Social mixing patterns relevant to infectious diseases spread by close contact in urban Blantyre, Malawi

Social contacts and mixing patterns in rural Gambia

BACKGROUND

Close contact between an infectious and susceptible person is an important factor in respiratory disease transmission. Information on social contacts and mixing patterns in a population is crucial to understanding transmission patterns and informing transmission models of respiratory infections. Although West Africa has one of the highest burdens of respiratory infections, there is a lack of data on interpersonal contact and mixing patterns in this region.

METHODS

Between January and November 2022, we conducted a cross-sectional, social contact survey within the population of the Central and Upper River Regions of The Gambia. Selected participants completed a questionnaire about their travel history and social contacts, detailing the number, intensity, location, frequency, duration, and age of contacts. We calculated age-standardized contact matrices to determine contact patterns in the population.

RESULTS

Overall, individuals made an average of 12.7 (95% CI: 12.4-13.0) contacts per day. Contact patterns were mostly age-assortative and 84.5% of all contacts were physical. School-aged children (5-14 years) had the highest mean number of physical contacts (11.3, 95% CI: 10.9-11.8) while the < 1-year age group had the fewest contacts (9.4, 95% CI: 9.1-9.8). A large proportion of contacts (78%) occurred at home. Daily number of contacts increased with household size. While we did not observe any effect of gender on contact patterns, there were seasonal variations in contact type. Non-physical contacts were higher during the dry season compared to the rainy season. In contrast, there were more physical contacts in the rainy season compared to the dry season.

CONCLUSIONS

In rural Gambia, social contact patterns were primarily driven by household mixing. Most contacts were physical and mostly age-assortative, particularly among school-aged children. Our data can improve infectious disease transmission models of respiratory pathogens in high-transmission settings, which are valuable for optimizing the delivery of different interventions.

The potential impact of new tuberculosis vaccines on the burden of tuberculosis in people with HIV in South Africa

BACKGROUND

People with HIV (PWH) are at an increased risk of tuberculosis (TB). New TB vaccines may help reduce this burden. New TB vaccine candidates are well tolerated and immunogenic in PWH. There are currently limited data on vaccine efficacy in this population.

METHODS

Using mathematical modeling, we explored the potential impact of a novel TB vaccine on TB burden in PWH in South Africa between 2030 and 2050. We compared the impact of a vaccine delivered irrespective of HIV status to vaccination of either PWH or people without HIV. We explored the impact of reduced vaccine efficacy and duration of protection in PWH relative to people without HIV on our model predictions.

RESULTS

Vaccination irrespective of HIV status, with a vaccine with equal efficacy and duration in PWH, could avert up to 1.01 (95% range: 0.96-1.22) million TB cases in PWH. Restricting vaccination to PWH or people without HIV would achieve 65% (60-70) and 48% (46-53) of the total impact, respectively. These results are strongly dependent on the assumed efficacy and duration of protection in PWH. Further information on these characteristics is important to identify the most efficient use of new vaccines to reduce TB burden in PWH.

CONCLUSION

Our results suggest that new vaccines could play an important role in reducing the TB burden in PWH. Vaccines targeted at people without HIV could provide significant indirect benefit to PWH, but vaccines which are well tolerated and effective in PWH will be critical to maximizing the impact in this population.

Within-country heterogeneity in patterns of social contact relevant for tuberculosis infection transmission, prevention, and care

