Improving Accuracy of Influenza-Associated Hospitalization Rate Estimates

Detection of Novel Influenza Viruses Through Community and Healthcare Testing: Implications for Surveillance Efforts in the United States

BACKGROUND

Novel influenza viruses pose a potential pandemic risk, and rapid detection of infections in humans is critical to characterizing the virus and facilitating the implementation of public health response measures.

METHODS

We use a probabilistic framework to estimate the likelihood that novel influenza virus cases would be detected through testing in different community and healthcare settings (urgent care, emergency department, hospital, and intensive care unit [ICU]) while at low frequencies in the United States. Parameters were informed by data on seasonal influenza virus activity and existing testing practices.

RESULTS

In a baseline scenario reflecting the presence of 100 novel virus infections with similar severity to seasonal influenza viruses, the median probability of detecting at least one infection per month was highest in urgent care settings (72%) and when community testing was conducted at random among the general population (77%). However, urgent care testing was over 15 times more efficient (estimated as the number of cases detected per 100,000 tests) due to the larger number of tests required for community testing. In scenarios that assumed increased clinical severity of novel virus infection, median detection probabilities increased across all healthcare settings, particularly in hospitals and ICUs (up to 100%) where testing also became more efficient.

CONCLUSIONS

Our results suggest that novel influenza virus circulation is likely to be detected through existing healthcare surveillance, with the most efficient testing setting impacted by the disease severity profile. These analyses can help inform future testing strategies to maximize the likelihood of novel influenza detection.

Epidemiology, Clinical Characteristics, and Outcomes of Influenza-Associated Hospitalizations in US Children Over 9 Seasons Following the 2009 H1N1 Pandemic

BACKGROUND

Recent population-based data are limited regarding influenza-associated hospitalizations in US children.

METHODS

We identified children <18 years hospitalized with laboratory-confirmed influenza during 2010-2019 seasons, through the Centers for Disease Control and Prevention's Influenza Hospitalization Surveillance Network. Adjusted hospitalization and in-hospital mortality rates were calculated, and multivariable logistic regression was conducted to evaluate risk factors for pneumonia, intensive care unit (ICU) admission, mechanical ventilation, and death.

RESULTS

Over 9 seasons, adjusted influenza-associated hospitalization incidence rates ranged from 10 to 375 per 100 000 persons each season and were highest among infants <6 months old. Rates decreased with increasing age. The highest in-hospital mortality rates were observed in children <6 months old (0.73 per 100 000 persons). Over time, antiviral treatment significantly increased, from 56% to 85% (P < .001), and influenza vaccination rates increased from 33% to 44% (P = .003). Among the 13 235 hospitalized children, 2676 (20%) were admitted to the ICU, 2262 (17%) had pneumonia, 690 (5%) required mechanical ventilation, and 72 (0.5%) died during hospitalization. Compared with those <6 months of age, hospitalized children ≥13 years old had higher odds of pneumonia (adjusted odds ratio, 2.7 [95% confidence interval, 2.1-3.4], ICU admission (1.6 [1.3-1.9]), mechanical ventilation (1.6 [1.1-2.2]), and death (3.3 [1.2-9.3]).

CONCLUSIONS

Hospitalization and death rates were greatest in younger children at the population level. Among hospitalized children, however, older children had a higher risk of severe outcomes. Continued efforts to prevent and attenuate influenza in children are needed.

Incidence of respiratory virus illness and hospitalizations in a Panama and El Salvador birth cohort, 2014-2018

Background

Respiratory viruses remain a key cause of early childhood illness, hospitalization, and death globally.The recent pandemic has rekindled interest in the control of respiratory viruses among paediatric populations. We estimate the burden of such viruses among children <2 years.

