Andrea Steffens

Incidence of medically attended influenza during pandemic and post-pandemic seasons through the Influenza Incidence Surveillance Project, 2009–13

BACKGROUND Since the introduction of pandemic influenza A (H1N1) to the USA in 2009, the Influenza Incidence Surveillance Project has monitored the burden of influenza in the outpatient setting through population-based surveillance. METHODS From Oct 1, 2009, to July 31, 2013, outpatient clinics representing 13 health jurisdictions in the USA reported counts of influenza-like illness (fever including cough or sore throat) and all patient visits by age. During four years, staff at 104 unique clinics (range 35-64 per year) with a combined median population of 368,559 (IQR 352,595-428,286) attended 35,663 patients with influenza-like illness and collected 13,925 respiratory specimens. Clinical data and a respiratory specimen for influenza testing by RT-PCR were collected from the first ten patients presenting with influenza-like illness each week. We calculated the incidence of visits for influenza-like illness using the size of the patient population, and the incidence attributable to influenza was extrapolated from the proportion of patients with positive tests each week. FINDINGS The site-median peak percentage of specimens positive for influenza ranged from 58.3% to 77.8%. Children aged 2 to 17 years had the highest incidence of influenza-associated visits (range 4.2-28.0 per 1000 people by year), and adults older than 65 years had the lowest (range 0.5-3.5 per 1000 population). Influenza A H3N2, pandemic H1N1, and influenza B equally co-circulated in the first post-pandemic season, whereas H3N2 predominated for the next two seasons. Of patients for whom data was available, influenza vaccination was reported in 3289 (28.7%) of 11,459 patients with influenza-like illness, and antivirals were prescribed to 1644 (13.8%) of 11,953 patients. INTERPRETATION Influenza incidence varied with age groups and by season after the pandemic of 2009 influenza A H1N1. High levels of influenza virus circulation, especially in young children, emphasise the need for additional efforts to increase the uptake of influenza vaccines and antivirals. FUNDING US Centers for Disease Control and Prevention.

Assessment of influenza vaccine effectiveness in a sentinel surveillance network 2010-13, United States.

BACKGROUND Influenza vaccines are now widely used to reduce the burden of annual epidemics of influenza virus infections. Influenza vaccine effectiveness (VE) is monitored annually to determine VE against each seasons circulating influenza strains in different groups such as children, adults and the elderly. Few prospective surveillance programs are available to evaluate influenza VE against medically attended illness for patients of all ages in the United States. METHODS We conducted surveillance of patients with acute respiratory illnesses in 101 clinics across the US during three consecutive influenza seasons. We analyzed laboratory testing results for influenza virus, self-reported vaccine history, and patient characteristics, defining cases as patients who tested positive for influenza virus and controls as patients who tested negative for influenza virus. Comparison of influenza vaccination coverage among cases versus controls, adjusted for potential confounders, was used to estimate VE as one minus the adjusted odds ratio multiplied by 100%. RESULTS We included 10,650 patients during three influenza seasons from August 2010 through December 2013, and estimated influenza VE in children 6m-5y of age (58%; 95% CI: 49%-66%), children 6-17y (45%; 95% CI: 34%-53%), adults 18-49y (36%; 95% CI: 24%, 46%), and adults ≥50y (34%, 95% CI: 13%, 51%). VE was higher against influenza A(H1N1) compared to A(H3N2) and B. CONCLUSIONS Our estimates of moderate influenza VE confirm the important role of vaccination in protecting against medically attended influenza virus infection.

Assessment of virus interference in a test-negative study of influenza vaccine effectiveness

Background: The observational test-negative study design is used to estimate vaccine effectiveness against influenza virus infection. An important assumption of the test-negative design is that vaccination does not affect the risk of infection with another virus. If such virus interference occurred, detection of other respiratory viruses would be more common among influenza vaccine recipients and vaccine effectiveness estimates could differ. We evaluated the potential for virus interference using data from the Influenza Incidence Surveillance Project. Methods: From 2010 to 2013, outpatients presenting to clinics in 13 US jurisdictions with acute respiratory infections were tested for influenza and other respiratory viruses. We investigated whether virus interference might affect vaccine effectiveness estimates by first evaluating the sensitivity of estimates using alternative control groups that include or exclude patients with other respiratory virus detections by age group and early/middle/late stage of influenza seasons. Second, we evaluated the association between influenza vaccination receipt and other respiratory virus detection among influenza test-negative patients. Results: Influenza was detected in 3,743/10,650 patients (35%), and overall vaccine effectiveness was 47% (95% CI: 42%, 52%). Estimates using each control group were consistent overall or when stratified by age groups, and there were no differences among early, middle, or late phase during influenza season. We found no associations between detection of other respiratory viruses and receipt of influenza vaccination. Conclusions: In this 3-year test-negative design study in an outpatient setting in the United States, we found no evidence of virus interference or impact on influenza vaccine effectiveness estimation.

