Artificial intelligence and machine learning in clinical trial design and application

In 2020, COVID-19 changed how health care was approached both in the United States and globally. In the early phases, the vast majority of energy and attention was devoted to containing the pandemic and treating the infected. Toward the end of 2020, that attention expanded to vaccinating people across the globe. What was not being considered at the time were challenges to future health and clinical trials that power new treatments for COVID-19 and non-COVID-19 treatments. © 2023 Elsevier Inc. All rights reserved.

A global federated real-world data and analytics platform for research.

Objective: This article describes a scalable, performant, sustainable global network of electronic health record data for biomedical and clinical research. Materials and Methods: TriNetX has created a technology platform characterized by a conservative security and governance model that facilitates collaboration and cooperation between industry participants, such as pharmaceutical companies and contract research organizations, and academic and community-based healthcare organizations (HCOs). HCOs participate on the network in return for access to a suite of analytics capabilities, large networks of de-identified data, and more sponsored trial opportunities. Industry participants provide the financial resources to support, expand, and improve the technology platform in return for access to network data, which provides increased efficiencies in clinical trial design and deployment. Results: TriNetX is a growing global network, expanding from 55 HCOs and 7 countries in 2017 to over 220 HCOs and 30 countries in 2022. Over 19 000 sponsored clinical trial opportunities have been initiated through the TriNetX network. There have been over 350 peer-reviewed scientific publications based on the network's data. Conclusions: The continued growth of the TriNetX network and its yield of clinical trial collaborations and published studies indicates that this academic-industry structure is a safe, proven, sustainable path for building and maintaining research-centric data networks.

Bottom-up and top-down paradigms of artificial intelligence research approaches to healthcare data science using growing real-world big data.

OBJECTIVES: As the real-world electronic health record (EHR) data continue to grow exponentially, novel methodologies involving artificial intelligence (AI) are becoming increasingly applied to enable efficient data-driven learning and, ultimately, to advance healthcare. Our objective is to provide readers with an understanding of evolving computational methods and help in deciding on methods to pursue. TARGET AUDIENCE: The sheer diversity of existing methods presents a challenge for health scientists who are beginning to apply computational methods to their research. Therefore, this tutorial is aimed at scientists working with EHR data who are early entrants into the field of applying AI methodologies. SCOPE: This manuscript describes the diverse and growing AI research approaches in healthcare data science and categorizes them into 2 distinct paradigms, the bottom-up and top-down paradigms to provide health scientists venturing into artificial intelligent research with an understanding of the evolving computational methods and help in deciding on methods to pursue through the lens of real-world healthcare data.

Vaccination in Patients with Inflammatory Rheumatic Diseases: Preliminary Results of an International, Real-World Life Panlar's Register

Objectives: To evaluate vaccination among patients with inflammatory rheumatic diseases initiating disease-modifying antirheumatic drugs (DMARD) Methods: Data from the real-world life PANLAR's register of consecutive patients diagnosed with RA, PsA, and axSpa (2010 ACR-EULAR /2006 CASPAR-2009 ASAS) from Dec 2021 to Dec 2022 were analyzed. Prevalence of recommended vaccinations were compared between different inflammatory rheumatic diseases. Categorical variables were expressed as %. Tables were analyzed with chi2 or Fisher tests, continuous variables (median, IQR)with the Kruskal-Wallis test, according with the variables type. A p value <=0.05 was considered significant. Result(s): 608 patients were included. Among patients with rheumatoid arthritis (RA), psoriatic arthritis (PsA), and axial Spondyloarthritis (axSpA) are presented in the table. RA and axSpA seemed to have lower vaccination rate of pneumococcal vaccines than PsA. (p = 0.045 for conjugate anti pneumococcal vaccine in RA vs PsA). A large percentage of the population was vaccinated against COVID-19. There was a high rate of influenza vaccination in all three diseases. Conclusion(s): In Latin America, anti-pneumococcal vaccination is low, especially in patients with RA and axSpA. For other vaccines there was an acceptable level of vaccination without differences between diseases.

