South Africa has experienced five waves of SARS-CoV-2 infection, with over 4 million laboratory-confirmed cases by August 2022 (National Institute for Communicable Diseases, 2022a). The true burden is highly underestimated, since based on seroprevalence data, after the third wave of infection, 43 to 83% of the 59.5 million South African inhabitants had already been infected, varying by age and setting (Kleynhans et al., 2022a; Bingham et al., 2022).

SARS-CoV-2 transmission is mainly via the respiratory route, through both droplet-mediated and airborne transmission (Meyerowitz et al., 2021; Wang et al., 2021). Infection from contaminated surfaces has also been described (Meyerowitz et al., 2021). Although infection risk is highest in symptomatic individuals (Madewell et al., 2020), with the most infectious period one day before symptom onset (Meyerowitz et al., 2021), asymptomatic individuals can still transmit SARS-CoV-2 (Liu, 2019; Cohen et al., 2022). Households are a focal point for SARS-CoV-2 transmission (Aleta et al., 2022; Hsu et al., 2021), especially during peaks of non-pharmaceutical intervention (NPI) restrictions, when movement outside of the household was limited (Aleta et al., 2022). Transmission within households can in turn lead to spillover to the community (Nande et al., 2021).

Prior to the widespread availability of SARS-CoV-2 vaccines, most countries relied on NPIs to reduce the transmission of the virus, including wearing face masks, social and physical distancing. While mobility and contact survey data showed that the implementation of NPIs led to a reduction in community contacts (Aleta et al., 2022; Liu et al., 2021) and in turn opportunity for infection, it is still unknown what the role of contact patterns are in the transmission of SARS-CoV-2 in the household. Most analysis relating contact patterns and SARS-CoV-2 transmission done to date has been based on low-resolution data collected from contact tracing (McAloon et al., 2021), mobility data (Aleta et al., 2022), and contact surveys (Liu et al., 2021). To obtain high-resolution contact data, devices broadcasting and receiving radio frequency waves can be used to measure the frequency and duration of close-proximity contacts. This has been used previously to collect contact data in among others, schools (Salathé et al., 2010), workplaces (Cattuto et al., 2010), hospitals (Voirin et al., 2015), and households (Kiti et al., 2016), which can, in turn, be used for modeling disease transmission. Specifically, for SARS-CoV-2 so far, high-resolution contact data were collected on cruise ships to identify areas of high contact and to investigate the usefulness of NPIs (Pung et al., 1956).

Understanding the drivers of SARS-CoV-2 transmission in the household, especially contact patterns, can help inform NPIs for future SARS-CoV-2 resurgences and potentially future emerging pathogens with pandemic potential. We aimed to assess the association of household close-range contact patterns with the transmission of SARS-CoV-2 in the household using proximity sensors deployed after the identification of SARS-CoV-2 in the household.

We screened 1531 individuals and identified 277 (18%) positive for SARS-CoV-2, of which 124 (45%) were enrolled and included in the household cumulative infection risk analysis (Kleynhans et al., 2022b), with 373 household contacts. After data cleaning, we had contact data for 88 (71%) index cases and 252 (68%) household contacts (Figure 1). Ninety-three individuals (19%, 36 index cases, and 73 household contacts) were excluded due to non-compliance, where no contacts were logged, or sensors were stationary for the period based on accelerometer data. We were more likely to have contact data for individuals from the Soweto site, from larger households, and with no household member reporting smoking indoors (Table 1). The median number of household members included in the analysis was 4 (interquartile range [IQR] 3–5), with a median of 3 (IQR 1–4) SARS-CoV-2 cases per household and a median of 67% (IQR 50–100%) of household members infected (including index cases). Sixty-six percent (225/340) of individuals included in the analysis lived in a household with 3–5 members, and 49% (168/340) lived in a home with only 1–2 rooms used for sleeping, a third (53/340) living in households where crowding was reported (>2 people per sleeping room Table 1).

Figure 1

Download asset Open asset

The overall median daily and maximum duration of close-range proximity events was 18 min (IQR 9–45 min) and 61 min (IQR 25–142 min), respectively. The average duration per close-range proximity event was 0.7 min (IQR 0.5–0.8 min), with a median of 26 (IQR 10–58) close-range proximity events per day amongst household members (Figures 2 and 3, Table 2). The highest median daily contact duration was observed between individuals within the <5 year, 5–12 year and 35–59 year groups (Figure 4A and D). Similar patterns were also seen for median daily close-range proximity duration and frequency in children aged 5–12 and 13–17 years (Figure 4B–F).

Figure 2

Download asset Open asset

Figure 3

Download asset Open asset

Figure 4

Download asset Open asset

We did not find any association between any of the contact parameters (either with the index case or all SARS-CoV-2 infected household members) and SARS-CoV-2 infection in the household using the Wilcoxon rank-sum test (p-values ranging from 0.1–0.8, Table 3).

When assessing factors associated with SARS-CoV-2 transmission from presumptive index cases and acquisition in household members, none of the contact parameters were associated with SARS-CoV-2 transmission on univariate analysis. Sleeping in the same room as the index case was also not associated with transmission (OR 0.94, 95% CI 0.47–1.88). On multivariable analysis after controlling for index age and SARS-CoV-2 infecting variant, factors significantly associated with higher SARS-CoV-2 transmission and acquisition was index case minimum C_t value <30 (aOR 16.8 95% CI 3.1–93.1) compared to C_t >35, and female contacts (aOR 2.5 95% CI 1.3–5.02). No contact parameters with the index case were associated with acquisition (Table 4). Similar results were observed in the sensitivity analysis when households with members seropositive at baseline were excluded (Table 5).

When not considering transmission from the presumptive index case, but rather all SARS-CoV-2 infected household members, factors significantly associated with SARS-CoV-2 acquisition on multivariable analysis after controlling for age and SARS-CoV-2 infecting variant were being obese (aOR 4.1 95% CI 1.5–11.1) compared to normal weight, and not currently smoking (aOR 3.2 95% CI 1.2–9.2). No contact parameters with SARS-CoV-2 infected household members were associated with acquisition (Table 6).

