Rapid Influenza Diagnostic Tests

If you want to learn more, please visit our website Accu-Tell.

Background

Rapid influenza diagnostic tests (RIDTs) are immunoassays that can identify the presence of influenza A and B viral nucleoprotein antigens in respiratory specimens, and display the result in a qualitative way (positive vs. negative) (1). In the United States, a number of RIDTs are commercially available. (See &#;Table 1: Influenza Virus Testing Methods&#; and &#;Table 2: Characteristics of Rapid Influenza Diagnostic Tests&#;.) The reference standards for laboratory confirmation of influenza virus infection in respiratory specimens are reverse transcription-polymerase chain reaction (RT-PCR) or viral culture. RIDTs can yield results in a clinically relevant time frame, i.e., less than approximately 15 minutes. However, RIDTs have limited sensitivity to detect influenza viruses in respiratory specimens compared to RT-PCR or viral culture and negative RIDT test results should be interpreted with caution given the potential for false negative results, especially during peak influenza activity in a community. Some RIDTs use analyzer reader devices to standardize result interpretation.

1 RIDTs do not include rapid molecular assays that have higher sensitivity to detect influenza viruses in respiratory specimens compared to RIDTs. See Guidance for Clinicians on the Use of RT-PCR and Other Molecular Assays for Diagnosis of Influenza Virus Infection for more information.

Advantages and Disadvantages of RIDTs

Advantages

• Produce quick result in less than approximately 15 minutes, simple to perform
• Some RIDTs are cleared for office/bedside use. RIDTs that have been CLIA waived can be used in settings that include point-of-care.

Disadvantages

• Sub-optimal test sensitivity, false negative results are common, especially when influenza activity is high
• Sensitivity of RIDTs to detect influenza B viral antigens is lower than for detection of influenza A viral antigens.
• Although specificity is high, false positive results can also occur, especially during times when influenza activity is low.
• Some RIDTs distinguish between influenza A or B viruses while others do not. RIDTs that provide results on the type of influenza virus (e.g., influenza A or B virus), do not provide information on influenza A virus subtype [e.g., A(H1N1)pdm09 versus A(H3N2)] or specific virus strain information (e.g., degree of similarity to vaccine strains). RIDTs cannot distinguish between seasonal influenza A virus infection and novel influenza A virus infection (due to infection with avian or variant influenza A viruses).

 Top of Page

Use of RIDTs in Clinical Decision-making

RIDTs may be used to help with diagnostic and treatment decisions for patients in clinical settings, such as whether to prescribe antiviral medications. However, due to the limited sensitivities, negative results of RIDTs do not exclude influenza virus infection in patients with signs and symptoms suggestive of influenza. Therefore, if clinically indicated, antiviral treatment should not be withheld from patients with suspected influenza, even if they test negative by RIDT, and further influenza testing of respiratory specimens by molecular assays may be indicated. More information about Antiviral Drugs and recommendations on their use.

Testing is not needed for all patients with signs and symptoms of influenza to make antiviral treatment decisions (See Figures 1-4). Once influenza activity has been documented in the community or geographic area, a clinical diagnosis of influenza can be made for outpatients with signs and symptoms consistent with suspected influenza, especially during periods of peak influenza activity in the community.

 Top of Page

Use of RIDTs for Public Health Purposes to Detect Influenza Outbreaks

RIDTs can be useful to identify influenza virus infection as a cause of respiratory outbreaks in any setting, but especially in institutions (i.e., nursing homes, chronic care facilities, and hospitals), cruise ships, summer camps, schools, etc. Positive RIDT results from one or more ill persons with suspected influenza can support decisions to promptly implement infection prevention and control measures for influenza outbreaks. However, negative RIDT results do not exclude influenza virus infection as a cause of a respiratory outbreak because of the limited sensitivity of these tests. Testing respiratory specimens from several persons with suspected influenza will increase the likelihood of detecting influenza virus infection if influenza virus is the cause of the outbreak, and use of molecular assays such as RT-PCR is recommended if the cause of the outbreak is not determined and influenza is suspected. Public health authorities should be notified promptly of any suspected institutional outbreak and respiratory specimens should be collected from ill persons (whether positive or negative by RIDT) and sent to a public health laboratory for more accurate influenza testing by molecular assays and viral culture.

