Affordable Access

Systems validation: application to statistical programs

BioMed Central
Publication Date
  • Debate
  • Computer Science

Abstract ral BMC Medical Research ss BioMed CentMethodology Open AcceDebate Systems validation: application to statistical programs Rickey E Carter* Address: Department of Biostatistics, Bioinformatics and Epidemiology, Medical University of South Carolina, Charleston, South Carolina, USA Email: Rickey E Carter* - [email protected] * Corresponding author Abstract Background: In 2003, the United States Food and Drug Administration (FDA) released a guidance document on the scope of "Part 11" enforcement. In this guidance document, the FDA indicates an expectation of a risk-based approach to determining which systems should undergo validation. Since statistical programs manage and manipulate raw data, their implementation should be critically reviewed to determine whether or not they should undergo validation. However, the concepts of validation are not often discussed in biostatistics curriculum. Discussion: This paper summarizes a "Plan, Do, Say" approach to validation that can be incorporated into statistical training so that biostatisticians can understand and implement validation principles in their research. Summary: Validation is a process that requires dedicated attention. The process of validation can be easily understood in the context of the scientific method. Background The United States Food and Drug Administration (FDA) has defined validation as "confirmation by examination and provision of objective evidence that computer system specifications conform to user needs and intended uses, and that all requirements can be consistently fulfilled" [1]. Validation is a process that begins with defining the requirements and ends with ensuring that the needs are being fulfilled consistently. Structured testing is only a part of the validation process. Interaction with researchers emphasized the notion that many people use validation and testing synonymously. To better illustrate the con- cepts of validation to these same researchers, the follow- ing mantra

There are no comments yet on this publication. Be the first to share your thoughts.


Seen <100 times