The timing of the large SARS-Cov-2 outbreaks in the Northern hemisphere, and the temporary delay in infections observed in the Southern hemisphere during the first half of the year has prompted questions related to potential seasonality of SARS-Cov-2. This study is designed to evaluate patterns of known viral respiratory diseases in different regions of the US to provide a benchmark for comparison with SARS-CoV-2 infections. Retrospective observational study using data from Optum PanTher. All patients admitted with COVID-19 diagnoses (starting April) or COVID-19-related severe diagnoses (acute severe respiratory distress - January to April) till end of June were identified. A time-series of hospitalization by region was established. For comparative purposes, time-series of hospital admissions for the flu, viral pneumonia and other respiratory diseases (ORD) for the same time period in 2018 and 2019 were analyzed. Descriptive analyses were used to compare the time series. A total of 70,829 patients with COVID-19 were compared to 46,056 flu, 168,185 pneumonia and and 680,888 ORD patients. Peak case volumes for ORD, flu and pneumonia were observed from January to March, with lowest case counts in July. From peak to lowest prevalence, there was 30-32% decrease in prevalence for ORD, a 41%-52% for pneumonia and a 98-99% for flu. Trends were most marked in the Midwest and least noticeable in the South. The significant increase of SARS-CoV-2 between January and April, with peak in April and slight decline in May, did not match timing of increased prevalence of the other known respiratory viral diseases. Available data to date did not suggest a seasonality for COVID-19 similar to that of other viral respiratory diseases. A limitation of our study is that the COVID-19 decline in some US states was also potentially affected by drastic government measures not present in prior years for flu, pneumonia or ORD.

A new diagnostic code was released in April 2020 for COVID-19 infection. Prior to that, database research used CDC-listed symptoms to identify COVID-19 cases. This study evaluates whether use of these symptoms can accurately identify patients with COVID-19, and at what time an increase in mortality was first observed using combinations of these symptoms. Patients from the Premier Healthcare Database with an inpatient event during which patients deceased, with COVID-19 related symptoms and a DRG (diagnostic related group) of respiratory infection or sepsis and, from April 2020 onwards, a COVID-19 diagnosis, were identified, from September 2019 to most recent. Comorbidities and symptoms of COVID-19 and related treatments were used as model variables. Patients with and without COVID-19 diagnosis were matched using propensity score matching (model: logit, method: nearest neighbor, caliper: 0.1). Time of discharge was analyzed to evaluate changes in mortality in the matched cohorts. Pre-match, 3,923 patients without (noDiag) and 4,903 patients with COVID-specific diagnosis codes (withDiag) were identified. In the noDiag group, there were 52.7% males and 11.8% African Americans (AA). In the withDiag group, there were 56.4% males and 21.1% AA. In the noDiag cohort, 71.2% patients had a DRG code of septicemia vs 52.9% in the withDiag group. After matching, there were 1,340 patients in each group, 60% males, 21.7% AA. In the NoDiag matched cohort, 59% death occurred in 2020, with a clear spike in March (30%). The remaining 41% were identified prior to 2020, as early as October 2020, suggesting potential contamination with mortality due to other causes. Due to the similarity of COVID-symptoms to those observed with other common respiratory diseases, and concurrent timing of COVID mortality with that of viral pneumonia, matching on symptoms and treatments identified patients prior to first US-reported case and therefore included patients with infections other than COVID-19.