 Top of Page

Factors Influencing Results of RIDTs

Many factors can influence the accuracy of RIDTs, including:

• Clinical signs and symptoms consistent with influenza
  • Having clinical signs and symptoms consistent with influenza increases the pre-test probability of influenza virus infection, which increases the reliability of a positive RIDT result.
• Prevalence of influenza activity in the population tested
  • Influenza activity varies seasonally, which directly affects the predictive values of RIDTs (See algorithms below [Figures 3 and 4], and Prevention Strategies for Seasonal Influenza in Heath Care Settings.)
• Time from illness onset to collection of respiratory specimens for testing
  • Testing specimens collected within 3-4 days of illness onset (when influenza viral shedding is highest) is more likely to yield positive RIDT results if the patient has influenza.
• Type of respiratory specimen tested
  • RIDTs have different specifications for acceptable specimens (e.g., nasopharyngeal, nasal or throat swab/aspirate). The package insert for the RIDT test used should be reviewed to ensure that an appropriate specimen is collected, and test procedures are followed. Some tests may require specimen collection using a special swab (some RIDTs must be used with a swab supplied with the test kit; some swab material can interfere with RIDT results).
  • RIDTs must also ensure that the appropriate viral transport media or other media is used, consistent with test specifications, if testing is done at a different location from where the specimen is collected from the patient.
  • Collection of good quality respiratory specimens (e.g., nasopharyngeal or nasal swab/aspirate/wash or combined nasal/throat swab specimens) also will increase the accuracy of RIDT results.
  • Some RIDTs require that the entire collected specimen be used in the test. Consider whether a second specimen should be collected for confirmatory testing using viral culture and/or RT-PCR.
• Accuracy of the test compared to a reference test (&#;gold standard&#; = RT-PCR or viral culture)
  • Sensitivity of the RIDT
    • Proportion of positive RIDT results of all positive &#;gold standard test&#; results (RT-PCR or viral culture)
    • Fixed characteristic of a test; generally low to moderate (50-70%) for RIDTs
      • An RIDT with low sensitivity will produce negative results in some patients with influenza (false negatives)
  • Specificity of the RIDT
    • Proportion of negative RIDT results of all negative &#;gold standard test&#; results (RT-PCR or viral culture)
    • Fixed characteristic of a test; generally very high for RIDTs (90-95%)
      • An RIDT with low specificity will produce positive results in some patients who do not have influenza (false positives )

 Top of Page

Interpretation of Rapid Influenza Diagnostic Test Results

Proper interpretation of RIDT results is very important for clinical management of patients and for assessing suspected influenza outbreaks. A number of factors can influence the results of RIDTs. The accuracy of RIDTs depends largely on the conditions under which they are used. Understanding some basic considerations can minimize being misled by false-positive or false-negative results.

Positive result

• A positive result means that the RIDT detected influenza viral antigen, but does not necessarily mean viable influenza virus is present or that the patient is contagious.
• The positive predictive value of an RIDT (the proportion of patients with positive results who have influenza) is highest when influenza activity is high in the population being tested (e.g. community).
• A positive result is most likely a true positive result if the respiratory specimen was collected close to illness onset (within 4 days) during periods of high influenza activity (e.g. winter) in the population being tested (e.g. community).
• A positive result in a person who recently received intranasal administration of live attenuated influenza virus vaccine (LAIV) may indicate detection of vaccine virus. LAIV contains influenza virus strains that undergo viral replication in respiratory tissues of lower temperature (e.g., nasal passages) than internal body temperature. Since the nasal passages are infected with live influenza virus vaccine strains during LAIV administration, sampling the nasal passages within a few days after LAIV vaccination can yield positive influenza testing results. It may be possible to detect LAIV vaccine strains up to 7 days after vaccination, and in rare situations, for longer periods.
• The positive predictive value of an RIDT (the proportion of patients with positive results who have influenza) is lowest when influenza activity is low in the population being tested (e.g. community).
• False-positive  results are more likely to occur when influenza  prevalence in the population tested (e.g. community) is low, which is generally at the beginning and end of the influenza season and during the summer.

Negative result

• A negative result means that the RIDT did not detect any influenza viral antigen.
• The negative predictive value of an RIDT (the proportion of patients with negative results who do not have influenza) is highest when influenza activity is low in the population being tested (e.g. community).
• A negative result is most likely a true negative result if the respiratory specimen was collected close to illness onset (within 4 days) during periods of low influenza activity (e.g. summer) in the population being tested (e.g. community).
• The negative predictive value of an RIDT (the proportion of patients with negative results who do not have influenza) is lowest when influenza activity is high in the population being tested (e.g. community).
  • • False-negative  results are more likely to occur when influenza  prevalence is high in the community.
• Sensitivities of RIDTs are generally approximately 50-70%, but a range of 10-80% has been reported compared to viral culture or RT-PCR. Specificities of RIDTs are approximately 90-95% (range 85-100%). Thus false negative results occur more commonly than false positive results.
• Negative results of RIDTs do not exclude influenza virus infection and influenza should still be considered in a patient if clinical suspicion is high based upon history, signs, symptoms and clinical examination.

