Quantifying previous SARS-CoV-2 infection through mixture modelling of antibody levels

Estimating SARS-CoV-2 infection probabilities with serological data and a Bayesian mixture model

The individual results of SARS-CoV-2 serological tests measured after the first pandemic wave of 2020 cannot be directly interpreted as a probability of having been infected. Plus, these results are usually returned as a binary or ternary variable, relying on predefined cut-offs. We propose a Bayesian mixture model to estimate individual infection probabilities, based on 81,797 continuous anti-spike IgG tests from Euroimmun collected in France after the first wave. This approach used serological results as a continuous variable, and was therefore not based on diagnostic cut-offs. Cumulative incidence, which is necessary to compute infection probabilities, was estimated according to age and administrative region. In France, we found that a "negative" or a "positive" test, as classified by the manufacturer, could correspond to a probability of infection as high as 61.8% or as low as 67.7%, respectively. "Indeterminate" tests encompassed probabilities of infection ranging from 10.8 to 96.6%. Our model estimated tailored individual probabilities of SARS-CoV-2 infection based on age, region, and serological result. It can be applied in other contexts, if estimates of cumulative incidence are available.

A Bayesian approach to estimating COVID-19 incidence and infection fatality rates

Naive estimates of incidence and infection fatality rates (IFR) of coronavirus disease 2019 suffer from a variety of biases, many of which relate to preferential testing. This has motivated epidemiologists from around the globe to conduct serosurveys that measure the immunity of individuals by testing for the presence of SARS-CoV-2 antibodies in the blood. These quantitative measures (titer values) are then used as a proxy for previous or current infection. However, statistical methods that use this data to its full potential have yet to be developed. Previous researchers have discretized these continuous values, discarding potentially useful information. In this article, we demonstrate how multivariate mixture models can be used in combination with post-stratification to estimate cumulative incidence and IFR in an approximate Bayesian framework without discretization. In doing so, we account for uncertainty from both the estimated number of infections and incomplete deaths data to provide estimates of IFR. This method is demonstrated using data from the Action to Beat Coronavirus erosurvey in Canada.

Estimating cutoff values for diagnostic tests to achieve target specificity using extreme value theory

BACKGROUND

Rapidly developing tests for emerging diseases is critical for early disease monitoring. In the early stages of an epidemic, when low prevalences are expected, high specificity tests are desired to avoid numerous false positives. Selecting a cutoff to classify positive and negative test results that has the desired operating characteristics, such as specificity, is challenging for new tests because of limited validation data with known disease status. While there is ample statistical literature on estimating quantiles of a distribution, there is limited evidence on estimating extreme quantiles from limited validation data and the resulting test characteristics in the disease testing context.

METHODS

We propose using extreme value theory to select a cutoff with predetermined specificity by fitting a Pareto distribution to the upper tail of the negative controls. We compared this method to five previously proposed cutoff selection methods in a data analysis and simulation study. We analyzed COVID-19 enzyme linked immunosorbent assay antibody test results from long-term care facilities and skilled nursing staff in Colorado between May and December of 2020.

RESULTS

We found the extreme value approach had minimal bias when targeting a specificity of 0.995. Using the empirical quantile of the negative controls performed well when targeting a specificity of 0.95. The higher target specificity is preferred for overall test accuracy when prevalence is low, whereas the lower target specificity is preferred when prevalence is higher and resulted in less variable prevalence estimation.

DISCUSSION

While commonly used, the normal based methods showed considerable bias compared to the empirical and extreme value theory-based methods.

CONCLUSIONS

When determining disease testing cutoffs from small training data samples, we recommend using the extreme value based-methods when targeting a high specificity and the empirical quantile when targeting a lower specificity.

