Associations between seasonal influenza and meteorological parameters in Costa Rica, Honduras and Nicaragua

Convolutional LSTM-LSTM model for predicting the daily number of influenza patients in South Korea using satellite images

OBJECTIVES

Influenza affects a considerable proportion of the global population each year, and meteorological conditions may have a significant impact on its transmission. In this study, we aimed to develop a prediction model for the number of influenza patients at the national level using satellite images and provide a basis for predicting influenza through satellite image data.

STUDY DESIGN

We developed an influenza incidence prediction model using satellite images and influenza patient data.

METHODS

We collected satellite images and daily influenza patient data from July 2014 to June 2019 and developed a convolutional long short-term memory (LSTM)-LSTM neural network model. The model with the lowest average of mean absolute error (MAE) was selected.

RESULTS

The final model showed a high correlation between the predicted and actual number of influenza patients, with an average MAE of 5.9010 per million population. The model performed best with a 2-week time sequence.

CONCLUSIONS

We developed a national-level prediction model using satellite images to predict influenza incidence. The model offers the advantage of nationwide analysis. These results may reduce the burden of influenza by enabling timely public health interventions.

Spatiotemporal effects of meteorological conditions on global influenza peaks

BACKGROUND

Numerous studies have suggested that meteorological conditions such as temperature and absolute humidity are highly indicative of influenza outbreaks. However, the explanatory power of meteorological factors on the seasonal influenza peaks varied widely between countries at different latitudes.

OBJECTIVES

We aimed to explore the modification effects of meteorological factors on the seasonal influenza peaks in multi-countries.

METHODS

Data on influenza positive rate (IPR) were collected across 57 countries and data on meteorological factors were collected from ECMWF Reanalysis v5 (ERA5). We used linear regression and generalized additive models to investigate the spatiotemporal associations between meteorological conditions and influenza peaks in cold and warm seasons.

RESULTS

Influenza peaks were significantly correlated with months with both lower and higher temperatures. In temperate countries, the average intensity of cold season peaks was stronger than that of warm season peaks. However, the average intensity of warm season peaks was stronfger than of cold season peaks in tropical countries. Temperature and specific humidity had synergistic effects on influenza peaks at different latitudes, stronger in temperate countries (cold season: R2=0.90; warm season: R2=0.84) and weaker in tropical countries (cold season: R2=0.64; warm season: R2=0.03). Furthermore, the effects could be divided into cold-dry and warm-humid modes. The temperature transition threshold between the two modes was 16.5-19.5 °C. During the transition from cold-dry mode to warm-humid mode, the average 2 m specific humidity increased by 2.15 times, illustrating that transporting a large amount of water vapor may compensate for the negative effect of rising temperatures on the spread of the influenza virus.

CONCLUSION

Differences in the global influenza peaks were related to the synergistic influence of temperature and specific humidity. The global influenza peaks could be divided into cold-dry and warm-humid modes, and specific thresholds of meteorological conditions were needed for the transition of the two modes.

Study of the influence of meteorological factors on HFMD and prediction based on the LSTM algorithm in Fuzhou, China

BACKGROUND

This study adopted complete meteorological indicators, including eight items, to explore their impact on hand, foot, and mouth disease (HFMD) in Fuzhou, and predict the incidence of HFMD through the long short-term memory (LSTM) neural network algorithm of artificial intelligence.

METHOD

A distributed lag nonlinear model (DLNM) was used to analyse the influence of meteorological factors on HFMD in Fuzhou from 2010 to 2021. Then, the numbers of HFMD cases in 2019, 2020 and 2021 were predicted using the LSTM model through multifactor single-step and multistep rolling methods. The root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE) and symmetric mean absolute percentage error (SMAPE) were used to evaluate the accuracy of the model predictions.

