Vaccine Effectiveness Against Hospitalization Among Adolescent and Pediatric SARS-CoV-2 Cases in Ontario, Canada

BackgroundVaccines against SARS-CoV-2 have been shown to reduce risk of infection, as well as severe disease among those with breakthrough infection, in adults. The latter effect is particularly important as Immune evasion by Omicron variants appears to have made vaccines less effective for prevention of infection. There is currently little available information on the protection conferred by vaccination against severe illness due to SARS-CoV-2 in children. MethodsTo minimize confounding by changing vaccination practices and dominant circulating viral variants, we performed an age- and time-matched nested case-control design. Reported SARS-CoV-2 case records in Ontario children and adolescents aged 4 to 17 were linked to vaccination records. We used multivariable logistic regression to estimate the effectiveness of one and two vaccine doses against hospitalization. ResultsWe identified 130 hospitalized SARS-CoV-2 cases and 1,300 non-hospitalized, age- and time-matched controls, with disease onset between May 28, 2021 and January 9, 2022. One vaccine dose was shown to be 34% effective against hospitalization among SARS-CoV-2 cases (aOR = 0.66 [95% CI: 0.34, 1.21]). In contrast, two doses were 56% (aOR = 0.44 [95% CI: 0.23, 0.83]) effective at preventing hospitalization among SARS-CoV-2 cases. Exploratory instrumental variable analyses, and calculation of E-values, suggested that these effects are unlikely to be explained by unmeasured confounding. ConclusionsEven with immune evasion by SARS-CoV-2 variants, two vaccine doses continue to provide protection against hospitalization among adolescent and pediatric SARS-CoV-2 cases, even when the vaccines do not prevent infection.

Severity of SARS-CoV-2 Infection in Pregnancy in Ontario: A Matched Cohort Analysis

BackgroundPregnancy represents a physiological state associated with increased vulnerability to severe outcomes from infectious diseases, both for the pregnant person and developing infant. The SARS-CoV-2 pandemic may have important health consequences for pregnant individuals, who may also be more reluctant than non-pregnant people to accept vaccination. We sought to estimate the degree to which increased severity of SARS-CoV-2 outcomes can be attributed to pregnancy. MethodsOur study made use of a population-based SARS-CoV-2 case file from Ontario, Canada. Due to both varying propensity to receive vaccination, and changes in dominant circulating viral strains over time, a time-matched cohort study was performed to evaluate the relative risk of severe illness in pregnant women with SARS-CoV-2 compared to other SARS-CoV-2 infected women of childbearing age (10 to 49 years old). Risk of severe SARS-CoV-2 outcomes (hospitalization or intensive care unit (ICU) admission) was evaluated in pregnant women and time-matched non-pregnant controls using multivariable conditional logistic regression. ResultsCompared to the rest of the population, non-pregnant women of childbearing age had an elevated risk of infection (standardized morbidity ratio (SMR) 1.28), while risk of infection was reduced among pregnant women (SMR 0.43). After adjustment for age, comorbidity, healthcare worker status, vaccination, and infecting viral variant, pregnant women had a markedly elevated risk of hospitalization (adjusted OR 4.96, 95% CI 3.86 to 6.37) and ICU admission (adjusted OR 6.58, 95% CI 3.29 to 13.18). The relative increase in hospitalization risk associated with pregnancy was greater in women without comorbidities than in those with comorbidities (P for heterogeneity 0.004). InterpretationA time-matched cohort study suggests that while pregnant women may be at a decreased risk of infection relative to the rest of the population, their risk of severe illness is markedly elevated if infection occurs. Given the safety of SARS-CoV-2 vaccines in pregnancy, risk-benefit calculus strongly favours SARS-CoV-2 vaccination in pregnant women.