Utilization of Health Resources During the Coronavirus Pandemic in Israel: Meuhedet Health Services' Real Data Analysis and Future Conclusions

Aim: This study aimed to examine the utilization of health resources during the first year of the COVID-19 pandemic in Israel through the analysis of Meuhedet Health Services' real-world database. Background(s): The history of COVID-19 in Israel comprises three waves: from February to May 2020, from May to November 2020, and from November 2020 to April 2021. Restrictions imposed on the Israeli population included travel limitations and even lockdowns. Meuhedet Health Services, the third largest health management organization in Israel, manages all its medical data through computerized electronic files and has collected various types of health services data from 2018 to 2020. This paper compared the consumption of Meuhedet Health Services over two consecutive years of the pandemic using a real-world database. Method(s): Electronic medical records from primary care physicians, laboratory tests, hospitalization medical histories, treatments in hospitals and institutes, visits to and treatments by community physicians, and prescriptions and medical equipment consumption were collected from 2018 to 2020. This research used aggregated, non-personalized, and decoded data from a cohort of insured individuals, and the research was approved by all the relevant institutional Helsinki Committees. The data analysis compared the corresponding data in a chosen month of the year with the data in the same month of the previous year. The differences were then scaled by the data corresponding to the month of the previous year, and the result was multiplied by 100 and plotted. To analyze drug consumption, we used the fixed price of every drug in a year multiplied by the difference in consumption of the drug in question between the month of the current year and the same month of the previous year, multiplied by 100. Result(s): A significant decrease was noted in hospitalization days, general hospital outpatient clinic visits, general hospital emergency room visits, and total numbers of visits to community physicians during the first lockdown in the first wave of the pandemic in comparison to 2019. At the end of the lockdown, however, a compensatory increase was noted in all services. In terms of drug consumption, the data showed no differences in the effects of the different waves. Our findings revealed that the first wave of COVID-19 was a shock, with a significant reduction in the consumption of health services, but this decrease attenuated with the second wave due to immediate management interventions and safety rules implemented in hospitals and clinics. Conclusion(s): People shun medical services during a fast-spreading epidemic that causes significant mortality. Since new variants of COVID-19 could be part of our lives for the next few years, we should learn how to continue living with the pandemic and develop alternative medical services to maintain healthy states. Digitization, remote services, telemedicine, and home care, including home hospitalization, should be part of the health services to cope with pandemic situations.Copyright © 2023 Klang et al.

Geographical differences in the use of oral corticosteroids in patients with severe asthma in Spain: heat map based on existing databases analyses.

BACKGROUND: Although there are currently alternative treatments to the long-term use of oral corticosteroids (OCS) in severe asthma, recent studies show excessive use depending on geography and differences in medical practice. The objective of the study was to describe the differences in OCS use for severe asthma across the Spanish geography. METHODS: This is a real-world study using existing databases (year 2019): longitudinal patient database (EMR), based on electronic medical records, and database of pharmacological consumption (Sell-in) in basic healthcare areas. With EMR, the percentage of OCS prescriptions corresponding to patients with severe asthma (ICD-9 "asthma" and prescription of biological treatment and/or high dose of inhaled corticosteroids/long-acting inhaled ß2 agonists) was calculated. This percentage was transferred to the OCS consumption of each basic healthcare area as reported in the Sell-in database and a national heat map was created. The estimation of OCS use in patients with severe asthma per 100,000 inhabitants for each region was calculated by grouping basic healthcare areas and the mean OCS use per patient for different regions in Spain was also estimated. RESULTS: Patients with severe asthma in Spain were mostly female (69.6%), with a mean age (SD) of 57.6 years (18.01). Median time (Pc25-Pc75) since asthma diagnosis was 83.1 months (34.65-131.56). Of all patients with OCS prescriptions in 2019 identified in EMR, 4.4% corresponded to patients with severe asthma. Regions with the highest OCS use were Asturias, Andalucía, and Galicia, whereas those with the lowest use were Navarra, Baleares, Madrid and País Vasco. The mean OCS use per patient with severe asthma in 2019 throughout Spain was 1099.85 mg per patient, ranging from 782.99 mg in Navarra to 1432.64 in Asturias. CONCLUSIONS: There are geographical differences between Spanish regions with respect to the use of OCS in patients with severe asthma. The national mean consumption of OCS per patient with severe asthma and year is above the limits that indicate good asthma control.

Use of population health data to promote equitable recruitment for a primary care practice implementation trial addressing unhealthy alcohol use.

