Modeling influenza seasonality in the tropics and subtropics

Discovering the climate dependent disease transmission mechanism through learning-explaining framework

There are evidence showing that meteorological factors, such as temperature and humidity, have critical effects on transmission of some infectious diseases, while quantifying the influence is challenging. In this study we develop a learning-explaining framework to discover the particular dependence of transmission mechanisms on meteorological factors based on multiple source data. The incidence rate based on the epidemic data and epidemic model is theoretically identified, and meanwhile the practical discovery of particular formula is feasible through deep neural networks (DNN), symbolic regression (SR) and sparse identification of nonlinear dynamics (SINDy). In particular, we initially learn the incidence rate in an SIRS model based on epidemic data, then use mechanism discovery methods to explore the possible explicit forms of the incidence rate, and consequently explore the possible relationship between transmission rate and meteorological factors. We finally use information criteria and a definition of evaluation score to make model selection, and hence suggest the optimal explicit formula. We illustrate the idea by derive the incidence rate and transmission rate of respiratory infectious diseases based on the case data on influenza-like illness (ILI) in Xi'an, Shaanxi Province of China and meteorological data from 1st January 2010 to 10th November 2016. The finding reveals that the influence of meteorological factors on transmission exhibits very strong nonlinearity, and modeling the effect should be of great care.

Forecasting influenza epidemics in China using transmission dynamic model with absolute humidity

Background

An influenza forecasting system is critical to influenza epidemic preparedness. Low temperature has long been recognized as a condition favoring influenza epidemic, yet it fails to justify the summer influenza peak in tropics/subtropics. Recent studies have suggested that absolute humidity (AH) had a U-shape relationship with influenza survival and transmission across climate zones, indicating that a unified influenza forecasting system could be established for China with various climate conditions.

Methods

Our study has generated weekly influenza forecasts by season and type/subtype in northern and southern China from 2011 to 2021, using a forecasting system combining an AH-driven susceptible-infected-recovered-susceptible (SIRS) model and the ensemble adjustment Kalman filter (EAKF). Model performance was assessed by sensitivity and specificity in predicting epidemics, and by accuracies in predicting peak timing and magnitude.

Results

Our forecast system can generally well predict seasonal influenza epidemics (mean sensitivity>87.5%; mean specificity >80%). The average forecast accuracies were 82% and 60% for peak timing and magnitude at 3-6 weeks ahead for northern China, higher than those of 42% and 20% for southern China. The accuracy was generally better when the forecast was made closer to the actual peak time.

Discussion

The established AH-driven forecasting system can generally well predict the occurrence of seasonal influenza epidemics in China.

The effect of particulate matter 2.5 on seasonal influenza transmission in 1,330 counties, China: A Bayesian spatial analysis based on Köppen Geiger climate zones classifications

Previous research has linked seasonal influenza transmission with particulate matters (PM2.5). However, the effect of PM2.5 on seasonal influenza transmission varied by region. This study aims explore how PM2.5 influenced seasonal influenza transmission in the elderly across 1330 counties in two Köppen Geiger climate zones in China, incorporating the socio-economic factors to enhance climate-driven early warning systems (EWS) for influenza. Data included weekly 2015-2019 influenza cases in those aged >65 from China's national influenza surveillance system for 1330 counties in two Köppen Geiger climate zones: Temperate, Hot Summer with Dry Winter (Cwa) and No Dry Season (Cfa). PM2.5 data from 2015 to 2019 were sourced from Copernicus Atmosphere Monitoring Services. Additional data on floating population, population density and Gross Domestic Product (GDP) per capita were collected from pertinent departments. A Bayesian spatial autoregressive model assessed the association of PM2.5 and influenza transmission after adjustment of socio-economic factors. Our research results showed PM2.5 (per 1 μg/m³ increase) was linked to increased influenza transmission in the Cwa zone during winter season (Relative Risk (RR) = 1.023, 95% Credible Interval (CI):1.008-1.040) but not in the Cfa winter (RR = 1.003, 95% CI: 0.992-1.015). Floating population significantly enhanced transmission in both zones (highest RR = 1.362, 95% CI:1.181-1.583), while GDP per capita growth was associated with reduced transmission risk (highest RR = 0.619, 95% CI: 0.445-0.861). The study identifies PM2.5 as a significant factor influencing influenza transmission in the elderly, with effects varying by climate zone, suggesting the need to incorporate PM2.5 and socio-economic factors into seasonal influenza EWS.

