Clinical labs have specific needs that set them apart from research and development (R&D) labs, quality control (QC) labs, or labs in the non-life science industries, which have traditionally been sample focused. The vast amounts of patient-specific, sensitive data generated in the process of administering a clinical trial or performing DNA screenings or blood typing require high levels of computational power and even more stringent security than that required to safeguard intellectual property.
The informatics solution for a clinical lab has traditionally been a laboratory information management system (LIMS) or laboratory information system (LIS), but the lines between a patient-focused LIS and a sample-focused LIMS are increasingly blurred. We recognize that clinical organizations have questions when making the cost–benefit analysis around informatics systems for their labs, so we will answer some of them in this blog, in the form of a Q&A session with our consultants.
here are five main areas where an informatics system outshines a spreadsheet:
To be honest, the first two are the key ones; Excel can handle a LOT—I’ve seen people do many integrations with Excel, it just takes a superuser to use the code within.
Data quality and data integrity are far lower in a spreadsheet than an informatics system because that’s not what a spreadsheet was designed for. Most informatics systems are 21 CFR Part 11 compliant right out of the box (OOTB). It is very difficult to get a spreadsheet to that point, although I have done it. The expanded memory and storage capabilities of an informatics system allows users to safeguard all data and to manage big data. Informatics systems are designed to hook up to storage databases, whereas spreadsheets aren’t intended for this—this distinction is especially useful and important as the number of records increase.
When you start to look at your choices of lab data management systems you have to ask, what are the potential upstream, downstream, and parallel uses and benefits? Are there questions you are asked that you have to repeatedly answer? If so, an informatics system allows requesters to get the answers themselves when they want them and with significant trust in the data, thereby reducing the need to respond to routine questions. An informatics system can send data automatically to users outside the lab or allow those users to access the system directly.
If your intellectual capital is specialized and valuable to your clinical organization, an informatics system reduces the risk that it will get lost or be hard to find. An informatics system isn’t always better, however. The choice (system or spreadsheet) still has to be about the fit for purpose. That purpose could be a budget, people, tech, or scale need. There are some instances where researchers labor in isolation; no one ever sees the work, uses it, or follows up on the topics they are tracking. In those cases, the testing that the researcher is doing can be well managed by a tool they’ve used all their life.
In every laboratory informatics system, the stored sensitive data is encrypted so external viewers (or unauthorized viewers at the server level) can’t see the data. In clinical laboratories, sensitive data refers to all the confidential information relating to a patient or client that should not be disclosed to a third party. The release of such information, knowingly or not, can result in penalties. Hence, there’s a need to safeguard the privacy and integrity of all four categories of sensitive patient information.
A clinical LIMS or LIS safeguards patient data in a number of specific ways. Clinical data is stored in a database, often on a server; access to the database is safe and restricted. The data is accessed and manipulated based on role access, with two-factor authentication. User permissions are set up so that not everyone can manipulate the data. It’s possible to encrypt data when it is sent from one system to another. It’s also possible to blind data when required, for example, when conducting a clinical trial.
Interfacing systems and instruments increases the productivity of your clinical lab and reduces the risk of human error. Examples of systems and instruments that can be interfaced with a clinical LIMS or LIS include the following:
A clinical LIMS or LIS can interface with many other systems like an enterprise resource planning system (ERP), a quality management system (QMS), or a manufacturing execution system (MES). The interface can be accomplished in a number of ways such as with Mirth Connect or via web services.
There are molecular diagnostics systems that can interface with your LIMS or LIS and then with the various assay instruments associated with it. There’s no reason why a clinical lab couldn’t have access to all of its data in once central location.
As far as what systems can be interfaced with a clinical LIS: if an interface is possible for a LIMS it can probably be done to some degree, particularly if it’s not a super old system (most newer systems are going to come with some kind of basic tools for interfacing). There are of course instrument integrations, where you hook it up so that you can pull results from the instrument into the LIMS. I’ve seen LIMS integrations with electronic medical record (EMR) systems, quality documentation systems, Oracle databases, custom phone applications, etc. The question is less “What can be interfaced” and more “What do you have that you want to interface, and what do you see that interaction doing for you?”
Choosing the right reports to maximize your lab’s efficiency is an art. When you know the specific information your stakeholders need to make wise business decisions, you can present that information in an easy-to-understand report, which can then be distributed automatically by email. Some common examples include the following:
Many reports are available in a clinical LIS or LIMS system; the most important in my opinion are the Patient Demographic report, Results reports, and Kit requisition forms.
In addition to the content of the reports, you can also choose the format. Examples I’ve seen include pdf, JSON, and csv.
Any data you put into your database can be pulled out into a report and arranged as you see fit. I’ve worked on reports that were emailed every Monday at 1 a.m., listing all samples processed in the previous week to give the lab’s lead an overview; reports for billing or staffing; certificates of analysis (CoAs); shipping inventory reports; storage inventory reports; all sorts of data analysis reports; etc. I’ve made reports for clients that can be run at will, reports that trigger when an action is completed, and reports that run on a schedule. I’ve made reports that download to your computer, reports that place themselves in a file in a file system, and reports that email themselves to the relevant contacts. LabVantage LIMS even has a feature called Dashboards that allows you to pull up data and display it on the screen rather than run a report (you can set it to refresh itself periodically to keep it relatively live, though you might want to keep an eye on system resources if you’re doing this with a huge amount of data and a high refresh rate—I’ve had a dashboard slow down system processes before).
Questions about features or setup often come from the client’s LIMS Admin. They are working with the setup of the LIMS or LIS and may be curious about specific features, or they may have encountered a problem. If I am not familiar with the topic, I will reach out to other consultants on our CSols team for help. Our consultants have a vast diversity of knowledge and although I may be the one who’s billable for that particular project, we all collaborate so that everyone is successful.
Knowing how a feature works requires training on the system and on that particular feature. Any clinical LIMS or LIS has a user guide that explains how to use the system, configure data, and enable process workflow. The level of scaffolding provided by the vendors does vary, so supplemental training is available from third-party clinical LIMS or LIS experts, like CSols.
The answer for this obviously depends on the actual work being done in the clinical lab, but the point I want to make is that when a team of informatics experts is involved the client needs to have dedicated resources ready to prevent delays. It’s important for the work to be done swiftly to minimize work process disruptions. So many of our projects are slow to start because we are waiting on training from the client, access to the network or system, documentation, or whatever else is needed to start the project. Also, not all on-boarding goes smoothly. Sometimes there are issues with connections to the client’s systems, so any initial delays in on-boarding could be further exacerbated by follow-on effects. The quicker the on-boarding, the quicker any kinks can be worked out.
A clinical LIMS or LIS is sold as a base-model Out of the Box (OOTB) version to every organization. The OOTB version offers multiple functionalities for processing a given Testing workflow. But sometimes there are specific requirements needed by different clients; in this case, customization can be applied to the OOTB system (with appropriate validation, of course!). Each requirement is different and can be done in a period of time estimated based on the complexity of the task. Some tasks can be completed in one hour and others can require a few days.
Perhaps the takeaway from all this is that the most appropriate system for your clinical lab will be the one that best fits your data needs and those of the end users, who may be patients, other researchers, or organizational stakeholders. A trusted consultant with decades of lab data management experience can help you tell the most appropriate data story for your clinical organization.
Do you have questions that weren’t answered here? Ask them below.