Risk of COVID-19 breakthrough infection and hospitalization in individuals with comorbidities.

BACKGROUND: The successful development of multiple COVID-19 vaccines has led to a global vaccination effort to reduce severe COVID-19 infection and mortality. However, the effectiveness of the COVID-19 vaccines wane over time leading to breakthrough infections where vaccinated individuals experience a COVID-19 infection. Here we estimate the risks of breakthrough infection and subsequent hospitalization in individuals with common comorbidities who had completed an initial vaccination series. METHODS: Our study population included vaccinated patients between January 1, 2021 to March 31, 2022 who are present in the Truveta patient population. Models were developed to describe 1) time from completing primary vaccination series till breakthrough infection; and 2) if a patient was hospitalized within 14 days of breakthrough infection. We adjusted for age, race, ethnicity, sex, and year-month of vaccination. RESULTS: Of 1,218,630 patients in the Truveta Platform who had completed an initial vaccination sequence between January 1, 2021 and March 31, 2022, 2.85, 3.42, 2.75, and 2.88 percent of patients with CKD, chronic lung disease, diabetes, or are in an immunocompromised state experienced breakthrough infection, respectively, compared to 1.46 percent of the population without any of these four comorbidities. We found an increased risk of breakthrough infection and subsequent hospitalization in individuals with any of the four comorbidities when compared to individuals without these four comorbidities. CONCLUSIONS: Vaccinated individuals with any of the studied comorbidities experienced an increased risk of breakthrough COVID-19 infection and subsequent hospitalizations compared to the people without any of the studied comorbidities. Individuals with immunocompromising conditions and chronic lung disease were most at risk of breakthrough infection, while people with CKD were most at risk of hospitalization following breakthrough infection. Patients with multiple comorbidities have an even greater risk of breakthrough infection or hospitalization compared to patients with none of the studied comorbidities. Individuals with common comorbidities should remain vigilant against infection even if vaccinated.

Evaluation of Phenolic Content Variability along with Antioxidant, Antimicrobial, and Cytotoxic Potential of Selected Traditional Medicinal Plants from India.

Plants have been used since ancient times as an important source of biologically active substances. The aim of the present study was to investigate the phytochemical constituents (flavonoids and phenolics), antioxidant potential, cytotoxicity against HepG2 (human hepato carcinoma) cancer cell lines, and the antimicrobial activity of the methanol extract of selected traditional medicinal plants collected from Mizoram, India. A number of phenolic compounds were detected using HPLC-DAD-ESI-TOF-MS, mainly Luteolin, Kaempferol, Myricetin, Gallic Acid, Quercetin and Rutin, some of which have been described for the first time in the selected plants. The total phenolic and flavonoid contents showed high variation ranging from 4.44 to 181.91 µg of Gallic Acid equivalent per milligram DW (GAE/mg DW) and 3.17 to 102.2 µg of Quercetin/mg, respectively. The antioxidant capacity was determined by DPPH (IC50 values ranges from 34.22 to 131.4 µg/mL), ABTS (IC50 values ranges from 24.08 to 513.4 µg/mL), and reducing power assays. Antimicrobial activity was assayed against gram positive (Staphylococcus aureus), gram negative (Escherichia coli, Pseudomonas aeruginosa), and yeast (Candida albicans) demonstrating that the methanol extracts of some plants were efficacious antimicrobial agents. Additionally, cytotoxicity was assessed on human hepato carcinoma (HepG2) cancer cell lines and found that the extracts of Albizia lebbeck, Dillenia indica, and Bombax ceiba significantly decreased the cell viability at low concentrations with IC50 values of 24.03, 25.09, and 29.66 µg/mL, respectively. This is the first report of detection of phenolic compounds along with antimicrobial, antioxidant and cytotoxic potential of selected medicinal plants from India, which indicates that these plants might be valuable source for human and animal health.

Early detection of sepsis in the emergency department using Dynamic Bayesian Networks.

Sepsis is a systemic inflammatory state due to an infection, and is associated with very high mortality and morbidity. Early diagnosis and prompt antibiotic and supportive therapy is associated with improved outcomes. Our objective was to detect the presence of sepsis soon after the patient visits the emergency department. We used Dynamic Bayesian Networks, a temporal probabilistic technique to model a system whose state changes over time. We built, trained and tested the model using data of 3,100 patients admitted to the emergency department, and measured the accuracy of detecting sepsis using data collected within the first 3 hours, 6 hours, 12 hours and 24 hours after admission. The area under the curve was 0.911, 0.915, 0.937 and 0.944 respectively. We describe the data, data preparation techniques, model, results, various statistical measures and the limitations of our experiments. We also briefly discuss techniques to improve accuracy, and the generalizability of our methods to other diseases.

