Early transmission dynamics of COVID-19 in a southern hemisphere setting: Lima-Peru: February 29th-March 30th, 2020

Abstract

The COVID-19 pandemic that emerged in Wuhan China has generated substantial morbidity and mortality impact around the world during the last four months. The daily trend in reported cases has been rapidly rising in Latin America since March 2020 with the great majority of the cases reported in Brazil followed by Peru as of April 15th, 2020. Although Peru implemented a range of social distancing measures soon after the confirmation of its first case on March 6th, 2020, the daily number of new COVID-19 cases continues to accumulate in this country. We assessed the early COVID-19 transmission dynamics and the effect of social distancing interventions in Lima, Peru. We estimated the reproduction number, R, during the early transmission phase in Lima from the daily series of imported and autochthonous cases by the date of symptoms onset as of March 30th, 2020. We also assessed the effect of social distancing interventions in Lima by generating short-term forecasts grounded on the early transmission dynamics before interventions were put in place. Prior to the implementation of the social distancing measures in Lima, the local incidence curve by the date of symptoms onset displays near exponential growth dynamics with the mean scaling of growth parameter, p, estimated at 0.96 (95% CI: 0.87, 1.0) and the reproduction number at 2.3 (95% CI: 2.0, 2.5). Our analysis indicates that school closures and other social distancing interventions have helped slow down the spread of the novel coronavirus, with the nearly exponential growth trend shifting to an approximately linear growth trend soon after the broad scale social distancing interventions were put in place by the government. While the interventions appear to have slowed the transmission rate in Lima, the number of new COVID-19 cases continue to accumulate, highlighting the need to strengthen social distancing and active case finding efforts to mitigate disease transmission in the region.

Keywords: COVID-19; Generalized growth model; Reproduction number; SARS-CoV-2; Short-term forecast; Transmission potential.

Fig. 1

Laboratory results of COVID-19 tests in Lima as of March 30^th, 2020. Blue color represents the negative test results and the yellow color represents the positive test results. The orange solid line denotes the COVID-19 positivity rate in Lima.

Fig. 2

Daily numbers of new local and imported confirmed COVID-19 cases in Lima by date of symptoms onset as of March 30^th, 2020.

Fig. 3

The reproduction number derived from the early growth phase in the number of COVID-19 cases in Lima after adjusting for imported cases with $\alpha =0.15$ using the GGM model as described in the text. The reproduction number based on the incidence curve by March 15^th, 2020 was estimated at 2.3 (95% CI: 2.0, 2.5).

Fig. 4

20-day ahead forecast of the COVID-19 epidemic in Lima by calibrating the GGM model until March 15^th, 2020 (vertical dashed line). Blue circles correspond to the data points, the red solid line indicates the model's mean fit and the red dashed lines represent the 95% prediction interval. The vertical black dashed line represents the time of the start of the forecast period. The forecast (March 16^th- March 30^th) suggests that social distancing interventions have slowed down the transmission rate.