In ED and POC settings, molecular tests improve treatment of infectious diseases

By Alan Wright, MD, MPH

The convenience care movement is quickly transforming how millions of Americans access primary care. According to the Robert Wood Johnson Foundation, more than 10 million Americans visit retail health clinics every year.1 CVS Health Corp alone operates more than 1,000 MinuteClinics in 33 states, handling everything from health screenings and vaccinations to treatment of minor injuries and illnesses.2 In addition, many thousands of patients are treated each year at as many as 500 freestanding emergency department (ED) facilities in 40 states.3

Alan Wright, MD, MPH, Roche.

Alan Wright, MD, MPH, Roche.

So far, advanced molecular diagnostics employing nucleic acid amplification technologies have not been a significant part of the movement, but there are reasons to believe this will change. Convenience care is booming because consumers are seeking health solutions that are more conducive to busy lifestyles. When an individual with flulike symptoms walks into an urgent care center, retail clinic, or ED facility, they expect to be provided with a reliable diagnosis that leads to appropriate treatment. Now, there is a growing body of evidence that molecular diagnostics could help point-of-care (POC) providers meet patients’ expectations regarding the speed and accuracy of their diagnoses.

Figure 1. Global prevalence of influenza A/B. Infographic courtesy Roche Diagnostics. Click to expand.

Figure 1. Global prevalence of influenza A/B. Infographic courtesy Roche Diagnostics. Click to expand.

In the past year, FDA has cleared several molecular diagnostic systems for use in physicians’ offices, retail settings, and ED facilities. Demand for such tests is especially strong in the area of flu, which affects approximately 5% to 20% of the US population each year, causing more than 200,000 hospitalizations (see Figure 1).4 In recent months, FDA has also approved molecular tests for streptococcal pharyngitis—known as Group A strep, or strep throat (see Figure 2)—and for respiratory syncytial virus (RSV) for use in the same settings.


The potential benefits of the newly approved molecular POC flu tests become clear when placed in the context of recent flu outbreaks. There have been four documented flu pandemics since 1918, each with a different profile. In 2009, the United States experienced its first influenza pandemic in nearly 40 years. By November of that year, 48 states had reported cases of “swine flu” caused by a new subtype of the H1N1 flu virus that leads to more severe symptoms in younger people. By April 2010, CDC estimated that up to 89 million people had contracted the flu virus, and as many as 18,300 deaths were related to H1N1 alone.5

Figure 2. Global prevalence of strep A. Infographic courtesy Roche Diagnostics. Click to expand.

Figure 2. Global prevalence of strep A. Infographic courtesy Roche Diagnostics. Click to expand.

A major genetic change in the influenza A virus made 2009 a particularly bad year for flu, but nearly every year the nation experiences epidemic-size outbreaks of the illness. These are caused by several distinct strains—the most common being two influenza A viruses (H1N1 and H3N2) and influenza B viruses—that accrue slight genetic changes over time.

A new subtype of H1N1 contributed to the medical crisis of 2009, but shortcomings in diagnosis may also have played a role. Although the rapid influenza diagnostic tests (RIDTs) used in hospitals and some POC settings produce results in less than 30 minutes, according to CDC such immunoassays also have some distinct disadvantages. Specifically, the suboptimal sensitivity of the tests means that false-negative results are common when flu activity is high.6 And despite high specificity rates for most RIDTs, false-positive results occur when flu activity is low (see Figure 3).

The challenges posed by false-negative results from RIDTs have been well publicized. ED physicians and nursing staff strive to follow CDC recommendations for early antiviral treatment of flu, especially when patients exhibit dire respiratory symptoms. But because of high false-negative rates for RIDTs and time delays with confirmatory tests, many physicians and ED staff are compelled to rely on their experience and instincts during flu outbreaks.

The results are not always satisfactory. In a 2015 study conducted by researchers at the Johns Hopkins University department of emergency medicine, sensitivity of clinician diagnosis was poor (36%) when responding to a prospective cohort of 270 adult subjects presenting with acute respiratory illness.7 In a setting such as the 2009 swine flu outbreak, this is precisely the population that would benefit most from fast, high-sensitivity molecular tests in ED and POC settings. As the Johns Hopkins study authors wrote in their conclusion: “Improved methods of influenza diagnosis are needed to help guide management in the clinical setting.”

The challenges of conventional methods of detecting influenza A and B and strep A. Infographic courtesy Roche Diagnostics. Click to expand.

The challenges of conventional methods of detecting influenza A and B and strep A. Infographic courtesy Roche Diagnostics. Click to expand.

In addition to these diagnostic challenges, flu vaccines are often not well tailored to viral strains. Researchers vary in how accurately they predict the genetic makeup of the viruses that will be prevalent in an upcoming flu season. In fact, flu vaccine effectiveness averages about 60%, with considerable yearly variation. In 2014–2015, for example, most of the influenza A (H3N2) viruses circulating were different from the H3N2 virus component of the commercial vaccine. The vaccine’s effectiveness in 2014–2015 was a dismal 19% overall, and was particularly low among people over 65.8 And while CDC recommends that everyone get a flu shot, fewer than half of Americans follow that advice.