Integrating dynamical modeling and phylogeographic inference to characterize global influenza circulation

Global seasonal influenza circulation involves a complex interplay between local (seasonality, demography, host immunity) and global factors (international mobility) shaping recurrent epidemic patterns. No studies so far have reconciled the two spatial levels, evaluating the coupling between national epidemics, considering heterogeneous coverage of epidemiological, and virological data, integrating different data sources. We propose a novel-combined approach based on a dynamical model of global influenza spread (GLEAM), integrating high-resolution demographic, and mobility data, and a generalized linear model of phylogeographic diffusion that accounts for time-varying migration rates. Seasonal migration fluxes across countries simulated with GLEAM are tested as phylogeographic predictors to provide model validation and calibration based on genetic data. Seasonal fluxes obtained with a specific transmissibility peak time and recurrent travel outperformed the raw air-transportation predictor, previously considered as optimal indicator of global influenza migration. Influenza A subtypes supported autumn-winter reproductive number as high as 2.25 and an average immunity duration of 2 years. Similar dynamics were preferred by influenza B lineages, with a lower autumn-winter reproductive number. Comparing simulated epidemic profiles against FluNet data offered comparatively limited resolution power. The multiscale approach enables model selection yielding a novel computational framework for describing global influenza dynamics at different scales-local transmission and national epidemics vs. international coupling through mobility and imported cases. Our findings have important implications to improve preparedness against seasonal influenza epidemics. The approach can be generalized to other epidemic contexts, such as emerging disease outbreaks to improve the flexibility and predictive power of modeling.

Improving influenza forecast in the tropics and subtropics: a case study of Hong Kong

Influenza forecasts could aid public health response as shown for temperate regions, but such efforts are more challenging in the tropics and subtropics due to more irregular influenza activities. Here, we built six forecast approaches for influenza in the (sub)tropics, with six model forms designed to model seasonal infection risk (i.e. seasonality) based on the dependence of virus survival on climate conditions and to flexibly account for immunity waning. We ran the models jointly with the ensemble adjustment Kalman filter to generate retrospective forecasts of influenza incidence in subtropical Hong Kong from January 1999 to December 2019 including the 2009 A(H1N1)pdm09 pandemic. In addition to short-term targets (one to four weeks ahead predictions), we also tested mid-range (one to three months) and long-range (four to six months) forecasts, which could be valuable for long-term planning. The largest improvement came from the inclusion of climate-modulated seasonality modelling, particularly for the mid- and long-range forecasts. The best-performing approach included a seasonal-trend-based climate modulation and assumed mixed immunity waning; the forecast accuracies, including peak week and intensity, were comparable to that reported for temperate regions including the USA. These findings demonstrate that incorporating mechanisms of climate modulation on influenza transmission can substantially improve forecast performance in the (sub)tropics.

Characterizing the seasonal influenza disease burden attributable to climate variability: A nationwide time-series modelling study in Japan, 2000-2019

BACKGROUND

Ambient temperature and humidity are established environmental stressors with regard to influenza infections; however, mapping disease burden is difficult owing to the complexities of the underlying associations and differences in vulnerable population distributions. In this study, we aimed to quantify the burden of influenza attributable to non-optimal ambient temperature and absolute humidity in Japan considering geographical differences in vulnerability.

METHODS

The exposure-lag-response relationships between influenza incidence, ambient temperature, and absolute humidity in all 47 Japanese prefectures for 2000-2019 were quantified using a distributed lag non-linear model for each prefecture; the estimates from all the prefectures were then pooled using a multivariate mixed-effects meta-regression model to derive nationwide average associations. Association between prefecture-specific indicators and the risk were also examined. Attributable risks were estimated for non-optimal ambient temperature and absolute humidity according to the exposure-lag-response relationships obtained before.

RESULTS

A total of 25,596,525 influenza cases were reported during the study period. Cold and dry conditions significantly increased influenza incidence risk. Compared with the minimum incidence weekly mean ambient temperature (29.8 °C) and the minimum incidence weekly mean absolute humidity (20.2 g/m3), the cumulative relative risks (RRs) of influenza in cold (2.5 °C) and dry (3.6 g/m3) conditions were 2.79 (95% confidence interval [CI]: 1.78-4.37) and 3.20 (95% CI: 2.37-4.31), respectively. The higher RRs for cold and dry conditions were associated with geographical and climatic indicators corresponding to the central and northern prefectures; demographic, socioeconomic, and health resources indicators showed no clear trends. Finally, 27.25% (95% empirical CI [eCI]: 5.54-36.35) and 32.35% (95% eCI: 22.39-37.87) of all cases were attributable to non-optimal ambient temperature and absolute humidity (6,976,300 [95% eCI: 1,420,068-9,306,128] and 8,280,981 [95% eCI: 8,280,981-9,693,532] cases), respectively.

CONCLUSIONS

These findings could help identify the most vulnerable populations in Japan and design adaptation policies to reduce the attributable burden of influenza due to climate variability.

