Improved prediction of new COVID-19 cases using a simple vector autoregressive model: Evidence from seven New York State counties

With the rapid spread of COVID-19, there is an urgent need for a framework to accurately predict COVID-19 transmission. Recent epidemiological studies have found that a prominent feature of COVID-19 is its ability to be transmitted before symptoms occur, which is generally not the case for seasonal influenza and SARS. Several COVID-19 predictive epidemiological models have been proposed; however, they share a common drawback-they are unable to capture the unique asymptomatic nature of COVID-19 transmission. Here, we propose vector autoregression (VAR) as an epidemiological county-level prediction model that captures this unique aspect of COVID-19 transmission by introducing newly infected cases in other counties as lagged explanatory variables. Using the number of new COVID-19 cases in seven New York State counties, we predicted new COVID-19 cases in the counties over the next 4 weeks. We then compared our prediction results with those of 11 other state-of-the-art prediction models proposed by leading research institutes and academic groups. The results showed that VAR prediction is superior to other epidemiological prediction models in terms of the root mean square error of prediction. Thus, we strongly recommend the simple VAR model as a framework to accurately predict COVID-19 transmission.

Prevalence of Anticoagulant Rodenticide Exposure in Red-tailed Hawks (Buteo jamaicensis) in New York State and Diagnostic Utility of Russell Viper Venom Test for Detecting Associated Coagulopathies

Anticoagulant rodenticides (ARs) continue to be used across the United States as a method for controlling unwanted rodent species. As a consequence, wild birds of prey are exposed to toxins by eating poisoned prey items. ARs prevent the hepatic recycling of vitamin K and thereby impede the post-translational processing of coagulation factors II, VII, IX, and X that is required for procoagulant complex assembly. Through this mechanism of action, ARs cause hemorrhage and death in their target species. Various studies have documented the persistence of these contaminants in birds of prey but few have attempted to use affordable and accessible diagnostic tests to diagnose exposure in free-ranging birds of prey. In our study free-ranging red-tailed hawks were found to be exposed to difethialone and brodifacoum. Eleven of sixteen (68%) livers tested for AR exposure were positive. Difethialone was found in 1/16 (6%) liver samples, and brodifacoum was detected in 15/16 (93%) liver samples. Difethialone was found at a concentration of 0.18 ppm and brodifacoum concentrations ranged from 0.003-0.234 ppm. Two out of 34 (6%) RTH assessed for blood rodenticide had brodifacoum in blood with measured concentrations of 0.003 and 0.006 ppm. The range of clotting times in the prothrombin time (PT) and Russell’s viper venom time assays for control RTH were 16.7 to 39.7 seconds and 11.5 to 91.8 seconds, respectively. No correlation was found between PT and RVVT in the control or free-range RTH, and there was no relationship found between the presence of liver anticoagulant residues and clotting times in the PT and RVVT.

Students as Community Vaccinators: Implementation of a Service-Learning COVID-19 Vaccination Program

Background: Service-learning is an integral component of medical education. While the COVID-19 pandemic has caused massive educational disruptions, it has also catalyzed innovation in service-learning as real-time responses to pandemic-related problems. For example, the limited number of qualified providers was a potential barrier to local and national SARS-CoV-2 vaccination efforts. Foreseeing this hurdle, New York State temporarily allowed healthcare professional trainees to vaccinate, enabling medical students to support an overwhelmed healthcare system and contribute to the community. Yet, it was the responsibility of medical schools to interpret these rules and implement the vaccination programs. Here the authors describe a service-learning vaccination program directed towards underserved communities. Methods: Weill Cornell Medicine (WCM) rapidly developed a faculty-led curriculum to prepare students to communicate with patients about the COVID-19 vaccines and to administer intramuscular injections. Qualified students were deployed to public vaccination clinics located in underserved neighborhoods across New York City in collaboration with an established community partner. The educational value of the program was evaluated with retrospective survey. Results: Throughout the program, which lasted from February to June 2021, 128 WCM students worked at 103 local events, helping to administer 26,889 vaccine doses. Analysis of student evaluations revealed this program taught fundamental clinical skills, increasing comfort giving intramuscular injection from 2% to 100% and increasing comfort talking to patients about the COVID-19 vaccine from 30% to 100%. Qualitatively participants described the program as a transformative service-learning experience. Conclusion: As new virus variants emerge, nations battle recurrent waves of infection, and vaccine eligibility expands to include children and boosters, the need for effective vaccination plans continues to grow. The program described here offers a novel framework that academic medical centers could adapt to increase vaccine access in their local community and provide students with a uniquely meaningful educational experience.