Relative Virulence of SARS-CoV-2 Among Vaccinated and Unvaccinated Individuals Hospitalized with SARS-CoV-2

BackgroundThe rapid development of safe and effective vaccines against the SARS-CoV-2 virus has been a singular scientific achievement. Confounding due to health seeking behaviours and differential testing by vaccination status may bias analyses towards an apparent increase in infection severity following vaccination. We sought to determine whether risks of intensive care unit (ICU) admission and death were diminished significantly by vaccination, even in individuals for whom vaccination failed to prevent hospitalization. MethodsWe used data from Ontario, Canadas Case and Contact Management database, merged to a provincial vaccination dataset (COVaxON) to create a time-matched cohort of individuals who were hospitalized with SARS-CoV-2 infection. Each vaccinated individual was matched to up to five unvaccinated individuals based on test date of positive SARS-CoV-2 infection. Risk of ICU admission and death were evaluated using multivariable conditional logistic regression. Unmatched exploratory analyses were performed to identify sources of heterogeneity in vaccine effects. ResultsIn 20,064 individuals (3,353 vaccinated and 16,711 unvaccinated) hospitalized with infection due to SARS-CoV-2 between January 1st, 2021 and January 5th, 2022, vaccination with 1, 2, or 3 doses significantly reduced the risk of ICU admission and death. An inverse dose-response relationship was observed between vaccine doses received and both outcomes (adjusted odds ratio (aOR) for ICU admission per additional dose: 0.66, 95% CI 0.62 to 0.71; aOR for death per additional dose: 0.78, 95% CI 0.72 to 0.84). The reduction in risk was greater for ICU admission than for death (P for heterogeneity <0.05), but no significant differences in risk were seen based on infecting variant of concern (VOC). InterpretationWe identified a decrease in the risk of ICU admission and death in vaccinated individuals compared to unvaccinated, time-matched controls, even when vaccines failed to prevent infection sufficiently severe to cause hospitalization. Even with diminished efficacy of vaccines against infection with novel VOCs, vaccines remain an important tool for reduction of ICU admission and mortality.

Timing of Breakthrough Infection Risk After Vaccination Against SARS-CoV-2

BackgroundProvision of safe and effective vaccines has been a remarkable public health achievement during the SARS-CoV-2 pandemic. The effectiveness and durability of protection of the first two doses of SARS-CoV-2 vaccines is an important area for study, as are questions related to optimal dose combinations and dosing intervals. MethodsWe performed a case-cohort study to generate real-world evidence on efficacy of first and second dose of SARS-CoV-2 vaccines, using a population-based case line list and vaccination database for the province of Ontario, Canada between December 2020 and October 2021. Risk of infection after vaccination was evaluated in all laboratory-confirmed vaccinated SARS-CoV-2 cases, and a 2% sample of vaccinated controls, evaluated using survival analytic methods, including construction of Cox proportional hazards models. Vaccination status was treated as a time-varying covariate. ResultsFirst and second doses of SARS-CoV-2 vaccine markedly reduced risk of infection (first dose efficacy 68%, 95% CI 67% to 69%; second dose efficacy 88%, 95% CI 87 to 88%). In multivariable models, extended dosing intervals were associated with lowest risk of breakthrough infection (HR for redosing 0.64 (95% CI 0.61 to 0.67) at 6-8 weeks). Heterologous vaccine schedules that mixed viral vector vaccine first doses with mRNA second doses were significantly more effective than mRNA only vaccines. Risk of infection largely vanished during the time period 4-6 months after the second vaccine dose, but rose markedly thereafter. InterpretationA case-cohort design provided an efficient means to identify strong protective effects associated with SARS-CoV-2 vaccination, particularly after the second dose of vaccine. However, this effect appeared to wane once more than 6 months had elapsed since vaccination. Heterologous vaccination and extended dosing intervals improved the durability of immune response.

Impact of Population Mixing Between a Vaccinated Majority and Unvaccinated Minority on Disease Dynamics. Implications for SARS-CoV-2

BackgroundThe speed of vaccine development has been a singular achievement during the SARS-CoV-2 pandemic, though uptake has not been universal. Vaccine opponents often frame their opposition in terms of the rights of the unvaccinated. Our objective was to explore the impact of mixing of vaccinated and unvaccinated populations on risk among vaccinated individuals. MethodsWe constructed a simple Susceptible-Infectious-Recovered (SIR) compartmental model of a respiratory infectious disease with two connected sub-populations: vaccinated individuals and unvaccinated individuals. We simulated a spectrum of patterns of mixing between vaccinated and unvaccinated groups that ranged from random mixing to like-with-like mixing (complete assortativity), where individuals preferentially have contact with others with the same vaccination status. We evaluated the dynamics of an epidemic within each subgroup, and in the population as a whole. ResultsThe relative risk of infection was markedly higher among unvaccinated individuals than among vaccinated individuals. However, the contact-adjusted contribution of unvaccinated individuals to infection risk during the epidemic was disproportionate, with unvaccinated individuals contributing to infections among the vaccinated at a rate higher than would have been expected based on contact numbers alone. As assortativity increased, attack rates among the vaccinated decreased, but the contact-adjusted contribution to risk among vaccinated individuals derived from contact with unvaccinated individuals increased. InterpretationWhile risk associated with avoiding vaccination during a virulent pandemic accrues chiefly to the unvaccinated, the choices of unvaccinated individuals impact the health and safety of vaccinated individuals in a manner disproportionate to the fraction of unvaccinated individuals in the population.

