Non-pharmaceutical interventions to reduce influenza transmission in households: a systematic review and meta-analysis

25

Influenza pandemic plans often recommend non-pharmaceutical 26 interventions (NPIs) in household settings, including hand hygiene and face masks. 27 We reviewed the evidence supporting the recommendations of these measures to 28 prevent the spread of influenza in households. 29

We performed systematic reviews between 26 May and 30 August 2022 in 30 Medline, PubMed, EMBASE, and CENTRAL to identify evidence for the 31 effectiveness of selected measures recommended by representative national influenza 32 pandemic plans. We prioritized evidence from randomized controlled trials. Fixed-33 effects models were used to estimate the overall effects. Systematic reviews were 34 registered in the OSF registry (https://osf.io/8kyth). 35

We selected 9 NPIs for evidence review. We identified 9 randomized-36 controlled trials related to hand hygiene and face masks in household settings. 2 37 studies reported that measures could delay the introduction of influenza virus 38 infections into households. However, we did not identify evidence from randomized 39 controlled trials that indicated a substantial effect of hand hygiene and face masks in 40 preventing the spread of pandemic influenza within households. 41

Limited evidence indicated that within-household measures may likely 42 be effective only when implemented before or as soon as possible after symptom 43 onset in an infected case. Improving the evidence base for NPIs in households and 44 elsewhere is a continuing priority. 45 Funding: World Health Organization and the Strategic Topic Grants Scheme 46

Influenza, Non-pharmaceutical interventions, Households 47 48 49 50 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 3

51 The threat posed by the next influenza A pandemic has not diminished in the wake of 52 the COVID-19 pandemic. It is important to adapt influenza pandemic plans in light of 53 experiences from the COVID-19 pandemic. Given the delays in the availability of 54 specific vaccines and limited stockpiles of influenza antivirals in many locations, non-55 pharmaceutical interventions (NPIs) – also referred to as public health and social 56 measures – will continue to provide the first line of defense in the next influenza 57 pandemic, just as they did at the start of the COVID-19 pandemic [1]. 58 59 Influenza virus infections spread mainly through inhalation of infectious respiratory 60 particles that can occur during close contact between individuals, and one of the 61 settings responsible for a considerable fraction of all influenza transmission is 62 households [2]. In the 2009 influenza A(H1N1)pdm09 pandemic, some studies 63 estimated that around one-third of all transmission events occurred in households [3]. 64 NPIs in households could, therefore, make a major contribution to containment or 65 mitigation efforts. We reviewed the scientific evidence supporting NPIs that might be 66 recommended to reduce influenza transmission in households. 67 68

69 Selection of NPIs 70 We reviewed the websites of national public health organizations from around the 71 world to determine which NPIs might be recommended in households during 72 influenza epidemics or pandemics (Table S1). Two to three countries were selected as 73 a sample from each continent to capture snapshots of country-specific 74 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 4 recommendations for NPIs to mitigate the spread of influenza in households. From 75 this, we identified a list of NPIs that could be assessed in evidence reviews. 76 77 Search strategy 78 We then conducted systematic reviews between 26 May and 30 August 2022 to 79 evaluate the effectiveness of these selected measures on influenza virus transmission 80 in the household setting. These systematic reviews followed the Preferred Reporting 81 Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The protocol 82 was registered in the Open Science Framework (OSF) registry under the registration 83 number https://osf.io/8kyth. Four databases (Medline, PubMed, EMBASE, and 84 CENTRAL) were searched for literature in all languages with specific search terms 85 (Table S2). 86 87 Study selection 88 For each review, two authors screened titles of all papers identified by the search 89 strategy independently. Abstracts for potentially relevant papers and the full texts of 90 manuscripts were assessed for eligibility. We aimed to identify studies of the efficacy 91 of each measure against laboratory-confirmed influenza outcomes in “private” 92 household settings, and defined a private household (denoted as “household” 93 hereafter) as two or more individuals living, not necessarily related, under the same 94 unit with common housekeeping (i.e. providing food for themselves) [4]. We 95 prioritized evidence from randomized controlled trials (RCTs) as they provide the 96 highest quality of evidence. For measures with a lack of RCTs with laboratory-97 confirmed influenza outcomes, we also searched for observational studies on 98 laboratory-confirmed influenza, influenza-like illness (ILI), and respiratory illness 99 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 5 outcomes (such as acute respiratory illness or ARI). If a published systematic review 100 was identified through our search, we updated the review using pre-defined search 101 terms and evaluated literature published after the search date of the previous review. 