Investigating the relationship between interventions, contact patterns, and SARS-CoV-2 transmissibility

COVID-19 from symptoms to prediction: A statistical and machine learning approach

During the COVID-19 pandemic, the analysis of patient data has become a cornerstone for developing effective public health strategies. This study leverages a dataset comprising over 10,000 anonymized patient records from various leading medical institutions to predict COVID-19 patient age groups using a suite of statistical and machine learning techniques. Initially, extensive statistical tests including ANOVA and t-tests were utilized to assess relationships among demographic and symptomatic variables. The study then employed machine learning models such as Decision Tree, Naïve Bayes, KNN, Gradient Boosted Trees, Support Vector Machine, and Random Forest, with rigorous data preprocessing to enhance model accuracy. Further improvements were sought through ensemble methods; bagging, boosting, and stacking. Our findings indicate strong associations between key symptoms and patient age groups, with ensemble methods significantly enhancing model accuracy. Specifically, stacking applied with random forest as a meta leaner exhibited the highest accuracy (0.7054). In addition, the implementation of stacking techniques notably improved the performance of K-Nearest Neighbors (from 0.529 to 0.63) and Naïve Bayes (from 0.554 to 0.622) and demonstrated the most successful prediction method. The study aimed to understand the number of symptoms identified in COVID-19 patients and their association with different age groups. The results can assist doctors and higher authorities in improving treatment strategies. Additionally, several decision-making techniques can be applied during pandemic, tailored to specific age groups, such as resource allocation, medicine availability, vaccine development, and treatment strategies. The integration of these predictive models into clinical settings could support real-time public health responses and targeted intervention strategies.

Effects of non-pharmaceutical interventions on COVID-19 transmission: rapid review of evidence from Italy, the United States, the United Kingdom, and China

Background

Prior to the development of COVID-19 vaccines, policymakers instituted various non-pharmaceutical interventions (NPIs) to limit transmission. Prior studies have attempted to examine the extent to which these NPIs achieved their goals of containment, suppression, or mitigation of disease transmission. Existing evidence syntheses have found that numerous factors limit comparability across studies, and the evidence on NPI effectiveness during COVID-19 pandemic remains sparse and inconsistent. This study documents the magnitude and variation in NPI effectiveness in reducing COVID-19 transmission (i.e., reduction in effective reproduction rate [Reff] and daily contact rate) in Italy, the United States, the United Kingdom, and China.

Methods

Our rapid review and narrative synthesis of existing research identified 126 studies meeting our screening criteria. We selected four contexts with >5 articles to facilitate a meaningful synthesis. This step yielded an analytic sample of 61 articles that used data from China, Italy, the United Kingdom, and the United States.

Results

We found wide variation and substantial uncertainty around the effectiveness of NPIs at reducing disease transmission. Studies of a single intervention or NPIs that are the least stringent had estimated Reff reductions in the 10-50% range; those that examined so-called "lockdowns" were associated with greater Reff reductions that ranged from 40 to 90%, with many in the 70-80% range. While many studies reported on multiple NPIs, only six of the 61 studies explicitly used the framing of "stringency" or "mild versus strict" or "tiers" of NPIs, concepts that are highly relevant for decisionmakers.

Conclusion

Existing evidence suggests that NPIs reduce COVID-19 transmission by 40 to 90 percent. This paper documents the extent of the variation in NPI effectiveness estimates and highlights challenges presented by a lack of standardization in modeling approaches. Further research on NPI effectiveness at different stringency levels is needed to inform policy responses to future pandemics.