Simple decision rules to predict local surges in COVID-19 hospitalizations during the winter and spring of 2022

Low rates of vaccination, emergence of novel variants of SARS-CoV-2, and increasing transmission relating to seasonal changes leave many U.S. communities at risk for surges of COVID-19 during the winter and spring of 2022 that might strain hospital capacity, as in previous waves. The trajectories of COVID-19 hospitalizations during this period are expected to differ across communities depending on their age distributions, vaccination coverage, cumulative incidence, and adoption of risk mitigating behaviors. Yet, existing predictive models of COVID-19 hospitalizations are almost exclusively focused on national- and state-level predictions. This leaves local policymakers in urgent need of tools that can provide early warnings about the possibility that COVID-19 hospitalizations may rise to levels that exceed local capacity. In this work, we develop simple decision rules to predict whether COVID-19 hospitalization will exceed the local hospitalization capacity within a 4- or 8-week period if no additional mitigating strategies are implemented during this time. These decision rules use real-time data related to hospital occupancy and new hospitalizations associated with COVID-19, and when available, genomic surveillance of SARS-CoV-2. We showed that these decision rules present reasonable accuracy, sensitivity, and specificity (all [≥]80%) in predicting local surges in hospitalizations under numerous simulated scenarios, which capture substantial uncertainties over the future trajectories of COVID-19 during the winter and spring of 2022. Our proposed decision rules are simple, visual, and straightforward to use in practice by local decision makers without the need to perform numerical computations. Significance StatementIn many U.S. communities, the risk of exceeding local healthcare capacity during the winter and spring of 2022 remains substantial since COVID-19 hospitalizations may rise due to seasonal changes, low vaccination coverage, and the emergence of new variants of SARS-CoV-2, such as the omicron variant. Here, we provide simple and easy-to-communicate decision rules to predict whether local hospital occupancy is expected to exceed capacity within a 4- or 8-week period if no additional mitigating measures are implemented. These decision rules can serve as an alert system for local policymakers to respond proactively to mitigate future surges in the COVID-19 hospitalization and minimize risk of overwhelming local healthcare capacity.

Age-Specific Changes in Virulence Associated with SARS-CoV-2 Variants of Concern

BackgroundNovel variants of concern (VOCs) have been associated with both increased infectivity and virulence of SARS-CoV-2. The virulence of SARS-CoV-2 is closely linked to age. Whether relative increases in virulence of novel VOCs is similar across the age spectrum, or is limited to some age groups, is unknown. MethodsWe created a retrospective cohort of people in Ontario, Canada testing positive for SARS-CoV-2 and screened for VOCs, with dates of test report between February 7 and August 30, 2021 (n=233,799). Cases were classified as N501Y-positive VOC, probable Delta VOC, or VOC undetected. We constructed age-specific logistic regression models to evaluate the effects of N501Y-postive or Delta VOC infections on infection severity, using hospitalization, intensive care unit (ICU) admission, and death as outcome variables. Models were adjusted for sex, time, health unit, vaccination status, comorbidities, immune compromise, long-term care residence, healthcare worker status, and pregnancy. ResultsInfection with either N501Y-positive or Delta VOCs was associated with significant elevations in risk of hospitalization, ICU admission, and death in younger and older adults, compared to infections where a VOC was not detected. Delta VOC increased hospitalization risk in children under 10 by a factor of 2.5 (adjusted odds ratio, 95% confidence interval: 1.2 to 5.1) compared to non-VOC. For most VOC-outcome combinations there was no heterogeneity in adverse outcomes by age. However, there was an inverse relationship between age and relative increase in risk of death with delta VOC, with younger age groups showing a greater relative increase in risk of death than older individuals. InterpretationSARS-CoV-2 VOCs appear to be associated with increased relative virulence of infection in all age groups, though low absolute numbers of outcomes in younger individuals make estimates in these groups imprecise.