Estimating SARS-CoV-2 seroprevalence

Governments and public health authorities use seroprevalence studies to guide responses to the COVID-19 pandemic. Seroprevalence surveys estimate the proportion of individuals who have detectable SARS-CoV-2 antibodies. However, serologic assays are prone to misclassification error, and non-probability sampling may induce selection bias. In this paper, non-parametric and parametric seroprevalence estimators are considered that address both challenges by leveraging validation data and assuming equal probabilities of sample inclusion within covariate-defined strata. Both estimators are shown to be consistent and asymptotically normal, and consistent variance estimators are derived. Simulation studies are presented comparing the estimators over a range of scenarios. The methods are used to estimate severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) seroprevalence in New York City, Belgium, and North Carolina.

A Mixture Model for Estimating SARS-CoV-2 Seroprevalence in Chennai, India

Serological assays used to estimate the prevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) often rely on manufacturers' cutoffs established on the basis of severe cases. We conducted a household-based serosurvey of 4,677 individuals in Chennai, India, from January to May 2021. Samples were tested for SARS-CoV-2 immunoglobulin G (IgG) antibodies to the spike (S) and nucleocapsid (N) proteins. We calculated seroprevalence, defining seropositivity using manufacturer cutoffs and using a mixture model based on measured IgG level. Using manufacturer cutoffs, there was a 5-fold difference in seroprevalence estimated by each assay. This difference was largely reconciled using the mixture model, with estimated anti-S and anti-N IgG seroprevalence of 64.9% (95% credible interval (CrI): 63.8, 66.0) and 51.5% (95% CrI: 50.2, 52.9), respectively. Age and socioeconomic factors showed inconsistent relationships with anti-S and anti-N IgG seropositivity using manufacturer cutoffs. In the mixture model, age was not associated with seropositivity, and improved household ventilation was associated with lower seropositivity odds. With global vaccine scale-up, the utility of the more stable anti-S IgG assay may be limited due to the inclusion of the S protein in several vaccines. Estimates of SARS-CoV-2 seroprevalence using alternative targets must consider heterogeneity in seroresponse to ensure that seroprevalence is not underestimated and correlates are not misinterpreted.

Serology Assays Used in SARS-CoV-2 Seroprevalence Surveys Worldwide: A Systematic Review and Meta-Analysis of Assay Features, Testing Algorithms, and Performance

Background: Many serological assays to detect SARS-CoV-2 antibodies were developed during the COVID-19 pandemic. Differences in the detection mechanism of SARS-CoV-2 serological assays limited the comparability of seroprevalence estimates for populations being tested. Methods: We conducted a systematic review and meta-analysis of serological assays used in SARS-CoV-2 population seroprevalence surveys, searching for published articles, preprints, institutional sources, and grey literature between 1 January 2020, and 19 November 2021. We described features of all identified assays and mapped performance metrics by the manufacturers, third-party head-to-head, and independent group evaluations. We compared the reported assay performance by evaluation source with a mixed-effect beta regression model. A simulation was run to quantify how biased assay performance affects population seroprevalence estimates with test adjustment. Results: Among 1807 included serosurveys, 192 distinctive commercial assays and 380 self-developed assays were identified. According to manufacturers, 28.6% of all commercial assays met WHO criteria for emergency use (sensitivity [Sn.] >= 90.0%, specificity [Sp.] >= 97.0%). However, manufacturers overstated the absolute values of Sn. of commercial assays by 1.0% [0.1, 1.4%] and 3.3% [2.7, 3.4%], and Sp. by 0.9% [0.9, 0.9%] and 0.2% [−0.1, 0.4%] compared to third-party and independent evaluations, respectively. Reported performance data was not sufficient to support a similar analysis for self-developed assays. Simulations indicate that inaccurate Sn. and Sp. can bias seroprevalence estimates adjusted for assay performance; the error level changes with the background seroprevalence. Conclusions: The Sn. and Sp. of the serological assay are not fixed properties, but varying features depending on the testing population. To achieve precise population estimates and to ensure the comparability of seroprevalence, serosurveys should select assays with high performance validated not only by their manufacturers and adjust seroprevalence estimates based on assured performance data. More investigation should be directed to consolidating the performance of self-developed assays.