Characterizing the interactions between influenza and respiratory syncytial viruses and their implications for epidemic control

Pathogen-pathogen interactions represent a critical but little-understood feature of infectious disease dynamics. In particular, experimental evidence suggests that influenza virus and respiratory syncytial virus (RSV) compete with each other, such that infection with one confers temporary protection against the other. However, such interactions are challenging to study using common epidemiologic methods. Here, we use a mathematical modeling approach, in conjunction with detailed surveillance data from Hong Kong and Canada, to infer the strength and duration of the interaction between influenza and RSV. Based on our estimates, we further utilize our model to evaluate the potential conflicting effects of live attenuated influenza vaccines (LAIV) on RSV burden. We find evidence of a moderate to strong, negative, bidirectional interaction, such that infection with either virus yields 40-100% protection against infection with the other for one to five months. Assuming that LAIV reduces RSV susceptibility in a similar manner, we predict that the impact of such a vaccine at the population level would likely depend greatly on underlying viral circulation patterns. More broadly, we highlight the utility of mathematical models as a tool to characterize pathogen-pathogen interactions.

Seasonal patterns of influenza incidence and the influence of meteorological and air pollution factors in Thailand during 2009-2019

Influenza, an acute respiratory illness, remains a significant public health challenge, contributing substantially to morbidity and mortality worldwide. Its seasonal prevalence exhibits diversity across regions with distinct climates. This study aimed to explore the seasonal patterns of influenza and their correlation with meteorological and air pollution factors across six regions of Thailand. We conducted an analysis of monthly average temperature, relative humidity, precipitation, PM10, NO2, O3 concentrations, and influenza incidence data from 2009 to 2019 using wavelet analysis. Our findings reveal inconsistent biannual influenza prevalence patterns throughout the study period. The biannual pattern emerged during 2010-2012 across all regions but disappeared during 2013-2016. However, post-2016, the biannual cycles resurfaced, with peaks occurring during the rainy and winter seasons in most regions, except for the southern region. Wavelet coherence reveals that relative humidity can be the main influencing factor for influenza incidence over a one-year period in the northern, northeastern, central, Bangkok-metropolitan, and eastern regions, not in the southern region during 2010-2012 and 2016-2018. Similarly, precipitation can drive the influenza incidence at the same period for the northeastern, central, Bangkok-metropolitan, and eastern regions. PM10 concentration can influence influenza incidence over a half-year period in the northeastern, central, Bangkok-metropolitan, and eastern regions of Thailand during certain years. These results enhance our understanding of the temporal dynamics of influenza seasonality influenced by weather conditions and air pollution over the past 11 years. Such knowledge is invaluable for resource allocation in clinical settings and informing public health strategies, particularly in navigating Thailand's climatic complexities.

Mathematical modeling for estimating influenza vaccine efficacy: A case study of the Valencian Community, Spain

Vaccine efficacy and its quantification is a crucial concept for the proper design of public health vaccination policies. In this work we proposed a mathematical model to estimate the efficacy of the influenza vaccine in a real-word scenario. In particular, our model is a SEIR-type epidemiological model, which distinguishes vaccinated and unvaccinated populations. Mathematically, its dynamics is governed by a nonlinear system of ordinary differential equations, where the non-linearity arises from the effective contacts between susceptible and infected individuals. Two key aspects of this study is that we use a vaccine distribution over time that is based on real data specific to the elderly people in the Valencian Community and the calibration process takes into account that over one influenza season a specific proportion of the population becomes infected with influenza. To consider the effectiveness of the vaccine, the model incorporates a parameter, the vaccine attenuation factor, which is related with the vaccine efficacy against the influenza virus. With this framework, in order to calibrate the model parameters and to obtain an influenza vaccine efficacy estimation, we considered the 2016-2017 influenza season in the Valencian Community, Spain, using the influenza reported cases of vaccinated and unvaccinated. In order to ensure the model identifiability, we choose to deterministically calibrate the parameters for different scenarios and we find the one with the minimum error in order to determine the vaccine efficacy. The calibration results suggest that the influenza vaccine developed for 2016-2017 influenza season has an efficacy of approximately 76.7%, and that the risk of becoming infected is five times higher for an unvaccinated individual in comparison with a vaccinated one. This estimation partially agrees with some previous studies related to the influenza vaccine. This study presents a new integrated mathematical approach to study the influenza vaccine efficacy and gives further insight into this important public health topic.