Multiple pandemic waves vs multi-period/multi-phasic epidemics: Global shape of the COVID-19 pandemic

The overall course of the COVID-19 pandemic in Western countries has been characterized by complex sequences of phases. In the period before the arrival of vaccines, these phases were mainly due to the alternation between the strengthening/lifting of social distancing measures, with the aim to balance the protection of health and that of the society as a whole. After the arrival of vaccines, this multi-phasic character was further emphasized by the complicated deployment of vaccination campaigns and the onset of virus' variants. To cope with this multi-phasic character, we propose a theoretical approach to the modeling of overall pandemic courses, that we term multi-period/multi-phasic, based on a specific definition of phase. This allows a unified and parsimonious representation of complex epidemic courses even when vaccination and virus' variants are considered, through sequences of weak ergodic renewal equations that become fully ergodic when appropriate conditions are met. Specific hypotheses on epidemiological and intervention parameters allow reduction to simple models. The framework suggest a simple, theory driven, approach to data explanation that allows an accurate reproduction of the overall course of the COVID-19 epidemic in Italy since its beginning (February 2020) up to omicron onset, confirming the validity of the concept.

Contact behaviour before, during and after the COVID-19 pandemic in the Netherlands: evidence from contact surveys, 2016 to 2017 and 2020 to 2023

BackgroundThe first wave of the COVID-19 pandemic in 2020 was largely mitigated by limiting contacts in the general population. In early 2022, most contact-reducing measures were lifted.AimTo assess whether the population has reverted to pre-pandemic contact behaviour and how this would affect transmission potential of a newly emerging pathogen.MethodsWe compared two studies on contact behaviour in the Netherlands: the PIENTER Corona study, conducted during and after the pandemic (held every 2-6 months from April 2020) and the PIENTER3 study (2016-17, as pre-pandemic baseline). In both, participants (ages 1-85 years) reported number and age group of all face-to-face persons contacted on the previous day in a survey. Transmission potential was examined using the next-generation matrix approach.ResultsWe found an average of 15.4 (95% CI: 14.3-16.4) community contacts per person per day after the pandemic in May 2023, 13% lower than baseline (17.8; 95% CI: 17.0-18.5). Among all ages, children (5-9 years) had the highest number of contacts, both pre- and post-pandemic. Mainly adults aged 20-59 years had not reverted to pre-pandemic behaviours, possibly because they more often work from home. Although the number of contacts is lower compared to the pre-pandemic period, the effect on transmission potential of a newly emerging respiratory pathogen is limited if all age groups were equally susceptible.ConclusionContinuous monitoring of contacts can signal changes in contact patterns and can define a 'new normal' baseline. Both aspects are needed to prepare for a future pandemic.

Social contact patterns following the COVID-19 pandemic: a snapshot of post-pandemic behaviour from the CoMix study

The COVID-19 pandemic led to unprecedented changes in behaviour. To estimate if these persisted, a final round of the CoMix social contact survey was conducted in four countries at a time when all societal restrictions had been lifted for several months. We conducted a survey on a nationally representative sample in the UK, Netherlands (NL), Belgium (BE), and Switzerland (CH). Participants were asked about their contacts and behaviours on the previous day. We calculated contact matrices and compared the contact levels to a pre-pandemic baseline to estimate R0. Data collection occurred from 17 November to 7 December 2022. 7477 participants were recruited. Some were asked to undertake the survey on behalf of their children. Only 14.4 % of all participants reported wearing a facemask on the previous day. Self-reported vaccination rates in adults were similar for each country at around 86 %. Trimmed mean recorded contacts were highest in NL with 9.9 (95 % confidence interval [CI] 9.0-10.8) contacts per person per day and lowest in CH at 6.0 (95 % CI 5.4-6.6). Contacts at work were lowest in the UK (1.4 contacts per person per day) and highest in NL at 2.8 contacts per person per day. Other contacts were also lower in the UK at 1.6 per person per day (95 % CI 1.4-1.9) and highest in NL at 3.4 recorded per person per day (95 % CI 43.0-4.0). The next-generation approach suggests that R0 for a close-contact disease would be roughly half pre-pandemic levels in the UK, 80 % in NL and intermediate in the other two countries. The pandemic appears to have resulted in lasting changes in contact patterns expected to have an impact on the epidemiology of many different pathogens. Further post-pandemic surveys are necessary to confirm this finding.