In this case-ascertained, prospective household transmission study we did not find an association between the duration and frequency of close-range proximity events with SARS-CoV-2 infected household members and transmission in the household.

High-resolution contact patterns have been previously used in the context of pathogen transmission. Examples include investigating influenza virus transmission routes in a hospital setting (Whitney, 2016), and contact surveys to show the association between contacts, locations, and influenza infection (Kwok et al., 2014). For bacterial infections, the high correlation between pneumococcal infection risk and contact behavior has been shown (Qian et al., 2022), and in the context of tuberculosis transmission, it was shown that contact with adults is more important than contact with children (Dodd et al., 2016). To our knowledge, there are few data available on the direct association of close-range proximity events and SARS-CoV-2, and none make use of high-resolution contact data. During contact tracing efforts early in the pandemic in Singapore, it was found that sharing a bedroom with an index case and speaking to the index case for 30 min or longer increased the risk for infection (Ng et al., 2021). We did not see similar results when assessing sharing a bedroom with the index case, and this may be due to the already high level of crowding in included households. Although we observed an increase in infection risk with higher average contact durations with the index case on univariate analysis, this association was no longer seen when adjusting for age and other index and contact factors associated with transmission/acquisition. Mobile device geolocation has also been used to predict contact events between individuals on population level, and was used in transmission models to predict case numbers (Crawford et al., 2022). However, we did not find close-range proximity events to be an important driver for household transmission.

There are several possible reasons why we did not observe an association between close-range proximity events and SARS-CoV-2 transmission; these can be classified as related to transmission dynamics or study limitations. One possibility is that along with droplet-mediated transmission during close-proximity contacts, airborne (Meyerowitz et al., 2021; Wang et al., 2021), and to a lesser extent, fomite-mediated transmission (Meyerowitz et al., 2021) may also play a role in the transmission of SARS-CoV-2 in the household. More evidence is becoming available showing that aerosol transmission may be a more important transmission route for SARS-CoV-2 than initially anticipated, especially so in poorly ventilated indoor environments (Wang et al., 2021; Duval et al., 2022). Households in these communities do not have central air-conditioning or heating (Mathee et al., 2021), and during the winter months ventilation may be poorer than in summer, although we did not measure this. Furthermore, sensors only measure face-to-face interactions, and if individuals were close to each other but not directly facing one another for extended durations, we would not have measured this, although sharing of the same air may have occurred. The ventilation within households should be considered in future studies, as this can be a target for intervention strategies to reduce secondary transmission. The high level of interaction in relatively crowded South African households may already be above the threshold for transmission risk, with host characteristics like index viral load and contact age being more important to determine infection risk in this context. It is of interest that close-range proximity patterns within the household did not fully account for the differences in transmission based on age; with teenagers and adults experiencing the highest infection risk, but children aged 5–17 years having the highest contacts.

Our study had limitations both in design and execution. Due to the nature of the case-ascertained study design, we would have missed the period when the index case was most infectious, just before symptom onset (Meyerowitz et al., 2021), and the close-range contact patterns measured during the study may have been different after the household members were aware of the index SARS-CoV-2 case (leading to reduced contact), and again once secondary cases were informed of their infection status (leading to increased contact). We also did not collect any information on possible NPI usage in the households, like wearing masks. A study from South Africa showed that individuals staying at home were less likely to wear a mask (Burger et al., 2022), but these data were not ascertained during a time when a household member was infected with SARS-CoV-2. We also did not consider where contacts took place (indoors or outdoors), which relates to ventilation and may have influenced transmission. We may have also misclassified the true index case if they were asymptomatic, and did not consider tertiary transmission chains in the index-directed analysis. To adjust for possible misclassification, we performed a grouped assessment investigating close-range proximity events with all SARS-CoV-2 infected household members. This grouped analysis may also have diluted possible associations with the true infector. Furthermore, although based on legislation, close contacts (including household contacts) of SARS-CoV-2 cases were supposed to quarantine, compliance was not monitored. Therefore, household contacts could have been exposed to non-household SARS-CoV-2 cases during the follow-up period. We did not consider multiple introductions of SARS-CoV-2 within the household, although we did exclude households with more than one SARS-CoV-2 variant detected. During the peaks of waves of infection in South Africa, one variant was responsible for the majority of the infections (National Institute for Communicable Diseases, 2022b), and the additional introductions within the household were likely to have been the same as the initial variant. Higher resolution sequencing data may be useful to more accurately identify chains of transmission within the household. Combining contact data with clinical and virological/bacteriological data has been shown to be useful to reconstruct transmission networks (Campbell et al., 2019), and we will consider this for future analyses. Our measurement of close-range contact patterns was also limited by compliance, as during the cleaning process we identified 73 sensors that were not worn, based on accelerometer data. We also had limited data in some households, where some individuals did not consent to the contact aspect of the study, or where we were unable to retrieve data due to hardware failure, lost, or damaged tags. The small sample size may have reduced our power to detect small differences in close proximity event parameters between those infected with SARS-CoV-2 and those not infected.

In conclusion, we did not observe an association between close-proximity contacts and SARS-CoV-2 transmission in the household. A case-ascertained, prospective household transmission study may not be well suited to investigate this question. A possible other study design to consider is randomly selected prospective household cohorts, but the deployment of sensors for extended periods of time may be logistically challenging and lead to participant fatigue, and households in a cohort may not experience infection episodes unless the community attack rate is very high. High-resolution contacts in other settings like schools or workplaces where contacts are less frequent could be useful to identify the type of contact events that may lead to SARS-CoV-2 transmission. If aerosol transmission plays a more important role in transmission than droplet-mediated transmission, ventilation within households can also be an important consideration for future studies. Increased ventilation could potentially be a method to reduce secondary transmission in households. Nevertheless, our study provides high-resolution household contact data that can be used to parametrize future transmission models, not only for SARS-CoV-2, but other pathogens as well.

