TEA-IS: A hybrid DEA-TOPSIS approach for assessing performance and synergy in Chinese health care

2023 | Decision Support Systems | Citations: 0

Authors: Antunes, Jorge; Hadi-Vencheh, Abdollah; Jamshidi, Ali; Tan, Yong; Wanke, Peter

Abstract: This paper presents an assessment of the Chinese healthcare system in 31 provinc ... Expand

Abstract: This paper presents an assessment of the Chinese healthcare system in 31 provinces for a 10-year period in light of relevant physical and human resource variables. First, a novel TEA-IS (Trigonometric Envelopment Analysis for Ideal Solutions) model is developed to assess healthcare efficiency at the province level. Machine learning methods are also employed to predict high-low performance and the synergistic Chinese healthcare province in terms of contextual variables. The results indicate that synergy has played a pivotal role in the Chinese healthcare systems, not only by triggering higher performance levels due to the progressive adoption of best practices over the course of time, but also by being closely related to different socioeconomic and demographic variables, such as the illiteracy rate. It is possible to claim that healthcare performance has remained stable in China over the past two decades, performance and synergy at the province level are still heterogeneous. Collapse

Topics: health information system IT productivity human resource management service quality productivity

Methods: TOPSIS trade-off analysis method machine learning artificial neural network literature study

Value Implications of Sourcing Electronic Health Records: The Role of Physician Practice Integration

2023 | Information Systems Research | Citations: 0

Authors: Bardhan, Indranil R.; Bao, Chenzhang; Ayabakan, Sezgin

Abstract: Should hospitals source electronic health records (EHR) systems from a single ve ... Expand

Abstract: Should hospitals source electronic health records (EHR) systems from a single vendor or multiple vendors to deliver high-value care? We study hospitals’ EHR sourcing strategies based on their degree of integration with physician practices and its impact on the value of healthcare delivered. We propose a novel framework to define healthcare value as the extent to which a hospital effectively expends clinical resources to deliver services that improve patient outcomes. Drawing on modular systems and transaction cost economics theories, we propose a moderated-mediation model that explores the pathways through which EHR sourcing strategies can create value in healthcare. We test our research hypotheses on a large, longitudinal sample of U.S. hospitals and observe that hospitals with EHR configurations closer to single sourcing strategies exhibit greater health information sharing compared with hospitals with multisourced EHR systems. Furthermore, we find that hospital-physician practice integration moderates the impact of single sourcing on health information sharing, which in turn, improves value. Specifically, tighter integration between hospitals and physician practices can create greater value if it is aligned with hospitals’ EHR sourcing strategies. As the healthcare industry moves toward value-based payment reform, our findings provide a useful roadmap to practitioners and policy makers to improve the performance of hospitals and healthcare providers. History: Rajiv Kohli, Senior Editor; Sunil Wattal, Associate Editor. Funding: I.R. Bardhan thanks the Foster Parker Centennial Professorship and the Dean’s Research Excellence Grant at the McCombs School of Business at UT Austin for generous financial support. C. Bao thanks the Spears Fellowship at Oklahoma State University for financial support. Supplemental Material: The online appendix is available at https://doi.org/10.1287/isre.2022.1183. Collapse

Topics: information exchange healthcare data health information system accounting database system

Methods: longitudinal research survey robustness check econometric modeling design artifact

Theories: transaction cost economics general systems theory institutional theory

How AI-Based Systems Can Induce Reflections: The Case of AI-Augmented Diagnostic Work

2023 | Management Information Systems Quarterly | Citations: 0

Authors: Abdel-Karim, Benjamin; Pfeuffer, Nicolas; Carl, K. Valerie; Hinz, Oliver

Abstract: This paper addresses a thus-far neglected dimension in human-artificial intelli ... Expand

Abstract: This paper addresses a thus-far neglected dimension in human-artificial intelligence (AI) augmentation: machine-induced reflections. By establishing a grounded theoretical-informed model of machine-induced reflection, we contribute to the ongoing discussion in information systems (IS) regarding AI and research on reflection theories. In our multistage study, physicians used a machine learning-based (ML) clinical decision support system (CDSS) to see if and how this interaction can stimulate reflective practice in the context of an X-ray diagnosis task. By analyzing verbal protocols, performance metrics, and survey data, we developed an integrative theoretical foundation to explain how ML-based systems can help stimulate reflective practice. Individuals engage in more critical or shallower modes depending on whether they perceive a conflict or agreement with these CDSS systems, which in turn leads to different levels of reflection depth. By uncovering the process of machine-induced reflections, we offer IS research a different perspective on how such AI-based systems can help individuals become more reflective, and consequently more effective, professionals. This perspective stands in stark contrast to the traditional, efficiency-focused view of MLbased decision support systems and also enriches theories on human-AI augmentation. Collapse

Topics: artificial intelligence decision making decision support system algorithmic transparency recommender system

Methods: cognitive walkthrough survey theory development qualitative interview Student's t-test

Theories: learning theory information systems theory theory of abduction

Self-Building Artificial Intelligence and Machine Learning to Empower Big Data Analytics in Smart Cities

2023 | Information Systems Frontiers | Citations: 0

Authors: Alahakoon, Damminda; Nawaratne, Rashmika; Xu, Yan; De Silva, Daswin; Sivarajah, Uthayasankar; Gupta, Bhumika

Abstract: The emerging information revolution makes it necessary to manage vast amounts of ... Expand