Progressive Increase in Virulence of Novel SARS-CoV-2 Variants in Ontario, Canada, February to June, 2021

BackgroundThe period from February to June 2021 was one during which initial wild-type SARS-CoV-2 strains were supplanted in Ontario, Canada, first by variants of concern (VOC) with the N501Y mutation (Alpha/B1.1.17, Beta/B.1.351 and Gamma/P.1 variants), and then by the Delta/B.1.617 variant. The increased transmissibility of these VOCs has been documented but data for increased virulence is limited. We used Ontarios COVID-19 case data to evaluate the virulence of these VOCs compared to non-VOC SARS-CoV-2 infections, as measured by risk of hospitalization, intensive care unit (ICU) admission, and death. MethodsWe created a retrospective cohort of people in Ontario testing positive for SARS-CoV-2 and screened for VOCs, with dates of test report between February 7 and June 27, 2021 (n=212,332). We constructed mixed effects logistic regression models with hospitalization, ICU admission, and death as outcome variables. Models were adjusted for age, sex, time, vaccination status, comorbidities, and pregnancy status. Health units were included as random intercepts. ResultsCompared to non-VOC SARS-CoV-2 strains, the adjusted elevation in risk associated with N501Y-positive variants was 52% (43-62%) for hospitalization; 89% (67-116%) for ICU admission; and 51% (30-74%) for death. Increases with Delta variant were more pronounced: 108% (80-138%) for hospitalization; 234% (164-331%) for ICU admission; and 132% (47-230%) for death. InterpretationThe progressive increase in transmissibility and virulence of SARS-CoV-2 VOCs will result in a significantly larger, and more deadly, pandemic than would have occurred in the absence of VOC emergence.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) seroprevalence: Navigating the absence of a gold standard

BackgroundMultiple anti-SARS-CoV-2 immunoassays are available, but no gold standard exists. We assessed four assays using various methodological approaches to estimate SARS-COV-2 seroprevalence during the first COVID-19 wave in Canada. MethodsThis serial cross-sectional study was conducted using plasma samples from healthy blood donors between April-September 2020. Qualitative assessment of SARS-CoV-2 IgG antibodies was based on four assays: Abbott Architect SARS-Cov-2 IgG assay (target nucleocapsid) (Abbott-NP) and three in-house IgG ELISA assays (target spike glycoprotein (Spike), spike receptor binding domain (RBD), and nucleocapsid (NP)). Seroprevalence was estimated using multiple composite reference standards (CRS) and by a series of Bayesian Latent Class Models (BLCM) (using uninformative, weakly, and informative priors). Results8999 blood samples were tested. The Abbott-NP assay consistently estimated seroprevalence to be lower than the ELISA-based assays. Discordance between assays was common, 13 unique diagnostic phenotypes were observed. Only 32 samples (0.4%) were positive by all four assays. BLCM using uninformative priors predicted seroprevalence increased from 0.7% (95% credible interval (CrI); 0.4, 1.0%) in April/May to 0.8% (95% CrI 0.5, 1.2%) in June/July to 1.1% (95% CrI 0.7, 1.6) in August/September. Results from CRS were very similar to the BLCM. Assay characteristics varied considerably over time. Overall spike had the highest sensitivity (89.1% (95% CrI 79.2, 96.9%), while the sensitivity of the Abbott-NP assay waned from 65.3% (95% CrI 43.6, 85.0%) in April/May to 45.9% (95% CrI 27.8, 65.6) by August/September. DiscussionWe found low SARS-CoV-2 seroprevalence rates at the end of the first wave and estimates derived from single assays may be biased. SummaryMultiple anti-SARS-CoV-2 immunoassays are available, but no gold standard exists. We used four unique assays to estimate very low SARS-COV-2 seroprevalence during the first COVID-19 wave in Canada. Caution should be exercised when interpretating seroprevalence estimates from single assays.

