Survival Analysis and Its Application of Discharge Time Likelihood Prediction Using Clinical Data on COVID-19 Patients-Machine-Learning Approaches

COVID-19 has significant fatality rate since its appearance in December 2019 as a respiratory ailment that is extremely contagious. As the number of cases in reduction zones rises, highly health officials are control that authorized treatment centers may become overrun with corona virus patients. Artificial neural networks (ANNs) are machine coding that can be used to find complicate relationships between datasets. They enable the detection of category in complicated biological datasets that would be impossible to identify with traditional linear statistical analysis. To study the survival characteristics of patients, several computational techniques are used. Men and older age groups had greater mortality rates than women, according to this study. COVID-19 patients discharge times were predicted;also, utilizing various machine learning and statistical tools applied technically. In medical research, survival analysis is a regularly used technique for identifying relevant predictors of adverse outcomes and developing therapy guidelines for patients. Historically, demographic statistics have been used to predict outcomes in such patients. These projections, on the other hand, have little meaning for the individual patient. We present the training of neural networks to predict outcomes for individual patients at one institution, as well as their predictive performance using data from another institution in a different region. The research output show that the Gradient boosting longevity model beats the all other different models, also in this research study for predicting patient longevity. This study aims to assist health officials in making more informed decisions during the outbreak. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Modeling time to event for epicenters of the pandemic COVID-19

The epidemic of COVID-19 has been the most mathematically informative pandemic. The unprecedented information gives rise to some unprecedented models, problems, and discussions. One of these new matters is modeling the epicenters of a pandemic. The present paper is the first attempt to model the waiting time to introduce a new epicenter during a pandemic. This modeling is conducted in terms of time-to-event, the number of epicenters, and the normalized time. We model the waiting time data by an exponential distribution, therefore, the number of epicenters can be represented through a Poisson process. Then, the parameters are estimated by the method of moments and maximum likelihood method. All the simulations are the result of 10,000 runs conducted on MATLAB R2015b. It is expected to encounter 12 and 14 (with probability 95%, 3-24 and 7-23) epicenters from 15th May to 13th June and from June 14 to July 12, 2020, respectively. We forecast that the cumulative number of confirmed cases for coming epicenters is over 10,000 when they join the existing epicenters. The paper suggests that the time to epicenter is a suitable criterion to compare the spreading speed of an epidemic in different periods or even different epidemics. Highlights: The study aims to model the time to the next epicenters during the pandemic COVID-19. The study introduces the time to epicenter as a criterion to study of spreading speed of an epidemic in different periods or compare different epidemics. The study deals with the number of cumulative confirmed cases at the time that a region become epicenter. The study proposes the Poisson process as the model to describe the number of epicenters. The study suggests that exponential distribution can model the time to event for the epicenters of COVID-19.Copyright © 2023 International Biometric Society-Chinese Region.

Time-to-Event Analysis of Factors Influencing Delay in Discharge from a Subacute Complex Discharge Unit during the First Year of the Pandemic (2020) in an Irish Tertiary Centre Hospital.

Our study aimed to analyse delaying factors amongst patients with a length of stay (LOS) > 15 days during the COVID-19 pandemic using time-to-event analysis. A total of 390 patients were admitted between March 2020-February 2021 to the subacute complex discharge unit in St James's Hospital: 326 (83.6%) were >65 years of age and 233 (59.7%) were female. The median (IQR) age was 79 (70-86) years with a median (IQR) of 19.4 (10-41) days. A total of 237 (60.7%) events were uncensored, with LOS > 15 days, of which 138 (58.2%) were female and 124 (52.32%) had >4 comorbidities; 153 (39.2%) were censored into LOS ≤ 15 days, and death occurred in 19 (4.8%). Kaplan-Meier's plot compared factors causing a delay in discharge to the single factors: age, gender, and multimorbidity. A multivariate Cox regression analysis adjusted to age, gender, and multimorbidity predicted factors affecting LOS. Further research is required to explore multimorbidity as a risk factor for mortality in patients with prolonged LOS within a complex discharge unit and target gender-specific frailty measures to achieve high-quality patient management.

A retrospective time-to-event analysis on COVID-19 patients using integrated medicine (Allopathy and Siddha medicine)