Minimize False Results

• Collect specimens as early in the illness as possible (ideally less than 4 days from illness onset).
• Follow manufacturer&#;s instructions, including acceptable specimens, and handling.
• Follow-up negative results with confirmatory tests (RT-PCR or viral culture) if a laboratory-confirmed influenza diagnosis is desired.

 Top of Page

Clinicians should contact their local or state health department for information about current influenza activity. For more information about influenza activity in the United States during the influenza season, visit the Weekly U.S. Influenza Surveillance Report (FluView).

When to Consider Further Influenza Testing

Consider sending respiratory specimens for influenza testing by viral culture or RT-PCR to confirm results of an RIDT when:

• A patient tests negative by RIDT when community influenza activity is high and laboratory confirmation of influenza is desired.
• A patient tests positive by RIDT and the community prevalence of influenza is low, and a false positive result is a consideration.
• A patient has had recent close exposure to pigs or poultry or other animals and novel influenza A virus infection is possible (e.g., influenza A viruses circulate widely among swine and birds, including poultry, and also can infect other animals such as horses and dogs) See Avian Influenza: Information for Health Professionals and Laboratorians for more information.

Hospitalized patients

Influenza testing is recommended for hospitalized patients with suspected influenza. Molecular assays such as RT-PCR are recommended for testing hospitalized patients. However, empiric antiviral treatment should be initiated as soon as possible for hospitalized patients with suspected influenza without the need to wait for any influenza testing results (see Antiviral Drugs, Information for Health Care Professionals). Antiviral treatment should not be stopped based on negative RIDT results given the limited sensitivities of RIDTs. Infection prevention and control measures should be implemented immediately upon admission for any hospitalized patient with suspected influenza even if RIDT results are negative (see Prevention Strategies for Seasonal Influenza in Heath Care Settings). Serology for influenza should not be performed for clinical management. Clinicians should understand that negative results of influenza testing do not exclude influenza virus infection, especially if the time from illness onset to collection of respiratory specimens is more than 3  days, or if upper respiratory tract specimens were tested and the patient has lower respiratory tract disease. If influenza is suspected, testing of clinical specimens collected from different respiratory sites can be done (e.g., upper and lower respiratory tract) and can be collected on more than one day to increase the likelihood of influenza virus detection; intubated patients should have endotracheal aspirate specimens tested if influenza is suspected, but not yet confirmed.

Detection of influenza virus infection and prompt implementation of infection prevention and control measures is critical to prevention of nosocomial influenza outbreaks. When there is influenza activity in the community, clinicians should consider influenza testing, including viral culture, for patients who develop signs and symptoms of influenza while they are in a health care facility. This should be done as part of a broader surveillance strategy for influenza as discussed in Prevention Strategies for Seasonal Influenza in Heath Care Settings.

Suspected influenza institutional outbreaks

For suspected influenza outbreaks in institutions, respiratory specimens should be collected from patients with suspected influenza as early as possible once the outbreak is suspected (See Figure 2). The use of influenza molecular assays is preferred. If RIDTs are used in these settings, clinical specimens should also be sent for influenza testing by viral culture and RT-PCR to provide detailed information on specific influenza A virus subtypes and strains, and antiviral susceptibility data and to verify RIDT test results. Active daily surveillance for suspected influenza illness and collection of specimens from patients with suspected influenza should continue through at least 2 weeks after implementation of control measures to assess effectiveness of the measures and to monitor for potential emergence of antiviral resistance. See Prevention Strategies for Seasonal Influenza in Heath Care Settings.

Influenza Surveillance

Laboratory-based surveillance for influenza viruses by viral culture is critically important to the selection of viruses for the next season&#;s influenza vaccine. Virus isolates are needed in order to characterize the circulating influenza A virus subtypes and influenza A and B virus strains and to determine how well they are matched antigenically to vaccine strains. Isolates are also needed for obtaining information on the emergence and prevalence of antiviral resistant strains, and the identification of human infection with novel influenza A viruses (e.g., an influenza A virus of animal origin that may sporadically cause illnesses in people) that may have pandemic potential. This information is needed from specimens sent for viral culture and RT-PCR year round for identification of novel influenza A virus strains or antigenically-drifted seasonal influenza virus strains, including during times of low influenza activity such as at the beginning and end of influenza seasonal activity. For more information about influenza activity in the United States during the influenza season, visit the Weekly U.S. Influenza Surveillance Report (FluView).

