Prediction of in-hospital mortality rate in COVID-19 patients with diabetes mellitus using machine learning methods.

Background: Since its emergence in December 2019, until June 2022, coronavirus 2019 (COVID-19) has impacted populations all around the globe with it having been contracted by ~ 535 M people and leaving ~ 6.31 M dead. This makes identifying and predicating COVID-19 an important healthcare priority. Method and Material: The dataset used in this study was obtained from Shahid Beheshti University of Medical Sciences in Tehran, and includes the information of 29,817 COVID-19 patients who were hospitalized between October 8, 2019 and March 8, 2021. As diabetes has been shown to be a significant factor for poor outcome, we have focused on COVID-19 patients with diabetes, leaving us with 2824 records. Results: The data has been analyzed using a decision tree algorithm and several association rules were mined. Said decision tree was also used in order to predict the release status of patients. We have used accuracy (87.07%), sensitivity (88%), and specificity (80%) as assessment metrics for our model. Conclusion: Initially, this study provided information about the percentages of admitted Covid-19 patients with various underlying disease. It was observed that diabetic patients were the largest population at risk. As such, based on the rules derived from our dataset, we found that age category (51-80), CPR and ICU residency play a pivotal role in the discharge status of diabetic inpatients.

Personalized predictions of adverse side effects of the COVID-19 vaccines.

Background: Misconceptions about adverse side effects are thought to influence public acceptance of the Coronavirus disease 2019 (COVID-19) vaccines negatively. To address such perceived disadvantages of vaccines, a novel machine learning (ML) approach was designed to generate personalized predictions of the most common adverse side effects following injection of six different COVID-19 vaccines based on personal and health-related characteristics. Methods: Prospective data of adverse side effects following COVID-19 vaccination in 19943 participants from Iran and Switzerland was utilized. Six vaccines were studied: The AZD1222, Sputnik V, BBIBP-CorV, COVAXIN, BNT162b2, and the mRNA-1273 vaccine. The eight side effects were considered as the model output: fever, fatigue, headache, nausea, chills, joint pain, muscle pain, and injection site reactions. The total input parameters for the first and second dose predictions were 46 and 54 features, respectively, including age, gender, lifestyle variables, and medical history. The performances of multiple ML models were compared using Area Under the Receiver Operating Characteristic Curve (ROC-AUC). Results: The total number of people receiving the first dose of the AZD1222, Sputnik V, BBIBP-CorV, COVAXIN, BNT162b2, and mRNA-1273 were 6022, 7290, 5279, 802, 277, and 273, respectively. For the second dose, the numbers were 2851, 5587, 3841, 599, 242 and 228. The Logistic Regression model for predicting different side effects of the first dose achieved ROC-AUCs of 0.620-0.686, 0.685-0.716, 0.632-0.727, 0.527-0.598, 0.548-0.655, 0.545-0.712 for the AZD1222, Sputnik V, BBIBP-CorV, COVAXIN, BNT162b2 and mRNA-1273 vaccines, respectively. The second dose models yielded ROC-AUCs of 0.777-0.867, 0.795-0.848, 0.857-0.906, 0.788-0.875, 0.683-0.850, and 0.486-0.680, respectively. Conclusions: Using a large cohort of recipients vaccinated with COVID-19 vaccines, a novel and personalized strategy was established to predict the occurrence of the most common adverse side effects with high accuracy. This technique can serve as a tool to inform COVID-19 vaccine selection and generate personalized factsheets to curb concerns about adverse side effects.

Identifying the preparedness components in COVID-19: Systematic literature review.

