Identifying psychological antecedents and predictors of vaccine hesitancy through machine learning: A cross sectional study among chronic disease patients of deprived urban neighbourhood, India.

COVID-19 vaccine hesitancy among chronic disease patients can severely impact individual health with the potential to impede mass vaccination essential for containing the pandemic. The present study was done to assess the COVID-19 vaccine antecedents and its predictors among chronic disease patients. This cross-sectional study was conducted among chronic disease patients availing care from a primary health facility in urban Jodhpur, Rajasthan. Factor and reliability analysis was done for the vaccine hesitancy scale to validate the 5 C scale. Predictors assessed for vaccine hesitancy were modelled with help of machine learning (ML). Out of 520 patients, the majority of participants were female (54.81%). Exploratory factor analysis revealed four psychological antecedents' "calculation"; "confidence"; "constraint" and "collective responsibility" determining 72.9% of the cumulative variance of vaccine hesitancy scale. The trained ML algorithm yielded an R2 of 0.33. Higher scores for COVID-19 health literacy and preventive behaviour, along with family support, monthly income, past COVID-19 screening, adherence to medications and age were associated with lower vaccine hesitancy. Behaviour changes communication strategies targeting COVID-19 health literacy and preventive behaviour especially among population sub-groups with poor family support, low income, higher age groups and low adherence to medicines may prove instrumental in this regard.

Early trends of socio-economic and health indicators influencing case fatality rate of COVID-19 pandemic.

Coronavirus disease 2019, i.e. COVID-19, started as an outbreak in a district of China and has engulfed the world in a matter of 3 months. It is posing a serious health and economic challenge worldwide. However, case fatality rates (CFRs) have varied amongst various countries ranging from 0 to 8.91%. We have evaluated the effect of selected socio-economic and health indicators to explain this variation in CFR. Countries reporting a minimum of 50 cases as on 14th March 2020, were selected for this analysis. Data about the socio-economic indicators of each country was accessed from the World bank database and data about the health indicators were accessed from the World Health Organisation (WHO) database. Various socioeconomic indicators and health indicators were selected for this analysis. After selecting from univariate analysis, the indicators with the maximum correlation were used to build a model using multiple variable linear regression with a forward selection of variables and using adjusted R-squared score as the metric. We found univariate regression results were significant for GDP (Gross Domestic Product) per capita, POD 30/70 (Probability Of Dying Between Age 30 And Exact Age 70 From Any of Cardiovascular Disease, Cancer, Diabetes or Chronic Respiratory Disease), HCI (Human Capital Index), GNI(Gross National Income) per capita, life expectancy, medical doctors per 10000 population, as these parameters negatively corelated with CFR (rho = -0.48 to -0.38 , p<0.05). Case fatality rate was regressed using ordinary least squares (OLS) against the socio-economic and health indicators. The indicators in the final model were GDP per capita, POD 30/70, HCI, life expectancy, medical doctors per 10,000, median age, current health expenditure per capita, number of confirmed cases and population in millions. The adjusted R-squared score was 0.306. Developing countries with a poor economy are especially vulnerable in terms of COVID-19 mortality and underscore the need to have a global policy to deal with this on-going pandemic. These trends largely confirm that the toll from COVID-19 will be worse in countries ill-equipped to deal with it. These analyses of epidemiological data are need of time as apart from increasing situational awareness, it guides us in taking informed interventions and helps policy-making to tackle this pandemic.

Extrapolation of mortality in COVID-19: Exploring the role of age, sex, co-morbidities and health-care related occupation.

