Big Data Management in US Hospitals: Benefits and Barriers

Incidence and mortality of prostate cancer in France from 2010 to 2021, using a real-life database (National Health Data System - SNDS) - the CaPCo Study

BACKGROUND AND OBJECTIVE

Prostate cancer (PCa) was the leading incident cancer and 3rd leading cause of cancer death in men in France in 2015 with inter-regional disparities. The objectives were to describe PCa incidence and mortality in France and by region, using real life data from the National Health Data System and to identify the factors associated with all-cause or PCa-specific mortality.

METHODS

Men aged ≥ 18years hospitalized and/or on long-term care for PCa (ICD-10 code C61) in France between 2010 and 2021 were included. An incident case was defined by the absence of any cancer in the five years preceding the first coding. Incidence and mortality estimates were age-standardized: France 2018f standardized rates (FSR), all-cause mortality (SMR) and PCa-specific mortality (SMRspe). Factors associated with death were identified using cause-specific Cox models.

RESULTS

The mean annual incidence was 47,081cases/year (FSR:179.6/100,000men), increasing over the period except 2020. All-cause mortality was 20,259 deaths/year (77.3/100,000men), and PCa-specific mortality was 7,265 deaths/year (27.7/100,000men). A PCa-specific mortality excess was found in Centre-Val-de-Loire (SMRspe = 1.21), Bretagne (1.18), Hauts-de-France (1.17) Normandie (1.15). After adjustment, significant PCa mortality excess was observed in Bretagne (HR = 1,29;95%IC[1.09-1.46]) and Hauts-de-France (HR = 1.19[1.03-1.34]). The other factors associated with death were an age ≥ 60years, an increasing comorbidity index, metastatic disease at onset (major weight in specific mortality with hazard ratio HR = 16.1[15.2-17.0]), precariousness, affiliation to the agricultural scheme, and the COVID period in all-cause mortality.

CONCLUSION

This study updated incidence and mortality data in France. It showed differences in mortality between regions in France. The adjustment moderates regional findings based on raw mortality data.

Data mining and safety analysis of traditional immunosuppressive drugs: a pharmacovigilance investigation based on the FAERS database

OBJECTIVE

The purpose of this study aimed to explore the new and serious adverse events(AEs) of Tacrolimus(FK506), cyclosporine(CsA), azathioprine(AZA), mycophenolate mofetil(MMF), cyclophosphamide(CTX) and methotrexate(MTX), which have not been concerned.

METHODS

The FAERS data from January 2016 and December 2022 were selected for disproportionality analysis to discover the potential risks of traditional immunosuppressive drugs.

RESULTS

Compared with CsA, FK506 has more frequent transplant rejection, and is more related to renal impairment, COVID-19, cytomegalovirus infection and aspergillus infection. However, CsA has a high infection-related fatality rate. In addition, we also found some serious and rare AE in other drugs which were rarely reported in previous studies. For example, AZA is closely related to hepatosplenic T-cell lymphoma with high fatality rate and MTX is strongly related to hypofibrinogenemia.

CONCLUSION

The AEs report on this study confirmed that the results were basically consistent with the previous studies, but there were also some important safety signals that were inconsistent with or not mentioned in previous published studies.

EXPERT OPINION

The opinion section discusses some of the limitations and shortcomings, proposing the areas where more effort should be invested in order to improve the safety of immunosuppressive drugs.

Platinum drugs-related safety profile: The latest five-year analysis from FDA adverse event reporting system data

Background

With the widespread application of platinum drugs in antitumor therapy, the incidence of platinum drug adverse events (ADEs) is always severe. This study aimed to explore the adverse event signals of Cisplatin, Carboplatin and Oxaliplatin, three widely used platinum-containing drugs, and to provide a reference for rational individualized clinical drug use.

Methods

The adverse event report data of the three platinum drugs from the first quarter of 2017 to the fourth quarter of 2021 were extracted from the FAERS database, and the data mining and risk factors for the relevant reports were carried out using the reporting odds ratio (ROR) method the proportional reporting ratio (PRR)and the comprehensive criteria (MHRA) method.

Results

A total of 1853 effective adverse event signals were obtained for the three platinum agents, including 558 effective signals for Cisplatin, 896 effective signals for Carboplatin, and 399 effective signals for Oxaliplatin. The signals involve 23 effective different system organs (SOCs). The adverse events of Cisplatin are mainly fixed on blood and lymphatic system diseases, gastrointestinal diseases, systemic diseases and various reactions at the administration site. The adverse events of Carboplatin are mainly focused on blood and lymphatic system diseases, respiratory system, thoracic and mediastinal diseases, while the adverse events of Oxaliplatin are mainly concentrated in respiratory system, thoracic and mediastinal diseases, various nervous system diseases, and gastrointestinal system diseases.