 Top of Page

References

Ali T, Scott N, Kallas W, Halliwell ME, Savino C, Rosenberg E, Ferraro M, Hohmann E. Detection of influenza antigen with rapid antibody-based tests after intranasal influenza vaccination (FluMist). Clin Infect Dis. Mar 1;38(5):760-2.

Balish A, Garten R, Klimov A, Villanueva J. Analytical detection of influenza A(H3N2)v and other A variant viruses from the USA by rapid influenza diagnostic tests. Influenza Other Respi Viruses. Jul;7(4):491-6. doi: 10./irv..

Block SL, Yogev R, Hayden FG, Ambrose CS, Zeng W, Walker RE. Shedding and immunogenicity of live attenuated influenza vaccine virus in subjects 5-49 years of age. Vaccine. Sep 8;26(38):-6.

Centers for Disease Control and Prevention (CDC). Evaluation of rapid influenza diagnostic tests for influenza A (H3N2)v virus and updated case count&#;United States, . MMWR Morb Mortal Wkly Rep. Aug 17;61(32):619-21.

Centers for Disease Control and Prevention (CDC). Evaluation of 11 commercially available rapid influenza diagnostic tests &#; United States, -. MMWR Morb Mortal Wkly Rep. Nov 2;61:873-6.

Centers for Disease Control and Prevention (CDC). Evaluation of rapid influenza diagnostic tests for detection of novel influenza A (H1N1) Virus&#;United States, . MMWR Morb Mortal Wkly Rep. Aug 7;58(30):826-9.Chartrand C, Leeflang MM, Minion J, Brewer T, Pai M. Accuracy of rapid influenza diagnostic tests: a meta-analysis. Ann Intern Med. Apr 3;156(7):500-11.

Committee on Infectious Diseases, American Academy of Pediatrics. Policy Statement&#;Recommendations for Prevention and Control of Influenza in Children, &#;.

Faix DJ, Sherman SS, Waterman SH. Rapid-test sensitivity for novel swine-origin influenza A (H1N1) virus in humans. N Engl J Med. Aug 13;361(7):728-9.

Grijalva CG, Poehling KA, Edwards KM, Weinberg GA, Staat MA, Iwane MK, Schaffner W, Griffin MR. Accuracy and interpretation of rapid influenza tests in children. Pediatrics. Jan;119(1):e6-11.

Harper SA, Bradley JS, Englund JA, File TM, Gravenstein S, Hayden FG et al. Seasonal influenza in adults and children&#;diagnosis, treatment, chemoprophylaxis, and institutional outbreak management: clinical practice guidelines of the Infectious Diseases Society of America. Clin Infect Dis. Apr 15;48(8):-32.

For more influenza swab techniqueinformation, please contact us. We will provide professional answers.

Moesker FM, van Kampen JJ, Aron G, Schutten M, van de Vijver DA, Koopmans MP, Osterhaus AD, Fraaij PL. Diagnostic performance of influenza viruses and RSV rapid antigen detection tests in children in tertiary care. J Clin Virol. Jun;79:12-7.

Ryu SW, Lee JH, Kim J, Jang MA, Nam JH, Byoun MS, Lim CS. Comparison of two new generation influenza rapid diagnostic tests with instrument-based digital readout systems for influenza virus detection. Br J Biomed Sci. Jul;73(3):115-120.

Uyeki TM. Influenza diagnosis and treatment in children: a review of studies on clinically useful tests and antiviral treatment for influenza. Pediatr Infect Dis J. Feb;22(2):164-77.

World Health Organization.em>WHO Guidelines for Pharmacological Management of Pandemic Influenza A(H1N1) and other Influenza Viruses. Revised February .

Mandell LA, Wunderink RG, Anzueto A, Bartlett JG, Campbell GD, Dean NC, Dowell SF, File TM Jr, Musher DM, Niederman MS, Torres A, Whitney CG; Infectious Diseases Society of America; American Thoracic Society. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults. Clin Infect Dis. Mar 1;44 Suppl 2:S27-72.