Background: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a single-stranded RNA virus that causes SARS. The virus was originally called SARS-CoV-2 named officially by the World Health Organization as COVID-19 and a global health emergency. Aim: The aim of this article is to assess various clinical findings, comorbidities, and outcomes including periods of hospital stay of asymptomatic/symptomatic COVID-19 patients admitted at the Government Medical College, Omandurar, Chennai District of Tamil Nadu state and those who were under Integrated Medical care of hydroxychloroquine and Kabasura Kudineer. Materials and Methods: This study is a retrospective cross-sectional study on 162 asymptomatic/symptomatic laboratory-confirmed SARS-CoV-2-infected patients from both sexes and all age groups, admitted to the Government Medical College, Omandurar, Chennai District of Tamil Nadu state, India from March 30, 2020 to April 30, 2020. Results: SARS-CoV-2 confirmed 162 patients: 114 (70.4%) males and 48 (29.6%) females participated in the study. About 37.7% of the participants belonged to the age group of 26–40 years, 32.7% belonged to the age group 41–60 years, 16% belonged to the age group 13–25 years, 9.9% belonged to the age group greater than 60 years, and 3.7% belonged to the age group of 0–12 years. Around 16.7% have comorbidities, whereas the remaining 83.3% were free from comorbid conditions. All participants were given Kabasura Kudineer, but the treatment of allopathy medicine intake varies from person to person depending on their health status and disease severity. Abdominal computed tomography scan or chest X-rays of 24 (14.8%) patients were done during the study period. Two patients died and two were referred to another hospital, remaining 97.5% completely recovered from the viral infection. Admission days varied from 1 to 23 days. The overall median length of stay was 16 days based on this study. Conclusion: This study reveals that 97.5% of the patients were completely recovered from the viral infection and both average and median hospital stay is 16 days.

Time-to-event modeling for hospital length of stay prediction for COVID-19 patients.

Providing timely patient care while maintaining optimal resource utilization is one of the central operational challenges hospitals have been facing throughout the pandemic. Hospital length of stay (LOS) is an important indicator of hospital efficiency, quality of patient care, and operational resilience. Numerous researchers have developed regression or classification models to predict LOS. However, conventional models suffer from the lack of capability to make use of typically censored clinical data. We propose to use time-to-event modeling techniques, also known as survival analysis, to predict the LOS for patients based on individualized information collected from multiple sources. The performance of six proposed survival models is evaluated and compared based on clinical data from COVID-19 patients.

Handling death as an intercurrent event in time to recovery analysis in COVID-19 treatment clinical trials.

In clinical trials with the objective to evaluate the treatment effect on time to recovery, such as investigational trials on therapies for COVID-19 hospitalized patients, the patients may face a mortality risk that competes with the opportunity to recover (e.g., be discharged from the hospital). Therefore, an appropriate analytical strategy to account for death is particularly important due to its potential impact on the estimation of the treatment effect. To address this challenge, we conducted a thorough evaluation and comparison of nine survival analysis methods with different strategies to account for death, including standard survival analysis methods with different censoring strategies and competing risk analysis methods. We report results of a comprehensive simulation study that employed design parameters commonly seen in COVID-19 trials and case studies using reconstructed data from a published COVID-19 clinical trial. Our research results demonstrate that, when there is a moderate to large proportion of patients who died before observing their recovery, competing risk analyses and survival analyses with the strategy to censor death at the maximum follow-up timepoint would be able to better detect a treatment effect on recovery than the standard survival analysis that treat death as a non-informative censoring event. The aim of this research is to raise awareness of the importance of handling death appropriately in the time-to-recovery analysis when planning current and future COVID-19 treatment trials.

Survival analysis: A primer for the clinician scientists.

Survival analysis is a collection of statistical procedures employed on time-to-event data. The outcome variable of interest is time until an event occurs. Conventionally, it dealt with death as the event, but it can handle any event occurring in an individual like disease, relapse from remission, and recovery. Survival data describe the length of time from a time of origin to an endpoint of interest. By time, we mean years, months, weeks, or days from the beginning of being enrolled in the study. The major limitation of time-to-event data is the possibility of an event not occurring in all the subjects during a specific study period. In addition, some of the study subjects may leave the study prematurely. Such situations lead to what is called censored observations as complete information is not available for these subjects. Life table and Kaplan-Meier techniques are employed to obtain the descriptive measures of survival times. The main objectives of survival analysis include analysis of patterns of time-to-event data, evaluating reasons why data may be censored, comparing the survival curves, and assessing the relationship of explanatory variables to survival time. Survival analysis also offers different regression models that accommodate any number of covariates (categorical or continuous) and produces adjusted hazard ratios for individual factor.

Communicating About Mortality in Health Decision Support: 'What and Why and When, and How and Where and Who'.

The Covid-19 pandemic has only accelerated the need and desire to deal more openly with mortality, because the effect on survival is central to the comprehensive assessment of harms and benefits needed to meet a 'reasonable patient' legal standard. Taking the view that this requirement is best met through a multi-criterial decision support tool, we offer our preferred answers to the questions of What should be communicated about mortality in the tool, and How, given preferred answers to Who for, Who by, Why, When, and Where. Summary measures, including unrestricted Life Expectancy and Restricted Mean Survival Time are found to be reductionist and relative, and not as easy to understand and communicate as often asserted. Full lifetime absolute survival curves should be presented, even if they cannot be 'evidence-based' beyond trial follow-up limits, along with equivalent measures for other criteria in the (necessarily) multi-criterial decision. A decision support tool should relieve the reasonable person of the resulting calculation burden.

Estimands, Handling of Missing Data and Impact on Assumed Effect Size and Power in Pivotal COVID-19 Treatment Trials.

Estimands play an important role for aligning study objectives, study design and analyses through a precise definition of the quantity of interest. For COVID-19 studies, apart from intercurrent events, high volume of missing data has been observed. We explore their impact on several estimands through a synthetic COVID-19 data generated from a discrete-time multi-state model. We compare estimators of these estimands based on their ability to closely match the true response rates and retain assumed power. The final choice of the estimand then needs to be aligned with clinically meaningful quantities of interest to patients, clinicians, regulators and payers.

