Outline

简要介绍
阶段划分&方法介绍
软件介绍
合规要求

Definition

A critical phase in clinical research, which leads to generation of high-quality, reliable, and statistically sound data from clinical trails.

Phases & Methodology

CDM activities are normally categorized into three different stages:

• Study start-up
  CRF and protocol alignment
  Data management plan, CRF annotation
  Database design/Edit check programming, data views
• Study conduct
  Double data entry
  Data validation & cleaning, query resolution
  MedDRA coding, SAE(Serious Adverse Event) reconciliation
• Study closeout
  Database lock

QC & documentation (study master data management file)

Key consideration

Case Report Form(CRF) Finaliztion

• Design based on study-specific requirements
  Study protocol and applicable regulatory requirements
• Facilitate the end users in data transcription
• Trivial inputs are not ignored
  units, time clock(24hrs or 12hrs. suffix with a.m./p.m.), data format etc.
• Incorporation of relevant inputs from all the team members
• CRF completion guidelines(CCGs); instructions and prompts
• Use of standard pages

Data Management Plan

Captures all the methods and measures carried out to handle the data management project to achieve consistent and credible data amenable to statistical analysis

Annotation of Case Report Form(CRF)

• Depicts the mapping between the database and CRF fields
• Assists in programming
• Provides metadata for the study
• Reflects applicable Standards for Data Acquisition & Submission
  For example:
  CDASH(Clinical Data Acquisition Standards Harmonization)
  CDISC(Clinical Data Interchange Standards Consort)
  SDTM(Study Data Tabulation Model)

Database

• Database, a clinical software application
  built to facilitate the CDM tasks to carry out multiple studies
• Database, serves as the warehouse for the data
  a structure set of rows and columns
• Database screens should be designed so that they are an exact replica of paper CRFs, to facilitate fast data entry with minimum errors

Edit Checks

Programming and Validation of Data

• Edit checks: programmed in the database for data cleaning
• Faciliate detection of possible errors in the data
• In accordance with the requirements mentioned in the protocol
• Programs are written according to the logic condition mentioned in the Data Validation Plan(DVP)
  For example:
  Range checks
  Missing data
  Data inconsistency checks
  Out of logical checks

Data Entry

• Done acording to the 'Data Entry Guidlines' prepared in consultant with study medical monitor
• Double Data Entry (DDE) for paper CRF
  the data transcribed on a form is captured twice by different entry operators and programmatically compared
• For e-CRFs, data directly entered in the database at the study site
  Reduce the chances of errors and discrepancies resolution is faster

Discrepancy Management

• Includes reviewing of discrepancies and resolving them by seeking the resolution with documentary proof
• Updation of resolutions in the database by designated CDM personnel
• Closure/resolution of all queries before database lock
• Main indicators for good CDM practices; discrepancy management
  Maintaining the TAT(Turn Around Time)
  Avoiding Query Aging

Database Lock

• Database lock is:
  prevent unauthorized data manipulation
  comply with the regulatory requirement for data security
• Database lock must not be done unless all the CDM processes have been completed
  Data entry - done
  Queries - resolved
  Data coding - done
  Quality check of all the activities - done
  SAE Reconciliation - done
  All relevent documentation complete

Data Extraction

Provides the data in the desired format for the purpose of QC, analysis, data transfer or to meet any other requirement for the study

Software Solution

Some widely used solutions

• Commercially available are:
  Medidata
  Phase Forward
  Oracle Clinical
• Open Source solutions:
  PhOSCo
  GCP BASE
  EpiData

Regulatory Perspective

Major regulations/guidelines/best practices apart from Good Clinical Practices (GCP)

• Code of Federal Regulations (CFR), 21CRF Part 11: demands use of validated systems
• CDISC(Clinical Data Interchange Standards Consort) provides standards to support acquisition, exchange, submission, and archival of trial data and metadata
• Following are the most relevent:
  Study Data Tabulation Model Implementation Guide for Human Clinical Trails (SDTMIG)
  Clinical Data Acquisition Standards Harmonization (CDASH) standards
• Good Clinical Data Management Practice (GCDMP) by Society for Clinical Data Management (SCDM)

Terms

ICH(The International Conference For Harmonization Of Technical Requirements For Pharmaceuticals For Human Use) 人用药物注册技术要求国际协调理事会
MedDRA 医学标准术语集

Reference:
知乎live - 临床试验第一讲：如何做好数据管理？- 王Nathan