Associations Between Reported Post-COVID-19 Symptoms and Subjective Well-Being, Israel, July 2021 -April 2022 (preprint)

The impact of individual symptoms reported post-COVID-19 on subjective well-being (SWB) is unknown. We described associations between SWB and selected reported symptoms following SARS-CoV-2 infection. We analysed reported symptoms and subjective well being from 2295 participants (of which 576 reporting previous infection) in an ongoing longitudinal cohort study taking place in Israel. We estimated changes in SWB associated with reported selected symptoms at three follow-up time points (3-6, 6-12, and 12-18 months post infection) among participants reporting previous SARS-CoV-2 infection, adjusted for key demographic variables, using linear regression. Our results suggest that the biggest and most sustained changes in SWB stems from non-specific symptoms (fatigue -7.7 percentage points (pp), confusion/ lack of concentration -10.7 pp, and sleep disorders -11.5pp, p<0.005), whereas the effect of system-specific symptoms, such as musculoskeletal symptoms (weakness in muscles and muscle pain) on SWB, are less profound and more transient. Taking a similar approach for other symptoms and following individuals over time to describe trends in SWB changes attributable to specific symptoms will help understand the post-acute phase of COVID-19 and how it should be defined and better managed.

Reasons underlying the intention to vaccinate children aged 5-11 against COVID-19: A cross-sectional study of parents in Israel, November 2021 (preprint)

BackgroundVaccination is a key tool to mitigate the impact of the COVID-19 pandemic. In Israel, COVID-19 vaccines became available to adults in December 2020 and to 5-11-year-old children in November 2021. Ahead of the vaccine roll-out in children, we aimed to determine whether parents intended to vaccinate their children and describe reasons for their intentions. MethodsWe recruited parents on social media and collected information on parental socio-demographic characteristics, COVID-19 vaccine history, intention to vaccinate their children against COVID-19, and reasons for parental decisions, using an anonymous online survey. We identified associations between parental characteristics and intention to vaccinate children using a logistic regression model and described reasons for intentions to vaccinate or not using proportions together with 95% confidence intervals (CI). Results1837 parents participated. Parental non-vaccination and having experienced major vaccination side effects were strongly associated with non-intention to vaccinate their children (OR 0.09 and 0.18 respectively, p<0.001). Compared with others, parents who were younger, lived in the socio-economically deprived periphery, and belonged to the Arab population had lower intentions to vaccinate their children. Commonly stated reasons for non-intention to vaccinate included vaccine safety and efficacy (53%, 95%CI 50-56) and the belief that COVID-19 was a mild disease (73%, 95%CI 73-79). The most frequently mentioned motives for intending to vaccine children was returning to normal social and educational life (89%, 95%CI 87-91). ConclusionParental socio-demographic background and their own vaccination experience was associated with intention to vaccinate their children aged 5-11. Intention to vaccinate was mainly for social and economic reasons rather than health, whereas non-intention to vaccinate mainly stemmed from health concerns. Understanding rationales for COVID-19 vaccine rejection or acceptance, as well as parental demographic data, can pave the way for intentional educational campaigns to encourage not only vaccination against COVID-19, but also regular childhood vaccine programming. HighlightsO_LIParental intention to vaccinate 5-11 children is much lower than vaccine coverage in parental age groups C_LIO_LIBeing unvaccinated and having experienced side effects following vaccination were the greatest negative predictors in parents of intention to vaccinate their children C_LIO_LIParents were more likely to accept a COVID-19 vaccine for their children to allow them to return to daily social life and to ensure economic security in the family C_LIO_LIParents were more likely to reject a COVID-19 vaccination for health reasons such as safety concerns or due the belief that COVID-19 was a mild disease in children C_LI

Pre-infection 25-hydroxyvitamin D3 levels and association with severity of COVID-19 illness (preprint)