Rural-Urban Differences in Diagnosed Cervical Artery Dissection in New York State

<b><i>Background:</i></b> Cervical artery dissection (CeAD) is a leading cause of stroke in young adults. Incidence estimates may be limited by under- or overdiagnosis. <b><i>Objective:</i></b> We aimed to investigate if CeAD diagnosis would be higher in urban centers compared to rural regions of New York State (NYS). <b><i>Methods:</i></b> For this ecological study, administrative codes were used to identify CeAD discharges in the NYS Statewide Planning and Research Cooperative System (SPARCS) from 2009 to 2014. Rural Urban Commuting Area (RUCA) codes were taken from the US Department of Agriculture and included the classifications metropolitan, micropolitan, small town, and rural. Negative binomial models were used to calculate effect estimates and 95% confidence limits (e^β; 95% CL) for the association between RUCA classification and the number of dissections per ZIP code. Models were further adjusted by population. <b><i>Results:</i></b> Population information was obtained from the US Census Bureau on 1,797 NYS ZIP codes (70.7% of NYS ZIP codes), 826 of which had at least 1 CeAD-related discharge from 2009 to 2014. Nonrural ZIP codes were more likely to report more CeAD cases relative to rural areas even after adjusting for population (metropolitan effect = e^β 5.00; 95% CI: 3.75–6.66; micropolitan effect 3.02; 95% CI: 2.16–4.23; small town effect 2.34; 95% CI: 1.58–3.47). <b><i>Conclusions:</i></b> CeAD diagnosis correlates with population density as defined by rural-urban status. Our results could be due to underdiagnosis in rural areas or overdiagnosis with increasing urbanicity.

Sex disparities in COVID-19 mortality vary considerably across time: The case of New York State

In order to characterize how sex disparities in COVID-19 mortality evolved over time in New York State (NY), we analyzed sex-disaggregated data from the US Gender/Sex COVID-19 Data Tracker from March 14, 2020 to August 28, 2021. We defined six different time periods and calculated mortality rates by sex and mortality rate ratios, both cumulatively and for each time period separately. As of August 28, 2021, 19 227 (44.2%) women and 24 295 (55.8%) men died from COVID-19 in NY. 72.7% of the cumulative difference in the number of COVID-19 deaths between women and men was accrued between March 14 and May 4, 2020. During this period, the COVID-19 mortality rate ratio for men compared to women was 1.56 (95% CI: 1.52-1.61). In the five subsequent time periods, the corresponding ratio ranged between 1.08 (0.98-1.18) and 1.24 (1.15-1.34). While the cumulative mortality rate ratio of men compared to women was 1.34 (1.31-1.37), the ratio equals 1.19 (1.16-1.22) if deaths during the initial COVID-19 surge are excluded from the analysis. This article shows that in NY the magnitude of sex disparities in COVID-19 mortality was not stable across time. While the initial surge in COVID-19 mortality was characterized by stark sex disparities, these were greatly attenuated after the introduction of public health controls.

Implementation of a statewide Advocacy Week: The New York State experience

Disclaimer In an effort to expedite the publication of articles, AJHP is posting manuscripts online as soon as possible after acceptance. Accepted manuscripts have been peer-reviewed and copyedited, but are posted online before technical formatting and author proofing. These manuscripts are not the final version of record and will be replaced with the final article (formatted per AJHP style and proofed by the authors) at a later time.