Estimated Deaths, Intensive Care Admissions and Hospitalizations Averted in Canada during the COVID-19 Pandemic

IntroductionNational responses to the SARS-CoV-2 pandemic have been highly variable, which may explain some of the heterogeneity in the pandemics health and economic impacts across the world. We sought to explore the effectiveness of the Canadian pandemic response relative to responses in four peer countries with similar political, economic and health systems, and with close historical and cultural ties to Canada (the United States, United Kingdom, France, and Australia) from March 2020 to May 2022. MethodsWe used reported age-specific mortality data to generate estimates of pandemic mortality standardized to the Canadian population. Age-specific case fatality, hospitalization, and intensive care admission probabilities for the Canadian province of Ontario were applied to estimated deaths in order to calculate hospitalizations and intensive care admissions averted by the Canadian response. The monetary value of averted hospitalizations was estimated using cost estimates from the Canadian Institute for Health Information. Age-specific quality-adjusted life-years (QALY) lost due to fatality were estimated using published estimates. QALY were monetized using a net expected benefit approach. ResultsRelative to the United States, United Kingdom, and France, the Canadian pandemic response was estimated to have averted 94,492, 64,306 and 13,641 deaths respectively, with more than 480,000 hospitalizations averted, and 1 million QALY saved, relative to the United States. A United States pandemic response applied to Canada would have resulted in more than $40 billion in economic losses due to healthcare expenditures and lost QALY; losses relative to the United Kingdom and France would have been $21 billion and $5 billion respectively. By contrast, an Australian pandemic response would have averted over 28,000 additional deaths and averted nearly $9 billion in costs in Canada. ConclusionsCanada outperformed peer countries that aimed for mitigation, rather than elimination, of SARS-CoV-2 in the first two years of the pandemic, likely because of a more stringent public health response to disease transmission. This resulted in substantial numbers of lives saved and economic costs averted. However, comparison with Australia demonstrates that an elimination focus would have allowed Canada to save tens of thousands of lives, and would have saved substantial economic costs.

Anticipating the curve: can online symptom-based data reflect COVID-19 case activity in Ontario, Canada?

BackgroundLimitations in laboratory diagnostic capacity and reporting delays have hampered efforts to mitigate and control the ongoing COVID-19 pandemic globally. Syndromic surveillance of COVID-19 is an important public health tool that can help detect outbreaks, mobilize a rapid response, and thereby reduce morbidity and mortality. The primary objective of this study was to determine whether syndromic surveillance through self-reported COVID-19 symptoms could be a timely proxy for laboratory-confirmed case trends in the Canadian province of Ontario. MethodsWe retrospectively analyzed self-reported symptoms data collected using an online tool - Outbreaks Near Me (ONM) - from April 20th to Oct 11th, 2020 in Ontario, Canada. We estimated the correlation coefficient between the weekly proportion of respondents reporting a COVID-like illness (CLI) to both the weekly number of PCR-confirmed COVID-19 cases and the percent positivity in the same period for the same week and with a one-week lag. ResultsThere were 314,686 responses from 188,783 unique respondents to the ONM symptom survey. Respondents were more likely to be female and be in the 40-59 age demographic compared to the Ontario general population. There was a strong positive correlation between the weekly number of reported cases in Ontario and the percent of respondents reporting CLI each week (r = 0.89, p <0.01) and with a one-week lag (r = 0.89, p <0.01). InterpretationWe demonstrate a strong positive and significant correlation (r = 0.89, p <0.01) between percent of self-reported COVID-like illness and the subsequent weeks COVID-19 cases reported, highlighting that a rise in CLI may precede official statistics by at least 1 week. This demonstrates the utility of syndromic surveillance in predicting near-future disease activity. Digital surveillance systems are low-cost tools that may help measure the burden of COVID-19 in a community if there is under-detection of cases through conventional laboratory diagnostic testing. This additional information can be used to guide a healthcare response and policy decisions.

Temporal Associations between Community Incidence of COVID-19 and Nursing Home Outbreaks in Ontario, Canada

The risk of nursing home COVID-19 outbreaks is strongly associated with the rate of infection in the communities surrounding homes, yet the temporal relationship between rising rates of community COVID-19 infection and the risk threshold for subsequent nursing home COVID-19 outbreaks is not well defined. This population-based cohort study included all COVID-19 cases in Canadas most populous Province of Ontario between March 1-July 16, 2020. We evaluated the temporal relationship between trends in the number of active community COVID-19 cases and the number of nursing home outbreaks. We found that the average lag time between community cases and nursing home outbreaks was 23 days for Ontario overall, with substantial variability across geographic regions. We also determined thresholds of community incidence of COVID-19 associated with a 75% probability of observing a nursing home outbreak 5, 10 and 15 days into the future. For the province overall, when daily active COVID-19 community cases are 2.30 per 100,000 population, there is a 75% probability of a nursing home outbreak occurring five days later.