Population-based Surveillance for Medically Attended Human Parainfluenza Viruses From the Influenza Incidence Surveillance Project, 2010-2014.

Background: Parainfluenza viruses (PIV) have been shown to contribute substantially to pediatric hospitalizations in the United States. However, to date, there has been no systematic surveillance to estimate the burden among pediatric outpatients. Methods: From August 2010 through July 2014, outpatient health care providers with enumerated patient populations in 13 states and jurisdictions participating in the Influenza Incidence Surveillance Project conducted surveillance of patients with influenza-like illness (ILI). Respiratory specimens were collected from the first 10 ILI patients each week with demographic and clinical data. Specimens were tested for multiple respiratory viruses, including PIV1–4, using reverse transcriptase–polymerase chain reaction assays. Cumulative incidence was calculated using provider patient population size as the denominator. Results: PIVs 1–3 were detected in 8.0% of 7716 ILI-related outpatient specimens: 30% were PIV1, 26% PIV2 and 44% PIV3. PIV circulation varied noticeably by year and type, with PIV3 predominating in 2010–2011 (incidence 110 per 100,000 children), PIV1 in 2011–2012 (89 per 100,000), dual predominance of PIV2 and PIV3 (88 and 131 per 100,000) in 2012–2013 and PIV3 (100 per 100,000) in 2013–2014. The highest incidence of PIV detections was among patients aged <5 years (259–1307 per 100,000). The median age at detection for PIV3 (3.4 years) was significantly lower than the median ages for PIV1 (4.5 years) and PIV2 (7.0 years; P < 0.05). Conclusions: PIVs 1–3 comprise a substantial amount of medically attended pediatric ILI, particularly among children aged <5 years. Distinct seasonal circulation patterns as well as significant differences in rates by age were observed between PIV types.

Comparison of respiratory pathogen detections from routine hospital testing and expanded systematic testing from the Minnesota Severe Acute Respiratory Illness surveillance program, 2015-2016

Abstract Background Hospital testing for respiratory pathogens is nonsystematic, leading to potential missed detection of clinically relevant pathogens. The Minnesota Severe Acute Respiratory Illness (SARI) surveillance program monitors hospitalizations due to acute respiratory illness and conducts systematic testing for several respiratory pathogens. We assessed viruses detected by the hospital and additional detections identified by expanded testing. Methods Residual upper respiratory specimens collected from patients hospitalized for suspected respiratory illness for routine diagnostic testing at three hospitals, including one children’s hospital, were submitted to the Minnesota Department of Health (MDH). Specimens were tested for 18 respiratory viruses by RT-PCR. Clinical and hospital test data were collected through medical record review. Results From September 2015 to August 2016, 2,351 hospitalized SARI patients were reported, with the following age distribution: 57% <5 years, 13% 5–17 years, 30% ≥18 years. Among all SARI patients, 97% (2,273) had hospital-based, clinician-directed testing for viral pathogens. Viruses were detected among 47% (1,077) of tested patients, among which testing methods included PCR (85%), rapid antigen (13%), and culture (2%); 74% were tested on the day of admission. Most common viruses detected by clinical testing included respiratory syncytial virus (41%), rhinovirus/enterovirus (31%), and influenza (15%) (Figure 1). Systematic RT–PCR testing at MDH identified 1,600 (68%) patients positive for ≥1 respiratory virus, identifying previously unknown detections among 35% (820) of SARI patients (Figure 2). Of 1,272 patients with no virus identified at the hospital, 46% (586) had a viral detection at MDH. Patients aged <18 years were significantly more likely to have an additional pathogen detected by MDH testing than those aged ≥18 years (P < 0.01), including rhinovirus/enterovirus, adenovirus, human metapneumovirus, and coronaviruses. Conclusion Systematic, expanded testing at MDH identified a higher proportion of respiratory pathogens among SARI patients compared with clinical laboratory testing. Additional testing for clinically relevant respiratory pathogens may inform medical decision-making.Figure 1.Figure 2. Disclosures All authors: No reported disclosures.

Partial Protection of Influenza Vaccine in a Primary Care Population—Wisconsin: 2012–2015

1Department of Family Medicine, University of Wisconsin-Madison, 2Wisconsin Division of Public Health, 3Wisconsin State Laboratory of Hygiene 4U.S. Centers for Disease Control and Prevention. The findings and conclusions in the this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention Jonathan Temte, MD/PhD jon.temte@fammed.wisc.edu 1100 Delaplaine Court Madison, WI 53715 Phone: 608-263-3111 Fax: 608-263-6663