The SARS-CoV-2 pandemic created significant stress on healthcare systems worldwide, and the impact thereof on patients' care is unknown. This study was designed to evaluate changes in healthcare delivery in the US during the COVID-19 era and evaluate whether patients that required surgical care during that time were at greater risk for mortality compared to similar patients treated in years prior. Patients from the Premier Healthcare Database undergoing surgical procedures from March 1st to May 31st, 2020 (covid_era) and March 1st to May 31st, 2019 (pre-covid_era) were identified and categorized by surgery type. Patients were characterized by demographic, comorbidities, and whether they had a diagnosis for COVID-19 (post April 1st). Length of hospital stay (LOS) and mortality rate for each surgical category were evaluated. Generalized linear models (GLM) were used to test for statistical differences in mortality (family: binomial, link: log), adjusting for demographics, comorbidities and presence of COVID-19 diagnosis. Twenty-four different surgical categories based on major anatomy were identified, such as cardiac and brain surgery. Overall, average LOS decreased from the pre_covid_era to the covid_era from 4.62 days (standard deviation (SD): 5.01) to 4.59 days (SD: 4.51, p = 0.013). Patients treated in the inpatient setting during the covid_era had more comorbidities than during the pre_covid_era (average Elixhauser score: 3.20 vs 3.00, p=0.000). Mortality between both eras in patients was comparable, except for brain surgery (OR: 1.29, 95%CI: 1.03-1.62) and gastrointestinal surgery (OR: 1.15, 95% CI: 1.02-1.30). Patients admitted for surgical procedures during the COVID_era were sicker and had shorter LOS than in prior years. However, adjusted mortality rates did not change for 22 of 24 surgical categories and only showed slight increases for 2 out of 24 categories. These modest changes might be due to the presence of underdiagnosed COVID-19 patients in surgical cohorts during the COVID-19 era.

Published mortality rates from SARS-CoV-2 infections have varied significantly due to differences in testing and disease tracking rates across countries. The impact of population density, family size, mobility and government actions on mortality due to SARS-CoV-2 infections is not well understood. This is a retrospective data analysis using the Worldwide WHO situation reports along with population density information and family size information (from Worldometer.com), government response data (7 policies: school closure (SC), workplace closure (WC), public event cancellation (PEC), restriction on gatherings (RG), public transport closure (PTC), stay-at-home (SAH), internal movement restrictions (IMR) - Oxford University) and mobility reports (Apple) to estimate variables associated with COVID-19 mortality from countries with complete information. Poisson regression models were developed to evaluate associations between mortality counts and variables mentioned herein. A 42-day lag was applied to mobility and government response metrics to evaluate the impact thereof on mortality. Nine countries were included in the analysis (Australia, Belgium, Chile, Italy, Peru, Saudi Arabia, Spain, UK, USA). The highest quartile of government response index had the greatest protective effect on mortality (incidence proportion ratio (IPR): 0.05 (95% confidence interval (CI): 0.01-0.32). There was no consistent association impact between walking, driving or transit mobility metric. Government actions with significant protective effect were PTC (IPR: 0.46 (0.28-0.75)) and PEC (IPR: 0.65 (0.47-0.91)). All other government actions did not show a statistically significant association with mortality. Our study identified population metrics and government actions potentially associated with mortality, across multiple geographies. Further research on key confounders (such as mask wearing) is required to evaluate further actions to mitigate the mortality from COVID-19.

Patients with severe SARS-CoV-2 infection require significant hospital resources. However, since the disease was first identified, treatments have evolved rapidly, reducing fatality rates. This study was designed to evaluate changes in healthcare utilization and cost in patients with in-hospital mortality, from earliest time of disease identification to May 2020. Patients from the Premier Healthcare Database with an inpatient event during which patients deceased, with COVID-19 related symptoms and a DRG (diagnostic related group) of respiratory infection or sepsis and, from April 2020 onwards, a COVID-19 diagnosis, were identified, from September 2019 to most recent. Comorbidities and symptoms of COVID-19 and related treatments were used as model variables. Patients with and without COVID-19 diagnosis were matched using propensity score matching (model: logit, method: nearest neighbor, caliper: 0.1). Cost of hospital admission was analyzed using a generalized linear model (family: Gamma, link: log). Least mean square methods were used to evaluate changes in cost over time. Hospital costs averaged $24,899 (95%CI: 21,980-28,205). Length of stay (LOS) increased non-significantly from 8.7 days (95%CI: 8.0-9.5) in December to 10.6 days (95%CI: 9.28-12.18) in May 2020. Comorbidities did not significantly affect patient costs and were therefore not predictors for higher costs. Age category (> 84) was associated with lower cost (mean: 17,350 (95%CI: 15,706-19,166) compared to all other age groups. Hospital costs showed a non-significant trend of decline from December 2019 (mean: $24,860, 95%CI: $21,659-$28,534) to May 2020 (mean: $23,328, 95%CI: $18,269-$29,787). The COVID-19 epidemic has prompted unprecedented medical research and fast-paced learnings. Our study focused on inpatient fatalities, as these represent the potentially most severe and complex cases. In this population, patient comorbidities did not affect cost. Older age was associated with lower cost. Trends of cost decline and LOS increase were observed over time, suggesting more efficient care and potentially longer survival.