A predictive paradigm for COVID-19 prognosis based on the longitudinal measure of biomarkers.

Novel coronavirus disease 2019 (COVID-19) is an emerging, rapidly evolving crisis, and the ability to predict prognosis for individual COVID-19 patient is important for guiding treatment. Laboratory examinations were repeatedly measured during hospitalization for COVID-19 patients, which provide the possibility for the individualized early prediction of prognosis. However, previous studies mainly focused on risk prediction based on laboratory measurements at one time point, ignoring disease progression and changes of biomarkers over time. By using historical regression trees (HTREEs), a novel machine learning method, and joint modeling technique, we modeled the longitudinal trajectories of laboratory biomarkers and made dynamically predictions on individual prognosis for 1997 COVID-19 patients. In the discovery phase, based on 358 COVID-19 patients admitted between 10 January and 18 February 2020 from Tongji Hospital, HTREE model identified a set of important variables including 14 prognostic biomarkers. With the trajectories of those biomarkers through 5-day, 10-day and 15-day, the joint model had a good performance in discriminating the survived and deceased COVID-19 patients (mean AUCs of 88.81, 84.81 and 85.62% for the discovery set). The predictive model was successfully validated in two independent datasets (mean AUCs of 87.61, 87.55 and 87.03% for validation the first dataset including 112 patients, 94.97, 95.78 and 94.63% for the second validation dataset including 1527 patients, respectively). In conclusion, our study identified important biomarkers associated with the prognosis of COVID-19 patients, characterized the time-to-event process and obtained dynamic predictions at the individual level.

Interstates of Infection: Preliminary Investigations of Human Mobility Patterns in the COVID-19 Pandemic.

PURPOSE: The COVID-19 pandemic has illuminated various heterogeneities between urban and rural environments in public health. The SARS-CoV-2 virus initially spread into the United States from international ports of entry and into urban population centers, like New York City. Over the course of the pandemic, cases emerged in more rural areas, implicating issues of transportation and mobility. Additionally, many rural areas developed into national hotspots of prevalence and transmission. Our aim was to investigate the preliminary impacts of road travel on the spread of COVID-19. This investigation has implications for future public health mitigation efforts and travel restrictions in the United States. METHODS: County-level COVID-19 data were analyzed for spatiotemporal patterns in time-to-event distributions using animated choropleth maps. Data were obtained from The New York Times and the Bureau of the Census. The arrival event was estimated by examining the number of days between the first reported national case (January 21, 2020) and the date that each county attained a given prevalence rate. Of the 3108 coterminous US counties, 2887 were included in the analyses. Data reflect cases accumulated between January 21, 2020, and May 17, 2020. FINDINGS: Animations revealed that COVID-19 was transmitted along the path of interstates. Quantitative results indicated rural-urban differences in the estimated arrival time of COVID-19. Counties that are intersected by interstates had an earlier arrival than non-intersected counties. The arrival time difference was the greatest in the most rural counties and implicates road travel as a factor of transmission into rural communities. CONCLUSION: Human mobility via road travel introduced COVID-19 into more rural communities. Interstate travel restrictions and road travel restrictions would have supported stronger mitigation efforts during the earlier stages of the COVID-19 pandemic and reduced transmission via network contact.

Statistical Analysis of Clinical COVID-19 Data: A Concise Overview of Lessons Learned, Common Errors and How to Avoid Them.

By definition, in-hospital patient data are restricted to the time between hospital admission and discharge (alive or dead). For hospitalised cases of COVID-19, a number of events during hospitalization are of interest regarding the influence of risk factors on the likelihood of experiencing these events. The same is true for predicting times from hospital admission of COVID-19 patients to intensive care or from start of ventilation (invasive or non-invasive) to extubation. This logical restriction of the data to the period of hospitalisation is associated with a substantial risk that inappropriate methods are used for analysis. Here, we briefly discuss the most common types of bias which can occur when analysing in-hospital COVID-19 data.

Assessing the Impact of COVID-19 on the Clinical Trial Objective and Analysis of Oncology Clinical Trials-Application of the Estimand Framework.

Abstract-Coronavirus disease 2019 (COVID-19) outbreak has rapidly evolved into a global pandemic. The impact of COVID-19 on patient journeys in oncology represents a new risk to interpretation of trial results and its broad applicability for future clinical practice. We identify key intercurrent events (ICEs) that may occur due to COVID-19 in oncology clinical trials with a focus on time-to-event endpoints and discuss considerations pertaining to the other estimand attributes introduced in the ICH E9 addendum. We propose strategies to handle COVID-19 related ICEs, depending on their relationship with malignancy and treatment and the interpretability of data after them. We argue that the clinical trial objective from a world without COVID-19 pandemic remains valid. The estimand framework provides a common language to discuss the impact of COVID-19 in a structured and transparent manner. This demonstrates that the applicability of the framework may even go beyond what it was initially intended for.