Patterns of mixing and the spread of pandemics

Incorporating social mix data into epidemic models can help policy makers better understand the spread of the epidemic. However, empirical data on admixture may not be readily available in most populations. In a recent work, a network model methodology is proposed to construct a micro-level social mix structure when empirical data is not available.

The ongoing COVID-19 pandemic has highlighted the importance of understanding infectious disease transmission dynamics, which are fundamentally determined by the social and physical mixing patterns of individuals in the population. Data on human mobility and social networks offer the possibility of directly observing patterns of admixture as well as understanding the underlying structure of social contacts. Indeed, mixing model data, such as contact rates between individuals and population flows between regions of interest, have been integrated into mathematical models to provide nowcasts and predictions of the pandemic. .1. These data provide unique opportunities to understand how patterns of diversity vary across socio-economic and demographic micro-segments, which can directly affect trends in social diversity and the spread of the epidemic.1,2,3. As indicated in this issue of Computational science of natureCao and Heydari developed a network model to show that the micro-level structure of person-to-person interactions, as measured by average household size and rate of in-person social contact, is a potential explanatory factor for variations in the patterns. of human mixing.4. As the pandemic transitions to endemicity, their findings provide important insights into explaining heterogeneities in regional and global dynamics of the COVID-19 pandemic.

Fundamentally, the social network perspective posits that the structure of social networks has significant effects on patterns of epidemic spread. For example, schoolchildren are the “bridge” linking family and community transmission5. In addition, social networks have behavioral, social and informational spillover effects that also have epidemiological implications: for example, US counties with more social ties to Italy and China tend to adhere more to mobility restrictions .1.6. Another level of complexity, pandemic and health policies (such as distancing) have affected social mix: the pandemic has weakened the weak ties that usually bind communities, while strengthening close and clustered networksseven. Recent empirical research modeling the spread of SARS-CoV-2 has examined how such nuances in the structural dynamics of social networks directly affect the dynamics of infection.8suggesting that intervention strategies based on structural network considerations can both directly and indirectly curb the social behaviors that lead to transmissions.

One of the best-known approaches to integrating social mix data into epidemic models is to estimate social contact matrices from POLYMOD-type diary-based contact surveys.1,2,9. However, representative empirical contact data is not always available.2. To fill these data gaps, synthetic contact matrices were constructed by inferring contact patterns from more widely available survey or census data on key socio-demographic characteristics.ten. To further capture contact changes over time, mobile-derived geolocation data (from sources including Facebook, Google, Apple, WeChat, Alipay, and Safegraph) has been widely used by researchers to study the COVID pandemic. -19, and have shown great potential in generating accurate nowcasts and near-term predictions of outbreaks, even when the population mix varies widely due to disruptive pandemic controls1.3.

Nevertheless, empirical data do not always accurately reflect social mix and underlying behaviors. For example, mobility measured by the use of public transport was reduced to low levels during the Chinese New Year holidays, while the transmission of COVID-19 was actually enhanced by an increased mixing of family gatherings. in Hong Kong1. Complementary to the data-driven approaches above, Cao and Heydari proposed an explicable model by claiming that social structure is a major determinant of contact mixing.4. In short: a representative network model is constructed as the basic social structure of disease transmission based on survey and census data; the transmission of the virus is approached by strong and weak links between individuals in the constructed network; and non-pharmaceutical interventions (NPIs) are modeled by sequential disconnection from weakest to strongest links in order of connectivity. Additionally, since empirical mixing data is not always available, this network model can also serve as a viable alternative for simulating outbreaks to inform decision-making and control measures. Ultimately, this approach has the potential to generate new insights into our understanding of human mixing patterns. During the early stages of a pandemic, especially when NPIs are the only measures available, the network model can also help identify the most cost-effective NPIs suited to the social structure of different populations.

However, the study itself has notable limitations. First, the framework has not been formally validated by calibration against empirical epidemiological data. Second, although a Government Stringency Index (GSI) could potentially be used as a proxy for the effects of interventions to estimate changes in effective reproductive number, such a relationship could be highly dependent on the specific interventions adopted and the compliance of the population over time.6. Model results should be interpreted with caution given potential temporal variations in the relationship between GSI and disease transmission. Third, although the model can reproduce the epidemic curves of the early phase of the pandemic, more data are needed to parameterize the model if it were to be applied to the prediction of later stages of the pandemic, given the heterogeneities of the vaccination coverage, infection history and circulation. variations in different populations.

Despite these limitations, the authors are among the first to develop explainable models with high potential for studies of social structure and population admixture in infectious disease modeling.4. Future research on this flow could develop models based on other network distribution assumptions and investigate the impact of network structural characteristics beyond network size and contact frequency. In particular, higher-order network interactions and topographies (e.g., network patterns and embedding structure) can be used to capture otherwise unobservable indirect relationships within specific social structures (e.g., families, nursing homes, schools, etc.) and provide epidemiological information. information beyond what traditional contact tracing methods allow. Interactions between social demographics, population mix, vaccination coverage, past waves of epidemics, and characteristics of circulating variants could also potentially be incorporated into the model by altering the degree of network connectivity (Fig. 1), thus providing more information on the epidemic. spread.

Fig. 1: Understanding patterns of human mixing and the spread of disease.

Cao and Heydari provide an explainable model to study heterogeneities in the regional and global dynamics of the COVID-19 pandemic. Interactions between social demographics, population mixing, vaccination coverage, past waves of epidemics, and characteristics of circulating variants could potentially be incorporated into their proposed model, which will provide even greater insight into the spread of disease.

About Tracy G. Larimore

Check Also

18 Adorable and Cozy Fall Knitting and Crochet Patterns for All Skill Levels

Photo: Image bank by GOFFKEIN.PRO/ShutterstockThis article may contain affiliate links. If you make a purchase, …