Ultimately, the threats posed by flu pandemics can be aggravated by this combination of disappointing vaccination rates, mediocre effectiveness of vaccines in some years, and the relatively poor accuracy of RIDTs at the point of care. The combined shortcomings are likely to fuel demand for advanced molecular tests in ED and POC environments.

In addition to human suffering, there’s an economic toll that could be reduced with better diagnostics. CDC estimates that flu causes upwards of $10 billion in direct medical costs each year.9 Early diagnosis of influenza can ensure prompt treatment with antiviral drugs when they are most effective—within 48 hours of the onset of flu symptoms. This can shorten the duration of illness and prevent complications in those most vulnerable—children under age 2, people over age 65, and those with chronic medical conditions.

Numerous studies have shown the sensitivity of today’s RIDTs is often suboptimal, and varies widely depending on the quality of sample, viral subtype and concentration, and other factors.10–12 The results of these rapid antigen or immunofluorescence tests often need to be sent to a laboratory for confirmation via polymerase chain reaction (PCR)-based molecular tests or viral culture, resulting in the loss of precious time when someone is sick and spreading the virus. Conversely, misdiagnosis can lead to the inappropriate use of antibiotics. A CDC-led study conducted over the 2012–2013 flu season found that only 16% of about 1,800 confirmed flu patients received antiviral treatment, while 30% received one of three common antibiotics.13


The new generation of compact, CLIA-waived molecular diagnostics for POC environments offers significant advantages—sometimes matching or exceeding the results of molecular assays that clinical labs develop on their own to confirm RIDT tests (see Figure 4).

Figure 4. The Cobas Liat analyzer by Roche Diagnostics.

Figure 4. The Cobas Liat analyzer by Roche Diagnostics.

In another 2015 study led by Matthew J. Binnicker, PhD, associate professor of laboratory medicine and pathology at the Mayo Clinic, researchers compared a PCR-based POC test—the Cobas Liat Flu A/B test from Roche Diagnostics, Indianapolis—to the Mayo Clinic’s own routine influenza A/B real-time PCR test.14

The Mayo researchers noted that real-time PCR is the test of choice for laboratory diagnosis of influenza A/B because of its high sensitivity, high specificity, and fast turnaround compared to cell culture. But to get the desired accuracy, there are tradeoffs. Most of the tests must be performed in moderate- or high-complexity labs, and require preanalytic DNA extraction and subsequent interpretation by trained staff. Such tradeoffs—including the need for transportation of samples to the testing lab—create challenges for ED physicians and retail clinic staff who need to make fast patient management decisions.

Figure 5. Inserting an influenza A/B test cartridge in the Cobas Liat analyzer.

Figure 5. Inserting an influenza A/B test cartridge in the Cobas Liat analyzer.

By contrast, the Cobas Liat POC platform used in the Mayo Clinic study is a closed, sample-to-result IVD system with CLIA-waived cartridges for flu, strep, and flu A/B+RSV (see Figure 5) that performs a PCR-based analysis in about 20 minutes. Based on 197 swabs, the POC system assays showed sensitivity of 99.2% for influenza A and 100% for influenza B. According to the researchers, the specificity of both tests was 100%—comparable to the in-house PCR test, with overall agreement of 99.5%. “The Cobas Liat platform requires less than 5 minutes of hands-on time and may help bridge the gap between rapid antigen testing and standard real-time PCR,” the Mayo team concluded.

Similarly, a recent study at the Medical University of South Carolina compared the performance of the Liat system and another POC nucleic acid amplification test to a lab reference standard for detecting influenza A and B viruses in 129 samples from adult and pediatric patients. The authors found that the sensitivities and specificities for flu A and B were all 100% for the Liat system. While such a level of accuracy and reliability exceeds the system’s claimed performance, it is clear that PCR-based POC tests can provide a much more definitive flu result than rapid antigen tests currently in widespread use at the point of care.15

Mounting evidence suggests that PCR-based tests and other molecular diagnostics adapted for ED and POC settings could lead to better treatment of flu and other infectious diseases (see Figure 6). In a study described at an industry workshop held in conjunction with the 2015 annual meeting of the Association for Molecular Pathology, researchers showed how the use of PCR-based POC testing affected real-time patient management during the flu season in early 2015, compared to the use of RIDTs.16 The clinical laboratory and diagnostic effectiveness (CLADE) study was conducted at the Hennepin County Medical Center.

The small footprint of the Cobas Liat analyzer makes it suitable for benchtop use in POC settings.

Figure 6. The small footprint of the Cobas Liat analyzer makes it suitable for benchtop use in POC settings.

For the study, led by Glen T. Hansen, PhD, the medical center’s director of clinical microbiology and molecular diagnostics, researchers collected nasopharyngeal samples from 289 unique patient encounters in the center’s ED during the 2015 flu season, and tested them on the Liat system.