Increasing intensity of enterovirus outbreaks projected with climate change

Pathogens of the enterovirus genus, including poliovirus and coxsackieviruses, typically circulate in the summer months suggesting a possible positive association between warmer weather and transmission. Here we evaluate the environmental and demographic drivers of enterovirus transmission, as well as the implications of climate change for future enterovirus circulation. We leverage pre-vaccination era data on polio in the US as well as data on two enterovirus A serotypes in China and Japan that are known to cause hand, foot, and mouth disease. Using mechanistic modeling and statistical approaches, we find that enterovirus transmission appears positively correlated with temperature although demographic factors, particularly the timing of school semesters, remain important. We use temperature projections from Coupled Model Intercomparison Project Phase 6 (CMIP6) to simulate future outbreaks under late 21st-century climate change for Chinese provinces. We find that outbreak size increases with climate change on average, though results differ across climate models depending on the degree of wintertime warming. In the worst-case scenario, we project peak outbreaks in some locations could increase by up to 40%.

Integrating dynamical modeling and phylogeographic inference to characterize global influenza circulation

Global seasonal influenza circulation involves a complex interplay between local (seasonality, demography, host immunity) and global factors (international mobility) shaping recurrent epidemic patterns. No studies so far have reconciled the two spatial levels, evaluating the coupling between national epidemics, considering heterogeneous coverage of epidemiological and virological data, integrating different data sources. We propose a novel combined approach based on a dynamical model of global influenza spread (GLEAM), integrating high-resolution demographic and mobility data, and a generalized linear model of phylogeographic diffusion that accounts for time-varying migration rates. Seasonal migration fluxes across global macro-regions simulated with GLEAM are tested as phylogeographic predictors to provide model validation and calibration based on genetic data. Seasonal fluxes obtained with a specific transmissibility peak time and recurrent travel outperformed the raw air-transportation predictor, previously considered as optimal indicator of global influenza migration. Influenza A subtypes supported autumn-winter reproductive number as high as 2.25 and an average immunity duration of 2 years. Similar dynamics were preferred by influenza B lineages, with a lower autumn-winter reproductive number. Comparing simulated epidemic profiles against FluNet data offered comparatively limited resolution power. The multiscale approach enables model selection yielding a novel computational framework for describing global influenza dynamics at different scales - local transmission and national epidemics vs. international coupling through mobility and imported cases. Our findings have important implications to improve preparedness against seasonal influenza epidemics. The approach can be generalized to other epidemic contexts, such as emerging disease outbreaks to improve the flexibility and predictive power of modeling.

Intensity and drivers of subtypes interference between seasonal influenza viruses in mainland China: A modeling study

Subtype interference has a significant impact on the epidemiological patterns of seasonal influenza viruses (SIVs). We used attributable risk percent [the absolute value of the ratio of the effective reproduction number (Rₑ) of different subtypes minus one] to quantify interference intensity between A/H1N1 and A/H3N2, as well as B/Victoria and B/Yamagata. The interference intensity between A/H1N1 and A/H3N2 was higher in southern China 0.26 (IQR: 0.11-0.46) than in northern China 0.17 (IQR: 0.07-0.24). Similarly, interference intensity between B/Victoria and B/Yamagata was also higher in southern China 0.14 (IQR: 0.07-0.24) than in norther China 0.10 (IQR: 0.04-0.18). High relative humidity significantly increased subtype interference, with the highest relative risk reaching 20.59 (95% CI: 6.12-69.33) in southern China. Southern China exhibited higher levels of subtype interference, particularly between A/H1N1 and A/H3N2. Higher relative humidity has a more pronounced promoting effect on subtype interference.

Epidemiological features of seasonal influenza transmission among 11 climate zones in Chinese Mainland

BACKGROUND

Previous studies provided some evidence of meteorological factors influence seasonal influenza transmission patterns varying across regions and latitudes. However, research on seasonal influenza activities based on climate zones are still in lack. This study aims to utilize the ecological-based Köppen Geiger climate zones classification system to compare the spatial and temporal epidemiological characteristics of seasonal influenza in Chinese Mainland and assess the feasibility of developing an early warning system.

METHODS

Weekly influenza cases number from 2014 to 2019 at the county and city level were sourced from China National Notifiable Infectious Disease Report Information System. Epidemic temporal indices, time series seasonality decomposition, spatial modelling theories including Moran's I and local indicators of spatial association were applied to identify the spatial and temporal patterns of influenza transmission.

RESULTS

All climate zones had peaks in Winter-Spring season. Arid, desert, cold (BWk) showed up the first peak. Only Tropical, savannah (Aw) and Temperate, dry winter with hot summer (Cwa) zones had unique summer peak. Temperate, no dry season and hot summer (Cfa) zone had highest average incidence rate (IR) at 1.047/100,000. The Global Moran's I showed that average IR had significant clustered trend (z = 53.69, P < 0.001), with local Moran's I identified high-high cluster in Cfa and Cwa. IR differed among three age groups between climate zones (0-14 years old: F = 26.80, P < 0.001; 15-64 years old: F = 25.04, P < 0.001; Above 65 years old: F = 5.27, P < 0.001). Age group 0-14 years had highest average IR in Cwa and Cfa (IR = 6.23 and 6.21) with unique dual peaks in winter and spring season showed by seasonality decomposition.