Simple Accurate Regression-Based Forecasting of Intensive Care Unit Admissions due to COVID-19 in Ontario, Canada

The pandemic caused by SARS-CoV-2 has proven challenging clinically, and at the population level, due to heterogeneity in both transmissibility and severity. Recent case incidence in Ontario, Canada (autumn 2020) has outstripped incidence in seen during the first (spring) pandemic wave; but has been associated with a lower incidence of intensive care unit (ICU) admissions and deaths. We hypothesized that differential ICU burden might be explained by increased testing volumes, as well as the shift in mean case age from older to younger. We constructed a negative binomial regression model using only three covariates, at a 2-week lag: log10(weekly cases); log10(weekly deaths); and mean weekly case age. This model reproduced observed ICU admission volumes, and demonstrated good preliminary predictive validity. Furthermore, when admissions were used in combination with ICU length of stay, our modeled estimates demonstrated excellent convergent validity with ICU occupancy data reported by the Canadian Institute for Health Information. Our approach needs external validation in other settings and at larger and smaller geographic scales, but appears to be a useful short-term forecasting tool for ICU resource demand; we also demonstrate that the virulence of SARS-CoV-2 infection has not meaningfully changed in Ontario between the first and second waves, but the demographics of those infected, and the fraction of cases identified, have.

Age-Specific SARS-CoV-2 Infection Fatality and Case Identification Fraction in Ontario, Canada

BackgroundSARS-CoV-2 is a novel pandemic pathogen that displays great variability in virulence across cases. Due to limitations in diagnostic testing only a subset of infections are identified. Underestimation of true infections makes calculation of infection fatality ratios (IFR) challenging. Seroepidemiology allows estimation of true cumulative incidence of infection in populations, for estimation of IFR. MethodsSeroprevalence estimates were derived using retention samples stored by Canadian Blood Services in May 2020. These were compared to non-long-term care-linked case and fatality data from the same period. Estimates were combined to generate IFR and case identification fraction estimates. ResultsOverall IFR was estimated to be 0.80% (0.75 to 0.85%), consistent with estimates from other jurisdictions. IFR increased exponentially with age from 0.01% (0.002 to 0.04%) in those aged 20-29 years, to 12.71% (4.43 to 36.50%) in those aged 70 and over. We estimated that 5.88 infections (3.70 to 9.21) occurred for every case identified, with a higher fraction of cases identified in those aged 70 and older (42.0%) than those aged 20-29 (9.4%). IFR estimates in those aged 60 and older were identical to pooled estimates from other countries. ConclusionsTo our knowledge these are the first Canadian estimates SARS-CoV-2 IFR and case identification fraction. Notwithstanding biases associated with donor sera they are similar to estimates from other countries, and approximately 80-fold higher than estimates for influenza A (H1N1) during the 2009 epidemic. Ontarios first COVID-19 pandemic wave is likely to have been accurately characterized due to a high case identification fraction.