Background: Recruiting underrepresented people and communities in research is essential for generalizable findings. Ensuring representative participants can be particularly challenging for practice-level dissemination and implementation trials. Novel use of real-world data about practices and the communities they serve could promote more equitable and inclusive recruitment. Methods: We used a comprehensive primary care clinician and practice database, the Virginia All-Payers Claims Database, and the HealthLandscape Virginia mapping tool with community-level socio-ecological information to prospectively inform practice recruitment for a study to help primary care better screen and counsel for unhealthy alcohol use. Throughout recruitment, we measured how similar study practices were to primary care on average, mapped where practices' patients lived, and iteratively adapted our recruitment strategies. Results: In response to practice and community data, we adapted our recruitment strategy three times; first leveraging relationships with residency graduates, then a health system and professional organization approach, followed by a community-targeted approach, and a concluding approach using all three approaches. We enrolled 76 practices whose patients live in 97.3% (1844 of 1907) of Virginia's census tracts. Our overall patient sample had similar demographics to the state for race (21.7% vs 20.0% Black), ethnicity (9.5% vs 10.2% Hispanic), insurance status (6.4% vs 8.0% uninsured), and education (26.0% vs 32.5% high school graduate or less). Each practice recruitment approach uniquely included different communities and patients. Discussion: Data about primary care practices and the communities they serve can prospectively inform research recruitment of practices to yield more representative and inclusive patient cohorts for participation.

Racial Disparities in Pulse Oximeter Device Inaccuracy and Estimated Clinical Impact on COVID-19 Treatment Course.

Arterial blood oxygen saturation measured by pulse oximetry (SpO2) may be differentially less accurate for people with darker skin pigmentation, which could potentially affect COVID-19 treatment course. We analyzed pulse oximeter accuracy and association with COVID-19 treatment outcomes using electronic health record (EHR) data from Sutter Health, a large, mixed-payer, integrated healthcare delivery system in northern California, United States (US). We analyzed two cohorts: (1) 43,753 concurrent arterial blood gas (ABG) oxygen saturation (SaO2)/SpO2 measurement pairs taken January 2020-February 2021 for Non-Hispanic white (NHW) or Non-Hispanic Black/African American (NHB) adults, and (2) 8,735 adults who went to the emergency department (ED) with COVID-19 July 2020-February 2021. Pulse oximetry systematically overestimated blood oxygenation by 1% more in NHB individuals than in NHW individuals. For people with COVID-19, this was associated with lower admission probability (-3.1 percentage-points), dexamethasone treatment (-3.1 percentage-points), and supplemental oxygen treatment (-4.5 percentage-points), as well as increased time-to-treatment: +37.2 minutes before dexamethasone initiation and +278.5 minutes before initiation of supplemental oxygen. These results call for additional investigation of pulse oximeters, and suggest that current guidelines for development, testing, and calibration of these devices should be revisited, investigated, and revised.

Worldwide Trends in Registering Real-World Studies at ClinicalTrials.gov: A Cross-Sectional Analysis.

Objective: The purpose of this study was to characterize real-world studies (RWSs) registered at ClinicalTrials.gov to help investigators better conduct relevant research in clinical practice. Methods: A retrospective analysis of 944 studies was performed on February 28, 2023. Results: A total of 944 studies were included. The included studies involved a total of 48 countries. China was the leading country in terms of the total number of registered studies (37.9%, 358), followed by the United States (19.7%, 186). Regarding intervention type, 42.4% (400) of the studies involved drugs, and only 9.1% (86) of the studies involved devices. Only 8.5% (80) of the studies mentioned both the detailed study design type and data source in the "Brief Summary". A total of 49.4% (466) of studies had a sample size of 500 participants and above. Overall, 63% (595) of the studies were single-center studies. A total of 213 conditions were covered in the included studies. One-third of the studies (32.7%, 309) involved neoplasms (or tumors). China and the United States were very different regarding the study of different conditions. Conclusion: Although the pandemic has provided new opportunities for RWSs, the rigor of scientific research still needs to be emphasized. Special attention needs to be given to the correct and comprehensive description of the study design in the Brief Summary of registered studies, thereby promoting communication and understanding. In addition, deficiencies in ClinicalTrials.gov registration data remain prominent.

The impact of COVID-19 on a large pragmatic clinical trial embedded in primary care.

