Multi-Source Agent-Based Modeling to Optimize Influenza Mitigation Strategies in Hong Kong

Multi-Source Agent-Based Modeling to Optimize Influenza Mitigation Strategies in Hong Kong | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Multi-Source Agent-Based Modeling to Optimize Influenza Mitigation Strategies in Hong Kong Tim Tsang, Liping Peng, Yiyang Guo, Nicole Tsang, Xiaotong Huang, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7574768/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Seasonal influenza control faces significant challenges from variable vaccine effectiveness and uncertain non-pharmaceutical intervention (NPI) performance across diverse contexts. While vaccination remains the primary strategy, effectiveness varies substantially between vaccine-matched and mismatched seasons. The COVID-19 pandemic has increased public acceptance of NPIs such as staying home when sick and mask use, enhancing feasibility for influenza control, yet optimal combination approaches remain poorly understood. We used an agent-based model to analyze influenza transmission across six seasons in Hong Kong (2009-2013) with varying vaccine effectiveness and epidemic characteristics. We integrated surveillance, serological, and school absenteeism data for calibration, enabling accurate estimation of reported and unreported infections. We evaluated age-targeted vaccination, staying home when sick, mask use, and school-based interventions across diverse real-world scenarios. Comparing with actual coverage levels, child vaccination consistently outperformed other strategies, with targeting those under 12 yielding the greatest population-level attack rate reduction (up to 2.17 percentage points absolute reduction and 8.5% relative reduction per 100,000 vaccinated individuals with coverage increases up to 50 percentage points). Among NPIs, 40% mask coverage reduced attack rates by 18-45%, comparable to 25% of symptomatic individuals staying home (17-49% reduction), though mask effectiveness remained stable while staying home declined with higher asymptomatic proportions. During vaccine-mismatched seasons, combining high-coverage mask use and staying home can reduce attack rates by 77-84%. School-based vaccination at high coverage was more effective than closures, reducing student attack rates by up to 86% versus 31% for 14-day closures. Our multi-source calibration approach provides robust evidence for prioritizing child vaccination and strategic NPI combinations. Health sciences/Medical research/Epidemiology Health sciences/Diseases/Infectious diseases influenza agent-based model vaccination staying home when sick school closure Mask-wearing Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 BACKGROUND Influenza remains a major global public health threat, particularly in subtropical and tropical regions, which typically experience multiple seasonal outbreaks each year 1 . After the COVID-19 pandemic, influenza has experienced a resurgence worldwide 2 . Hong Kong, a subtropical region, is significantly affected by seasonal influenza, with peaks commonly occurring from January to March/April and from July to August. In the post-COVID-19 period, influenza transmission in Hong Kong has returned to pre-pandemic levels, with outbreaks lasting even longer, up to 28 weeks, in 2024 3 . Vaccination and non-pharmaceutical interventions (NPIs) are key tools against influenza, but their uptake and overall effectiveness remain uncertain. In Hong Kong, free seasonal influenza vaccination 4 and a government subsidy program 5 were introduced after the 2009 H1N1 pandemic for high-risk groups, including children under 6 and adults aged 65 and above. However, coverage remained low before 2014, with uptake rates of 12.9% and 32.7%, respectively 6 . The program has since expanded to include children under 18 and adults over 50. However, low perceived risk, safety concerns 7 , and shifts in public attitudes influenced by COVID-19 vaccination policies 8 may have hindered progress in increasing influenza vaccine uptake in Hong Kong. Vaccine effectiveness is also challenged by mismatches with circulating strains 9 . Between 1996 and 2012, fewer than half of H3N2 seasons in Hong Kong had vaccine strains closely matching circulating viruses 10 , resulting in reduced protection. Non-pharmaceutical interventions (NPIs) such as social distancing 11 , mask use 12 , and school closures 13 can reduce transmission, with COVID-19 experience having increased public acceptance and familiarity with measures such as mask use and work from home 14 . However, implementation remains constrained by economic costs 15 , logistical challenges 16 , and varying adherence across contexts 17 , underscoring the need for tailored strategies 18 . Given these challenges with individual interventions, agent-based models 19 , offer valuable tools to assess optimal combination strategies by simulating transmission at the individual level. However, model reliability depends critically on calibration approaches that accurately capture true transmission dynamics. Common methods using surveillance data alone often underestimate infections due to underreporting and asymptomatic cases 20 , 21 , 22 . These limitations necessitate integrated calibration using multiple data sources, including serological data 23 , to improve model calibration and better capture true transmission dynamics. To support the WHO’s Influenza Strategy 2030 24 , we adapt an agent-based model to simulate influenza transmission in Hong Kong across six seasons (2009–2013). By integrating surveillance, serological, and school absenteeism data, we reconstruct baseline age-specific transmission dynamics and evaluate targeted interventions, including age-specific vaccination, staying home when sick, mask use, and school-based measures. This study provides context-specific evidence to guide seasonal influenza control and strengthen preparedness for future outbreaks. RESULTS Overviews This study aimed to identify optimal strategies for reducing influenza attack rates in Hong Kong across diverse seasonal epidemic conditions, particularly vaccine-matched versus vaccine-mismatched scenarios. We evaluated the effectiveness of pharmaceutical and non-pharmaceutical interventions in reducing overall and age-specific infection rates during six influenza seasons (2009–2013), which varied in circulating virus subtypes, vaccine effectiveness, epidemic magnitude, and seasonal timing. Our analytical framework prioritized comparative assessment of targeted vaccination strategies, staying home when sick, mask use, and school-based interventions, both individually and in strategic combinations, to determine strategy effectiveness for different scenarios. We used a stochastic agent-based model 25 to estimate influenza transmission dynamics in Hong Kong. The model integrates diverse data streams to capture the heterogeneity of social contacts and infection progression within the population. A synthetic population was constructed using detailed demographic and social structure data, including age-stratified population figures, employment rates, enrolment statistics, and school capacity records, which informed the creation of realistic contact networks across multiple layers (households, schools, workplaces, and community settings). Infection time series derived from serological data, which captures both reported and unreported infections, avoiding the healthcare-seeking bias inherent in clinical surveillance, along with school absenteeism data were used for model calibration, with the latter enhancing the representation of contact dynamics among school-age children. Epidemiological characteristics of the six influenza seasons The six modeled seasons (2009–2013) exhibited distinct epidemiological profiles (Table 1 ). Seasons 1, 2, and 6 coincided with summer holidays and showed predominantly H1N1 (season 1) or H3N2 (seasons 2 and 6) circulation. Seasons 3, 4, and 5 occurred outside summer holiday periods, with seasons 3 and 5 featuring well-matched vaccines and season 4 experiencing vaccine mismatch. Cumulative incidence of infections derived from serological data ranged from 0.29 million (season 6) to 1.25 million (season 4), with peak daily case numbers between 4,518 (season 5) and 27,095 (season 4). The 0–24 age group had the highest or near-highest attack rates in five of six seasons, ranging from 4.5–34.6%. Table 1 Information on the six modeled seasons Season Season 1 Season 2 Season 3 Season 4 Season 5 Season 6 Start date 2009/06/20 2010/06/25 2010/12/04 2012/02/20 2013/01/01 2013/05/31 End date 2009/11/21 2010/10/16 2011/02/26 2012/06/23 2013/04/20 2013/10/12 Summer holiday Yes Yes No No No Yes Variant H1N1 H3N2 H1N1 H3N2 H1N1 H3N2 Vaccine matching Mismatched Matched Matched Mismatched Matched Matched Cumulative infections (million) 1.24 0.96 0.75 1.27 0.30 0.29 Peak case number 27,090 18,127 21,456 27,095 4,518 4,532 Attack rate (%) 0–24 34.6 12.4 11.5 21.3 4.9 4.5 25–44 14.4 13.5 12.2 16.0 4.0 5.2 45–64 10.7 14.3 9.4 17.0 4.0 3.4 ≥ 65 9.4 14.9 8.7 17.2 4.0 2.7 Peak influenza-related absences (per 100,000 students aged 6–18) 47 12 14 7 3 3 Smart-card based school absenteeism monitoring revealed distinct temporal patterns that closely tracked influenza activity (Fig. 2 c). During high-transmission periods, peak influenza-related absenteeism among students aged 6–18 ranged from 30 to 470 per 100,000 student-days across seasons. Over the course of each influenza season, average daily absenteeism ranged from 8 to 53 per 100,000 student-days (Table 1 ). The calibrated baseline model demonstrated an acceptable fit (Fig. 2 ), achieving good agreement with observed data across all target metrics. Across the six seasons, an estimated 1.8–5.0% of symptomatic individuals stayed home due to illness. We estimated a higher relative susceptibility among older adults (≥ 65 years), with odds ratios of 1.4 to 3.1, compared to younger age groups (Supplementary Table 1). Age-targeted vaccination Across six seasons, vaccinating children under 12 consistently resulted in the largest reductions in population attack rate, with absolute reductions ranging from 0.52 to 2.17 percentage points and relative reductions from 6.4–8.5% per 100,000 vaccinated individuals (Fig. 3 , Supplementary Fig. 2). Targeting individuals under 18 showed moderate effectiveness (absolute reductions: 0.35 to 1.67 percentage points; relative reductions: 5.1–6.8% per 100,000 vaccinated), while vaccination of those aged ≥ 65 yielded smaller reductions (absolute reductions: 0.09 to 1.12 percentage points; relative reductions: 0.5–5.5% per 100,000 vaccinated). Universal vaccination was the least efficient strategy, yielding only modest reductions (absolute reductions: 0.03 to 0.39 percentage points; relative reductions: 2.2–2.6% per 100,000 vaccinated). Further analysis showed that targeting vaccination to children under 12 not only substantially reduced attack rates within this group (absolute reductions: 1.03 to 6.13 percentage points; relative reductions: 7.1%-9.3%, per 100,000 vaccinated), but also conferred indirect protection to adolescents (< 18 years: absolute reductions 0.84 to 4.47 percentage points ; relative reductions: 6.8–9.0% per 100,000 vaccinated) and older adults (≥ 65 years: absolute reductions 0.56 to 2.5 percentage points ; relative reductions 6.0%-8.4% per 100,000 vaccinated) (Supplementary Fig. 3). Targeting individuals under 18 also benefited older adults, though to a lesser extent (absolute reductions: 0.28 to 1.88 percentage points; relative reductions: 4.9%-6.9% per 100,000 vaccinated). In contrast, prioritizing vaccination in adults aged ≥ 65 had limited indirect impact on younger age groups, with reductions among children < 18 observed in three of six seasons (absolute reductions: 0.19 to 0.73 percentage points; relative reductions: 3.4%-4.1% per 100,000 vaccinated) and no significant effect in the remaining seasons. Non-pharmaceutical interventions (NPIs) Increasing the daily probability of adherence to staying home when sick by 10–50% (resulting in 6–31% of symptomatic cases staying home) reduced the attack rate by 2–49% compared to the baseline across all six seasons. In contrast, increasing population mask usage by 10–50% (equivalent to 20–60% of the population using masks) achieved a 8–64% reduction in attack rate (Fig. 4 a). We estimated that population-wide masking (40% coverage, Mask + 30% scenario) reduces influenza transmission as effectively as 20–31% of symptomatic individuals staying home (Stay-home + 50% scenario). Across six seasons, both strategies achieved comparable reductions (18–45% for masking vs. 17–49% for staying home). However, their relative effectiveness varied with asymptomatic infection proportions. While mask effectiveness remained stable, the effectiveness of staying home declined as asymptomatic proportion increased. Under the Stay-home + 50% scenario, increasing the asymptomatic infection probability from 20% (baseline) to 60% resulted in an absolute increase of 0.3–4.0% in the overall attack rate across seasons. (Fig. 4 b). Intervention effectiveness was sensitive to implementation timing, with staying home when sick more affected by delays than mask use. Higher coverage generally preserved effectiveness under short delays, but as delays increased, the difference between coverage levels narrowed (Supplementary Fig. 4). Combination strategies during mismatched seasons During seasons with vaccine mismatch and low vaccination coverage (Seasons 1 and 4), combining vaccination with NPIs was essential. In these two seasons, a 50% increase in the daily probability of adherence to staying home when sick (resulting in 21–31% of symptomatic cases staying home) reduced the attack rate by 41–47% (Fig. 5 ). When both staying home when sick and mask use were intensified simultaneously, each increasing by 10–50% over baseline, the attack rate was reduced by 22–84% in Season 1 and 26–77% in Season 4. School closure and targeted strategies for school-aged children We assessed three school-based interventions across seasons without summer holidays (seasons 3, 4 and 5) (Fig. 6 ). When coverage increased from 10–50% for each intervention, student vaccination consistently achieved the greatest reductions in both student attack rates (56–86%) and overall population attack rates (33–78%). Staying home when sick among symptomatic students showed variable effectiveness, reducing student rates by 23–55% and overall rates by 15–45%. School closures had the smallest impact, with student reductions of 15–31% and overall reductions of 18–25%. Sensitivity analyses showed that the key conclusions remained robust under joint variation of multiple parameters (Supplementary Results, Supplementary Fig. 5–20). The relative effectiveness of interventions and the overall patterns of attack rate reduction were broadly consistent across both lower-bound and upper-bound parameter settings. DISCUSSION We calibrated the model using infection time series from surveillance and serological data, and incorporated school absenteeism to inform school-based intervention. By simulating six influenza seasons in Hong Kong (2009–2013) across varied virus subtypes and vaccine matches, our model quantifies the differential impact of vaccination and non-pharmaceutical interventions. Child vaccination consistently reduced transmission, underscoring children’s pivotal role as infection amplifiers in dense urban settings and supporting targeted immunization policies. Among non-pharmaceutical measures, mask use showed robust and stable effectiveness comparable to moderate levels of staying home when sick, highlighting its value as a reliable control strategy. Importantly, during vaccine-mismatched seasons, combined interventions were crucial to maintaining epidemic control. Furthermore, school-based vaccination demonstrated superior effectiveness and sustainability compared to reactive closure strategies. Vaccination strategies targeting children consistently outperformed alternatives, reflecting both strong direct protection and substantial indirect benefits at population level. This finding is well-supported by modeling studies from the US and UK, and notably by a community-based trial in Hong Kong where increasing child coverage halved child attack rates while also reducing adult rates, demonstrating important indirect protection 26 – 28 . Children are highly exposed in school settings and typically exhibit stronger immune responses to influenza vaccines than adults, leading to higher effectiveness 29 . Vaccinating this group also reduces onward transmission, as children play a central role in community spread due to their high contact rates 30 and elevated transmissibility 31 , These findings support the expansion of Hong Kong’s Seasonal Influenza Vaccination (SIV) School Outreach Programme to include adolescents under 18, which proved more efficient than universal or elderly-targeted strategies. As the benefits of child vaccination are largely indirect, uptake may remain suboptimal without targeted incentives 32 . Beyond vaccination, other school-based interventions provide moderate mitigation. Encouraging symptomatic students to stay home reduces transmission meaningfully, consistent with CDC recommendations showing modest reductions under partial compliance 33 , emphasizing the importance of promoting voluntary. In contrast, short-term reactive school closures produced limited effects in our simulations, aligning with systematic reviews showing that brief closures achieve only modest reductions 34 , unless repeated or extended under high compliance 35 , 36 . Given their limited epidemiological benefit and considerable social and economic costs, school closures should be carefully weighed against more targeted and sustainable interventions. Mask use provides population-level control that does not rely on symptom recognition or individual adherence decisions. Even widespread but not universal use can meaningfully reduce transmission through network effects, where preventing early transmission events disproportionately constrains total outbreak size 37 . Laboratory and meta-analytic studies support substantial per-contact efficacy of surgical or cloth masks 12 , 38 . Compared to stay-home policies, masks maintain more consistent effectiveness across seasons because they protect regardless of symptom awareness. However, effectiveness still varies with baseline transmissibility and intervention timing 37 , 39 . Our simulations indicate that voluntary stay-home policies among symptomatic individuals can achieve meaningful transmission reductions, even with modest voluntary levels. While baseline compliance is typically low due to mild symptoms and limited paid sick leave 40 , increasing participation to feasible levels