Mycobacterium tuberculosis (Mtb) transmission is driven by variable social, environmental, and biological factors, including the number and duration of indoor contacts. Social contact data can provide information on potential transmission patterns, but is underutilised outside the field of mathematical modelling. We explore three contexts where contact data can provide valuable insights: 1) household contact tracing; 2) infection prevention and control measures (IPC) in congregate settings; and 3) contamination in cluster randomised trials (CRTs). A social contact survey was conducted in three communities with comparable population sizes in South Africa: an urban township and peri-urban and rural clinic catchment areas. Participants reported congregate settings visited over 24-hours, visit durations, and estimated numbers of people present. To correspond with the three contexts, we estimated the proportion of contact hours occurring 1) within the home; 2) in congregate settings outside the home; and 3) outside the participants' communities. Participants reported a mean of 27.0 (rural), 55.2 (peri-urban), and 73.0 (urban) contact hours. The proportion within the home was similar among rural and peri-urban participants (76.8% and 71.7%), compared to urban (48.6%). Congregate settings visited varied; urban participants spent the most contact hours in retail/office settings (19.9%), peri-urban participants in community-service buildings (20.4%), and rural participants in other peoples' homes (25.5%). Urban participants reported the highest proportion of contact outside the community (67.0%) compared to rural (38.8%) and peri-urban (21.5%) participants. The heterogeneity in contact patterns has implications for TB interventions. Household contact tracing may be most effective in the rural community where household contact was highest. The diverse range of congregate settings visited suggests that prioritising IPC measures in these locations may enhance their overall efficacy. Considering contact patterns when designing clusters may reduce contamination risk in CRTs. Tailored interventions, informed by local contexts, are essential to reduce TB burden.

Social Contact Patterns in Rural and Urban Settings, Mozambique, 2021-2022

Few sources have reported empirical social contact data from resource-poor settings. To address this shortfall, we recruited 1,363 participants from rural and urban areas of Mozambique during the COVID-19 pandemic, determining age, sex, and relation to the contact for each person. Participants reported a mean of 8.3 (95% CI 8.0-8.6) contacts per person. The mean contact rates were higher in the rural site compared with the urban site (9.8 vs 6.8; p<0.01). Using mathematical models, we noted higher vaccine effects in the rural site when comparing empirical (32%) with synthetic (29%) contact matrices and lower corresponding vaccine effects in the urban site (32% vs 35%). Those effects were prominent in younger (0-9 years) and older (≥60 years) persons. Our work highlights the importance of empirical data, showing differences in contact rates and patterns between rural and urban sites in Mozambique and their nonnegligible effects in modeling potential effects of vaccine interventions.

Directly observed social contact patterns among school children in rural Gambia

INTRODUCTION

School-aged children play a major role in the transmission of many respiratory pathogens due to high rate of close contacts in schools. The validity and accuracy of proxy-reported contact data may be limited, particularly for children when attending school. We observed social contacts within schools and assessed the accuracy of proxy-reported versus observed physical contact data among students in rural Gambia.

METHODS

We enrolled school children who had also been recruited to a survey of Streptococcus pneumoniae carriage and social contacts. We visited participants at school and observed their contact patterns within and outside the classroom for two hours. We recorded the contact type, gender and approximate age of the contactee, and class size. We calculated age-stratified contact matrices to determine in-school contact patterns. We compared proxy-reported estimated physical contacts for the subset of participants (18 %) randomised to be observed on the same day for which the parent or caregiver reported the school contacts.

RESULTS

We recorded 3822 contacts for 219 participants from 114 schools. The median number of contacts was 15 (IQR: 11-20). Contact patterns were strongly age-assortative, and mainly involved physical touch (67.5 %). Those aged 5-9 years had the highest mean number of contacts [19.0 (95 %CI: 16.7-21.3)] while the ≥ 15-year age group had fewer contacts [12.8 (95 %CI: 10.9-14.7)]. Forty (18 %) participants had their school-observed contact data collected on the same day as their caregiver reported their estimated physical contacts at school; only 22.5 % had agreement within ±2 contacts between the observed and reported contacts. Fifty-eight percent of proxy-reported contacts were under-estimates.

CONCLUSIONS

Social contact rates observed among pupils at schools in rural Gambia were high, strongly age-assortative, and physical. Reporting of school contacts by proxies may underestimate the effect of school-age children in modelling studies of transmission of infections. New approaches are needed to quantify contacts within schools.