INTRODUCTION: The COVID-19 pandemic had significant impact on clinical care and clinical trial operations, but the impact on decentralized pragmatic trials is unclear. The Diuretic Comparison Project (DCP) is a Point-of Care (POC) pragmatic trial testing whether chlorthalidone is superior to hydrochlorothiazide in preventing major cardiovascular (CV) events and non-cancer death. DCP utilized telephone consent, data collection from the electronic health record and Medicare, forwent study visits, and limited provider commitment beyond usual care. We assessed the impact of COVID-19 on recruitment, follow-up, data collection, and outcome ascertainment in DCP. METHODS: We compared data from two 8-month periods: Pre-Pandemic (July 2019-February 2020) and Mid-Pandemic (July 2020-February 2021). Consent and randomization rates, diuretic adherence, blood pressure (BP) and electrolyte follow-up rates, records of CV events, hospitalization, and death rates were compared. RESULTS: Providers participated at a lower rate mid-pandemic (65%) than pre-pandemic (71%), but more patients were contacted (7622 vs. 5363) and consented (3718 vs. 3048) mid-pandemic than pre-pandemic. Patients refilled medications and remained on their randomized diuretic equally (90%) in both periods. Overall, rates of BP, electrolyte measurements, and hospitalizations decreased mid-pandemic while deaths increased. CONCLUSIONS: While recruitment, enrollment, and adherence did not suffer during the pandemic, documented blood pressure checks and laboratory evaluations decreased, likely due to fewer in-person visits. VA hospitalizations decreased, despite a considerable number of COVID-related hospitalizations. This suggests changes in clinical care during the pandemic, but the limited impact on DCP's operations during a global pandemic is an important strength of POC trials. CLINICAL TRIAL REGISTRATION: NCT02185417.

A Path to Real-World Evidence in Critical Care Using Open-Source Data Harmonization Tools.

COVID-19 highlighted the need for use of real-world data (RWD) in critical care as a near real-time resource for clinical, research, and policy efforts. Analysis of RWD is gaining momentum and can generate important evidence for policy makers and regulators. Extracting high quality RWD from electronic health records (EHRs) requires sophisticated infrastructure and dedicated resources. We sought to customize freely available public tools, supporting all phases of data harmonization, from data quality assessments to de-identification procedures, and generation of robust, data science ready RWD from EHRs. These data are made available to clinicians and researchers through CURE ID, a free platform which facilitates access to case reports of challenging clinical cases and repurposed treatments hosted by the National Center for Advancing Translational Sciences/National Institutes of Health in partnership with the Food and Drug Administration. This commentary describes the partnership, rationale, process, use case, impact in critical care, and future directions for this collaborative effort.

Decentralized clinical trials and rare diseases: a Drug Information Association Innovative Design Scientific Working Group (DIA-IDSWG) perspective.

BACKGROUND: Traditional clinical trials require tests and procedures that are administered in centralized clinical research sites, which are beyond the standard of care that patients receive for their rare and chronic diseases. The limited number of rare disease patients scattered around the world makes it particularly challenging to recruit participants and conduct these traditional clinical trials. MAIN BODY: Participating in clinical research can be burdensome, especially for children, the elderly, physically and cognitively impaired individuals who require transportation and caregiver assistance, or patients who live in remote locations or cannot afford transportation. In recent years, there is an increasing need to consider Decentralized Clinical Trials (DCT) as a participant-centric approach that uses new technologies and innovative procedures for interaction with participants in the comfort of their home. CONCLUSION: This paper discusses the planning and conduct of DCTs, which can increase the quality of trials with a specific focus on rare diseases.

Augmented Intelligence for Clinical Discovery in Hypertensive Disorders of Pregnancy Using Outlier Analysis.

