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.

Assessing the burden of COVID-19 in developing countries: systematic review, meta-analysis and public policy implications.

INTRODUCTION: The infection fatality rate (IFR) of COVID-19 has been carefully measured and analysed in high-income countries, whereas there has been no systematic analysis of age-specific seroprevalence or IFR for developing countries. METHODS: We systematically reviewed the literature to identify all COVID-19 serology studies in developing countries that were conducted using representative samples collected by February 2021. For each of the antibody assays used in these serology studies, we identified data on assay characteristics, including the extent of seroreversion over time. We analysed the serology data using a Bayesian model that incorporates conventional sampling uncertainty as well as uncertainties about assay sensitivity and specificity. We then calculated IFRs using individual case reports or aggregated public health updates, including age-specific estimates whenever feasible. RESULTS: In most locations in developing countries, seroprevalence among older adults was similar to that of younger age cohorts, underscoring the limited capacity that these nations have to protect older age groups.Age-specific IFRs were roughly 2 times higher than in high-income countries. The median value of the population IFR was about 0.5%, similar to that of high-income countries, because disparities in healthcare access were roughly offset by differences in population age structure. CONCLUSION: The burden of COVID-19 is far higher in developing countries than in high-income countries, reflecting a combination of elevated transmission to middle-aged and older adults as well as limited access to adequate healthcare. These results underscore the critical need to ensure medical equity to populations in developing countries through provision of vaccine doses and effective medications.

Ethnoracial Disparities in SARS-CoV-2 Seroprevalence in a Large Cohort of Individuals in Central North Carolina from April to December 2020.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused millions of deaths around the world within the past 2 years. Transmission within the United States has been heterogeneously distributed by geography and social factors with little data from North Carolina. Here, we describe results from a weekly cross-sectional study of 12,471 unique hospital remnant samples from 19 April to 26 December 2020 collected by four clinical sites within the University of North Carolina Health system, with a majority of samples from urban, outpatient populations in central North Carolina. We employed a Bayesian inference model to calculate SARS-CoV-2 spike protein immunoglobulin prevalence estimates and conditional odds ratios for seropositivity. Furthermore, we analyzed a subset of these seropositive samples for neutralizing antibodies. We observed an increase in seroprevalence from 2.9 (95% confidence interval [CI], 1.8 to 4.5) to 12.8 (95% CI, 10.6 to 15.2) over the course of the study. Latinx individuals had the highest odds ratio of SARS-CoV-2 exposure at 6.56 (95% CI, 4.66 to 9.44). Our findings aid in quantifying the degree of asymmetric SARS-CoV-2 exposure by ethnoracial grouping. We also find that 49% of a subset of seropositive individuals had detectable neutralizing antibodies, which was skewed toward those with recent respiratory infection symptoms. IMPORTANCE PCR-confirmed SARS-CoV-2 cases underestimate true prevalence. Few robust community-level SARS-CoV-2 ethnoracial and overall prevalence estimates have been published for North Carolina in 2020. Mortality has been concentrated among ethnoracial minorities and may result from a high likelihood of SARS-CoV-2 exposure, which we observe was particularly high among Latinx individuals in North Carolina. Additionally, neutralizing antibody titers are a known correlate of protection. Our observation that development of SARS-CoV-2 neutralizing antibodies may be inconsistent and dependent on severity of symptoms makes vaccination a high priority despite prior exposure.

Multivariate Bayesian clustering using covariate-informed components with application to boreal vegetation sensitivity.

Climate change is impacting both the distribution and abundance of vegetation, especially in far northern latitudes. The effects of climate change are different for every plant assemblage and vary heterogeneously in both space and time. Small changes in climate could result in large vegetation responses in sensitive assemblages but weak responses in robust assemblages. But, patterns and mechanisms of sensitivity and robustness are not yet well understood, largely due to a lack of long-term measurements of climate and vegetation. Fortunately, observations are sometimes available across a broad spatial extent. We develop a novel statistical model for a multivariate response based on unknown cluster-specific effects and covariances, where cluster labels correspond to sensitivity and robustness. Our approach utilizes a prototype model for cluster membership that offers flexibility while enforcing smoothness in cluster probabilities across sites with similar characteristics. We demonstrate our approach with an application to vegetation abundance in Alaska, USA, in which we leverage the broad spatial extent of the study area as a proxy for unrecorded historical observations. In the context of the application, our approach yields interpretable site-level cluster labels associated with assemblage-level sensitivity and robustness without requiring strong a priori assumptions about the drivers of climate sensitivity.