RESULTS

Overall, the effect of daily precipitation on HFMD was not significant. Low (4 hPa) and high (≥ 21 hPa) daily air pressure difference (PRSD) and low (< 7 °C) and high (> 12 °C) daily air temperature difference (TEMD) were risk factors for HFMD. The RMSE, MAE, MAPE and SMAPE of using the weekly multifactor data to predict the cases of HFMD on the following day, from 2019 to 2021, were lower than those of using the daily multifactor data to predict the cases of HFMD on the following day. In particular, the RMSE, MAE, MAPE and SMAPE of using weekly multifactor data to predict the following week's daily average cases of HFMD were much lower, and similar results were also found in urban and rural areas, which indicating that this approach was more accurate.

CONCLUSION

This study's LSTM models combined with meteorological factors (excluding PRE) can be used to accurately predict HFMD in Fuzhou, especially the method of predicting the daily average cases of HFMD in the following week using weekly multifactor data.

Study on the influence of meteorological factors on influenza in different regions and predictions based on an LSTM algorithm

BACKGROUND

Influenza epidemics pose a threat to human health. It has been reported that meteorological factors (MFs) are associated with influenza. This study aimed to explore the similarities and differences between the influences of more comprehensive MFs on influenza in cities with different economic, geographical and climatic characteristics in Fujian Province. Then, the information was used to predict the daily number of cases of influenza in various cities based on MFs to provide bases for early warning systems and outbreak prevention.

METHOD

Distributed lag nonlinear models (DLNMs) were used to analyse the influence of MFs on influenza in different regions of Fujian Province from 2010 to 2021. Long short-term memory (LSTM) was used to train and model daily cases of influenza in 2010-2018, 2010-2019, and 2010-2020 based on meteorological daily values. Daily cases of influenza in 2019, 2020 and 2021 were predicted. The root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE) and symmetric mean absolute percentage error (SMAPE) were used to quantify the accuracy of model predictions.

RESULTS

The cumulative effect of low and high values of air pressure (PRS), air temperature (TEM), air temperature difference (TEMD) and sunshine duration (SSD) on the risk of influenza was obvious. Low (< 979 hPa), medium (983 to 987 hPa) and high (> 112 hPa) PRS were associated with a higher risk of influenza in women, children aged 0 to 12 years, and rural populations. Low (< 9 °C) and high (> 23 °C) TEM were risk factors for influenza in four cities. Wind speed (WIN) had a more significant effect on the risk of influenza in the ≥ 60-year-old group. Low (< 40%) and high (> 80%) relative humidity (RHU) in Fuzhou and Xiamen had a significant effect on influenza. When PRS was between 1005-1015 hPa, RHU > 60%, PRE was low, TEM was between 10-20 °C, and WIN was low, the interaction between different MFs and influenza was most obvious. The RMSE, MAE, MAPE, and SMAPE evaluation indices of the predictions in 2019, 2020 and 2021 were low, and the prediction accuracy was high.

CONCLUSION

All eight MFs studied had an impact on influenza in four cities, but there were similarities and differences. The LSTM model, combined with these eight MFs, was highly accurate in predicting the daily cases of influenza. These MFs and prediction models could be incorporated into the influenza early warning and prediction system of each city and used as a reference to formulate prevention strategies for relevant departments.

Global Geographical and Temporal Patterns of Seasonal Influenza and Associated Climatic Factors

Understanding geographical and temporal patterns of seasonal influenza can help strengthen influenza surveillance to early detect epidemics and inform influenza prevention and control programs. We examined variations in spatiotemporal patterns of seasonal influenza in different global regions and explored climatic factors that influence differences in influenza seasonality, through a systematic review of peer-reviewed publications. The literature search was conducted to identify original studies published between January 2005 and November 2016. Studies were selected using predetermined inclusion and exclusion criteria. The primary outcome was influenza cases; additional outcomes included seasonal or temporal patterns of influenza seasonality, study regions (temperate or tropical), and associated climatic factors. Of the 2,160 records identified in the selection process, 36 eligible studies were included. There were significant differences in influenza seasonality in terms of the time of onset, duration, number of peaks, and amplitude of epidemics between temperate and tropical/subtropical regions. Different viral types, cocirculation of influenza viruses, and climatic factors, especially temperature and absolute humidity, contributed to the variations in spatiotemporal patterns of seasonal influenza. The findings reported in this review could inform global surveillance of seasonal influenza and influenza prevention and control measures such as vaccination recommendations for different regions.

