Clinical Data Sources Research Articles

FUPEC study, a prospective open-cohort on severe pre-eclampsia and cardiovascular risk factors based in the Netherlands

PurposeThe FUPEC (Follow-Up Pre-EClampsia) study aims to investigate the presence and development of cardiovascular risk factors, cardiovascular disease, as well as cardiovascular health following a pregnancy complicated by severe pre-eclampsia.ParticipantsThe...

A StarGAN and transformer-based hybrid classification-regression model for multi-institution VMAT patient-specific quality assurance.

The field of artificial intelligence (AI)-based patient-specific quality assurance (PSQA) for volumetric modulated arc therapy (VMAT) faces challenges in terms of developing general models across institutions due to the prevalence of multi-institution data collection and multivariate heterogeneity. Building a general model that is capable of handling diverse multi-institution data is critical for enabling large-scale integration and analysis. This study aims to develop a star generative adversarial network (StarGAN) and transformer-based hybrid classification-regression PSQA framework to address unification of heterogeneous data from different institutions. A StarGAN and transformer-based hybrid classification-regression model was developed as a general PSQA framework to predict gamma passing rates (GPRs) and classify quality assurance (QA) results as "Pass" or "Fail" at multiple institutions. A total of 1815 VMAT plans were collected from eight institutions to develop the general PSQA framework and perform clinical commissioning and implementation. Among them, 20 independent clinical plans from each of eight institutions, for a total of 160 plans, were used for the clinical commissioning, and 205 new clinical plans from eight institutions were used for clinical implementation. For the 3%/3, 3%/2, and 2%/2mm gamma criteria, the sensitivity of the proposed PSQA framework with pretraining was 90.13%, 92.03%, and 95.84%, respectively, while the specificity was 76.01%, 76.12%, and 85.34%, respectively. The mean absolute errors (MAEs) of the proposed PSQA framework with pretraining were 1.36%, 2.37%, and 3.96%, respectively, while the root-mean-square errors (RMSEs) were 2.31%, 3.89%, and 5.17%, respectively. The results demonstrated visible improvement at multiple institutions. For clinical commissioning, the deviations between the predicted and measured results were all within 3% for 3%/3 and 3%/2mm at eight institutions. For clinical implementation, all failure plans were correctly identified by the proposed PSQA framework. The general PSQA framework enables diverse clinical data sources to be handled to achieve enhanced model performance and generalizability, and provides a solution to the unification of heterogeneous data from different institutions to construct robust QA models. This approach can be clinically deployed for VMAT QA.

Identify the most appropriate imputation method for handling missing values in clinical structured datasets: a systematic review

Background and objectivesComprehending the research dataset is crucial for obtaining reliable and valid outcomes. Health analysts must have a deep comprehension of the data being analyzed. This comprehension allows them to suggest practical solutions for handling missing data, in a clinical data source. Accurate handling of missing values is critical for producing precise estimates and making informed decisions, especially in crucial areas like clinical research. With data's increasing diversity and complexity, numerous scholars have developed a range of imputation techniques. To address this, we conducted a systematic review to introduce various imputation techniques based on tabular dataset characteristics, including the mechanism, pattern, and ratio of missingness, to identify the most appropriate imputation methods in the healthcare field.Materials and methodsWe searched four information databases namely PubMed, Web of Science, Scopus, and IEEE Xplore, for articles published up to September 20, 2023, that discussed imputation methods for addressing missing values in a clinically structured dataset. Our investigation of selected articles focused on four key aspects: the mechanism, pattern, ratio of missingness, and various imputation strategies. By synthesizing insights from these perspectives, we constructed an evidence map to recommend suitable imputation methods for handling missing values in a tabular dataset.ResultsOut of 2955 articles, 58 were included in the analysis. The findings from the development of the evidence map, based on the structure of the missing values and the types of imputation methods used in the extracted items from these studies, revealed that 45% of the studies employed conventional statistical methods, 31% utilized machine learning and deep learning methods, and 24% applied hybrid imputation techniques for handling missing values.ConclusionConsidering the structure and characteristics of missing values in a clinical dataset is essential for choosing the most appropriate data imputation technique, especially within conventional statistical methods. Accurately estimating missing values to reflect reality enhances the likelihood of obtaining high-quality and reusable data, contributing significantly to precise medical decision-making processes. Performing this review study creates a guideline for choosing the most appropriate imputation methods in data preprocessing stages to perform analytical processes on structured clinical datasets.