Online bilateral matching methodology for anti-epidemic resources based on spatial transmission risk

This paper focuses on the online anti-epidemic resource allocation between demanders and suppliers considering the epidemic spatial spread. The spatial crowdsourcing with sharing platform is an effective way for anti-epidemic resource allocation, and a reasonable online matching strategy can improve the efficiency of resource utilization. The paper proposes online matching heuristic strategy (HSTRF-LRLUF strategy) and designs an online batch bilateral matching algorithm for anti-epidemic resources, which considers the impact of grid spatial aggregation and diffusion risk of emerging infectious diseases. The population distribution within grids and the commuting patterns between grids can provide decision support for selecting online matching strategies of anti-epidemic resources. A larger matching time window focusing on the spatial transmission risk (TR) of the epidemic can obtain better matching results. However, with a smaller matching time window, the decision makers can focus on spatial agglomeration risk (OR) or spatial diffusion risk (IR). The paper effectively combines the spatial crowdsourcing model with the anti-epidemic resource allocation to achieve the allocation of emergency resources to individuals. A combined anti-epidemic resource online matching heuristic strategies is designed from the spatial agglomeration risk and the spatial diffusion risk. Decision makers can dynamically adjust the online matching strategies of anti-epidemic resources by evaluating the spatial agglomeration risk, the spatial diffusion risk, and the overall spatial transmission risk based on the real-time spread of the epidemic and the supply of anti-epidemic resources.

Estimating pneumococcal carriage dynamics in adults living with HIV in a mature infant pneumococcal conjugate vaccine programme in Malawi, a modelling study

BACKGROUND

Adults living with human immunodeficiency virus (ALWHIV) receiving antiretroviral therapy (ART) exhibit higher pneumococcal carriage prevalence than adults without HIV (HIV-). To assess factors influencing high pneumococcal carriage in ALWHIV, we estimated pneumococcal carriage acquisition and clearance rates in a high transmission and disease-burdened setting at least 10 years after introducing infant PCV13 in routine immunisation.

METHODS

We collected longitudinal nasopharyngeal swabs from individuals aged 18-45 in Blantyre, Malawi. The study group included both HIV- individuals and those living with HIV, categorised based on ART duration as either exceeding 1 year (ART > 1y) or less than 3 months (ART < 3 m). Samples were collected at baseline and then weekly for 16 visits. To detect pneumococcal carriage, we used classical culture microbiology, and to determine pneumococcal serotypes, we used latex agglutination. We modelled trajectories of serotype colonisation using multi-state Markov models to capture pneumococcal carriage dynamics, adjusting for age, sex, number of under 5 year old (< 5y) children, social economic status (SES), and seasonality.

RESULTS

We enrolled 195 adults, 65 adults in each of the study groups. 51.8% were females, 25.6% lived with more than one child under 5 years old, and 41.6% lived in low socioeconomic areas. The median age was 33 years (IQR 25-37 years). The baseline pneumococcal carriage prevalence of all serotypes was 31.3%, with non-PCV13 serotypes (NVT) at 26.2% and PCV13 serotypes (VT) at 5.1%. In a multivariate longitudinal analysis, pneumococcal carriage acquisition was higher in females than males (hazard ratio [HR], NVT [1.53]; VT [1.96]). It was also higher in low than high SES (NVT [1.38]; VT [2.06]), in adults living with 2 + than 1 child < 5y (VT [1.78]), and in ALWHIV on ART > 1y than HIV- adults (NVT [1.43]). Moreover, ALWHIV on ART > 1y cleared pneumococci slower than HIV- adults ([0.65]). Residual VT 19F and 3 were highly acquired, although NVT remained dominant.

CONCLUSIONS

The disproportionately high point prevalence of pneumococcal carriage in ALWHIV on ART > 1y is likely due to impaired nasopharyngeal clearance, which results in prolonged carriage. Our findings provide baseline estimates for comparing pneumococcal carriage dynamics after implementing new PCV strategies in ALWHIV.

