Determining the Seasonality of Respiratory Syncytial Virus in the United States: The Impact of Increased Molecular Testing

Abstract

Background: In the United States, the seasonality of respiratory syncytial virus (RSV) has traditionally been defined on the basis of weeks during which antigen-based tests detect RSV in >10% of specimens (hereafter, the "10% threshold"). Because molecular testing has become more widely used, we explored the extent of polymerase chain reaction (PCR)-based RSV testing and its impact on determining the seasonality of RSV.

Methods: We assessed antigen- and PCR-based RSV reports submitted to the National Respiratory and Enteric Virus Surveillance System during July 2005-June 2015. To characterize RSV seasons by using PCR-based reports, we assessed the traditional 10% threshold; subsequently, we developed 3 methods based on either PCR-based detections or the percentage of positive test results.

Results: The annual number of PCR-based reports increased 200-fold during 2005-2015, while the annual number of antigen-based reports declined. The weekly percentage of specimens positive for RSV by PCR was less than that for antigen-detection tests; accordingly, the 10% threshold excluded detections by PCR and so was imprecise for characterizing RSV seasons. Among our PCR-specific approaches, the most sensitive and consistent method captured 96%-98% of annual detections within a season, compared with 82%-94% captured using the traditional method.

Conclusions: PCR-based reports are increasingly relevant for RSV surveillance and determining the seasonality of RSV. These PCR-specific methods provide a more comprehensive understanding of RSV trends, particularly in settings where testing and reporting are most active. Diagnostic practices will vary by locality and should be understood before choosing which method to apply.

Keywords: RSV season; RSV surveillance; antigen diagnostic tests; molecular diagnostic tests; season threshold.

Conflict of interest statement

Potential conflicts of interest. All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

Published by Oxford University Press for the Infectious Diseases Society of America 2017. This work is written by (a) US Government employee(s) and is in the public domain in the US.

Figures

Figure 1

Reporting of respiratory syncytial virus (RSV) diagnostic data by National Respiratory and Enteric Virus Surveillance System (NREVSS) laboratories during July 2005–June 2015. A, Numbers of all reporting laboratories (bars) and those consistently reporting data for each diagnostic method (lines). B and C, RSV tests from consistent reporters (B) and RSV detections from consistent reporters (C), by surveillance year and diagnostic method. NREVSS laboratories that consistently reported data are defined here as those that reported data for an average of ≥10 RSV tests per week annually for at least 30 weeks of NREVSS-defined year. PCR, polymerase chain reaction.

Figure 2

Comparison of respiratory syncytial virus (RSV) data from polymerase chain reaction (PCR) and antigen-detection tests, based on weekly numbers of RSV tests (A) and detections (B) and the weekly percentage of tests positive for RSV (C) across 6 surveillance years during July 2009–June 2015, by diagnostic method. The dashed line in panel C represents the 10% threshold. All data are limited to National Respiratory and Enteric Virus Surveillance System (NREVSS) laboratories that consistently reported data, defined here as those that reported data for an average of ≥10 RSV tests per week annually for at least 30 weeks of NREVSS-defined year. Surveillance started at week 27 for all years except 2009–2010, during which it started at week 26. Calendar years ended at week 52 for all years except 2014, which ended at week 53; there is therefore a gap in some of the data curves.

Figure 3

Comparison of respiratory syncytial virus (RSV) tests, detections, and percentage of tests positive for RSV during July 2009–June 2010, by diagnostic method. RSV tests are represented on a different scale to either of the axes depicted; the peak numbers of RSV tests are instead shown on the figures. Data are limited to National Respiratory and Enteric Virus Surveillance System (NREVSS) laboratories that consistently reported data, defined here as those that reported data for an average of ≥10 RSV tests per week annually for at least 30 weeks of NREVSS-defined year. The 2009–2010 surveillance year started at week 26 in 2009 and ended in week 26 in 2010. The 2009 calendar year ended at week 52.

Figure 4

Methods for defining respiratory syncytial virus (RSV) season, using RSV detections reported on the basis of polymerase chain reaction (PCR) diagnostic assays. A, Generation of onset week, using the retrospective slope 10 (RS10) method. The normalized centered 5-week moving average of weekly RSV detections by PCR during July–June are shown for each surveil-lance year. The onset week for each season is depicted by a solid circle of the same color. B, Generation of onset week, using the 10-fold baseline (10FB) method. The 4-week moving average of weekly RSV detections by PCR, relative to the 4-week moving average at week 29, are shown for each surveillance year during July–June. The horizontal line represents the 10-fold increase in detections relative to week 29. The onset week for each season is a solid circle of the same color. C, Generation of onset week, using the 3% threshold method. The weekly percentage of tests positive for RSV is shown for each surveillance year during July–June. The horizontal line represents the 3% threshold. The onset week for each season is a solid circle of the same color. All data are limited to National Respiratory and Enteric Virus Surveillance System (NREVSS) laboratories that consistently reported data, defined here as those that reported data for an average of ≥10 RSV tests per week annually for at least 30 weeks of NREVSS-defined year. Surveillance started at week 27 for all years except 2009–2010, during which it started at week 26. Calendar years ended at week 52 for all years except 2014, which ended at week 53; there is therefore a gap in some of the data curves.

Figure 5

Comparison of respiratory syncytial virus (RSV) seasons, generated by 4 different methods. A, National estimates for season onset and offset generated by the 4 methods overlaid on the weekly raw (unadjusted) RSV detections. Each graph represents a different surveillance year, as labeled. Weekly detection of RSV by both polymerase chain reaction (PCR; black data curve) and antigen-detection (yellow data curve) tests are depicted. Season onset and offset are shown for (1) the 10% threshold method (ie, the traditional antigen-based method; vertical yellow lines), (2) the 3% threshold method (vertical green lines), (3) the 10-fold baseline (10FB) method (vertical blue lines), and (4) the retrospective slope 10 (RS10) method (vertical black lines). B, Comparison of season characteristics for the 4 methods during July 2009–June 2015. Depicted are season onset, peak (circles), and offset for 4 methods: (1) the 10% threshold method (yellow bar and circle), (2) the 3% threshold method (green bar and circle), (3) the 10FB method (blue bar and circle), and (4) the RS10 method (black bar and circle). All data are limited to National Respiratory and Enteric Virus Surveillance System (NREVSS) laboratories that consistently reported data, defined here as those that reported data for an average of ≥10 RSV tests per week annually for at least 30 weeks of NREVSS-defined year. Surveillance started at week 27 for all years except 2009–2010, during which it started at week 26. Calendar years ended at week 52 for all years except 2014, which ended at week 53.