Methods

Enrolled neonates were followed until two years of age. Weekly active symptom monitoring for the development of acute respiratory illnesses (ARI) defined as cough, rhinorrhoea, difficulty breathing, asthenia, anorexia, irritability, or vomiting was conducted. When the child had ARI and fever, nasopharyngeal swabbing was performed, and samples were tested through singleplex RT-PCR. Incidence of respiratory viruses was calculated by dividing the number of laboratory-confirmed detections by the person-time accrued during weeks when that virus was detectable through national surveillance then corrected for under-ascertainment among untested children.

Findings

During December 2014-November 2017, 1567 enrolled neonates contributed 2,186.9 person-years (py). Six in ten (64·4%) children developed ARI (total 2493 episodes). Among children <2 years, incidence of respiratory syncytial virus (RSV)-associated ARI episodes (21·0, 95%CI 19·3-22·8, per 100py) and rhinovirus-associated (20·5, 95%CI 20·4-20·7) were similar and higher than parainfluenza 1-3-associated (14·2, 95%CI 12·2-16·1), human metapneumovirus-associated (9·2, 95%CI 7·7-10·8), influenza-associated (5·9, 95%CI 4·4-7·5), and adenovirus-associated ARI episodes (5·1, 95%CI 5·0-5·2). Children aged <3 months had the highest rates of RSV ARI (49·1, 95%CI 44·0-54·1 per 100py) followed by children aged 3-5 (25·1, 95%CI 20·1-30·0), 6-11 (17·6, 95%CI 13·2-21·9), and 12-23 months (11·9, 95%CI 10·8-12·9). One in ten children with RSV was referred to the hospital (2·5, 95%CI 2·1-2·8, per 100py).

Interpretation

Children frequently developed viral ARI and a substantive proportion required hospital care. Such findings suggest the importance of exploring the value of new interventions and increasing uptake of existing prevention measures to mitigate burden of epidemic-prone respiratory viruses.

Funding

The study was supported by the Centers for Disease Control and Prevention.

Distinguishing viruses responsible for influenza-like illness

The many respiratory viruses that cause influenza-like illness (ILI) are reported and tracked as one entity, defined by the CDC as a group of symptoms that include a fever of 100 degrees Fahrenheit, a cough, and/or a sore throat. In the United States alone, ILI impacts 9-49 million people every year. While tracking ILI as a single clinical syndrome is informative in many respects, the underlying viruses differ in parameters and outbreak properties. Most existing models treat either a single respiratory virus or ILI as a whole. However, there is a need for models capable of comparing several individual viruses that cause respiratory illness, including ILI. To address this need, here we present a flexible model and simulations of epidemics for influenza, RSV, rhinovirus, seasonal coronavirus, adenovirus, and SARS/MERS, parameterized by a systematic literature review and accompanied by a global sensitivity analysis. We find that for these biological causes of ILI, their parameter values, timing, prevalence, and proportional contributions differ substantially. These results demonstrate that distinguishing the viruses that cause ILI will be an important aspect of future work on diagnostics, mitigation, modeling, and preparation for future pandemics.

Mortality Associated With Influenza and Respiratory Syncytial Virus in the US, 1999-2018

IMPORTANCE

Respiratory syncytial virus (RSV) mortality estimates have not been updated since 2009, and no study has assessed changes in influenza mortality after the 2009 pandemic. Updated burden estimates are needed to characterize long-term changes in the epidemiology of these viruses.

OBJECTIVE

To evaluate excess mortality from RSV and influenza in the US from 1999 to 2018.

DESIGN, SETTING, AND PARTICIPANTS

This cross-sectional study used data from 50.3 million US death certificates from 1999 to 2018 to create age-specific linear regression models and assess weekly mortality fluctuations above a seasonal baseline associated with RSV and influenza. Statistical analysis was performed for 1043 weeks from January 3, 1999, to December 29, 2018.

MAIN OUTCOMES AND MEASURES

Excess mortality associated with RSV and influenza estimated from the difference between observed and expected underlying respiratory mortality each season.