 Top of Page

Of the 72 paired specimens analyzed, 25 were positive for influenza A or B RNA by at least one of the collection methods (34.7% positivity rate). When the 14 patients who had prior health care training were excluded, the qualitative agreement between collection methods was 94.8% (55 of 58). Two of the 58 specimens (3.4%) from patients without health care training were positive only by HCW collection, and 1 of 58 (1.7%) was positive only by patient self-collection. A total of 53.4% of patients (31 of 58) preferred the self-collection method over the HCW collection, and 25.9% (15 of 58) had no preference.

We enrolled adult patients presenting with influenzalike illness to the Emergency Department at Mayo Clinic, Rochester, Minnesota, from January 28, , through April 30, . Patients self-collected a midturbinate nasal flocked swab from their right nostril following written instructions. A second swab was then collected by an HCW from the left nostril. Swabs were tested for influenza A and B viruses by real-time reverse transcription&#;polymerase chain reaction, and percent concordance between collection methods was determined.

Few published studies examine the utility of self- or parent-collected swabs for bacteria and respiratory viral detection. 9-14 These studies suggest that the self- or parent-performed swab collection method is an efficient, well-tolerated, and sensitive method for laboratory testing. However, there are no studies directly comparing the sensitivity of patient- and HCW-collected nasal or nasopharyngeal swabs for influenza A and B molecular testing. Therefore, our aim was to determine the degree of concordance between paired midturbinate nasal flocked swabs collected by the patient and an HCW using a real-time reverse transcription&#;polymerase chain reaction (rRT-PCR) assay for influenza A and B viruses. Furthermore, we conducted a survey to determine the patients' preference for self- vs HCW-collected specimens and their perceived degree of difficulty in collecting their specimen with a nasal flocked swab.

A simple approach to prevent close contact in the health care setting and decrease unnecessary utilization of health care services during influenza season is to have select patient groups (eg, previously healthy adults with no underlying medical conditions) collect their own nasal swab and deliver it to a convenient drop-off location for testing. This could be facilitated by a well-designed or online triage system that would obviate the need for a health care visit in selected patients. Use of this strategy would require that patient-collected samples have similar sample quality and test efficacy for detection of influenza A and B virus RNA compared with those collected by a trained HCW.

Surveillance studies have identified health care setting&#;acquired influenza outbreaks. 5 Influenza A and B viruses can spread rapidly between patients and health care workers (HCWs) in health care settings and may have serious and devastating consequences to high-risk populations. In particular, the elderly, critically ill, young children, and immunosuppressed patients including those receiving antineoplastic chemotherapy and/or solid organ or bone marrow transplant are at highest risk of developing serious influenza complications. The risks of infection-driven complications coupled with the clustering of these susceptible patients in the health care setting emphasize the need to minimize both patient and provider influenza virus exposures. 1,6-8 The overall burden of health care facility&#;acquired influenza is unknown, but the potential economic and clinical impact is significant. 3,5,7

Influenza A and B are highly contagious respiratory viruses transmitted directly between infected and healthy individuals by large, virus-containing respiratory droplets generated by coughing or sneezing. Infection can also be transmitted indirectly by contact with contaminated surfaces followed by self-inoculation of ocular, nasal, or oral mucosa. Adults are typically infectious to others from 1 day before developing symptoms to approximately 5 days after symptom onset. Young children and immunocompromised persons may shed virus for even longer, with shedding reported for 10 or more days after symptom onset. 1-4

Percentage agreement between the patient- and HCW-collected nasal swabs was estimated with 95% score confidence intervals, and degree of agreement was summarized with the κ statistic. Viral detection level between patient- and HCW-collected samples was compared using the Cp values with a 2-sided paired t test, and the average difference between the methods was summarized with a 95% confidence interval. Analyses were conducted using SAS version 9 or JMP version 8 (SAS Institute, Cary, NC).

Paired specimens were processed in parallel in the clinical laboratory as part of routine patient specimen testing by technologists who were blinded to the type of collection technique. RNA was extracted by the MagNA Pure 2.0 instrument (Roche Applied Science, Indianapolis, IN), and then influenza A and B RNA was detected using the multiplex rRT-PCR Prodesse ProFlu+ assay (Gen-Probe, San Diego, CA). The assay was modified for use on the LightCycler 480 (Roche Applied Science). A positive result is produced by this assay when amplification of viral RNA generates a fluorescent signal above the instrument detection threshold. The number of rRT-PCR cycles at which the signal crosses the threshold is known as the crossing point (Cp), which is inversely correlated to the logarithm of the initial copy number; higher Cps correlate with lower levels of viral RNA. Discordant results were resolved using previously described laboratory-developed influenza A and B rRT-PCR assays. 15