Field evaluation of rapid diagnostic tests to determine dengue serostatus in Timor-Leste

The live attenuated tetravalent CYD-TDV vaccine (Dengvaxia) is effective but has scarcely been used due to safety concerns among seronegative recipients. Rapid diagnostic tests (RDTs) which can accurately determine individual dengue serostatus are needed for use in pre-vaccination screening. This study aimed to determine the performance of existing RDTs (which have been designed to detect levels of immunoglobulin G, IgG, associated with acute post-primary dengue) when repurposed for detection of previous dengue infection (where concentrations of IgG are typically lower). A convenience sample of four-hundred-and-six participants (including 217 children) were recruited in the community. Whole blood was collected by phlebotomy and tested using Bioline Dengue IgG/IgM (Abbott) and Standard Q Dengue IgM/IgG (SD Biosensor) RDTs in the field. Serum samples from the same individuals were also tested at National Health Laboratory. The Panbio indirect IgG ELISA was used as a reference test. Reference testing determined that 370 (91.1%) participants were dengue IgG seropositive. Both assays were highly specific (100.0%) but had low sensitivity (Bioline = 21.1% and Standard Q = 4.6%) when used in the field. Sensitivity was improved when RDTs were used under laboratory conditions, and when assays were allowed to run beyond manufacturer recommendations (and read at a delayed time-point), but specificity was reduced. Efforts to develop RDTs with high sensitivity and specificity for prior dengue infection which can be operationalised for pre-vaccination screening are ongoing. Performance of forthcoming candidate assays should be tested under field conditions with blood samples, as well as in the laboratory.

Informing healthcare operations with integrated pathology, clinical, and epidemiology data: Lessons from a single institution in Kenya during COVID-19 waves

Pathology, clinical care teams, and public health experts often operate in silos. We hypothesized that large data sets from laboratories when integrated with other healthcare data can provide evidence that can be used to optimize planning for healthcare needs, often driven by health-seeking or delivery behavior. From the hospital information system, we extracted raw data from tests performed from 2019 to 2021, prescription drug usage, and admission patterns from pharmacy and nursing departments during the COVID-19 pandemic in Kenya (March 2020 to December 2021). Proportions and rates were calculated. Regression models were created, and a t-test for differences between means was applied for monthly or yearly clustered data compared to pre-COVID-19 data. Tests for malaria parasite, Mycobacterium tuberculosis, rifampicin resistance, blood group, blood count, and histology showed a statistically significant decrease in 2020, followed by a partial recovery in 2021. This pattern was attributed to restrictions implemented to control the spread of COVID-19. On the contrary, D-dimer, fibrinogen, CRP, and HbA1c showed a statistically significant increase (p-value <0.001). This pattern was attributed to increased utilization related to the clinical management of COVID-19. Prescription drug utilization revealed a non-linear relationship to the COVID-19 positivity rate. The results from this study reveal the expected scenario in the event of similar outbreaks. They also reveal the need for increased efforts at diabetes and cancer screening, follow-up of HIV, and tuberculosis patients. To realize a broader healthcare impact, pathology departments in Africa should invest in integrated data analytics, for non-communicable diseases as well.

SARS-CoV-2 seroprevalence in three Kenyan health and demographic surveillance sites, December 2020-May 2021

BACKGROUND

Most of the studies that have informed the public health response to the COVID-19 pandemic in Kenya have relied on samples that are not representative of the general population. We conducted population-based serosurveys at three Health and Demographic Surveillance Systems (HDSSs) to determine the cumulative incidence of infection with SARS-CoV-2.

METHODS

We selected random age-stratified population-based samples at HDSSs in Kisumu, Nairobi and Kilifi, in Kenya. Blood samples were collected from participants between 01 Dec 2020 and 27 May 2021. No participant had received a COVID-19 vaccine. We tested for IgG antibodies to SARS-CoV-2 spike protein using ELISA. Locally-validated assay sensitivity and specificity were 93% (95% CI 88-96%) and 99% (95% CI 98-99.5%), respectively. We adjusted prevalence estimates using classical methods and Bayesian modelling to account for the sampling scheme and assay performance.