102 Because the relative importance of modes of influenza transmission might vary in 103 different household settings, studies that were conducted in “institutional” households 104 (such as dormitories for students and homes for the elderly) whose need for shelter 105 and subsistence is being provided by a common authority were excluded. 106 107 Statistical analysis 108 Meta-analyses were performed for interventions with a sufficient number of studies. 109 The efficacy or effectiveness of measures in preventing laboratory-confirmed 110 influenza was measured by risk ratios (RRs). Overall effects were estimated in pooled 111 analyses with fixed-effects models. No overall effect was generated if there was 112 considerable heterogeneity based on an estimated I2 statistic ≥ 75%. The Appendix 113 includes additional details of the search strategies (Tables S1 and S2), selection of 114 articles (Figures S1-S9), and summaries of the selected articles (Tables S3 and S4). 115 116

117 National public health guidance on NPIs in households 118 We reviewed the websites of national public health organizations from 15 countries, 119 specifically: Ghana, Nigeria and South Africa in Africa; China, Singapore and South 120 Korea in Asia; Germany, Italy and United Kingdom in Europe; Canada and United 121 States in North America; Australia and New Zealand in Oceania; and Peru and Brazil 122 in South America (Table 1). NPIs that were implemented could be broadly 123 categorized as personal protective measures, environmental measures or other 124 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 6 measures which included measures such as hand hygiene, surface disinfection or 125 physical distancing respectively. For personal protective measures, all selected 126 countries except Germany recommended hand hygiene and respiratory etiquette in 127 household settings, while around half of the countries (e.g., China, South Korea and 128 Italy) recommended the use of face masks. None of the sampled countries 129 recommended face shields. Similarly, around half of the countries (e.g., South Africa, 130 China and Germany) recommended surface and object cleaning or ventilation or both 131 as environmental measures in household settings, and none recommended 132 humidification. Finally, all countries recommended the isolation of sick individuals 133 and physical distancing in household settings during influenza epidemics or 134 pandemics. 135 136 Country-specific recommendations on NPIs during influenza epidemics or pandemics 137 were generally disseminated through national health agency websites in the form of 138 general health information or formal guidelines for influenza (Table S1) [5, 6]. 139 Recommendations in four countries were updated after the A(H1N1)pdm09 pandemic 140 [6-8], while recommendations for the other 11 countries were updated during the 141 COVID-19 pandemic (Table S1) [9, 10]. 142 143 Systematic review of intervention studies 144 From the review of national recommendations, we constructed a list of 9 NPIs 145 including those that have been recommended and some that have not (Table 2). We 146 identified a total of 23,001 articles for title and abstract screening across the 9 NPIs 147 and 800 full-text articles were retrieved and reviewed (Figures S1–S9). For hand 148 hygiene, 576 articles were reviewed, 62 full-text articles were screened, and 7 149 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 7 intervention studies were identified for the meta-analysis. For face masks, 1,890 150 articles were reviewed, 151 full-text articles were screened, and 7 intervention studies 151 were identified for the meta-analysis. No intervention studies were identified for the 152 other 7 NPIs. After removing duplicates for studies based on hand hygiene and face 153 masks, 9 unique intervention studies were included in the review (Tables 2, S3–S4). 154 155 Personal protective measures: hand hygiene, respiratory etiquette, face masks, 156 and face shields 157 We identified seven RCTs, six of which were included in the meta-analysis, to assess 158 the efficacy of hand hygiene against transmission of laboratory-confirmed influenza 159 in household settings with at least one case, with 5,118 participants (Figure 1; Tables 160 S3) [11-13]. The study by Levy et al [12] was excluded in the meta-analysis because 161 the number of secondary influenza virus infections were reported in terms of number 162 of households instead of number of individuals. An overall pooled effect of hand 163 hygiene only, hand hygiene combined with face masks, and hand hygiene with or 164 without face masks was estimated. Results from our meta-analysis on RCTs did not 165 provide evidence to support a protective effect of hand hygiene only against 166 transmission of laboratory-confirmed influenza (RR: 1.07; 95% CI: 0.85-1.35; p-167 value: 0.58; I2 =48%). Although the pooled analysis did not identify a significant 168 effect of hand hygiene on laboratory-confirmed influenza overall, some household 169 transmission studies reported that initiating hand hygiene intervention earlier after 170 symptom onset in the index case might be more effective in preventing secondary 171 cases in the household settings [11, 13]. 