The contact network, selected individual characteristics and analysis script are available at https://github.com/crdm-nicd/sashts (copy archived at Kleynhans, 2023).

1. 1. Aleta A
   2. Martín-Corral D
   3. Bakker MA
   4. Pastore Y Piontti A
   5. Ajelli M
   6. Litvinova M
   7. Chinazzi M
   8. Dean NE
   9. Halloran ME
   10. Longini IM Jr
   11. Pentland A
   12. Vespignani A
   13. Moreno Y
   14. Moro E

   (2022) Quantifying the importance and location of SARS-Cov-2 transmission events in large metropolitan areas

   PNAS 119:e2112182119.

   https://doi.org/10.1073/pnas.2112182119

   • PubMed
   • Google Scholar
2. Preprint

   1. Bingham J
   2. Cable R
   3. Coleman C
   4. Glatt TN
   5. Grebe E
   6. Mhlanga L
   7. Nyano C
   8. Pieterson N
   9. Swanevelder R
   10. Swarts A
   11. Sykes W
   12. van den Berg K
   13. Vermeulen M
   14. Welte A

   (2022) Estimates of prevalence of anti-sars-cov-2 antibodies among blood donors in south africa in march 2022

   Research Square.

   https://doi.org/10.21203/rs.3.rs-1687679/v1

   • Google Scholar
3. 1. Burger R
   2. Christian C
   3. English R
   4. Maughan-Brown B
   5. Rossouw L

   (2022) Predictors of mask-wearing during the advent of the COVID-19 pandemic: evidence from South Africa

   Translational Behavioral Medicine 12:ibab132.

   https://doi.org/10.1093/tbm/ibab132

   • Google Scholar
4. 1. Campbell F
   2. Cori A
   3. Ferguson N
   4. Jombart T

   (2019) Bayesian inference of transmission chains using timing of symptoms, pathogen Genomes and contact data

   PLOS Computational Biology 15:e1006930.

   https://doi.org/10.1371/journal.pcbi.1006930

   • PubMed
   • Google Scholar
5. 1. Cattuto C
   2. Van den Broeck W
   3. Barrat A
   4. Colizza V
   5. Pinton J-F
   6. Vespignani A

   (2010) Dynamics of person-to-person interactions from distributed RFID sensor networks

   PLOS ONE 5:e11596.

   https://doi.org/10.1371/journal.pone.0011596

   • PubMed
   • Google Scholar
6. 1. Cohen C
   2. Kleynhans J
   3. von Gottberg A
   4. McMorrow ML
   5. Wolter N
   6. Bhiman JN
   7. Moyes J
   8. du Plessis M
   9. Carrim M
   10. Buys A
   11. Martinson NA
   12. Kahn K
   13. Tollman S
   14. Lebina L
   15. Wafawanaka F
   16. du Toit JD
   17. Gómez-Olivé FX
   18. Dawood FS
   19. Mkhencele T
   20. Sun K
   21. Viboud C
   22. Tempia S
   23. PHIRST-C Group

   (2022) SARS-Cov-2 incidence, transmission, and Reinfection in a rural and an urban setting: results of the PHIRST-C cohort study, South Africa, 2020-21

   The Lancet. Infectious Diseases 22:821–834.

   https://doi.org/10.1016/S1473-3099(22)00069-X

   • PubMed
   • Google Scholar
7. 1. Crawford FW
   2. Jones SA
   3. Cartter M
   4. Dean SG
   5. Warren JL
   6. Li ZR
   7. Barbieri J
   8. Campbell J
   9. Kenney P
   10. Valleau T
   11. Morozova O

   (2022) Impact of close interpersonal contact on COVID-19 incidence: evidence from 1 year of mobile device data

   Science Advances 8:eabi5499.

   https://doi.org/10.1126/sciadv.abi5499

   • PubMed
   • Google Scholar
8. 1. Dodd PJ
   2. Looker C
   3. Plumb ID
   4. Bond V
   5. Schaap A
   6. Shanaube K
   7. Muyoyeta M
   8. Vynnycky E
   9. Godfrey-Faussett P
   10. Corbett EL
   11. Beyers N
   12. Ayles H
   13. White RG

   (2016) Age- and sex-specific social contact patterns and incidence of Mycobacterium tuberculosis infection

   American Journal of Epidemiology 183:156–166.

   https://doi.org/10.1093/aje/kwv160

   • PubMed
   • Google Scholar
9. 1. Duval D
   2. Palmer JC
   3. Tudge I
   4. Pearce-Smith N
   5. O’Connell E
   6. Bennett A
   7. Clark R

   (2022) Long distance airborne transmission of SARS-Cov-2: rapid systematic review

   BMJ 377:e068743.

   https://doi.org/10.1136/bmj-2021-068743

   • PubMed
   • Google Scholar
10. 1. Hsu C-Y
    2. Wang J-T
    3. Huang K-C
    4. Fan AC-H
    5. Yeh Y-P
    6. Chen SL-S

    (2021) Household transmission but without the community-acquired outbreak of COVID-19 in Taiwan

    Journal of the Formosan Medical Association = Taiwan Yi Zhi 120 Suppl 1:S38–S45.

    https://doi.org/10.1016/j.jfma.2021.04.021

    • PubMed
    • Google Scholar
11. 1. Kiti MC
    2. Tizzoni M
    3. Kinyanjui TM
    4. Koech DC
    5. Munywoki PK
    6. Meriac M
    7. Cappa L
    8. Panisson A
    9. Barrat A
    10. Cattuto C
    11. Nokes DJ

    (2016) Quantifying social contacts in a household setting of rural Kenya using Wearable proximity sensors

    EPJ Data Science 5:21.