RESULTS

There were 50.3 million death certificates (50.1% women and 49.9% men; mean [SD] age at death, 72.7 [18.6] years) included in this analysis, 1.0% for children younger than 1 year and 73.4% for adults aged 65 years or older. A mean of 6549 (95% CI, 6140-6958) underlying respiratory deaths were associated with RSV annually, including 96 (95% CI, 92-99) deaths among children younger than 1 year. For influenza, there were 10 171 (95% CI, 9652-10 691) underlying respiratory deaths per year, with 23 deaths (95% CI, 19-27) among children younger than 1 year. The highest mean mortality rate per 100 000 population for both viruses was among adults aged 65 years or older at 14.7 (95% CI, 13.8-15.5) for RSV and 20.5 (95% CI, 19.4-21.5) for influenza. A lower proportion of influenza deaths occurred among those aged 65 years or older compared with earlier estimates (75.1% [95% CI, 67.4%-82.8%]). Influenza mortality was highest among those aged 65 years or older in seasons when A/H3N2 predominated (18 739 [95% CI, 16 616-21 336] deaths in 2017-2018) and among those aged 5 to 49 years when A/H1N1pdm2009 predominated (1683 [95% CI, 1583-1787] deaths in 2013-2014). Results were sensitive to the choice of mortality outcome and method, with the broadest outcome associated with annual means of 23 352 (95% CI, 21 814-24 891) excess deaths for RSV and 27 171 (95% CI, 25 142-29 199) for influenza.

CONCLUSIONS AND RELEVANCE

This study suggests that RSV poses a greater risk than influenza to infants, while both are associated with substantial mortality among elderly individuals. Influenza has large interannual variability, affecting different age groups depending on the circulating virus. The emergence of the influenza A/H1N1pdm2009 pandemic virus in 2009 shifted mortality toward middle-aged adults, a trend still observed to date. This study's estimates provide a benchmark to evaluate the mortality benefits associated with interventions against respiratory viruses, including new or improved immunization strategies.

Variability in published rates of influenza-associated hospitalizations: A systematic review, 2007-2018

Background

Influenza burden estimates help provide evidence to support influenza prevention and control programs at local and international levels.

Methods

Through a systematic review, we aimed to identify all published articles estimating rates of influenza-associated hospitalizations, describe methods and data sources used, and identify regions of the world where estimates are still lacking. We evaluated study heterogeneity to determine if we could pool published rates to generate global estimates of influenza-associated hospitalization.

Results

We identified 98 published articles estimating influenza-associated hospitalization rates from 2007-2018. Most articles (65%) identified were from high-income countries, with 34 of those (53%) presenting estimates from the United States. While we identified fewer publications (18%) from low- and lower-middle-income countries, 50% of those were published from 2015-2018, suggesting an increase in publications from lower-income countries in recent years. Eighty percent (n = 78) used a multiplier approach. Regression modelling techniques were only used with data from upper-middle or high-income countries where hospital administrative data was available. We identified variability in the methods, case definitions, and data sources used, including 91 different age groups and 11 different categories of case definitions. Due to the high observed heterogeneity across articles (I²>99%), we were unable to pool published estimates.

Conclusions

The variety of methods, data sources, and case definitions adapted locally suggests that the current literature cannot be synthesized to generate global estimates of influenza-associated hospitalization burden.

Quantifying the annual incidence and underestimation of seasonal influenza: A modelling approach

BACKGROUND

Seasonal influenza poses a significant public health and economic burden, associated with the outcome of infection and resulting complications. The true burden of the disease is difficult to capture due to the wide range of presentation, from asymptomatic cases to non-respiratory complications such as cardiovascular events, and its seasonal variability. An understanding of the magnitude of the true annual incidence of influenza is important to support prevention and control policy development and to evaluate the impact of preventative measures such as vaccination.