172 173 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 8 In our systematic review, we identified seven RCTs that reported estimates of the 174 effectiveness of face masks in reducing laboratory-confirmed influenza virus 175 infections in household settings (Table S4) [11, 13]. Five of these trials investigated 176 the masking of all household members, regardless of symptom presentation, and we 177 were therefore unable to distinguish the potential effects of face masks worn by 178 infected vs uninfected individuals [11, 13]. Despite results not being statistically 179 significant, a trial on face masks reported a lower risk of ILI and laboratory-confirmed 180 influenza infection among those with medical mask use, and similar results were 181 reported in an earlier study. In the pooled analysis, there was no statistically 182 significant reduction in influenza transmission with the use of face masks only (RR: 183 0.59; 95% CI: 0.32-1.10; p-value 0.10; I2 =16%) (Figure 2). Study designs in the 184 seven household studies were slightly different: one trial provided face masks and P2 185 respirators for household members only, another trial evaluated the use of face masks 186 as source control for infected individuals only, and the remaining five trials provided 187 face masks for the infected individuals as well as their household members (Table S4) 188 [11, 13]. Only two household studies reported a statistically significant reduction in 189 secondary laboratory-confirmed influenza virus infections, when face masks were 190 worn within 36 hours of symptom onset [11, 13]. Most household studies were 191 underpowered due to small sample sizes, and some studies reported suboptimal 192 adherence in the face mask group. 193 194 We did not identify any published intervention studies on the effectiveness of 195 respiratory etiquette and face shields in reducing the risk of laboratory-confirmed 196 influenza in household settings. 197 198 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 9 Environmental measures: surface and object cleaning, ventilation, 199 humidification 200 We did not identify any published intervention studies that quantified the 201 effectiveness of modifying humidity, surface and object cleaning, or ventilation in 202 reducing influenza transmission in household settings. 203 204 Isolation of sick individuals and physical distancing 205 We did not identify any published intervention studies on the effectiveness of 206 isolation policies for sick individuals and physical distancing measures in reducing 207 the risk of laboratory-confirmed influenza in household settings. 208 209

210 Prevention and control of respiratory virus infections in households is an important 211 yet relatively underexplored area of research. Guidelines for infection prevention and 212 control of seasonal and pandemic influenza in healthcare settings are well established 213 [14]. During the COVID-19 pandemic, several guidelines on infection control and 214 prevention in households using NPIs were issued by health authorities alongside 215 guidance for self-care and family care. For example, the World Health Organization 216 Q&A webpage on “Home care for families and caregivers” recommends donning 217 medical masks while sharing a space with someone with COVID-19, staying at least 1 218 meter away from the sick person, and opening windows to bring fresh air into the sick 219 person’s room where possible [15]. Although the feasibility of these measures may 220 depend on living conditions, forward planning for the possibility of having a 221 household member who is sick with an infectious disease is prudent even in inter-222 pandemic periods [16]. 223 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 10 224 Among household settings, hand hygiene, face masks, respiratory etiquette, surface 225 and object cleaning and ventilation are feasible NPIs to implement during an 226 influenza epidemic or pandemic. With hand hygiene and face masks as recommended 227 hygiene practices to limit the spread of respiratory virus infections within the 228 household, the effectiveness of such measures could be enhanced through public 229 health campaigns that boost compliance [17]. Similarly, respiratory etiquette should 230 be highly feasible in household settings, and an improvement in compliance has been 231 demonstrated among school children after piloting an educational intervention in one 232 study in elementary schools [18]. It should also be feasible to implement surface and 233 object cleaning in the household due to the low cost of implementation and 234 accessibility of common household cleaning agents. Given the potential for aerosol 235 transmission of respiratory viruses including influenza [19], improving ventilation 236 should be considered except perhaps for households in areas with poor outdoor air 237 quality or when this would substantially increase heating costs. When household 238 members are sick, it should often be feasible to isolate those sick individuals and 239 increase physical distancing, for example by avoiding spending time in the same 240 rooms or eating separately with them [20], although it may be more challenging in 241 households with crowded living conditions. 242 243 In this review, we did not find evidence to support a substantial protective effect of 244 personal protective measures, environmental measures, isolation of sick individuals or 245 physical distancing measures in reducing influenza transmission in household settings. 