Latitudes mediate the association between influenza activity and meteorological factors: A nationwide modelling analysis in 45 Japanese prefectures from 2000 to 2018

BACKGROUND

Cold and dry conditions were well-documented as a major determinant of influenza seasonality in temperate countries but the association may not be consistent when the climate in temperate areas is closer to that in sub-tropical areas. We hypothesized latitudes may mediate the association between influenza activity and meteorological factors in 45 Japanese prefectures.

METHODS

We used the weekly incidence of influenza-like illness of 45 prefectures from 2000 to 2018 as a proxy for influenza activity in Japan, a temperate country lying off the east coast of Asia. A combination of generalized additive model and distributed lag nonlinear model was adopted to investigate the associations between meteorological factors (average temperature, relative humidity, total rainfall, and actual vapour pressure, a proxy for absolute humidity) and the influenza incidence. Kendall's tau b (τ) and Spearman correlation coefficient (r_s) between latitude and the adjusted relative risk (ARR) of each meteorological factor were also assessed.

RESULTS

A higher vapour pressure was significantly associated with a lower influenza risk but the ARR strongly weakened along with a lower latitude (τ = -0.23, p-value = 0.02; r_s = -0.33, p-value = 0.03). Lower temperature and lower relatively humidity were significantly associated with higher influenza risks in over 65% and around 40% of the prefectures respectively but the strength and significance of the correlations between their ARRs and latitude were weaker than that from vapour pressure.

CONCLUSION

Even though the range of latitudes in Japan is small (26°N-43°N), the relationships between meteorological factors and influenza activity were mediated by the latitude. Our study echoed absolute humidity played a more important role in relating influenza risk, but we on the other hand showed its effect on influenza activity could be hampered in a low-latitude temperate region, which have a warmer climate. These findings thus offer a high-resolution characterization of the role of meteorological factors on influenza seasonality.

Circulation of influenza virus from 2009 to 2018 in Cameroon: 10 years of surveillance data

Since the recent emergence of several subtypes of influenza viruses with pandemic potentials, there has been growing interest on the control of this infection worldwide. This study aimed to describe the 10 years of influenza activity in Cameroon between January 2009 and December 2018. Respiratory samples were collected from sentinel sites responsible for influenza surveillance in Cameroon and analyzed for the presence of influenza. Globally, 9 of the 10 administrative regions of the country were represented with at least 1 year of data. A total of 11816 respiratory samples were collected and influenza virus detection rate was 24.0%. The most represented age group was the 0-1 years representing more than 40% of the collected samples and possessing the lowest proportion of influenza cases (16.2%). Meanwhile higher proportions of influenza positive cases was found in the 2-4, 5-14 and 15-49 years age group at ≥29%. Among outpatients, the frequency of influenza virus was 24.8% while in hospitalized patients, 18.7% of samples were positive for influenza virus. We noted year-round circulation of influenza virus in Cameroon with 2 peaks in activity: a major peak in the months of September to December and a minor peak in the months of March to July. Antigenic characterization of influenza isolates showed 37.5% (6/16) vaccine match between the predominant Cameroon strains and the Northern hemisphere vaccine strains with majority of vaccine match observed in influenza B/Victoria subtype (4/6; 66.7%). Data collected from this surveillance system is essential to add to global information on the spread of influenza.