Laboratory-Based SARS-CoV-2 Receptor Binding Domain Serologic Assays Perform with Equivalent Sensitivity and Specificity to Commercial FDA-EUA Approved Tests.

During early phases of the SARS-CoV-2 epidemic, many research laboratories repurposed their efforts towards developing diagnostic testing that could aid public health surveillance while commercial and public diagnostic laboratories developed capacity and validated large scale testing methods. Simultaneously, the rush to produce point-of-care and diagnostic facility testing resulted in FDA Emergency Use Authorization with scarce and poorly validated clinical samples. Here, we review serologic test results from 186 serum samples collected in early phases of the pandemic (May 2020) from skilled nursing facilities tested with six laboratory-based and two commercially available assays. Serum neutralization titers were used to set cut-off values using positive to negative ratio (P/N) analysis to account for batch effects. We found that laboratory-based receptor binding domain (RBD) binding assays had equivalent or superior sensitivity and specificity compared to commercially available tests. We also determined seroconversion rate and compared with qPCR outcomes. Our work suggests that research laboratory assays can contribute reliable surveillance information and should be considered important adjuncts to commercial laboratory testing facilities during early phases of disease outbreaks.

Durable Antibody Responses in Staff at Two Long-Term Care Facilities, during and Post SARS-CoV-2 Outbreaks.

SARS-CoV-2 has had a disproportionate impact on nonhospital health care settings, such as long-term-care facilities (LTCFs). The communal nature of these facilities, paired with the high-risk profile of residents, has resulted in thousands of infections and deaths and a high case fatality rate. To detect presymptomatic infections and identify infected workers, we performed weekly surveillance testing of staff at two LTCFs, which revealed a large outbreak at one of the sites. We collected serum from staff members throughout the study and evaluated it for binding and neutralization to measure seroprevalence, seroconversion, and type and functionality of antibodies. At the site with very few incident infections, we detected that over 40% of the staff had preexisting SARS-CoV-2 neutralizing antibodies, suggesting prior exposure. At the outbreak site, we saw rapid seroconversion following infection. Neutralizing antibody levels were stable for many weeks following infection, suggesting a durable, long-lived response. Receptor-binding domain antibodies and neutralizing antibodies were strongly correlated. The site with high seroprevalence among staff had two unique introductions of SARS-CoV-2 into the facility through seronegative infected staff during the period of study, but these did not result in workplace spread or outbreaks. Together, our results suggest that a high seroprevalence rate among staff can contribute to immunity within a workplace and protect against subsequent infection and spread within a facility. IMPORTANCE Long-term care facilities (LTCFs) have been disproportionately impacted by COVID-19 due to their communal nature and high-risk profile of residents. LTCF staff have the ability to introduce SARS-CoV-2 into the facility, where it can spread, causing outbreaks. We tested staff weekly at two LTCFs and collected blood throughout the study to measure SARS-CoV-2 antibodies. One site had a large outbreak and infected individuals rapidly generated antibodies after infection. At the other site, almost half the staff already had antibodies, suggesting prior infection. The majority of these antibodies bind to the receptor-binding domain of the SARS-CoV-2 spike protein and are potently neutralizing and stable for many months. The non-outbreak site had two unique introductions of SARS-CoV-2 into the facility, but these did not result in workplace spread or outbreaks. Our results reveal that high seroprevalence among staff can contribute to immunity and protect against subsequent infection and spread within a facility.

Disparities in SARS-CoV-2 seroprevalence among individuals presenting for care in central North Carolina over a six-month period.

Background: Robust community-level SARS-CoV-2 prevalence estimates have been difficult to obtain in the American South and outside of major metropolitan areas. Furthermore, though some previous studies have investigated the association of demographic factors such as race with SARS-CoV-2 exposure risk, fewer have correlated exposure risk to surrogates for socioeconomic status such as health insurance coverage. Methods: We used a highly specific serological assay utilizing the receptor binding domain of the SARS-CoV-2 spike-protein to identify SARS-CoV-2 antibodies in remnant blood samples collected by the University of North Carolina Health system. We estimated the prevalence of SARS-CoV-2 in this cohort with Bayesian regression, as well as the association of critical demographic factors with higher prevalence odds. Findings: Between April 21st and October 3rd of 2020, a total of 9,624 unique samples were collected from clinical sites in central NC and we observed a seroprevalence increase from 2·9 (1·7, 4·3) to 9·1 (7·2, 11·1) over the study period. Individuals who identified as Latinx were associated with the highest odds ratio of SARS-CoV-2 exposure at 7·77 overall (5·20, 12·10). Increased odds were also observed among Black individuals and individuals without public or private health insurance. Interpretation: Our data suggests that for this care-accessing cohort, SARS-CoV-2 seroprevalence was significantly higher than cumulative total cases reported for the study geographical area six months into the COVID-19 pandemic in North Carolina. The increased odds of seropositivity by ethnoracial grouping as well as health insurance highlights the urgent and ongoing need to address underlying health and social disparities in these populations.