COVID-19 Case-Age Distribution: Correction for Differential Testing by Age

BackgroundSARS-CoV-2 is a novel pathogen and is currently the cause of a global pandemic. Despite expected universal susceptibility to a novel pathogen, the pandemic to date has been characterized by higher observed incidence in the oldest individuals and lower incidence in children and adolescents. Differential testing by age group may explain some of these observed differences, but datasets linking case counts to public health testing volumes are uncommon. MethodsWe used data from Ontario, Canada. Case data were obtained from Ontarios provincial line, while testing data were obtained from an information system with complete SARS-CoV-2 testing data for public, hospital, and private laboratories. Demographic and temporal patterns in reported case incidence, testing rates, and test positivity were explored using negative binomial regression models. Standardized morbidity and testing ratios (SMR, STR), and standardized test positivity (STP) were calculated by dividing age- and sex-specific rates by overall rates; demographic and temporal patterns in standardized ratios were explored using meta-regression. Testing adjusted SMR were estimated using linear regression models. ResultsObserved disease incidence and testing rates were highest in oldest individuals and markedly lower in those aged < 20. Temporal trends in disease incidence and testing were observed, but standardizing morbidity and testing ratios eliminated temporal trends (i.e., relative patterns by age and sex remained identical regardless of epidemic phase). After adjustment for testing frequency, SMR were lowest in children and adults aged 70 and older, approximately the same in adolescents as in the population as a whole and elevated in young adults (aged 20-29 years), providing a markedly different picture of the epidemic than seen with crude SMR or case-based incidence. Test-adjusted SMR were validated using seroprevalence data (Pearson correlation coefficient 0.82, P = 0.04). ConclusionsSurveillance for SARS-CoV-2 infection is typically performed using only test-positive case data, without adjustment for testing frequency. Older adults are tested more frequently, likely due to increased disease severity, while children are under-tested. Adjustment for testing frequency results in a very different picture of SARS-CoV-2 infection risk by age, one that is consistent with estimates obtained through serological testing.

Derivation and Validation of Clinical Prediction Rule for COVID-19 Mortality in Ontario, Canada

BackgroundSARS-CoV-2 is currently causing a high mortality global pandemic. However, the clinical spectrum of disease caused by this virus is broad, ranging from asymptomatic infection to cytokine storm with organ failure and death. Risk stratification of individuals with COVID-19 would be desirable for management, prioritization for trial enrollment, and risk stratification. We sought to develop a prediction rule for mortality due to COVID-19 in individuals with diagnosed infection in Ontario, Canada. MethodsData from Ontarios provincial iPHIS system were extracted for the period from January 23 to May 15, 2020. Both logistic regression-based prediction rules, and a rule derived using a Cox proportional hazards model, were developed in half the study and validated in remaining patients. Sensitivity analyses were performed with varying approaches to missing data. Results21,922 COVID-19 cases were reported. Individuals assigned to the derivation and validation sets were broadly similar. Age and comorbidities (notably diabetes, renal disease and immune compromise) were strong predictors of mortality. Four point-based prediction rules were derived (base case, smoking excluded as a predictor, long-term care excluded as a predictor, and Cox model based). All rules displayed excellent discrimination (AUC for all rules > 0.92) and calibration (both by graphical inspection and P > 0.50 by Hosmer-Lemeshow test) in the derivation set. All rules performed well in the validation set and were robust to random replacement of missing variables, and to the assumption that missing variables indicated absence of the comorbidity or characteristic in question. ConclusionsWe were able to use a public health case-management data system to derive and internally validate four accurate, well-calibrated and robust clinical prediction rules for COVID-19 mortality in Ontario, Canada. While these rules need external validation, they may be a useful tool for clinical management, risk stratification, and clinical trials.

Reduced COVID-19-Related Critical Illness and Death, and High Risk of Epidemic Resurgence, After Physical Distancing in Ontario, Canada

BackgroundInsights from epidemiological models have helped to both guide and better understand COVID-19 mitigation policies that have been adopted across the globe. Many early models focussed on initial control options and were less reliant on fitting to observed data. As the pandemic progresses, models can be used to quantify the impact that control measures have had and what may unfold when such measures are relaxed. ObjectiveTo explore the impact of physical distancing measures on COVID-19 transmission in the population of Ontario, Canada.

Failing our Most Vulnerable: COVID-19 and Long-Term Care Facilities in Ontario

BackgroundThe COVID-19 epidemic has taken a fearsome toll on individuals residing in long-term care facilities (LTC). As of April 10, 2020 half of Canadas COVID-19 deaths had occurred in LTC. We sought to better understand trends and risk factors for COVID-19 death in LTC in Ontario. MethodsWe analyzed a COVID-19 outbreak database created by the Ontario Ministry of Health, for the period March 29-April 7, 2020. Mortality incidence rate ratios for LTC were calculated with community living Ontarians aged > 69 used as the comparator group. Count-based regression methods were used to model temporal trends and identify associations between infection risk in staff and residents, and subsequent LTC resident death. ResultsConfirmed or suspected cases of COVID-19 were identified in 272/627 LTC by April 7, 2020. The incidence rate ratio for COVID-19 death was 13.1 (9.9-17.3) relative to community living adults over 69. Incidence rate ratio increased over time and was 87.28 (90% CrI 9.98 to 557.08) by April 7, 2020. Lagged infection in staff was a strong predictor of death in residents (e.g., adjusted IRR for death per infected staff member 1.17, 95% CI 1.11 to 1.26 at a 6-day lag). InterpretationMortality risk in elders in Ontario is currently concentrated in LTC, and this risk has increased sharply over a short period of time. Early identification of risk requires a focus on testing and provision of personal protective equipment to staff, and restructuring the LTC workforce to prevent movement of COVID-19 between LTC. FundingThe research was supported by a grant to DNF from the Canadian Institutes for Health Research (2019 COVID-19 rapid researching funding OV4-170360).