produced notable mitigation, consistent with workplace studies showing substantial infection reductions with partial sick leave uptake 41 , and community simulations demonstrating comparable benefits 42 , 43 . However, financial disincentives such as income loss 44 often discourage adherence, especially in the absence of supportive policies. The widespread adoption of remote work arrangements following the COVID-19 pandemic may have increased the feasibility of staying home when sick, particularly for office workers. However, when economic constraints prevent staying home, public health authorities may need to recommend less effective but more accessible alternatives such as mask use. Unlike mask use, which provides population-wide protection regardless of infection awareness, staying home when sick depends on individuals recognizing symptoms and voluntarily avoiding contact with others. Its effectiveness is therefore influenced by the timeliness of symptom detection and the consistency of individual adherence. The presence of asymptomatic infections, which are common in influenza and vary in estimated prevalence across studies 45 , further limits the effectiveness of staying home when sick. Without complementary strategies such as contact tracing, this strategy remains modest in effect and operationally challenging to implement. Vaccine mismatch seasons, common in influenza epidemics 9 , 46 , highlight the need for early and sustained non-pharmaceutical interventions to control transmission effectively. This aligns with previous studies showing that vaccination alone is insufficient under high-transmission conditions and that multiple interventions are needed to suppress 47 . However, our assumption of constant adherence likely overestimates real-world impact, as behavioral responses often weaken over time 48 , 49 . Moreover, interactions between interventions—such as reduced NPI adherence after vaccination or lower vaccine uptake when NPIs are widespread—may influence overall impact 50 . Future models should account for these behavioral dynamics to better inform integrated strategies. This study offers several methodological strengths. First, analyzing six influenza seasons under varying epidemiological conditions enhanced the generalizability of our findings. Second, calibration using serology-based attack rates, rather than clinical surveillance data alone, enabled more accurate estimation of total infections, including asymptomatic cases. While some prior studies incorporated serological data 51 52 , these focused exclusively on the 2009 H1N1 pandemic and used serology only to estimate seasonal attack rates without integrating surveillance data, lacking temporal resolution and sometimes using data from different seasons or subtypes. In contrast, we utilized contemporaneous, population-specific serological data integrated with surveillance systems to reconstruct age-specific epidemic curves, capturing both reported and unreported infections with daily resolution. Third, incorporating school absenteeism data enabled us to model illness-related behaviors in school-aged children, a key driver of influenza transmission, and to establish a more realistic behavioral baseline often omitted in models. This comprehensive approach improved model realism and strengthened the reliability of intervention impact assessments. Several limitations should be noted. First, our analyses focused on infections and did not capture clinical severity or mortality, limiting its ability to reflect total disease burden. Second, antiviral treatment was not considered in the analyses due to low uptake in Hong Kong. Third, some parameters, such as the probability of symptomatic infection, were not age-specific, potentially underestimating population heterogeneity. Mask use and health-seeking were assumed constant over time, which may not reflect real-world dynamics. Lastly, the model did not capture behavioral interactions between interventions, such as reduced NPI adherence following vaccination. In summary, this study used a multi-season, age-structured model calibrated with serological, surveillance and school absenteeism data to evaluate vaccination and NPIs for influenza control in Hong Kong. By integrating diverse data sources and capturing behavioral and seasonal variability, the model provides a robust tool for evaluating targeted strategies. The findings support child-targeted vaccination and emphasize the complementary role of NPIs, particularly during vaccine-mismatched seasons. School-based interventions further highlight the advantages of proactive measures like student vaccination over reactive strategies such as closures. This framework offers actionable insights for influenza control and broader respiratory preparedness. METHODS Data sources Data for this study were obtained to capture a comprehensive picture of influenza activity over six seasons in Hong Kong (2009–2013) (Table 1 ). Multiple, high-quality sources provided the basis for the diverse datasets used in this analysis. Demographic and social structure data, including age-stratified population figures, employment statistics, enrolment numbers, and school capacity records were sourced from governmental agencies and public institutions. These datasets offered detailed insights into the population structure and social mixing patterns in Hong Kong, to generate the contact matrix. Influenza activity was monitored through surveillance systems that recorded the percentage of outpatient visits attributed to influenza-like illness (ILI) alongside the proportion of laboratory-confirmed influenza cases from public health laboratories. Weekly influenza activity proxy was derived by multiplying these two indicators. Serology data were collected from two community-based randomized controlled trials (RCTs) for evaluating direct and indirect benefits of influenza vaccination 53 , 54 . In the RCTs conducted in 2008/09 and 2009/ 10, 119 and 796 households were recruited. Serum specimens were collected at the start of the study, and after 6 and 12 months from all participants. In the sub- sequent observational follow-up of the same cohort participants from late 2010 to late 2013 without intervention 55 , serum specimens were collected from all participants in each autumn (October to December), and also each spring (April to May). Receipt of influenza vaccine outside of the trial was recorded annually. School absenteeism data were collected through a smartcard-based monitoring system covering 66 primary and 41 secondary schools across all 18 districts of Hong Kong, encompassing 75,052 students 56 . This system tracked daily attendance patterns and class sizes, providing data on all-cause absenteeism. To estimate influenza-attributable absenteeism, we adjusted the all-cause absenteeism data using ILI consultation rates and specimen positivity rates from the established surveillance systems. Intervention-related data, including effectiveness, baseline coverage, and implementation timing of various interventions were obtained through a literature review and from official government sources (Supplementary Methods, Supplementary Table 4). Model details Covasim was originally developed for SARS-CoV-2, hence, we refined the model to capture the transmission characteristics and epidemiology of influenza. It explicitly represents the progression of the disease through distinct states within the population. In the model, individuals transition from a susceptible state to exposed, then progress to either symptomatic or asymptomatic infection, with an 80% probability of developing symptoms, before ultimately recovering (Fig. 1 ). To mirror the approximately 7 million inhabitants of Hong Kong, we constructed a synthetic population of about 70,000 agents using a scale-up factor of 100 to reduce complexity while maintain accuracy 25 . This population was designed to reflect Hong Kong’s demographic composition by incorporating comprehensive data, such as age distributions, employment statistics, and school enrolment figures. These data underpin the development of realistic contact networks that span households, schools, workplaces, and community settings (Supplementary Methods and Supplementary Table 2), capturing the heterogeneous nature of social interactions in an urban environment. Initial parameter values were sourced from existing literature and established assumptions. The transmission model incorporates empirically-derived parameter values obtained through literature review and calibration to Hong Kong surveillance data. This includes age-specific susceptibility profiles, transmissibility factors, and disease progression timelines. Importantly, the model integrates actual NPI measures and documented vaccine coverage data from the study period. By incorporating real-world values for interventions such as staying home when sick and mask use, alongside empirical vaccination coverage across various age groups, the model accurately mirrors the public health strategies implemented during the observed influenza seasons. Detailed numerical assumptions and parameter values governing these processes are summarized in Supplementary Table 1. Calibration The calibration target was the age-specific infection time series for six influenza seasons. To reconstruct age-specific infection time series for six influenza seasons, we combined this activity proxy with age-specific cumulative incidence estimated by method in Tsang et al. 57 with using serological data, which can relax the 4-fold rise assumption. To capture the unique epidemiological dynamics of influenza in Hong Kong, our calibration process focused on several key parameters, while fixing others using the best available epidemiological evidence to prevent overfitting and ensure model identifiability. The calibrated parameters included the initial number of infections, the per-contact transmission probability, age group relative susceptibility, and the probability of staying home when sick among symptomatic individuals who seek healthcare. By systematically exploring a wide range of parameter sets, we minimized the mean squared error between the observed data and the model’s daily projections. Specifically, the calibration targeted three outcomes: cumulative infections per day, daily new infections stratified by age group, and influenza-related absences among primary and secondary school students (aged 6–18). We employed the Optuna 3.2.0 hyperparameter optimization framework in Python and conducted 40,000 simulation runs for each influenza season to ensure robust parameter estimation. Intervention Strategies We systematically assessed age-targeted vaccination approaches, staying home when sick, community mask-wearing, and school-based interventions. Table 2 summarized the strategy specifications and parameter configurations employed in our primary analysis. Table 2 Intervention strategies Intervention Target Contact layers Intervention level Efficacy Baseline coverage Scenarios Vaccination • < 12 y • < 18 y • ≥ 65 y • Both < 18 y and ≥ 65 y • Universal All Individual Matched: 70% Mismatched: 30% • 0–5 y: 7.5%-12.9% • 6–11 y: 10.8%-18.4% • 12–17 y: 5.1%-8.8% • 18–39 y:2.7%-3.3% • 40–61 y: 8.4%-9.8% • ≥ 65 y: 28.1%-32.7% Coverage + 10% increments from baseline Staying home when sick Symptomatic cases School, workplace, and community Individual 80% • Probability of symptomatic presentation: 80% • Daily healthcare-seeking probability: ≤15: 18.1%, 16–54: 6.5%, and ≥ 55: 9.5% • Daily stay-at-home adherence probability: calibrated Daily adherence probability + 10% increments from baseline Mask use Universal School, workplace, and community Population (contact layer level) 25% 10% Coverage + 10% increments from baseline School closure School-aged children (3–17 y) School Population (contact layer level) 100% 0% Coverage + 10% increments from baseline; 14-day duration We explored the impact of expanding influenza vaccination coverage across specific population segments, starting from documented baseline coverage levels (Supplementary Table 3). Five distinct targeting approaches were evaluated: 1) children under 12 years; 2) individuals under 18 years; 3) adults aged 65 and above; 4) combined targeting of both young (< 18) and elderly (≥ 65) populations; 5) universal vaccination across all age groups. Additionally, we simulated a school-based vaccination program specifically targeting children aged 3–17 years. Vaccine efficacy parameters reflected documented seasonal variation, with 70% efficacy during antigenically-matched seasons and 30% during mismatched seasons 58 . Immunity was assumed to be acquired following vaccination or natural infection. Natural infection conferred complete protection for the remainder of the same season, whereas vaccination provided partial protection, with breakthrough infections allowed and immunity waning at a rate of 14% per year 57 . In the baseline scenario, staying home when sick reflected existing behaviour, where symptomatic individuals remained at home due to illness severity. Enhanced interventions represented increased work-from-home policies and higher public adherence to stay-at-home recommendations. This behaviour was modelled using a probability function: $$\:{p}_{sh}\left(i\right)={p}_{sym}*\left[1-{\left(1-{p}_{hs}\left(i\right)*{p}_{ad}\right)}^{{T}_{rec}-1}\right]$$ where \(\:{p}_{sym}\) is the probability of symptomatic presentation, \(\:{p}_{hs}\left(i\right)\) is the age-specific daily probability of healthcare-seeking behaviour among symptomatic individuals, \(\:{p}_{ad}\) is the daily probability of adherence to stay-at-home behaviour following healthcare-seeking among symptomatic individuals, and \(\:{T}_{rec}\) is the duration from symptom onset to recovery in days. We assumed a one-day delay between symptom onset and initiation of stay-at-home behaviour. Intervention scenarios simulated enhanced adherence by increasing \(\:{p}_{ad}\) relative to the baseline, which was calibrated to observed behaviour. In addition to population-wide measures, we simulated a targeted strategy focusing on school-aged children (aged 3–17), reflecting potential guidance encouraging symptomatic children to stay home until recovery. Staying home when sick reduced transmission in community, workplace, and school settings, while household transmission remained unchanged as infected individuals continued interactions with household members. Mask-wearing was implemented at the contact layer level, modeled as a reduction in per-contact transmission probability across school, workplace, and community settings. The baseline scenario assumed 10% population coverage of mask and 25% per-contact effectiveness 59 . This corresponds to an average 2.5% reduction in transmission probability at the population level, assuming homogeneous mixing. No transmission reduction was applied within households, where mask use was considered impractical. Intervention scenarios progressively increased the proportion of mask users in the population. School closures were modeled by reducing per-contact transmission within the school contact layer. Coverage levels corresponded to proportional reductions in transmission relative to baseline: 0% indicated no closure and 100% represented full closure. Seasons overlapping with the summer holiday period (July 15 to August 31) were excluded, as the holiday inherently reduces school-based transmission. For seasons without such overlap, we modeled 14-day reactive school closures 13 triggered by elevated absenteeism rates among school-aged, defined as daily absenteeism exceeding 200 students. Intervention timing We assumed vaccination occurred prior to influenza season onset. Implementation of staying home when sick and mask use aligned with HK CHP’s announcements of the onset of influenza seasons (Supplementary Table 4). We assessed timing sensitivity by simulating 1–5 week delays in NPIs implementation after season onset. Comparative Analysis Framework To optimize mitigation approaches, we implemented a layered analysis framework that progressively added interventions to baseline vaccination scenarios. This structured approach allowed for systematic assessment of the incremental benefit of each additional public health measure. Specifically, we assessed: 1) age-targeted vaccination strategies; 2) the comparative effectiveness of staying home when sick and mask use; 3) combined NPIs in vaccine-mismatched seasons; and 4) the necessity of school closures by comparing their effectiveness with other school-based interventions—including vaccination and staying home when sick among school-age children—given the substantial societal costs associated with school closures as a mitigation measure 60 . Sensitivity analysis To examine the robustness of our findings under parameter uncertainty, we conducted a multi-way sensitivity analysis by jointly varying key epidemiological parameters to their respective lower and upper bounds (Supplementary Table 1). These parameters included: latent and infectious periods, age-specific transmissibility odds ratios, symptom probability, per-contact transmission weights across settings, daily contact rates, age-specific healthcare-seeking behavior, and the effectiveness of staying home when sick and mask use. Following parameter adjustment, we re-calibrated the model for each extreme scenario and repeated the main analyses using the re-calibrated models. For all analyses, each scenario underwent 30 simulations using the top 30 best-fitting parameter sets from calibration. Results are presented as median estimates with 80% projection intervals 61 to appropriately characterize uncertainty in intervention outcomes. Declarations Data availability The demographic data, social structure data, and weekly influenza activity data used in this study are publicly available from open-access sources, as described in the Supplementary Methods. Serological data and school absenteeism records are available from the corresponding author upon reasonable request. Code availability The code developed in the study to perform the main analysis is available in the GitHub directory at https://github.com/Liping-Peng/Influenza_ABM_HK. Acknowledgements This project was supported by the Theme-based Research Scheme (Project No. T11-712/19-N), General Research Fund (Project No. 17104220, 17106424 to TKT) of the Research Grants Council of the Hong Kong SAR Government, and HMRF Research Fellowship Scheme (Project No. 05190097 to TKT) from Health Bureau of the Hong Kong SAR Government. BJC is supported by an RGC Senior Research Fellowship (grant number: HKU SRFS2021-7S03) and the AIR@innoHK program of the Innovation and Technology Commission of the Hong Kong SAR Government. Potential conflicts of interest. BJC reports honoraria from AstraZeneca, GlaxoSmithKline, Moderna, Roche