The burden of hand, foot, and mouth disease (HFMD) in children under five is substantial, with the greatest burden in China. Most cases are mild, although some are severe and even fatal. A vaccine against EV-A71, the pathogen most commonly associated with severe HFMD, was licensed in China in Dec 2015 but not introduced into the National Immunisation Programme (NIP). It was hence not covered by routine national vaccine surveillance and its coverage remained unknown for the years following its initial licensure. Here we report the results of a novel data collection and analysis approach to address this knowledge gap. Local public health entities were invited to report county-specific numbers of EV-A71 vaccine doses administered between 2016 and 2019 in mainland China. A cohort model was then used to estimate vaccine coverage. The association between county-level factors (epidemiological, socioeconomic, demographic, and environmental) and vaccine coverage was assessed using zero-inflated beta regression models. We received responses from 2,248 out of 3,252 counties in 23 of 31provinces in mainland China. The median county-level EV-A71 vaccine coverage was 10.17% [IQR: 3.50%, 19.39%] in 2018 and 16.70% [IQR: 8.78%, 27.45%] in 2019. However, the median absolute differences in coverage (i.e., max-min) within-prefecture were ~30%. Results from the regression model indicate that low vaccine coverage was associated with low socioeconomic status, small populations and high proportions of young children. Coverage of EV-A71 vaccines was low in China prior to the COVID-19 pandemic, with substantial geographical disparities over 4 years after the initial vaccine licensure. Our results illustrate the private market response to a new childhood vaccine that is licensed but not centrally funded. Despite economic growth, vaccine coverage among marginalised populations will likely be low without targeted policy and financial support.

We aimed to estimate the incidence rate of rotavirus-associated diarrhea in Pudong New Area, Shanghai, China, from 2018 to 2022. Data from the hospital information systems of eight sentinel hospitals and the Healthcare Utilization and Attitudes Survey were used to calculate the incidence rate of rotavirus-associated diarrhea in Monte Carlo simulation. The findings revealed a decline in the positivity rate of rotavirus among medically-attended patients with diarrhea, from 13.5

Introduction Tuberculosis (TB) remains a globally concerning infectious disease, and significant challenges persist in attaining the 2030 targets set by the WHO. With the rapid advancements in computer-aided detection (CAD) technology, CAD-assisted Chest X-Ray (CAD-CXR) has been applied in TB patients triaging, but the practical application value of the CAD-CXR system in real-world primary healthcare settings in China for TB prevention and control has not been fully elucidated. This protocol reports a design of a cluster randomised controlled trial (CRCT), which aims to evaluate the effectiveness and clinical pathway of CAD-CXR in enhancing TB diagnostic yield in primary healthcare settings, thereby contributing to global TB elimination strategies.Methods and analysis Scheduled for September 2025, this CRCT will recruit 22 townships in Yichang of Hubei Province, China. These townships will be randomly allocated at a 1:1 ratio to either the CAD-CXR system intervention group or the control group. In the intervention group, healthcare providers will use the CAD-CXR analysis system to assist in TB screening, whereas the control group will rely solely on conventional CXR interpretation by radiologists. The primary outcome of the study is the TB diagnostic yield; the secondary outcomes include diagnostic delay duration and the accuracy of the CAD-CXR system. These metrics will be comprehensively evaluated to assess the effectiveness of the CAD-CXR intervention. Findings from this study are anticipated to offer evidence-based recommendations regarding the optimal application scenarios and implementation pathways for CAD-CXR.Ethics and dissemination This study was approved by the Ethics Committee of the Peking Union Medical College (CAMS&PUMC-IEC-2025-044). Findings of this study will be disseminated through traditional academic pathways, including peer-reviewed publications and conference presentations.Trial registration number NCT06963606.

