Global-scale modeling of early factors and country-specific trajectories of COVID-19 incidence: a cross-sectional study of the first 6 months of the pandemic

<p><span>Studies examining factors responsible for COVID-19 incidence </span><span>are</span><span> mostly focused at the national or sub-national level. A global-level characterization of contributing factors and temporal trajectories of disease incidence is lacking. Here we conducted a global-scale analysis of COVID-19 infections to identify key factors associated with early disease incidence. Additionally, we compared longitudinal trends of COVID-19 incidence at a per-country level and classified countries based on COVID-19 incidence trajectories and effects of lockdown responses. </span><span>Univariate analysis identified</span> <span>eleven variables as independently associated with COVID-19 infections at a global level (p&lt;1e-05). Multivariable analysis identified a 4-variable model as optimal for explaining global variations in COVID-19 (p&lt;0.01). COVID-19 case trajectories for most countries were best captured by a log-logistic model, as determined by AIC estimates. Six predominant country clusters were identified when characterizing the effects of lockdown intervals on variations in COVID-19 new cases per country.</span> <span>Globally, economic and meteorological factors are important determinants of early COVID-19 incidence. Analysis of longitudinal trends and lockdown effects on COVID-19 highlights important nuances in country-specific responses to infections. These results </span><span>provide valuable insights into disease incidence at a per-country level, possibly allowing for more informed decision making by individual governments in future disease outbreaks.</span></p>