Environmental surveillance for Salmonella Typhi in rivers and wastewater from an informal sewage network in Blantyre, Malawi

Environmental surveillance for Salmonella Typhi may provide information on the community-level dynamics of typhoid fever in resource poor regions experiencing high disease burden. Many knowledge gaps concerning the feasibility of ES remain, especially in areas lacking formal sewage systems. We implemented protocols for S. Typhi ES, including site selection and catchment population estimation, sample concentration and testing using qPCR for S. Typhi specific gene targets. Between May 2021 and May 2022, we collected grab samples and Moore swabs from 43 sites in Blantyre, Malawi. Catchment characteristics, water quality, and human faecal contamination (qPCR for Bacteroides HF183) were also recorded. Their association with S. Typhi detection was investigated using a logistic mixed-effects regression analysis. Prevalence of S. Typhi in ES samples was 2.1% (1.1-4.0%) and 3.9% (1.9-7.9%) for grab and Moore swab samples, respectively. HF183 was associated S. Typhi positivity, with a unit increase in log genome copies/microlitre increasing the odds of detection of S. Typhi by 1.56 (95% CI: 1.29-1.89) and 1.33 (1.10-1.61) in Moore swabs and grab samples, respectively. The location and timing of S. Typhi detection through ES was not associated with the incidence of typhoid fever reported in associated catchment populations. During this period of relatively low typhoid fever incidence, wastewater surveillance continued to detect S. Typhi in human sewage and wastewater suggesting that ES using natural river systems can be a sensitive indicator of transmission.

Comprehensive profiling of social mixing patterns in resource poor countries: A mixed methods research protocol

BACKGROUND

Low-and-middle-income countries (LMICs) bear a disproportionate burden of communicable diseases. Social interaction data inform infectious disease models and disease prevention strategies. The variations in demographics and contact patterns across ages, cultures, and locations significantly impact infectious disease dynamics and pathogen transmission. LMICs lack sufficient social interaction data for infectious disease modeling.

METHODS

To address this gap, we will collect qualitative and quantitative data from eight study sites (encompassing both rural and urban settings) across Guatemala, India, Pakistan, and Mozambique. We will conduct focus group discussions and cognitive interviews to assess the feasibility and acceptability of our data collection tools at each site. Thematic and rapid analyses will help to identify key themes and categories through coding, guiding the design of quantitative data collection tools (enrollment survey, contact diaries, exit survey, and wearable proximity sensors) and the implementation of study procedures. We will create three age-specific contact matrices (physical, nonphysical, and both) at each study site using data from standardized contact diaries to characterize the patterns of social mixing. Regression analysis will be conducted to identify key drivers of contacts. We will comprehensively profile the frequency, duration, and intensity of infants' interactions with household members using high resolution data from the proximity sensors and calculating infants' proximity score (fraction of time spent by each household member in proximity with the infant, over the total infant contact time) for each household member.

DISCUSSION

Our qualitative data yielded insights into the perceptions and acceptability of contact diaries and wearable proximity sensors for collecting social mixing data in LMICs. The quantitative data will allow a more accurate representation of human interactions that lead to the transmission of pathogens through close contact in LMICs. Our findings will provide more appropriate social mixing data for parameterizing mathematical models of LMIC populations. Our study tools could be adapted for other studies.

Social contacts patterns relevant to the transmission of infectious diseases in Suzhou, China following the COVID-19 epidemic

BACKGROUND

The COVID-19 pandemic has profoundly affected human social contact patterns, but there is limited understanding regarding the post-pandemic social contact patterns. Our objective is to quantitatively assess social contact patterns in Suzhou post-COVID-19.

METHODS

We employed a diary design and conducted social contact surveys from June to October 2023, utilizing paper questionnaires. A generalized linear model was utilized to analyze the relationship between individual contacts and covariates. We examined the proportions of contact type, location, duration, and frequency. Additionally, age-related mixed matrices were established.

