Statistical Computing Environments and the Practice of Statistics in the Biopharmaceutical Industry

A structured statistical computing environment (SCE) enhances rigor in operational implementation of statistical analyses of clinical studies through process transparency, allowing reproducibility of results by independent reviewers. Desirable features and associated benefits of an SCE system are de...

Full description

Saved in:
Bibliographic Details
Published in:Drug information journal Vol. 44; no. 1; pp. 29 - 42
Main Authors: Hopkins, Alan, Duke, Susan, Dubman, Sue
Format: Journal Article
Language:English
Published: Los Angeles, CA SAGE Publications 01.01.2010
Springer International Publishing
Springer Nature B.V
Subjects:
Biostatistics
Drug Safety and Pharmacovigilance
Metadata
Pharmacotherapy
Pharmacy
Science
Software
Studies
Technological change
Statistical programming
Metadata
Data standards
IT architecture
Good statistical practice
Statistical operations
ISSN:2168-4790, 0092-8615, 2164-9200, 2168-4804
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A structured statistical computing environment (SCE) enhances rigor in operational implementation of statistical analyses of clinical studies through process transparency, allowing reproducibility of results by independent reviewers. Desirable features and associated benefits of an SCE system are described. Minimum SCE requirements discussed in detail consist of a structured programming environment, an operational analysis data repository, and a metadata-driven architecture containing information about data and status of various processes. The metadata provide a foundation for connecting multiple processes and systems, thereby allowing the creation of tools that largely automate the analysis process. Standards drive productivity enhancement for creating statistical deliverables based on metadata obtained from the development plan, protocols, and analysis plans. Not all the features discussed are available today in commercial systems. In the future, nearly all information about clinical trial analytics can be driven by a standards-based, metadata-driven architecture. To accomplish this goal, metadata need to be available about all the processes used to collect, transform, and analyze the patient data. Further standards development will be necessary to fully describe the entire statistical analysis process. Upon completion of this article, participants should be able to do the following: Learning Objectives • Describe the elements of good statistical practice that contribute to establishing the credibility of clinical trial results. • Describe the fundamental concept of statistical computing environment (SCE); the SCE as a programming environment; and as a clinical data platform and repository driven by a metadata architecture. Target Audience This article is informative for medical doctors working in the pharmaceutical industry; biostatisticians, statistical programmers, clinical data managers, and IT professionals.
Bibliography:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ISSN:2168-4790
0092-8615
2164-9200
2168-4804
DOI:10.1177/009286151004400104