Objectives Clinical discoveries are heralded by observing unique and unusual clinical cases. The effort of identifying such cases rests on the shoulders of busy clinicians. We assess the feasibility and applicability of an augmented intelligence framework to accelerate the rate of clinical discovery in preeclampsia and hypertensive disorders of pregnancy-an area that has seen little change in its clinical management. Methods We conducted a retrospective exploratory outlier analysis of participants enrolled in the folic acid clinical trial (FACT, N=2,301) and the Ottawa and Kingston birth cohort (OaK, N=8,085). We applied two outlier analysis methods: extreme misclassification contextual outlier and isolation forest point outlier. The extreme misclassification contextual outlier is based on a random forest predictive model for the outcome of preeclampsia in FACT and hypertensive disorder of pregnancy in OaK. We defined outliers in the extreme misclassification approach as mislabelled observations with a confidence level of more than 90%. Within the isolation forest approach, we defined outliers as observations with an average path length z score less or equal to -3, or more or equal to 3. Content experts reviewed the identified outliers and determined if they represented a potential novelty that could conceivably lead to a clinical discovery. Results In the FACT study, we identified 19 outliers using the isolation forest algorithm and 13 outliers using the random forest extreme misclassification approach. We determined that three (15.8%) and 10 (76.9%) were potential novelties, respectively. Out of 8,085 participants in the OaK study, we identified 172 outliers using the isolation forest algorithm and 98 outliers using the random forest extreme misclassification approach; four (2.3%) and 32 (32.7%), respectively, were potential novelties. Overall, the outlier analysis part of the augmented intelligence framework identified a total of 302 outliers. These were subsequently reviewed by content experts, representing the human part of the augmented intelligence framework. The clinical review determined that 49 of the 302 outliers represented potential novelties.  Conclusions Augmented intelligence using extreme misclassification outlier analysis is a feasible and applicable approach for accelerating the rate of clinical discoveries. The use of an extreme misclassification contextual outlier analysis approach has resulted in a higher proportion of potential novelties than using the more traditional point outlier isolation forest approach. This finding was consistent in both the clinical trial and real-world cohort study data. Using augmented intelligence through outlier analysis has the potential to speed up the process of identifying potential clinical discoveries. This approach can be replicated across clinical disciplines and could exist within electronic medical records systems to automatically identify outliers within clinical notes to clinical experts.

[Exploration and example interpretation of real-world herbal prescription classification based on similarity matching algorithm].

In observational studies, herbal prescriptions are usually studied in the form of "similar prescriptions". At present, the classification of prescriptions is mainly based on clinical experience judgment, but there are some problems in manual judgment, such as lack of unified criteria, labor consumption, and difficulty in verification. In the construction of a database of integrated traditional Chinese and western medicine for the treatment of coronavirus disease 2019(COVID-19), our research group tried to classify real-world herbal prescriptions using a similarity matching algorithm. The main steps include 78 target prescriptions are determined in advance; four levels of importance labeling shall be carried out for the drugs of each target prescription; the combination, format conversion, and standardization of drug names of the prescriptions to be identified in the herbal medicine database; calculate the similarity between the prescriptions to be identified and each target prescription one by one; prescription discrimination is performed based on the preset criteria; remove the name of the prescriptions with "large prescriptions cover the small". Through the similarity matching algorithm, 87.49% of the real prescriptions in the herbal medicine database of this study can be identified, which preliminarily proves that this method can complete the classification of herbal prescriptions. However, this method does not consider the influence of herbal dosage on the results, and there is no recognized standard for the weight of drug importance and criteria, so there are some limitations, which need to be further explored and improved in future research.

Prevention and cost control in the German healthcare sector: Routine data analysis and effective health communication as advantages for health insurance competitiveness?

Background: The German healthcare system is struggling with increasing costs. Besides the current extra burden due to the corona pandemic, the vast majority of Germans pursue unhealthy lifestyles which will lead to additional morbidity and costs in the future. Objectives: This contribution sketches out an idea, on how analyses on claims data from the Statutory Health Insurance (SHI) in Germany may contribute to better usage of preventive health services to counteract the onset and progress of morbidities and hence ensure stable premium income from the insured. Effective health communication may further enable demand for preventive measures. Materials and methods: An idea is developed and discussed in which, in addition to the existing possibilities of the SHI to work towards preventive health behavior, results of secondary data analysis may be used for preventive measures and behavior. Results and conclusions: A machine-learning-based analysis is the core of a class of prediction models for prevention of illnesses. The models exploit the information from routine data and provide recommendations for prevention services, which in turn may be promoted to the insured via targeted, tailored, and personalized communication, e.g., via mHealth apps. The high potential for cost reductions as well as the possibilities to exploit them via data analytics provide a promising perspective for sustained cost control in the healthcare sector. © 2023, The Author(s).

SARS-CoV-2 infection and COVID19 vaccination across eight immune-mediated inflammatory disorders: A prospective, real-life Belgian cohort study - the BELCOMID study.