Serial population-based serosurveys for COVID-19 in two neighbourhoods of Karachi, Pakistan.

OBJECTIVE: To determine population-based estimates of coronavirus disease 2019 (COVID-19) in a densely populated urban community of Karachi, Pakistan. METHODS: Three cross-sectional surveys were conducted in April, June and August 2020 in low- and high-transmission neighbourhoods. Participants were selected at random to provide blood for Elecsys immunoassay for detection of anti-severe acute respiratory syndrome coronavirus-2 antibodies. A Bayesian regression model was used to estimate seroprevalence after adjusting for the demographic characteristics of each district. RESULTS: In total, 3005 participants from 623 households were enrolled in this study. In Phase 2, adjusted seroprevalence was estimated as 8.7% [95% confidence interval (CI) 5.1-13.1] and 15.1% (95% CI 9.4-21.7) in low- and high-transmission areas, respectively, compared with 0.2% (95% CI 0-0.7) and 0.4% (95% CI 0-1.3) in Phase 1. In Phase 3, it was 12.8% (95% CI 8.3-17.7) and 21.5% (95% CI 15.6-28) in low- and high-transmission areas, respectively. The conditional risk of infection was 0.31 (95% CI 0.16-0.47) and 0.41 (95% CI 0.28-0.52) in low- and high-transmission neighbourhoods, respectively, in Phase 2. Similar trends were observed in Phase 3. Only 5.4% of participants who tested positive for COVID-19 were symptomatic. The infection fatality rate was 1.66%, 0.37% and 0.26% in Phases 1, 2 and 3, respectively. CONCLUSION: Continuing rounds of seroprevalence studies will help to improve understanding of secular trends and the extent of infection during the course of the pandemic.

Modeling Bronchiolitis Incidence Proportions in the Presence of Spatio-Temporal Uncertainty.

Bronchiolitis (inflammation of the lower respiratory tract) in infants is primarily due to viral infection and is the single most common cause of infant hospitalization in the United States. To increase epidemiological understanding of bronchiolitis (and, subsequently, develop better prevention strategies), this research analyzes data on infant bronchiolitis cases from the U.S. Military Health System between the years 2003-2013 in Norfolk, Virginia, USA. For privacy reasons, child home addresses, birth dates, and diagnosis dates were randomized (jittered) creating spatio-temporal uncertainty in the geographic location and timing of bronchiolitis incidents. Using spatio-temporal point patterns, we created a modeling strategy that accounts for the jittering to estimate and quantify the uncertainty for the incidence proportion (IP) of bronchiolitis. Additionally, we regress the IP onto key covariates including pollution where we adequately account for uncertainty in the pollution levels (i.e., covariate uncertainty) using a land use regression model. Our analysis results indicate that the IP is positively associated with sulfur dioxide and population density. Further, we demonstrate how scientific conclusions may change if various sources of uncertainty (either spatio-temporal or covariate uncertainty) are not accounted for. Code submitted with this article was checked by an Associate Editor for Reproducibility and is available as an online supplement.

Estimating seasonal onsets and peaks of bronchiolitis with spatially and temporally uncertain data.

RSV bronchiolitis (an acute lower respiratory tract viral infection in infants) is the most common cause of infant hospitalizations in the United States (US). The only preventive intervention currently available is monthly injections of immunoprophylaxis. However, this treatment is expensive and needs to be administered simultaneously with seasonal bronchiolitis cycles in order to be effective. To increase our understanding of bronchiolitis timing, this research focuses on identifying seasonal bronchiolitis cycles (start times, peaks, and declinations) throughout the continental US using data on infant bronchiolitis cases from the US Military Health System Data Repository. Because this data involved highly personal information, the bronchiolitis dates in the dataset were "jittered" in the sense that the recorded dates were randomized within a time window of the true date. Hence, we develop a statistical change point model that estimates spatially varying seasonal bronchiolitis cycles while accounting for the purposefully introduced jittering in the data. Additionally, by including temperature and humidity data as regressors, we identify a relationship between bronchiolitis seasonality and climate. We found that, in general, bronchiolitis seasons begin earlier and are longer in the southeastern states compared to the western states with peak times lasting approximately 1 month nationwide.