A national Australian Congenital Heart Disease registry; methods and initial results

BackgroundAlthough several National Data Registries for Congenital Heart Disease (CHD) exist, few are comprehensive and contemporary. A National Australian CHD Registry has been developed that aims to redress this by creating the first comprehensive data collection for CHD children and adults, initially across Australia. MethodsWe defined and collected a minimum dataset of demographics, diagnoses, and procedures from people with CHD presenting at participating quaternary CHD services Australia-wide. Data were collected from a range of clinical data sources. Diagnoses and procedures were standardised to the European Paediatric Congenital Code – Short List. Methodological limitations were carefully documented. ResultsFrom 8 participating institutions, an initial 359,084 patient records were assessed for eligibility and 68,234 unique individuals with structural CHD have been included in the current dataset. There were 20,395 (30 %) people with mild CHD, 25,157 (37 %) with moderate CHD, and 13,530 (20 %) with severe CHD (6 % unknown complexity). The most common diagnoses were Ventricular Septal Defect (16,781, 25 %), Atrial Septal Defect (6,607, 10 %), Aortic Valve Disorders (5516 8 %), Coarctation of the Aorta (5,321, 8 %), Tetralogy of Fallot (4,489, 7 %), Transposition of the Great Arteries (4,009, 6 %). ConclusionThe data presented here represents the most comprehensive cohort collected for the Australian CHD population thus far and is comparable with the largest contemporary CHD registries around the world. This Registry represents a key resource for improved understanding of the CHD population and will drive better care and outcomes for people living with CHD.

A Comparison of Active Pharmacovigilance Strategies Used to Monitor Adverse Events to Antiviral Agents: A Systematic Review.

The safety of antiviral agents in real-world clinical settings is crucial, as pre-marketing studies often do not capture all adverse events (AE). Active pharmacovigilance strategies are essential for detecting and characterising these AE comprehensively. The aim of this study was to identify and characterise active pharmacovigilance strategies used in real-world clinical settings for patients under systemic antiviral agents, focusing on the frequency of AE and the clinical data sources used. We conducted a systematic review by searching three electronic bibliographic databases targeting observational prospective active pharmacovigilance studies, phase IV clinical trials for post-marketing safety surveillance, and interventional studies assessing active pharmacovigilance strategies, focusing on individuals exposed to systemic antiviral agents. We included 36 primary studies, predominantly using Drug Event Monitoring (DEM), with a minority employing sentinel sites and registries. Human immunodeficiency virus (HIV) was the most common condition, with the majority using DEM. Within the DEM, there was a wide range of incidences of patients experiencing at least one AE, and most of these studies used one or two data sources. Sentinel site studies were less common, with two on hepatitis C virus (HCV) and one on HIV, each relying on one or two data sources. The single study using a registry focusing on HIV therapy reported using just one data source. Patient interviews were the most common data source, followed by medical records and laboratory tests. The quality of the studies was considered 'good' in 18/36, 'fair' in 1/36, and 'poor' in 17/36 studies. DEM was the predominant pharmacovigilance strategy, employing multiple data sources, and appears to increase the likelihood of detecting higher AE incidence. Establishing such a framework would facilitate a more detailed and consistent approach across different studies and settings.

The potential immunological mechanisms of sepsis.