METHODS

We use a dynamic disease transmission model, laboratory-confirmed influenza surveillance data, and randomized-controlled trial (RCT) data to quantify the underestimation factor, expansion factor, and symptomatic influenza illnesses in the US and Canada during the 2011-2012 and 2012-2013 influenza seasons.

RESULTS

Based on 2 case definitions, we estimate between 0.42-3.2% and 0.33-1.2% of symptomatic influenza illnesses were laboratory-confirmed in Canada during the 2011-2012 and 2012-2013 seasons, respectively. In the US, we estimate between 0.08-0.61% and 0.07-0.33% of symptomatic influenza illnesses were laboratory-confirmed in the 2011-2012 and 2012-2013 seasons, respectively. We estimated the symptomatic influenza illnesses in Canada to be 0.32-2.4 million in 2011-2012 and 1.8-8.2 million in 2012-2013. In the US, we estimate the number of symptomatic influenza illnesses to be 4.4-34 million in 2011-2012 and 23-102 million in 2012-2013.

CONCLUSIONS

We illustrate that monitoring a representative group within a population may aid in effectively modelling the transmission of infectious diseases such as influenza. In particular, the utilization of RCTs in models may enhance the accuracy of epidemiological parameter estimation.

Very high sensitivity of a rapid influenza diagnostic test in adults and elderly individuals within 48 hours of the onset of illness

During influenza epidemics, Japanese clinicians routinely perform rapid influenza diagnostic tests (RIDTs) in the examination of patients who have an influenza-like illness, and patients with positive test results, including otherwise healthy individuals, are treated with anti-influenza drugs. However, it was recently reported that the sensitivity of RIDTs was extremely low in adult patients. We examined the sensitivity and specificity of an RIDT that is widely used in Japan, ImunoAce Flu (TAUNS, Shizuoka, Japan), in comparison to reverse transcriptase polymerase chain reaction (RT-PCR). The sensitivity and specificity of the ImunoAce Flu test were 97.1% (95%CI: 93.8–98.9) and 89.2% (95%CI: 84.1–93.1), respectively. The ImunoAce Flu test is designed to not only detect influenza A or B, but also to detect H1N1pdm09 with the use of an additional test kit (Linjudge FluA/pdm). Its sensitivity and specificity for A/H1N1pdm09 were 97.6% (95%CI: 87.4–99.9) and 92.6% (95%CI: 82.1–97.9), respectively. Thus, by consecutively testing patients with the ImunoAce Flu test followed by the Linjudge FluA/pdm test, we are able to diagnose whether a patient has A/H1N1pdm09 or A/H3N2 infection within a short time. The reliability of rapid test results seems to be much higher in Japan than in other countries, because approximately 90% of influenza patients are tested and treated within 48 hours after the onset of illness, when the influenza viral load in the upper respiratory tract is high. From the Japanese experience, RIDTs are sufficiently sensitive and highly useful, if patients are tested within 48 hours after the onset of illness.

Determining the Cause of Death Among Children Hospitalized With Respiratory Illness in Kenya: Protocol for Pediatric Respiratory Etiology Surveillance Study (PRESS)

BACKGROUND

In sub-Saharan Africa, where the burden of respiratory disease-related deaths is the highest, information on the cause of death remains inadequate because of poor access to health care and limited availability of diagnostic tools. Postmortem examination can aid in the ascertainment of causes of death. This manuscript describes the study protocol for the Pediatric Respiratory Etiology Surveillance Study (PRESS).

OBJECTIVE

This study protocol aims to identify causes and etiologies associated with respiratory disease-related deaths among children (age 1-59 months) with respiratory illness admitted to the Kenyatta National Hospital (KNH), the largest public hospital in Kenya, through postmortem examination coupled with innovative approaches to laboratory investigation.