We used a publicly available data of 44,672 patients reported by China's centre for disease control to study the role of age, sex, co-morbidities and health-care related occupation on COVID-19 mortality. The data is in the form of absolute numbers and proportions. Using the percentages, retrospective synthetic data of 100 survivors and 100 deaths were generated using random number libraries so that proportions of ages, genders, co-morbidities, and occupations were constant as in the original data. Logistic regression of the four predictor factors of age, sex, co-morbidities and occupation revealed that only age and comorbidities significantly affected mortality. Sex and occupation when adjusted for other factors in the equation were not significant predictors of mortality. Age and presence of co-morbidities correlated negatively with survival with co-efficient of -1.23 and -2.33 respectively. Odds ratio (OR) for dying from COVID-19 for every 10-year increase in age was 3.4 compared to the previous band of 10 years. OR for dying of COVID-19 was 10.3 for the presence of any of the co-morbidities. Our findings could help in triaging the patients in the emergency room and emphasize the need to protect the elderly and those with comorbidities from getting exposed.

Assessing the accuracy of pleural puncture sites in patients with pleural effusion as determined by clinical examination versus ultrasound-A single-centre prospective study.

INTRODUCTION: This study aimed to ascertain the accuracy of clinical examination for the determination of pleural puncture sites as compared to the use of ultrasonography in patients with pleural effusion. MATERIAL AND METHODS: A single-centre, prospective, observational study was carried out amongst 115 patients with pleural effusion in a tertiary care hospital in western India. Patients were subjected to clinical assessment for determination of pleural puncture sites and the same were confirmed with ultrasonography. All physicians were blinded to the marking of the previous physician to prevent any influence on their assessment. RESULTS: The study had 345 physician observations. The overall accuracy of the clinical examination was 94.8%. Multivariate logistic regression of the factors responsible for the accuracy of clinical examination demonstrated a significant role of higher body mass index (BMI) (OR-1.19) and lower zone pleural effusions (OR-4.99) when adjusted for age, gender, side of effusion, and experience of examining doctors. When the effusions were classified according to their location, lower zone pleural effusions and loculated pleural effusions had an error rate of 15.9% and 8.33%, respectively. CONCLUSION: An ultrasound is the standard of care to assess all pleural effusions and guide the best point for aspiration.

Development of Electronic Health Record-Based Machine Learning Models to Predict Barrett's Esophagus and Esophageal Adenocarcinoma Risk.

INTRODUCTION: Screening for Barrett's esophagus (BE) is suggested in those with risk factors, but remains underutilized. BE/esophageal adenocarcinoma (EAC) risk prediction tools integrating multiple risk factors have been described. However, accuracy remains modest (area under the receiver-operating curve [AUROC] ≤0.7), and clinical implementation has been challenging. We aimed to develop machine learning (ML) BE/EAC risk prediction models from an electronic health record (EHR) database. METHODS: The Clinical Data Analytics Platform, a deidentified EHR database of 6 million Mayo Clinic patients, was used to predict BE and EAC risk. BE and EAC cases and controls were identified using International Classification of Diseases codes and augmented curation (natural language processing) techniques applied to clinical, endoscopy, laboratory, and pathology notes. Cases were propensity score matched to 5 independent randomly selected control groups. An ensemble transformer-based ML model architecture was used to develop predictive models. RESULTS: We identified 8,476 BE cases, 1,539 EAC cases, and 252,276 controls. The BE ML transformer model had an overall sensitivity, specificity, and AUROC of 76%, 76%, and 0.84, respectively. The EAC ML transformer model had an overall sensitivity, specificity, and AUROC of 84%, 70%, and 0.84, respectively. Predictors of BE and EAC included conventional risk factors and additional novel factors, such as coronary artery disease, serum triglycerides, and electrolytes. DISCUSSION: ML models developed on an EHR database can predict incident BE and EAC risk with improved accuracy compared with conventional risk factor-based risk scores. Such a model may enable effective implementation of a minimally invasive screening technology.

Identification and risk stratification of coronary disease by artificial intelligence-enabled ECG.