CONCLUSIONS

Seasonal influenza exhibited distinct spatial and temporal patterns in different climate zones. Seasonal influenza primarily emerged in BWk, subsequently in Cfa and Cwa. Cfa, Cwa and BSk pose high risk for seasonal influenza epidemics. The research finds will provide scientific evidence for developing seasonal influenza early warning system based on climate zones.

The impact of current and future climates on spatiotemporal dynamics of influenza in a tropical setting

Although the drivers of influenza have been well studied in high-income settings in temperate regions, many open questions remain about the burden, seasonality, and drivers of influenza dynamics in the tropics. In temperate climates, the inverse relationship between specific humidity and transmission can explain much of the observed temporal and spatial patterns of influenza outbreaks. Yet, this relationship fails to explain seasonality, or lack there-of, in tropical and subtropical countries. Here, we analyzed eight years of influenza surveillance data from 12 locations in Bangladesh to quantify the role of climate in driving disease dynamics in a tropical setting with a distinct rainy season. We find strong evidence for a nonlinear bimodal relationship between specific humidity and influenza transmission in Bangladesh, with highest transmission occurring for relatively low and high specific humidity values. We simulated influenza burden under current and future climate in Bangladesh using a mathematical model with a bimodal relationship between humidity and transmission, and decreased transmission at very high temperatures, while accounting for changes in population immunity. The climate-driven mechanistic model can accurately capture both the temporal and spatial variation in influenza activity observed across Bangladesh, highlighting the usefulness of mechanistic models for low-income countries with inadequate surveillance. By using climate model projections, we also highlight the potential impact of climate change on influenza dynamics in the tropics and the public health consequences.

Impact of meteorological and demographic factors on the influenza epidemic in Japan: a large observational database study

Factors affecting the start date of the influenza epidemic season and total number of infected persons per 1,000,000 population in 47 prefectures of Japan were evaluated. This retrospective observational study (September 2014-August 2019; N = 472,740-883,804) evaluated data from a Japanese health insurance claims database. Single and multiple regression analyses evaluated the time to start of the epidemic or total infected persons per 1,000,000 population with time to absolute humidity (AH) or number of days with AH (≤ 5.5, ≤ 6.0, ≤ 6.5, and ≤ 7.0), total visitors (first epidemic month or per day), and total population. For the 2014/15, 2015/16, and 2016/17 seasons, a weak-to-moderate positive correlation (R2: 0.042-0.417) was observed between time to start of the epidemic and time to first day with AH below the cutoff values. Except in the 2016/17 season (R2: 0.089), a moderate correlation was reported between time to start of the epidemic and the total population (R2: 0.212-0.401). For all seasons, multiple regression analysis showed negative R2 for time to start of the epidemic and total visitors and population density (positive for time to AH ≤ 7.0). The earlier the climate becomes suitable for virus transmission and the higher the human mobility (more visitors and higher population density), the earlier the epidemic season tends to begin.

Repeatability and timing of tropical influenza epidemics

Much of the world experiences influenza in yearly recurring seasons, particularly in temperate areas. These patterns can be considered repeatable if they occur predictably and consistently at the same time of year. In tropical areas, including southeast Asia, timing of influenza epidemics is less consistent, leading to a lack of consensus regarding whether influenza is repeatable. This study aimed to assess repeatability of influenza in Vietnam, with repeatability defined as seasonality that occurs at a consistent time of year with low variation. We developed a mathematical model incorporating parameters to represent periods of increased transmission and then fitted the model to data collected from sentinel hospitals throughout Vietnam as well as four temperate locations. We fitted the model for individual (sub)types of influenza as well as all combined influenza throughout northern, central, and southern Vietnam. Repeatability was evaluated through the variance of the timings of peak transmission. Model fits from Vietnam show high variance (sd = 64-179 days) in peak transmission timing, with peaks occurring at irregular intervals and throughout different times of year. Fits from temperate locations showed regular, annual epidemics in winter months, with low variance in peak timings (sd = 32-57 days). This suggests that influenza patterns are not repeatable or seasonal in Vietnam. Influenza prevention in Vietnam therefore cannot rely on anticipation of regularly occurring outbreaks.

Association of climatic variables with risk of transmission of influenza in Guangzhou, China, 2005-2021

BACKGROUND

Climatic variables constitute important extrinsic determinants of transmission and seasonality of influenza. Yet quantitative evidence of independent associations of viral transmissibility with climatic factors has thus far been scarce and little is known about the potential effects of interactions between climatic factors on transmission.

OBJECTIVE

This study aimed to examine the associations of key climatic factors with risk of influenza transmission in subtropical Guangzhou.