    https://doi.org/10.1140/epjds/s13688-016-0084-2

    • PubMed
    • Google Scholar
12. 1. Kleynhans J
    2. Tempia S
    3. Wolter N
    4. von Gottberg A
    5. Bhiman JN
    6. Buys A
    7. Moyes J
    8. McMorrow ML
    9. Kahn K
    10. Gómez-Olivé FX
    11. Tollman S
    12. Martinson NA
    13. Wafawanaka F
    14. Lebina L
    15. du Toit JD
    16. Jassat W
    17. Neti M
    18. Brauer M
    19. Cohen C
    20. PHIRST-C Group

    (2022a) SARS-Cov-2 Seroprevalence after third wave of infections, South Africa

    Emerging Infectious Diseases 28:1055–1058.

    https://doi.org/10.3201/eid2805.220278

    • PubMed
    • Google Scholar
13. 1. Kleynhans J
    2. Walaza S
    3. Martinson NA
    4. Neti M
    5. Gottberg AV
    6. Bhiman JN
    7. Toi D
    8. Amoako DG
    9. Buys A
    10. Ndlangisa K
    11. Wolter N
    12. Genade L
    13. Maloma L
    14. Chewparsad J
    15. Lebina L
    16. de Gouveia L
    17. Kotane R
    18. Tempia S
    19. Cohen C

    (2022b) Household transmission of SARS-Cov-2 from adult index cases living with and without HIV in South Africa, 2020-2021: A case-ascertained, prospective observational household transmission study

    Epidemiology 1:ciac640.

    https://doi.org/10.1101/2022.04.08.22273160

    • Google Scholar
14. Software

    1. Kleynhans J

    (2023) Sashts, version swh:1:rev:7b43d0ceebc66b4fe879ff6aa314e9d725ca4bd1

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:8f451277d4d6305b260019815d5b6cfcd411417e;origin=https://github.com/crdm-nicd/sashts;visit=swh:1:snp:901bc34776c6ea0fab2695a1b08078c06682e9f1;anchor=swh:1:rev:7b43d0ceebc66b4fe879ff6aa314e9d725ca4bd1
15. 1. Kwok KO
    2. Cowling BJ
    3. Wei VWI
    4. Wu KM
    5. Read JM
    6. Lessler J
    7. Cummings DA
    8. Peiris JSM
    9. Riley S

    (2014) Social contacts and the locations in which they occur as risk factors for influenza infection

    Proceedings of the Royal Society B 281:20140709.

    https://doi.org/10.1098/rspb.2014.0709

    • Google Scholar
16. Preprint

    1. Liu T

    (2019) Transmission dynamics of 2019 novel coronavirus

    bioRxiv.

    https://doi.org/10.1101/2020.01.25.919787

    • Google Scholar
17. 1. Liu CY
    2. Berlin J
    3. Kiti MC
    4. Del Fava E
    5. Grow A
    6. Zagheni E
    7. Melegaro A
    8. Jenness SM
    9. Omer SB
    10. Lopman B
    11. Nelson K

    (2021) Rapid review of social contact patterns during the COVID-19 pandemic

    Epidemiology 32:781–791.

    https://doi.org/10.1097/EDE.0000000000001412

    • PubMed
    • Google Scholar
18. 1. Madewell ZJ
    2. Yang Y
    3. Longini IM
    4. Halloran ME
    5. Dean NE

    (2020) Household transmission of SARS-Cov-2: A systematic review and meta-analysis

    JAMA Network Open 3:e2031756.

    https://doi.org/10.1001/jamanetworkopen.2020.31756

    • PubMed
    • Google Scholar
19. 1. Mathee A
    2. Moyes J
    3. Mkhencele T
    4. Kleynhans J
    5. Language B
    6. Piketh S
    7. Moroe E
    8. Wafawanaka F
    9. Martinson N
    10. McMorrow M
    11. Tempia S
    12. Kahn K
    13. Cohen C

    (2021) Housing quality in a rural and an urban settlement in South Africa

    International Journal of Environmental Research and Public Health 18:2240.

    https://doi.org/10.3390/ijerph18052240

    • PubMed
    • Google Scholar
20. 1. McAloon CG
    2. Wall P
    3. Butler F
    4. Codd M
    5. Gormley E
    6. Walsh C
    7. Duggan J
    8. Murphy TB
    9. Nolan P
    10. Smyth B
    11. O’Brien K
    12. Teljeur C
    13. Green MJ
    14. O’Grady L
    15. Culhane K
    16. Buckley C
    17. Carroll C
    18. Doyle S
    19. Martin J
    20. More SJ

    (2021) Numbers of close contacts of individuals infected with SARS-Cov-2 and their association with government intervention strategies

    BMC Public Health 21:2238.

    https://doi.org/10.1186/s12889-021-12318-y

    • PubMed
    • Google Scholar
21. 1. Meyerowitz EA
    2. Richterman A
    3. Gandhi RT
    4. Sax PE

    (2021) Transmission of SARS-Cov-2: A review of viral, host, and environmental factors

    Annals of Internal Medicine 174:69–79.

    https://doi.org/10.7326/M20-5008

    • PubMed
    • Google Scholar
22. 1. Nande A
    2. Adlam B
    3. Sheen J
    4. Levy MZ
    5. Hill AL

    (2021) Dynamics of COVID-19 under social distancing measures are driven by transmission network structure

    PLOS Computational Biology 17:e1008684.

    https://doi.org/10.1371/journal.pcbi.1008684

    • PubMed
    • Google Scholar
23. Report

    1. National Institute for Communicable Diseases

    (2022a) COVID-19 weekly epidemiology brief

    National Institute for Communicable Diseases.

    https://www.nicd.ac.za/diseases‐a‐z‐index/disease‐index‐covid‐19/surveillance‐reports/weekly‐epidemiological‐brief/

    • Google Scholar
24. Website

    1. National Institute for Communicable Diseases

    (2022b) Tracking SARS-CoV-2 Variants

    Accessed July 15, 2022.