Background: Atherosclerotic cardiovascular disease (ASCVD) is the leading cause of death worldwide, driven primarily by coronary artery disease (CAD). ASCVD risk estimators such as the pooled cohort equations (PCE) facilitate risk stratification and primary prevention of ASCVD but their accuracy is still suboptimal. Methods: Using deep electronic health record data from 7,116,209 patients seen at 70+ hospitals and clinics across 5 states in the USA, we developed an artificial intelligence-based electrocardiogram analysis tool (ECG-AI) to detect CAD and assessed the additive value of ECG-AI-based ASCVD risk stratification to the PCE. We created independent ECG-AI models using separate neural networks including subjects without known history of ASCVD, to identify coronary artery calcium (CAC) score ≥300 Agatston units by computed tomography, obstructive CAD by angiography or procedural intervention, and regional left ventricular akinesis in ≥1 segment by echocardiogram, as a reflection of possible prior myocardial infarction (MI). These were used to assess the utility of ECG-AI-based ASCVD risk stratification in a retrospective observational study consisting of patients with PCE scores and no prior ASCVD. The study period covered all available digitized EHR data, with the first available ECG in 1987 and the last in February 2023. Findings: ECG-AI for identifying CAC ≥300, obstructive CAD, and regional akinesis achieved area under the receiver operating characteristic (AUROC) values of 0.88, 0.85, and 0.94, respectively. An ensembled ECG-AI identified 3, 5, and 10-year risk for acute coronary events and mortality independently and additively to PCE. Hazard ratios for acute coronary events over 3-years in patients without ASCVD that tested positive on 1, 2, or 3 versus 0 disease-specific ECG-AI models at cohort entry were 2.41 (2.14-2.71), 4.23 (3.74-4.78), and 11.75 (10.2-13.52), respectively. Similar stratification was observed in cohorts stratified by PCE or age. Interpretation: ECG-AI has potential to address unmet need for accessible risk stratification in patients in whom PCE under, over, or insufficiently estimates ASCVD risk, and in whom risk assessment over time periods shorter than 10 years is desired. Funding: Anumana.

Air flow limitation in smokers - A cause of concern.

Introduction: In India smoking is a common habit prevalent in both urban and rural areas irrespective of mode of smoking i.e., cigarettes, bidis, pipes, cigar, hookah etc., Spirometry can be helpful to determine effects of smoking on pulmonary functions. We aimed to study the effect of smoking on the pulmonary function tests. Materials and Methods: This study was conducted on 300 subjects including 150 smokers and 150 non-smokers aged between 25 and 60 years attending a tertiary health care center in northern part of our country. Quantification of tobacco smoking was performed by calculating smoking index. All the study subjects underwent spirometry. Results: All the spirometric variables (FVC, FEV1, PEFR, FEF 25-75%) were lower in smokers as compared to non-smokers and this difference was statistically significant. 76% of the smokers had obstructive, 10.7% had normal, 6.7% had restrictive, and 6.7% had mixed pattern on spirometry. 65.3% of the non-smokers had normal, 28.7% had obstructive and 6% had restrictive pattern on spirometry. Conclusion: Almost all the pulmonary function parameters were significantly reduced in smokers compared to non-smokers and obstructive impairment was common amongst smokers. As early quitting is associated with improved survival, it is important that these asymptomatic smokers are identified early and helped to quit. Primary care physicians being the first point of contact, can play a major role.

Emotional distress among health professionals involved in care of inpatients with COVID-19: a survey based cross-sectional study.

INTRODUCTION: Health care workers (HCWs) are directly involved in processes linked with diagnosis, management, and assistance of coronavirus disease-19 (COVID-19) patients which could have direct implications on their physical and emotional health. Emotional aspects of working in an infectious pandemic situation is often neglected in favour of the more obvious physical ramifications. This single point assessment study aimed to explore the factors related to stress, anxiety and depression among HCWs consequent to working in a pandemic. MATERIAL AND METHODS: This was a cross-sectional study involving healthcare workers who were working in COVID-19 inpatient ward, COVID-19 screening area, suspect ward, suspect intensive care unit (ICU) and COVID-19 ICU across four hospitals in India. A web-based survey questionnaire was designed to elicit responses to daily challenges faced by HCWs. The questionnaire was regressed using machine-learning algorithm (Cat Boost) against the standardized Depression, Anxiety and Stress Scale - 21 (DASS 21) which was used to quantify emotional distress experienced by them. RESULTS: A total of 156 participants were included in this study. As per DASS-21 scoring, severe stress was seen in ∼17% of respondents. We could achieve an R² of 0.28 using our machine-learning model. The major factors responsible for stress were decreased time available for personal needs, increasing age, being posted out of core area of expertise, setting of COVID-19 care, increasing duty hours, increasing duty days, marital status and being a resident physician. CONCLUSIONS: Factors elicited in this study that are associated with stress in HCWs need to be addressed to provide wholesome emotional support to HCWs battling the pandemic. Targeted interventions may result in increased emotional resilience of the health-care system.