Background: The risks and impact of COVID19 disease and vaccination in patients with Immune Mediated Inflammatory Diseases (IMID) remain incompletely understood. IMID patients and particularly patients receiving immunosuppressive treatment were excluded from the original, registrational phase-3 COVID19 vaccination efficacy and safety trials. Real-world observational data can help to fill this gap in knowledge. The BELCOMID study aims to explore the interaction between IMIDs, immune-modulating treatment modalities and SARS-CoV-2 infection and vaccination in a real-life patient cohort. Methods: A multidisciplinary, prospective, observational cohort study was set up. Consecutive patients with IMIDs of the gut, joints and skin followed at two high-volume referral centers were invited. Both patients under conventional treatment or targeted immune modulating therapies were included. Patient data and serological samples were collected at 3 predefined periods (before COVID19 vaccination, before booster vaccination, after booster vaccination). Primary endpoints were positive PCR-test and SARS-CoV-2 serology reflecting previous SARS-CoV-2 infection or vaccination. Associations with IMID treatment modality and IMID disease activity were assessed. Results of the first two inclusion periods (before booster vaccination) are reported. Results: At the first inclusion period data was assessed of 2165 IMID-patients before COVID19 vaccination. At the second inclusion period, data of 2065 patients was collected of whom 1547 had received complete baseline COVID19 vaccination and 222 were partially vaccinated. SARS-CoV-2 infection rate remained low in both groups. No significant increase in IMID flare-up rate was noted in patients with prior SARS-CoV-2 infection. Multiple logistic regression analyses did not show a significant influence of IMID-treatment modality or IMID activity on SARS-CoV-2 infection risk (based on PCR positivity or N-serology). Patients treated with conventional immunomodulators, systemic steroids, and patients on advanced therapies such as biologics or small molecules, had reduced S-antibody seroconversion. S-antibody response was also lower in patients without prior SARS-CoV-2 infection and in active smokers. A subset of patients (4.1%) had no S- nor N-antibody seroconversion following complete baseline vaccination. Conclusion: The BELCOMID study results confirm the benign course of COVID19 infection and vaccination in a large real-life IMID-population. However, our results underscore the need for repeated vaccination and smoking cessation in patients with IMIDs treated with immune-modulating therapies or systemic steroids during the pandemic.

Development of a COVID-19 vaccine effectiveness and safety assessment system in Japan: The VENUS study.

BACKGROUND: There are currently no COVID-19 vaccine assessment systems in Japan that allow for the active surveillance of both vaccinated and unvaccinated persons. Herein, we describe the development of Japan's first COVID-19 vaccine effectiveness and safety assessment system with active surveillance capabilities. METHODS: The Vaccine Effectiveness, Networking, and Universal Safety (VENUS) Study was developed as a multi-source database that links four data types at the individual resident level: Basic Resident Register (base population information), Vaccination Record System (vaccination-related information), Health Center Real-time Information-sharing System on COVID-19 (HER-SYS; information on COVID-19 occurrence), and health care claims data (information on diagnoses, hospitalizations, diagnostic tests, and treatments). These data were obtained from four municipalities. Individual residents were linked across the data types using five matching algorithms based on names, birth dates, and sex; the data were anonymized after linkage. To ascertain the viability of the VENUS Study's database for COVID-19 vaccine assessments, we examined the trends in COVID-19 vaccinations, COVID-19 cases, and polymerase chain reaction (PCR) test numbers. We also evaluated the linkage rates across the data types. RESULTS: Our multi-source database was able to monitor COVID-19 vaccinations, COVID-19 cases, and PCR test numbers throughout the pandemic. Using the five algorithms, the data linkage rates between the COVID-19 occurrence information in the HER-SYS and the Basic Resident Register ranged from 85·4% to 91·7%. CONCLUSION: If used judiciously with an understanding of each data source's characteristics, the VENUS Study can provide a viable data platform that facilitates active surveillance and comparative analyses for population-based research on COVID-19 vaccine effectiveness and safety in Japan.

Factors Associated with Self-Reported Post/Long-COVID-A Real-World Data Study.

Evidence suggests that Post/Long-COVID (PLC) is associated with a reduced health-related quality of life, however little knowledge exists on the risk factors that contribute to PLC. The objective of this prospective real-world data study was to evaluate factors associated with PLC using national online survey data. Adjusted multivariable regression analyses were performed using the software R. Between 14 April and 15 June 2021, 99 registered individuals reported to have suffered from PLC symptoms and the most common PLC symptoms reported were fatigue, dyspnoea, decreased strength, hyposmia, and memory loss. The odds of individuals suffering from COVID-19-associated anxiety, hyposmia, or heart palpitations developing PLC were eight times (OR 8.28, 95% CI 1.43−47.85, p < 0.01), five times (OR 4.74, 95% CI 1.59−14.12, p < 0.005), or three times (OR 2.62, 95% CI 1.72−3.99, p < 0.01) higher, respectively, than of those who had not experienced these symptoms. Individuals who experienced fatigue while having COVID-19 were seven times more likely to develop PLC fatigue than those who had not (OR 6.52, 95% CI: 4.29−9.91, p < 0.0001). Our findings revealed that 13% of the individuals who had previously suffered from COVID-19 subsequently reported having PLC. Furthermore, COVID-19-associated anxiety, hyposmia, heart palpitations, and fatigue were, among others, significant determinants for the development of PLC symptoms. Hyposmia has not previously been reported as an independent predictive factor for PLC. We suggest closely monitoring patients with COVID-19-induced fatigue, heart palpitations, and anxiety, as these symptoms may be predictors of PLC symptoms, including fatigue.