Participants first collected a midturbinate specimen with a nasal flocked swab (Diagnostic Hybrids, Athens, OH) from the right nostril following written instructions in English ( ). All patients were observed during self-collection by the HCW for adherence to instructions. Language interpreters were used when necessary to read the instructions to the participant; however, no collection guidance was provided by the HCW or interpreter. A second midturbinate nasal flocked swab was collected immediately after the patient collection by a trained HCW from the left nostril. Each swab was placed in an individually labeled container with universal transport media (Diagnostic Hybrids, Athens, OH) and sent to the clinical microbiology laboratory for rRT-PCR testing. All specimens were transported within 2 hours of collection using the hospital pneumatic tube system and stored in the laboratory at 4 ο C until processed. After swab collection, the participants were asked to complete a questionnaire that subjectively rated the degree of difficulty in obtaining the self-collected swab (very difficult, difficult, easy, or very easy) and their preference for collection technique (self-collected, HCW-collected, or no preference). Demographic information including age, sex, race, occupation, and influenza vaccination status was also obtained. Patients who were dependent on oxygen via nasal cannula or were unable to provide verbal consent were excluded from the study. All positive results (regardless of collection method) were reported to the health care provider (primary care physician, nurse practitioner or physician assistant) to ensure appropriate patient management.

The study was approved by the Mayo Clinic Institutional Review Board, and verbal consent was obtained from all participants. Patients comprised adults aged 18 years and older presenting to the Saint Marys Emergency Department (ED), an academic ED that is part of Mayo Clinic, Rochester, Minnesota, with influenzalike illness (ILI) from January 28, , to April 30, . Influenzalike illness was defined by the presence of fever (measured at &#;37.7 ° C or reported by the patient) and either cough or sore throat per Centers for Disease Control and Prevention guidelines ( http://www.acha.org/ILI_Project/ILI_case_definition_CDC.pdf ). Additional symptoms such as runny nose, nasal congestion, irritability, chills, body or muscle ache, lethargy, weakness, and vomiting were also recorded.

The results of the patient satisfaction survey to assess perceived level of comfort of self-collection and preference for collection method are shown in . To avoid bias, the patients with prior health care training have been excluded. Only 6 (10.3%) of the patients graded the self-collection method as difficult, whereas the remaining 52 patients (89.7%) found it either easy or very easy. Most participants (53.4%) preferred the self-collection technique, a smaller fraction (20.7%) preferred the HCW collection, and 25.9% had no preference.

The Cp data were compared between collection methods for the subset of patients without prior health care training. Among the 20 positive results for influenza A and B by at least one of the collection methods, 17 were positive by both methods and thus had Cp data that could be compared. Of concordant positive results, the HCW-collected specimens tended to have lower Cp ± SD as compared with patient-collected specimens (Cp, 24.8±4.2 vs 26.6±4.5, respectively; P=.02), indicating slightly higher RNA yield from the HCW-collected specimens ( ).

In the entire data set of 72 paired specimens, there were a total of 4 discordant results. Two paired specimens (1 obtained by a patient with prior health care training) were positive by the patient collection method but negative by the HCW collection method. Confirmatory testing on these discrepant specimens was performed in triplicate using an influenza A and B rRT-PCR laboratory-developed test, which showed that both collections were positive in at least 1 of the 3 replicates (data not shown), indicating that the discordance was due to low viral RNA levels in the specimen. There were also 2 paired specimens that were positive by the HCW collection and negative by the patient collection. The results did not change on repeat testing. As described earlier in this article, these 2 specimens were collected by patients who were nonadherent to the written instructions. When patients with prior health care training were removed from the analysis, there were 3 discordant results.

Twenty-five of the 72 total paired specimens (34.7%) were positive for influenza A or B RNA by at least one of the collection methods. Overall, 14 patients (19.4%) had prior health care training. To remove the potential bias of this training, these participants were excluded from the subsequent comparison analyses. In the subset without prior health care training, 20 of the 58 paired specimens collected (34.5%) were positive for influenza A or B RNA by at least one of the collection methods ( ). Seventeen of the 20 positive specimens (85.0%) were positive for influenza A virus and 3 (15.0%) were positive for influenza B virus by at least one of the collection methods. The overall qualitative agreement between collection methods was 94.8% (95% confidence interval, 85.9%-98.2%), with a similar rate of influenza detection by both methods (κ=0.88).