Standardization and Age-Distribution of COVID-19: Implications for Variability in Case Fatality and Outbreak Identification

BackgroundEpidemiological data from the COVID-19 pandemic has demonstrated variability in attack rates by age, and country-to-country variability in case fatality ratio (CFR). ObjectiveTo use direct and indirect standardization for insights into the impact of age-specific under-reporting on between-country variability in CFR, and apparent size of COVID-19 epidemics. DesignPost-hoc secondary data analysis ("case studies"), and mathematical modeling. SettingChina, global. InterventionsNone. MeasurementsData were extracted from a sentinel epidemiological study by the Chinese Center for Disease Control (CCDC) that describes attack rates and CFR for COVID-19 in China prior to February 12, 2020. Standardized morbidity ratios (SMR) were used to impute missing cases and adjust CFR. Age-specific attack rates and CFR were applied to different countries with differing age structures (Italy, Japan, Indonesia, and Egypt), in order to generate estimates for CFR, apparent epidemic size, and time to outbreak recognition for identical age-specific attack rates. ResultsSMR demonstrated that 50-70% of cases were likely missed during the Chinese epidemic. Adjustment for under-recognition of younger cases decreased CFR from 2.4% to 0.8% (assuming 50% case ascertainment in older individuals). Standardizing the Chinese epidemic to countries with older populations (Italy, and Japan) resulted in larger apparent epidemic sizes, higher CFR and earlier outbreak recognition. The opposite effect was demonstrated for countries with younger populations (Indonesia, and Egypt). LimitationsSecondary data analysis based on a single country at an early stage of the COVID-19 pandemic, with no attempt to incorporate second order effects (ICU saturation) on CFR. ConclusionDirect and indirect standardization are simple tools that provide key insights into between-country variation in the apparent size and severity of COVID-19 epidemics. FundingThe research was supported by a grant to DNF from the Canadian Institutes for Health Research (2019 COVID-19 rapid researching funding OV4-170360).

Mathematical modeling of COVID-19 transmission and mitigation strategies in the population of Ontario, Canada

BackgroundWe evaluated how non-pharmaceutical interventions could be used to control the COVID-19 pandemic and reduce the burden on the healthcare system. MethodsUsing an age-structured compartmental model of COVID-19 transmission in the population of Ontario, Canada, we compared a base case with limited testing, isolation, and quarantine to scenarios with: enhanced case finding; restrictive social distancing measures; or a combination of enhanced case finding and less restrictive social distancing. Interventions were either implemented for fixed durations or dynamically cycled on and off, based on projected ICU bed occupancy. We present median and credible intervals (CrI) from 100 replicates per scenario using a two-year time horizon. ResultsWe estimated that 56% (95% CrI: 42-63%) of the Ontario population would be infected over the course of the epidemic in the base case. At the epidemic peak, we projected 107,000 (95% CrI: 60,760-149,000) cases in hospital and 55,500 (95% CrI: 32,700-75,200) cases in ICU. For fixed duration scenarios, all interventions were projected to delay and reduce the height of the epidemic peak relative to the base case, with restrictive social distancing estimated to have the greatest effect. Longer duration interventions were more effective. Dynamic interventions were projected to reduce the proportion of the population infected at the end of the two-year period. Dynamic social distancing interventions could reduce the median number of cases in ICU below current estimates of Ontarios ICU capacity. InterpretationWithout significant social distancing or a combination of moderate social distancing with enhanced case finding, we project that ICU resources would be overwhelmed. Dynamic social distancing could maintain health system capacity and also allow periodic psychological and economic respite for populations.