Conclusion

It was found that the main systems involved in common adverse events of platinum drugs are different, and the correlation strength of platinum drugs with the certain adverse events of each system is different.

Big data curation framework: Curation actions and challenges

Big data curation represents an emerging topic of inquiry but still in an early phase along its adoption curve. The term big data itself is a nebulous concept, and the differences between small data curation and big data curation are nuanced. The goal of this research is to provide a theoretical framework that identifies big data curation actions and associated curation challenges. This study is based on the practices of big data research and data curation by systematically examining literature. The outcome of the study includes the big data curation framework that provides overview of curation activities and concerns that are essential to perform such activities. The study also provides practical implications for libraries, archives, data repositories and other information organisations that concerns the issue of big data curation as big data presents a multidimensional array of exigencies in relation to the mission of those organisations.

Theory and Practice of Integrating Machine Learning and Conventional Statistics in Medical Data Analysis

The practice of medical decision making is changing rapidly with the development of innovative computing technologies. The growing interest of data analysis with improvements in big data computer processing methods raises the question of whether machine learning can be integrated with conventional statistics in health research. To help address this knowledge gap, this paper presents a review on the conceptual integration between conventional statistics and machine learning, focusing on the health research. The similarities and differences between the two are compared using mathematical concepts and algorithms. The comparison between conventional statistics and machine learning methods indicates that conventional statistics are the fundamental basis of machine learning, where the black box algorithms are derived from basic mathematics, but are advanced in terms of automated analysis, handling big data and providing interactive visualizations. While the nature of both these methods are different, they are conceptually similar. Based on our review, we conclude that conventional statistics and machine learning are best to be integrated to develop automated data analysis tools. We also strongly believe that machine learning could be explored by health researchers to enhance conventional statistics in decision making for added reliable validation measures.

Time interval from last visit to imaging diagnosis influences outcome in pancreatic adenocarcinoma: A regional population-based study on linked medico-administrative and clinical data

Background:

Excessive waiting time intervals for the diagnosis and treatment of patients with pancreatic cancer can influence their prognosis but they remain unclear. The objective was to describe time intervals from the medical visit to diagnostic imaging and to treatment and their prognostic impact in pancreatic cancer in one French region.

Methods:

This retrospective observational multicentre study included all patients with pancreatic cancer seen for the first time in 2017 in multidisciplinary team meetings (MTMs), where clinical data were collected. A probabilistic matching with the medico-administrative data from the French national healthcare database (Système National des Données de Santé) was performed to define the care pathway from clinical presentation to the beginning of treatment. Median key time intervals were estimated for both resected and unresected tumours. Factors associated with 1-year survival were studied using Cox model.

Results:

A total of 324 patients (88% of total patients with MTM presentation) were matched and included: male 54%, mean age 72 years ±9.2, Eastern Cooperative Oncology Group (ECOG) PS > 1 19.5%, metastatic disease at diagnosis 47.4%, tumour resection 16%. At 1 year, 57% had died (65% in the unresected group and 17% in the resected group). The median time interval from the medical visit to diagnostic imaging was 15 days [Q1–Q3: 8–44]. After imaging, median time intervals to definite diagnosis and to first treatment were 11 and 20 days, respectively. Significant prognostic factors associated with the risk of death at 1 year were ECOG PS > 1 (hazard ratio (HR) 2.1 [1.4–3.0]), metastasis (HR 2.7 [1.9–3.9]), no tumour resection (HR 2.7 [1.3–5.6]) and time interval between the medical visit and diagnostic imaging ⩾25 days (HR 1.7 [1.2–2.3]).

Conclusion:

Delay in access to diagnostic imaging impacted survival in patients with pancreatic cancer, regardless of whether tumour resection had been performed.

Identification and analysis of adoption barriers of disruptive technologies in the logistics industry

Purpose

Recently, disruptive technologies (DTs) have proposed several innovative applications in managing logistics and promise to transform the entire logistics sector drastically. Often, this transformation is not successful due to the existence of adoption barriers to DTs. This study aims to identify the significant barriers that impede the successful adoption of DTs in the logistics sector and examine the interrelationships amongst them.