Population Behavior Changes Underlying Phasic Shifts of SARS-CoV-2 Exposure Settings Across 3 Omicron Epidemic Waves in Hong Kong: Prospective Cohort Study

BACKGROUND

Exposure risk was shown to have affected individual susceptibility and the epidemic spread of COVID-19. The dynamics of risk by and across exposure settings alongside the variations following the implementation of social distancing interventions are understudied.

OBJECTIVE

This study aims to examine the population's trajectory of exposure risk in different settings and its association with SARS-CoV-2 infection across 3 consecutive Omicron epidemic waves in Hong Kong.

METHODS

From March to June 2022, invitation letters were posted to 41,132 randomly selected residential addresses for the recruitment of households into a prospective population cohort. Through web-based monthly surveys coupled with email reminders, a representative from each enrolled household self-reported incidents of SARS-CoV-2 infections, COVID-19 vaccination uptake, their activity pattern in the workplace, and daily and social settings in the preceding month. As a proxy of their exposure risk, the reported activity trend in each setting was differentiated into trajectories based on latent class growth analyses. The associations of different trajectories of SARS-CoV-2 infection overall and by Omicron wave (wave 1: February-April; wave 2: May-September; wave 3: October-December) in 2022 were evaluated by using Cox proportional hazards models and Kaplan-Meier analysis.

RESULTS

In total, 33,501 monthly responses in the observation period of February-December 2022 were collected from 5321 individuals, with 41.7% (2221/5321) being male and a median age of 46 (IQR 34-57) years. Against an expanding COVID-19 vaccination coverage from 81.9% to 95.9% for 2 doses and 20% to 77.7% for 3 doses, the cumulative incidence of SARS-CoV-2 infection escalated from <0.2% to 25.3%, 32.4%, and 43.8% by the end of waves 1, 2, and 3, respectively. Throughout February-December 2022, 52.2% (647/1240) of participants had worked regularly on-site, 28.7% (356/1240) worked remotely, and 19.1% (237/1240) showed an assorted pattern. For daily and social settings, 4 and 5 trajectories were identified, respectively, with 11.5% (142/1240) and 14.6% (181/1240) of the participants gauged to have a high exposure risk. Compared to remote working, working regularly on-site (adjusted hazard ratio [aHR] 1.47, 95% CI 1.19-1.80) and living in a larger household (aHR 1.12, 95% CI 1.06-1.18) were associated with a higher risk of SARS-CoV-2 infection in wave 1. Those from the highest daily exposure risk trajectory (aHR 1.46, 95% CI 1.07-2.00) and the second highest social exposure risk trajectory (aHR 1.52, 95% CI 1.18-1.97) were also at an increased risk of infection in waves 2 and 3, respectively, relative to the lowest risk trajectory.

CONCLUSIONS

In an infection-naive population, SARS-CoV-2 transmission was predominantly initiated at the workplace, accelerated in the household, and perpetuated in the daily and social environments, as stringent restrictions were scaled down. These patterns highlight the phasic shift of exposure settings, which is important for informing the effective calibration of targeted social distancing measures as an alternative to lockdown.

Importance of social inequalities to contact patterns, vaccine uptake, and epidemic dynamics

Individuals' socio-demographic and economic characteristics crucially shape the spread of an epidemic by largely determining the exposure level to the virus and the severity of the disease for those who got infected. While the complex interplay between individual characteristics and epidemic dynamics is widely recognised, traditional mathematical models often overlook these factors. In this study, we examine two important aspects of human behaviour relevant to epidemics: contact patterns and vaccination uptake. Using data collected during the COVID-19 pandemic in Hungary, we first identify the dimensions along which individuals exhibit the greatest variation in their contact patterns and vaccination uptake. We find that generally higher socio-economic groups of the population have a higher number of contacts and a higher vaccination uptake with respect to disadvantaged groups. Subsequently, we propose a data-driven epidemiological model that incorporates these behavioural differences. Finally, we apply our model to analyse the fourth wave of COVID-19 in Hungary, providing valuable insights into real-world scenarios. By bridging the gap between individual characteristics and epidemic spread, our research contributes to a more comprehensive understanding of disease dynamics and informs effective public health strategies.