246 Although these measures have mechanistic plausibility of reducing influenza 247 transmission based on our knowledge of how influenza is transmitted between 248 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 11 individuals [21, 22], randomized trials of hand hygiene and face marks in household 249 settings have not demonstrated protection against laboratory-confirmed influenza. 250 There were no RCTs on respiratory etiquette, face shields, modifying humidity, 251 ventilation, isolation policies for sick individuals and physical distancing in household 252 settings. 253 254 Despite a lack of intervention studies on measures other than face masks and hand 255 hygiene, we identified an observational study on the association between indoor 256 humidity and influenza transmission, suggesting a potential role of humidification in 257 controlling transmission of influenza [23] although there are also potential harms of 258 humidification which would need to be considered, such as increasing mold. Other 259 studies suggested that surface and object cleaning using common household agents, 260 indoor ventilation and voluntary self-isolation were effective measures in reducing 261 influenza transmission by inactivating influenza viruses in the environment or 262 decreasing the transmission risk [24]. Another retrospective cohort study found that 263 daily use of chlorine or ethanol-based disinfectant was effective (OR: 0.23, 95% CI: 264 0.07, 0.84) in reducing COVID-19 household transmission, and similarly for face 265 mask use (OR: 0.21; 95% CI: 0.06, 0.79) and surface disinfection when the measures 266 were implemented before symptom onset of the primary case [20]. The disinfection of 267 surfaces also has an established impact on prevention of other infectious diseases such 268 as gastrointestinal diseases [25]. 269 270 When devising strategies to reduce influenza transmission in households, it is 271 important to understand the basic transmission dynamics of influenza virus infections. 272 In the next pandemic, important information on transmission dynamics of the novel 273 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 12 strain could be provided by timely First Few Hundred studies [26] and household 274 transmission studies [27]. If the transmission dynamics of the new pandemic strain are 275 similar to that of H1N1pdm09 and current interpandemic strains, we can note the 276 following four properties. First, infectiousness is thought to peak at around the same 277 time as when symptoms appear [28]. Second, infectiousness likely declines rapidly 278 within a few days after peak based on viral culture data [29] despite viral RNA 279 continuing to be detectable by PCR typically for more than a week [30]. Third, only a 280 fraction of influenza virus infections result in fever, and while fever and cough may 281 be a relatively more specific syndrome for influenza, it is not particularly sensitive in 282 the general community as contrasted with its higher sensitivity in individuals who 283 seek medical attention with respiratory symptoms [31]. Fourth, the role of 284 asymptomatic and pre-symptomatic transmission has been controversial but recent 285 reports from South Africa [32] and Hong Kong [33] indicate that these may comprise 286 a substantial fraction of all influenza transmission, with asymptomatic and pre-287 symptomatic transmission also playing an important role in COVID-19 transmission 288 [34]. This fundamental knowledge of infectiousness profiles would imply that early 289 intervention is essential to reduce transmission, and early intervention should not be 290 limited to individuals with a fever and cough but could be triggered by other less 291 specific symptoms. Rapid antigen tests done in the household could help to 292 distinguish influenza from other viral infections and might even be considered for use 293 in exposed individuals to identify influenza virus infection before any symptoms 294 appear. 295 296 There are a number of limitations to our review. First, in our analysis of the 297 effectiveness of face masks and hand hygiene we did not review observational data as 298 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 13 a higher level of evidence from randomized controlled trials were available. Other 299 studies have reviewed observational data and concluded that these two measures 300 likely have small to moderate effects on transmission [35]. Second, we focused on 301 measures to prevent the spread of influenza within the household in this review. There 302 is limited evidence on the degree of reductions in transmission in households when 303 personal protective measures (e.g., wearing face masks plus frequent hand hygiene) 304 are used in combination with other measures like isolation of sick household members. 