and Sanofi Pasteur. The authors report no other potential conflicts of interest. References Hirve, S. et al. Influenza Seasonality in the Tropics and Subtropics - When to Vaccinate? PLoS One 11 , e0153003, doi:10.1371/journal.pone.0153003 (2016). Zhao, C. et al. Characterising the asynchronous resurgence of common respiratory viruses following the COVID-19 pandemic. Nat Commun 16 , 1610, doi:10.1038/s41467-025-56776-z (2025). Center of Health Protection of The Government of the Hong Kong Special Administrative Region. Epidemiology of seasonal influenza in Hong Kong and use of seasonal influenza vaccines , (2024). Department of Health of The Government of the Hong Kong Special Administrative Region. Vaccination programmes 2010/11 to be launched in November , (2010). Department of Health of The Government of the Hong Kong Special Administrative Region. Vaccination subsidy schemes launched , (2009). The Legislative Council of the Hong Kong Special Administrative Region. Seasonal influenza vaccination , (2018). Sun, K. S. et al. Seasonal influenza vaccine uptake among Chinese in Hong Kong: barriers, enablers and vaccination rates. Hum Vaccin Immunother 16 , 1675-1684, doi:10.1080/21645515.2019.1709351 (2020). Yuan, J. et al. Parental vaccine hesitancy and influenza vaccine type preferences during and after the COVID-19 Pandemic. Commun Med (Lond) 4 , 165, doi:10.1038/s43856-024-00585-w (2024). Choi, Y. J. et al. Real-world effectiveness of influenza vaccine over a decade during the 2011-2021 seasons-Implications of vaccine mismatch. Vaccine 42 , 126381, doi:10.1016/j.vaccine.2024.126381 (2024). Chan, M. C. W. et al. Frequent Genetic Mismatch between Vaccine Strains and Circulating Seasonal Influenza Viruses, Hong Kong, China, 1996-2012. Emerg Infect Dis 24 , 1825-1834, doi:10.3201/eid2410.180652 (2018). Fong, M. W. et al. Nonpharmaceutical Measures for Pandemic Influenza in Nonhealthcare Settings-Social Distancing Measures. Emerg Infect Dis 26 , 976-984, doi:10.3201/eid2605.190995 (2020). Liang, M. et al. Efficacy of face mask in preventing respiratory virus transmission: A systematic review and meta-analysis. Travel Med Infect Dis 36 , 101751, doi:10.1016/j.tmaid.2020.101751 (2020). Bin Nafisah, S., Alamery, A. H., Al Nafesa, A., Aleid, B. & Brazanji, N. A. School closure during novel influenza: A systematic review. J Infect Public Health 11 , 657-661, doi:10.1016/j.jiph.2018.01.003 (2018). Rashid, H. et al. Evidence compendium and advice on social distancing and other related measures for response to an influenza pandemic. Paediatr Respir Rev 16 , 119-126, doi:10.1016/j.prrv.2014.01.003 (2015). Skarp, J. E. et al. A Systematic Review of the Costs Relating to Non-pharmaceutical Interventions Against Infectious Disease Outbreaks. Appl Health Econ Health Policy 19 , 673-697, doi:10.1007/s40258-021-00659-z (2021). Haldane, V. et al. Strengthening the basics: public health responses to prevent the next pandemic. BMJ 375 , e067510, doi:10.1136/bmj-2021-067510 (2021). Zweig, S. A., Zapf, A. J., Beyrer, C., Guha-Sapir, D. & Haar, R. J. Ensuring Rights while Protecting Health: The Importance of Using a Human Rights Approach in Implementing Public Health Responses to COVID-19. Health Hum Rights 23 , 173-186 (2021). Faherty, L. J. et al. Effects of non-pharmaceutical interventions on COVID-19 transmission: rapid review of evidence from Italy, the United States, the United Kingdom, and China. Front Public Health 12 , 1426992, doi:10.3389/fpubh.2024.1426992 (2024). Willem, L., Verelst, F., Bilcke, J., Hens, N. & Beutels, P. Lessons from a decade of individual-based models for infectious disease transmission: a systematic review (2006-2015). BMC Infect Dis 17 , 612, doi:10.1186/s12879-017-2699-8 (2017). Zhang, H. et al. Combinational Recommendation of Vaccinations, Mask-Wearing, and Home-Quarantine to Control Influenza in Megacities: An Agent-Based Modeling Study With Large-Scale Trajectory Data. Front Public Health 10 , 883624, doi:10.3389/fpubh.2022.883624 (2022). Guo, D. et al. Multi-scale modeling for the transmission of influenza and the evaluation of interventions toward it. Sci Rep 5 , 8980, doi:10.1038/srep08980 (2015). Shaman, J., Karspeck, A., Yang, W., Tamerius, J. & Lipsitch, M. Real-time influenza forecasts during the 2012-2013 season. Nat Commun 4 , 2837, doi:10.1038/ncomms3837 (2013). Van Kerkhove, M. D., Hirve, S., Koukounari, A., Mounts, A. W. & group, H. N. p. s. w. Estimating age-specific cumulative incidence for the 2009 influenza pandemic: a meta-analysis of A(H1N1)pdm09 serological studies from 19 countries. Influenza Other Respir Viruses 7 , 872-886, doi:10.1111/irv.12074 (2013). World Health Organization. Global Influenza Strategy 2019–2030 , (2019). Kerr, C. C. et al. Covasim: An agent-based model of COVID-19 dynamics and interventions. PLoS Comput Biol 17 , e1009149, doi:10.1371/journal.pcbi.1009149 (2021). Bambery, B. et al. Influenza Vaccination Strategies Should Target Children. Public Health Ethics 11 , 221-234, doi:10.1093/phe/phx021 (2018). Tsang, T. K. & Cowling, B. J. Optimal age groups to target for influenza vaccination to reduce the impact of influenza in Hong Kong: abridged secondary publication. Hong Kong Med J 31 Suppl 3 , 30-33 (2025). King, J. C., Jr. et al. Effectiveness of school-based influenza vaccination. N Engl J Med 355 , 2523-2532, doi:10.1056/NEJMoa055414 (2006). Zhu, S. et al. Estimating Influenza Vaccine Effectiveness Against Laboratory-Confirmed Influenza Using Linked Public Health Information Systems, California, 2023-2024 Season. J Infect Dis , doi:10.1093/infdis/jiaf248 (2025). Mousa, A. et al. Social contact patterns and implications for infectious disease transmission – a systematic review and meta-analysis of contact surveys. eLife 10 , e70294, doi:10.7554/eLife.70294 (2021). Viboud, C. et al. Risk factors of influenza transmission in households. Br J Gen Pract 54 , 684-689 (2004). Chapman, G. B. et al. Using game theory to examine incentives in influenza vaccination behavior. Psychol Sci 23 , 1008-1015, doi:10.1177/0956797612437606 (2012). Burns, A. A. C. & Gutfraind, A. Effectiveness of isolation policies in schools: evidence from a mathematical model of influenza and COVID-19. PeerJ 9 , e11211, doi:10.7717/peerj.11211 (2021). Jackson, C., Mangtani, P., Hawker, J., Olowokure, B. & Vynnycky, E. The effects of school closures on influenza outbreaks and pandemics: systematic review of simulation studies. PLoS One 9 , e97297, doi:10.1371/journal.pone.0097297 (2014). Martinez, D. L. & Das, T. K. Design of non-pharmaceutical intervention strategies for pandemic influenza outbreaks. BMC Public Health 14 , 1328, doi:10.1186/1471-2458-14-1328 (2014). Fumanelli, L., Ajelli, M., Merler, S., Ferguson, N. M. & Cauchemez, S. Model-Based Comprehensive Analysis of School Closure Policies for Mitigating Influenza Epidemics and Pandemics. PLoS Comput Biol 12 , e1004681, doi:10.1371/journal.pcbi.1004681 (2016). Brienen, N. C., Timen, A., Wallinga, J., van Steenbergen, J. E. & Teunis, P. F. The effect of mask use on the spread of influenza during a pandemic. Risk Anal 30 , 1210-1218, doi:10.1111/j.1539-6924.2010.01428.x (2010). Offeddu, V., Yung, C. F., Low, M. S. F. & Tam, C. C. Effectiveness of Masks and Respirators Against Respiratory Infections in Healthcare Workers: A Systematic Review and Meta-Analysis. Clin Infect Dis 65 , 1934-1942, doi:10.1093/cid/cix681 (2017). Tracht, S. M., Del Valle, S. Y. & Hyman, J. M. Mathematical modeling of the effectiveness of facemasks in reducing the spread of novel influenza A (H1N1). PLoS One 5 , e9018, doi:10.1371/journal.pone.0009018 (2010). Kumar, S., Quinn, S. C., Kim, K. H., Daniel, L. H. & Freimuth, V. S. The impact of workplace policies and other social factors on self-reported influenza-like illness incidence during the 2009 H1N1 pandemic. Am J Public Health 102 , 134-140, doi:10.2105/AJPH.2011.300307 (2012). Kumar, S., Grefenstette, J. J., Galloway, D., Albert, S. M. & Burke, D. S. Policies to reduce influenza in the workplace: impact assessments using an agent-based model. Am J Public Health 103 , 1406-1411, doi:10.2105/AJPH.2013.301269 (2013). Glass, R. J., Glass, L. M., Beyeler, W. E. & Min, H. J. Targeted social distancing design for pandemic influenza. Emerg Infect Dis 12 , 1671-1681, doi:10.3201/eid1211.060255 (2006). Wu, J. T., Riley, S., Fraser, C. & Leung, G. M. Reducing the impact of the next influenza pandemic using household-based public health interventions. PLoS Med 3 , e361, doi:10.1371/journal.pmed.0030361 (2006). Blanchet Zumofen, M. H., Frimpter, J. & Hansen, S. A. Impact of Influenza and Influenza-Like Illness on Work Productivity Outcomes: A Systematic Literature Review. Pharmacoeconomics 41 , 253-273, doi:10.1007/s40273-022-01224-9 (2023). Leung, N. H., Xu, C., Ip, D. K. & Cowling, B. J. Review Article: The Fraction of Influenza Virus Infections That Are Asymptomatic: A Systematic Review and Meta-analysis. Epidemiology 26 , 862-872, doi:10.1097/EDE.0000000000000340 (2015). Darvishian, M., Bijlsma, M. J., Hak, E. & van den Heuvel, E. R. Effectiveness of seasonal influenza vaccine in community-dwelling elderly people: a meta-analysis of test-negative design case-control studies. Lancet Infect Dis 14 , 1228-1239, doi:10.1016/S1473-3099(14)70960-0 (2014). Zhang, H. et al. Combinational Recommendation of Vaccinations, Mask-Wearing, and Home-Quarantine to Control Influenza in Megacities: An Agent-Based Modeling Study With Large-Scale Trajectory Data. Frontiers in Public Health 10 , doi:10.3389/fpubh.2022.883624 (2022). Glaubitz, A. & Fu, F. Social dilemma of non-pharmaceutical interventions. arXiv preprint arXiv:2404.07829 (2024). Gao, H. et al. Pandemic fatigue and attenuated impact of avoidance behaviours against COVID-19 transmission in Hong Kong by cross-sectional telephone surveys. BMJ Open 11 , e055909, doi:10.1136/bmjopen-2021-055909 (2021). Funk, S., Andrews, M. A. & Bauch, C. T. Disease Interventions Can Interfere with One Another through Disease-Behaviour Interactions. PLOS Computational Biology 11 , doi:10.1371/journal.pcbi.1004291 (2015). Ajelli, M., Poletti, P., Melegaro, A. & Merler, S. The role of different social contexts in shaping influenza transmission during the 2009 pandemic. Scientific Reports 4 , 7218, doi:10.1038/srep07218 (2014). Halder, N., Kelso, J. K. & Milne, G. J. Analysis of the effectiveness of interventions used during the 2009 A/H1N1 influenza pandemic. BMC Public Health 10 , 168, doi:10.1186/1471-2458-10-168 (2010). Cowling, B. J. et al. Protective efficacy of seasonal influenza vaccination against seasonal and pandemic influenza virus infection during 2009 in Hong Kong. Clin Infect Dis 51 , 1370-1379, doi:10.1086/657311 (2010). Cowling, B. J. et al. Protective efficacy against pandemic influenza of seasonal influenza vaccination in children in Hong Kong: a randomized controlled trial. Clin Infect Dis 55 , 695-702, doi:10.1093/cid/cis518 (2012). Cowling, B. J. et al. Incidence of influenza virus infections in children in Hong Kong in a 3-year randomized placebo-controlled vaccine study, 2009-2012. Clin Infect Dis 59 , 517-524, doi:10.1093/cid/ciu356 (2014). Ip, D. K. M. et al. A Smart Card-Based Electronic School Absenteeism System for Influenza-Like Illness Surveillance in Hong Kong: Design, Implementation, and Feasibility Assessment. JMIR Public Health Surveill 3 , e67, doi:10.2196/publichealth.6810 (2017). Tsang, T. K. et al. Reconstructing antibody dynamics to estimate the risk of influenza virus infection. Nat Commun 13 , 1557, doi:10.1038/s41467-022-29310-8 (2022). Orrico-Sánchez, A., Valls-Arévalo, Á., Garcés-Sánchez, M., Álvarez Aldeán, J. & Ortiz de Lejarazu Leonardo, R. Efficacy and effectiveness of influenza vaccination in healthy children. A review of current evidence. Enfermedades Infecciosas y Microbiología Clínica 41 , 396-406, doi:https://doi.org/10.1016/j.eimc.2022.02.005 (2023). Xiong, W., Cowling, B. J. & Tsang, T. K. Influenza Resurgence after Relaxation of Public Health and Social Measures, Hong Kong, 2023. Emerg Infect Dis 29 , 2556-2559, doi:10.3201/eid2912.230937 (2023). Wong, Z. S., Goldsman, D. & Tsui, K. L. Economic Evaluation of Individual School Closure Strategies: The Hong Kong 2009 H1N1 Pandemic. PLoS One 11 , e0147052, doi:10.1371/journal.pone.0147052 (2016). Panovska-Griffiths, J. et al. Determining the optimal strategy for reopening schools, the impact of test and trace interventions, and the risk of occurrence of a second COVID-19 epidemic wave in the UK: a modelling study. Lancet Child Adolesc Health 4 , 817-827, doi:10.1016/S2352-4642(20)30250-9 (2020). Additional Declarations Yes there is potential Competing Interest. BJC reports honoraria from AstraZeneca, GlaxoSmithKline, Moderna, Roche and Sanofi Pasteur. The authors report no other potential conflicts of interest. Supplementary Files FigureS1Calibparamtop30.pdf Supplementary Figure 1 FigureS4NPIdelayandrelaxARmedian.pdf Supplementary Figure 4 FigureS5lbCalibrationsmooth7.pdf Supplementary Figure 5 FigureS7lbintvNPIisopersymmean.pdf Supplementary Figure 7 FigureS9lbSchoolbasedintvscatter.pdf Supplementary Figure 9 FigureS11lbintvVaccinegroupARreductionARmedian.pdf Supplementary Figure 11 FigureS15ubintvNPIisopersymmean.pdf Supplementary Figure 15 FigureS12lbNPIdelayandrelaxARmedian.pdf Supplementary Figure 12 FigureS8lbintvcombination2024ARmedianrange100.pdf Supplementary Figure 8 FigureS16ubintvcombination2024ARmedianrange100.pdf Supplementary Figure 16 appendix.docx Appendix FigureS17ubSchoolbasedintvscatter.pdf Supplementary Figure 17 FigureS20ubNPIdelayandrelaxARmedian.pdf Supplementary Figure 20 FigureS14ubintvVaccineARmedian.pdf Supplementary Figure 14 FigureS2intvVaccineARreductionARmedian.pdf Supplementary Figure 2 FigureS3intvVaccinegroupARreductionARmedian.pdf Supplementary Figure 3 FigureS6lbintvVaccineARmedian.pdf Supplementary Figure 6 FigureS10lbintvVaccineARreductionARmedian.pdf Supplementary Figure 10 FigureS13ubCalibrationsmooth7.pdf Supplementary Figure 13 FigureS19ubintvVaccinegroupARreductionARmedian.pdf Supplementary Figure 19 FigureS18ubintvVaccineARreductionARmedian.pdf Supplementary Figure 18 Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7574768","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":519101729,"identity":"e513f54d-58b4-483a-8d1e-1368cd2ddd40","order_by":0,"name":"Tim Tsang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAr0lEQVRIiWNgGAWjYHACZoYHDAfkDHhArAPE6WBmSGA4YEy6lsQNRGvRbT9/2CCh5k76dp4zBswFZ4jQYnYmmTkh4diz3J29PQbMM24Qo+VAMvOBBLbDuRvO8xgw83wgRsv5x0At/w6nGxCv5QbQYYlthxMMzgIdxkOUw248NjZI7HtmuOHMsYLDPER5/3ziY4kP3+7IG5xJ3viY5xgRWlDAAVI1jIJRMApGwSjAAQDmaj4t9eO3dQAAAABJRU5ErkJggg==","orcid":"","institution":"The University of Hong Kong","correspondingAuthor":true,"prefix":"","firstName":"Tim","middleName":"","lastName":"Tsang","suffix":""},{"id":519101730,"identity":"c57e9db2-f273-4994-b2f2-7456de7b1de0","order_by":1,"name":"Liping Peng","email":"","orcid":"","institution":"The University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Liping","middleName":"","lastName":"Peng","suffix":""},{"id":519101731,"identity":"d99ce1ac-58f6-4bad-a87e-2e5e25a42296","order_by":2,"name":"Yiyang Guo","email":"","orcid":"","institution":"The University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Yiyang","middleName":"","lastName":"Guo","suffix":""},{"id":519101732,"identity":"0da160e3-856a-440a-b48f-5bdbef981826","order_by":3,"name":"Nicole Tsang","email":"","orcid":"","institution":"The University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Nicole","middleName":"","lastName":"Tsang","suffix":""},{"id":519101733,"identity":"cb7a510e-fdd9-41d6-aa44-6fc8143a76e8","order_by":4,"name":"Xiaotong Huang","email":"","orcid":"","institution":"The University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Xiaotong","middleName":"","lastName":"Huang","suffix":""},{"id":519101734,"identity":"3284486f-a088-4e01-a609-7d2397bbfd97","order_by":5,"name":"Hau Chi So","email":"","orcid":"https://orcid.org/0000-0003-3119-1527","institution":"University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Hau","middleName":"Chi","lastName":"So","suffix":""},{"id":519101735,"identity":"bd5e49e1-e5de-4086-be30-27f78fc1ab21","order_by":6,"name":"Benjamin Cowling","email":"","orcid":"https://orcid.org/0000-0002-6297-7154","institution":"University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Benjamin","middleName":"","lastName":"Cowling","suffix":""},{"id":519101736,"identity":"f253482e-4124-4f52-a330-4de8818fc6f8","order_by":7,"name":"Dennis Ip","email":"","orcid":"https://orcid.org/0000-0003-3592-8709","institution":"The University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Dennis","middleName":"","lastName":"Ip","suffix":""}],"badges":[],"createdAt":"2025-09-09 14:21:27","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7574768/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7574768/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92089960,"identity":"7dfe5b63-7335-4715-9aa7-85c1a0dc7d13","added_by":"auto","created_at":"2025-09-24 13:28:37","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":183942,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/d854d171e62a913602a6c8dd.docx"},{"id":92086601,"identity":"9882d1a5-b68a-460b-870b-8233539ef672","added_by":"auto","created_at":"2025-09-24 13:04:37","extension":"json","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9421,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS2571825.json","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/15784cdaecec60f56a61727b.json"},{"id":92088482,"identity":"5df1e578-475d-4c8d-a50b-b0b4d81851a7","added_by":"auto","created_at":"2025-09-24 13:12:37","extension":"pdf","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":40371,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS10lbintvVaccineARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/d959c6d983c6dc59493d0e66.pdf"},{"id":92088855,"identity":"c1e5244a-3881-42c7-a59c-6a2e1d3d3a62","added_by":"auto","created_at":"2025-09-24 13:20:37","extension":"pdf","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54293,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS11lbintvVaccinegroupARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/e902e0dec9cd128627414cc8.pdf"},{"id":92088484,"identity":"e18f2ef8-034a-48e4-bc62-6b5523733186","added_by":"auto","created_at":"2025-09-24 13:12:37","extension":"pdf","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21154,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS12lbNPIdelayandrelaxARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/59c249fbed268b457289032e.pdf"},{"id":92086616,"identity":"44f3b4ea-6fba-46d5-8489-d197c75de028","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"pdf","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":109786,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS13ubCalibrationsmooth7.