The Future Is Here! But How Do We Deal With It?

At a Glance

• Molecular diagnostics give pathologists more information on patients and diseases than ever, but more isn’t always better
• The increased volume of data can lead to confusion, prompt unnecessary testing, or discourage pathologists from updating their knowledge
• Each unique laboratory needs to balance the pros and cons of technological advancement and prioritize the most efficient advancements
• To best understand and make use of new technology, we should establish strong lines of communication with other labs and professional sectors

Change is everywhere in pathology. With each newsletter, magazine or journal article, we’re seeing new technology and new applications for it – faster sequencing, more efficient sample preparation, cheaper software, and better ways of acquiring a lot of information at minimal cost. But while all of these advances are valuable, I think that value may be reaching a plateau. What does that mean? To me, it means that we’re generating information that is so detailed and so extensive that we’re unable to take full advantage of it. In essence, it means that we now have access to more information than we can use – so if more data isn’t the answer, how do we keep advancing? We need to look beyond technological improvements to determine how laboratories can make the best and most efficient use of the wealth of information now available to them.

We’ve seen huge developments in molecular diagnostics in the past decade. Many technologies that were originally exclusively used within research have now become more refined and user-friendly. At the same time, there’s been a marked decrease in the cost of those technologies – a combination that has made them more applicable to the needs of a routine clinical laboratory. When these changes were new, clinical virology laboratories were natural early adopters because the advantages (in terms of time, cost and labor savings) offered by molecular techniques were so overwhelming when compared to traditional techniques like cell culture. But since then, the other areas of clinical microbiology have also begun using genetic and molecular testing to inform their conclusions. And the technology is useful – it shows increased sensitivity and specificity over traditional phenotypic diagnostic methods, and often yields better reproducibility. In fact, “black box” technologies that automate laboratory tasks have shown advantages in reduced hands-on time, quick turnaround time to result and reliability – advantages that have real benefits for patient care and the quality of the support we can provide. The data also help us to better understand infection and prevention in hospital and community settings, as well as giving us the tools to continually improve and challenge ourselves. But if we opt to employ new technologies, especially ones that automate laboratory tasks that were previously our responsibility, we need to ensure that we aren’t sacrificing a complete and detailed understanding of the diagnostic process.

More information, even with regard to identifying and characterizing disease, doesn’t always mean a better outcome for us or the patient. It forces us to ask ourselves, “What does this information mean?” And sometimes, having so much data to interpret can even be a disadvantage. Where we might previously have obtained a negative test result, we might now get a positive; where we might previously have had a single positive result, we now have multiple results to interpret and weigh against one another. In adopting new technologies, we risk reducing our understanding of the reasons we conduct a given test. The initial stages of clinical diagnosis, especially the selection of appropriate tests, have been an important part of our role. But if people don’t fully understand how and why certain testing is done, it becomes much harder to identify what’s truly needed, and we run the risk of “screen” testing, which – if inappropriate – can be as dangerous as not testing at all. Risks like these might make laboratory professionals want to go back to older methods where results are more familiar and interpretation simpler.

A balancing act

Rather than sacrifice the technological gains we’ve made since molecular diagnostics came to the fore, it’s important to work out a balance – and there’s no one right way of doing this. Each lab, and indeed each person working in the lab, has different needs and limitations. So if we can’t identify a way forward that works for everyone, what can we do instead? Some areas under consideration do apply to every lab – questions of cost, speed, quality, accuracy and precision, and the use of a multidisciplinary approach to take the best possible advantage of the technologies each lab chooses to pursue.

At first glance, speed seems an obvious positive – who wouldn’t want a faster turnaround time on lab tests? But the fastest route to completion isn’t always the most efficient; not every result needs to be available immediately. There are certainly situations in which rapid testing is valuable, but in many cases, doctors don’t review lab test results in a timely manner – or, in some cases, at all (1)! In other cases, even if a test is rapidly completed, doctors may not be able to act on the result with the same speed; for instance, community and general practice centers that request 24- or 48-hour turnaround times for initial results of urine tests may not have the capacity to address the findings in a similar timeframe. So if a provider needs a result in a week’s time, why expend resources and delay potentially more critical tests in order to turn it around in half an hour? Though there’s no question that speed is a valuable advancement, it’s up to us to distinguish the situations where it’s truly worth applying from those in which it might not be necessary.