METHODS

Influenza epidemics were identified over a 17-year period using the moving epidemic method (MEM) from a dataset of N = 295,981 clinically- and laboratory-confirmed cases of influenza in Guangzhou. Data on eight key climatic variables were collected from China Meteorological Data Service Centre. Generalized additive model combined with the distributed lag non-linear model (DLNM) were developed to estimate the exposure-lag-response curve showing the trajectory of instantaneous reproduction number (R_t) across the distribution of each climatic variable after adjusting for depletion of susceptible, inter-epidemic effect and school holidays. The potential interaction effects of temperature, humidity and rainfall on influenza transmission were also examined.

RESULTS

Over the study period (2005-21), 21 distinct influenza epidemics with varying peak timings and durations were identified. Increasing air temperature, sunshine, absolute and relative humidity were significantly associated with lower R_t, while the associations were opposite in the case of ambient pressure, wind speed and rainfall. Rainfall, relative humidity, and ambient temperature were the top three climatic contributors to variance in transmissibility. Interaction models found that the detrimental association between high relative humidity and transmissibility was more pronounced at high temperature and rainfall.

CONCLUSION

Our findings are likely to help understand the complex role of climatic factors in influenza transmission, guiding informed climate-related mitigation and adaptation policies to reduce transmission in high density subtropical cities.

Development of Accurate Long-lead COVID-19 Forecast

Coronavirus disease 2019 (COVID-19) will likely remain a major public health burden; accurate forecast of COVID-19 epidemic outcomes several months into the future is needed to support more proactive planning. Here, we propose strategies to address three major forecast challenges, i.e., error growth, the emergence of new variants, and infection seasonality. Using these strategies in combination we generate retrospective predictions of COVID-19 cases and deaths 6 months in the future for 10 representative US states. Tallied over >25,000 retrospective predictions through September 2022, the forecast approach using all three strategies consistently outperformed a baseline forecast approach without these strategies across different variant waves and locations, for all forecast targets. Overall, probabilistic forecast accuracy improved by 64% and 38% and point prediction accuracy by 133% and 87% for cases and deaths, respectively. Real-time 6-month lead predictions made in early October 2022 suggested large attack rates in most states but a lower burden of deaths than previous waves during October 2022 -March 2023; these predictions are in general accurate compared to reported data. The superior skill of the forecast methods developed here demonstrate means for generating more accurate long-lead forecast of COVID-19 and possibly other infectious diseases.

Development of Accurate Long-lead COVID-19 Forecast

Coronavirus disease 2019 (COVID-19) will likely remain a major public health burden; accurate forecast of COVID-19 epidemic outcomes several months into the future is needed to support more proactive planning. Here, we propose strategies to address three major forecast challenges, i.e., error growth, the emergence of new variants, and infection seasonality. Using these strategies in combination we generate retrospective predictions of COVID-19 cases and deaths 6 months in the future for 10 representative US states. Tallied over >25,000 retrospective predictions through September 2022, the forecast approach using all three strategies consistently outperformed a baseline forecast approach without these strategies across different variant waves and locations, for all forecast targets. Overall, probabilistic forecast accuracy improved by 64% and 38% and point prediction accuracy by 133% and 87% for cases and deaths, respectively. Real-time 6-month lead predictions made in early October 2022 suggested large attack rates in most states but a lower burden of deaths than previous waves during October 2022 - March 2023; these predictions are in general accurate compared to reported data. The superior skill of the forecast methods developed here demonstrate means for generating more accurate long-lead forecast of COVID-19 and possibly other infectious diseases.

AUTHOR SUMMARY

Infectious disease forecast aims to reliably predict the most likely future outcomes during an epidemic. To date, reliable COVID-19 forecast remains elusive and is needed to support more proactive planning. Here, we pinpoint the major challenges facing COVID-19 forecast and propose three strategies. Comprehensive testing shows the forecast approach using all three strategies consistently outperforms a baseline approach without these strategies across different variant waves and locations in the United States for all forecast targets, improving the probabilistic forecast accuracy by ∼50% and point prediction accuracy by ∼100%. The superior skills of the forecast methods developed here demonstrate means for generating more accurate long-lead COVID-19 forecasts. The methods may be also applicable to other infectious diseases.

ONE SENTENCE SUMMARY

To support more proactive planning, we develop COVID-19 forecast methods that substantially improve accuracy with lead time up to 6 months.