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/b92a43b58a50764178de5b40.pdf"},{"id":92086612,"identity":"e026fea0-f67e-4977-88cb-7dbb09b14872","added_by":"auto","created_at":"2025-09-24 13:04:37","extension":"pdf","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37258,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS14ubintvVaccineARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/4c8b492436e7584fdd3f9e30.pdf"},{"id":92088497,"identity":"12a3ec79-ef92-49df-a13a-3d801c75950e","added_by":"auto","created_at":"2025-09-24 13:12:38","extension":"pdf","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":26943,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS15ubintvNPIisopersymmean.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/a86caf9e980fa8c30cdef6d7.pdf"},{"id":92086613,"identity":"f2b41808-158a-46bd-b9ea-4cb535776947","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"pdf","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20405,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS16ubintvcombination2024ARmedianrange100.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/617febcc9b5c8719b6ef383f.pdf"},{"id":92086618,"identity":"884290b4-ce2e-4ec1-b8c9-c910885e5695","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"pdf","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":85622,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS17ubSchoolbasedintvscatter.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/3e8dd3b9b29b1f24ca8c8c8a.pdf"},{"id":92086630,"identity":"ac8cdbaa-42c5-46f4-a6aa-033931b0e7db","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"pdf","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":40623,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS18ubintvVaccineARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/c6b11675fcbe25a625d0e4fc.pdf"},{"id":92086633,"identity":"c38103b1-d2ef-4037-be0a-e9bb6b65ea56","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"pdf","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54567,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS19ubintvVaccinegroupARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/84fb6fdb06b88f22ddea01b8.pdf"},{"id":92088494,"identity":"8ba892e2-9041-4d7f-9403-fe58c7e8b279","added_by":"auto","created_at":"2025-09-24 13:12:38","extension":"pdf","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":136086,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS1Calibparamtop30.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/9b81a90897bb4da7050927b4.pdf"},{"id":92086617,"identity":"0ea43abe-8757-4ca2-86cb-bb19a8191541","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"pdf","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21609,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS20ubNPIdelayandrelaxARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/5903b78da222b488f38fa1d5.pdf"},{"id":92088487,"identity":"e4dbb373-00fd-4dc5-8c71-95dde4859bb5","added_by":"auto","created_at":"2025-09-24 13:12:38","extension":"pdf","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":40254,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS2intvVaccineARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/c5d4525099520840c78cc25a.pdf"},{"id":92086638,"identity":"aea9e745-602e-4884-a52d-cff911a217d1","added_by":"auto","created_at":"2025-09-24 13:04:39","extension":"pdf","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54697,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS3intvVaccinegroupARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/f81c80971eeec2ce54d0e2fa.pdf"},{"id":92086624,"identity":"3849cdc5-ba82-485f-8761-bdea4d2e62bf","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"pdf","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21167,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS4NPIdelayandrelaxARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/94ee9cfbcaab1a7e14942d56.pdf"},{"id":92086648,"identity":"0f98c273-1709-452d-a3ac-e692e1303fc8","added_by":"auto","created_at":"2025-09-24 13:04:39","extension":"pdf","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":107642,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS5lbCalibrationsmooth7.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/33102c90201b72ef99ac67d0.pdf"},{"id":92086635,"identity":"30beeb6c-13fd-4759-883c-f55d6e227d6d","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"pdf","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37230,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS6lbintvVaccineARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/3f5d96066ac4f4ec20e92e3e.pdf"},{"id":92088498,"identity":"56dbefb0-a4a2-4383-a741-fa26da9c49d8","added_by":"auto","created_at":"2025-09-24 13:12:38","extension":"pdf","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":26814,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS7lbintvNPIisopersymmean.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/d56dc76065dd548241fbc4e8.pdf"},{"id":92088862,"identity":"4836a4c4-98a9-4686-89b8-e0a21034e878","added_by":"auto","created_at":"2025-09-24 13:20:38","extension":"pdf","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20716,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS8lbintvcombination2024ARmedianrange100.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/17f640a2ac73e467f3998b78.pdf"},{"id":92086643,"identity":"90ddba38-38b9-4603-ab27-b3fed8f695d7","added_by":"auto","created_at":"2025-09-24 13:04:39","extension":"pdf","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":86240,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS9lbSchoolbasedintvscatter.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/245e4f9ce271685767f086b7.pdf"},{"id":92086622,"identity":"3ee8e3c2-6215-4105-8c9b-36388e005694","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"docx","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":99982,"visible":true,"origin":"","legend":"","description":"","filename":"appendix.docx","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/6d20dc6218e9e18c2b28920f.docx"},{"id":92088861,"identity":"755fbb96-9212-4107-a024-be40e28c7e7f","added_by":"auto","created_at":"2025-09-24 13:20:38","extension":"xml","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":156249,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS25718250enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/22323d0d843e1c28d900a43a.xml"},{"id":92088860,"identity":"2510b64b-c95a-4e94-a79e-362ba532bf9e","added_by":"auto","created_at":"2025-09-24 13:20:38","extension":"pdf","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":215420,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1Modelframework.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/449eec7dc0e0d34d657fa457.pdf"},{"id":92086646,"identity":"4f7b61f7-7931-40bf-b60e-0b4b311d343a","added_by":"auto","created_at":"2025-09-24 13:04:39","extension":"pdf","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":109375,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2Calibrationsmooth7.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/f9a41bcafe25bf7fa97e971d.pdf"},{"id":92088859,"identity":"e995ad01-e4e0-4506-ae24-4d7c9867517d","added_by":"auto","created_at":"2025-09-24 13:20:38","extension":"pdf","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37210,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3intvVaccineARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/e76b52dafef8eb776ac138d8.pdf"},{"id":92088858,"identity":"23c3ac48-c084-49c2-a00b-04fc4de6e32d","added_by":"auto","created_at":"2025-09-24 13:20:38","extension":"pdf","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":36802,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4intvNPIisopersymmean.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/3909b9bee2c9309c5a9c6415.pdf"},{"id":92088492,"identity":"a6401ae4-a3dc-4dee-a737-3555615c0d06","added_by":"auto","created_at":"2025-09-24 13:12:38","extension":"pdf","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20555,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5intvcombination2024ARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/c69c465764c76ff310581eba.pdf"},{"id":92088502,"identity":"e4b8ee98-40c7-46ff-a694-adb6cb9f72ac","added_by":"auto","created_at":"2025-09-24 13:12:39","extension":"pdf","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":77295,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6Schoolbasedintvscatter.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/194ba5ef70fe09ce620e2b42.pdf"},{"id":92088504,"identity":"dd97fe48-83a0-48e9-9c51-6fae3f2e7424","added_by":"auto","created_at":"2025-09-24 13:12:39","extension":"xml","order_by":36,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":154725,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS25718250structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/2ffdf63d119ed5deaab28baa.xml"},{"id":92086642,"identity":"5b102472-4891-45c0-897b-ecd1ee0c8a87","added_by":"auto","created_at":"2025-09-24 13:04:39","extension":"html","order_by":37,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":168972,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/88d884dc4e5c47d63d2138e4.html"},{"id":92086592,"identity":"d2b8b317-2982-4314-bc8e-2a54e187bbf3","added_by":"auto","created_at":"2025-09-24 13:04:37","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":161321,"visible":true,"origin":"","legend":"\u003cp\u003eModel structure.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/21c4ca95b47247569a40b625.png"},{"id":92086596,"identity":"087dc4b7-c257-42eb-b5d6-25c6dc3b2c98","added_by":"auto","created_at":"2025-09-24 13:04:37","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":247259,"visible":true,"origin":"","legend":"\u003cp\u003eModel calibration results across six influenza seasons.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003eNew infection across six modelled influenza seasons. \u003cstrong\u003eb\u003c/strong\u003e Cumulative infection numbers. \u003cstrong\u003ec\u003c/strong\u003e Influenza-related school absences among students aged 6-18. \u003cstrong\u003ed\u003c/strong\u003e Age group distribution of infections. In panels a, b, and c, observed data are smoothed using a 7-day rolling average. Blue lines (or bars) represent the median across 30 calibration simulations; error bars indicate the range (minimum to maximum). \"OR\" denotes the median calibrated relative susceptibility of each age group over 30 simulations.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/3615fba946dba314b9e4b5e5.png"},{"id":92088473,"identity":"322e1fb5-5550-4ee9-ad2a-1517d978c615","added_by":"auto","created_at":"2025-09-24 13:12:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":122756,"visible":true,"origin":"","legend":"\u003cp\u003eImpact of age-targeted vaccination strategies on attack rate.\u003c/p\u003e\n\u003cp\u003eEach color represents a distinct age-targeted vaccination strategy. The x-axis denotes the number of individuals vaccinated under each strategy across varying coverage levels. For each strategy, five vaccination expansion levels are shown (10% to 50% increases in coverage, in 10% increments). Error bars represent the 10th and 90th percentiles across 30 simulation runs.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/3489b654f39213b24f7824cf.png"},{"id":92086593,"identity":"d39461ae-a424-4b00-bedf-ab93503f6828","added_by":"auto","created_at":"2025-09-24 13:04:37","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":82102,"visible":true,"origin":"","legend":"\u003cp\u003eImpact of Non-Pharmaceutical Interventions (NPIs) strategies on attack rate.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e Effect of varying coverage levels of staying home when sick and mask use on overall attack rate. Coverage for staying home refers to the proportion of symptomatic individuals adhering to staying home; coverage for mask use refers to the proportion of the total population wearing masks. \u003cstrong\u003eb\u003c/strong\u003e Comparative effectiveness of staying home when sick and mask use under different probabilities of asymptomatic infection. Points represent median attack rates across 30 stochastic simulations; error bars represent the 10th and 90th percentiles.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/0dbb03056bc99ed2e4b67b56.png"},{"id":92088849,"identity":"9b9c2650-e04e-4e23-a75f-7aa9d6e0a2bf","added_by":"auto","created_at":"2025-09-24 13:20:37","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":68402,"visible":true,"origin":"","legend":"\u003cp\u003eImpact of combination strategies for vaccine-mismatched seasons.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003eAttack rate under the combined implementation of staying home when sick and mask use in Season 1. \u003cstrong\u003eb\u003c/strong\u003e Attack rate under the combined implementation of staying home when sick and mask use in Season 4. Simulations assume baseline vaccination coverage. Accompanying tables show baseline age-specific vaccination rates for each season. Increases in staying home reflect higher probability of adherence to staying home when sick.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/220404ac55e5a417ec7eb3cd.png"},{"id":92088479,"identity":"ee5f90c3-6e60-4610-bff5-3bbfe643b51a","added_by":"auto","created_at":"2025-09-24 13:12:37","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":281385,"visible":true,"origin":"","legend":"\u003cp\u003eImpact of interventions targeting school-age children.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e Effectiveness of three interventions on the student attack rate. \u003cstrong\u003eb\u003c/strong\u003e Effectiveness of three interventions on the overall population attack rate. Scatter points represent outcomes from individual simulations, while the curves correspond to linear regression fits summarizing the simulation trends.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/c151b414422a458d542591a4.png"},{"id":92090407,"identity":"ffdbedb5-0d1f-48c1-902a-78020e3dcca3","added_by":"auto","created_at":"2025-09-24 13:36:39","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1657257,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/fabacc08-df96-4da9-8f70-8e35b9e4000e.pdf"},{"id":92086590,"identity":"991dade9-dd4e-4899-8989-9177841b706d","added_by":"auto","created_at":"2025-09-24 13:04:37","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":136086,"visible":true,"origin":"","legend":"Supplementary Figure 1","description":"","filename":"FigureS1Calibparamtop30.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/f60f0bf1d25a7db7b6a3c7f9.pdf"},{"id":92089958,"identity":"04168bc3-e990-4cc4-835b-ad7619072800","added_by":"auto","created_at":"2025-09-24 13:28:37","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":21167,"visible":true,"origin":"","legend":"Supplementary Figure 4","description":"","filename":"FigureS4NPIdelayandrelaxARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/d7f6742b6fa5da9a8c3d4632.pdf"},{"id":92088851,"identity":"66c84f9e-9cfb-427a-839f-fdd52ff690eb","added_by":"auto","created_at":"2025-09-24 13:20:37","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":107642,"visible":true,"origin":"","legend":"Supplementary Figure 5","description":"","filename":"FigureS5lbCalibrationsmooth7.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/33969b1b4d6ad627c5e8df99.pdf"},{"id":92088476,"identity":"7e536020-c60d-487b-8159-5e428eb1bdc5","added_by":"auto","created_at":"2025-09-24 13:12:37","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":26814,"visible":true,"origin":"","legend":"Supplementary Figure 7","description":"","filename":"FigureS7lbintvNPIisopersymmean.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/0bf46ae10fdf2a47fbfef809.pdf"},{"id":92086599,"identity":"68b0fc03-6a4f-495c-8078-2fb69f22840f","added_by":"auto","created_at":"2025-09-24 13:04:37","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":86240,"visible":true,"origin":"","legend":"Supplementary Figure 9","description":"","filename":"FigureS9lbSchoolbasedintvscatter.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/5e03aa2b15f30cd774a00fd5.pdf"},{"id":92089959,"identity":"f42cb8d2-5869-4072-bd6b-92333f64932d","added_by":"auto","created_at":"2025-09-24 13:28:37","extension":"pdf","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":54293,"visible":true,"origin":"","legend":"Supplementary Figure 11","description":"","filename":"FigureS11lbintvVaccinegroupARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/7456e35accff25852dc8427c.pdf"},{"id":92088853,"identity":"6527047a-4dc2-479a-8b8d-730e811d9fcd","added_by":"auto","created_at":"2025-09-24 13:20:37","extension":"pdf","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":26943,"visible":true,"origin":"","legend":"Supplementary Figure 15","description":"","filename":"FigureS15ubintvNPIisopersymmean.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/6403fb91f474de585914f535.pdf"},{"id":92088489,"identity":"c2328a89-4fc7-446d-8f6e-bd58910f918b","added_by":"auto","created_at":"2025-09-24 13:12:38","extension":"pdf","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":21154,"visible":true,"origin":"","legend":"Supplementary Figure 12","description":"","filename":"FigureS12lbNPIdelayandrelaxARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/f5c05e413776260f451a7676.pdf"},{"id":92088864,"identity":"c8adc198-3318-478b-a78f-6db69b2f34e5","added_by":"auto","created_at":"2025-09-24 13:20:39","extension":"pdf","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":20716,"visible":true,"origin":"","legend":"Supplementary Figure 8","description":"","filename":"FigureS8lbintvcombination2024ARmedianrange100.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/c281ce8e1aa5de5e544a9ceb.pdf"},{"id":92086603,"identity":"f0b4d382-b994-41be-8469-813778edf3bf","added_by":"auto","created_at":"2025-09-24 13:04:37","extension":"pdf","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":20405,"visible":true,"origin":"","legend":"Supplementary Figure 16","description":"","filename":"FigureS16ubintvcombination2024ARmedianrange100.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/a75e29168e8b943cef46363b.pdf"},{"id":92086609,"identity":"fd653d53-e3e2-459a-9027-ba78f1377092","added_by":"auto","created_at":"2025-09-24 13:04:37","extension":"docx","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":99982,"visible":true,"origin":"","legend":"Appendix","description":"","filename":"appendix.docx","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/5fc6a82efb45c777006d2307.docx"},{"id":92088486,"identity":"07467a48-3b3d-4a78-9bf9-111a35327024","added_by":"auto","created_at":"2025-09-24 13:12:38","extension":"pdf","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":85622,"visible":true,"origin":"","legend":"Supplementary Figure 17","description":"","filename":"FigureS17ubSchoolbasedintvscatter.