Objective: Studies have demonstrated a potential link between low vitamin D levels and both an increased risk of infection with SARS-CoV-2 and poorer clinical outcomes but have not established temporality. This retrospective study examined if, and to what degree, a relationship exists between pre-infection serum vitamin D levels and disease severity and mortality of SARS-CoV-19. Design and patients: The records of individuals admitted between April 7th, 2020 and February 4th, 2021 to the Galilee Medical Center (GMC) in Nahariya, Israel with positive polymerase chain reaction (PCR) tests for SARS-CoV-2 were searched for vitamin D (VitD) levels measured 14 to 730 days prior to the positive PCR test. Measurements: Patients admitted to GMC with COVID-19 were categorized according to disease severity and VitD level. Association between pre-infection VitD levels and COVID-19 severity was ascertained utilizing a multivariate regression analysis. Results: Of 1176 patients admitted, 253 had VitD levels prior to COVID-19 infection. Compared with mildly or moderately diseased patients, those with severe or critical COVID-19 disease were more likely to have pre-infection vitamin D deficiency of less than 20 ng/mL (OR=14.30, 95%, 4.01-50.9; p < .001); be older (OR=1.039 for each year, 95% CI for OR, 1.017-1.061; p< .01), and have diabetes (OR=2.031, 95% CI for OR, 1.04-3.36; p= 0.038). Vitamin D deficiency was associated with higher rates of mortality (p<0.001) and comorbidities including COPD (p=0.006), diabetes (p=0.026), and hypertension (p =0.016). Conclusions: Among hospitalized COVID-19 patients, pre-infection deficiency of vitamin D was associated with increased disease severity and mortality.

Point of Care Image Analysis for COVID-19 (preprint)

Early detection of COVID-19 is key in containing the pandemic. Disease detection and evaluation based on imaging is fast and cheap and therefore plays an important role in COVID-19 handling. COVID-19 is easier to detect in chest CT, however, it is expensive, non-portable, and difficult to disinfect, making it unfit as a point-of-care (POC) modality. On the other hand, chest X-ray (CXR) and lung ultrasound (LUS) are widely used, yet, COVID-19 findings in these modalities are not always very clear. Here we train deep neural networks to significantly enhance the capability to detect, grade and monitor COVID-19 patients using CXRs and LUS. Collaborating with several hospitals in Israel we collect a large dataset of CXRs and use this dataset to train a neural network obtaining above 90% detection rate for COVID-19. In addition, in collaboration with ULTRa (Ultrasound Laboratory Trento, Italy) and hospitals in Italy we obtained POC ultrasound data with annotations of the severity of disease and trained a deep network for automatic severity grading.

COVID-19 Classification of X-ray Images Using Deep Neural Networks (preprint)

In the midst of the coronavirus disease 2019 (COVID-19) outbreak, chest X-ray (CXR) imaging is playing an important role in the diagnosis and monitoring of patients with COVID-19. Machine learning solutions have been shown to be useful for X-ray analysis and classification in a range of medical contexts. The purpose of this study is to create and evaluate a machine learning model for diagnosis of COVID-19, and to provide a tool for searching for similar patients according to their X-ray scans. In this retrospective study, a classifier was built using a pre-trained deep learning model (ReNet50) and enhanced by data augmentation and lung segmentation to detect COVID-19 in frontal CXR images collected between January 2018 and July 2020 in four hospitals in Israel. A nearest-neighbors algorithm was implemented based on the network results that identifies the images most similar to a given image. The model was evaluated using accuracy, sensitivity, area under the curve (AUC) of receiver operating characteristic (ROC) curve and of the precision-recall (P-R) curve. The dataset sourced for this study includes 2362 CXRs, balanced for positive and negative COVID-19, from 1384 patients (63 +/- 18 years, 552 men). Our model achieved 89.7% (314/350) accuracy and 87.1% (156/179) sensitivity in classification of COVID-19 on a test dataset comprising 15% (350 of 2326) of the original data, with AUC of ROC 0.95 and AUC of the P-R curve 0.94. For each image we retrieve images with the most similar DNN-based image embeddings; these can be used to compare with previous cases.

Efficient culture of SARS-CoV-2 in human hepatoma cells enhances viability of the virus in human lung cancer cell lines permitting the screening of antiviral compounds (preprint)

