Getting Started with Analytics and Business Intelligence in the Lab

Posted by Adam Hasler on Feb 19, 2016 3:06:42 PM

In recent years, companies across industries, and of all sizes, have embraced analytics and real-time data as essential to understanding and growing their business. Clinical labs have joined the movement as well. A variety of business intelligence software that connects with existing LIS has come on the market and, for lab managers, it’s no longer enough to think about only QA and audits. Lab managers must now think about growing the lab’s business as a whole, finding new efficiencies in lab workflows, informing partners or sales reps in real time, and reducing costly errors. Like many small businesses, small or startup labs view business analytics as the domain of the “big guys”, for those with resources that far exceed their own. However, it is even more incumbent upon small labs to take a hard look at their lab’s performance in near real time—less wiggle room in the budget means every mistake and inefficiency matters even more. What are some easy steps that labs without expensive business intelligence software can take to optimize their business, make more money, and perform better on important metrics, like turn around time?

Know your error rate, and set goals for improving it

According to Robert Hawkins, M.D. in his paper entitled “Managing the Pre- and Post-analytical Phases of the Total Testing Process” (2012), healthcare and laboratory testing is an exceptionally high risk area as it relates to error rates, as compared to other industries. First and foremost, it is essential for labs to know their error rate. But what does error rate mean for a lab? In many ways, that’s up to you. I would suggest thinking about all the times the lab has to rerun a sample for a single tube repeat, or repeat the entire workflow again with a backup swab or patient sample. In both of these cases, there was an ideal scenario (the sample ran through the workflow on its first pass), and resources that the lab has to expend if the ideal scenario fails (manhours spent on single-tube repeat workflows, or running a backup sample as part of a new batch). Define you error rate, and then measure it...constantly. Choose a time scale, such as a day or a week, and reassess the error rate with the passing of each time period. Then, be clear about your goals based on the baseline error rate. If you have found that your lab has a 5.4% overall error rate for all the workflows that you perform, set a goal that within a particular period of time you’ll have it down to 3.8%. Then, make targeted interventions and measure the performance of those interventions based on their impact on the error rate. Maybe it’s a particular test you run that has an exceptionally high error rate, which contributes to a higher overall error rate. Spend your time and attention there. Dig down into each workflow’s activities, and shake things up a bit: add new QC checks or verify more than usual. And don’t be afraid to write the current error rate, and the goal error rate, on a white board for everyone on staff to see. The more buy in you get for making improvements from your team, the better.

 

Define your workflows and analyze their performance

What tests do you run? What workflows of activities comprise those tests? Do you know how long it takes (not how long you hope it takes!) for a workflow to run, from start to finish? Or how much it actually costs, in terms of man hours and materials, to run a full workflow? Now’s the time to figure out some of those numbers. Here, of course, is where the ever important turnaround time enters the picture, and can be used as a valuable metric for making workflows more efficient. Operations specialists in manufacturing use the term “bottlenecks” to identify spots where the workflow takes an inordinate amount of time, or in the case of a clinical lab, samples begin to collect in a backlog waiting to go through an especially time consuming activity. The first key to improving efficiency is to identify those backlogs or activities that take a good bit longer than others, and find ways, if possible, to reduce that time and thereby improve turnaround time. But you won’t be able to find these answers without some data….so be ready to time how long a process takes SEVERAL TIMES in order to find an average. Don’t be afraid to cross reference those numbers with the cost of human resources expended to perform the tasks, and put a dollar amount on the performance of each activity. Soon, you’ll have a better picture of your Cost of Revenue and be able to see, based on what your lab is making per sample, how much you make off of each test.  Based on this, you can set some goals for improving the profit margin.

 

Know your sales representatives (and what they bring into your business)

One of the best ways to manage the volume of samples coming into your lab is to gain critical insight into the people making those inflows happen: the sales representatives or marketing groups out in the field. One of the best tools for managing these relationships, and maintaining a close watch on the best performers, is with a customer relationship management system, or CRM. With a CRM, you can maintain a list of your sales reps, the clinics they’re responsible for, the samples that have come in from those clinics, and ultimately the revenue that a sales rep has brought into your business. Know who your top tier sales reps are and make sure they know they’re appreciated, and similarly know who your worst performers are and make interventions where necessary to improve that relationship.  Also, keep in mind that these sales reps will need real-time status updates on samples they routed to your lab, so you may want to consider having a Sales Portal that gives them access to that information. You can measure the value of your time, and the energy you put into the relationships with sales reps, in dollars! So be smart and targeted, and manage that resource accordingly.

 

Great information doesn’t have to cost an arm and leg, if you’re making good choices now

I hear from lab staff members and lab software experts what a pain it is to get any information out of their LIMS or from their billing provider. I’ve heard the LIMS referred to as a data graveyard, and heard horror stories of lab leadership who, when they need information to run their business better, has to request a particular report with very specific parameters from the IT staff they keep on hand, and even then access it only from a particular computer. I’m happy to say that the era of that kind of relationship with technology is coming to an end, as LIMS and billing providers have become cloud-based and open, with easily queryable APIs and great interfaces for accessing information. Be sure when you’re making a decision on which LIMS or billing provider you choose to use, or if you’re considering upgrading from your current system, you make information access, and the system’s ability to communicate with other software, a top priority. All of this analytical work can stop being worth it if you’re spending endless amounts of your valuable time just collecting the information that you need in order to make a rational analysis.


This post on business analytics for clinical labs has been intentionally brief—it’s a favorite topic of ours here at Ovation, as we try to provide tools that help lab’s grow their business, in addition to providing the essential services that any lab needs, like compliance, document management, and sample tracking. We’re curious though: how are you bringing business analytics into your lab, whether to improve your error rate or increase efficiency? Tweet at us at @OvationSL or send us an e-mail at info@ovation.io. We’re always learning more about this, and we’d love to start a conversation about it.

Get the whitepaper