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/e26f4b41966e2b4d52b7828c.pdf"},{"id":92088856,"identity":"9cb8da7e-508f-4bf1-b025-497b94ed224b","added_by":"auto","created_at":"2025-09-24 13:20:37","extension":"pdf","order_by":13,"title":"","display":"","copyAsset":false,"role":"supplement","size":21609,"visible":true,"origin":"","legend":"Supplementary Figure 20","description":"","filename":"FigureS20ubNPIdelayandrelaxARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/429606700c9f24ce9445145c.pdf"},{"id":92086605,"identity":"2208db4a-0958-4995-962b-9efe379734bc","added_by":"auto","created_at":"2025-09-24 13:04:37","extension":"pdf","order_by":14,"title":"","display":"","copyAsset":false,"role":"supplement","size":37258,"visible":true,"origin":"","legend":"Supplementary Figure 14","description":"","filename":"FigureS14ubintvVaccineARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/9e8ad8bcd518430e3cf37c84.pdf"},{"id":92088857,"identity":"9d056477-b0cf-44b5-81f1-66e3dc9c6318","added_by":"auto","created_at":"2025-09-24 13:20:38","extension":"pdf","order_by":15,"title":"","display":"","copyAsset":false,"role":"supplement","size":40254,"visible":true,"origin":"","legend":"Supplementary Figure 2","description":"","filename":"FigureS2intvVaccineARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/9cc1a70e492f4d9302762611.pdf"},{"id":92088488,"identity":"17e4e8a9-52d4-49c3-b4aa-8819d9e96b7c","added_by":"auto","created_at":"2025-09-24 13:12:38","extension":"pdf","order_by":16,"title":"","display":"","copyAsset":false,"role":"supplement","size":54697,"visible":true,"origin":"","legend":"Supplementary Figure 3","description":"","filename":"FigureS3intvVaccinegroupARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/27c8483da4a8be4ead9a4e7e.pdf"},{"id":92086625,"identity":"9e6f198b-9101-4545-8bd9-62dec622063f","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"pdf","order_by":17,"title":"","display":"","copyAsset":false,"role":"supplement","size":37230,"visible":true,"origin":"","legend":"Supplementary Figure 6","description":"","filename":"FigureS6lbintvVaccineARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/686e884cb37ef21875bae66c.pdf"},{"id":92086621,"identity":"3af85957-4275-43ec-b342-fb0251bc74ca","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"pdf","order_by":18,"title":"","display":"","copyAsset":false,"role":"supplement","size":40371,"visible":true,"origin":"","legend":"Supplementary Figure 10","description":"","filename":"FigureS10lbintvVaccineARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/4b939b247b5055b7abe51f4b.pdf"},{"id":92086619,"identity":"a586faab-699e-4bd2-a2a9-7d0eee85d25c","added_by":"auto","created_at":"2025-09-24 13:04:38","extension":"pdf","order_by":19,"title":"","display":"","copyAsset":false,"role":"supplement","size":109786,"visible":true,"origin":"","legend":"Supplementary Figure 13","description":"","filename":"FigureS13ubCalibrationsmooth7.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/c1f56c75111789eb5f95f616.pdf"},{"id":92088491,"identity":"f74c862c-2385-4a0d-b3b4-74a1bc8ff59f","added_by":"auto","created_at":"2025-09-24 13:12:38","extension":"pdf","order_by":20,"title":"","display":"","copyAsset":false,"role":"supplement","size":54567,"visible":true,"origin":"","legend":"Supplementary Figure 19","description":"","filename":"FigureS19ubintvVaccinegroupARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/a130cf93c1fc80f748dcb0e9.pdf"},{"id":92088505,"identity":"701ac8e5-440e-4097-af67-fe6067af310f","added_by":"auto","created_at":"2025-09-24 13:12:39","extension":"pdf","order_by":21,"title":"","display":"","copyAsset":false,"role":"supplement","size":40623,"visible":true,"origin":"","legend":"Supplementary Figure 18","description":"","filename":"FigureS18ubintvVaccineARreductionARmedian.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7574768/v1/d36daeb49609c83744e53a70.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nBJC reports honoraria from AstraZeneca, GlaxoSmithKline, Moderna, Roche and Sanofi Pasteur. The authors report no other potential conflicts of interest.","formattedTitle":"Multi-Source Agent-Based Modeling to Optimize Influenza Mitigation Strategies in Hong Kong","fulltext":[{"header":"BACKGROUND","content":"\u003cp\u003eInfluenza remains a major global public health threat, particularly in subtropical and tropical regions, which typically experience multiple seasonal outbreaks each year \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. After the COVID-19 pandemic, influenza has experienced a resurgence worldwide \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Hong Kong, a subtropical region, is significantly affected by seasonal influenza, with peaks commonly occurring from January to March/April and from July to August. In the post-COVID-19 period, influenza transmission in Hong Kong has returned to pre-pandemic levels, with outbreaks lasting even longer, up to 28 weeks, in 2024 \u003csup\u003e3\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eVaccination and non-pharmaceutical interventions (NPIs) are key tools against influenza, but their uptake and overall effectiveness remain uncertain. In Hong Kong, free seasonal influenza vaccination \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e and a government subsidy program \u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e were introduced after the 2009 H1N1 pandemic for high-risk groups, including children under 6 and adults aged 65 and above. However, coverage remained low before 2014, with uptake rates of 12.9% and 32.7%, respectively \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. The program has since expanded to include children under 18 and adults over 50. However, low perceived risk, safety concerns \u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e, and shifts in public attitudes influenced by COVID-19 vaccination policies \u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e may have hindered progress in increasing influenza vaccine uptake in Hong Kong. Vaccine effectiveness is also challenged by mismatches with circulating strains \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Between 1996 and 2012, fewer than half of H3N2 seasons in Hong Kong had vaccine strains closely matching circulating viruses \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e, resulting in reduced protection.\u003c/p\u003e\u003cp\u003eNon-pharmaceutical interventions (NPIs) such as social distancing \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e, mask use \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e, and school closures \u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e can reduce transmission, with COVID-19 experience having increased public acceptance and familiarity with measures such as mask use and work from home\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. However, implementation remains constrained by economic costs \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e, logistical challenges \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e, and varying adherence across contexts\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, underscoring the need for tailored strategies \u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eGiven these challenges with individual interventions, agent-based models\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e, offer valuable tools to assess optimal combination strategies by simulating transmission at the individual level. However, model reliability depends critically on calibration approaches that accurately capture true transmission dynamics. Common methods using surveillance data alone often underestimate infections due to underreporting and asymptomatic cases \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e,\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. These limitations necessitate integrated calibration using multiple data sources, including serological data\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e, to improve model calibration and better capture true transmission dynamics.\u003c/p\u003e\u003cp\u003eTo support the WHO\u0026rsquo;s Influenza Strategy 2030 \u003csup\u003e24\u003c/sup\u003e, we adapt an agent-based model to simulate influenza transmission in Hong Kong across six seasons (2009\u0026ndash;2013). By integrating surveillance, serological, and school absenteeism data, we reconstruct baseline age-specific transmission dynamics and evaluate targeted interventions, including age-specific vaccination, staying home when sick, mask use, and school-based measures. This study provides context-specific evidence to guide seasonal influenza control and strengthen preparedness for future outbreaks.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eOverviews\u003c/h2\u003e\u003cp\u003eThis study aimed to identify optimal strategies for reducing influenza attack rates in Hong Kong across diverse seasonal epidemic conditions, particularly vaccine-matched versus vaccine-mismatched scenarios. We evaluated the effectiveness of pharmaceutical and non-pharmaceutical interventions in reducing overall and age-specific infection rates during six influenza seasons (2009\u0026ndash;2013), which varied in circulating virus subtypes, vaccine effectiveness, epidemic magnitude, and seasonal timing. Our analytical framework prioritized comparative assessment of targeted vaccination strategies, staying home when sick, mask use, and school-based interventions, both individually and in strategic combinations, to determine strategy effectiveness for different scenarios.\u003c/p\u003e\u003cp\u003eWe used a stochastic agent-based model \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e to estimate influenza transmission dynamics in Hong Kong. The model integrates diverse data streams to capture the heterogeneity of social contacts and infection progression within the population. A synthetic population was constructed using detailed demographic and social structure data, including age-stratified population figures, employment rates, enrolment statistics, and school capacity records, which informed the creation of realistic contact networks across multiple layers (households, schools, workplaces, and community settings). Infection time series derived from serological data, which captures both reported and unreported infections, avoiding the healthcare-seeking bias inherent in clinical surveillance, along with school absenteeism data were used for model calibration, with the latter enhancing the representation of contact dynamics among school-age children.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eEpidemiological characteristics of the six influenza seasons\u003c/h3\u003e\n\u003cp\u003eThe six modeled seasons (2009\u0026ndash;2013) exhibited distinct epidemiological profiles (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Seasons 1, 2, and 6 coincided with summer holidays and showed predominantly H1N1 (season 1) or H3N2 (seasons 2 and 6) circulation. Seasons 3, 4, and 5 occurred outside summer holiday periods, with seasons 3 and 5 featuring well-matched vaccines and season 4 experiencing vaccine mismatch. Cumulative incidence of infections derived from serological data ranged from 0.29\u0026nbsp;million (season 6) to 1.25\u0026nbsp;million (season 4), with peak daily case numbers between 4,518 (season 5) and 27,095 (season 4). The 0\u0026ndash;24 age group had the highest or near-highest attack rates in five of six seasons, ranging from 4.5\u0026ndash;34.6%.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eInformation on the six modeled seasons\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSeason\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSeason 1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSeason 2\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eSeason 3\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSeason 4\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eSeason 5\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eSeason 6\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eStart date\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2009/06/20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2010/06/25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2010/12/04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2012/02/20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2013/01/01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e2013/05/31\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEnd date\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2009/11/21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2010/10/16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2011/02/26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2012/06/23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2013/04/20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e2013/10/12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSummer holiday\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariant\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eH1N1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eH3N2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eH1N1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eH3N2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eH1N1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eH3N2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVaccine matching\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMismatched\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMatched\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eMatched\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMismatched\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eMatched\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eMatched\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCumulative infections\u003c/p\u003e\u003cp\u003e(million)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.29\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePeak case number\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e27,090\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e18,127\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e21,456\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e27,095\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4,518\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e4,532\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAttack rate (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e0\u0026ndash;24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e34.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e12.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e11.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e21.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e4.5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e25\u0026ndash;44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e14.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e13.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e12.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e16.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e5.2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e45\u0026ndash;64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e10.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e14.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e9.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e17.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e3.4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u0026ge;\u0026thinsp;65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e14.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e8.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e17.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e2.7\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePeak influenza-related absences (per 100,000 students aged 6\u0026ndash;18)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eSmart-card based school absenteeism monitoring revealed distinct temporal patterns that closely tracked influenza activity (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003ec). During high-transmission periods, peak influenza-related absenteeism among students aged 6\u0026ndash;18 ranged from 30 to 470 per 100,000 student-days across seasons. Over the course of each influenza season, average daily absenteeism ranged from 8 to 53 per 100,000 student-days (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe calibrated baseline model demonstrated an acceptable fit (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e), achieving good agreement with observed data across all target metrics. Across the six seasons, an estimated 1.8\u0026ndash;5.0% of symptomatic individuals stayed home due to illness. We estimated a higher relative susceptibility among older adults (\u0026ge;\u0026thinsp;65 years), with odds ratios of 1.4 to 3.1, compared to younger age groups (Supplementary Table\u0026nbsp;1).\u003c/p\u003e\n\u003ch3\u003eAge-targeted vaccination\u003c/h3\u003e\n\u003cp\u003eAcross six seasons, vaccinating children under 12 consistently resulted in the largest reductions in population attack rate, with absolute reductions ranging from 0.52 to 2.17 percentage points and relative reductions from 6.4\u0026ndash;8.5% per 100,000 vaccinated individuals (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Supplementary Fig.\u0026nbsp;2). Targeting individuals under 18 showed moderate effectiveness (absolute reductions: 0.35 to 1.67 percentage points; relative reductions: 5.1\u0026ndash;6.8% per 100,000 vaccinated), while vaccination of those aged\u0026thinsp;\u0026ge;\u0026thinsp;65 yielded smaller reductions (absolute reductions: 0.09 to 1.12 percentage points; relative reductions: 0.5\u0026ndash;5.5% per 100,000 vaccinated). Universal vaccination was the least efficient strategy, yielding only modest reductions (absolute reductions: 0.03 to 0.39 percentage points; relative reductions: 2.2\u0026ndash;2.6% per 100,000 vaccinated).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFurther analysis showed that targeting vaccination to children under 12 not only substantially reduced attack rates within this group (absolute reductions: 1.03 to 6.13 percentage points; relative reductions: 7.1%-9.3%, per 100,000 vaccinated), but also conferred indirect protection to adolescents (\u0026lt;\u0026thinsp;18 years: absolute reductions 0.84 to 4.47 percentage points ; relative reductions: 6.8\u0026ndash;9.0% per 100,000 vaccinated) and older adults (\u0026ge;\u0026thinsp;65 years: absolute reductions 0.56 to 2.5 percentage points ; relative reductions 6.0%-8.4% per 100,000 vaccinated) (Supplementary Fig.