With the rapid proliferation of information and communication technologies (ICT), epidemic dynamics are now inseparably entangled with information diffusion: public health messages alter individual behavior and thereby reshape transmission pathways, while rising case counts in turn drive demand for epidemic-related information. Despite growing recognition of this bidirectional coupling, most existing models either assume ad hoc behavioral parameters or confine analysis to a single intervention dimension, limiting their practical utility for guiding official communication campaigns. In this paper, we present an empirically parameterized agent-based model that couples information diffusion with epidemic transmission on a synthetic population of Yichang, China. The information-to-epidemic and epidemic-to-information coupling parameters are calibrated from real-world survey data. Using this framework, we conduct four sets of controlled experiments evaluating official health communication strategies across four dimensions: release timing, spatial targeting, release frequency, and target population. Results show that release timing is the most consequential dimension, reducing the final attack rate by up to 7.8 percentage points when the advisory is issued on Day 1 versus Day 45. Target population is the second most important lever: seeding the advisory among workers yields the lowest attack rate, whereas targeting the elderly-a group often prioritized in practice-produces the highest, because low-degree nodes in the information network diffuse the advisory too slowly to outpace the epidemic. Release frequency has a moderate effect, and spatial targeting produces negligible differences at the city scale tested. These findings offer concrete, evidence-grounded guidance for designing official health communication strategies during epidemic outbreaks.

Introduction: Since 2022, the global monkeypox (mpox) epidemic has undergone clade replacement from clade IIb to clade Ib, accompanied by a geographic shift in the epidemic's epicenter. However, systematic analyses spanning the entire epidemic cycle from 2022 to 2025 remain limited. Integrating global epidemiological and genomic data is essential for determining spatiotemporal transmission dynamics. Methods: Global mpox surveillance data, mpox virus whole-genome sequences, and demographic data were collected. Retrospective spatiotemporal scan statistics were used to identify significant transmission clusters. Phylogenetic trees were used to characterize evolutionary relationships among clades, and haplotype network analysis was used to infer transmission pathways. Results: A total of 165,244 mpox cases were reported globally, showing a two-phase pattern. From January 2022 to August 2023, clade IIb predominated, accounting for 96.18% of cases in Europe and the Americas. From September 2023 to September 2025, the epicenter shifted to Africa, where clade Ib became dominant; during this phase, Africa accounted for 71.48% of global cases. Spatiotemporal scan analysis identified 10 significant clusters, with the highest relative risk observed in the West African cluster in 2025. Phylogenetic analyses linked clade IIb to the European and American phases, whereas clade Ib was associated with the African phase. The haplotype network suggested that Africa is the core transmission region of clade Ib, with Europe serving as a dissemination hub. Conclusion: Mpox evolved from IIb-driven global dissemination to Ib-dominated regional clustering in Africa, with clade replacement driving the geographic shifts. The current risk is concentrated in Africa, underscoring the need for strengthened surveillance, genomic tracing, equitable access to resources, and targeted interventions.

Despite continuous viral evolution, it remains unclear whether SARS-CoV-2 has transitioned to transmission dynamics resembling those of other endemic respiratory pathogens in the post-pandemic era. During 2023–2025, we compared post-pandemic SARS-CoV-2 transmissibility with influenza virus (IFV) and common human coronaviruses (HCoV) transmissibility and SARS-CoV-2 transmissibility during and after the pandemic. Among 7,399 participants in 1,764 households, there were 704 acute-respiratory-infection index cases and 2,684 household contacts. Healthcare-seeking was higher for SARS-CoV-2 (17.6%) and IFV (28.1%) than HCoVs (9.8%). Post-pandemic SARS-CoV-2 household secondary attack risk (HSAR) (7.9%) was comparable to IFV (7.0%) and higher than HCoV (4.4%;P<0.05). Older adults had highest SARS-CoV-2 and HCoV transmissibility; children had highest IFV transmissibility, correlated with peak viral loads. Post-pandemic SARS-CoV-2 HSAR was lower than pandemic-period Omicron HSAR (adjusted odds ratio:3.30;95%CI:1.82–5.98) and Delta-period HSAR (aOR:2.46;95%CI:1.27–4.79). Pre-symptomatic transmission was observed for all three pathogens, highest for SARS-CoV-2 (35.7%). Our findings indicate that SARS-CoV-2 maintains high household transmissibility and requires substantial medical resources in the post-pandemic era, and non-pharmaceutical interventions and vaccination strategies that account for age-specific transmissibility and susceptibility merit study and consideration.