RESULTS

We recruited 2,559 individuals from the three HDSS sites, median age (IQR) 27 (10-78) years and 52% were female. Seroprevalence at all three sites rose steadily during the study period. In Kisumu, Nairobi and Kilifi, seroprevalences (95% CI) at the beginning of the study were 36.0% (28.2-44.4%), 32.4% (23.1-42.4%), and 14.5% (9.1-21%), and respectively; at the end they were 42.0% (34.7-50.0%), 50.2% (39.7-61.1%), and 24.7% (17.5-32.6%), respectively. Seroprevalence was substantially lower among children (<16 years) than among adults at all three sites (p≤0.001).

CONCLUSION

By May 2021 in three broadly representative populations of unvaccinated individuals in Kenya, seroprevalence of anti-SARS-CoV-2 IgG was 25-50%. There was wide variation in cumulative incidence by location and age.

Serology as a Tool to Assess Infectious Disease Landscapes and Guide Public Health Policy

Understanding the local burden and epidemiology of infectious diseases is crucial to guide public health policy and prioritize interventions. Typically, infectious disease surveillance relies on capturing clinical cases within a healthcare system, classifying cases by etiology and enumerating cases over a period of time. Disease burden is often then extrapolated to the general population. Serology (i.e., examining serum for the presence of pathogen-specific antibodies) has long been used to inform about individuals past exposure and immunity to specific pathogens. However, it has been underutilized as a tool to evaluate the infectious disease burden landscape at the population level and guide public health decisions. In this review, we outline how serology provides a powerful tool to complement case-based surveillance for determining disease burden and epidemiology of infectious diseases, highlighting its benefits and limitations. We describe the current serology-based technologies and illustrate their use with examples from both the pre- and post- COVID-19-pandemic context. In particular, we review the challenges to and opportunities in implementing serological surveillance in low- and middle-income countries (LMICs), which bear the brunt of the global infectious disease burden. Finally, we discuss the relevance of serology data for public health decision-making and describe scenarios in which this data could be used, either independently or in conjunction with case-based surveillance. We conclude that public health systems would greatly benefit from the inclusion of serology to supplement and strengthen existing case-based infectious disease surveillance strategies.

Sero-surveillance for IgG to SARS-CoV-2 at antenatal care clinics in three Kenyan referral hospitals: Repeated cross-sectional surveys 2020-21

INTRODUCTION

The high proportion of SARS-CoV-2 infections that have remained undetected presents a challenge to tracking the progress of the pandemic and estimating the extent of population immunity.

METHODS

We used residual blood samples from women attending antenatal care services at three hospitals in Kenya between August 2020 and October 2021and a validated IgG ELISA for SARS-Cov-2 spike protein and adjusted the results for assay sensitivity and specificity. We fitted a two-component mixture model as an alternative to the threshold analysis to estimate of the proportion of individuals with past SARS-CoV-2 infection.

RESULTS

We estimated seroprevalence in 2,981 women; 706 in Nairobi, 567 in Busia and 1,708 in Kilifi. By October 2021, 13% of participants were vaccinated (at least one dose) in Nairobi, 2% in Busia. Adjusted seroprevalence rose in all sites; from 50% (95%CI 42-58) in August 2020, to 85% (95%CI 78-92) in October 2021 in Nairobi; from 31% (95%CI 25-37) in May 2021 to 71% (95%CI 64-77) in October 2021 in Busia; and from 1% (95% CI 0-3) in September 2020 to 63% (95% CI 56-69) in October 2021 in Kilifi. Mixture modelling, suggests adjusted cross-sectional prevalence estimates are underestimates; seroprevalence in October 2021 could be 74% in Busia and 72% in Kilifi.

CONCLUSIONS

There has been substantial, unobserved transmission of SARS-CoV-2 in Nairobi, Busia and Kilifi Counties. Due to the length of time since the beginning of the pandemic, repeated cross-sectional surveys are now difficult to interpret without the use of models to account for antibody waning.