Life Sciences

How Many Environments Do You REALLY Need?

A couple of weeks ago, a client of ours discovered a problem in the production (PROD) environment of one of their clinical systems. In the initial system implementation, they had opted for a single environment (without a sandbox, development, training, test, or validation environment), so they found themselves with no separate environment in which to safely investigate the problem and test potential solutions.

When the problem arose, they still had the option to set up and qualify a validation (VAL) environment for testing, but, instead, chose to just makeclinical-systems-environments a temporary copy of PROD. We understood their reasons: it was faster and cheaper. However, the risk was certainly higher.

Sure enough, as soon as the temporary copy was made, PROD itself crashed and users were suddenly unable to do their jobs. The temporary copy was dismantled and, thankfully, PROD came back online.

Life Sciences - How Artificial Intelligence Can Enhance the Clinical Data Review and Cleaning Process
How Artificial Intelligence Can Enhance the Clinical Data Review and Cleaning Process

This guide analyzes how artificial intelligence – including machine learning – can be used by pharmaceutical and medical device companies to improve the clinical data review and cleansing process.

Get the Guide

But, now, our client is in a worse position than they were before: they are spending resources investigating why the temporary environment caused PROD to crash, as well as trying to determine whether PROD survived the crash 100% intact, and they still have the original problem to investigate with no safe place to do so. They are now beginning to understand the value in allotting the time and money to set up and qualify a permanent VAL environment…

This experience reminded me just how important it is to make the investment in maintaining a separate, qualified, and controlled environment that mirrors PROD. While it can be tempting to cut corners and opt for just one environment, when you consider the importance of the system and the data it contains, it really is worth the investment to be prepared to handle these kind of “what if” situations.

Here are the minimum environment recommendations that we make to all of our clients:

  • For systems with no customizations (used out-of-the-box), install at least two environments: validation (VAL) and production (PROD). VAL should mirror PROD.
  • For systems with customizations, install at least three environments: development (DEV), validation (VAL), and production (PROD). VAL should mirror PROD, but DEV does not have to.
  • In all cases, we recommend a disaster recovery (DR) environment, for the utmost insurance.

If you have questions about the number and types of environments you should have for any of your clinical or safety systems, drop us a line. We would be happy to give you our best advice.

About the Author

Marin joined the life sciences industry in 2001. Over the course of her tenure, she has held roles in clinical finance, IT, quality assurance, and validation. The diversity of her experience provides her with a unique perspective on the interconnectedness of this complex, multi-faceted industry. Marin Richeson is a lead business consultant in Perficient's life sciences practice.

More from this Author

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Subscribe to the Weekly Blog Digest:

Sign Up