Four of the 72 patients (5.5%) showed gross nonadherence to the written instructions by inserting the swab only into the superficial nares and leaving it in place for 1 to 2 seconds. All 4 of these patient-collected specimens were negative for influenza A and B RNA by rRT-PCR. The HCW-collected swabs from 2 of these patients were positive for influenza RNA.

Eighty-seven patients with a clinical diagnosis of ILI were considered for participation in this study. Of these, 10 (11.5%) were determined to be ineligible or declined to participate because of the presence of a nasal oxygen cannula, altered mental status, or lack of interest in the study. An additional 5 participants (5.7%) declined to participate because they were not comfortable obtaining their own nasal sample. Ultimately, paired midturbinate nasal flocked swabs were obtained from 72 patients. Median age was 39.5 years (range, 18-92 years), and 45.8% (33/72) were males ( ). Seventy-five percent of the study participants (54/72) were white. The summary of presenting symptoms is provided in . In addition to fever (required for inclusion), cough and body or muscle aches were the predominant symptoms. Only 50.0% of the study participants (36 of 72) had received the annual influenza vaccine.

Discussion

Our study demonstrates that HCW- and patient-collection techniques using midturbinate nasal flocked swabs were comparable for influenza A and B RNA detection by rRT-PCR. There was no significant difference in the overall positivity rate by either collection method. Although there were slight differences in the observed rRT-PCR Cps between patient- and HCW-collected specimens, this did not change the qualitative result. Further, most participants stated that the self-collection technique was easy or very easy to perform and preferred this method over the HCW collection method.

To the best of our knowledge, no studies have been conducted to directly compare the patient self-swabbing technique with HCW swabbing for respiratory virus molecular testing. However, several studies have shown in pediatric patients that parental collection of midturbinate, nasal, and/or throat specimens for testing viruses such as human metapneumovirus, influenza A virus, influenza B virus, respiratory syncytial virus, parainfluenza viruses, and adenoviruses is an efficient and acceptable method of conducting vaccine efficacy studies and other community-based respiratory virus research.10,11,14 Esposito et al14 directly compared parent-collected midturbinate nasal swabs with pediatrician-collected swabs for influenza detection by rRT-PCR and demonstrated moderately high sensitivity (89.3%) and high specificity (97.7%) for the parental collection technique. Further, they demonstrated that direct involvement of parents in the collection process increased the child's acceptance of the midturbinate nasal flocked swab. Our study results are consistent with those in the Esposito study in that we found similar concordance between patient- and HCW-collected specimens with a patient preference for the self-collection method.

Like Esposito et al, we used the recently introduced midturbinate nasal flocked swabs for sample collection. Unlike traditional swabs that are constructed by wrapping absorbent rayon fibers around a sturdy applicator, flocked swabs have no internal mattress core to entrap the sample. As a result, the entire sample stays close to the surface and elutes quickly and completely into testing media. Both swabs have a similar diameter. Recent studies have indicated that midturbinate nasal flocked swabs have high sensitivity and specificity for detection of common respiratory viruses by direct immunofluorescent antibody and molecular assays compared with nasopharyngeal swabs, while only being inserted to half the depth.14,16-18 Thus, the midturbinate nasal flocked swabs provide a less invasive yet sensitive alternative to nasopharyngeal swabs and are more amenable to patient self-collection. The swabs used in this study have a safety collar to indicate the depth of insertion and are also available in a pediatric version for children aged 2 years or younger.

In our patient subset without prior health care training, we observed 3 discordant results. The 2 paired specimens that were positive by HCW collection and negative by patient collection were anticipated at the time of collection because both patients showed gross nonadherence to the written instructions. This indicates that quality of collection is important for influenza virus molecular testing. An additional paired specimen was positive by the patient collection but negative by the HCW collection. Confirmatory testing on this third specimen indicated that the discrepancy was due to the low level of virus present, as evidenced by a high Cp.

A potential bias in this study is that we did not randomize the participants for collection by nostril side. However, there is no evidence in the literature that virus shedding is affected by the side of sample collection. Our study also had a few limitations. First, the study had a relatively small number of patients with influenza B infection compared with those with influenza A, although it is unlikely that the method of collection would have any influence on detection of influenza by type. Second, most study participants were white, and hence, the results may not be generalizable to broader populations with disparate socioeconomic, educational, and racial backgrounds. Third, since midturbinate nasal flocked swabs were employed in this study, extrapolation of results to commercially available molecular systems that require use of nasopharyngeal swabs may require further validation studies. Finally, low level of virus in specimens can result in discrepant results between HCW and self-collection methods, as observed in this study.