Design/methodology/approach

Initially, 12 critical barriers were identified through an extensive literature review on disruptive logistics management, and the barriers were screened to ten relevant barriers with the help of Fuzzy Delphi Method (FDM). Further, an Interpretive Structural Modelling (ISM) approach was built with the inputs from logistics experts working in the various departments of warehouses, inventory control, transportation, freight management and customer service management. ISM approach was then used to generate and examine the interrelationships amongst the critical barriers. Matrics d’Impacts Croises-Multiplication Applique a Classement (MICMAC) analysed the barriers based on the barriers' driving and dependence power.

Findings

Results from the ISM-based technique reveal that the lack of top management support (B6) was a critical barrier that can influence the adoption of DTs. Other significant barriers, such as legal and regulatory frameworks (B1), infrastructure (B3) and resistance to change (B2), were identified as the driving barriers, and industries need to pay more attention to them for the successful adoption of DTs in logistics. The MICMAC analysis shows that the legal and regulatory framework and lack of top management support have the highest driving powers. In contrast, lack of trust, reliability and privacy/security emerge as barriers with high dependence powers.

Research limitations/implications

The authors' study has several implications in the light of DT substitution. First, this study successfully analyses the seven DTs using Adner and Kapoor's framework (2016a, b) and the Theory of Disruptive Innovation (Christensen, 1997; Christensen et al., 2011) based on the two parameters as follows: emergence challenge of new technology and extension opportunity of old technology. Second, this study categorises these seven DTs into four quadrants from the framework. Third, this study proposes the recommended paths that DTs might want to follow to be adopted quickly.

Practical implications

The authors' study has several managerial implications in light of the adoption of DTs. First, the authors' study identified no autonomous barriers to adopting DTs. Second, other barriers belonging to any lower level of the ISM model can influence the dependent barriers. Third, the linkage barriers are unstable, and any preventive action involving linkage barriers would subsequently affect linkage barriers and other barriers. Fourth, the independent barriers have high influencing powers over other barriers.

Originality/value

The contributions of this study are four-fold. First, the study identifies the different DTs in the logistics sector. Second, the study applies the theory of disruptive innovations and the ecosystems framework to rationalise the choice of these seven DTs. Third, the study identifies and critically assesses the barriers to the successful adoption of these DTs through a strategic evaluation procedure with the help of a framework built with inputs from logistics experts. Fourth, the study recognises DTs adoption barriers in logistics management and provides a foundation for future research to eliminate those barriers.

Longitudinal Analysis of Continuous Pulse Oximetry as Prognostic Factor in Neonatal Respiratory Distress

OBJECTIVE

Analysis of longitudinal data can provide neonatologists with tools that can help predict clinical deterioration and improve outcomes. The aim of this study is to analyze continuous monitoring data in newborns, using vital signs to develop predictive models for intensive care admission and time to discharge.

STUDY DESIGN

We conducted a retrospective cohort study, including term and preterm newborns with respiratory distress patients admitted to the neonatal ward. Clinical and epidemiological data, as well as mean heart rate and saturation, at every minute for the first 12 hours of admission were collected. Multivariate mixed, survival and joint models were developed.

RESULTS

A total of 56,377 heart rate and 56,412 oxygen saturation data were analyzed from 80 admitted patients. Of them, 73 were discharged home and 7 required transfer to the intensive care unit (ICU). Longitudinal evolution of heart rate (p < 0.01) and oxygen saturation (p = 0.01) were associated with time to discharge, as well as birth weight (p < 0.01) and type of delivery (p < 0.01). Longitudinal heart rate evolution (p < 0.01) and fraction of inspired oxygen at admission at the ward (p < 0.01) predicted neonatal ICU (NICU) admission.

CONCLUSION

Longitudinal evolution of heart rate can help predict time to transfer to intensive care, and both heart rate and oxygen saturation can help predict time to discharge. Analysis of continuous monitoring data in patients admitted to neonatal wards provides useful tools to stratify risks and helps in taking medical decisions.

KEY POINTS

· Continuous monitoring of vital signs can help predict and prevent clinical deterioration in neonatal patients.. · In our study, longitudinal analysis of heart rate and oxygen saturation predicted time to discharge and intensive care admission.. · More studies are needed to prospectively prove that these models can helpmake clinical decisions and stratify patients' risks..