Efforts to mitigate COVID-19 include screening of existing antiviral molecules that could be re-purposed to treat SARS-CoV-2 infections. Although SARS-CoV-2 propagates efficiently in African green monkey kidney (Vero) cells, antivirals such as nucleos(t)ide analogs (nucs) often exhibit decreased activity in these cells due to inefficient metabolization. Limited SARS-CoV-2 replication and propagation occurs in human cells, which are the most relevant testing platforms. By performing serial passages of a SARS-CoV-2 isolate in the human hepatoma cell line clone Huh7.5, we selected viral populations with improved viability in human cells. Culture adaptation led to the emergence of a significant number of high frequency changes (>90% of the viral population) in the region coding for the spike glycoprotein, including a deletion of nine amino acids in the N-terminal domain and 3 amino acid changes (E484D, P812R, and Q954H). We demonstrated that the Huh7.5-adapted virus exhibited a >3-Log10 increase in infectivity titers (TCID50) in Huh7.5 cells, with titers of ~8 Log10TCID50/mL, and >2-Log10 increase in the human lung cancer cell line Calu-1, with titers of ~6 Log10TCID50/mL. Culture adaptation in Huh7.5 cells further permitted efficient infection of the otherwise SARS-CoV-2 refractory human lung cancer cell line A549, with titers of ~6 Log10TCID50/mL. The enhanced ability of the virus to replicate and propagate in human cells permitted screening of a panel of nine nucs, including broad-spectrum compounds. Remdesivir, EIDD-2801 and to a limited extent galidesivir showed antiviral effect across these human cell lines, whereas sofosbuvir, uprifosbuvir, valopicitabine, mericitabine, ribavirin, and favipiravir had no apparent activity. Importance: The cell culture adapted variant of the SARS-CoV-2 virus obtained in the present study, showed significantly enhanced replication and propagation in various human cell lines, including lung derived cells otherwise refractory for infection with the original virus. This SARS-CoV-2 variant will be a valuable tool permitting investigations across human cell types, and studies of identified mutations could contribute to our understanding of viral pathogenesis. In particular, the adapted virus can be a good model for investigations of viral entry and cell tropism for SARS-CoV-2, in which the spike glycoprotein plays a central role. Further, as shown here with the use of remdesivir and EIDD-2801, two nucs with significant inhibitory effect against SARS-CoV-2, large differences in the antiviral activity are observed depending on the cell line. Thus, it is essential to select the most relevant target cells for pre-clinical screenings of antiviral compounds, facilitated by using a virus with broader tropism.

COVID-19 Classification of X-ray Images Using Deep Neural Networks (preprint)

In the midst of the coronavirus disease 2019 (COVID-19) outbreak, chest X-ray (CXR) imaging is playing an important role in the diagnosis and monitoring of patients with COVID-19. Machine learning solutions have been shown to be useful for X-ray analysis and classification in a range of medical contexts. The purpose of this study is to create and evaluate a machine learning model for diagnosis of COVID-19, and to provide a tool for searching for similar patients according to their X-ray scans. In this retrospective study, a classifier was built using a pre-trained deep learning model (ReNet50) and enhanced by data augmentation and lung segmentation to detect COVID-19 in frontal CXR images collected between January 2018 and July 2020 in four hospitals in Israel. A nearest-neighbors algorithm was implemented based on the network results that identifies the images most similar to a given image. The model was evaluated using accuracy, sensitivity, area under the curve (AUC) of receiver operating characteristic (ROC) curve and of the precision-recall (P-R) curve. The dataset sourced for this study includes 2362 CXRs, balanced for positive and negative COVID-19, from 1384 patients (63 +/- 18 years, 552 men). Our model achieved 89.7% (314/350) accuracy and 87.1% (156/179) sensitivity in classification of COVID-19 on a test dataset comprising 15% (350 of 2326) of the original data, with AUC of ROC 0.95 and AUC of the P-R curve 0.94. For each image we retrieve images with the most similar DNN-based image embeddings; these can be used to compare with previous cases.

A lesson from COVID-19 on inaccessibility of web-based information for disabled populations worldwide (preprint)

Many government websites and mobile content are inaccessible for people with vision, hearing, and cognitive disabilities. The COVID-19 pandemic highlighted these disparities when health authority website information, critical in providing resources for curbing the spread of the virus, remained inaccessible for disabled populations. The Web Content Accessibility Guidelines provide comparatively universally accepted guidelines for website accessibility. We utilized these parameters to examine the number of countries with or without accessible health authority websites. The resulting data indicate a dearth of countries with websites accessible for persons with disabilities. Methods of information dissemination must take into consideration individuals with disabilities, particularly in times of global health crises.

Vaccine Hesitancy: The Next Challenge in the Fight Against COVID-19 (preprint)

Vaccine hesitancy remains a barrier to full population inoculation against highly infectious diseases. With rapid developments in a potential COVID-19 vaccine by scientists across the globe, public concerns over the safety and side effects of such a vaccine may contribute to vaccine hesitancy. We analyzed anonymous questionnaire answers regarding acceptance of a potential COVID-19 vaccine posed to healthcare workers and the general population throughout Israel with a total respondent count of 1941. Our results demonstrate higher rates of COVID-19 vaccine hesitancy among various groups: parents, nurses, and medical workers not caring for SARS-CoV-2 positive patients. Healthcare staff involved in the care of COVID-19 positive patients and individuals who consider themselves at higher risk of disease were more likely to self-report acquiescence to obtain a COVID-19 vaccine if and when it becomes available. Interventional educational campaigns targeted towards populations at risk of vaccine hesitancy, therefore, are urgently needed to combat misinformation and resultant low inoculation rates.