\u0026nbsp;3). Targeting individuals under 18 also benefited older adults, though to a lesser extent (absolute reductions: 0.28 to 1.88 percentage points; relative reductions: 4.9%-6.9% per 100,000 vaccinated). In contrast, prioritizing vaccination in adults aged\u0026thinsp;\u0026ge;\u0026thinsp;65 had limited indirect impact on younger age groups, with reductions among children\u0026thinsp;\u0026lt;\u0026thinsp;18 observed in three of six seasons (absolute reductions: 0.19 to 0.73 percentage points; relative reductions: 3.4%-4.1% per 100,000 vaccinated) and no significant effect in the remaining seasons.\u003c/p\u003e\n\u003ch3\u003eNon-pharmaceutical interventions (NPIs)\u003c/h3\u003e\n\u003cp\u003eIncreasing the daily probability of adherence to staying home when sick by 10\u0026ndash;50% (resulting in 6\u0026ndash;31% of symptomatic cases staying home) reduced the attack rate by 2\u0026ndash;49% compared to the baseline across all six seasons. In contrast, increasing population mask usage by 10\u0026ndash;50% (equivalent to 20\u0026ndash;60% of the population using masks) achieved a 8\u0026ndash;64% reduction in attack rate (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003ea).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe estimated that population-wide masking (40% coverage, Mask\u0026thinsp;+\u0026thinsp;30% scenario) reduces influenza transmission as effectively as 20\u0026ndash;31% of symptomatic individuals staying home (Stay-home\u0026thinsp;+\u0026thinsp;50% scenario). Across six seasons, both strategies achieved comparable reductions (18\u0026ndash;45% for masking vs. 17\u0026ndash;49% for staying home). However, their relative effectiveness varied with asymptomatic infection proportions. While mask effectiveness remained stable, the effectiveness of staying home declined as asymptomatic proportion increased. Under the Stay-home\u0026thinsp;+\u0026thinsp;50% scenario, increasing the asymptomatic infection probability from 20% (baseline) to 60% resulted in an absolute increase of 0.3\u0026ndash;4.0% in the overall attack rate across seasons. (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003eb).\u003c/p\u003e\u003cp\u003eIntervention effectiveness was sensitive to implementation timing, with staying home when sick more affected by delays than mask use. Higher coverage generally preserved effectiveness under short delays, but as delays increased, the difference between coverage levels narrowed (Supplementary Fig.\u0026nbsp;4).\u003c/p\u003e\n\u003ch3\u003eCombination strategies during mismatched seasons\u003c/h3\u003e\n\u003cp\u003eDuring seasons with vaccine mismatch and low vaccination coverage (Seasons 1 and 4), combining vaccination with NPIs was essential. In these two seasons, a 50% increase in the daily probability of adherence to staying home when sick (resulting in 21\u0026ndash;31% of symptomatic cases staying home) reduced the attack rate by 41\u0026ndash;47% (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003e). When both staying home when sick and mask use were intensified simultaneously, each increasing by 10\u0026ndash;50% over baseline, the attack rate was reduced by 22\u0026ndash;84% in Season 1 and 26\u0026ndash;77% in Season 4.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eSchool closure and targeted strategies for school-aged children\u003c/h2\u003e\u003cp\u003eWe assessed three school-based interventions across seasons without summer holidays (seasons 3, 4 and 5) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003e). When coverage increased from 10\u0026ndash;50% for each intervention, student vaccination consistently achieved the greatest reductions in both student attack rates (56\u0026ndash;86%) and overall population attack rates (33\u0026ndash;78%). Staying home when sick among symptomatic students showed variable effectiveness, reducing student rates by 23\u0026ndash;55% and overall rates by 15\u0026ndash;45%. School closures had the smallest impact, with student reductions of 15\u0026ndash;31% and overall reductions of 18\u0026ndash;25%.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eSensitivity analyses showed that the key conclusions remained robust under joint variation of multiple parameters (Supplementary Results, Supplementary Fig.\u0026nbsp;5\u0026ndash;20). The relative effectiveness of interventions and the overall patterns of attack rate reduction were broadly consistent across both lower-bound and upper-bound parameter settings.\u003c/p\u003e\u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eWe calibrated the model using infection time series from surveillance and serological data, and incorporated school absenteeism to inform school-based intervention. By simulating six influenza seasons in Hong Kong (2009\u0026ndash;2013) across varied virus subtypes and vaccine matches, our model quantifies the differential impact of vaccination and non-pharmaceutical interventions. Child vaccination consistently reduced transmission, underscoring children\u0026rsquo;s pivotal role as infection amplifiers in dense urban settings and supporting targeted immunization policies. Among non-pharmaceutical measures, mask use showed robust and stable effectiveness comparable to moderate levels of staying home when sick, highlighting its value as a reliable control strategy. Importantly, during vaccine-mismatched seasons, combined interventions were crucial to maintaining epidemic control. Furthermore, school-based vaccination demonstrated superior effectiveness and sustainability compared to reactive closure strategies.\u003c/p\u003e\u003cp\u003eVaccination strategies targeting children consistently outperformed alternatives, reflecting both strong direct protection and substantial indirect benefits at population level. This finding is well-supported by modeling studies from the US and UK, and notably by a community-based trial in Hong Kong where increasing child coverage halved child attack rates while also reducing adult rates, demonstrating important indirect protection \u003csup\u003e\u003cspan additionalcitationids=\"CR27\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Children are highly exposed in school settings and typically exhibit stronger immune responses to influenza vaccines than adults, leading to higher effectiveness \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. Vaccinating this group also reduces onward transmission, as children play a central role in community spread due to their high contact rates \u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e and elevated transmissibility \u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e, These findings support the expansion of Hong Kong\u0026rsquo;s Seasonal Influenza Vaccination (SIV) School Outreach Programme to include adolescents under 18, which proved more efficient than universal or elderly-targeted strategies. As the benefits of child vaccination are largely indirect, uptake may remain suboptimal without targeted incentives \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eBeyond vaccination, other school-based interventions provide moderate mitigation. Encouraging symptomatic students to stay home reduces transmission meaningfully, consistent with CDC recommendations showing modest reductions under partial compliance \u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e, emphasizing the importance of promoting voluntary. In contrast, short-term reactive school closures produced limited effects in our simulations, aligning with systematic reviews showing that brief closures achieve only modest reductions \u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e, unless repeated or extended under high compliance \u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. Given their limited epidemiological benefit and considerable social and economic costs, school closures should be carefully weighed against more targeted and sustainable interventions.\u003c/p\u003e\u003cp\u003eMask use provides population-level control that does not rely on symptom recognition or individual adherence decisions. Even widespread but not universal use can meaningfully reduce transmission through network effects, where preventing early transmission events disproportionately constrains total outbreak size \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Laboratory and meta-analytic studies support substantial per-contact efficacy of surgical or cloth masks \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Compared to stay-home policies, masks maintain more consistent effectiveness across seasons because they protect regardless of symptom awareness. However, effectiveness still varies with baseline transmissibility and intervention timing \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eOur simulations indicate that voluntary stay-home policies among symptomatic individuals can achieve meaningful transmission reductions, even with modest voluntary levels. While baseline compliance is typically low due to mild symptoms and limited paid sick leave \u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e, increasing participation to feasible levels produced notable mitigation, consistent with workplace studies showing substantial infection reductions with partial sick leave uptake \u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e, and community simulations demonstrating comparable benefits \u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. However, financial disincentives such as income loss \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e often discourage adherence, especially in the absence of supportive policies. The widespread adoption of remote work arrangements following the COVID-19 pandemic may have increased the feasibility of staying home when sick, particularly for office workers. However, when economic constraints prevent staying home, public health authorities may need to recommend less effective but more accessible alternatives such as mask use.\u003c/p\u003e\u003cp\u003eUnlike mask use, which provides population-wide protection regardless of infection awareness, staying home when sick depends on individuals recognizing symptoms and voluntarily avoiding contact with others. Its effectiveness is therefore influenced by the timeliness of symptom detection and the consistency of individual adherence. The presence of asymptomatic infections, which are common in influenza and vary in estimated prevalence across studies \u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e, further limits the effectiveness of staying home when sick. Without complementary strategies such as contact tracing, this strategy remains modest in effect and operationally challenging to implement.\u003c/p\u003e\u003cp\u003eVaccine mismatch seasons, common in influenza epidemics \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e, highlight the need for early and sustained non-pharmaceutical interventions to control transmission effectively. This aligns with previous studies showing that vaccination alone is insufficient under high-transmission conditions and that multiple interventions are needed to suppress \u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. However, our assumption of constant adherence likely overestimates real-world impact, as behavioral responses often weaken over time \u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e,\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. Moreover, interactions between interventions\u0026mdash;such as reduced NPI adherence after vaccination or lower vaccine uptake when NPIs are widespread\u0026mdash;may influence overall impact \u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. Future models should account for these behavioral dynamics to better inform integrated strategies.\u003c/p\u003e\u003cp\u003eThis study offers several methodological strengths. First, analyzing six influenza seasons under varying epidemiological conditions enhanced the generalizability of our findings. Second, calibration using serology-based attack rates, rather than clinical surveillance data alone, enabled more accurate estimation of total infections, including asymptomatic cases. While some prior studies incorporated serological data\u003csup\u003e51 52\u003c/sup\u003e, these focused exclusively on the 2009 H1N1 pandemic and used serology only to estimate seasonal attack rates without integrating surveillance data, lacking temporal resolution and sometimes using data from different seasons or subtypes. In contrast, we utilized contemporaneous, population-specific serological data integrated with surveillance systems to reconstruct age-specific epidemic curves, capturing both reported and unreported infections with daily resolution. Third, incorporating school absenteeism data enabled us to model illness-related behaviors in school-aged children, a key driver of influenza transmission, and to establish a more realistic behavioral baseline often omitted in models. This comprehensive approach improved model realism and strengthened the reliability of intervention impact assessments.\u003c/p\u003e\u003cp\u003eSeveral limitations should be noted. First, our analyses focused on infections and did not capture clinical severity or mortality, limiting its ability to reflect total disease burden. Second, antiviral treatment was not considered in the analyses due to low uptake in Hong Kong. Third, some parameters, such as the probability of symptomatic infection, were not age-specific, potentially underestimating population heterogeneity. Mask use and health-seeking were assumed constant over time, which may not reflect real-world dynamics. Lastly, the model did not capture behavioral interactions between interventions, such as reduced NPI adherence following vaccination.\u003c/p\u003e\u003cp\u003eIn summary, this study used a multi-season, age-structured model calibrated with serological, surveillance and school absenteeism data to evaluate vaccination and NPIs for influenza control in Hong Kong. By integrating diverse data sources and capturing behavioral and seasonal variability, the model provides a robust tool for evaluating targeted strategies. The findings support child-targeted vaccination and emphasize the complementary role of NPIs, particularly during vaccine-mismatched seasons. School-based interventions further highlight the advantages of proactive measures like student vaccination over reactive strategies such as closures. This framework offers actionable insights for influenza control and broader respiratory preparedness.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eData sources\u003c/h2\u003e\u003cp\u003eData for this study were obtained to capture a comprehensive picture of influenza activity over six seasons in Hong Kong (2009\u0026ndash;2013) (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Multiple, high-quality sources provided the basis for the diverse datasets used in this analysis.\u003c/p\u003e\u003cp\u003eDemographic and social structure data, including age-stratified population figures, employment statistics, enrolment numbers, and school capacity records were sourced from governmental agencies and public institutions. These datasets offered detailed insights into the population structure and social mixing patterns in Hong Kong, to generate the contact matrix.\u003c/p\u003e\u003cp\u003eInfluenza activity was monitored through surveillance systems that recorded the percentage of outpatient visits attributed to influenza-like illness (ILI) alongside the proportion of laboratory-confirmed influenza cases from public health laboratories. Weekly influenza activity proxy was derived by multiplying these two indicators.\u003c/p\u003e\u003cp\u003eSerology data were collected from two community-based randomized controlled trials (RCTs) for evaluating direct and indirect benefits of influenza vaccination \u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e,\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. In the RCTs conducted in 2008/09 and 2009/ 10, 119 and 796 households were recruited. Serum specimens were collected at the start of the study, and after 6 and 12 months from all participants. In the sub- sequent observational follow-up of the same cohort participants from late 2010 to late 2013 without intervention \u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e, serum specimens were collected from all participants in each autumn (October to December), and also each spring (April to May). Receipt of influenza vaccine outside of the trial was recorded annually.\u003c/p\u003e\u003cp\u003eSchool absenteeism data were collected through a smartcard-based monitoring system covering 66 primary and 41 secondary schools across all 18 districts of Hong Kong, encompassing 75,052 students \u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. This system tracked daily attendance patterns and class sizes, providing data on all-cause absenteeism. To estimate influenza-attributable absenteeism, we adjusted the all-cause absenteeism data using ILI consultation rates and specimen positivity rates from the established surveillance systems.\u003c/p\u003e\u003cp\u003eIntervention-related data, including effectiveness, baseline coverage, and implementation timing of various interventions were obtained through a literature review and from official government sources (Supplementary Methods, Supplementary Table\u0026nbsp;4).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eModel details\u003c/h2\u003e\u003cp\u003eCovasim was originally developed for SARS-CoV-2, hence, we refined the model to capture the transmission characteristics and epidemiology of influenza. It explicitly represents the progression of the disease through distinct states within the population. In the model, individuals transition from a susceptible state to exposed, then progress to either symptomatic or asymptomatic infection, with an 80% probability of developing symptoms, before ultimately recovering (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTo mirror the approximately 7\u0026nbsp;million inhabitants of Hong Kong, we constructed a synthetic population of about 70,000 agents using a scale-up factor of 100 to reduce complexity while maintain accuracy \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. This population was designed to reflect Hong Kong\u0026rsquo;s demographic composition by incorporating comprehensive data, such as age distributions, employment statistics, and school enrolment figures. These data underpin the development of realistic contact networks that span households, schools, workplaces, and community settings (Supplementary Methods and Supplementary Table\u0026nbsp;2), capturing the heterogeneous nature of social interactions in an urban environment. Initial parameter values were sourced from existing literature and established assumptions.