dataset structures, variables, and standard solutions to ADaM implementation issues. 

The ADaM standard data structures in the implementation guide include the following datasets: 

  • Subject-Level Analysis Dataset (ADSL)  

The ADSL dataset has one record per subject and variables like subject-level population flags, demographic information, and important dates. ADSL and related metadata are a CDISC requirement for clinical trial submission. 

  • Basic Data Structure (BDS) 

The BDS dataset has one or more records per analysis timepoint, per analysis parameter, and per subject. It bears a central set of variables representing the actual data under analysis. Furthermore, this dataset should include variables PARAM (Parameter) and AVAL (Analysis Value) and/or AVALC (Analysis Value (C)).

Common Mistakes In ADaM Implementation 

The ADaMIG can lead to confusion and misinterpretation, resulting in non-compliant datasets. That’s why it’s important to fully understand the implementation guide to avoid causing dataset problems and ensure CDISC compliance. 

Moreover, below are the common ADaM implementation mistakes and recommended solutions: 

  • Creating ADaM Dataset For Listing Generation  

ADaM isn’t suitable for generating a listing. Creating listings involves using the corresponding SDTM domain. You only have to create an ADaM data set if the listing comes from a single data source but needs another calculation based on analysis or treatment period date. This is so because FDA reviewers aren’t programmers. Hence, they may find it uncomfortable to combine datasets and derive variables. 

  • Incomplete Analysis 

It could lead to long hours of re-work if you don’t know what you’re analyzing when developing ADaM data set specifications. It could also cause analysis violations because programmers find it easier to update the table/figure (TF) programs instead of going back and updating the ADaM dataset specifications and programs. You must go through each dataset to see the required data and analyze and create annotations to define the necessary ADaM datasets. 

  • Traceability Issues 

Traceability instills confidence in the clinical trial results, linking back to the source data for transparency between the analysis results of the clinical study and the ADaM datasets and SDTM domains. With this, use data point traceability to determine the predecessor record quickly. In addition, include metadata traceability so the reviewer can easily understand the relationship between the clinical study analysis data and the source data. 

  • Incorporating Baseline Values 

Incorporating baseline values in ADaM Subject Level Analysis Datasets (ADSL) isn’t always a good idea. Reviewers may think all the analysis results on a figure or table must come from one dataset. If including baseline values isn’t a requirement other than summarizing baseline characteristics, you can save time and effort by not doing it.  

Conclusion 

ADaM standards are valuable in clinical research. For one, being compliant with ADaM standards can hasten a clinical trial’s review and approval process. That said, companies carrying out clinical studies must be equipped with the right information about the set standards in the ADaM Implementation Guide for easier compliance.

 

We need your support so Novinite.com can keep delivering news and information about Bulgaria! Thank you!


Back  

Search

Search