Sepsis is described as a life-threatening organ dysfunction and a heterogeneous syndrome that is a leading cause of morbidity and mortality in intensive care settings. Severe sepsis could incite an uncontrollable surge of inflammatory cytokines, and the host immune system's immunosuppression could respond to counter excessive inflammatory responses, characterized by the accumulated anti-inflammatory cytokines, impaired function of immune cells, over-proliferation of myeloid-derived suppressor cells and regulatory T cells, depletion of immune effector cells by different means of death, etc. In this review, we delve into the underlying pathological mechanisms of sepsis, emphasizing both the hyperinflammatory phase and the associated immunosuppression. We offer an in-depth exploration of the critical mechanisms underlying sepsis, spanning from individual immune cells to a holistic organ perspective, and further down to the epigenetic and metabolic reprogramming. Furthermore, we outline the strengths of artificial intelligence in analyzing extensive datasets pertaining to septic patients, showcasing how classifiers trained on various clinical data sources can identify distinct sepsis phenotypes and thus to guide personalized therapy strategies for the management of sepsis. Additionally, we provide a comprehensive summary of recent, reliable biomarkers for hyperinflammatory and immunosuppressive states, facilitating more precise and expedited diagnosis of sepsis.

Domain base dynamic convolution and distance map guidance for anterior mediastinal lesion segmentation

The automatic segmentation of anterior mediastinal lesions in the enhanced CT image plays a significant role in clinical diagnosis. Due to the low incidence rate of anterior mediastinal disease, the data for training segmentation network may come from different hospitals, different equipment, and different operators, which leads to lower generalization and robustness of the trained network. This paper proposes an anterior mediastinal lesion segmentation method based on domain-adaptive dynamic convolution and distance map guidance. Considering the diversity of clinical data sources and equipment differences, the domain prior knowledge is added to the global average-pooling layer of the network, and the dynamic convolution forms specific network parameters for specific domains to enhance domain-specific information. Meanwhile, the Singed Distance Map (SDM) head is involved in the network according to the shape, size and position of the anterior mediastinal lesions to constraint the boundary and assist in localization. Ablation experiments demonstrate that the distance map can effectively constrain the segmentation target and reduce false positives. Both qualitative and quantitative experimental results indicate that our method can achieve more accurate anterior mediastinal lesion segmentation with greater generalization ability. Our approach achieves an overall Dice coefficient of 88.33%, which is 2.31% higher than existing state-of-the-art methods, and has achieved good performance in terms of ASSD evaluation for segmentation edges.

Correlation Aware Relevance-Based Semantic Index for Clinical Big Data Repository.

In this paper, we focus on indexing mechanisms for unstructured clinical big integrated data repository systems. Clinical data is unstructured and heterogeneous, which comes in different files and formats. Accessing data efficiently and effectively are critical challenges. Traditional indexing mechanisms are difficult to apply on unstructured data, especially by identifying correlation information between clinical data elements. In this research work, we developed a correlation-aware relevance-based index that retrieves clinical data by fetching most relevant cases efficiently. In our previous work, we designed a methodology that categorizes medical data based on the semantics of data elements and merges them into an integrated repository. We developed a data integration system for medical data sources that combines heterogeneous medical data and provides access to knowledge-based database repositories to different users. In this research work, we designed an indexing system using semantic tags extracted from clinical data sources and medical ontologies that retrieves relevant data from database repositories and speeds up the process of data retrieval. Our objective is to provide an integrated biomedical database repository that can be used by radiologists as a reference, or for patient care, or by researchers. In this paper, we focus on designing a technique that performs data processing for data integration, learn the semantic properties of data elements, and develop a correlation-aware topic index that facilitates efficient data retrieval. We generated semantic tags by identifying key elements from integrated clinical cases using topic modeling techniques. We investigated a technique that identifies tags for merged categories and provides an index to fetch data from an integrated database repository. We developed a topic coherence matrix that shows how well a topic is supported by a corpus from clinical cases and medical ontologies. We were able to find more relevant results using an annotation index from an integrated database repository, and there was a 61% increase in a recall. We evaluated results with the help of experts and compared them with naive index (index with all terms from the corpus). Our approach improved data retrieval quality by providing most relevant results and reduced data retrieval time as we applied correlation-aware index on an integrated data repository. Topic indexing approach proposed in this research work identifies tags based on a correlation between different data elements, improves data retrieval time, and provides most relevant cases as an outcome of this system.

Could the mitotic count improve personalized prognosis in melanoma patients?