Effects and interaction of meteorological factors on influenza: Based on the surveillance data in Shaoyang, China

BACKGROUND

Previous studies have demonstrated that meteorological factors influence the incidence of influenza. However, little is known regarding the interactions of meteorological factors on the risk of influenza in China.

OBJECTIVE

The study aimed to evaluate the associations between meteorological factors and influenza in Shaoyang of southern China, and explore the interaction of temperature with humidity and rainfall.

METHODS

Weekly meteorological data and disease surveillance data of influenza in Shaoyang were collected from 2009 to 2012. According to the incubation period and infectious period of influenza virus, the maximum lag period was set as 3 weeks. A generalized additive model was conducted to evaluate the effect of meteorological factors on the weekly number of influenza cases and a stratification model was applied to investigate the interaction.

RESULTS

During the study period, the total number of influenza cases that were notified in the study area was 2506, with peak times occurring from December to March. After controlling for the confounders, each 5 °C decrease in minimum temperature was related to 8% (95%CI: 1-15%) increase in the number of influenza cases at a 1-week lag. There was an interaction between minimum temperature and relative humidity and the risk of influenza was higher in cold and less humid conditions than other conditions. The interaction between minimum temperature and rainfall was not statistically significant in our study.

CONCLUSIONS

The study suggests that minimum temperature is inversely associated with influenza in the study area of China, and the effect can be modified by relative humidity. Meteorological variables could be integrated in current public health surveillance system to better prepare for the risks of influenza.

Seasonal characteristics of influenza vary regionally across US

Given substantial regional differences in absolute humidity across the US and our understanding of the relationship between absolute humidity and influenza, we may expect important differences in regional seasonal influenza activity. Here, we assessed cross-seasonal influenza activity by comparing counts of positive influenza A and B rapid test results during the influenza season versus summer baseline periods for the 2016/2017 and 2017/2018 influenza years. Our analysis indicates significant regional patterns in cross-seasonal influenza activity, with relatively fewer influenza cases during the influenza season compared to summertime baseline periods in humid areas of the US, particularly in Florida and Hawaii. The cross-seasonal ratios vary from year-to-year and influenza type, but the geographic patterning of the ratios is relatively consistent. Mixed-effects regression models indicated absolute humidity during the influenza season was the strongest predictor of cross-seasonal influenza activity, suggesting a relationship between absolute humidity and cross-seasonal influenza activity. There was also evidence that absolute humidity during the summer plays a role, as well. This analysis suggests that spatial variation in seasonal absolute humidity levels may generate important regional differences in seasonal influenza activity and dynamics in the US.

Respiratory Viral Infection: An Underappreciated Cause of Acute Febrile Illness Admissions in Southern Sri Lanka

The contribution of respiratory viruses to acute febrile illness (AFI) burden is poorly characterized. We describe the prevalence, seasonality, and clinical features of respiratory viral infection among AFI admissions in Sri Lanka. We enrolled AFI patients ≥ 1 year of age admitted to a tertiary care hospital in southern Sri Lanka, June 2012-October 2014. We collected epidemiologic/clinical data and a nasal or nasopharyngeal sample that was tested using polymerase chain reaction (Luminex NxTAG, Austin, TX). We determined associations between weather data and respiratory viral activity using the Spearman correlation and assessed respiratory virus seasonality using a Program for Appropriate Technology definition. Bivariable and multivariable regression analyses were conducted to identify features associated with respiratory virus detection. Among 964 patients, median age was 26.2 years (interquartile range 14.6-39.9) and 646 (67.0%) were male. One-fifth (203, 21.1%) had respiratory virus detected: 13.9% influenza, 1.4% human enterovirus/rhinovirus, 1.4% parainfluenza virus, 1.1% respiratory syncytial virus, and 1.1% human metapneumovirus. Patients with respiratory virus identified were younger (median 9.8 versus 27.7 years, P < 0.001) and more likely to have respiratory signs and symptoms. Influenza A and respiratory viral activity peaked in February-June each year. Maximum daily temperature was associated with influenza and respiratory viral activity (P = 0.03 each). Patients with respiratory virus were as likely as others to be prescribed antibiotics (55.2% versus 52.6%, P = 0.51), and none reported prior influenza vaccination. Respiratory viral infection was a common cause of AFI. Improved access to vaccines and respiratory diagnostics may help reduce disease burden and inappropriate antibiotic use.