RESULTS

The participants reported an average of 11.51 (SD 5.96) contact numbers and a total of 19.78 (SD 20.94) contact numbers per day, respectively. The number of contacts was significantly associated with age, household size, and the type of week. Compared to the 0-9 age group, those in the 10-19 age group reported a higher number of contacts (IRR = 1.12, CI: 1.01-1.24), while participants aged 20 and older reported fewer (IRR range: 0.54-0.67). Larger households (5 or more) reported more contacts (IRR = 1.09, CI: 1.01-1.18) and fewer contacts were reported on weekends (IRR = 0.95, CI: 0.90-0.99). School had the highest proportion of contact durations exceeding 4 h (49.5%) and daily frequencies (90.4%), followed by home and workplace. The contact patterns exhibited clear age-assortative mixing, with Q indices of 0.27 and 0.28.

CONCLUSIONS

We assessed the characteristics of social contact patterns in Suzhou, which are essential for parameterizing models of infectious disease transmission. The high frequency and intensity of contacts among school-aged children should be given special attention, making school intervention policies a crucial component in controlling infectious disease transmission.

Estimating household contact matrices structure from easily collectable metadata

Contact matrices are a commonly adopted data representation, used to develop compartmental models for epidemic spreading, accounting for the contact heterogeneities across age groups. Their estimation, however, is generally time and effort consuming and model-driven strategies to quantify the contacts are often needed. In this article we focus on household contact matrices, describing the contacts among the members of a family and develop a parametric model to describe them. This model combines demographic and easily quantifiable survey-based data and is tested on high resolution proximity data collected in two sites in South Africa. Given its simplicity and interpretability, we expect our method to be easily applied to other contexts as well and we identify relevant questions that need to be addressed during the data collection procedure.

Transmission matrices used in epidemiologic modelling

Mixing matrices are included in infectious disease models to reflect transmission opportunities between population strata. These matrices were originally constructed on the basis of theoretical considerations and most of the early work in this area originates from research on sexually transferred diseases in the 80s, in response to AIDS. Later work in the 90s populated these matrices on the basis of survey data gathered to capture transmission risks for respiratory diseases. We provide an overview of developments in the construction of matrices for capturing transmission opportunities in populations. Such transmission matrices are useful for epidemiologic modelling to capture within and between stratum transmission and can be informed from theoretical mixing assumptions, informed by empirical evidence gathered through investigation as well as generated on the basis of data. Links to summary measures and threshold conditions are also provided.

Risk factors for pneumococcal carriage in adults living with HIV on antiretroviral therapy in the infant pneumococcal vaccine era in Malawi

OBJECTIVE

Adults living with HIV (ALWHIV) on antiretroviral therapy (ART) are at high risk of pneumococcal carriage and disease. To help evaluate carriage risk in African ALWHIV at least 4 years after infant pneumococcal conjugate vaccination introduction in 2011, we assessed association between pneumococcal carriage and potential risk factors.

METHODS

Nasopharyngeal swabs were collected from adults aged 18-40 years attending an ART clinic during rolling, cross-sectional surveys in Blantyre, Malawi between 2015 and 2019. We fitted generalized additive models to estimate the risk of sex, social economic status (SES), living with a child less than 5 years, and ART duration on carriage.

RESULTS

Of 2067 adults, median age was 33 years (range 28-37), 1427 (69.0%) were women, 1087 (61.4%) were in low-middle socioeconomic-status (SES), 910 (44.0%) were living with a child less than 5 years, and median ART duration was 3 years (range 0.004-17). We estimated 38.2 and 60.6% reductions in overall and vaccine-serotype carriage prevalence. Overall carriage was associated with low SES, living with a child less than 5 years and shorter duration on ART. By contrast, vaccine-type carriage was associated with living without a child less than 5 years and male sex.

CONCLUSION

Despite temporal reductions in overall and vaccine-serotype carriage, there is evidence of incomplete vaccine-serotype indirect protection. A targeted-vaccination campaign should be considered for ALWHIV, along with other public health measures to further reduce vaccine-serotype carriage and therefore disease.