\u003c/p\u003e\u003cp\u003eThe transmission model incorporates empirically-derived parameter values obtained through literature review and calibration to Hong Kong surveillance data. This includes age-specific susceptibility profiles, transmissibility factors, and disease progression timelines. Importantly, the model integrates actual NPI measures and documented vaccine coverage data from the study period. By incorporating real-world values for interventions such as staying home when sick and mask use, alongside empirical vaccination coverage across various age groups, the model accurately mirrors the public health strategies implemented during the observed influenza seasons. Detailed numerical assumptions and parameter values governing these processes are summarized in Supplementary Table\u0026nbsp;1.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eCalibration\u003c/h2\u003e\u003cp\u003eThe calibration target was the age-specific infection time series for six influenza seasons.\u003c/p\u003e\u003cp\u003eTo reconstruct age-specific infection time series for six influenza seasons, we combined this activity proxy with age-specific cumulative incidence estimated by method in Tsang et al. \u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e with using serological data, which can relax the 4-fold rise assumption.\u003c/p\u003e\u003cp\u003eTo capture the unique epidemiological dynamics of influenza in Hong Kong, our calibration process focused on several key parameters, while fixing others using the best available epidemiological evidence to prevent overfitting and ensure model identifiability. The calibrated parameters included the initial number of infections, the per-contact transmission probability, age group relative susceptibility, and the probability of staying home when sick among symptomatic individuals who seek healthcare. By systematically exploring a wide range of parameter sets, we minimized the mean squared error between the observed data and the model\u0026rsquo;s daily projections. Specifically, the calibration targeted three outcomes: cumulative infections per day, daily new infections stratified by age group, and influenza-related absences among primary and secondary school students (aged 6\u0026ndash;18).\u003c/p\u003e\u003cp\u003eWe employed the Optuna 3.2.0 hyperparameter optimization framework in Python and conducted 40,000 simulation runs for each influenza season to ensure robust parameter estimation.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eIntervention Strategies\u003c/h2\u003e\u003cp\u003eWe systematically assessed age-targeted vaccination approaches, staying home when sick, community mask-wearing, and school-based interventions. Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e summarized the strategy specifications and parameter configurations employed in our primary analysis.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eIntervention strategies\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIntervention\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTarget\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eContact layers\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eIntervention level\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eEfficacy\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eBaseline coverage\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eScenarios\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVaccination\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u0026bull; \u0026lt;\u0026thinsp;12 y\u003c/p\u003e\u003cp\u003e\u0026bull; \u0026lt;\u0026thinsp;18 y\u003c/p\u003e\u003cp\u003e\u0026bull; \u0026ge;\u0026thinsp;65 y\u003c/p\u003e\u003cp\u003e\u0026bull; Both \u0026lt;\u0026thinsp;18 y and \u0026ge;\u0026thinsp;65 y\u003c/p\u003e\u003cp\u003e\u0026bull; Universal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAll\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eIndividual\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMatched: 70%\u003c/p\u003e\u003cp\u003eMismatched: 30%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026bull; 0\u0026ndash;5 y: 7.5%-12.9%\u003c/p\u003e\u003cp\u003e\u0026bull; 6\u0026ndash;11 y: 10.8%-18.4%\u003c/p\u003e\u003cp\u003e\u0026bull; 12\u0026ndash;17 y: 5.1%-8.8%\u003c/p\u003e\u003cp\u003e\u0026bull; 18\u0026ndash;39 y:2.7%-3.3%\u003c/p\u003e\u003cp\u003e\u0026bull; 40\u0026ndash;61 y: 8.4%-9.8%\u003c/p\u003e\u003cp\u003e\u0026bull; \u0026ge;\u0026thinsp;65 y: 28.1%-32.7%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eCoverage\u0026thinsp;+\u0026thinsp;10% increments from baseline\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eStaying home when sick\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSymptomatic cases\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSchool, workplace, and community\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eIndividual\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e80%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026bull; Probability of symptomatic presentation: 80%\u003c/p\u003e\u003cp\u003e\u0026bull; Daily healthcare-seeking probability: \u0026le;15: 18.1%, 16\u0026ndash;54: 6.5%, and \u0026ge;\u0026thinsp;55: 9.5%\u003c/p\u003e\u003cp\u003e\u0026bull; Daily stay-at-home adherence probability: calibrated\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eDaily adherence probability\u0026thinsp;+\u0026thinsp;10% increments from baseline\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMask use\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUniversal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSchool, workplace, and community\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePopulation (contact layer level)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e25%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e10%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eCoverage\u0026thinsp;+\u0026thinsp;10% increments from baseline\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSchool closure\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSchool-aged children\u003c/p\u003e\u003cp\u003e(3\u0026ndash;17 y)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSchool\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePopulation (contact layer level)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e100%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eCoverage\u0026thinsp;+\u0026thinsp;10% increments from baseline; 14-day duration\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eWe explored the impact of expanding influenza vaccination coverage across specific population segments, starting from documented baseline coverage levels (Supplementary Table\u0026nbsp;3). Five distinct targeting approaches were evaluated: 1) children under 12 years; 2) individuals under 18 years; 3) adults aged 65 and above; 4) combined targeting of both young (\u0026lt;\u0026thinsp;18) and elderly (\u0026ge;\u0026thinsp;65) populations; 5) universal vaccination across all age groups. Additionally, we simulated a school-based vaccination program specifically targeting children aged 3\u0026ndash;17 years. Vaccine efficacy parameters reflected documented seasonal variation, with 70% efficacy during antigenically-matched seasons and 30% during mismatched seasons \u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e. Immunity was assumed to be acquired following vaccination or natural infection. Natural infection conferred complete protection for the remainder of the same season, whereas vaccination provided partial protection, with breakthrough infections allowed and immunity waning at a rate of 14% per year \u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn the baseline scenario, staying home when sick reflected existing behaviour, where symptomatic individuals remained at home due to illness severity. Enhanced interventions represented increased work-from-home policies and higher public adherence to stay-at-home recommendations. This behaviour was modelled using a probability function:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:{p}_{sh}\\left(i\\right)={p}_{sym}*\\left[1-{\\left(1-{p}_{hs}\\left(i\\right)*{p}_{ad}\\right)}^{{T}_{rec}-1}\\right]$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{p}_{sym}\\)\u003c/span\u003e\u003c/span\u003e is the probability of symptomatic presentation, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{p}_{hs}\\left(i\\right)\\)\u003c/span\u003e\u003c/span\u003e is the age-specific daily probability of healthcare-seeking behaviour among symptomatic individuals, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{p}_{ad}\\)\u003c/span\u003e\u003c/span\u003e is the daily probability of adherence to stay-at-home behaviour following healthcare-seeking among symptomatic individuals, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{T}_{rec}\\)\u003c/span\u003e\u003c/span\u003eis the duration from symptom onset to recovery in days. We assumed a one-day delay between symptom onset and initiation of stay-at-home behaviour. Intervention scenarios simulated enhanced adherence by increasing \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{p}_{ad}\\)\u003c/span\u003e\u003c/span\u003e relative to the baseline, which was calibrated to observed behaviour.\u003c/p\u003e\u003cp\u003eIn addition to population-wide measures, we simulated a targeted strategy focusing on school-aged children (aged 3\u0026ndash;17), reflecting potential guidance encouraging symptomatic children to stay home until recovery. Staying home when sick reduced transmission in community, workplace, and school settings, while household transmission remained unchanged as infected individuals continued interactions with household members.\u003c/p\u003e\u003cp\u003eMask-wearing was implemented at the contact layer level, modeled as a reduction in per-contact transmission probability across school, workplace, and community settings. The baseline scenario assumed 10% population coverage of mask and 25% per-contact effectiveness \u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. This corresponds to an average 2.5% reduction in transmission probability at the population level, assuming homogeneous mixing. No transmission reduction was applied within households, where mask use was considered impractical. Intervention scenarios progressively increased the proportion of mask users in the population.\u003c/p\u003e\u003cp\u003eSchool closures were modeled by reducing per-contact transmission within the school contact layer. Coverage levels corresponded to proportional reductions in transmission relative to baseline: 0% indicated no closure and 100% represented full closure. Seasons overlapping with the summer holiday period (July 15 to August 31) were excluded, as the holiday inherently reduces school-based transmission. For seasons without such overlap, we modeled 14-day reactive school closures \u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e triggered by elevated absenteeism rates among school-aged, defined as daily absenteeism exceeding 200 students.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eIntervention timing\u003c/h2\u003e\u003cp\u003eWe assumed vaccination occurred prior to influenza season onset. Implementation of staying home when sick and mask use aligned with HK CHP\u0026rsquo;s announcements of the onset of influenza seasons (Supplementary Table\u0026nbsp;4). We assessed timing sensitivity by simulating 1\u0026ndash;5 week delays in NPIs implementation after season onset.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003eComparative Analysis Framework\u003c/h2\u003e\u003cp\u003eTo optimize mitigation approaches, we implemented a layered analysis framework that progressively added interventions to baseline vaccination scenarios. This structured approach allowed for systematic assessment of the incremental benefit of each additional public health measure. Specifically, we assessed: 1) age-targeted vaccination strategies; 2) the comparative effectiveness of staying home when sick and mask use; 3) combined NPIs in vaccine-mismatched seasons; and 4) the necessity of school closures by comparing their effectiveness with other school-based interventions\u0026mdash;including vaccination and staying home when sick among school-age children\u0026mdash;given the substantial societal costs associated with school closures as a mitigation measure \u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003eSensitivity analysis\u003c/h2\u003e\u003cp\u003eTo examine the robustness of our findings under parameter uncertainty, we conducted a multi-way sensitivity analysis by jointly varying key epidemiological parameters to their respective lower and upper bounds (Supplementary Table\u0026nbsp;1). These parameters included: latent and infectious periods, age-specific transmissibility odds ratios, symptom probability, per-contact transmission weights across settings, daily contact rates, age-specific healthcare-seeking behavior, and the effectiveness of staying home when sick and mask use. Following parameter adjustment, we re-calibrated the model for each extreme scenario and repeated the main analyses using the re-calibrated models.\u003c/p\u003e\u003cp\u003eFor all analyses, each scenario underwent 30 simulations using the top 30 best-fitting parameter sets from calibration. Results are presented as median estimates with 80% projection intervals \u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e to appropriately characterize uncertainty in intervention outcomes.\u003c/p\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe demographic data, social structure data, and weekly influenza activity data used in this study are publicly available from open-access sources, as described in the Supplementary Methods. Serological data and school absenteeism records are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe code developed in the study to perform the main analysis is available in the GitHub directory at https://github.com/Liping-Peng/Influenza_ABM_HK.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis project was supported by the Theme-based Research Scheme (Project No. T11-712/19-N), General Research Fund (Project No. 17104220, 17106424 to TKT) of the Research Grants Council of the Hong Kong SAR Government, and HMRF Research Fellowship Scheme (Project No. 05190097 to TKT) from Health Bureau of the Hong Kong SAR Government. BJC is supported by an RGC Senior Research Fellowship (grant number: HKU SRFS2021-7S03) and the AIR@innoHK program of the Innovation and Technology Commission of the Hong Kong SAR Government.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePotential conflicts of interest.\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBJC reports honoraria from AstraZeneca, GlaxoSmithKline, Moderna, Roche and Sanofi Pasteur. The authors report no other potential conflicts of interest.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eHirve, S.\u003cem\u003e et al.\u003c/em\u003e Influenza Seasonality in the Tropics and Subtropics - When to Vaccinate? \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, e0153003, doi:10.1371/journal.pone.0153003 (2016).\u003c/li\u003e\n\u003cli\u003eZhao, C.\u003cem\u003e et al.\u003c/em\u003e Characterising the asynchronous resurgence of common respiratory viruses following the COVID-19 pandemic. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 1610, doi:10.1038/s41467-025-56776-z (2025).\u003c/li\u003e\n\u003cli\u003eCenter of Health Protection of The Government of the Hong Kong Special Administrative Region. \u003cem\u003eEpidemiology of seasonal influenza in Hong Kong and use of seasonal influenza vaccines\u003c/em\u003e, \u0026lt;https://www.chp.gov.hk/files/pdf/epi_of_seasonal_flu_in_hk_and_use_of_siv.pdf\u0026gt; (2024).\u003c/li\u003e\n\u003cli\u003eDepartment of Health of The Government of the Hong Kong Special Administrative Region. \u003cem\u003eVaccination programmes 2010/11 to be launched in November\u003c/em\u003e, \u0026lt;https://www.dh.gov.hk/english/press/2010/100916-2.html\u0026gt; (2010).\u003c/li\u003e\n\u003cli\u003eDepartment of Health of The Government of the Hong Kong Special Administrative Region. \u003cem\u003eVaccination subsidy schemes launched\u003c/em\u003e, \u0026lt;https://www.chp.gov.hk/en/features/19094.html\u0026gt; (2009).\u003c/li\u003e\n\u003cli\u003eThe Legislative Council of the Hong Kong Special Administrative Region. \u003cem\u003eSeasonal influenza vaccination\u003c/em\u003e, \u0026lt;https://www.legco.gov.hk/research-publications/english/essentials-1718ise06-seasonal-influenza-vaccination.htm\u0026gt; (2018).\u003c/li\u003e\n\u003cli\u003eSun, K. S.\u003cem\u003e et al.\u003c/em\u003e Seasonal influenza vaccine uptake among Chinese in Hong Kong: barriers, enablers and vaccination rates. \u003cem\u003eHum Vaccin Immunother\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 1675-1684, doi:10.1080/21645515.2019.1709351 (2020).\u003c/li\u003e\n\u003cli\u003eYuan, J.\u003cem\u003e et al.\u003c/em\u003e Parental vaccine hesitancy and influenza vaccine type preferences during and after the COVID-19 Pandemic. \u003cem\u003eCommun Med (Lond)\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 165, doi:10.1038/s43856-024-00585-w (2024).\u003c/li\u003e\n\u003cli\u003eChoi, Y. J.\u003cem\u003e et al.\u003c/em\u003e Real-world effectiveness of influenza vaccine over a decade during the 2011-2021 seasons-Implications of vaccine mismatch. \u003cem\u003eVaccine\u003c/em\u003e \u003cstrong\u003e42\u003c/strong\u003e, 126381, doi:10.1016/j.vaccine.2024.126381 (2024).\u003c/li\u003e\n\u003cli\u003eChan, M. C. W.\u003cem\u003e et al.\u003c/em\u003e Frequent Genetic Mismatch between Vaccine Strains and Circulating Seasonal Influenza Viruses, Hong Kong, China, 1996-2012. \u003cem\u003eEmerg Infect Dis\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 1825-1834, doi:10.3201/eid2410.180652 (2018).\u003c/li\u003e\n\u003cli\u003eFong, M. W.\u003cem\u003e et al.\u003c/em\u003e Nonpharmaceutical Measures for Pandemic Influenza in Nonhealthcare Settings-Social Distancing Measures. \u003cem\u003eEmerg Infect Dis\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 976-984, doi:10.3201/eid2605.190995 (2020).\u003c/li\u003e\n\u003cli\u003eLiang, M.\u003cem\u003e et al.\u003c/em\u003e Efficacy of face mask in preventing respiratory virus transmission: A systematic review and meta-analysis. \u003cem\u003eTravel Med Infect Dis\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 101751, doi:10.1016/j.tmaid.2020.101751 (2020).