Hansen also asked ED physicians to fill out surveys before and after the tests, in order to elicit the doctors’ assessments of their patients’ condition and their care plans prior to PCR testing, as well as the impact that PCR-confirmed diagnosis had on patient management. The second part of the survey, post-PCR testing, covered plans for admission or discharge and course of treatment, including the use of antivirals. The responses for this part of the survey were confirmed against the patients’ electronic medical records.

According to Hansen, the Cobas Liat flu test demonstrated sensitivity of 96.6% and “led to changes in patient management for 67% of 105 cases evaluated for flu.” For 80% of the cases examined, these changes included alterations to the physicians’ initial antiinfective therapy. And for 18% of the cases examined, the changes included some alteration to patients’ admission or discharge orders.


The simplicity, turnaround time, and high sensitivity of molecular tests approved for CLIA-waived environments give them the potential to strengthen both the convenience care movement and patient care in general. In addition, their ability to provide highly accurate IVD testing at the point of care, while also reducing the need for confirmatory lab testing, will have significant ramifications for cutting healthcare costs.

Like all emerging technologies, the new POC molecular devices will have to win the confidence of payors and the caregivers who use them. If peer-reviewed studies continue to demonstrate that they can streamline the testing process and help healthcare providers make more confident treatment decisions, molecular diagnostics aimed at POC settings could have a significant impact on managing outbreaks of flu and other illnesses.

Alan Wright, MD, MPH, is chief medical officer at Roche Diagnostics Corporation.


  1. Bachrach D, Frohlich J, Garcimonde A, Nevitt K. Building a culture of health: the value proposition of retail clinics. Princeton, NJ: Robert Wood Johnson Foundation, 2015. Available at: Accessed August 2016.
  1. Minute Clinic: about us [online]. Woonsocket, RI: MinuteClinic LLC, 2016. Available at: Accessed August 2016.
  1. Royse D. Free-standing ERs eye lobbying to win state approval for growth [online]. Modern Healthcare. July 4, 2015. Available at: Accessed August 2016.
  1. Influenza (flu): seasonal influenza-associated hospitalizations in the United States [online]. Atlanta: Centers for Disease Control and Prevention, 2016. Available at: Accessed August 2016.
  1. Pandemic flu history [online]. Washington, DC:, US Department of Health and Human Services, 2016. Available at: Accessed August 2016.
  1. Influenza (flu): guidance for clinicians on the use of rapid influenza diagnostic tests [online]. Atlanta: Centers for Disease Control and Prevention, 2016. Available at: Accessed August 2016.
  1. Dugas AF, Valsamakis A, Atreya MR, et al. Clinical diagnosis of influenza in the ED. Am J Emerg Med. 2015;33(6):770–775; doi: 10.1016/j.ajem.2015.03.008.
  1. Influenza (flu): CDC presents updated estimates of flu vaccine effectiveness for the 2014–2015 season. Atlanta: Centers for Disease Control and Prevention, 2015. Available at: Accessed August 2016.
  1. CDC Washington: influenza: perspective on current season and update on preparedness. Atlanta: Centers for Disease Control and Prevention, 2013. Available at: Accessed August 2016.
  1. Evaluation of 11 commercially available rapid influenza diagnostic tests, United States, 2011–2012. MMWR. 2012;61(43):873–876. Available at: Accessed August 2016.
  1. Chan KH, Chan KM, Ho YL, et al. Quantitative analysis of four rapid antigen assays for detection of pandemic H1N1 2009 compared with seasonal H1N1 and H3N2 influenza viruses on nasopharyngeal aspirates from patients with influenza. J Virol Methods. 2012;186(1–2):184–188; doi: 10.1016/j.viromet.2012.09.001.
  1. Yang JR, Lo J, Ho YL, Wu HS, Liu MT. Pandemic H1N1 and seasonal H3N3 influenza infection in the human population show different distribution of viral loads, which substantially affect the performance of rapid influenza tests. Virus Res. 2011;155(1):163–167; doi: 10.1016/j.virusres.2010.09.015.
  1. Havers F, Thaker S, Clippard JR, et al. Use of influenza antiviral agents by ambulatory care clinicians during the 2012–2013 influenza season. Clin Infect Dis. 2014;59(6):774–782; doi: 10.1093/cid/ciu422.
  1. Binnicker MJ, Espy MJ, Irish CL, Vetter EA. Direct detection of influenza A and B viruses in less than 20 minutes using a commercially available rapid PCR assay. J Clin Microbiol. 2015;53(7):2353–2354; doi: 10.1128/jcm.00791-15.
  1. Nolte FS, Gauld L, Barrett SB. Direct comparison of Alere i and Cobas Liat influenza A and B tests for rapid detection of influenza virus infection. J Clin Microbiol. Posted online August 31, 2016; doi: 10.1128/jcm.01586-16.
  1. Hansen G. Experience with the Roche Cobas Liat rapid influenza assay during influenza season: analysis of test performance and qualification on the impact of patient management in the emergency department setting. Poster presented at industry workshop in conjunction with the 2015 annual meeting of the Association for Molecular Pathology. Austin, Tex: Association for Molecular Pathology, 4–7 November 2015.