Data quality considerations for evaluating COVID-19 treatments using real world data: learnings from the National COVID Cohort Collaborative (N3C).

BACKGROUND: Multi-institution electronic health records (EHR) are a rich source of real world data (RWD) for generating real world evidence (RWE) regarding the utilization, benefits and harms of medical interventions. They provide access to clinical data from large pooled patient populations in addition to laboratory measurements unavailable in insurance claims-based data. However, secondary use of these data for research requires specialized knowledge and careful evaluation of data quality and completeness. We discuss data quality assessments undertaken during the conduct of prep-to-research, focusing on the investigation of treatment safety and effectiveness. METHODS: Using the National COVID Cohort Collaborative (N3C) enclave, we defined a patient population using criteria typical in non-interventional inpatient drug effectiveness studies. We present the challenges encountered when constructing this dataset, beginning with an examination of data quality across data partners. We then discuss the methods and best practices used to operationalize several important study elements: exposure to treatment, baseline health comorbidities, and key outcomes of interest. RESULTS: We share our experiences and lessons learned when working with heterogeneous EHR data from over 65 healthcare institutions and 4 common data models. We discuss six key areas of data variability and quality. (1) The specific EHR data elements captured from a site can vary depending on source data model and practice. (2) Data missingness remains a significant issue. (3) Drug exposures can be recorded at different levels and may not contain route of administration or dosage information. (4) Reconstruction of continuous drug exposure intervals may not always be possible. (5) EHR discontinuity is a major concern for capturing history of prior treatment and comorbidities. Lastly, (6) access to EHR data alone limits the potential outcomes which can be used in studies. CONCLUSIONS: The creation of large scale centralized multi-site EHR databases such as N3C enables a wide range of research aimed at better understanding treatments and health impacts of many conditions including COVID-19. As with all observational research, it is important that research teams engage with appropriate domain experts to understand the data in order to define research questions that are both clinically important and feasible to address using these real world data.

Changing patterns and clinical outcomes of hospitalized patients with COVID-19 severe pneumonia treated with remdesivir according to vaccination status: results from a real-world retrospective study.

Since the beginning of Coronavirus Disease 2019 (COVID-19) pandemic, many drugs have been purposed for the treatment of severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2). Remdesivir emerged as an encouraging antiviral drug for patients with documented severe COVID-19-related pneumonia. Although several studies about remdesivir effectiveness exist, no study investigated the effect of the combination of remdesivir with the vaccination status. The aim of this study was to assess whether the administration of remdesivir could show some differences in terms of clinical outcomes in patients vaccinated against SARS-CoV-2 versus those who were not. The primary outcome was the in-hospital mortality. The secondary outcomes were 30-days mortality, the need for ICU admission and for oxygen supplementation. This is a retrospective cohort study including all consecutive adult patients hospitalized for severe COVID-19 at the Padua University Hospital (Italy), between September 1st, 2020, and January 31st, 2022, and who received a 5-days course of remdesivir. A total of 708 patients were included, 467 (66%) were male, and the median age was 67 (IQR: 56-79) years. To better estimate the outcomes of interest, a propensity score weighted approach was implemented for vaccination status. A total of 605/708 patients (85.4%) did not complete the vaccination schedule. In-hospital mortality rate was 5.1% (n = 36), with no statistically significant difference between the unvaccinated (n=29, 4.8%) and vaccinated (n=7, 6.8%; p=0.4) patients. After propensity score matching, mortality between the two groups remained similar. However, both the need for ICU and oxygen supplementation were significantly lower in the vaccinated group. Our finding suggests that a complete vaccination course could have an impact in reducing the need for transfer in ICU and for high-flow therapy in moderate-to-severe COVID-19 patients treated with remdesivir.