Addressing Data Integrity in the Age of Sticky Notes

By: Steve Ferrell, Chief Compliance Officer
data integrity

In 1992, I entered my first lab: a land of Bunsen burners, pipettes, test tubes and a kaleidoscopic selection of oddly colored sample jars. Within 20 minutes, my lab partner and I had spilled 4 molar hydrofluoric acid into an electrical outlet and my scientific career went up in literal flames. For reasons you might guess, it was not until almost 10 years later that I found myself in the lab again. This time I was validating 21 CFR Part 11 compliance solutions that had been installed across all the PC’s in the lab – less fun and dangerous.

In a lab replete with Mass Spectrometers and HPLC systems, there was one tool that stood above the rest; a tool so powerful that its absence would have disabled the lab indefinitely; the only tool that every scientist used consistently.

Yes, it was the yellow, green, orange and pink sticky notes adorning almost every instrument and PC. It was on these stickies that chicken-scratched username and passwords were proudly on display. Some of them held such importance that they were scotch taped to defy the ravages of time, while some were the final legacy of a long forgotten lab tech.

I was told it was convenient and necessary; that there was no way anyone could remember all those passwords, and if they didn’t have the admin credentials they wouldn’t have the resources to do their jobs properly. Today, I have no doubt that quality science was being conducted in that lab but the data they produced, regardless of its qualitative aspects, was fully lacking in Data Integrity (DI).

To that end, one of the most interesting outcomes of the FDA and MHRA’s release of new DI guidance is that they uniformly expect facilities to achieve DI by applying existing rules and standards. In other words, there has been no legislative change driving the growing DI fever, but rather a realization of what the convergence of people, policy, and technology means to the principles of integrity.

 

People

IT can no longer afford to consider the lab an island of unwashed PCs. The lab can no longer afford to play IT or download who-knows-what to their lab PC’s.

To achieve real DI, lab and IT resources must come together to develop a common framework for data governance. An example of a first step could be banning of the sticky note from the lab; just as White Out was long ago. DI and data governance take work and they require a cross-departmental team approach and diligence. Pressure has to be applied to instrument manufacturers and lab software developers to offer instrument connective Cloud based solutions (like we do) or at the least offer PC-agnostic deployments.

 

Policy

Lab SOP’s have to be written not just from a work or process flow perspective. Instead, they must consider the quality and integrity aspects of the data they are seeking to produce and analyze. For example, a lab will buy an instrument that comes with its own PC, in turn IT regards this PC as unworthy or unclean and will refuse to install it, or if they do install the PC, they do so adding a virus scanner and some other security management tools. The instrument manufacturer is then reticent to guarantee that their software will work with all of the additional software processes running.

In our example scenario and dozens of others, after the lab manager navigates all of those constraints, they find themselves with a validated instrument and PC and a set of operational SOP’s, but still no considerations for the administrative aspects of the system and the governance of its data. Which brings us to:

 

Technology

If you are a lab technician, manager, supervisor, director, there are a couple of points I can make about your LIMS platform:

  • If you are millennial, your LIMS system is probably older than you are
  • Your LIMS only works when supported by a whole mess of spreadsheet-based reports and calculations
  • Your end-to-end workflows require manual intervention, paper, and lifting and shifting of data
  • The Infrastructure on which it resides is a little old, and a little dusty

Many of these apply to not just LIMS but to many other types of lab software you depend on. The good news is that all of these implied problems can be mitigated with a system that is designed to manage lab data, as it exists today, correctly. A system that reduces human error, increases compliance, and offers a cloud based deployment is the update that most labs need on their journey to DI and data governance.

I am proud to say that RURO’s platforms are built for the lab of today and are infinitely configurable. As such, they vanquish the sticky notes, spreadsheets, and manual workarounds. Can your grandma’s LIMS do that?

Achieving progress in these areas of Data Integrity will ensure your lab data is inflammable.

In my next blog post, we’ll break down the first piece of the ALCOA puzzle: Making Lab Data Attributable.