Introduction: Monkeypox (mpox) has re-emerged as a significant global public health threat. This study developed a practical, rapid early-warning tool to quantify country-level mpox importation risk and identify potential sources of cross-border transmission. Methods: We obtained weekly confirmed mpox case data and international passenger volume data spanning January 2022 through November 2025. We constructed an importation risk index (RI) that integrated epidemic intensity in source countries, passenger flows, and public health response capacity. Using a simplified ordinary differential equation framework, we derived a regional RI to estimate time-varying growth and acceleration parameters from surveillance time series. We then computed both realtime and baseline RI values and evaluated their lagged correlations with reported case counts. Results: The RI demonstrated consistently strong lagged correlations with subsequent reported case counts. The real-time RI provided earlier alerts in 93.3% of countries, with a mean lead time of 4.23 weeks - approximately 3.7 weeks ahead of the baseline RI. The dominant sources of importation risk varied across countries, suggesting that this model can inform tailored, country-specific prevention and control strategies. Conclusion: This cross-border risk assessment model provides an actionable framework for early warning and source attribution, supporting targeted surveillance and proactive allocation of public health resources. These findings underscore the critical importance of global health collaboration and timely data sharing.

BACKGROUND:COVID-19 non-pharmaceutical interventions disrupted respiratory syncytial virus (RSV) transmission globally. In Australia, RSV surged back after restrictions eased, whether this reflected immunity debt repayment or epidemic amplification is untested. METHODS:We conducted a national interrupted time-series analysis using Australian Institute of Health and Welfare hospitalisation data from financial years (FY) 2010/11-2023/24. The primary outcome was RSV-attributable fraction (RSV-AF) of bronchiolitis in infants and children 1-4 years. Pre-pandemic counterfactual trends were estimated by linear extrapolation of FY2010/11-2019/20 training data, this was appropriate given the stable, low-gradient pre-pandemic RSV-AF trends in both groups. For adults ≥ 65 years, a flat counterfactual (pre-pandemic mean) to bound estimates where a secular ascertainment rise made linear extrapolation inappropriate. Formal amplification test: one-sided Z-test of whether rebound exceeded suppression deficit. Influenza-attributable fraction used as a pathogen-specific comparator to isolate RSV from general respiratory rebound. Non-RSV bronchiolitis decomposed to rule out general post-pandemic denominator effects. RESULTS:In infants, pre-pandemic RSV-AF was 32.3% (SD 1.8%). In FY 2022/23, it reached 52.0% (z = 10.78; p < 0.001), 19.7 %age points above the pre-pandemic upper control limit. RSV bronchiolitis accounted for 95.1% of the total bronchiolitis increase; non-RSV bronchiolitis remained below counterfactual. In 1-4-year-olds, RSV-AF of bronchiolitis reached 49.1%, against a pre-pandemic mean of 23.2%. Formal testing found rebound significantly exceeded deficit in infants, 1-4-year-olds, and 5-19-year-olds (all p < 0.001). Adults 20-64 years showed significantly incomplete discharge (β=0.71, meaning the rebound recovered only 71% of the accumulated suppression deficit; p = 0.020). CONCLUSIONS:The RSV rebound in young children significantly exceeded the suppression deficit, consistent with epidemic amplification potentially reflecting accumulated primary-infection susceptibility in unvaccinated cohorts. After explanations (testing intensity, admission thresholds, demography, and coding changes) were accounted for. The FY2022/23 baseline provides the pre-vaccination reference for evaluating Australia's RSV immunisation program.