Mortality and Its Predictors in COVID-19 Patients With Pre-existing Interstitial Lung Disease.

Background The data on the impact of coronavirus disease 2019 (COVID-19) on interstitial lung disease (ILD) is still limited. To the best of our knowledge, there has been no study from India to date to assess the impact of COVID-19 in patients with preexisting ILD. We undertook this study to assess the clinical outcome of ILD patients admitted to our hospital with COVID-19. Methods In this retrospective observational study, records of reverse transcription polymerase chain reaction (RT-PCR)-confirmed COVID-19 patients with preexisting ILD who were admitted to the hospital in the period from May 1, 2020, to April 30, 2021, were obtained from the hospital database. The clinical outcomes of the patients were recorded. Univariate analysis was performed to find relation between various predetermined risk factors for mortality and those with significant p values (p<0.05) were subjected to multiple logistic regression to determine independent risk factors. Results In our study of 28 patients, the overall mortality was 35.7%. On comparing the parameters associated with increased mortality, there was no effect of age, gender, comorbidities, type of ILD, CT thorax findings on diagnosis, use of corticosteroids and antifibrotics in the past, spirometric findings on mortality. On multivariate analysis, the significant parameters were interleukin 6 (IL-6), p=0.02, OR=1.020 (1.006-1.043) and D-dimer, p=0.04, OR=2.14 (5.55-1.14). Conclusion COVID-19 in patients with pre-existing ILD has a comparatively higher mortality. D-dimer and IL-6 are significant predictors of mortality in ILD patients infected with COVID-19.

D-dimer, disease severity, and deaths (3D-study) in patients with COVID-19: a systematic review and meta-analysis of 100 studies.

Hypercoagulability and the need for prioritizing coagulation markers for prognostic abilities have been highlighted in COVID-19. We aimed to quantify the associations of D-dimer with disease progression in patients with COVID-19. This systematic review and meta-analysis was registered with PROSPERO, CRD42020186661.We included 113 studies in our systematic review, of which 100 records (n = 38,310) with D-dimer data) were considered for meta-analysis. Across 68 unadjusted (n = 26,960) and 39 adjusted studies (n = 15,653) reporting initial D-dimer, a significant association was found in patients with higher D-dimer for the risk of overall disease progression (unadjusted odds ratio (uOR) 3.15; adjusted odds ratio (aOR) 1.64). The time-to-event outcomes were pooled across 19 unadjusted (n = 9743) and 21 adjusted studies (n = 13,287); a strong association was found in patients with higher D-dimers for the risk of overall disease progression (unadjusted hazard ratio (uHR) 1.41; adjusted hazard ratio (aHR) 1.10). The prognostic use of higher D-dimer was found to be promising for predicting overall disease progression (studies 68, area under curve 0.75) in COVID-19. Our study showed that higher D-dimer levels provide prognostic information useful for clinicians to early assess COVID-19 patients at risk for disease progression and mortality outcomes. This study, recommends rapid assessment of D-dimer for predicting adverse outcomes in COVID-19.

Cotrimoxazole-induced SIADH - a unique challenge during treatment of pulmonary nocardiosis.