Automation is another gain that seems unambiguous on the surface. But look a little deeper and it isn’t always the right approach. The advantages are clear: machines can handle more samples at a time than humans, performing more complex analyses on all of them simultaneously. Handing testing over to an automated platform can save time, preserve resources, and ease the hiring burden on labs already struggling to find staff with the skills they need. And machines can run 24 hours a day, every day, if necessary – something that would seriously cut into pathologists’ personal lives! But even with all of these benefits, automation isn’t a panacea for laboratory challenges. The recent push toward automated bacteriology, for instance, may not always provide significant benefits compared to manual work – so upgrading might not be worth the time and resources it would take. And not every laboratory has sufficient testing volume to justify purchasing automation technology, especially in the case of newer – and therefore still expensive – devices and software. As with speed, it’s our job to determine when and how automation can be useful, and then to make our case to the people who control the purse strings. Solutions for any given lab might take the form of purchasing new machines, or might involve improving our use of technologies we already own, teaming up with other laboratories to share devices, or outsourcing certain tests.

It’s true that the costs of molecular diagnostics are dropping rapidly. For some time, they followed a Moore’s-law-esque trend of steady decrease; however, after next-generation sequencing eclipsed Sanger-based methods in 2008, the drop was precipitous (2). But as I said earlier, our priorities shouldn’t be focused on getting as much data as possible – because with that as our goal, we may end up getting so much information that we can’t translate it all into clinical services. It’s important to ask ourselves what clinical value any given piece of information might have – and then to make the case for acquiring that information to the people responsible for funding its acquisition. This is especially true when these costs go beyond a single test; for instance, buying a new piece of equipment or hiring a new staff member with specialized skills. Unfortunately, as pathologists, we’re often tasked with striking a balance between cost and our other priorities, so it’s important for us to be able to identify our needs, figure out how best to meet them, and explain our conclusions to the people in charge of funding.

Step out of the silo

It should be known that these challenges aren’t unique to any one lab – you’re not alone! It’s difficult to establish strategies for advancement when there’s no good one-size-fits-all solution to present, but there’s a lot to be gained by communicating with other pathologists and learning what they’re doing. For labs that are trying to increase or improve their use of molecular diagnostics, I recommend dropping the “silo mentality” we tend to prefer and instead developing a strong network. Share best practices, collaborate with other pathology centers, and break down the divisions between clinical and academic environments. Over the past four years, I’ve been working closely with colleagues both at my home Nottingham University Hospitals trust and at the University Hospitals of Leicester to implement a consolidated laboratory model (in which local labs reconfigure collaboratively to create greater efficiency) – and it’s had significant benefits. Consolidation is difficult, complicated and time-consuming, but sharing best practices and common challenges has helped us to devise solutions together. The same applies to liaisons between healthcare, academia and industry – three sectors that have historically treated one another with some trepidation. There’s a lot we can learn from one another, and a lot of opportunities to be gained from making connections between different professional groups.

As a profession, we’re getting better at discussion and collaboration, but the culture of isolation and competition is a long-established one and it’s difficult to move to a more open way of working. We face the same problem with finances; those pathways aren’t currently very transparent, and making them more so would allow us to understand the cost of the complete patient pathway and work toward more efficient improvement by investing in areas that will ultimately result in greater savings or patient benefit. The key is to keep the patient at the center of everything we do. As we start to share more of how we operate – the good, the bad and the ugly – seeing the improvements to patient care should encourage others to participate as well.

The rise of molecular diagnostics, and our need to manage and streamline large amounts of information, has presented us with unique challenges we’re still learning to address. But this isn’t the first time pathology has tackled such a radical change. In the beginnings of the profession, even identifying the organism associated with a clinical presentation was difficult and confusing – but we didn’t step back; we carried on trying to understand, and I see the same thing happening today. As long as we remember that more doesn’t always mean better, and focus on using collaboration and prioritization to make things better for our patients, then I think molecular diagnostics will prove to be not only an impressive, but also a useful, tool.