OBJECTIVES:This study aimed to investigate scarlet fever epidemiology in Liaoning Province, China (2005-2024), and assess COVID-19 pandemic impacts. METHODS:Individual-level data on scarlet fever cases from the National Notifiable Infectious Disease Reporting System (NNIDRS) were analyzed for temporal, demographic and spatiotemporal features across pre-, peri- and post-pandemic stages. RESULTS:A total of 73,102 scarlet fever cases were reported. Incidence remained stable during the pre-pandemic period, from 2005 (8.9 per 100,000 population per year) to 2019 (10.8), sharply declined during the pandemic period (2020-2022; range: 1.4-0.6), and rebounded markedly post-pandemic (2023: 1.9; 2024: 7.7). Most cases (79.4%) occurred in children under 15 years of age from urban areas. During 2005-2019, incidence significantly increased among children aged 3-6 years (AAPC: 7.4%; 95% CI: 3.0%-11.9%) and in rural populations (AAPC: 4.3%; 95% CI: 1.1%-7.6%). Compared with the pre-pandemic period (2005-2019), the proportion of cases among children aged 7-14 years significantly increased during the post-pandemic period (2023-2024) (31.4% vs. 54.1%, p < 0.01). CONCLUSION:Scarlet fever incidence in Liaoning Province showed a resurgence after the pandemic, with an age distribution shifting toward older children. These evolving epidemiological patterns highlight the need for enhanced, age-targeted surveillance and adaptive prevention strategies.

Background During the COVID-19 pandemic, seasonal influenza virus circulation was heavily suppressed worldwide. In Australia, since the virus re-emerged in 2022, shifts in seasonal influenza patterns have been observed. Both the 2022 and 2023 seasons started earlier than pre-pandemic norms and were categorised as moderate to severe, highlighting the renewed importance of prevention strategies for seasonal influenza. Methods We analysed influenza notification data from the Australian National Notifiable Diseases Surveillance System (2012–2022) and virological surveillance data from the FluNet database (2012–2023). Using generalised additive models, we compared predicted weekly influenza case counts during 2020–2022 with observed counts. Epidemic weeks were detected using a negative binomial threshold, and epidemic onset was estimated with a Bayesian Poisson count detection algorithm. Trends in epidemic magnitude and onset timing across influenza virus types and subtypes were compared for pre-, during, and post-COVID-19 periods. Results Seasonal influenza activity was nearly absent in 2020 and 2021 but rebounded significantly in 2022 and 2023. Epidemic detection confirmed suppressed seasonal influenza circulation during the pandemic. While influenza A subtypes returned to pre-pandemic onset timings in 2022, influenza B exhibited a significantly delayed onset. The 2022 and 2023 seasons were moderate to severe, with earlier-than-average season starts, underscoring the ongoing changes in influenza dynamics post-pandemic. Conclusions This study provided a detailed analysis of the disruptions and subsequent shifts in seasonal influenza patterns in Australia during and after the COVID-19 pandemic. The rapid resurgence of influenza activity in 2022 and 2023, combined with altered onset timings, highlights the importance of ongoing surveillance and adaptive forecasting models to address the evolving complexity of influenza epidemiology in the post-pandemic era.

Background and Aims:Human psittacosis is a zoonotic disease mainly transmitted by contact with birds. However, the recently identified potential human-to-human transmission indicated an emerging threat of this disease. This study aimed to estimate the incubation period of confirmed psittacosis cases. Methods:We retrospectively collected epidemiological information on confirmed psittacosis cases in Hangzhou City, Zhejiang Province, China, during 2022-2024. Incubation period distributions were estimated from interval-censored exposure and illness onset data using Bayesian parametric models and the Turnbull nonparametric maximum likelihood method. Results:The median incubation period of confirmed psittacosis cases was 11.3 (95% CrI: 8.5-14.2) days with a 95th percentile of 21.8 (95% CrI: 17.8-28.3) days. Conclusion:These estimates could inform clinical diagnosis and would be an important input into mathematical models.