    https://www.nicd.ac.za/diseases-a-z-index/disease-index-covid-19/sars-cov-2-genomic-surveillance-update/
25. 1. Ng OT
    2. Marimuthu K
    3. Koh V
    4. Pang J
    5. Linn KZ
    6. Sun J
    7. De Wang L
    8. Chia WN
    9. Tiu C
    10. Chan M
    11. Ling LM
    12. Vasoo S
    13. Abdad MY
    14. Chia PY
    15. Lee TH
    16. Lin RJ
    17. Sadarangani SP
    18. Chen MI-C
    19. Said Z
    20. Kurupatham L
    21. Pung R
    22. Wang L-F
    23. Cook AR
    24. Leo Y-S
    25. Lee VJ

    (2021) SARS-Cov-2 Seroprevalence and transmission risk factors among high-risk close contacts: a retrospective cohort study

    The Lancet. Infectious Diseases 21:333–343.

    https://doi.org/10.1016/S1473-3099(20)30833-1

    • PubMed
    • Google Scholar
26. 1. Pung R
    2. Firth JA
    3. Spurgin LG
    4. Chang A
    5. Kong J
    6. Wong J
    7. Jin OJ
    8. Selvaraj D
    9. Yong D
    10. Lang J
    11. Sivalingam A
    12. Procter SR
    13. Flasche S
    14. Waites W
    15. Prem K
    16. Pearson CAB
    17. Gibbs HP
    18. Sherratt K
    19. Villabona-Arenas CJ
    20. Wong KLM
    21. Liu Y
    22. Mee P
    23. Chapman LAC
    24. Atkins KE
    25. Quaife M
    26. Munday JD
    27. Funk S
    28. Eggo RM
    29. Huè S
    30. Davies NG
    31. Hodgson D
    32. Abbas K
    33. McCarthy CV
    34. Hellewell J
    35. Abbott S
    36. Bosse NI
    37. Brady O
    38. Barnard RC
    39. Jit M
    40. Tully DC
    41. Medley G
    42. Sun FY
    43. Jarvis CI
    44. Lowev R
    45. O’Reilly K
    46. Meakin SR
    47. Endo A
    48. Sandmann FG
    49. Edmunds WJ
    50. Koltai M
    51. Finch E
    52. Gimma A
    53. Rosello A
    54. Quilty BJ
    55. Jafari Y
    56. Knight GM
    57. Clifford S
    58. Russell TW
    59. Lee VJ
    60. Kucharski AJ
    61. Singapore CruiseSafe working group
    62. CMMID COVID-19 working group

    (1956) Using high-resolution contact networks to evaluate SARS-Cov-2 transmission and control in large-scale multi-day events

    Nature Communications 13:29522-y.

    https://doi.org/10.1038/s41467-022-29522-y

    • Google Scholar
27. 1. Qian G
    2. Toizumi M
    3. Clifford S
    4. Le LT
    5. Papastylianou T
    6. Satzke C
    7. Quilty B
    8. Iwasaki C
    9. Kitamura N
    10. Takegata M
    11. Bui MX
    12. Nguyen HAT
    13. Dang DA
    14. van Hoek AJ
    15. Yoshida LM
    16. Flasche S

    (2022) Association of Pneumococcal carriage in infants with the risk of carriage among their contacts in Nha Trang, Vietnam: A nested cross-sectional survey

    PLOS Medicine 19:e1004016.

    https://doi.org/10.1371/journal.pmed.1004016

    • PubMed
    • Google Scholar
28. 1. Salathé M
    2. Kazandjieva M
    3. Lee JW
    4. Levis P
    5. Feldman MW
    6. Jones JH

    (2010) A high-resolution human contact network for infectious disease transmission

    PNAS 107:22020–22025.

    https://doi.org/10.1073/pnas.1009094108

    • Google Scholar
29. 1. Torresi J
    2. Edeling MA
    3. Nolan T
    4. Godfrey DI

    (2022) A complementary Union of SARS-Cov-2 natural and vaccine induced immune responses

    Frontiers in Immunology 13:914167.

    https://doi.org/10.3389/fimmu.2022.914167

    • PubMed
    • Google Scholar
30. 1. Voirin N
    2. Payet C
    3. Barrat A
    4. Cattuto C
    5. Khanafer N
    6. Régis C
    7. Kim B-A
    8. Comte B
    9. Casalegno J-S
    10. Lina B
    11. Vanhems P

    (2015) Combining high-resolution contact data with virological data to investigate influenza transmission in a tertiary care hospital

    Infection Control and Hospital Epidemiology 36:254–260.

    https://doi.org/10.1017/ice.2014.53

    • PubMed
    • Google Scholar
31. 1. Wang CC
    2. Prather KA
    3. Sznitman J
    4. Jimenez JL
    5. Lakdawala SS
    6. Tufekci Z
    7. Marr LC

    (2021) Airborne transmission of respiratory viruses

    Science 373:eabd9149.

    https://doi.org/10.1126/science.abd9149

    • PubMed
    • Google Scholar
32. 1. Whitney CG

    (2016) Changing epidemiology of Pneumococcal disease in the era of conjugate vaccines

    Current Epidemiology Reports 3:125–135.

    https://doi.org/10.1007/s40471-016-0077-5

    • Google Scholar
33. 1. Wibmer CK
    2. Ayres F
    3. Hermanus T
    4. Madzivhandila M
    5. Kgagudi P
    6. Oosthuysen B
    7. Lambson BE
    8. de Oliveira T
    9. Vermeulen M
    10. van der Berg K
    11. Rossouw T
    12. Boswell M
    13. Ueckermann V
    14. Meiring S
    15. von Gottberg A
    16. Cohen C
    17. Morris L
    18. Bhiman JN
    19. Moore PL

    (2021) SARS-Cov-2 501Y.V2 escapes neutralization by South African COVID-19 donor plasma

    Nature Medicine 27:622–625.

    https://doi.org/10.1038/s41591-021-01285-x

    • PubMed
    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "Association of close-range contact patterns with SARS-CoV-2: a household transmission study" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by Neil Ferguson as the Senior Editor. The following individual involved in the review of your submission has agreed to reveal their identity: Virginia E. Pitzer (Reviewer #3).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

This is a strong and generally well-presented study. We request revisions to help strengthen the impact of the work.