In 2019, the COVID-19 pandemic posed a major challenge to the world. Since the world is constantly exposed to communicable diseases, comprehensive preparedness of countries is required. Therefore, the present systematic review is aimed at identifying the preparedness components in COVID-19. In this systematic literature review, PubMed, Scopus, Web of Science, ProQuest, Science Direct, Iran Medex, Magiran, and Scientific Information Database were searched from 2019 to 2021 to identify preparedness components in COVID-19. Thematic content analysis method was employed for data analysis. Out of 11,126 journals retrieved from searches, 45 studies were included for data analysis. Based on the findings, the components of COVID-19 preparedness were identified and discussed in three categories: governance with three subcategories of characteristics, responsibilities, and rules and regulations; society with two subcategories of culture and resilience; and services with three subcategories of managed services, advanced technology, and prepared health services. Among these, the governance and its subcategories had the highest frequency in studies. Considering the need to prepare for the next pandemic, countries should create clear and coherent structures and responsibilities for crisis preparedness through legal mechanisms, strengthening the infrastructure of the health system, coordination between organizations through analysis and identification of stakeholders, culture building and attracting social participation, and service management for an effective response.

Migrant Health and COVID-19 Pandemic: A Cross-sectional Study of Characteristics, Clinical Features, and Health Outcome from Iran.

BACKGROUND: As the pandemic unfolds, major concerns remain with those in disadvantaged positions who may be disproportionately affected. This paper aimed to present the characteristics of COVID-19 immigrant patients and investigate whether they were disproportionately affected by COVID-19 pandemic. METHODS: A cross-sectional study was performed using data on 589,146 patients diagnosed with COVID-19 in Iran. Descriptive analyses were used to summarize the study population's characteristics. Chi-squared test and logistic regression model were applied. RESULTS: After accounting for possible confounding covariates, being an immigrant was significantly associated with increased risk of death due to COVID-19 (OR 1.64, CI 1.568-1.727). When compared to Iranian-born patients, the prevalence of low blood oxygen levels on admission was higher among immigrant patients (53.9% versus 47.7%, P value < 0.001). Moreover, greater proportions of immigrants who were diagnosed with COVID-19 were admitted to an ICU (17% versus 15.8%, P value < 0.001). Patients aged 65 and above were the largest age category in both populations. However, there was a significant difference between the age profiles of patients, with children under the age of eighteen presenting 16% of immigrant patients vs 6.6% of Iranian-born patients (P value < 0.001). In both groups, more men were affected by COVID-19 than women, yet the sex bias was more prominent for migrant patients (P value < 0.001). CONCLUSION: The evidence from this study revealed that immigrant patients infected with COVID-19 were more likely to suffer from severe health outcome of the disease compared to Iranian-born patients.

Assessment of the relationship between exposure to air pollutants and COVID-19 pandemic in Tehran city, Iran.

The COVID-19 disease caused by the SARS-CoV-2 virus first identified in December 2019 has resulted in millions of deaths so far around the world. Controlling the spread of the disease requires a good understanding of the factors (e.g. air pollutants) that influence virus transmission and the conditions under which it spreads. This study analyzed the relationships between COVID-19 cases and both short-term (6-month) and long-term (60-month) exposures to eight air pollutants (NO, NO2, NOx, CO, SO2, O3, PM2.5 and PM10) in Tehran city, Iran, by integrating geostatistical interpolation models, regression analysis, and an innovated COVID-19 incidence rate calculation (Q-index) that considered the spatial distributions of both population and air pollution. The results show that the higher COVID-19 incidence rate was significantly associated with the exposure to higher concentrations of CO, NO, and NOx during the short-term period; the higher COVID-19 incidence rate was significantly related to the exposure to higher concentrations of PM2.5 during the long-term period; while COVID-19 incidence rate was not significantly associated with the concentrations of O3, SO2, PM10 and NO2 in either period. This study indicates that exposure to air pollutants can effect an increase in the number of infected people by transmitting the virus through the air or by predisposing people to the disease over time. The Q-index calculation method developed in this study can be also used by other studies to calculate more accurate disease rates that consider the spatial distribution of both population and air pollution.

Relationship Between Income Level and Hospitalization Rate in COVID-19 Cases; an Example of Social Factors Affecting Health.