A number of studies have indicated that the mitotic rate may be a predictive factor for poor prognosis in melanoma patients. The aim of this study was to investigate whether the mitotic rate is associated with other prognostic clinical and anatomopathological characteristics. After adjusting for other anatomopathological characteristics, we then verified the prognostic value of the number of mitoses, determining in which population subgroup this variable may have greater prognostic significance on 3-year mortality. The Veneto Cancer Registry (Registro Tumori del Veneto-RTV), a high-resolution population-based dataset covering the regional population of approximately 4.9 million residents, served as the clinical data source for the analysis. Inclusion criteria included all incident cases of invasive cutaneous malignant melanoma recorded in the RTV in 2015 (1,050 cases) and 2017 (1,205 cases) for which the number of mitoses was available. Mitotic classes were represented by Kaplan-Meier curves for short-term overall survival. Cox regression calculated hazard ratios in multivariable models to evaluate the independent prognostic role of different mitotic rate cut-offs. The results indicate that the mitotic rate is associated with other survival prognostic factors: the variables comprising the TNM stage (e.g., tumor thickness, ulceration, lymph node status and presence of metastasis) and the characteristics that are not included in the TNM stage (e.g., age, site of tumor, type of morphology, growth pattern and TIL). Moreover, this study demonstrated that, even after adjusting for these prognostic factors, mitoses per mm2 are associated with higher mortality, particularly in T2 patients. In conclusion, these findings revealed the need to include the mitotic rate in the histological diagnosis because it correlates with the prognosis as an independent factor. The mitotic rate can be used to develop a personalized medicine approach in the treatment and follow-up monitoring of melanoma patients.

Organizational Breast Cancer Data Mart: A Solution for Assessing Outcomes of Imaging and Treatment.

In the United States, a comprehensive national breast cancer registry (CR) does not exist. Thus, care and coverage decisions are based on data from population subsets, other countries, or models. We report a prototype real-world research data mart to assess mortality, morbidity, and costs for breast cancer diagnosis and treatment. With institutional review board approval and Health Insurance Portability and Accountability Act (HIPPA) compliance, a multidisciplinary clinical and research data warehouse (RDW) expert group curated demographic, risk, imaging, pathology, treatment, and outcome data from the electronic health records (EHR), radiology (RIS), and CR for patients having breast imaging and/or a diagnosis of breast cancer in our institution from January 1, 2004, to December 31, 2020. Domains were defined by prebuilt views to extract data denormalized according to requirements from the existing RDW using an export, transform, load pattern. Data dictionaries were included. Structured query language was used for data cleaning. Five-hundred eighty-nine elements (EHR 311, RIS 211, and CR 67) were mapped to 27 domains; all, except one containing CR elements, had cancer and noncancer cohort views, resulting in a total of 53 views (average 12 elements/view; range, 4-67). EHR and RIS queries returned 497,218 patients with 2,967,364 imaging examinations and associated visit details. Cancer biology, treatment, and outcome details for 15,619 breast cancer cases were imported from the CR of our primary breast care facility for this prototype mart. Institutional real-world data marts enable comprehensive understanding of care outcomes within an organization. As clinical data sources become increasingly structured, such marts may be an important source for future interinstitution analysis and potentially an opportunity to create robust real-world results that could be used to support evidence-based national policy and care decisions for breast cancer.

Emergence and spread of SARS-CoV-2 variants of concern in Canada: a retrospective analysis from clinical and wastewater data

BackgroundThe spread of SARS-CoV-2 has been studied at unprecedented levels worldwide. In jurisdictions where molecular analysis was performed on large scales, the emergence and competition of numerous SARS-CoV-2lineages have been observed in near real-time. Lineage identification, traditionally performed from clinical samples, can also be determined by sampling wastewater from sewersheds serving populations of interest. Variants of concern (VOCs) and SARS-CoV-2 lineages associated with increased transmissibility and/or severity are of particular interest.MethodHere, we consider clinical and wastewater data sources to assess the emergence and spread of VOCs in Canada retrospectively.ResultsWe show that, overall, wastewater-based VOC identification provides similar insights to the surveillance based on clinical samples. Based on clinical data, we observed synchrony in VOC introduction as well as similar emergence speeds across most Canadian provinces despite the large geographical size of the country and differences in provincial public health measures.ConclusionIn particular, it took approximately four months for VOC Alpha and Delta to contribute to half of the incidence. In contrast, VOC Omicron achieved the same contribution in less than one month. This study provides significant benchmarks to enhance planning for future VOCs, and to some extent for future pandemics caused by other pathogens, by quantifying the rate of SARS-CoV-2 VOCs invasion in Canada.