1) Please see the numerous small questions and points of confusion described by all three reviewers. Please revise the manuscript accordingly, unless you believe the peer review (which included cross-review) reflects an unusual degree of misreading.

2) Please consider some of the alternative metrics for close contacts suggested by reviewers 2 and 3 or expand the discussion of how alternate contact parameters might lead to different results, potentially supporting a greater role for droplet transmission (or not).

3) Please choose more reasonable units for the ORs (see reviewer 3) and interpret them in the text.

Reviewer #1 (Recommendations for the authors):

Excellent job making the code and data available.

It was not clear to me when seropositive individuals were excluded if they truly "disappeared" from the analysis, e.g., excluded from the household size statistics.

As I alluded to in the public review, I'm not clear on what adjustments for age (or even variant) are really doing here. Gender presumably could be a proxy of time at home. How much confidence is there in the identification of previously infected individuals? As I also alluded to in the public review, I am curious if the impact of proximity becomes more pronounced in wholly susceptible households, or if any of the other associations change substantially (although I recognize there will be less power).

I was surprised to read the statement that droplet transmission is "thought to be the most important [respiratory transmission] route." Maybe (infamously) by the WHO? The claim seems to ignore major work by Jimenez, Prather, Marr, Milton et al. by the time of submission; early outbreak investigations (planes, workplaces, etc.); and decades of prior work demonstrating a very large role for aerosol transmission with measles, tuberculosis, influenza, etc. I do not think the study would at all be weakened by acknowledging the strong evidence for aerosol transmission.

Ll. 153-154: I was confused by the 5-min time slice exclusion. If people were still, the contact was excluded.

l. 195: I found "(number of nodes)" confusing.

Reviewer #2 (Recommendations for the authors):

I have the following comments and suggestions:

1. On the 3rd page, Methods, the authors mentioned "The resulting total sample size was 440 exposed household members". However, on the 6th page, Results, the authors mentioned "identified 277 (18%) positive for SARS-CoV-2, of which 124 (45%) were enrolled and included in the household cumulative infection risk analysis".

This comparison may suggest that the actual number of study participants was much smaller than the target sample size. The authors may need to clarify this.

2. Lines 167 – 177, the authors mainly considered three types of contact parameters: including (1) duration (median daily cumulative time in contact in seconds), (2) frequency (median daily number of contacts with the index/infected individuals over the deployment period), and (3) average duration (cumulative time in contact divided by the cumulative number of close-range proximity events over full deployment period).

I'm not sure if the median measurements of the contact duration, frequency, and average duration are representative of the daily human behaviour within the household settings. For example, the duration of a formal lunch or dinner can be longer than one hour. Could you include more contact parameters such as long-duration contacts in your analysis?

3. It's also not clear if the household members could contact other people within the same community. For example, did you require the study participants to stay at home during the study's 2-week period? Could you adjust the community force of infection in your regression model?

4. In lines 109-110, the authors mentioned "with {greater than or equal to}2 household contacts of whom none reported symptoms prior to index case onset". It's not clear why the authors needed this requirement.

5. In lines 163-165, the authors mentioned "For the analysis, we only considered close-range proximity events that occurred one day after deployment and one day before collection".

Do you mean the proximity events occurred between deployment and collection?

Essential revisions:

This is a strong and generally well-presented study. We request revisions to help strengthen the impact of the work.

We thank the editors for the positive feedback.

1) Please see the numerous small questions and points of confusion described by all three reviewers. Please revise the manuscript accordingly, unless you believe the peer review (which included cross-review) reflects an unusual degree of misreading.

Please see our responses to individual comments below in green. Line numbers refer to the tracked version of the manuscript.

2) Please consider some of the alternative metrics for close contacts suggested by reviewers 2 and 3 or expand the discussion of how alternate contact parameters might lead to different results, potentially supporting a greater role for droplet transmission (or not).

We added four additional parameters: The maximum frequency and duration of close contact events – to capture the extreme end of the spectrum; the daily average close proximity event frequency (cumulative number of events / deployment days), and the cumulative time in contact (normalised by deployment days).

3) Please choose more reasonable units for the ORs (see reviewer 3) and interpret them in the text.

All time-based close proximity event contact parameters are now displayed in minutes, instead of seconds.

Reviewer #1 (Recommendations for the authors):

Excellent job making the code and data available.

It was not clear to me when seropositive individuals were excluded if they truly "disappeared" from the analysis, e.g., excluded from the household size statistics.

To clarify, we updated the final sentence of the Screening, enrolment and follow-up Methods section (line 130-132, page 4): “Individuals were considered seropositive if they tested positive on either assay. Individuals sero-positive at the start of follow-up with no rRT-PCR confirmed SARS-CoV-2 infection during follow-up were excluded from the risk factor analysis for household SARS-CoV-2 acquisition as they may have been protected from infection, but were still considered in the household size parameter.”

As I alluded to in the public review, I'm not clear on what adjustments for age (or even variant) are really doing here. Gender presumably could be a proxy of time at home. How much confidence is there in the identification of previously infected individuals? As I also alluded to in the public review, I am curious if the impact of proximity becomes more pronounced in wholly susceptible households, or if any of the other associations change substantially (although I recognize there will be less power).

We added another sensitivity analysis where any households where individuals were excluded due to seropositivity was excluded from the analysis. The results (Table 5) showed that none of the contact patterns were significantly associated with SARS-CoV-2 transmission in the household.

I was surprised to read the statement that droplet transmission is "thought to be the most important [respiratory transmission] route." Maybe (infamously) by the WHO? The claim seems to ignore major work by Jimenez, Prather, Marr, Milton et al. by the time of submission; early outbreak investigations (planes, workplaces, etc.); and decades of prior work demonstrating a very large role for aerosol transmission with measles, tuberculosis, influenza, etc. I do not think the study would at all be weakened by acknowledging the strong evidence for aerosol transmission.