305 The effectiveness of different cleaning products at different concentrations in 306 deactivating or eliminating influenza virus in household environments remains 307 unclear. Third, increased influenza activity is associated with cold temperatures, low 308 indoor humidity and rainy seasons [36]. Further investigation could clarify the 309 effectiveness of NPIs by different seasonal patterns (such as indoor crowding during 310 colder months). Finally, we observed low to moderate levels of heterogeneity in our 311 meta-analyses of hand hygiene and face masks (Figures 1-2). We could determine 312 whether these differences were artefactual or real, perhaps related to differences in the 313 adherence of measures in various populations or the time delay between symptom 314 onset of an infected case and the implementation of a measure [37]. Further work 315 could attempt to identify additional factors that explain this heterogeneity, for 316 example, by exploring very different estimates of effectiveness of measures based on 317 the same population during similar time periods, or conducting subgroup analyses by 318 the time delay between symptom onset and measure implementation. 319 320 Improved evidence is needed on all of the measures included in our review. Given the 321 effect sizes in our meta-analysis of hand hygiene and face masks (Figures 1 and 2), 322 any future RCTs of these interventions in households would likely need to be very 323 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 14 large to be adequately powered to detect a relative reduction in the risk of infection of 324 approximately 10% [38]. To avoid contamination of interventions, cluster randomized 325 trials, in which each household is randomized to receive either the intervention or 326 control, could be used to assess the effect of the intervention in reducing the 327 transmission of influenza in households [11]. A promising area for randomized trials 328 or cluster-randomized trials in the household setting is the effect of physical 329 distancing on influenza transmission, either by alternating within-home isolation 330 strategies or comparing the feasibility and effectiveness of physical distancing in 331 housing areas with different population densities. Surveys about the feasibility of each 332 measure in local contexts are also important to inform national-level 333 recommendations on home care and/or voluntary self-isolation or quarantine [39]. 334 335 In conclusion, although our study found limited evidence to support a substantial 336 protective effect of personal protective measures, environmental measures, isolation 337 of sick persons or physical distancing measures in controlling influenza transmission 338 in the household setting, these measures have mechanistic plausibility based on our 339 knowledge of person-to-person transmission of influenza [21, 22]. Future 340 investigations on transmission dynamics of influenza would be helpful in preparing 341 guidelines and evidence-based recommendations for household transmission in the 342 next pandemic. Although our review focused on NPIs to be used during influenza 343 pandemics, these results could also be applicable to intense seasonal influenza 344 epidemics. 345 346 347 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 15 ACKNOWLEDGMENTS 348 The authors thank Julie Au for administrative support. 349 350 DATA SHARING STATEMENT 351 The data that support the findings of this study are available upon request. 352 353 FUNDING SOURCES 354 This project was supported by a grant from the World Health Organization and the 355 Strategic Topic Grants Scheme (Project No. STG4/M-701/23-N) of the Research 356 Grants Council of the Hong Kong Special Administrative Region, China. 357 358 POTENTIAL CONFLICTS OF INTEREST 359 B.J.C. has consulted for AstraZeneca, Fosun Pharma, GlaxoSmithKline, Haleon, 360 Moderna, Novavax, Pfizer, Roche, and Sanofi Pasteur. All other authors report no 361 potential conflicts of interest. 362 363 CONTRIBUTIONS 364 All authors meet the ICMJE criteria for authorship. The study was conceived by BJC, 365 and JYW. JYW and JKC analyzed the data. JYW wrote the first draft of the 366 manuscript. All authors provided critical review and revision of the text and approved 367 the final version. 368 369 370 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 16

371 [1] Lai S, Ruktanonchai NW, Zhou L, Prosper O, Luo W, Floyd JR, et al. Effect 372 of non-pharmaceutical interventions to contain COVID-19 in China. Nature. 373 2020;585(7825):410-3. 374 [2] Hayden FG, Belshe R, Villanueva C, Lanno R, Hughes C, Small I, et al. 375 Management of influenza in households: a prospective, randomized comparison of 376 oseltamivir treatment with or without postexposure prophylaxis. J Infect Dis. 377 2004;189(3):440-9. 378 [3] Ghani A, Baguelin M, Griffin J, Flasche S, van Hoek AJ, Cauchemez S, et al. 379 The Early Transmission Dynamics of H1N1pdm Influenza in the United Kingdom. 380 PLoS Curr. 2009;1:RRN1130. 381 [4] The Organisation for Economic Co-operation and Development. Household 382 definitions in other statistical standards [Available from: https://www.oecd-383 ilibrary.org/docserver/9789264194878-18-384 en.pdf?expires=1721030289&id=id&accname=guest&checksum=A47802DDF2046D385 9A7DBC0F2529D80432]. (Accessed 16 July 2024). 386 [5] Ministry of Health. Influenza [Available from: 387 https://www.moh.gov.sg/diseases-updates/influenza]. (Accessed 26 January 2024). 388 [6] Nigeria Centre for Disease Control and Prevention. Nigeria National 389 Pandemic Influenza Preparedness and Response Plan [Available from: 390 https://ncdc.gov.ng/themes/common/docs/protocols/16_1490369515.pdf]. (Accessed 391 28 January 2024). 