Modification of the Conventional Influenza Epidemic Models Using Environmental Parameters in Iran

Objectives

The association between the spread of infectious diseases and climate parameters has been widely studied in recent decades. In this paper, we formulate, exploit, and compare three variations of the susceptible-infected-recovered (SIR) model incorporating climate data. The SIR model is a well-studied model to investigate the dynamics of influenza viruses; however, the improved versions of the classic model have been developed by introducing external factors into the model.

Methods

The modification models are derived by multiplying a linear combination of three complementary factors, namely, temperature (T), precipitation (P), and humidity (H) by the transmission rate. The performance of these proposed models is evaluated against the standard model for two outbreak seasons.

Results

The values of the root-mean-square error (RMSE) and the Akaike information criterion (AIC) improved as they declined from 8.76 to 7.05 and from 98.12 to 93.01 for season 2013/14, respectively. Similarly, for season 2014/15, the RMSE and AIC decreased from 8.10 to 6.45 and from 117.73 to 107.91, respectively. The estimated values of R(t) in the framework of the standard and modified SIR models are also compared.

Conclusions

Through simulations, we determined that among the studied environmental factors, precipitation showed the strongest correlation with the transmission dynamics of influenza. Moreover, the SIR+P+T model is the most efficient for simulating the behavioral dynamics of influenza in the area of interest.

Influence of meteorological parameters in the seasonality of influenza viruses circulating in Northern Cameroon

BACKGROUND

Several studies have demonstrated the role of meteorological parameters in the seasonality of influenza viruses in tropical and subtropical regions, most importantly temperature, humidity, and rainfall.

OBJECTIVES

This study aimed to describe the influence of meteorological parameters in the seasonality of influenza viruses in Northern Cameroon, a region characterized by high temperatures.

METHODS

This was a retrospective study performed in Garoua Cameroon from January 2014 to December 2016. Monthly proportions of confirmed influenza cases from six sentinel sites were considered as dependent variables, whereas monthly values of mean temperature, average relative humidity, and accumulated rainfall were considered as independent variables. A vector error correction model was used to determine the relationship between influenza activity and the meteorological variables.

RESULTS AND CONCLUSION

Analysis showed that there was a statistically significant association between overall influenza activity and influenza A activity with respect to average relative humidity. A unit increase in humidity within a given month leads to more than 85% rise in the overall influenza and influenza A activity 2 months later. Meanwhile, none of the three meteorological variables could explain influenza B activity. This observation is essential in filling the gap of knowledge and could help in the prevention and control strategies to strengthen influenza surveillance program in Cameroon.

The effect of ambient temperature on the activity of influenza and influenza like illness in Jiangsu Province, China

OBJECTIVE

We aimed to evaluate and quantify the association between ambient temperature and activity of influenza like illness (ILI) and influenza in Jiangsu Province, China.

METHOD

Daily data of meteorology, influenza-like illness and detected influenza virus from 1 April 2013 to 27 March 2016 were collected. Distributed lag non-linear model (DLNM) was used to quantify the exposure-lag-response of ILI and influenza activity to daily average temperature.

RESULT

Influenza A virus (Flu-A) circulated throughout the year with two peaks at -4 °C and 28 °C respectively, while influenza B (Flu-B) viruses were usually tested positive in winter or early spring and peaked at 5 °C. The lag-response curves revealed that the RR of ILI increased with time and peaked 1 day later at low temperature (3 °C), however, the maximum RR of ILI caused by high temperature (26 °C) appeared immediately on day 0, the similar phenomena of immediate effect to ILI at high temperature were also observed in the lag-response curve for Flu-A or Flu-B.

