ABSTRACT
COVID-19 epidemic dynamics are driven by a complex interplay of factors including population behaviour, government interventions, new variants, vaccination campaigns and immunity from prior infections. We aimed to quantify the epidemic drivers of SARS-CoV-2 dynamics in the Dominican Republic, an upper-middle income country of 10.8 million people, and assess the impact of the vaccination campaign implemented in February 2021 in saving lives and averting hospitalisations. We used an age-structured, multi-variant transmission dynamic model to characterise epidemic drivers in the Dominican Republic and explore counterfactual scenarios around vaccination coverage and population mobility. We fit the model to reported deaths, hospital bed occupancy, ICU bed occupancy and seroprevalence data until December 2021 and simulated epidemic trajectories under different counterfactual vaccination scenarios. We estimate that vaccination averted 5040 hospital admissions (95% CrI: 4750 - 5350), 1500 ICU admissions (95% CrI: 1420 - 1590) and 544 deaths (95% CrI: 488 - 606) in the first 6 months of the campaign. We also found that early vaccination with Sinovac-CoronaVac was preferable to delayed vaccination using a product with higher efficacy. We investigated the trade-off between changes in vaccination coverage and population mobility to understand how much relaxation of social distancing measures vaccination was able to 'buy' in the later stages of a pandemic. We found that if no vaccination had occurred, an additional decrease of 10-20% in population mobility would have been required to maintain the same death and hospitalisation outcomes. We found SARS-CoV-2 transmission dynamics in the Dominican Republic were driven by substantial accumulation of immunity during the first two years of the pandemic but that, despite this, vaccination was essential in enabling a return to pre-pandemic mobility levels without incurring considerable additional morbidity and mortality.
Subject(s)
COVID-19 , DeathABSTRACT
While individual level immune responses to SARS-CoV-2 are well characterized, population immunity and the factors that drive population immune markers are largely undescribed. In this study, we examined spike antibody responses that track with infection risk amongst a household cohort in the northwest and southeast of the Dominican Republic. Our sampling period was from Aug 2021 to Nov 2022, capturing sequential waves of Delta, BA.1, BA.2, and BA.4/5 transmission. We show that population antibody levels normalized from a highly irregular to a Gaussian distribution, driven by accrued infections and antibody boosting among individuals with lower baseline immunity and waning among those with higher immunity, irrespective of interval vaccination. Using a limited number of predictor variables and out-of-sample validation methods we were able to predict S-antibody changes at the Nov 2022 timepoint with a high degree of accuracy (Pearson’s correlation coefficient 0.95 for predicted vs observed change). S-antibody level at the baseline sampling timepoint was by far the most influential predictor, demonstrated by a strong association when used as the only predictor variable (Pearson’s correlation coefficient 0.92). Findings were stable across geographically distinct study regions, suggesting drivers of immune dynamics apply equally across the Dominican Republic, and likely other countries with comparable transmission profiles. Our results suggest that given sufficient transmission, generalizable and discernable principles underly population immune dynamics. We propose that these findings can be used to delineate immune dynamics in other settings, inform transmission modeling, and guide public health priorities for SARS-CoV-2, and potentially other non-immune sterilizing emerging pathogens.
ABSTRACT
We monitored SARS-CoV-2 antibody changes following implementation of a national COVID-19 vaccination campaign and assessed implications for immunological protection against variants of concern. Between March 2021 and August 2022, we prospectively enrolled 2,300 patients seeking care for undifferentiated febrile illnesses across two study sites in the Dominican Republic. Sera was tested for total anti-spike antibodies (anti-S) and simultaneously collected nasopharyngeal samples by RT-PCR for acute SARS-CoV-2 infection. Geometric mean anti-S titers increased from 6.6 BAU/ml (95% CI 5.1-8.7) to 1,332 BAU/ml (CI 1055-1,682) during the study period. Multivariable binomial odds ratios for acute SARS-CoV-2 infection were 0.55 (0.40-0.74), 0.38 (0.27-0.55), and 0.27 (0.18-0.40) for the second, third, and fourth versus the first anti-S quartile, with similar findings by viral strain. Integrated serological and virological screening present an opportunity to rethink existing surveillance platforms by simultaneously monitoring population-level immunological markers and implications for emerging variant transmission.