Introduction: Considering the population's socioeconomic status and clinical features is essential in planning and performing interventions related to disease control. The main purpose of this study was to investigate the relationship between income level and hospitalization rate of COVID-19 patients|. Methods: A cross-sectional study was performed on 198,944 hospitalized COVID-19 patients in Tehran province between March 2020 and March 2021. Data of hospitalized COVID-19 patients was obtained from the Hospital Intelligent Management System (HIM). The income data of patients were obtained from the Iranian Database on Targeted Subsidies belonging to the Ministry of Cooperatives, Labor, and Social Welfare. Data analyses were performed using SPSS software. Results: About 2.5% of the inpatients were from the first decile, while 20.6% were from the tenth. The share of the lower three deciles of total hospitalization was about 11%, while the share of the upper three deciles was 50%. There was a big difference between the upper- and lower-income deciles regarding death rates. In the first decile, 30% of inpatients died, while the proportion was 10% in the tenth decile. There was a significant and positive relationship between income decline and hospitalization (r = 0.75; p = 0.02). Also, there was a significant and negative relationship between income decline and death rate (r = -0.90; p = 0.01). Conclusion: Low-income groups use fewer inpatient services, are more prone to severe illness and death from COVID-19|, and treatment in this group has a lower chance of success. Using a systemic approach to address socioeconomic factors in healthcare planning is crucial.

Urban and sub-urban disparities in health outcomes among patients with COVID-19; a cross-sectional study of 234 418 patients in Iran.

BACKGROUND: It remains crucial to understand socio-demographic determinants of COVID-19 infection to improve access to care and recovery rates from the disease. This study aimed to investigate the urban and sub-urban disparities associated with COVID-19 in patients visiting healthcare facilities in the province of Tehran, Iran. METHODS: Data from 234 418 patients who were diagnosed with COVID-19 infection from March 2020 to March 2021 in the province of Tehran were used in this analysis. Descriptive statistics were used to describe the characteristics of the study population. Chi-Squared test was applied to examine the association of study variables with residing area. Independent samples t-test was performed to compare mean age of patients in urban and sub-urban areas. Multiple Logistic Regression model was applied to examine the association of study variables with disease outcome. RESULTS: Overall, most patients resided in the urban settings (73%). Mean age of patients was significantly lower in sub-urban areas compared to their counterparts in urban settings (49 ± 23.1 years versus 53 ± 21.1 years, P < 0.001). Positive PCR test results were more common in urban areas (48.5% versus 41.3%, P < 0.001). Yet, sub-urban settings had higher rates of positive chest CT scan reports (62.8% versus 53.4%, P < 0.001). After accounting for age and sex covariates, residing in urban areas was associated with higher likelihood of being admitted to an ICU (OR = 1.27, CI: 1.240-1.305). Yet, a greater vulnerability to fatal outcome of COVID-19 infection was shown in patients living in sub-urban areas (OR = 1.13, CI: 1.105-1.175). CONCLUSIONS: This study revealed a clear disparity in the health outcome of patients infected with COVID-19 between urban and sub-urban areas.

A description of spatial-temporal patterns of the novel COVID-19 outbreak in the neighbourhoods' scale in Tehran, Iran.

Background: Analyzing and monitoring the spatial-temporal patterns of the new coronavirus disease (COVID-19) pandemic can assist local authorities and researchers in detecting disease outbreaks in the early stages. Because of different socioeconomic profiles in Tehran's areas, we will provide a clear picture of the pandemic distribution in Tehran's neighbourhoods during the first months of its spread from February to July 2020, employing a spatial-temporal analysis applying the geographical information system (GIS). Disease rates were estimated by location during the 5 months, and hot spots and cold spots were highlighted. Methods: This study was performed using the COVID-19 incident cases and deaths recorded in the Medical Care Monitoring Centre from February 20, to July 20, 2020. The local Getis-Ord Gi* method was applied to identify the hotspots where the infectious disease distribution had significantly clustered spatially. A statistical analysis for incidence and mortality rates and hot spots was conducted using ArcGIS 10.7 software. Results: The addresses of 43,000 Tehrani patients (15,514 confirmed COVID-19 cases and 27,486 diagnosed as probable cases) were changed in its Geo-codes in the GIS. The highest incidence rate from February to July 2020 was 48 per 10,000 and the highest 5-month incidence rate belonged to central and eastern neighbourhoods. According to the Cumulative Population density of patients, the higher number is estimated by more than 2500 people in the area; however, the lower number is highlighted by about 500 people in the neighborhood. Also, the results from the local Getis-Ord Gi* method indicate that COVID-19 has formed a hotspot in the eastern, southeast, and central districts in Tehran since February. We also observed a death rate hot spot in eastern areas. Conclusion: Because of the spread of COVID-19 disease throughout Tehran's neighborhoods with different socioeconomic status, it seems essential to pay attention to health behaviors to prevent the next waves of the disease. The findings suggest that disease distribution has formed a hot spot in Tehran's eastern and central regions.