In resource-limited areas, severe shortages of radiologist contribute to high rates of missed pulmonary tuberculosis (PTB) cases when relying solely on conventional chest X-ray (CXR). Although artificial intelligence (AI)-powered computer-aided detection (CAD) has shown effective in PTB diagnosis, its real-world clinical utility and scalability in primary healthcare settings remained underexplored. To evaluate the real-world performance of CAD technology for triaging of PTB patients in primary healthcare facilities in high-incidence areas, and to assess its potential for optimizing radiological resource allocation. We conducted a retrospective paired-design diagnostic accuracy study using CXR images collected from 7 county- and 32 township-level healthcare facilities in Yichang city between 2022 and 2024. All images were retrospectively reprocessed with CAD software (JF CXR-1 v2), and the original radiology reports interpreted by radiologists were extracted. CAD and radiologist performances were compared using two primary evaluation indicators: diagnostic yield among diagnosed cases (DYD) and positive predictive value (PPV). Subgroup analysis (by region, age, sex, healthcare facility tier, and patients category) and sensitivity analysis were conducted to assess the robustness of the results. Among 93,319 enrolled study patients (including 273 bacteriologically confirmed PTB cases), CAD demonstrated a substantially higher DYD (83.88%, 229/273) than radiologists (25.64%, 70/273), though with much lower PPV (1.70% vs. 10.31%). This high-sensitivity performance achieved an 85.52% reduction (only 13,515 instead of 93,319 CXRs) in radiologist workload via selective review of CAD-positive images, without missing any radiologist-identified PTB cases. Furthermore, among CAD-positive images, probability scores >0.75 was a key threshold for identifying high-risk PTB patients, prioritizing for radiologist review. Subgroup analysis further revealed that CAD outperformed radiologists in identifying PTB cases across all scenarios, despite some heterogeneity. In township healthcare facilities, CAD demonstrated significantly better performance than in county-level facilities, with DYD of 86.72% and 62.50%, and PPV of 2.00% and 0.65%, respectively. CAD technology demonstrates valuable PTB screening performance in primary healthcare facilities. Combined with a tiered "AI pre-screening with selective human review" strategy, this approach effectively alleviates workload of radiologist in resource-constrained regions, offering a scalable solution for tuberculosis prevention and control.