METHODS

We prospectively followed children hospitalized with respiratory illness until the end of clinical care or death. In case of death, parents or guardians were offered grief counseling, and postmortem examination was offered. Lung tissue specimens were collected using minimally invasive tissue sampling and conventional autopsy where other tissues were collected. Tissues were tested using histopathology, immunohistochemistry, and multipathogen molecular-based assays to identify pathogens. For each case, clinical and laboratory data were reviewed by a team of pathologists, clinicians, laboratorians, and epidemiologists to assign a cause of and etiology associated with death.

RESULTS

We have enrolled pediatric cases of respiratory illness hospitalized at the KNH at the time of this submission; of those, 14.8% (140/945) died while in the hospital. Both analysis and interpretation of laboratory results and writing up of findings are expected in 2019-2020.

CONCLUSIONS

Postmortem studies can help identify major pathogens contributing to respiratory-associated deaths in children. This information is needed to develop evidence-based prevention and treatment policies that target important causes of pediatric respiratory mortality and assist with the prioritization of local resources. Furthermore, PRESS can provide insights into the interpretation of results using multipathogen testing platforms in resource-limited settings.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/10854.

Mapping of the US Domestic Influenza Virologic Surveillance Landscape

Influenza virologic surveillance is critical each season for tracking influenza circulation, following trends in antiviral drug resistance, detecting novel influenza infections in humans, and selecting viruses for use in annual seasonal vaccine production. We developed a framework and process map for characterizing the landscape of US influenza virologic surveillance into 5 tiers of influenza testing: outpatient settings (tier 1), inpatient settings and commercial laboratories (tier 2), state public health laboratories (tier 3), National Influenza Reference Center laboratories (tier 4), and Centers for Disease Control and Prevention laboratories (tier 5). During the 2015–16 season, the numbers of influenza tests directly contributing to virologic surveillance were 804,000 in tiers 1 and 2; 78,000 in tier 3; 2,800 in tier 4; and 3,400 in tier 5. With the release of the 2017 US Pandemic Influenza Plan, the proposed framework will support public health officials in modeling, surveillance, and pandemic planning and response.

Cost-effectiveness analysis of different seasonal influenza vaccines in the elderly Italian population

In the perspective of reaching at least 75% influenza vaccination coverage in the elderly and substantial budget constraints, Italian decision makers are facing important challenges in determining an optimal immunization strategy for this growing and particularly vulnerable population. Four different influenza vaccines are currently available for Italian older adults aged 65 years or above, namely trivalent inactivated vaccines (TIVs), MF59-adjuvanted TIV (MF59-TIV), intradermal TIV (ID-TIV) and quadrivalent inactivated vaccines (QIVs). The present study is the first to compare the cost-effectiveness profiles of virtually all possible public health strategies, including the aforementioned four vaccine formulations as well non-vaccination. For this purpose, a decision tree model was built ex novo; the analysis was conducted from the third-payer perspective in the timeframe of one year. All available vaccines were cost-effective compared with non-vaccination. However, MF59-TIV had the most favorable economic profile in the Italian elderly population. Indeed, compared with non-vaccination, it was deemed highly cost-effective with an incremental cost-effectiveness ratio (ICER) of €10,750 per quality-adjusted life year (QALY). The ICER was much lower (€4,527/QALY) when MF59-TIV was directly compared with TIV. ID-TIV and QIV were dominated by MF59-TIV as the former comparators were associated with greater total costs and lower health benefits. Both deterministic and probabilistic sensitivity analyses confirmed robustness of the base case results. From the economic perspective, MF59-TIV should be considered as a preferential choice for Italian older adults aged 65 years or above.

Age- and risk-related appropriateness of the use of available influenza vaccines in the Italian elderly population is advantageous: results from a budget impact analysis

INTRODUCTION

Nowadays, four different types of influenza vaccines are available in Italy: trivalent (TIV), quadrivalent (QIV), MF59-adjuvanted (aTIV) and intradermal TIV (idTIV) inactivated vaccines. Recently, a concept of the appropriateness (i.e. according to the age and risk factors) of the use of different vaccines has been established in Italy. We conducted a budget impact analysis of switching to a policy, in which the Italian elderly (who carry the major disease burden) received the available vaccines according to their age and risk profile.