\u003c/li\u003e\n\u003cli\u003eBin Nafisah, S., Alamery, A. H., Al Nafesa, A., Aleid, B. \u0026amp; Brazanji, N. A. School closure during novel influenza: A systematic review. \u003cem\u003eJ Infect Public Health\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 657-661, doi:10.1016/j.jiph.2018.01.003 (2018).\u003c/li\u003e\n\u003cli\u003eRashid, H.\u003cem\u003e et al.\u003c/em\u003e Evidence compendium and advice on social distancing and other related measures for response to an influenza pandemic. \u003cem\u003ePaediatr Respir Rev\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 119-126, doi:10.1016/j.prrv.2014.01.003 (2015).\u003c/li\u003e\n\u003cli\u003eSkarp, J. E.\u003cem\u003e et al.\u003c/em\u003e A Systematic Review of the Costs Relating to Non-pharmaceutical Interventions Against Infectious Disease Outbreaks. \u003cem\u003eAppl Health Econ Health Policy\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 673-697, doi:10.1007/s40258-021-00659-z (2021).\u003c/li\u003e\n\u003cli\u003eHaldane, V.\u003cem\u003e et al.\u003c/em\u003e Strengthening the basics: public health responses to prevent the next pandemic. \u003cem\u003eBMJ\u003c/em\u003e \u003cstrong\u003e375\u003c/strong\u003e, e067510, doi:10.1136/bmj-2021-067510 (2021).\u003c/li\u003e\n\u003cli\u003eZweig, S. A., Zapf, A. J., Beyrer, C., Guha-Sapir, D. \u0026amp; Haar, R. J. Ensuring Rights while Protecting Health: The Importance of Using a Human Rights Approach in Implementing Public Health Responses to COVID-19. \u003cem\u003eHealth Hum Rights\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 173-186 (2021).\u003c/li\u003e\n\u003cli\u003eFaherty, L. J.\u003cem\u003e et al.\u003c/em\u003e Effects of non-pharmaceutical interventions on COVID-19 transmission: rapid review of evidence from Italy, the United States, the United Kingdom, and China. \u003cem\u003eFront Public Health\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 1426992, doi:10.3389/fpubh.2024.1426992 (2024).\u003c/li\u003e\n\u003cli\u003eWillem, L., Verelst, F., Bilcke, J., Hens, N. \u0026amp; Beutels, P. Lessons from a decade of individual-based models for infectious disease transmission: a systematic review (2006-2015). \u003cem\u003eBMC Infect Dis\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 612, doi:10.1186/s12879-017-2699-8 (2017).\u003c/li\u003e\n\u003cli\u003eZhang, H.\u003cem\u003e et al.\u003c/em\u003e Combinational Recommendation of Vaccinations, Mask-Wearing, and Home-Quarantine to Control Influenza in Megacities: An Agent-Based Modeling Study With Large-Scale Trajectory Data. \u003cem\u003eFront Public Health\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 883624, doi:10.3389/fpubh.2022.883624 (2022).\u003c/li\u003e\n\u003cli\u003eGuo, D.\u003cem\u003e et al.\u003c/em\u003e Multi-scale modeling for the transmission of influenza and the evaluation of interventions toward it. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 8980, doi:10.1038/srep08980 (2015).\u003c/li\u003e\n\u003cli\u003eShaman, J., Karspeck, A., Yang, W., Tamerius, J. \u0026amp; Lipsitch, M. Real-time influenza forecasts during the 2012-2013 season. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 2837, doi:10.1038/ncomms3837 (2013).\u003c/li\u003e\n\u003cli\u003eVan Kerkhove, M. D., Hirve, S., Koukounari, A., Mounts, A. W. \u0026amp; group, H. N. p. s. w. Estimating age-specific cumulative incidence for the 2009 influenza pandemic: a meta-analysis of A(H1N1)pdm09 serological studies from 19 countries. \u003cem\u003eInfluenza Other Respir Viruses\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 872-886, doi:10.1111/irv.12074 (2013).\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. \u003cem\u003eGlobal Influenza Strategy 2019\u0026ndash;2030\u003c/em\u003e, \u0026lt;https://www.who.int/publications/i/item/9789241515320\u0026gt; (2019).\u003c/li\u003e\n\u003cli\u003eKerr, C. C.\u003cem\u003e et al.\u003c/em\u003e Covasim: An agent-based model of COVID-19 dynamics and interventions. \u003cem\u003ePLoS Comput Biol\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, e1009149, doi:10.1371/journal.pcbi.1009149 (2021).\u003c/li\u003e\n\u003cli\u003eBambery, B.\u003cem\u003e et al.\u003c/em\u003e Influenza Vaccination Strategies Should Target Children. \u003cem\u003ePublic Health Ethics\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 221-234, doi:10.1093/phe/phx021 (2018).\u003c/li\u003e\n\u003cli\u003eTsang, T. K. \u0026amp; Cowling, B. J. Optimal age groups to target for influenza vaccination to reduce the impact of influenza in Hong Kong: abridged secondary publication. \u003cem\u003eHong Kong Med J\u003c/em\u003e \u003cstrong\u003e31 Suppl 3\u003c/strong\u003e, 30-33 (2025).\u003c/li\u003e\n\u003cli\u003eKing, J. C., Jr.\u003cem\u003e et al.\u003c/em\u003e Effectiveness of school-based influenza vaccination. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e355\u003c/strong\u003e, 2523-2532, doi:10.1056/NEJMoa055414 (2006).\u003c/li\u003e\n\u003cli\u003eZhu, S.\u003cem\u003e et al.\u003c/em\u003e Estimating Influenza Vaccine Effectiveness Against Laboratory-Confirmed Influenza Using Linked Public Health Information Systems, California, 2023-2024 Season. \u003cem\u003eJ Infect Dis\u003c/em\u003e, doi:10.1093/infdis/jiaf248 (2025).\u003c/li\u003e\n\u003cli\u003eMousa, A.\u003cem\u003e et al.\u003c/em\u003e Social contact patterns and implications for infectious disease transmission \u0026ndash; a systematic review and meta-analysis of contact surveys. \u003cem\u003eeLife\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, e70294, doi:10.7554/eLife.70294 (2021).\u003c/li\u003e\n\u003cli\u003eViboud, C.\u003cem\u003e et al.\u003c/em\u003e Risk factors of influenza transmission in households. \u003cem\u003eBr J Gen Pract\u003c/em\u003e \u003cstrong\u003e54\u003c/strong\u003e, 684-689 (2004).\u003c/li\u003e\n\u003cli\u003eChapman, G. B.\u003cem\u003e et al.\u003c/em\u003e Using game theory to examine incentives in influenza vaccination behavior. \u003cem\u003ePsychol Sci\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 1008-1015, doi:10.1177/0956797612437606 (2012).\u003c/li\u003e\n\u003cli\u003eBurns, A. A. C. \u0026amp; Gutfraind, A. Effectiveness of isolation policies in schools: evidence from a mathematical model of influenza and COVID-19. \u003cem\u003ePeerJ\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, e11211, doi:10.7717/peerj.11211 (2021).\u003c/li\u003e\n\u003cli\u003eJackson, C., Mangtani, P., Hawker, J., Olowokure, B. \u0026amp; Vynnycky, E. The effects of school closures on influenza outbreaks and pandemics: systematic review of simulation studies. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, e97297, doi:10.1371/journal.pone.0097297 (2014).\u003c/li\u003e\n\u003cli\u003eMartinez, D. L. \u0026amp; Das, T. K. Design of non-pharmaceutical intervention strategies for pandemic influenza outbreaks. \u003cem\u003eBMC Public Health\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 1328, doi:10.1186/1471-2458-14-1328 (2014).\u003c/li\u003e\n\u003cli\u003eFumanelli, L., Ajelli, M., Merler, S., Ferguson, N. M. \u0026amp; Cauchemez, S. Model-Based Comprehensive Analysis of School Closure Policies for Mitigating Influenza Epidemics and Pandemics. \u003cem\u003ePLoS Comput Biol\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, e1004681, doi:10.1371/journal.pcbi.1004681 (2016).\u003c/li\u003e\n\u003cli\u003eBrienen, N. C., Timen, A., Wallinga, J., van Steenbergen, J. E. \u0026amp; Teunis, P. F. The effect of mask use on the spread of influenza during a pandemic. \u003cem\u003eRisk Anal\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 1210-1218, doi:10.1111/j.1539-6924.2010.01428.x (2010).\u003c/li\u003e\n\u003cli\u003eOffeddu, V., Yung, C. F., Low, M. S. F. \u0026amp; Tam, C. C. Effectiveness of Masks and Respirators Against Respiratory Infections in Healthcare Workers: A Systematic Review and Meta-Analysis. \u003cem\u003eClin Infect Dis\u003c/em\u003e \u003cstrong\u003e65\u003c/strong\u003e, 1934-1942, doi:10.1093/cid/cix681 (2017).\u003c/li\u003e\n\u003cli\u003eTracht, S. M., Del Valle, S. Y. \u0026amp; Hyman, J. M. Mathematical modeling of the effectiveness of facemasks in reducing the spread of novel influenza A (H1N1). \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, e9018, doi:10.1371/journal.pone.0009018 (2010).\u003c/li\u003e\n\u003cli\u003eKumar, S., Quinn, S. C., Kim, K. H., Daniel, L. H. \u0026amp; Freimuth, V. S. The impact of workplace policies and other social factors on self-reported influenza-like illness incidence during the 2009 H1N1 pandemic. \u003cem\u003eAm J Public Health\u003c/em\u003e \u003cstrong\u003e102\u003c/strong\u003e, 134-140, doi:10.2105/AJPH.2011.300307 (2012).\u003c/li\u003e\n\u003cli\u003eKumar, S., Grefenstette, J. J., Galloway, D., Albert, S. M. \u0026amp; Burke, D. S. Policies to reduce influenza in the workplace: impact assessments using an agent-based model. \u003cem\u003eAm J Public Health\u003c/em\u003e \u003cstrong\u003e103\u003c/strong\u003e, 1406-1411, doi:10.2105/AJPH.2013.301269 (2013).\u003c/li\u003e\n\u003cli\u003eGlass, R. J., Glass, L. M., Beyeler, W. E. \u0026amp; Min, H. J. Targeted social distancing design for pandemic influenza. \u003cem\u003eEmerg Infect Dis\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 1671-1681, doi:10.3201/eid1211.060255 (2006).\u003c/li\u003e\n\u003cli\u003eWu, J. T., Riley, S., Fraser, C. \u0026amp; Leung, G. M. Reducing the impact of the next influenza pandemic using household-based public health interventions. \u003cem\u003ePLoS Med\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, e361, doi:10.1371/journal.pmed.0030361 (2006).\u003c/li\u003e\n\u003cli\u003eBlanchet Zumofen, M. H., Frimpter, J. \u0026amp; Hansen, S. A. Impact of Influenza and Influenza-Like Illness on Work Productivity Outcomes: A Systematic Literature Review. \u003cem\u003ePharmacoeconomics\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 253-273, doi:10.1007/s40273-022-01224-9 (2023).\u003c/li\u003e\n\u003cli\u003eLeung, N. H., Xu, C., Ip, D. K. \u0026amp; Cowling, B. J. Review Article: The Fraction of Influenza Virus Infections That Are Asymptomatic: A Systematic Review and Meta-analysis. \u003cem\u003eEpidemiology\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 862-872, doi:10.1097/EDE.0000000000000340 (2015).\u003c/li\u003e\n\u003cli\u003eDarvishian, M., Bijlsma, M. J., Hak, E. \u0026amp; van den Heuvel, E. R. Effectiveness of seasonal influenza vaccine in community-dwelling elderly people: a meta-analysis of test-negative design case-control studies. \u003cem\u003eLancet Infect Dis\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 1228-1239, doi:10.1016/S1473-3099(14)70960-0 (2014).\u003c/li\u003e\n\u003cli\u003eZhang, H.\u003cem\u003e et al.\u003c/em\u003e Combinational Recommendation of Vaccinations, Mask-Wearing, and Home-Quarantine to Control Influenza in Megacities: An Agent-Based Modeling Study With Large-Scale Trajectory Data. \u003cem\u003eFrontiers in Public Health\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, doi:10.3389/fpubh.2022.883624 (2022).\u003c/li\u003e\n\u003cli\u003eGlaubitz, A. \u0026amp; Fu, F. Social dilemma of non-pharmaceutical interventions. \u003cem\u003earXiv preprint arXiv:2404.07829\u003c/em\u003e (2024).\u003c/li\u003e\n\u003cli\u003eGao, H.\u003cem\u003e et al.\u003c/em\u003e Pandemic fatigue and attenuated impact of avoidance behaviours against COVID-19 transmission in Hong Kong by cross-sectional telephone surveys. \u003cem\u003eBMJ Open\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, e055909, doi:10.1136/bmjopen-2021-055909 (2021).\u003c/li\u003e\n\u003cli\u003eFunk, S., Andrews, M. A. \u0026amp; Bauch, C. T. Disease Interventions Can Interfere with One Another through Disease-Behaviour Interactions. \u003cem\u003ePLOS Computational Biology\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, doi:10.1371/journal.pcbi.1004291 (2015).\u003c/li\u003e\n\u003cli\u003eAjelli, M., Poletti, P., Melegaro, A. \u0026amp; Merler, S. The role of different social contexts in shaping influenza transmission during the 2009 pandemic. \u003cem\u003eScientific Reports\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 7218, doi:10.1038/srep07218 (2014).\u003c/li\u003e\n\u003cli\u003eHalder, N., Kelso, J. K. \u0026amp; Milne, G. J. Analysis of the effectiveness of interventions used during the 2009 A/H1N1 influenza pandemic. \u003cem\u003eBMC Public Health\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 168, doi:10.1186/1471-2458-10-168 (2010).\u003c/li\u003e\n\u003cli\u003eCowling, B. J.\u003cem\u003e et al.\u003c/em\u003e Protective efficacy of seasonal influenza vaccination against seasonal and pandemic influenza virus infection during 2009 in Hong Kong. \u003cem\u003eClin Infect Dis\u003c/em\u003e \u003cstrong\u003e51\u003c/strong\u003e, 1370-1379, doi:10.1086/657311 (2010).\u003c/li\u003e\n\u003cli\u003eCowling, B. J.\u003cem\u003e et al.\u003c/em\u003e Protective efficacy against pandemic influenza of seasonal influenza vaccination in children in Hong Kong: a randomized controlled trial. \u003cem\u003eClin Infect Dis\u003c/em\u003e \u003cstrong\u003e55\u003c/strong\u003e, 695-702, doi:10.1093/cid/cis518 (2012).\u003c/li\u003e\n\u003cli\u003eCowling, B. J.\u003cem\u003e et al.\u003c/em\u003e Incidence of influenza virus infections in children in Hong Kong in a 3-year randomized placebo-controlled vaccine study, 2009-2012. \u003cem\u003eClin Infect Dis\u003c/em\u003e \u003cstrong\u003e59\u003c/strong\u003e, 517-524, doi:10.1093/cid/ciu356 (2014).\u003c/li\u003e\n\u003cli\u003eIp, D. K. M.\u003cem\u003e et al.\u003c/em\u003e A Smart Card-Based Electronic School Absenteeism System for Influenza-Like Illness Surveillance in Hong Kong: Design, Implementation, and Feasibility Assessment. \u003cem\u003eJMIR Public Health Surveill\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, e67, doi:10.2196/publichealth.6810 (2017).\u003c/li\u003e\n\u003cli\u003eTsang, T. K.\u003cem\u003e et al.\u003c/em\u003e Reconstructing antibody dynamics to estimate the risk of influenza virus infection. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 1557, doi:10.1038/s41467-022-29310-8 (2022).\u003c/li\u003e\n\u003cli\u003eOrrico-S\u0026aacute;nchez, A., Valls-Ar\u0026eacute;valo, \u0026Aacute;., Garc\u0026eacute;s-S\u0026aacute;nchez, M., \u0026Aacute;lvarez Alde\u0026aacute;n, J. \u0026amp; Ortiz de Lejarazu Leonardo, R. Efficacy and effectiveness of influenza vaccination in healthy children. A review of current evidence. \u003cem\u003eEnfermedades Infecciosas y Microbiolog\u0026iacute;a Cl\u0026iacute;nica\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 396-406, doi:https://doi.org/10.1016/j.eimc.2022.02.005 (2023).\u003c/li\u003e\n\u003cli\u003eXiong, W., Cowling, B. J. \u0026amp; Tsang, T. K. Influenza Resurgence after Relaxation of Public Health and Social Measures, Hong Kong, 2023. \u003cem\u003eEmerg Infect Dis\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 2556-2559, doi:10.3201/eid2912.230937 (2023).\u003c/li\u003e\n\u003cli\u003eWong, Z. S., Goldsman, D. \u0026amp; Tsui, K. L. Economic Evaluation of Individual School Closure Strategies: The Hong Kong 2009 H1N1 Pandemic. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, e0147052, doi:10.1371/journal.pone.0147052 (2016).\u003c/li\u003e\n\u003cli\u003ePanovska-Griffiths, J.\u003cem\u003e et al.\u003c/em\u003e Determining the optimal strategy for reopening schools, the impact of test and trace interventions, and the risk of occurrence of a second COVID-19 epidemic wave in the UK: a modelling study. \u003cem\u003eLancet Child Adolesc Health\u003c/em\u003e\u003cstrong\u003e4\u003c/strong\u003e, 817-827, doi:10.1016/S2352-4642(20)30250-9 (2020).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"influenza, agent-based model, vaccination, staying home when sick, school closure, Mask-wearing ","lastPublishedDoi":"10.21203/rs.3.rs-7574768/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7574768/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSeasonal influenza control faces significant challenges from variable vaccine effectiveness and uncertain non-pharmaceutical intervention (NPI) performance across diverse contexts. While vaccination remains the primary strategy, effectiveness varies substantially between vaccine-matched and mismatched seasons. The COVID-19 pandemic has increased public acceptance of NPIs such as staying home when sick and mask use, enhancing feasibility for influenza control, yet optimal combination approaches remain poorly understood. We used an agent-based model to analyze influenza transmission across six seasons in Hong Kong (2009-2013) with varying vaccine effectiveness and epidemic characteristics. We integrated surveillance, serological, and school absenteeism data for calibration, enabling accurate estimation of reported and unreported infections. We evaluated age-targeted vaccination, staying home when sick, mask use, and school-based interventions across diverse real-world scenarios. Comparing with actual coverage levels, child vaccination consistently outperformed other strategies, with targeting those under 12 yielding the greatest population-level attack rate reduction (up to 2.17 percentage points absolute reduction and 8.5% relative reduction per 100,000 vaccinated individuals with coverage increases up to 50 percentage points). Among NPIs, 40% mask coverage reduced attack rates by 18-45%, comparable to 25% of symptomatic individuals staying home (17-49% reduction), though mask effectiveness remained stable while staying home declined with higher asymptomatic proportions. During vaccine-mismatched seasons, combining high-coverage mask use and staying home can reduce attack rates by 77-84%. School-based vaccination at high coverage was more effective than closures, reducing student attack rates by up to 86% versus 31% for 14-day closures. Our multi-source calibration approach provides robust evidence for prioritizing child vaccination and strategic NPI combinations.\u003c/p\u003e","manuscriptTitle":"Multi-Source Agent-Based Modeling to Optimize Influenza Mitigation Strategies in Hong Kong","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-24 13:04:32","doi":"10.21203/rs.3.rs-7574768/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"ccae0c34-7bb4-40f4-a420-6294b24b4e0e","owner":[],"postedDate":"September 24th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":55160591,"name":"Health sciences/Medical research/Epidemiology"},{"id":55160592,"name":"Health sciences/Diseases/Infectious diseases"}],"tags":[],"updatedAt":"2026-05-13T04:00:30+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-24 13:04:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7574768","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7574768","identity":"rs-7574768","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}