Seasonal transmission dynamics of varicella in Japan: The role of temperature and school holidays

In Japan, major and minor bimodal seasonal patterns of varicella have been observed. To investigate the underlying mechanisms of seasonality, we evaluated the effects of the school term and temperature on the incidence of varicella in Japan. We analyzed epidemiological, demographic and climate datasets of seven prefectures in Japan. We fitted a generalized linear model to the number of varicella notifications from 2000 to 2009 and quantified the transmission rates as well as the force of infection, by prefecture. To evaluate the effect of annual variation in temperature on the rate of transmission, we assumed a threshold temperature value. In northern Japan, which has large annual temperature variations, a bimodal pattern in the epidemic curve was observed, reflecting the large deviation in average weekly temperature from the threshold value. This bimodal pattern was diminished with southward prefectures, gradually shifting to a unimodal pattern in the epidemic curve, with little temperature deviation from the threshold. The transmission rate and force of infection, considering the school term and temperature deviation from the threshold, exhibited similar seasonal patterns, with a bimodal pattern in the north and a unimodal pattern in the south. Our findings suggest the existence of preferable temperatures for varicella transmission and an interactive effect of the school term and temperature. Investigating the potential impact of temperature elevation that could reshape the epidemic pattern of varicella to become unimodal, even in the northern part of Japan, is required.

COVID-19 pandemic dynamics in South Africa and epidemiological characteristics of three variants of concern (Beta, Delta, and Omicron)

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants of concern (VOCs) have been key drivers of new coronavirus disease 2019 (COVID-19) pandemic waves. To better understand variant epidemiologic characteristics, here we apply a model-inference system to reconstruct SARS-CoV-2 transmission dynamics in South Africa, a country that has experienced three VOC pandemic waves (i.e. Beta, Delta, and Omicron BA.1) by February 2022. We estimate key epidemiologic quantities in each of the nine South African provinces during March 2020 to February 2022, while accounting for changing detection rates, infection seasonality, nonpharmaceutical interventions, and vaccination. Model validation shows that estimated underlying infection rates and key parameters (e.g. infection-detection rate and infection-fatality risk) are in line with independent epidemiological data and investigations. In addition, retrospective predictions capture pandemic trajectories beyond the model training period. These detailed, validated model-inference estimates thus enable quantification of both the immune erosion potential and transmissibility of three major SARS-CoV-2 VOCs, that is, Beta, Delta, and Omicron BA.1. These findings help elucidate changing COVID-19 dynamics and inform future public health planning.

Prevalence of respiratory viruses among paediatric patients in acute respiratory illnesses in Malaysia

Objectives

Acute respiratory infections (ARIs) are one of the leading causes of childhood morbidity and mortality worldwide. However, there is limited surveillance data on the epidemiological burden of respiratory pathogens in tropical countries like Malaysia. This study aims to estimate the prevalence of respiratory pathogens causing ARIs among children aged <18 years old in Malaysia and their epidemiological characteristics.

Methods

Nasopharyngeal swab specimens received at 12 laboratories located in different states of Malaysia from 2015–2019 were studied. Detection of 18 respiratory pathogens were performed using multiplex PCR.

Results

Data from a total of 23,306 paediatric patients who presented with ARI over a five-year period was studied. Of these, 18538 (79.5%) were tested positive. The most prevalent respiratory pathogens detected in this study were enterovirus/ rhinovirus (6837/ 23000; 29.7%), influenza virus (5176/ 23000; 22.5%) and respiratory syncytial virus (RSV) (3652/ 23000; 15.9%). Throughout the study period, RSV demonstrated the most pronounce seasonality; peak infection occurred during July to September. Whereas the influenza virus was detected year-round in Malaysia. No seasonal variation was noted in other respiratory pathogens. The risk of RSV hospitalisation was found to be significantly higher in children aged less than two years old, whereas hospitalisation rates for the influenza virus peaked at children aged between 3–6 years old.

Conclusion

This study provides insight into the epidemiology and the seasonality of the causative pathogens of ARI among the paediatric population in Malaysia. Knowledge of seasonal respiratory pathogens epidemiological dynamics will facilitate the identification of a target window for vaccination.

The double-edged impacts of COVID-19 epidemic on influenza vaccination uptake in nurses in Hong Kong

BACKGROUND

Worldwide, the COVID-19 pandemic has disrupted influenza seasonality and impacted influenza vaccination behaviours. This study examines the patterns of influenza vaccination uptake in nurses during the pre-pandemic seasons and amidst the pandemic in 2020/21 in subtropical Hong Kong.

METHODS

In April 2021, a cross-sectional survey was conducted in practising and student nurses. Six patterns were differentiated by their influenza vaccination status in 2020/21 and preceding five seasons: (a) resistance throughout, (b) negative change, (c) decreased tendency, (d) increased tendency (e) positive change, and (f) adherence throughout. Participants' characteristics and reasons for vaccination/non-vaccination and their associations with each vaccination pattern were explored using multivariable multinomial regression.