Background: Several studies have examined the effect of non-pharmaceutical interventions (NPIs) on COVID-19 and other infectious diseases in Australia and globally. However, to our knowledge none have sufficiently explored their impact on other infectious diseases with robust time series model. In this study, we aimed to use Bayesian Structural Time Series model (BSTS) to systematically assess the impact of NPIs on 64 National Notifiable Infectious Diseases (NNIDs) by conducting a comprehensive and comparative analysis across eight disease categories within each Australian state and territory, as well as nationally. Methods: Monthly data on 64 NNIDs from eight categories were obtained from the Australian National Notifiable Disease Surveillance System. The incidence rates for each infectious disease in 2020 were compared with the 2015-2019 average and then with the expected rates in 2020 using a BSTS model. The study investigated the causal effects of 2020 interventions and analysed the impact of government policy restrictions at the national level from January 2020 to December 2022. Results: During the COVID-19 pandemic interventions in Australia, there was a 38 % (95 % Credible Interval [CI] [9 %, 54 %]) overall relative reduction in incidence reported across all disease categories compared to the 2015-2019 average. Significant reductions were observed in bloodborne diseases: 20 % (95 % CI [10 %, 29 %]), respiratory diseases: 79 % (95 % CI [52 %, 91 %]), and zoonoses: 8 % (95 % CI [1 %, 17 %]). Conversely, vectorborne diseases increased by 9 % over the same period. Reductions and intervention effects varied by state and territory, with higher policy stringency linked to fewer cases for some diseases. Conclusions: COVID-19 NPIs also impacted the transmission of other infectious diseases, with varying effects across regions reflecting diverse outcomes in response strategies throughout Australia. The findings could inform public health strategies and provide scientific evidence to support the development of early warning systems for future disease outbreaks. (c) 2025 Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Purpose We are conducting a longitudinal cohort study—the Community Burden of Acute Respiratory Infections in Shanghai—to assess age-stratified incidence, healthcare utilisation and risk factors of influenza virus, respiratory syncytial virus (RSV) and SARS-CoV-2 associated acute respiratory infections (ARIs) in Shanghai, China.Participants Study participants were enrolled by family doctors in all 47 community health services centres in Pudong New Area District, Shanghai, China. All permanent residents 6 months and older living in Pudong for at least 6 months were eligible for enrolment; residents who planned to leave Pudong for more than 1 month in the first study year were excluded. During enrolment, study staff conducted baseline assessments of sociodemographics, underlying medical conditions, vaccination history and household and self-rated health status. Study participants are being followed for ARIs for 3 years. Nasopharyngeal and oropharyngeal swab specimens are being obtained from suspected ARI cases. Influenza virus, RSV, SARS-CoV-2 and other respiratory pathogens are tested for by multiplex respiratory pathogen real-time quantitative PCR assays. Illness courses and clinical recoveries of ARI cases are assessed through weekly contact with ARI cases for 28 days post ascertainment.Findings to date Between 14 October 2024 and 22 November 2024, we enrolled 5387 community residents into the cohort, including 233 children aged from 6 months to 2 years, 278 preschool children aged 3–6 years, 575 school-age children aged 7–18 years, 2150 adults aged 19–64 years and 2151 older adults aged 65+years. All finished baseline assessment and started follow-up. Surveillance of ARI symptoms, collection of specimens and laboratory testing are ongoing.Future plans Findings from this study will be used to provide valuable scientific data to inform ongoing control efforts and future pandemic preparedness for respiratory diseases in China. Planned analyses include analysis of annual pathogen-specific incidence by age group and exploration of healthcare seeking behaviour and factors associated with ARIs and severe ARIs. We will also assess transmission dynamics of common respiratory pathogens in a household transmission subcohort.

Background Patients with chronic respiratory disease (CRD) face an increased risk of severe influenza complications. However, limited studies offer estimates of vaccine effectiveness (VE) against influenza within these CRD populations in China.Methods A multicenter, retrospective, test-negative, case-control study was conducted to estimate the VE in 37 medical institutions in Shanghai, China, during the 2023/2024 and 2024/2025 seasons. We included patients with CRD who presented with acute respiratory infections and received nucleic acid amplification tests and/or rapid antigen tests. Patients with CRD with a positive test were assigned to the case group, and those with a negative test were assigned to the control group. Multivariable unconditional logistic regression was used to control potential confounders and to determine 95% confidence intervals (CIs).Results A total of 10 711 participants, including 1650 influenza cases (5.8% vaccinated) and 9061 influenza-negative controls (7.9% vaccinated), were eligible for analysis. The combined VE over the 2 seasons was estimated to be 44.23% (95% CI: 30.45-55.75) for the study population. The VE was 42.91% (95% CI: 27.13-55.78) in the 2023/2024 season and 51.51% (95% CI: 17.70-73.61) in the 2024/2025 season. The combined VE for influenza subtypes A and B were 38.49% (95% CI: 21.75-52.27) and 62.64% (95% CI: 39.91-78.28), respectively.Conclusions Influenza vaccination provides consistent and moderate protection to patients with CRD against medically attended influenza, regardless of the dominant circulating subtypes. Nevertheless, vaccination coverage remains suboptimal, underscoring the need to improve annual influenza vaccination uptake among patients with CRD , even amid ongoing viral antigenic evolution through shifts and drifts.