METHODS

A novel budget impact model was constructed with a time horizon of one influenza season. In the reference scenario the cohort of Italian elderly individuals could receive either available vaccine according to 2017/18 season market share. The alternative scenario envisaged the administration of TIV/QIV to people aged 65-74 years and at low risk of developing influenza-related complications, while aTIV/idTIV were allocated to high-risk 65-74-year-olds and all subjects aged ≥ 75 years.

RESULTS

Switching to the alternative scenario would result in both significant health benefits and net budget savings. Particularly, it would be possible to prevent an additional 8201 cases of laboratory-confirmed influenza, 988 complications, 355 hospitalizations and 14 deaths. Despite the alternative strategy being associated with slightly higher vaccination costs, the total savings derived from fewer influenza events completely resets this increase with net budget savings of € 0.13 million.

CONCLUSIONS

An immunization policy in which influenza vaccines are administered according to the age and risk profile of Italian elderly individuals is advisable.

Evaluation of the point-of-care Becton Dickinson Veritor™ Rapid influenza diagnostic test in Kenya, 2013-2014

BACKGROUND

We evaluated the performance of the Becton Dickinson Veritor™ System Flu A + B rapid influenza diagnostic test (RIDT) to detect influenza viruses in respiratory specimens from patients enrolled at five surveillance sites in Kenya, a tropical country where influenza seasonality is variable.

METHODS

Nasal swab (NS) and nasopharyngeal (NP)/oropharyngeal (OP) swabs were collected from patients with influenza like illness and/or severe acute respiratory infection. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the RIDT using NS specimens were evaluated against nasal swabs tested by real time reverse transcription polymerase chain reaction (rRT-PCR). The performance parameter results were expressed as 95% confidence intervals (CI) calculated using binomial exact methods, with P < 0.05 considered significant. Two-sample Z tests were used to test for differences in sample proportions. Analysis was performed using SAS software version 9.3.

RESULTS

From July 2013 to July 2014, 3,569 patients were recruited, of which 78.7% were aged <5 years. Overall, 14.4% of NS specimens were influenza-positive by RIDT. RIDT overall sensitivity was 77.1% (95% CI 72.8-81.0%) and specificity was 94.9% (95% CI 94.0-95.7%) compared to rRT-PCR using NS specimens. RIDT sensitivity for influenza A virus compared to rRT-PCR using NS specimens was 71.8% (95% CI 66.7-76.4%) and was significantly higher than for influenza B which was 43.8% (95% CI 33.8-54.2%). PPV ranged from 30%-80% depending on background prevalence of influenza.

CONCLUSION

Although the variable seasonality of influenza in tropical Africa presents unique challenges, RIDTs may have a role in making influenza surveillance sustainable in more remote areas of Africa, where laboratory capacity is limited.

Effect of Culture-Independent Diagnostic Tests on Future Emerging Infections Program Surveillance

The Centers for Disease Control and Prevention Emerging Infections Program (EIP) network conducts population-based surveillance for pathogens of public health importance. Central to obtaining estimates of disease burden and tracking microbiological characteristics of these infections is accurate laboratory detection of pathogens. The use of culture-independent diagnostic tests (CIDTs) in clinical settings presents both opportunities and challenges to EIP surveillance. Because CIDTs offer better sensitivity than culture and are relatively easy to perform, their use could potentially improve estimates of disease burden. However, changes in clinical testing practices, use of tests with different sensitivities and specificities, and changes to case definitions make it challenging to monitor trends. Isolates are still needed for performing strain typing, antimicrobial resistance testing, and identifying other molecular characteristics of organisms. In this article, we outline current and future EIP activities to address issues associated with adoption of CIDTs, which may apply to other public health surveillance.