Updated statement as follows (Introduction, line 70-71, page 3): “SARS-CoV-2 transmission is mainly via the respiratory route, through both droplet-mediated and airborne transmission.”

Ll. 153-154: I was confused by the 5-min time slice exclusion. If people were still, the contact was excluded.

Added line 157-157, page 5 to clarify: “Accelerometers are very sensitive and even a slight movement will be detected, therefore contacts that occurred while individuals were sitting/standing still will still be included.”

l. 195: I found "(number of nodes)" confusing.

Fragment of sentence removed, line 212 page 6: “For the analysis with close-range proximity events with all SARS-CoV-2 infected household members, we included an offset term in the model to account for the number of SARS-CoV-2 infected members in contact with.

Reviewer #2 (Recommendations for the authors):

I have the following comments and suggestions:

1. On the 3rd page, Methods, the authors mentioned "The resulting total sample size was 440 exposed household members". However, on the 6th page, Results, the authors mentioned "identified 277 (18%) positive for SARS-CoV-2, of which 124 (45%) were enrolled and included in the household cumulative infection risk analysis".

This comparison may suggest that the actual number of study participants was much smaller than the target sample size. The authors may need to clarify this.

We have acknowledged this in the limitations: “Discussion, line 345-347, page 9: “The small sample size may have reduced our power to detect small differences in close proximity event parameters between those infected with SARS-CoV-2 and those not infected.”

2. Lines 167 – 177, the authors mainly considered three types of contact parameters: including (1) duration (median daily cumulative time in contact in seconds), (2) frequency (median daily number of contacts with the index/infected individuals over the deployment period), and (3) average duration (cumulative time in contact divided by the cumulative number of close-range proximity events over full deployment period).

I'm not sure if the median measurements of the contact duration, frequency, and average duration are representative of the daily human behaviour within the household settings. For example, the duration of a formal lunch or dinner can be longer than one hour. Could you include more contact parameters such as long-duration contacts in your analysis?

We added four additional parameters: The maximum frequency and duration of close-range proximity events – to capture the extreme end of the spectrum; the daily average close-range proximity event frequency (cumulative number of events / deployment days), and the cumulative time in contact (normalised by deployment days). The methods section, line 173-185 (page 5) has been updated with the details: “We assessed the following close-range proximity event parameters: (1) median daily duration (median of cumulative duration of close-range proximity events for each day of deployment, in minutes), (2) maximum duration (longest duration of a close-range proximity event during deployment, in minutes), (3) median frequency (median of number of close proximity events for each day of deployment), (4) maximum frequency (highest number of close proximity events in one day during deployment), (5) median average daily duration (median of cumulative duration of close-range proximity events in the day divided by the cumulative number of close-range proximity events during that day), (6) daily average contact frequency (cumulative duration of close-range proximity events over deployment period divided by the cumulative number of close-range proximity events during the deployment period), and (7) cumulative time in contact (cumulative duration of close-range proximity events over deployment period divided by number of days sensor worn).”

3. It's also not clear if the household members could contact other people within the same community. For example, did you require the study participants to stay at home during the study's 2-week period? Could you adjust the community force of infection in your regression model?

To clarify on participants staying at home during the period we added to the Discussion section line 331-334, page 8-9: “Furthermore, although based on legislation, close contacts (including household contacts) of SARS-CoV-2 cases were supposed to quarantine, compliance was not monitored. Therefore, household contacts could have been exposed to non-household SARS-CoV-2 cases during the follow-up period.”

Our current analytic framework does not allow for the adjustment of the community force of infection. We will be considering this in future analyses.

4. In lines 109-110, the authors mentioned "with {greater than or equal to}2 household contacts of whom none reported symptoms prior to index case onset". It's not clear why the authors needed this requirement.

We updated lines 108-112 (page 3) of the methods to clarify: “We enrolled household contacts of SARS-CoV-2 infected individuals identified through screening (presumptive index) with ≥2 household contacts (for efficient investigation of risk factors for transmission in the household, weighting cost of household visits and data collected) of whom none reported symptoms prior to index case onset (reducing probability of previous recent SARS-CoV-2 infection in the household).”

5. In lines 163-165, the authors mentioned "For the analysis, we only considered close-range proximity events that occurred one day after deployment and one day before collection".

Do you mean the proximity events occurred between deployment and collection?

To clarify we updated line 168-171 (page 5) of the methods section: “For the analysis, we only considered close-range proximity events that occurred one day after sensors were issued and one day before collection, hence excluding any false events logged when sensors were prepared, handed out and collected in the household.”

Author details

1. Jackie Kleynhans

   1. Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
   2. School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review and editing

   For correspondence

   jackiel@nicd.ac.za

   Competing interests

   reports institutional grant funding from UK Foreign, Commonwealth and Development Office, Wellcome Trust, WHO AFRO, Africa Pathogen Genomics Initiative (Africa PGI) through African Society of Laboratory Medicine (ASLM) and Africa CDC, US Centers for Disease Control and Prevention, South African Medical Research Council (SAMRC) and UK Department of Health and Social Care and managed by the Fleming Fund: SEQAFRICA project. The author has no other competing interests to declare

   "This ORCID iD identifies the author of this article:" 0000-0001-7081-6273

2. Lorenzo Dall'Amico

   ISI Foundation, Turin, Italy

   Contribution

   Conceptualization, Formal analysis, Methodology, Writing – review and editing

   Competing interests

   received support from the European Union's Horizon 2a020 research and innovation programme under grant agreements No. 101003688 (EpiPose) and No. 101016233 (PERISCOPE). The author acknowledges support from the Lagrange Project of ISI Foundation funded by CRT Foundation. They were awarded a PhD contract with Grenoble INP, and received support for attending meetings/travel from the ISP Foundation and Grenoble INP. The author has no other competing interests to declare