392 [7] Communicable Diseases Network Australia. Seasonal Influenza Infection 393 CDNA National Guidelines for Public Health Units [Available from: 394 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 17 https://www.health.gov.au/sites/default/files/documents/2022/06/influenza-infection-395 flu-cdna-national-guidelines-for-public-health-units.pdf]. (Accessed 5 April 2024). 396 [8] Ghana Food and Drugs Authority. Frequently Asked Questions on the H1N1 397 Influenza Virus [Available from: 398 https://www.fdaghana.gov.gh/img/reports/H1N1%20FAQs.pdf]. (Accessed 5 April 399 2024). 400 [9] Public Health Agency of Canada. Flu (influenza): Prevention and risks 401 [Available from: https://www.canada.ca/en/public-health/services/diseases/flu-402 influenza/prevention-risks.html]. (Accessed 26 January 2024). 403 [10] Centers for Disease Control and Prevention. Preventive Actions to Help 404 Protect Against Flu [Available from: https://www.cdc.gov/flu/prevent/actions-405 prevent-flu.htm]. (Accessed 5 April 2024). 406 [11] Cowling BJ, Chan KH, Fang VJ, Cheng CK, Fung RO, Wai W, et al. 407 Facemasks and hand hygiene to prevent influenza transmission in households: a 408 cluster randomized trial. Ann Intern Med. 2009;151(7):437-46. 409 [12] Levy JW, Suntarattiwong P, Simmerman JM, Jarman RG, Johnson K, Olsen 410 SJ, et al. Increased hand washing reduces influenza virus surface contamination in B 411 angkok households, 2009–2010. Influenza Other Respir Viruses. 2014;8(1):13–6. 412 [13] Suess T, Remschmidt C, Schink SB, Schweiger B, Nitsche A, Schroeder K, et 413 al. The role of facemasks and hand hygiene in the prevention of influenza 414 transmission in households: results from a cluster randomised trial; Berlin, Germany, 415 2009-2011. BMC Infect Dis. 2012;12:26. 416 [14] Centers for Disease Control and Prevention. Prevention Strategies for 417 Seasonal Influenza in Healthcare Settings [Available from: 418 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 18 https://www.cdc.gov/flu/professionals/infectioncontrol/healthcaresettings.htm]. 419 (Accessed 25 January 2024). 420 [15] World Health Organization. Coronavirus disease (COVID-19): Home care for 421 families and caregivers [Available from: https://www.who.int/news-room/questions-422 and-answers/item/coronavirus-disease-covid-19-home-care-for-families-and-423 caregivers]. (Accessed 25 January 2024). 424 [16] Geddes L. COVID-19 in the house: How to reduce the risk of transmission 425 [Available from: https://www.gavi.org/vaccineswork/covid-19-house-how-reduce-426 risk-427 transmission?gclid=EAIaIQobChMIzdmHvsOfgwMV_thMAh1WsQ2oEAMYASAA428 EgJT7PD_BwE]. (Accessed 25 January 2024). 429 [17] Islam M, Benjamin-Chung J, Sultana S, Unicomb L, Alam M, Rahman M, et 430 al. Effectiveness of Mass Media Campaigns to Improve Handwashing-Related 431 Behavior, Knowledge, and Practices in Rural Bangladesh. Am J Trop Med Hyg. 432 2021;104(4):1546-53. 433 [18] Sultana F, Nizame FA, Southern DL, Unicomb L, Winch PJ, Luby SP. Pilot of 434 an Elementary School Cough Etiquette Intervention: Acceptability, Feasibility, and 435 Potential for Sustainability. Am J Trop Med Hyg. 2017;97(6):1876-85. 436 [19] Tellier R, Li Y, Cowling BJ, Tang JW. Recognition of aerosol transmission of 437 infectious agents: a commentary. BMC Infect Dis. 2019;19(1):101. 438 [20] Wang Y, Tian H, Zhang L, Zhang M, Guo D, Wu W, et al. Reduction of 439 secondary transmission of SARS-CoV-2 in households by face mask use, disinfection 440 and social distancing: a cohort study in Beijing, China. BMJ Glob Health. 2020;5(5). 441 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 19 [21] Leung NHL, Chu DKW, Shiu EYC, Chan KH, McDevitt JJ, Hau BJP, et al. 442 Respiratory virus shedding in exhaled breath and efficacy of face masks. Nat Med. 443 2020;26(5):676-80. 444 [22] Grayson ML, Melvani S, Druce J, Barr IG, Ballard SA, Johnson PD, et al. 445 Efficacy of soap and water and alcohol-based hand-rub preparations against live 446 H1N1 influenza virus on the hands of human volunteers. Clin Infect Dis. 447 2009;48(3):285-91. 448 [23] Petrie JG, Ohmit SE, Cowling BJ, Johnson E, Cross RT, Malosh RE, et al. 449 Influenza transmission in a cohort of households with children: 2010-2011. PLoS One. 450 2013;8(9):e75339. 451 [24] Aganovic A, Bi Y, Cao G, Kurnitski J, Wargocki P. Modeling the impact of 452 indoor relative humidity on the infection risk of five respiratory airborne viruses. Sci 453 Rep. 2022;12(1):11481. 454 [25] Sandora TJ, Shih MC, Goldmann DA. Reducing absenteeism from 455 gastrointestinal and respiratory illness in elementary school students: a randomized, 456 controlled trial of an infection-control intervention. Pediatrics. 2008;121(6):e1555-62. 457 [26] Marcato AJ, Black AJ, Walker CR, Morris D, Meagher N, Price DJ, et al. 458 Learnings from the Australian first few X household transmission project for COVID-459 19. Lancet Reg Health West Pac. 2022;28:100573. 460 [27] Tsang TK, Lau LLH, Cauchemez S, Cowling BJ. Household Transmission of 461 Influenza Virus. Trends Microbiol. 2016;24(2):123-33. 462 [28] Tsang TK, Cowling BJ, Fang VJ, Chan KH, Ip DK, Leung GM, et al. 463 Influenza A Virus Shedding and Infectivity in Households. J Infect Dis. 464 2015;212(9):1420-8. 465 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 20 [29] Ip DKM, Lau LLH, Chan KH, Fang VJ, Leung GM, Peiris MJS, et al. The 466 Dynamic Relationship Between Clinical Symptomatology and Viral Shedding in 467 Naturally Acquired Seasonal and Pandemic Influenza Virus Infections. Clin Infect 468 Dis. 2016;62(4):431-7. 