The relationship between post-traumatic stress disorder and gastrointestinal disease in United States Military Veterans.

This study examined the relationship between gastrointestinal disease and post-traumatic stress disorder in U.S. military Veterans. Based on literature and clinical practice data sources from the U.S. Veterans Administration, gastrointestinal disease and post-traumatic stress disorder were hypothesized to be positively correlated in Veterans. This study aimed to determine the frequency with which gastrointestinal disease and post-traumatic stress disorder are diagnosed comorbidities, a diagnosis of gastrointestinal disease accompanies a diagnosis of post-traumatic stress disorder, and a diagnosis of post-traumatic stress disorder accompanies a diagnosis of a gastrointestinal disease. The methodology was a retrospective, correlational design using data collected from the U.S. Department of Veterans Affairs patient database. The results were that post-traumatic stress disorder is bi-directionally correlated with the gastrointestinal diseases of gastroesophageal reflux disease, peptic ulcer disease, functional dyspepsia, Crohn's disease, diverticular disease, irritable bowel syndrome, and the symptoms of constipation and nausea/vomiting within Veterans who served during wartime periods. The study also found that post-traumatic stress disorder is not correlated with ulcerative colitis in Veterans. The conclusions are that clinicians who see a presentation of post-traumatic stress disorder should be screening for gastrointestinal disease, while primary care and gastroenterology providers treating gastrointestinal disease should be screening for a history of trauma, as improved diagnosis rates may lead to improved treatment.

The Quebec Congenital Heart Disease Registry: A Model of Prospective Databank to Facilitate Research in Congenital Cardiology

BackgroundA national registry of congenital heart disease (CHD) would facilitate project initiation, decrease costs, increase statistical power, and avoid duplication. Establishing such registries poses numerous challenges, but the current Canadian research ecosystem in CHD is well positioned to meet them. We assessed the feasibility of building a province-wide CHD registry by automatically identifying people with CHD and extracting their native cardiac anatomy from multiple clinical data sources, without the need for manual data entry. MethodsWe designed a CHD registry of all fetuses and children with at least 1 echocardiographic report confirming CHD since 2000. We interfaced the registry with several clinical and echocardiography data sources from all paediatric cardiology programmes in Québec. ResultsWe extracted 885,287 echocardiogram reports and 70,121 clinical records. We identified CHD in 43,452 children and 4682 fetuses. There were 1128 (2.3%) cases with files in multiple institutions, and patients with more complex CHD were 3 times more likely to be seen in more than 1 institution. So far, the registry has been used to build and link CHD cohorts for 7 distinct projects. ConclusionsWe demonstrated the feasibility of a baseline CHD registry in Québec without the need for manual data entry, in which other CHD research projects could be nested. This could serve as a blueprint to expand the registry and to develop an integrated approach where data gathered in caring for patients with CHD serve as data layers that incrementally contribute to a national cohort, for which data remain easily accessible and usable.