Background:In resource-limited areas, severe shortages of radiologists contribute to high rates of missed pulmonary tuberculosis (PTB) cases when relying solely on conventional chest X-ray (CXR). Although artificial intelligence-powered computer-aided detection (CAD) has proven effective in PTB diagnosis, its real-world performance remains underexplored. Objective:This study aimed to evaluate the real-world diagnostic yield of CAD technology as a triage tool for detecting PTB in primary health care facilities in high-burden areas. Methods:We conducted a retrospective paired-design diagnostic yield study using CXR images collected from 7 county- and 32 township-level health care facilities in Yichang city between 2022 and 2024 year. All images were retrospectively reprocessed with CAD software (JF CXR-1), and the original reports interpreted by radiologists at the time of patient admission were extracted. CAD and radiologist performances were compared using 2 primary evaluation indicators-diagnostic yield among diagnosed cases (DYD) and positive predictive value (PPV). Subgroup analysis (by region, age, sex, health care facility tier, and patient category) and sensitivity analysis were conducted to assess the robustness of the results. Results:Among 93,319 enrolled study patients, including 273 (0.3%) bacteriologically confirmed PTB cases, CAD demonstrated a substantially higher DYD (229/273, 83.9%) than radiologists (70/273, 25.6%), although the PPV was much lower (1.70% vs 10.31%). This high-sensitivity performance achieved an 85.5% (79,804/93,319) reduction (only 13,515 instead of 93,319 CXRs) in radiologist workload via selective review of CAD-positive images, without missing any radiologist-identified PTB cases. Furthermore, probability scores greater than 0.75 were a key threshold for identifying high-risk patients with PTB, and these patients were prioritized for radiologist review. Subgroup analysis further revealed that CAD outperformed radiologists in identifying PTB cases across all scenarios, despite some heterogeneity. CAD performance was significantly better in township-level medical facilities (DYD: 86.7%; PPV: 2%) than in county-level hospitals (DYD: 62.5%; PPV: 0.6%). Conclusions:CAD technology is valuable for detecting PTB in primary health care facilities. Combined with a tiered artificial intelligence prescreening with selective human review strategy, this approach effectively alleviates the workload of radiologists in resource-constrained regions, offering a scalable solution for tuberculosis prevention and control.

What is already known about this topic?:School-aged children represent a particularly vulnerable population for influenza transmission due to their dense social interactions and limited awareness of protective measures. Since 2019, Shenzhen has provided free influenza immunizations to this demographic, with vaccination campaigns typically initiated during the autumn months. What is added by this report?:This study utilized influenza surveillance data from Shenzhen to develop an age-stratified compartmental model for epidemiological simulations, evaluating the disease burden prevented by influenza vaccinations among school-aged children during the 2023-2024 season. Additionally, an optimization framework was developed to design strategic vaccination schedules while considering the importance of maintaining stable public health policies over time. What are the implications for public health practice?:The findings suggest concentrating vaccination efforts during November and December; however, optimal strategies may vary depending on specific influenza transmission patterns. A more robust approach involves implementing a generalized strategy optimized using historical seasonal data with comparable transmission characteristics.

Early diagnosis of pulmonary tuberculosis (PTB) is essential for individual case treatment and community transmission control. However, the impact of the COVID-19 pandemic on PTB diagnosis remains inadequately understood. In this study, we aimed to investigate the diagnostic delay in patients with PTB before, during and after the COVID-19 pandemic. We conducted a longitudinal study of PTB in Yichang City from 2005 to 2023, utilizing data from the Tuberculosis Information Management System of China. The distribution of diagnostic delay (DD) was analyzed across three periods: pre-pandemic, during the pandemic, and post-pandemic. Multivariate mixed-effects logistic regression models were employed to identify factors associated with prolonged DD, defined as a delay exceeding 28 days. A total of 58,774 patients with PTB were included in this study. The average annual number of cases was 3,293 pre-pandemic, 2,319 during the pandemic, and 2,426 post-pandemic. The fitted median DD in the pre-pandemic period (31.7 days, interquartile range [IQR] = 13.8–72.8) was significantly longer than that in the pandemic period (23.8 days, IQR = 11.3–50.3) and the post-pandemic period (20.6 days, IQR = 9-47.1) (p < 0.01). Elder patients aged 65 years and older had a longer median DD (32 days, IQR = 14.2–72.0) than patients aged 18–64 years (median: 30.1 days, IQR = 13.1–68.9) and patients under 18 years (median: 19.5 days, IQR = 8.6–44.2) (p < 0.01). Patients residing in rural areas also had a longer median DD (31 days, IQR = 14.2–72.0) compared to those in urban (median: 29.4 days, IQR = 13.7–70.2) (p < 0.01). Older age (adjusted Odds Ratio [aOR] = 2.20, 95