Applicability and potentiality in the use of Business Intelligence tools in Primary Health Care

Data management tools, called Business Intelligence (BI), can be important to provide complete and customizable information for the demands of health management. The objective of the article is to present the evaluation of the applicability and potential of a BI tool in the planning of management actions of Primary Health Care. Exploratory study, with a quantitative approach, using the dimensions of efficiency and optimization as attributes of quality. A Family Clinic was selected in the city of Rio de Janeiro. Data from the territory, from the Bolsa Família Program register and some "Care Lines" were inserted in the BI, in order to explore the possibilities of combining and generating indicators. In this article, we present the use of Form A and the pregnant woman's Care Line. As a result, greater range of detailed indicators compared to a common tab, and optimization in obtaining lists and perform monitoring tasks by the teams and the manager. Regarding efficiency, its low cost and easy handling reduces the costs of creation and necessary professionals. As a conclusion, the BI tool enables greater organization and planning, facilitating the Family Health Clinic management, mainly for the monitoring of indicators and evaluation processes.

Clinical Decision Support for Worker Health: A Five-Site Qualitative Needs Assessment in Primary Care Settings

BACKGROUND

Although patients who work and have related health issues are usually first seen in primary care, providers in these settings do not routinely ask questions about work. Guidelines to help manage such patients are rarely used in primary care. Electronic health record (EHR) systems with worker health clinical decision support (CDS) tools have potential for assisting these practices.

OBJECTIVE

This study aimed to identify the need for, and barriers and facilitators related to, implementation of CDS tools for the clinical management of working patients in a variety of primary care settings.

METHODS

We used a qualitative design that included analysis of interview transcripts and observational field notes from 10 clinics in five organizations.

RESULTS

We interviewed 83 providers, staff members, managers, informatics and information technology experts, and leaders and spent 35 hours observing. We identified eight themes in four categories related to CDS for worker health (operational issues, usefulness of proposed CDS, effort and time-related issues, and topic-specific issues). These categories were classified as facilitators or barriers to the use of the CDS tools. Facilitators related to operational issues include current technical feasibility and new work patterns associated with the coordinated care model. Facilitators concerning usefulness include users' need for awareness and evidence-based tools, appropriateness of the proposed CDS for their patients, and the benefits of population health data. Barriers that are effort-related include additional time this proposed CDS might take, and other pressing organizational priorities. Barriers that are topic-specific include sensitive issues related to health and work and the complexities of information about work.

CONCLUSION

We discovered several themes not previously described that can guide future CDS development: technical feasibility of the proposed CDS within commercial EHRs, the sensitive nature of some CDS content, and the need to assist the entire health care team in managing worker health.

Open Source Infrastructure for Health Care Data Integration and Machine Learning Analyses

PURPOSE

We have created a cloud-based machine learning system (CLOBNET) that is an open-source, lean infrastructure for electronic health record (EHR) data integration and is capable of extract, transform, and load (ETL) processing. CLOBNET enables comprehensive analysis and visualization of structured EHR data. We demonstrate the utility of CLOBNET by predicting primary therapy outcomes of patients with high-grade serous ovarian cancer (HGSOC) on the basis of EHR data.

MATERIALS AND METHODS

CLOBNET is built using open-source software to make data preprocessing, analysis, and model training user friendly. The source code of CLOBNET is available in GitHub. The HGSOC data set was based on a prospective cohort of 208 patients with HGSOC who were treated at Turku University Hospital, Finland, from 2009 to 2019 for whom comprehensive clinical and EHR data were available.

RESULTS

We trained machine learning (ML) models using clinical data, including a herein developed dissemination score that quantifies the disease burden at the time of diagnosis, to identify patients with progressive disease (PD) or a complete response (CR) on the basis of RECIST (version 1.1). The best performance was achieved with a logistic regression model, which resulted in an area under receiver operating characteristic curve (AUROC) of 0.86, with a specificity of 73% and a sensitivity of 89%, when it classified between patients who experienced PD and CR.

CONCLUSION

We have developed an open-source computational infrastructure, CLOBNET, that enables effective and rapid analysis of EHR and other clinical data. Our results demonstrate that CLOBNET allows predictions to be made on the basis of EHR data to address clinically relevant questions.

Designing a Performance Measurement System for Accountability, Quality Improvement, and Innovation

The purpose of this article is to detail a system for the design of performance measures that will be used to assess the achievement of a health care organization's strategic goals and its need for change. The article begins by emphasizing the importance of accountability and the need for the presence of a dynamic learning culture that is premised on a foundation of accountability, continuous improvement, learning, and innovation. This is followed by describing the importance of utilizing an interdisciplinary team with physician and patient involvement to guide the design and implementation of the performance measurement system. The goals of the system are then outlined and followed by a description of the process for the determination of the framework, scope, domains, measures, and reporting mechanisms for displaying the performance measures. Lastly, guidelines for the design of valid, reliable, and cost-effective performance measures are discussed with the aim of maximizing their utility by health care professionals, managers, and administrators.