Electronic health record impact on data quality: An integrated review

Few studies have assessed the impact of Electronic Health Record (EHR) on data quality and how it has impacted timeliness, accuracy, availability, integrity, consistency and uniqueness of health data. Assessing the current gaps between electronic health record utilisation in relation to enhancing data quality can help to identify and recommend key measures to improve healthcare delivery. This review examines the impact of electronic health record on data quality, and proposes recommendations desirable to improving healthcare delivery. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and Integrative Review Frameworks (IRF) were adapted as a sign post for the study. These frameworks guided literature search, data analysis and presentation of findings. Boolean AND/OR operators were combined with search terms to increase relevant search results. Six (6) electronic databases that were used for literature search are Scopus, Web of Science, Science Direct, PubMed, ProQuest and Google Scholar. Searched articles were between 2012 to 2022, with the exception of two articles on the Technology Acceptance Model (TAM) published in 1986 and 1975. EHR implementation impact key health outcomes and contribute to timely and complete data submission. Implementation of Electronic Health Systems (EHS) embellish consistency, completeness, accuracy and integrity of clinical data significantly through embedded interoperability solutions. By reducing the scope of manual reviews, integrating a technology that automates data integrity and quality checks into EHR may potentially lessen the strain of maintaining data quality. Our findings depict that interoperability of clinical data sources will ensure consistent, timely and available health data when using EHR. The automated data integrity and consistency checks in the EHR improve data quality. The study contributes to advancing the theory of Technology Acceptance Model (TAM) in electronic health records and its impact on data quality in healthcare delivery.

Multiple Myeloma Disease Model to Predict PFS Outcomes across Lines of Therapy, an Analysis Based on Commpass Observational Study

Introduction Multiple Myeloma (MM) remains incurable despite advances in treatment. Model-based approaches based on tumor dynamics have successfully been used to optimize drug-development in oncology. Model-derived tumor dynamics also called tumor growth inhibition (TGI) metrics capture the treatment effect as tumor biomarkers and are predictive of the overall survival (OS) benefit, based on TGI-OS models, this mathematical framework has been shown to be drug-independent in most cases. M-protein is measured longitudinally during clinical trials, it reflects tumor burden and is one of the criteria used to assess clinical response according to the International Myeloma Working Group Uniform Response Criteria. A drug-independent link between M-protein dynamic metrics and OS was previously established and confirmed for refractory MM patients. The primary endpoint in MM clinical trials is usually progression free survival (PFS). New immunotherapies are currently in development and there is a need for early and robust drug development decision making. Therefore there is strong interest to expand this modeling framework to leverage M-protein dynamics to predict PFS outcomes. Methods CoMMpass (https://www.themmrf.org/finding-a-cure/personalized-treatment-approaches/) is a prospective, longitudinal, observational study in newly diagnosed symptomatic patients with multiple myeloma. It represents a unique source of clinical data to investigate the link between M-protein dynamic and survival outcomes across lines of therapy and treatments. Individual M-protein and PFS Data from 804 patients within the CoMMpass study were used. Given patients are enrolled at their initial treatment regimen and observed over 8 years, they can receive several lines of therapy. Due to the characteristics of the modeling framework, allowing to analyze only one line of therapy for each patient, the latest line of therapy for each subject was selected for which they had at least 2 M-protein assessments available. M-protein dynamic metrics (M-protein ratio to baseline at different time points, time to growth, growth rate, shrinkage rate) derived using an empirical bi-exponential model, 42 prognostic factors and potential treatment effects were first tested in a univariate analysis using a Cox proportional hazards regression model. Statistically significant covariates in the univariate analysis (p<0.05) were included in a full multivariate parametric distribution survival model. A backward deletion stepwise procedure was performed (p<0.01) to retain covariates in the final PFS model. Model qualification was performed using posterior predictive checks based on Kaplan Meier plots. Results Among 804 patients, 490 patients received only 1 line of therapy (LoT), 157 received 2 LoTs, 93 received 3 LoTs, 31 received 4 LoTs,18 received 5 LoTs, 10 received 6 LoTs, finally 7 patients got up to 7 to 11 LoTs. PFS distribution followed a log-normal distribution. Table 1 shows the parameter estimates of the PFS model. Among all tested M-protein dynamic metrics and prognostic factors, the logarithmic value of the growth rate (log(KG)), in red in Table 1, was found to be the best predictor of PFS. Patients with slower growth rate, who received a transplantation, with higher albumin and lower lactate dehydrogenase values had longer PFS. Those findings are consistent with previous modeling works conducted in MM. Also, the PFS model was confirmed to be drug-independent. As shown in Figure 1, prediction intervals derived from model-based simulations well captured observed PFS data for each line of treatment although the line of therapy is not included as a predictor in the PFS model. Similar results were obtained across treatments. In a next step, external validations assessing the model performance to reproduce Phase 3 clinical trial outcomes will be performed. Conclusions A model linking M-protein dynamic to PFS in MM was developed across a wide range of lines of therapy and treatments. The model can support drug development in MM: enabling early predictions of clinical trial outcomes and inform study design e.g. by simulating the probability of success of upcoming pivotal clinical trials.