The most places at risk surrounding the COVID-19 treatment hospitals in an urban environment- case study: Tehran city.

Investigations on the spatial patterns of COVID-19 spreading indicate the possibility of the virus transmission by moving infected people in an urban area. Hospitals are the most susceptible locations due to the COVID-19 contaminations in metropolises. This paper aims to find the riskiest places surrounding the hospitals using an MLP-ANN. The main contribution is discovering the influence zone of COVID-19 treatment hospitals and the main spatial factors around them that increase the prevalence of COVID-19. The innovation of this paper is to find the most relevant spatial factors regarding the distance from central hospitals modeling the risk level of the study area. Therefore, eight hospitals with two service areas for each of them are computed with [0-500] and [500-1000] meters distance. Besides, five spatial factors have been considered, consist of the location of patients' financial transactions, the distance of streets from hospitals, the distance of highways from hospitals, the distance of the non-residential land use from the hospitals, and the hospital patient number. The implementation results revealed a meaningful relation between the distance from the hospitals and patient density. The RMSE and R measures are 0.00734 and 0.94635 for [0-500 m] while these quantities are 0.054088 and 0.902725 for [500-1000 m] respectively. These values indicate the role of distance to central hospitals for COVID-19 treatment. Moreover, a sensitivity analysis demonstrated that the number of patients' transactions and the distance of the non-residential land use from the hospitals are two dominant factors for virus propagation. The results help urban managers to begin preventative strategies to decrease the community incidence rate in high-risk places.

Symptom Prediction and Mortality Risk Calculation for COVID-19 Using Machine Learning.

Background: Early prediction of symptoms and mortality risks for COVID-19 patients would improve healthcare outcomes, allow for the appropriate distribution of healthcare resources, reduce healthcare costs, aid in vaccine prioritization and self-isolation strategies, and thus reduce the prevalence of the disease. Such publicly accessible prediction models are lacking, however. Methods: Based on a comprehensive evaluation of existing machine learning (ML) methods, we created two models based solely on the age, gender, and medical histories of 23,749 hospital-confirmed COVID-19 patients from February to September 2020: a symptom prediction model (SPM) and a mortality prediction model (MPM). The SPM predicts 12 symptom groups for each patient: respiratory distress, consciousness disorders, chest pain, paresis or paralysis, cough, fever or chill, gastrointestinal symptoms, sore throat, headache, vertigo, loss of smell or taste, and muscular pain or fatigue. The MPM predicts the death of COVID-19-positive individuals. Results: The SPM yielded ROC-AUCs of 0.53-0.78 for symptoms. The most accurate prediction was for consciousness disorders at a sensitivity of 74% and a specificity of 70%. 2,440 deaths were observed in the study population. MPM had a ROC-AUC of 0.79 and could predict mortality with a sensitivity of 75% and a specificity of 70%. About 90% of deaths occurred in the top 21 percentile of risk groups. To allow patients and clinicians to use these models easily, we created a freely accessible online interface at www.aicovid.net. Conclusion: The ML models predict COVID-19-related symptoms and mortality using information that is readily available to patients as well as clinicians. Thus, both can rapidly estimate the severity of the disease, allowing shared and better healthcare decisions with regard to hospitalization, self-isolation strategy, and COVID-19 vaccine prioritization in the coming months.

Sociodemographic determinants and clinical risk factors associated with COVID-19 severity: a cross-sectional analysis of over 200,000 patients in Tehran, Iran.