RESULTS

Among 1323 nurses (88% female with a median age of 39 years), no substantial change in vaccination uptake was noted from 2019/20 (41%) to 2020/21 (42%). Some 4% were positive-changers who were older at age of ≥50 years (aOR 4.77) and more likely to anticipate a reduced risk of severe disease with COVID-19/influenza co-infection following vaccination (aOR 8.76). Negative-changers made up 3% and were more inclined to perceive an unlikelihood of widespread influenza outbreaks amidst the pandemic (aOR 3.67). Some 26 and 43% remained adherent and resistant to influenza vaccination throughout respectively notwithstanding the COVID-19 outbreak.

CONCLUSIONS

In contrast to the increased uptake elsewhere, the stable influenza vaccination coverage among nurses in Hong Kong could be explained by the ubiquity of strict physical distancing regulations, which have deterred vaccination in some nurses that offsets the new uptakes induced by the mild COVID-19 outbreaks.

COVID-19 pandemic dynamics in South Africa and epidemiological characteristics of three variants of concern (Beta, Delta, and Omicron)

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants of concern (VOCs) have been key drivers of new coronavirus disease 2019 (COVID-19) pandemic waves. To better understand variant epidemiologic characteristics, here we apply a model-inference system to reconstruct SARS-CoV-2 transmission dynamics in South Africa, a country that has experienced three VOC pandemic waves (i.e. Beta, Delta, and Omicron). We estimate key epidemiologic quantities in each of the nine South African provinces during March 2020 â€" Feb 2022, while accounting for changing detection rates, infection seasonality, nonpharmaceutical interventions, and vaccination. Model validation shows that estimated underlying infection rates and key parameters (e.g., infection-detection rate and infection-fatality risk) are in line with independent epidemiological data and investigations. In addition, retrospective predictions capture pandemic trajectories beyond the model training period. These detailed, validated model-inference estimates thus enable quantification of both the immune erosion potential and transmissibility of three major SARS-CoV-2 VOCs, i.e., Beta, Delta, and Omicron. These findings help elucidate changing COVID-19 dynamics and inform future public health planning.

The Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) Pandemic: Are Africa's Prevalence and Mortality Rates Relatively Low?

Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), the cause of coronavirus disease 19 (COVID-19), has been rapidly spreading since December 2019, and within a few months, it turned out to be a global pandemic. The disease affects primarily the lungs, but its pathogenesis spreads to other organs as well. However, its mortality rates vary, and in the majority of infected people, there are no serious consequences. Many factors including advanced age, preexisting health conditions, and genetic predispositions are believed to exacerbate outcomes of COVID-19. The virus contains several structural proteins including the spike (S) protein with subunits for binding, fusion, and internalization into host cells following interaction with host cell receptors and proteases (ACE2 and TMPRSS2, respectively) to cause the subsequent pathology. Although the pandemic has spread into all countries, most of Africa is thought of as having relatively less prevalence and mortality. Several hypotheses have been forwarded as reasons for this and include warmer weather conditions, vaccination with BCG (i.e., trained immunity), and previous malaria infection. From genetics or metabolic points of view, it has been proposed that most African populations could be protected to some degree because they lack some genetic susceptibility risk factors or have low-level expression of allelic variants, such as ACE2 and TMPRSS2 that are thought to be involved in increased infection risk or disease severity. The frequency of occurrence of α-1 antitrypsin (an inhibitor of a tissue-degrading protease, thereby protecting target host tissues including the lung) deficiency is also reported to be low in most African populations. More recently, infections in Africa appear to be on the rise. In general, there are few studies on the epidemiology and pathogenesis of the disease in African contexts, and the overall costs and human life losses due to the pandemic in Africa will be determined by all factors and conditions interacting in complex ways.

Development of a model-inference system for estimating epidemiological characteristics of SARS-CoV-2 variants of concern

To support COVID-19 pandemic planning, we develop a model-inference system to estimate epidemiological properties of new SARS-CoV-2 variants of concern using case and mortality data while accounting for under-ascertainment, disease seasonality, non-pharmaceutical interventions, and mass-vaccination. Applying this system to study three variants of concern, we estimate that B.1.1.7 has a 46.6% (95% CI: 32.3-54.6%) transmissibility increase but nominal immune escape from protection induced by prior wild-type infection; B.1.351 has a 32.4% (95% CI: 14.6-48.0%) transmissibility increase and 61.3% (95% CI: 42.6-85.8%) immune escape; and P.1 has a 43.3% (95% CI: 30.3-65.3%) transmissibility increase and 52.5% (95% CI: 0-75.8%) immune escape. Model simulations indicate that B.1.351 and P.1 could outcompete B.1.1.7 and lead to increased infections. Our findings highlight the importance of preventing the spread of variants of concern, via continued preventive measures, prompt mass-vaccination, continued vaccine efficacy monitoring, and possible updating of vaccine formulations to ensure high efficacy.