Unexpected and Rare Sites of Metastasis in Oncologic Patients.

Case studies of rare oncologic metastases are an important source of clinical data for health care professionals and researchers. While infrequent, the knowledge base and clinical recommendations derived from such cases aid in advancements in the field. As such, we aim to add five cases to the growing body of literature. The first two male patients, aged 69 and 73, were diagnosed with colon adenocarcinoma, suspected to be a second primary prostate carcinoma, following positron emission tomography-computer tomography (PET-CT). This suspicion was ruled out by prostatectomy and histopathological investigations, which instead found adenocarcinoma of colonic origin. The next two male patients, ages 63 and 68, were diagnosed, respectively, with metastatic pancreatic adenocarcinoma with cardiac metastases and metastatic melanoma with distant metastases to the pancreas. The final patient was a 73-year-old male diagnosed with metastatic breast cancer after a radiological investigation of suspected renal cell carcinoma.

Using administrative healthcare data to evaluate drug repurposing opportunities for cancer: the possibility of using beta-blockers to treat breast cancer.

Introduction: Cancer registries and hospital electronic medical records are commonly used to investigate drug repurposing candidates for cancer. However, administrative data are often more accessible than data from cancer registries and medical records. Therefore, we evaluated if administrative data could be used to evaluate drug repurposing for cancer by conducting an example study on the association between beta-blocker use and breast cancer mortality. Methods: A retrospective cohort study of women aged ≥50years with incident breast cancer was conducted using a linked dataset with statewide hospital admission data and nationwide medication claims data. Women receiving beta blockers and first-line anti-hypertensives prior to and at diagnosis were compared. Breast cancer molecular subtypes and metastasis status were inferred by algorithms from commonly prescribed breast cancer antineoplastics and hospitalization diagnosis codes, respectively. Subdistribution hazard ratios (sHR) and corresponding 95% confidence intervals (CIs) for breast cancer mortality were estimated using Fine and Gray's competing risk models adjusted for age, Charlson comorbidity index, congestive heart failure, myocardial infraction, molecular subtype, presence of metastasis at diagnosis, and breast cancer surgery. Results: 2,758 women were hospitalized for incident breast cancer. 604 received beta-blockers and 1,387 received first-line antihypertensives. In total, 154 breast cancer deaths were identified over a median follow-up time of 2.7years. We found no significant association between use of any beta-blocker and breast-cancer mortality (sHR 0.86, 95%CI 0.58-1.28), or when stratified by beta-blocker type (non-selective, sHR 0.42, 95%CI 0.14-1.25; selective, sHR 0.95, 95%CI 0.63-1.43). Results were not significant when stratified by molecular subtypes (e.g., triple negative breast cancer (TNBC), any beta blocker, sHR 0.16, 95%CI 0.02-1.51). Discussion: It is possible to use administrative data to explore drug repurposing opportunities. Although non-significant, an indication of an association was found for the TNBC subtype, which aligns with previous studies using registry data. Future studies with larger sample size, longer follow-up are required to confirm the association, and linkage to clinical data sources are required to validate our methodologies.