Background: Seasonal influenza is a major global public health concern, leading to escalated morbidity and mortality rates. Traditional early warning models rely on binary (0/1) classification methods, which issue alerts only when predefined thresholds are crossed. However, these models exhibit inflexibility, often leading to false alarms or missed warnings and failing to provide granular risk assessments essential for decision-making. Therefore, we propose a probability-based early warning system using machine learning to mitigate these limitations and to offer continuous risk estimations of alerts (0-1 variable) instead of rigid threshold-based alerts. Based on probabilistic prediction, public health experts can make more flexible decisions in combination with the actual situation, significantly reducing the uncertainty and pressure in the decision-making process and reducing the waste of public health resources and the risk of social panic. Objective: The main aim of this study is to devise an innovative approach for early warning systems focused on influenza-like cases. Therefore, a Dense Residual Network (Dense ResNet), a supervised deep learning model, was developed. The model's training involved fitting the influenza-like illness positive rate, which enabled the early detection and warning of signals of changes occurring in the activity level of influenza-like cases. This departure from conventional methodologies underscores the transformative potential of machine learning, particularly in providing advanced capabilities for timely and proactive warnings in the context of influenza outbreaks. Methods: We developed a Dense ResNet machine learning model trained on influenza surveillance data from Northern and Southern China (2014-2024). This model generates early warning signals 3, 5, and 7 days in advance, providing a probability-based risk assessment represented as a continuous variable ranging from 0 to 1, in contrast to the traditional binary (0/1) warning systems. We evaluated the performance of this model using area under the curve scores, accuracy, recall, and F1-scores, then compared it with support vector machine (SVM), random forests, XGBoost (Extreme Gradient Boosting), and LSTM (long short-term memory) models. Results: The Dense ResNet model demonstrated the best performance, characterized by 5-day lead warnings and a 50th percentile probability threshold, achieving area under the curve scores of 0.94 (Northern China) and 0.95 (Southern China). Relative to traditional models, probability-based warning signals improved early detection, reduced false alarms, and facilitated tiered public health responses. Conclusions: This study presented a novel probability-based machine learning model essential for early warning signals of influenza, demonstrating superior accuracy, flexibility, and practical applicability compared to other techniques. This approach enhances preparedness for influenza among the population and promotes the use of automated artificial intelligence-driven public health responses by replacing binary warnings with probability-driven risk assessments. Future research should integrate real-time surveillance data and dynamic transmission models to improve the precision of early warning.

Dengue fever poses significant public health concern in tropical and subtropical regions, including areas of China-Myanmar border, in which the population-representative evidence on dengue seroprevalence is limited. A community-based cross-sectional survey was conducted in December 2015 in Ruili, a border city in southwestern China. Participants were selected using a stratified random sampling method. Demographic information, previous clinical and travel history, and environment-related factors regarding dengue were collected by structured interview, with serum samples tested for IgG antibodies using ELISA. Age- and gender-weighted seroprevalence was estimated based on the local demographic distribution. Associated factors for seropositivity were analyzed using logistic regression. The study included 1,616 participants aged 1–95 years, with 58.42