Readers of this newsletter will likely agree that the statistical sciences should be central to global efforts to understand and manage the COVID-19 pandemic. The available data paints an incomplete picture of the current and past state of the pandemic for a variety of reasons, including incompleteness and non-representativeness of data, unobserved contact networks, and unknown infection times. Uncertainty is ubiquitous and rigorous statistical inference is essential. Sadly, biostatisticians have received only a tiny fraction of the limelight (and of the research funding) that has been bestowed on clinicians, mathematical modellers, and infectious disease epidemiologists. It is therefore notable and encouraging that a team of SSC regulars obtained funding from the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Public Health Agency of Canada (PHAC) as part of the Emerging Infectious Disease Management program aimed at strengthening Canada's capacity to respond to COVID-19 and future pandemics.
The team is led by Patrick Brown (SSC treasurer), and includes Erica Moodie (current CRM-SSC Prize recipient), Paul Gustafson (current Gold Medal recipient), Rob Deardon (SSC Biostats president-elect), Cindy Feng (Atlantic regional representative), Alexandra Schmidt (Quebec regional representative), Grace Yi (president-elect), Mahmoud Torabi (former Survey Methods president), Ed Susko (2011 CRM-SSC Prize recipient), Charmaine Dean (former SSC president), David Stephens (former CJS editor), Laura Cowen (current WNAR president and SSC regular), and Lam Ho (new investigators committee).
Three avenues of biostatistical methods research will be pursued.
First is forecasting and mapping the spatiotemporal distribution of infectious disease risk; estimation of prevalence and mortality rate by subgroup using population level data. This will help quantify the uncertainty in the future trajectory of cases and deaths, and understand the spatial distribution of infections and deaths in the population. As data that are available during a pandemic can be incomplete and biased, predicting infections in subpopulations, either geographic, ethnic, or occupational, requires careful analysis.
Second is estimating immunity in the population (and subpopulations) using seroprevalence surveys involving dried blood spot samples. Two large studies have collected dried blood spots from the Canadian population at a national level to measure COVID-19 antibodies, one of which is the Action-to-Beat Coronavirus (Ab-C) study which is currently collecting the third round of longitudinal samples. The tests are imperfect, and sensitivity declines over time, which combined with the non-representativeness of the sample means statistical methods for measurement error and survey sampling are required.
The third area of research is methods that can infer transmissibility and recovery rates for individual-level models (ILM) from population-representative data. ILM's explicitly model transmission and recover rates, and even the simplest of these models are computationally challenging to fit when data are incomplete, or times interval censored.
The project, Statistical Methods for Managing Emerging Infectious Diseases (SMMEID), has received an NSERC award of over $750,000 to pursue these objectives. A team of postdoctoral fellows and graduate students, each teamed up with several co-investigators, will be recruited to create an interdisciplinary network called the SMMEID network. The SMMEID network will work closely with public health agencies, partner organizations, scientists, engineers, and other experts in the field to create an arsenal of software tools, educational materials, and statistical methods that will augment Canada's capacity to respond to emerging infectious diseases.
By the end of the two-year funding period, the SMMEID group will have developed the foundations to support Canada’s capacity for EID management and control. Prospective postdocs are welcomed and encouraged to apply. A list of projects and instructions to apply can be found here.