CONCLUSION

ILI and Flu-A experienced two peaks of circulates at both low and high temperature in Jiangsu. The influenza viruses activity did drive up the rising of ILI%, particularly the activity of Flu-A which circulated throughout the year played a crucial role. Regional homogeneity was the relatively mainstream in aspects of cumulative association between influenza activity and temperature in Jiangsu Province.

Associations between meteorological parameters and influenza activity in a subtropical country: Case of five sentinel sites in Yaoundé-Cameroon

Influenza is associated with highly contagious respiratory infections. Previous research has found that influenza transmission is often associated with climate variables especially in temperate regions. This study was performed in order to fill the gap of knowledge regarding the relationship between incidence of influenza and three meteorological parameters (temperature, rainfall and humidity) in a tropical setting. This was a retrospective study performed in Yaoundé-Cameroon from January 2009 to November 2015. Weekly proportions of confirmed influenza cases from five sentinel sites were considered as dependent variables, whereas weekly values of mean temperature, average relative humidity and accumulated rainfall were considered as independent variables. A univariate linear regression model was used in determining associations between influenza activity and weather covariates. A time-series method was used to predict on future values of influenza activity. The data was divided into 2 parts; the first 71 months were used to calibrate the model, and the last 12 months to test for prediction. Overall, there were 1173 confirmed infections with influenza virus. Linear regression analysis showed that there was no statistically significant association observed between influenza activity and weather variables. Very weak relationships (-0.1 < r < 0.1) were observed. Three prediction models were obtained for the different viral types (overall positive, Influenza A and Influenza B). Model 1 (overall influenza) and model 2 (influenza A) fitted well during the estimation period; however, they did not succeed to make good forecasts for predictions. Accumulated rainfall was the only external covariate that enabled good fit of both models. Based on the stationary R², 29.5% and 41.1% of the variation in the series can be explained by model 1 and 2, respectively. This study laid more emphasis on the fact that influenza in Cameroon is characterized by year-round activity. The meteorological variables selected in this study did not enable good forecast of future influenza activity and certainly acted as proxies to other factors not considered, such as, UV radiation, absolute humidity, air quality and wind.

Construction and evaluation of two computational models for predicting the incidence of influenza in Nagasaki Prefecture, Japan

It remains challenging to forecast local, seasonal outbreaks of influenza. The goal of this study was to construct a computational model for predicting influenza incidence. We built two computational models including an Autoregressive Distributed Lag (ARDL) model and a hybrid model integrating ARDL with a Generalized Regression Neural Network (GRNN), to assess meteorological factors associated with temporal trends in influenza incidence. The modelling and forecasting performance of these two models were compared using observations collected between 2006 and 2015 in Nagasaki Prefecture, Japan. In both the training and forecasting stages, the hybrid model showed lower error rates, including a lower residual mean square error (RMSE) and mean absolute error (MAE) than the ARDL model. The lag of log-incidence, weekly average barometric pressure, and weekly average of air temperature were 4, 1, and 3, respectively in the ARDL model. The ARDL-GRNN hybrid model can serve as a tool to better understand the characteristics of influenza epidemic, and facilitate their prevention and control.