BACKGROUND: Defining socio-demographic factors, clinical presentations and underlying diseases associated with COVID-19 severity could be helpful in its management. This study aimed to further clarify the determinants and clinical risk factors of the disease severity in patients infected with COVID-19. METHODS: A multi-centre descriptive study on all patients who have been diagnosed with COVID-19 in the province of Tehran from March 2020 up to Dec 2020 was conducted. Data on socio-demographic characteristics, clinical presentations, comorbidities, and the health outcomes of 205,654 patients were examined. Characteristics of the study population were described. To assess the association of study variables with the disease severity, the Chi-Squared test and Multiple Logistic Regression model were applied. RESULTS: The mean age of the study population was 52.8 years and 93,612 (45.5%) were women. About half of the patients have presented with low levels of blood oxygen saturation. The ICU admission rate was 17.8% and the overall mortality rate was 10.0%. Older age, male sex, comorbidities including hypertension, cancer, chronic respiratory diseases other than asthma, chronic liver diseases, chronic kidney diseases, chronic neurological disorders, and HIV/AIDS infection were risk markers of poor health outcome. Clinical presentations related with worse prognosis included fever, difficulty breathing, impaired consciousness, and cutaneous manifestations. CONCLUSION: These results might alert physicians to pay attention to determinants and risk factors associated with poor prognosis in patients with COVID-19. In addition, our findings aid decision makers to emphasise on vulnerable groups in the public health strategies that aim at preventing the spread of the disease and its mortalities.

Spatio-temporal patterns of the COVID-19 pandemic, and place-based influential factors at the neighborhood scale in Tehran.

Since its emergence in late 2019, the COVID-19 pandemic has attracted the attention of researchers in various fields, including urban planning and design. However, the spreading patterns of the disease in cities are still not clear. Historically, preventing and controlling pandemics in cities has always been challenging due to various factors such as higher population density, higher mobility of people, and higher contact frequency. To shed more light on the spread patterns of the pandemic, in this study we analyze 43,000 confirmed COVID-19 cases at the neighborhood level in Tehran, the capital of Iran. To examine spatio-temporal patterns and place-based factors contributing to the spread of the pandemic, we used exploratory spatial data analysis and spatial regression. We developed a geo-referenced database composed of 12 quantitative place-based variables related to physical attributes, land use and public transportation facilities, and demographic status. We also used the geographically weighted regression model for the local examination of spatial non-stationarity. According to the results, population density (R2 = 0.88) and distribution of neighborhood centers (R2 = 0.59), drugstores (R2 = 0.64), and chain stores (R2 = 0.59) are the main factors contributing to the spread of the disease. Additionally, density of public transportation facilities showed a varying degree of contribution. Overall, our findings suggest that demographic composition and major neighborhood-level physical attributes are important factors explaining high rates of infection and mortality. Results contribute to gaining a better understanding of the role of place-based attributes that may contribute to the spread of the pandemic and can inform actions aimed at achieving Sustainable Development Goals, particularly Goals 3 and 11.

Resilient urban form to pandemics: Lessons from COVID-19.

Background: The worldwide emergence of future pandemics emphasizes the need to assess the pandemic resilient urban form to prevent infectious disease transmission during this epidemic. According to the lessons of the COVID-19 outbreak, this study aimed to review the current strategies of responding to pandemics through disaster risk management (DRM) to develop a pandemic-resilient urban form in phases of response, mitigation, and preparedness. Methods: The research method is developed through desk study was used to explore the current literature of urban form responded to COVID-19 pandemic and for the text analysis; qualitative content analysis was applied developing a conceptual framework. Results: To create pandemic resilient urban form, this study proposes principles to enhance the urban form resiliency in 3 scales of housing, neighborhoods/public spaces, and cities. These principles focus on the concept of resilient urban form from new perspectives focusing on the physical and nonphysical aspects of resilient urban form, which develops a new understanding of pandemics as a disaster and health-related emergency risks. The physical aspect of resiliency to epidemic outbreaks includes urban form, access, infrastructure, land use, and natural environment factors. Moreover, the nonphysical aspect can be defined by the sociocultural, economic, and political (including good governance) factors. By providing and enhancing the physical and nonphysical prerequisites, several benefits can be gained and the effectiveness of all response, mitigation, and preparedness activities can be supported. Conclusion: As the pandemic's disruptions influence the citizens' lifestyle dramatically, the prominent role of place characteristics in the outbreak of pandemics, policymakers, urban planners, and urban designers should be pulled together to make urban areas more resilient places for epidemics and pandemics.