   "This ORCID iD identifies the author of this article:" 0000-0002-7493-6421

3. Laetitia Gauvin

   1. ISI Foundation, Turin, Italy
   2. Institute for Research on Sustainable Development, Aubervilliers, France

   Contribution

   Conceptualization, Formal analysis, Methodology, Writing – review and editing

   Competing interests

   has received Payment or honoraria for lectures from the University of Torino, Italy. They received support from the Lagrange Project of ISI Foundation funded by CRT Foundation. The authors has no other competing interests to declare

4. Michele Tizzoni

   1. ISI Foundation, Turin, Italy
   2. Department of Sociology and Social Research, University of Trento, Trento, Italy

   Contribution

   Conceptualization, Formal analysis, Methodology, Writing – review and editing

   Competing interests

   acknowledges support from the Lagrange Project of ISI Foundation funded by CRT Foundation

5. Lucia Maloma

   Perinatal HIV Research Unit, University of the Witwatersrand, Johannesburg, South Africa

   Contribution

   Data curation, Investigation, Methodology, Project administration, Writing – review and editing

   Competing interests

   No competing interests declared

6. Sibongile Walaza

   1. Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
   2. School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa

   Contribution

   Conceptualization, Resources, Formal analysis, Investigation, Methodology, Writing – review and editing

   Competing interests

   reports institutional grant funding from WHO AFRO, Africa Pathogen Genomics Initiative (Africa PGI) through African Society of Laboratory Medicine (ASLM) and Africa CDC, US Centers for Disease Control and Prevention, South African Medical Research Council (SAMRC), UK Department of Health and Social Care and managed by the Fleming Fund: SEQAFRICA project and the Wellcome Trust. The author has no other competing interests to declare

7. Neil A Martinson

   1. Perinatal HIV Research Unit, University of the Witwatersrand, Johannesburg, South Africa
   2. Johns Hopkins University Center for TB Research, Baltimore, United States

   Contribution

   Conceptualization, Resources, Investigation, Project administration, Writing – review and editing

   Competing interests

   received grant funds from Pfizer. They participated on the data safety monitoring board of a TB HDT trial, and on the Scientific Advisory Board of a trial of an electronic reminder to take daily TB treatment. They hold an unpaid leadership role at the Setshaba Research Center. The authors has no other competing interests to declare

8. Anne von Gottberg

   1. Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
   2. School of Pathology, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa

   Contribution

   Conceptualization, Resources, Investigation, Methodology, Writing – review and editing

   Competing interests

   reports receiving grant funds from Sanofi Pasteur and Bill & Melinda Gates Foundation. They act as chairperson of the National Advisory Group on Immunisation. The author has no other competing interests to declare

   "This ORCID iD identifies the author of this article:" 0000-0002-0243-7455

9. Nicole Wolter

   1. Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
   2. School of Pathology, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa

   Contribution

   Conceptualization, Investigation, Methodology, Writing – review and editing

   Competing interests

   received grants from Sanofi Pasteur and Bill & Melinda Gates Foundation, and from US Centers for Disease Control and Prevention. The author has no other competing interests to declare

10. Mvuyo Makhasi

    1. Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
    2. School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa

    Contribution

    Conceptualization, Writing – review and editing

    Competing interests

    No competing interests declared

11. Cheryl Cohen

    1. Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
    2. School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa

    Contribution

    Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Investigation, Methodology, Writing – review and editing

    Competing interests

    has received grant support from Sanofi Pasteur, US Centers for Disease Control and Prevention, Welcome Trust, Programme for Applied Technologies in Health (PATH), Bill & Melinda Gates Foundation and South African Medical Research Council (SA-MRC). The author has no other competing interests to declare

12. Ciro Cattuto

    1. ISI Foundation, Turin, Italy
    2. Department of Informatics, University of Turin, Turin, Italy

    Contribution

    Conceptualization, Resources, Formal analysis, Supervision, Methodology, Writing – review and editing

    Competing interests

    received support from the European Union's Horizon 2020 research and innovation programme under grant agreements No. 101003688 (EpiPose) and No. 101016233 (PERISCOPE), and from Fondation Botnar, EPFL COVID-19 Real Time Epidemiology I-DAIR Pathfinder. They also received support from the Lagrange Project of ISI Foundation funded by CRT Foundation, and received equipment from the European Space Agency. They were issued the patent US8660490B2. They received support for attending meetings/travel from Gulbenkian Foundation, and participate as a Steering Board Member at CRT Foundation, on the Advisory Board at The Data Guild and on the Advisory Board in GovLab's 100 Questions Initiative and Open Data Policy Lab. The authors has no other competing interests to declare

13. Stefano Tempia

    1. Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
    2. School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa

    Contribution

    Conceptualization, Formal analysis, Supervision, Funding acquisition, Methodology, Writing – review and editing

    Competing interests

    No competing interests declared

14. SA-S-HTS Group

    Competing interests

    reports institutional grant funding from UK Foreign, Commonwealth and Development Office, Wellcome Trust, WHO AFRO, Africa Pathogen Genomics Initiative (Africa PGI) through African Society of Laboratory Medicine (ASLM) and Africa CDC, US Centers for Disease Control and Prevention, South African Medical Research Council (SAMRC) and UK Department of Health and Social Care and managed by the Fleming Fund: SEQAFRICA project. The author has no other competing interests to declare

    1. Amelia Buys, Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
    2. Daniel G Amoako, Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
    3. Dylan Toi, Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
    4. Jinal N Bhiman, Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
    5. Juanita Chewparsad, Perinatal HIV Research Unit (PHRU), University of the Witwatersrand, Johannesburg, South Africa
    6. Kedibone Ndlangisa, Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
    7. Leisha Genade, Perinatal HIV Research Unit (PHRU), University of the Witwatersrand, Johannesburg, South Africa
    8. Limakatso Lebina, Perinatal HIV Research Unit (PHRU), University of the Witwatersrand, Johannesburg, South Africa
    9. Linda de Gouveia, Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
    10. Mzimasi Neti, Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa
    11. Retshidisitswe Kotane, Centre for Respiratory Diseases and Meningitis, National Institute for Communicable Diseases of the National Health Laboratory Service, Johannesburg, South Africa

• 463

  Page views

• 60

  Downloads

• 0

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.