A 62 year old male non-smoker diagnosed with pulmonary nocardiosis was initiated on Cotrimoxazole therapy at a dose of 20 mg/kg per day in three divided doses. He developed hyponatremia (serum sodium 105 mEq/L) on day 3 of therapy. The potential causes of hyponatremia were evaluated. After ruling out other causes, the cause was suspected to be Cotrimoxazole-induced syndrome of inappropriate anti-diuretic hormone secretion (SIADH). We subsequently re-initiated therapy with Cotrimoxazole and the hyponatremia (serum sodium 110 mEq/L) recurred. Upon discontinuation of therapy, serum sodium levels returned to normal. The patient was started on Amoxycillin-Clavulanic Acid as an alternative therapy for pulmonary nocardiosis which resulted in resolution of the hyponatremia. Cotrimoxazole-induced SIADH is a rare occurrence. This case is representative of a patient with Cotrimoxazole-induced SIADH and the causal relationship was confirmed once resumption of therapy with the offending medi-cation resulted in hyponatremia. Clinicians should be aware of this rare adverse effect of Cotrimoxazole and should monitor serum electrolytes during therapy, especially in the elderly and in those receiving high doses.

Recurrent pneumothorax, skin lesions and frequent urination.

The causes of cystic lung diseases are varied. Proper evaluation is required for appropriate management. http://bit.ly/37J7dvE.

Low dose radiotherapy for COVID pneumonia: Irradiate to Eradicate - Will it be possible?

.

Using a simple open-source automated machine learning algorithm to forecast COVID-19 spread: A modelling study.

INTRODUCTION: Machine learning algorithms have been used to develop prediction models in various infectious and non-infectious settings including interpretation of images in predicting the outcome of diseases. We demonstrate the application of one such simple automated machine learning algorithm to a dataset obtained about COVID-19 spread in South Korea to better understand the disease dynamics. MATERIAL AND METHODS: Data from 20th January 2020 (when the first case of COVID-19 was detected in South Korea) to 4th March 2020 was accessed from Korea's centre for disease control (KCDC). A future time-series of specified length (taken as 7 days in our study) starting from 5th March 2020 to 11th March 2020 was generated and fed to the model to generate predictions with upper and lower trend bounds of 95% confidence intervals. The model was assessed for its ability to reliably forecast using mean absolute percentage error (MAPE) as the metric. RESULTS: As on 4th March 2020, 145,541 patients were tested for COVID-19 (in 45 days) in South Korea of which 5166 patients tested positive. The predicted values approximated well with the actual numbers. The difference between predicted and observed values ranged from 4.08% to 12.77% . On average, our predictions differed from actual values by 7.42% (MAPE) over the same period. CONCLUSION: Open source and automated machine learning tools like Prophet can be applied and are effective in the context of COVID-19 for forecasting spread in naïve communities. It may help countries to efficiently allocate healthcare resources to contain this pandemic.

Clinical outcome, viral response and safety profile of chloroquine in COVID-19 patients - initial experience.

INTRODUCTION: Chloroquine and its analogues are currently being investigated for the treatment and post exposure prophylaxis of COVID-19 due to its antiviral activity and immunomodulatory activity. MATERIAL AND METHODS: Confirmed symptomatic cases of COVID-19 were included in the study. Patients were supposed to receive chloroquine (CQ) 500 mg twice daily for 7 days. Due to a change in institutional protocol, initial patients received chloroquine and subsequent patients who did not receive chloroquine served as negative controls. Clinical effectiveness was determined in terms of timing of symptom resolution and conversion rate of reverse transcriptase polymerase chain reaction (RT-PCR) on day 14 and day 15 of admission. RESULTS: Twelve COVID-19 patients formed the treatment arm and 17 patients were included in the control arm. The duration of symptoms among the CQ treated group (6.3 ± 2.7 days) was significantly (p-value = 0.009) lower than that of the control group (8.9 ± 2.2 days). There was no significant difference in the rate of RT-PCR negativity in both groups. 2 patients out of 12 developed diarrhea in the CQ therapy arm. CONCLUSION: The duration of symptoms among the treated group (with chloroquine) was significantly lower than that of the control group. RT-PCR conversion was not significantly different between the 2 groups.