Many governments have adopted strict social distancing and stay-at-home orders to slow the spread of the Coronavirus-19 (COVID-19) in early 2020. The resulting growth in new infection cases has varied between geographies. This study was designed to measure the association between government action, mobility responses and new daily reported infection cases. This is a retrospective data analysis using the Worldwide WHO situation reports along with population density information and family size information (from Worldometer.com), government response data (7 policies: school closure (SC), workplace closure (WC), public event cancellation (PEC), restriction on gatherings (RG), public transport closure (PTC), stay-at-home (SAH), internal movement restrictions (IMR) - Oxford University) and mobility reports (Apple). Poisson regression models were developed to evaluate associations between new reported daily cases of COVID-19 and all other available variables, for a list of 9 countries that had more than 50K tests per million people. A 42-day lag was applied to mobility and government response variables to evaluate the impact thereof on new COVID-19 cases. Nine countries were included in the analysis (Australia, Belgium, Chile, Italy, Peru, Saudi Arabia, Spain, UK, USA). From all the government measures, highest association was observed for RG (Incidence proportion ratio (IPR): 0.30 (95%CI: 0.19-0.47), followed by PTC (IPR: 0.38 (95%CI: 0.31-0.45)) and WC (IPR: 0.58 (95%CI: 0.46-0.72)). Reduction in mobility due to transit was only slightly protective, and only for the top quartile of transit mobility reduction (IPR: 0.91 (95%CI: 0.84-98)). Walking and driving mobility restriction were not associated with new cases. Unprecedented government responses have been deployed to try and contain the COVID-19 pandemic. Whereas the actual effectiveness of each measure is unknown, this study highlights key metrics that might have proven useful in containing the COVID-19 pandemic.

As COVID-19 cases increased in early 2020, most governments imposed social distancing and mobility restrictions. This study was designed to determine associations between government actions and new cases of fatalities, and the differences between the effectiveness of these actions across five different countries. The World Health Organization daily situation reports (Feb 1 - June 30, 2020) were used to obtain counts of new fatalities, for 5 countries with testing rate > 50K per million inhabitants and with complete social distancing data (Australia, Belgium, Italy, United Kingdom, United States). Open-source mobility data from Apple and government action data (7 policies: school closure (SC), workplace closure (WC), public event cancellation (PEC), restriction on gatherings (RG), public transport closure (PTC), stay-at-home (SAH), internal movement restrictions (IMR)) from Oxford University were used as explanatory variables. Poisson regression models were built to evaluate the association between new cases and new fatalities, mobility, and government actions. A 6-week lag time was incorporated to evaluate the impact of actions taken 42-day prior on new cases of fatalities. Across countries, there was little consistency in any of the associations. In Australia, the UK and the US, none of the government actions or mobility metrics were associated with new fatalities. In Belgium, WC and PEC were protective vs new fatalities (Incidence proportion ratio (IPR) for WC = 0.66 95%CI: 0.51-0.84; IPR for PEC = 0.76, 95% CI = 0.65-0.89). In Italy, PTC and SC were associated with lowering counts of fatalities (IPR for PTC =0.59, 95% CI = 0.48-0.75, IPR for SC = 0.81, 95%CI: 0.68-0.97). Unprecedented government actions have been instituted to maintain the spread of COVID-19. We did not identify associations consistent across multiple countries between government measures and new cases of fatalities within 6 weeks of measure implementation.