Modeling Glucose Homeostasis and Insulin Dosing in an Intensive Care Unit using Dynamic Bayesian Networks.

Adequate control of serum glucose in critically ill patients is a complex problem requiring continuous monitoring and intervention, which have a direct effect on clinical outcomes. Understanding temporal relationships can help to improve our knowledge of complex disease processes and their response to treatment. We discuss a Dynamic Bayesian Network (DBN) model that we created using the open-source Projeny toolkit to represent various clinical variables and the temporal and atemporal relationships underlying insulin and glucose homeostasis. We evaluated this model by comparing the DBN model's insulin dose predictions against those of a rule-based protocol (eProtocol-insulin) currently used in the ICU. The results suggest that the DBN model's predictions are as effective as or better than those of the rule-based protocol. The limitations of our methods are discussed, with a brief note on their generalizability.

Collaborative authoring of biomedical terminologies using a semantic Wiki.

To be broadly useful, biomedical terminologies need to capture the knowledge and expertise of multiple experts, research groups and end users. Consequently, the construction of such a terminology requires collaboration among multiple participants. This paper summarizes several terminology projects which employ a collaborative authoring mode in development, along with the various tools used by these projects to support collaboration, such as mailing list, issue tracking system, client-side GUI software and Wiki system. We then discuss the essential functional requirements for a collaborative authoring tool in the context of terminology development, and analyze the inherent features of Semantic Wiki that enable it to be a competent tool used in this type of effort. To demonstrate, we describe a prototype system of collaborative authoring for health care terminologies built upon the Semantic Wiki technology. Finally, we discuss the potential issues that might be associated with this open Semantic Wiki platform.

Vocabulary metadata service for terminology servers to handle variations in design of various biomedical terminologies.

Terminology servers provide access to various biomedical terminologies for authoring and maintenance, in addition to automated use by various clinical and administrative applications and interface engines. HL7 Common Terminology Services (CTS) intends to provide standard interfaces for accessing terminology services within and across organizations. However, variations in designs of different terminologies make this hard to achieve. We describe a novel solution to this issue, and hope to integrate this into the HL7 CTS2 standard.

Practical issues in using SNOMED CT as a reference terminology.

SNOMED CT was created by the merger of SNOMED RT (Reference Terminology) and Read Codes Version 3 (also known as Clinical Terms Version 3). SNOMED CT is considered to be among the most extensive and comprehensive biomedical vocabularies available today. It is considered for use as the Reference Terminology of various institutions. We review the adequacy of SNOMED CT as a Reference Terminology and discuss the issues in its use as such. We discuss issues with content coverage of various clinical domains, data integrity and validity, and the update frequency of SNOMED CT, and why SNOMED CT alone is not adequate to serve as the Reference Terminology of a healthcare organization.

Best practices for distribution of controlled medical terminologies.

A health care organization wanting to implement electronic capture and storage needs to implement several controlled and legacy terminologies. The organization has control over its legacy terminologies, its data format and update process. But it has to deal with various aspects of third-party controlled medical terminologies which are out of its control. These aspects include distribution formats, notification of updates and so on. We discuss the issues we have encountered while trying to use several biomedical terminologies that are created and distributed by third-party organizations, and recommend best practices for distribution.

Generalizability of hybrid search algorithms to map multiple biomedical vocabulary domains.

Hybrid text matching algorithms similar to those used for DNA sequencing were developed by 3M Health Information Systems to map a noisy legacy codeset to the 3M Healthcare Data Dictionary (3M HDD). Applying these techniques to map other biomedical vocabularies was briefly introduced in an earlier paper describing the algorithms. We now present results from successfully utilizing them to map different vocabularies across multiple biomedical domains, proving their generalizability.

Applying hybrid algorithms for text matching to automated biomedical vocabulary mapping.

Several biomedical vocabularies are often used by clinical applications due to their different domain(s) of coverage, intended use, etc. Mapping them to a reference terminology is essential for inter-systems interoperability. Manual vocabulary mapping is labor-intensive and allows room for inconsistencies. It requires manual searching for synonyms, abbreviation expansions, variations, etc., placing additional burden on the mappers. Furthermore, local vocabularies may use non-standard words and abbreviations, posing additional problems. However, much of this process can be automated to provide decision-support, allowing mappers to focus on steps that absolutely need their expertise. We developed hybrid algorithms comprising of rules, permutations, sequence alignment and cost algorithms that utilize the UMLS SPECIALIST Lexicon, a custom knowledgebase and a search engine to automatically find probable matches, allowing mappers to select the best match from this list. We discuss the techniques, results from assisting to map a local codeset, and scope for generalizability.