Assessing Needs, Avoiding Pitfalls
She cautions to avoid, for example, LIMS that must be recoded for every new feature. Labs must absorb programming themselves, call the vendor back in, or outsource coding to third-party software firms that lack sufficient familiarity with the package.
“People treat a LIMS deployment like the end of a race,” Meek says, “but the process is more like buying a house, where you redo the kitchen one year, replace the roof the next year. You’re constantly extending your LIMS implementation because your business keeps changing.” LIMS should be flexible and support those changes.
Since one goal of lab informatics systems is to reduce human error, potential purchasers should examine the flow points or rate-limiting steps within their workflows and see to it that their LIMS can execute those tasks. Tasks may be as simple as documenting which materials are being weighed and the numerical value, or as complex as routing an out-ofspecification value to the correct manager for action.
Just a few years ago, lab information products were sold and operated in stand-alone mode. Next came interoperability and data exchange. The convergence and overlap of major laboratory IT products have been followed by a blurring of terms. Thanks to a consensus decision by ASTM, an ELN for QA/QC is now referred to as an LES (laboratory execution system). By no means is that the limit of integration. Thermo Fisher includes an SDMS (scientific data management system) within the same platform as its LIMS/LES. “LES is the correct way to think of and refer to an ELN in a QA/QC environment because you’re executing predefined processes versus designing and executing experiments as you’d do in an R&D organization,” Meek tells Lab Manager.
Variable but remarkably similar needs
For Garrett Mullen, manager for NuGenesis Lab Management Systems at Waters (Milford, MA), companies should base LIMS purchase decisions on key business drivers—the need to create and manage inputs and make sound decisions based on them.
“Those high-level requirements then push down into laboratories and the information they generate and use,” he says.
Needs vary for different industries and laboratory sizes, for regulated and unregulated businesses. Many LIMS are designed for food or environmental testing. Some are traditional sample management tools, and others integrate with electronic laboratory notebooks or supervisory software, while others take on the functions of a laboratory execution system.
So what would trigger undertaking a LIMS deployment? Mullen rejects the notion that workflows, sample volumes, or company size should dictate.
LIMS options and opportunities for small to mediumsized labs have never been better. Quality and traceability options are in no way less critical for these customers than for multinational corporations. “They also need to manage sampling, register samples, allocate work, manage results capture, measure standards, and create reports,” Mullen says. Today’s LIMS can be designed specifically for tasks that small labs need to manage, and at a level that suits them.
This leads to the most common mistake lab managers make when embarking on a LIMS project, what Mullen refers to as “boiling the ocean.”
Personnel managing a LIMS installation usually create a wish list of capabilities they’re seeking. That’s natural. But insisting that the LIMS plug all sample management and data holes, and do so from day one, is unreasonable. Projects that begin in that manner too often have unpleasant consequences, sometimes extending beyond the tenures of key personnel, through mergers, and after the original mission no longer applies.
“Consider a phased approach,” Mullen advises. “Don’t try to solve every single problem. Your lab may not need a full-blown LIMS.”
With luck you may eventually “boil the ocean,” but the odds of success are greater if you approach the problem one bucket at a time.
For additional resources on LIMS, including useful articles and a list of manufacturers, visit www.labmanager.com/LIMS