A realistic two-strain model for MERS-CoV infection uncovers the high risk for epidemic propagation
Open Access
- 14 February 2020
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS Neglected Tropical Diseases
- Vol. 14 (2) , e0008065
Abstract
Middle East Respiratory Syndrome Coronavirus (MERS-CoV) causes severe acute respiratory illness with a case fatality rate (CFR) of 35,5%. The highest number of MERS-CoV cases are from Saudi-Arabia, the major worldwide hotspot for this disease. In the absence of neither effective treatment nor a ready-to-use vaccine and with yet an incomplete understanding of its epidemiological cycle, prevention and containment measures can be derived from mathematical models of disease epidemiology. We constructed 2-strain models to predict past outbreaks in the interval 2012–2016 and derive key epidemiological information for Macca, Madina and Riyadh. We approached variability in infection through three different disease incidence functions capturing social behavior in response to an epidemic (e.g. Bilinear, BL; Non-monotone, NM; and Saturated, SAT models). The best model combination successfully anticipated the total number of MERS-CoV clinical cases for the 2015–2016 season and accurately predicted both the number of cases at the peak of seasonal incidence and the overall shape of the epidemic cycle. The evolution in the basic reproduction number (R0) warns that MERS-CoV may easily take an epidemic form. The best model correctly captures this feature, indicating a high epidemic risk (1≤R0≤2,5) in Riyadh and Macca and confirming the alleged co-circulation of more than one strain. Accurate predictions of the future MERS-CoV peak week, as well as the number of cases at the peak are now possible. These results indicate public health agencies should be aware that measures for strict containment are urgently needed before new epidemics take off in the region. There is currently no way to anticipate MERS-CoV epidemic outbreaks and strategies for disease prediction and containment are largely undermined by the limited knowledge of its epidemiological cycle. Not an effective treatment nor a vaccine for MERS-CoV exist to date. Instead, using three two-strain mathematical models that incorporate human social behavior as different disease incidence functions (e.g. bilinear, non-monotone and saturated), the best model combinations successfully anticipate the occurrence of the peak week in the season and the incidence at the peak. Our results confirm there are currently 2 strains co-circulating in the most populated regions in Saudi Arabia and highlight the high risk for large epidemic outbreaks, while the role of super-spreaders appears irrelevant for disease spread.Keywords
Funding Information
- Generalitat de Catalunya (SLT002/16/00466)
- UGC-DAE Consortium for Scientific Research, University Grants Commission (UGC-JRF23402016)
This publication has 39 references indexed in Scilit:
- Real-time influenza forecasts during the 2012–2013 seasonNature Communications, 2013
- Hospital Outbreak of Middle East Respiratory Syndrome CoronavirusNew England Journal of Medicine, 2013
- Forecasting seasonal outbreaks of influenzaProceedings of the National Academy of Sciences of the United States of America, 2012
- Isolation of a Novel Coronavirus from a Man with Pneumonia in Saudi ArabiaNew England Journal of Medicine, 2012
- Estimating the reproductive numbers for the 2008–2009 cholera outbreaks in ZimbabweProceedings of the National Academy of Sciences, 2011
- Incubation periods of acute respiratory viral infections: a systematic reviewThe Lancet Infectious Diseases, 2009
- Global analysis of an epidemic model with nonmonotone incidence rateMathematical Biosciences, 2007
- DRAM: Efficient adaptive MCMCStatistics and Computing, 2006
- A generalization of the Kermack-McKendrick deterministic epidemic modelMathematical Biosciences, 1978
- `Mean Crowding'Journal of Animal Ecology, 1967