469 [30] Hirose R, Daidoji T, Naito Y, Watanabe Y, Arai Y, Oda T, et al. Long-term 470 detection of seasonal influenza RNA in faeces and intestine. Clin Microbiol Infect. 471 2016;22(9):813 e1- e7. 472 [31] Monto AS, Gravenstein S, Elliott M, Colopy M, Schweinle J. Clinical signs 473 and symptoms predicting influenza infection. Arch Intern Med. 2000;160(21):3243-7. 474 [32] Cohen C, Kleynhans J, Moyes J, McMorrow ML, Treurnicht FK, Hellferscee 475 O, et al. Asymptomatic transmission and high community burden of seasonal 476 influenza in an urban and a rural community in South Africa, 2017-18 (PHIRST): a 477 population cohort study. Lancet Glob Health. 2021;9(6):e863-e74. 478 [33] Tsang TK, Wang C, Fang VJ, Perera R, So HC, Ip DKM, et al. Reconstructing 479 household transmission dynamics to estimate the infectiousness of asymptomatic 480 influenza virus infections. Proc Natl Acad Sci U S A. 2023;120(33):e2304750120. 481 [34] Gao W, Lv J, Pang Y, Li LM. Role of asymptomatic and pre-symptomatic 482 infections in covid-19 pandemic. BMJ. 2021;375:n2342. 483 [35] Gozdzielewska L, Kilpatrick C, Reilly J, Stewart S, Butcher J, Kalule A, et al. 484 The effectiveness of hand hygiene interventions for preventing community 485 transmission or acquisition of novel coronavirus or influenza infections: a systematic 486 review. BMC Public Health. 2022;22(1):1283. 487 [36] Tamerius J, Nelson MI, Zhou SZ, Viboud C, Miller MA, Alonso WJ. Global 488 influenza seasonality: reconciling patterns across temperate and tropical regions. 489 Environ Health Perspect. 2011;119(4):439-45. 490 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 21 [37] Trevas S, Manuel K, Malkani R, Hoelscher D. Mask Adherence and Social 491 Distancing in Houston, TX from January to April 2021. Int J Environ Res Public 492 Health. 2023;20(3). 493 [38] Cowling BJ, Leung GM. Face masks and COVID-19: don't let perfect be the 494 enemy of good. Euro Surveill. 2020;25(49). 495 [39] Ahmed F, Qualls N, Kowalczyk S, Randolph Cunningham S, Zviedrite N, 496 Uzicanin A. Feasibility, acceptability, and barriers to implementing select non-497 pharmaceutical interventions to reduce the transmission of pandemic influenza - 498 United States, 2019. Disaster Med Public Health Prep. 2022;17:e209. 499 500 . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 22 Table 1: Recommendations of household-related non-pharmaceutical interventions in different countries. Continent Country Personal protective measures Environmental measures Other measures Hand hygiene Respiratory etiquette Face masks Face shields Surface and object cleaning Ventilation Humidification Isolation of sick individuals Physical distancing Africa Ghana /checkbld /checkbld /checkbld /checkbld Nigeria /checkbld /checkbld /checkbld /checkbld South Africa /checkbld /checkbld /checkbld /checkbld /checkbld /checkbld Asia China /checkbld /checkbld /checkbld /checkbld /checkbld /checkbld /checkbld Singapore /checkbld /checkbld /checkbld /checkbld South Korea /checkbld /checkbld /checkbld /checkbld /checkbld Europe Germany /checkbld /checkbld /checkbld /checkbld Italy /checkbld /checkbld /checkbld /checkbld /checkbld United Kingdom /checkbld /checkbld /checkbld /checkbld North America Canada /checkbld /checkbld /checkbld /checkbld /checkbld United States /checkbld /checkbld /checkbld /checkbld /checkbld /checkbld /checkbld Oceania Australia /checkbld /checkbld /checkbld /checkbld /checkbld New Zealand /checkbld /checkbld /checkbld /checkbld /checkbld South America Brazil /checkbld /checkbld /checkbld /checkbld /checkbld Peru /checkbld /checkbld /checkbld /checkbld /checkbld /checkbld . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 23 Table 2: Summary of literature searches for systematic review on non-pharmaceutical interventions in household settings for influenza. Type of measures No. of studies identified Main findings Hand hygiene 7 The evidence from the RCTs suggested that hand hygiene intervention only did not exert substantial effects on influenza household transmission. However, implementing hand hygiene and face mask at early symptom onset of index patients is effective in reducing secondary transmission of viruses. Respiratory etiquette 0 No study examining the effectiveness of respiratory etiquette on influenza transmission in household settings was found. Face masks 7 The evidence from the RCTs suggested that wearing face masks had an effect on reducing influenza household transmission when it was implemented before or at early symptom onset of index patients. Face shields 0 No study examining the effectiveness of face shields on influenza transmission in household settings was found. Surface and object cleaning 0 No study examining the effectiveness of surface and object cleaning on influenza transmission in household settings was found. Ventilation 0 No study examining the effectiveness of ventilation on influenza transmission in household settings was found. Humidification 0 No study examining the effectiveness of humidification on influenza household transmission was found. . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 24 Isolation of sick individuals 0 No study examining the effectiveness of isolation of sick individuals on influenza household transmission was found. Physical distancing 0 No study examining the effectiveness of physical distancing on influenza household transmission was found. RCT: randomized controlled trial. . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint 25 FIGURE LEGENDS Figure 1: Meta-analysis of risk ratios for the effect of hand hygiene with or without face mask use on laboratory-confirmed influenza from 6 randomized controlled trials with 5,118 participants. (A) Hand hygiene alone; (B) Hand hygiene and face mask; (C) Hand hygiene with or without face mask. Pooled estimates were not generated if there was high heterogeneity ( I 2 /g341075%). Squares indicate risk ratio for each of the included studies, horizontal line indicates 95% CIs, dashed vertical line indicates pooled estimation of risk ratio, and diamond indicates pooled estimation of risk ratio. Diamond width corresponds to the 95% CI. The study by Levy et al was excluded in the meta-analysis but included in the review as its number of secondary infections are measured in households instead of participants [12]. Figure 2: Meta-analysis of risk ratios for the effect of face mask use with or without hand hygiene on laboratory-confirmed influenza from 7 randomized controlled trials with 4,247 participants. (A) Face mask use alone; (B) Face mask and hygiene; (C) Face mask with or without hand hygiene. Pooled estimates were not generated if there was high heterogeneity (I 2/g341075%). Squares indicate risk ratio for each of the included studies, horizontal line indicates 95% CIs, dashed vertical line indicates pooled estimation of risk ratio, and diamond indicates pooled estimation of risk ratio. Diamond width corresponds to the 95% CI. . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint Author (Year) Fixed effect model Heterogeneity: I2 = 48%, τ2 = 0.0807, p = 0.11 Test for overall effect: z = 0.55 (p = 0.58) Cowling (2008) Cowling (2009) Larson (2010) Ram (2015) Simmerman (2011) Events 5 14 29 9 66 Total 1675 84 257 946 96 292 Hand hygiene Events 12 28 24 4 58 Total 1807 205 279 904 117 302 Control Weight 100.0% 5.9% 22.6% 20.6% 3.0% 47.9% Risk Ratio 1.07 1.02 0.54 1.15 2.74 1.18 95% C.I. [0.85; 1.35] [0.37; 2.80] [0.29; 1.01] [0.68; 1.97] [0.87; 8.63] [0.86; 1.61] 0.2 0.5 1 2 5 Risk Ratio Favors Hand Hygiene Favors Control Hand hygiene only Author (Year) Fixed effect model Heterogeneity: I2 = 28%, τ2 = 0.0324, p = 0.24 Test for overall effect: z = −0.27 (p = 0.79) Cowling (2009) Larson (2010) Simmerman (2011) Suess (2012) Events 18 25 66 10 Total 1554 258 938 291 67 Hand hygiene Events 28 24 58 19 Total 1567 279 904 302 82 Control Weight 100.0% 21.5% 19.5% 45.4% 13.6% Risk Ratio 0.97 0.70 1.00 1.18 0.64 95% C.I. [0.77; 1.22] [0.39; 1.23] [0.58; 1.74] [0.86; 1.62] [0.32; 1.29] 0.5 1 2 Risk Ratio Favors Hand Hygiene Favors Control Hand hygiene with mask Author (Year) Fixed effect model Heterogeneity: I2 = 49%, τ2 = 0.0616, p = 0.08 Test for overall effect: z = 0.16 (p = 0.87) Cowling (2008) Cowling (2009) Larson (2010) Ram (2015) Simmerman (2011) Suess (2012) Events 5 32 54 9 132 10 Total 3229 84 515 1884 96 583 67 Hand hygiene Events 12 28 24 4 58 19 Total 1889 205 279 904 117 302 82 Control Weight 100.0% 4.0% 21.0% 18.8% 2.1% 44.2% 9.9% Risk Ratio 1.02 1.02 0.62 1.08 2.74 1.18 0.64 95% C.I. [0.84; 1.24] [0.37; 2.80] [0.38; 1.01] [0.67; 1.73] [0.87; 8.63] [0.89; 1.55] [0.32; 1.29] 0.2 0.5 1 2 5 Risk Ratio Favors Hand Hygiene Favors Control Hand hygiene with or without mask . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint Author (Year) Fixed effect model Heterogeneity: I2 = 16%, τ2 = 0.2375, p = 0.31 Test for overall effect: z = −1.66 (p = 0.10) Cowling (2008) MacIntyre (2009) MacIntyre (2016) Suess (2012) Events 4 1 0 6 Total 526 61 94 302 69 Mask use Events 12 0 1 19 Total 682 205 100 295 82 Control Weight 100.0% 22.1% 1.9% 6.1% 69.8% Risk Ratio 0.59 1.12 3.19 0.33 0.38 95% C.I. [0.32; 1.10] [0.37; 3.35] [0.13; 77.36] [0.01; 7.96] [0.16; 0.89] 0.1 0.5 1 2 10 Risk Ratio Favors Mask Favors Control Mask only Author (Year) Fixed effect model Heterogeneity: I2 = 28%, τ2 = 0.0324, p = 0.24 Test for overall effect: z = −0.27 (p = 0.79) Cowling (2009) Larson (2010) Simmerman (2011) Suess (2012) Events 18 25 66 10 Total 1554 258 938 291 67 Mask use Events 28 24 58 19 Total 1567 279 904 302 82 Control Weight 100.0% 21.5% 19.5% 45.4% 13.6% Risk Ratio 0.97 0.70 1.00 1.18 0.64 95% C.I. [0.77; 1.22] [0.39; 1.23] [0.58; 1.74] [0.86; 1.62] [0.32; 1.29] 0.5 1 2 Risk Ratio Favors Mask Favors Control Mask with hand hygiene Author (Year) Fixed effect model Heterogeneity: I2 = 27%, τ2 = 0.0640, p = 0.22 Test for overall effect: z = −0.58 (p = 0.56) Cowling (2008) Cowling (2009) Larson (2010) MacIntyre (2009) MacIntyre (2016) Simmerman (2011) Suess (2012) Events 4 18 25 1 0 66 16 Total 2080 61 258 938 94 302 291 136 Mask use Events 12 28 24 0 1 58 19 Total 2167 205 279 904 100 295 302 82 Control Weight 100.0% 3.9% 19.3% 17.5% 0.3% 1.1% 40.8% 17.0% Risk Ratio 0.94 1.12 0.70 1.00 3.19 0.33 1.18 0.51 95% C.I. [0.75; 1.17] [0.37; 3.35] [0.39; 1.23] [0.58; 1.74] [0.13; 77.36] [0.01; 7.96] [0.86; 1.62] [0.28; 0.93] 0.1 0.5 1 2 10 Risk Ratio Favors Mask Favors Control Mask with or without hand hygiene . CC-BY-NC 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.(which was not certified by peer review)preprint The copyright holder for thisthis version posted September 10, 2024. ; https://doi.org/10.1101/2024.09.10.24313390doi: medRxiv preprint