Influenza transmission during extreme indoor conditions in a low-resource tropical setting

Influenza transmission occurs throughout the planet across wide-ranging environmental conditions. However, our understanding of the environmental factors mediating transmission is evaluated using outdoor environmental measurements, which may not be representative of the indoor conditions where influenza is transmitted. In this study, we examined the relationship between indoor environment and influenza transmission in a low-resource tropical population. We used a case-based ascertainment design to enroll 34 households with a suspected influenza case and then monitored households for influenza, while recording indoor temperature and humidity data in each household. We show that the indoor environment is not commensurate with outdoor conditions and that the relationship between indoor and outdoor conditions varies significantly across homes. We also show evidence of influenza transmission in extreme indoor environments. Specifically, our data suggests that indoor environments averaged 29 °C, 18 g/kg specific humidity, and 68 % relative humidity across 15 transmission events observed. These indoor settings also exhibited significant temporal variability with temperatures as high as 39 °C and specific and relative humidity increasing to 22 g/kg and 85 %, respectively, during some transmission events. However, we were unable to detect differences in the transmission efficiency by indoor temperature or humidity conditions. Overall, these results indicate that laboratory studies investigating influenza transmission and virus survival should increase the range of environmental conditions that they assess and that observational studies investigating the relationship between environment and influenza activity should use caution using outdoor environmental measurements since they can be imprecise estimates of the conditions that mediate transmission indoors.

Spatial and Temporal Spread of Acute Viral Respiratory Infections in Young Children Living in High-altitude Rural Communities: A Prospective Household-based Study

BACKGROUND

Few studies have described patterns of transmission of viral acute respiratory infections (ARI) in children in developing countries. We examined the spatial and temporal spread of viral ARI among young children in rural Peruvian highland communities. Previous studies have described intense social interactions in those communities, which could influence the transmission of viral infections.

METHODS

We enrolled and followed children <3 years of age for detection of ARI during the 2009 to 2011 respiratory seasons in a rural setting with relatively wide geographic dispersion of households and communities. Viruses detected included influenza, respiratory syncytial virus (RSV), human metapneumovirus and parainfluenza 2 and 3 viruses (PIV2, PIV3). We used geospatial analyses to identify specific viral infection hot spots with high ARI incidence. We also explored the local spread of ARI from index cases using standard deviational ellipses.

RESULTS

Geospatial analyses revealed hot spots of high ARI incidence around the index cases of influenza outbreaks and RSV outbreak in 2010. Although PIV3 in 2009 and PIV2 in 2010 showed distinct spatial hot spots, clustering was not in proximity to their respective index cases. No significant aggregation around index cases was noted for other viruses. Standard deviational ellipse analyses suggested that influenza B and RSV in 2010, and human metapneumovirus in 2011 spread temporally in alignment with the major road network.

CONCLUSIONS

Despite the geographic dispersion of communities in this rural setting, we observed a rapid spread of viral ARI among young children. Influenza strains and RSV in 2010 had distinctive outbreaks arising from their index cases.

Temporal Patterns of Influenza A and B in Tropical and Temperate Countries: What Are the Lessons for Influenza Vaccination?

Introduction

Determining the optimal time to vaccinate is important for influenza vaccination programmes. Here, we assessed the temporal characteristics of influenza epidemics in the Northern and Southern hemispheres and in the tropics, and discuss their implications for vaccination programmes.

Methods

This was a retrospective analysis of surveillance data between 2000 and 2014 from the Global Influenza B Study database. The seasonal peak of influenza was defined as the week with the most reported cases (overall, A, and B) in the season. The duration of seasonal activity was assessed using the maximum proportion of influenza cases during three consecutive months and the minimum number of months with ≥80% of cases in the season. We also assessed whether co-circulation of A and B virus types affected the duration of influenza epidemics.

Results

212 influenza seasons and 571,907 cases were included from 30 countries. In tropical countries, the seasonal influenza activity lasted longer and the peaks of influenza A and B coincided less frequently than in temperate countries. Temporal characteristics of influenza epidemics were heterogeneous in the tropics, with distinct seasonal epidemics observed only in some countries. Seasons with co-circulation of influenza A and B were longer than influenza A seasons, especially in the tropics.

Discussion

Our findings show that influenza seasonality is less well defined in the tropics than in temperate regions. This has important implications for vaccination programmes in these countries. High-quality influenza surveillance systems are needed in the tropics to enable decisions about when to vaccinate.