Our study results have a number of practical implications. First, the patient-collection strategy may reduce time spent in the outpatient clinic or emergency department (ED) and thus reduce influenza exposure risks to other vulnerable patients and HCWs. We view this as a considerable advantage because outpatient clinic and ED waiting rooms often collectively group immunocompetent and immunocompromised patients together, and waiting times are commonly prolonged. Patient protection is a priority, and minimizing unintentional exposures to influenza virus is a primary objective of this testing approach. Second, self-collection may provide a more time-efficient testing approach, which could lead to an earlier diagnosis of influenza infection and initiation of time-sensitive antiviral treatment. Third, this strategy could decrease the burden on busy outpatient clinics and EDs during peak influenza season if combined with a triage method to determine which patients should be tested and if self-collection is an appropriate strategy. Finally, patient collection may be useful for large epidemiological or vaccine efficacy studies in which patient or parent collection may ensure better compliance. Of note, this method potentially could be used for detection of other respiratory viruses in addition to influenza A and B, such as respiratory syncytial virus, parainfluenza virus, rhinovirus, and coronavirus. This should be examined further using laboratory-developed or commercially available individual and multiplex molecular respiratory platforms.

Despite the advantages of self-collection, we encountered potential barriers for successful implementation of this technique. Five potential participants were not comfortable obtaining a midturbinate nasal flocked swab themselves and opted not to participate in the study for this reason. Therefore, self-collection programs may wish to provide an alternative mechanism for HCW collection when necessary. It is also important to note that some patients may encounter difficulties in reading or interpreting written instructions or opening the swab packaging. Consideration should be given to providing instructions in multiple languages and providing easy-to-open packaging. Picture-based instructions may be created for a wider variety of patients, including illiterate patients, patients with limited reading skills, children, and those for whom English is not the first language. Further, pictures may improve adherence to instructions by avoiding misinterpretations of written instructions even among literate patients.

Self-collection in lieu of an office or ED visit is not appropriate for all patients, such as those with complicated illness or risk factors for severe disease. It may also not be useful to test all patients who present with an ILI if treatment is going to be prescribed regardless of the laboratory result. Diagnosing influenza, however, can avoid unnecessary antibacterial therapy in the appropriate early settings. Practice-specific algorithms can be used to direct patients to appropriate testing (or nontesting) options.

On the basis of our experience, we propose a model for &#;personalized patient care&#; by providing a structured system for patient-collected nasal swabs ( ). Patients could first be screened over the by a trained HCW using a standardized triage questionnaire to determine whether self-collection is appropriate for that particular patient according to an assessment of risk factors and disease severity. If treatment decisions will not be influenced by the test result, then testing may not be indicated. If the patient is eligible for self-collection, then the patient or a caregiver would be offered the option of obtaining a swab kit with instructions from an easy access point such as a community-based clinic, pharmacy, drive-through facility, or even a specialized vending machine. If patients are not comfortable obtaining their own specimen, then they may be offered an office visit. After self-collection, the patient or caregiver would drop the swab off at a convenient location designed to maximize testing turnaround time. After testing is performed, the laboratory result could be communicated to the patient or a health care provider (primary care physician, nurse practitioner or physician assistant) who could offer further instruction and decide whether treatment is appropriate. At our institution, a prescription for oseltamivir may be generated by a health care provider and sent to the laboratory along with the patient's swab specimen.15 If the test result is positive for influenza, then the laboratory will fax the prescription to the patient's pharmacy. If the test result is negative, then the prescription is not faxed. This serves to link the prescription to the positive test result and avoid unnecessary drug use. The patients obtain their test results by calling a registered number and providing their unique patient identification number. The automated service informs the patient when a prescription has been sent to the pharmacy on the basis of a positive laboratory result. It is easy to envision how this system could be further modified to suit the needs of the clinical practice and the patients it serves. This same approach has been used for many years at our institution for detection of group A streptococci in patients suspected to have streptococcal pharyngitis.19

In addition to clinical benefits, patient self-collection may alleviate cost burdens to the health care system and patient. At our institution, a visit to the ED for an uncomplicated ILI costs the health care system approximately 1.8 times more than an outpatient office visit, which in turn costs 4 times more than a visit to an express care clinic. In comparison, the salary and indirect costs incurred during a 10-minute call with a registered nurse cost the health care system 5 times less than a visit to an express care clinic. Therefore, the cost incurred to the health care system is significantly reduced when the nurse-triage model is used. One would expect that the charge to the patient would be similarly decreased.

If you are looking for more details, kindly visit hiv1 2.