Clinical validation of in-hospital stroke diagnosis

ObjectiveIn-hospital stroke cases occur during hospitalization for another diagnosis and reflect a clinically distinct cohort from community-onset stroke. The objective was to validate the diagnostic accuracy of in-hospital stroke identification in stroke audit data at a large teaching hospital. MethodsA retrospective clinical validation of in-hospital stroke diagnoses from two linked data sources was completed for a 2-year period from January 1st 2020 to December 31st 2021. The linked data sources include the Hospital Inpatient Enquiry system which assigns coded stroke diagnoses at discharge and/or the local stroke audit coordinators who work clinically in stroke teams and input additional specific clinical data. Diagnostic sensitivity, specificity and the level of agreement using an unweighted Cohen's Kappa were calculated. ResultsThere were 597 strokes admitted during the 2-year period. The median age was 72 years and 55% occurred in males. In total, 88 cases of in-hospital stroke were clinically validated yielding an in-hospital stroke rate of 15%. The clinical audit coordinator identified in-hospital stroke with higher sensitivity (86%; 95% CI 77%-93%) whereas the coding process was more specific at 96% (95% CI 85% to 99%). Levels of agreement with the clinically validated gold standard sample were moderate for the audit coordinator and coding process with κ = 0.57 and K = 0.42 respectively. When both data sources were combined the level of agreement was substantial (κ = 0.65; p < .000). ConclusionsClinical validation studies are required to reinforce data quality within stroke registers. Combining clinical and administrative data sources improves diagnostic accuracy for in-hospital stroke identification.

Association of select psychiatric disorders with incident brain aneurysm and subarachnoid hemorrhage among veterans.

Brain aneurysms represent a significant cause of hemorrhagic stroke. Prior research has demonstrated links between stress and stroke, including brain aneurysms. We aimed to determine relationships between select psychiatric disorders and aneurysms and aneurysmal SAH. We performed retrospective, case-control study of a National Veterans Affairs population with two experimental groups (aneurysm-only and aneurysmal SAH) and 10-fold controls per group matched by age, date, and clinical data source. The studied the presence of 4 psychiatric disorders: Posttraumatic stress disorder (PTSD), major depressive disorder (MDD), generalized anxiety disorder (GAD), and other mood disorders. Our main outcomes Unadjusted and multivariable adjusted ORs of PTSD, MDD, GAD, and mood disorders within aneurysm-only and aSAH groups. In 6,320,789 US Veterans who were enrolled for at least 5 years in Medicare and/or the Veterans Health Administration, we identified 35,094 cases of aneurysm without SAH and 5,749 cases of aneurysm with SAH between 1/2005 and 12/2019. In analyses adjusted for sex, hypertension, and tobacco use, patients with aneurysm were more likely than matched controls to have a history of PTSD (OR 1.48), MDD (OR 1.33), GAD (OR 1.26), and other mood disorders (OR 1.34) (all p-values < 0.0001). Similarly, patients with aSAH were more likely than controls to have a history of PTSD (OR 1.35), MDD (OR 1.38), GAD (OR 1.18), and other mood disorders (OR 1.30) (all p-values < 0.0001). The study, the largest of its kind, further suggests links between psychiatric disorders and stroke. This is important as patients with aneurysms are not routinely screened for such psychiatric risk factors. Additional research on this topic could lead to novel strategies to improve stroke prevention.

An AI Approach to Generating MIDD Assets Across the Drug Development Continuum

Model-informed drug development involves developing and applying exposure-based, biological, and statistical models derived from preclinical and clinical data sources to inform drug development and decision-making. Discrete models are generated from individual experiments resulting in a single model expression that is utilized to inform a single stage-gate decision. Other model types provide a more holistic view of disease biology and potentially disease progression depending on the appropriateness of the underlying data sources for that purpose. Despite this awareness, most data integration and model development approaches are still reliant on internal (within company) data stores and traditional structural model types. An AI/ML-based MIDD approach relies on more diverse data and is informed by past successes and failures including data outside a host company (external data sources) that may enhance predictive value and enhance data generated by the sponsor to reflect more informed and timely experimentation. The AI/ML methodology also provides a complementary approach to more traditional modeling efforts that support MIDD and thus yields greater fidelity in decision-making. Early pilot studies support this assessment but will require broader adoption and regulatory support for more evidence and refinement of this paradigm. An AI/ML-based approach to MIDD has the potential to transform regulatory science and the current drug development paradigm, optimize information value, and increase candidate and eventually product confidence with respect to safety and efficacy. We highlight early experiences with this approach using the AI compute platforms as representative examples of how MIDD can be facilitated with an AI/ML approach.Graphical abstract