COVID-19 Pandemic Risk Assessment: Systematic Review

Background

The COVID-19 pandemic presents the possibility of future large-scale infectious disease outbreaks. In response, we conducted a systematic review of COVID-19 pandemic risk assessment to provide insights into countries' pandemic surveillance and preparedness for potential pandemic events in the post-COVID-19 era.

Objective

We aim to systematically identify relevant articles and synthesize pandemic risk assessment findings to facilitate government officials and public health experts in crisis planning.

Methods

This study followed the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines and included over 620,000 records from the World Health Organization COVID-19 Research Database. Articles related to pandemic risk assessment were identified based on a set of inclusion and exclusion criteria. Relevant articles were characterized based on study location, variable types, data-visualization techniques, research objectives, and methodologies. Findings were presented using tables and charts.

Results

Sixty-two articles satisfying both the inclusion and exclusion criteria were identified. Among the articles, 32.3% focused on local areas, while another 32.3% had a global coverage. Epidemic data were the most commonly used variables (74.2% of articles), with over half of them (51.6%) employing two or more variable types. The research objectives covered various aspects of the COVID-19 pandemic, with risk exposure assessment and identification of risk factors being the most common theme (35.5%). No dominant research methodology for risk assessment emerged from these articles.

Conclusion

Our synthesized findings support proactive planning and development of prevention and control measures in anticipation of future public health threats.

Effectiveness of social distancing measures and lockdowns for reducing transmission of COVID-19 in non-healthcare, community-based settings

Social distancing measures (SDMs) are community-level interventions that aim to reduce person-to-person contacts in the community. SDMs were a major part of the responses first to contain, then to mitigate, the spread of SARS-CoV-2 in the community. Common SDMs included limiting the size of gatherings, closing schools and/or workplaces, implementing work-from-home arrangements, or more stringent restrictions such as lockdowns. This systematic review summarized the evidence for the effectiveness of nine SDMs. Almost all of the studies included were observational in nature, which meant that there were intrinsic risks of bias that could have been avoided were conditions randomly assigned to study participants. There were no instances where only one form of SDM had been in place in a particular setting during the study period, making it challenging to estimate the separate effect of each intervention. The more stringent SDMs such as stay-at-home orders, restrictions on mass gatherings and closures were estimated to be most effective at reducing SARS-CoV-2 transmission. Most studies included in this review suggested that combinations of SDMs successfully slowed or even stopped SARS-CoV-2 transmission in the community. However, individual effects and optimal combinations of interventions, as well as the optimal timing for particular measures, require further investigation. This article is part of the theme issue 'The effectiveness of non-pharmaceutical interventions on the COVID-19 pandemic: the evidence'.

Tiered social distancing policies and epidemic control

Tiered social distancing policies have been adopted by many governments to mitigate the harmful consequences of COVID-19. Such policies have a number of well-established features, i.e. they are short-term, adaptive (to the changing epidemiological conditions), and based on a multiplicity of indicators of the prevailing epidemic activity. Here, we use ideas from Behavioural Epidemiology to represent tiered policies in an SEIRS model by using a composite information index including multiple indicators of current and past epidemic activity mimicking those used by governments during the COVID-19 pandemic, such as transmission intensity, infection incidence and hospitals’ occupancy. In its turn, the dynamics of the information index is assumed to endogenously inform the governmental social distancing interventions. The resulting model is described by a hereditary system showing a noteworthy property, i.e. a dependency of the endemic levels of epidemiological variables from initial conditions. This is a consequence of the need to normalize the different indicators to pool them into a single index. Simulations suggest a rich spectrum of possible results. These include policy suggestions and identify pitfalls and undesired outcomes, such as a worsening of epidemic control, that can arise following such types of approaches to epidemic responses.