Correlation Between Low-Dose Chest Computed Tomography and RT-PCR Results for the Diagnosis of COVID-19: A Report of 27,824 Cases in Tehran, Iran.

RATIONALE AND OBJECTIVES: Real-time polymerase chain reaction (RT-PCR) remains the gold standard for confirmation of Coronavirus Disease 2019 (COVID-19) despite having many disadvantages. Here, we investigated the diagnostic performance of chest computed tomography (CT) as an alternative to RT-PCR in patients with clinical suspicion of COVID-19 infection. METHODS: In this descriptive cross-sectional study, 27,824 patients with clinical suspicion of COVID-19 infection who underwent unenhanced low-dose chest CT from 20 February, 2020 to 21 May, 2020 were evaluated. Patients were recruited from seven specifically designated hospitals for patients with COVID-19 infection affiliated to Shahid Beheshti University of Medical Sciences. In each hospital, images were interpreted by two independent radiologists. CT findings were considered as positive/negative for COVID-19 infection based on RSNA diagnostic criteria. Then, the correlation between the number of daily positive chest CT scans and number of daily PCR-confirmed cases and COVID-19-related deaths in Tehran province during this three-month period was assessed. The trends of admission rate and patients with positive CT scans were also evaluated. RESULTS: A strong positive correlation between the numbers of daily positive CT scans and daily PCR-confirmed COVID-19 cases (r = 0.913, p < 0.001) was observed. Furthermore, in hospitals located in regions with a lower socioeconomic status, the admission rate and number of positive cases within this three-month period was higher as compared to other hospitals. CONCLUSION: Low-dose chest CT is a safe, rapid and reliable alternative to RT-PCR for the diagnosis of COVID-19 in high-prevalence regions. In addition, our study provides further evidence for considering patients' socioeconomic status as an important risk factor for COVID-19.

Baseline Characteristics and Associated Factors of Mortality in COVID-19 Patients;an Analysis of 16000 Cases in Tehran, Iran

Introduction: Given the importance of evidence-based decision-making, this study aimed to evaluate epidemiological and clinical characteristics as well as associate factors of mortality among admitted COVID-19 cases. Methods: This multicenter, cross-sectional study was conducted on confirmed and suspected COVID-19 cases who were hospitalized in 19 public hospitals affiliated to Shahid Beheshti University of Medical Sciences (SBMU), Tehran, Iran, between February 19 and May 12, 2020. Epidemiological and clinical characteristics of the infected cases were compared between the deceased and survivors after discharge. Case fatality rates (CFRs) were calculated across all study variables. Single and multiple logistic regressions were used to explore the risk factors associated with COIVD-19 mortality. Results: Out of the 16035 cases that referred to the hospitals affiliated to SBMU, 16016 patients (99.93% of Confirmed and 99.83% of suspected cases) were hospitalized. 1612 patients died with median hospitalization days of 5 (interquartile range (IQR): 2-9) and 3 (1-7) for confirmed and suspected COVID-19 cases, respectively. The highest death rate was observed among ages>65 (63.4% of confirmed cases, 62.3% of suspected cases) and intensive care unit (ICU)/critical care unit (CCU) patients (62.7% of confirmed cases, 52.2% of suspected cases). Total case fatality rate (CFR) was 10.05% (13.52% and 6.37% among confirmed and suspected cases, respectively). The highest total CFR was observed in patients with age>65 years (25.32%), underlying comorbidities (25.55%), and ICU/CCU patients (41.7%). The highest CFR was reported for patients who had diabetes and cardiovascular diseases (38.46%) as underlying non-communicable diseases (NCDs), and patients with cancer (35.79%). Conclusion: This study showed a high CFR among suspected and confirmed COVID-19 cases, and highlighted the main associated risk factors including age, sex, underlying NCDs, and ICU/CCU admission affecting survival of COVID-19 patients.