Abstract: The emerging information revolution makes it necessary to manage vast amounts of unstructured data rapidly. As the world is increasingly populated by IoT devices and sensors that can sense their surroundings and communicate with each other, a digital environment has been created with vast volumes of volatile and diverse data. Traditional AI and machine learning techniques designed for deterministic situations are not suitable for such environments. With a large number of parameters required by each device in this digital environment, it is desirable that the AI is able to be adaptive and self-build (i.e. self-structure, self-configure, self-learn), rather than be structurally and parameter-wise pre-defined. This study explores the benefits of self-building AI and machine learning with unsupervised learning for empowering big data analytics for smart city environments. By using the growing self-organizing map, a new suite of self-building AI is proposed. The self-building AI overcomes the limitations of traditional AI and enables data processing in dynamic smart city environments. With cloud computing platforms, the self-building AI can integrate the data analytics applications that currently work in silos. The new paradigm of the self-building AI and its value are demonstrated using the IoT, video surveillance, and action recognition applications. Collapse

Topics: artificial intelligence smart city big data cloud computing distributed system

Methods: computational algorithm experiment unsupervised learning machine learning longitudinal research

Managing compliance with privacy regulations through translation guardrails: A health information exchange case study

2023 | Information and Organization | Citations: 0

Authors: Anderson, Chad; Baskerville, Richard; Kaul, Mala

Abstract: Information privacy is increasingly important in our digitally connected world, ... Expand

Abstract: Information privacy is increasingly important in our digitally connected world, particularly in healthcare, and privacy regulations are ramping up to promote appropriate privacy practices. As a digital platform that enables healthcare providers to exchange protected health information (PHI), a health information exchange (HIE) is governed by health information privacy regula­ tions. The challenge for HIEs is to operate in a way that will maximize information exchange while maintaining compliance with regulations that may constrain the sharing of PHI. Regula­ tions impose a measure of universality through compliance requirements, while being flexible to allow adaptation to the local context. However, our longitudinal case study into the privacy policies of an HIE, demonstrates that the journey of privacy ideas from their original formulation in regulations, to their ultimate enactment in an organizational setting, is accompanied by translations, such that the final implementation may vary extensively from its original form. Such variability often results in interpretations that differ from what the regulators intended. Conse­ quently, translation guardrails are necessary to protect against problematic translations of reg­ ulatory ideas which could lead to compliance issues and loss of platform participation. Our findings offer two contributions. First, we contribute to the compliance literature by explaining how guardrails can balance the use of permission and obligation schemas which are necessary to translate regulations into effective organizational policies for the success of HIEs and other in­ formation exchange platforms. Second, we contribute to extending translation theory by explaining how pragmatic reasoning schemas function as the mechanism through which trans­ lation of regulations occurs. Collapse

Topics: privacy information privacy law data security information exchange legal context

Methods: qualitative interview longitudinal research case study longitudinal case study literature study

Theories: organizational behavior theory

Blockchain innovation for consent self-management in health information exchanges

2023 | Decision Support Systems | Citations: 0

Authors: Anderson, Chad; Carvalho, Arthur; Kaul, Mala; Merhout, Jeffrey W.

Abstract: With the increasing digitalization of health data, patients need to make informe ... Expand

Abstract: With the increasing digitalization of health data, patients need to make informed decisions about the online sharing of their protected health information. This necessitates a robust technical infrastructure that enables patients to self-manage consent and the trusted exchange of this information across sharing entities. Unfortu­ nately, current health information exchange systems in the U.S. are limited in both these regards. While there is recent work on digital patient consent management, there is limited work that provides effective solutions for patient self-management of consent. Further, interoperability issues in the way health information exchanges are currently architected and differences in regulations across localities exacerbate the challenges of consent man­ agement. In this research, we survey potential patients' willingness to self-manage healthcare-related consent. Having established the desire for consent self-management, we propose a solution that enables the seamless sharing of patient consent across different healthcare providers and health information exchanges. Specifically, we use a rigorous design science approach to create a blockchain-based, self-managed patient consent system and we evaluate the design through an instantiated prototype. The results of our study should be useful to researchers in healthcare information management as well as to practitioners designing consent management systems. Our research contributes to design science research with an innovative, rigorously evaluated, design principles-based artifact that addresses a critical problem of sharing protected health information. Collapse

Topics: blockchain mobile application information exchange privacy Ethereum

Methods: survey design artifact design science design process design principle

Theories: evidence-based design

Does Analytics Help Resolve Equivocality in the Healthcare Context? Contrasting the Effects of Analyzability and Differentiation

2023 | Journal of the Association for Information Systems | Citations: 0

Authors: Adjerid, Idris; Angst, Corey; Devaraj, Sarv; Berente, Nicholas

Abstract: Organizations are increasingly using data analytics to help make decisions and ... Expand

Abstract: Organizations are increasingly using data analytics to help make decisions and drive positive outcomes. But organizational scholarship has warned us that the sort of information processing associated with analytic capabilities, while effective for uncertainty reduction, may be less effective in equivocal contexts. Equivocality is evident when tasks are not easily analyzable (task analyzability) or when organizational departments are highly differentiated (differentiation). We hypothesize that analytics will be less effective in driving positive outcomes when equivocality is high because of low task analyzability. However, when an organization is more differentiated, resulting in high equivocality, we anticipate that analytics will be more effective in driving positive outcomes. To test this theory, we studied how clinical healthcare analytics influenced experiential quality (akin to patient satisfaction) in over 3,000 hospitals across nine years. Our results show that analytics capabilities, on average, do improve outcomes in terms of patient experiential quality, suggesting that analytics can reduce uncertainty, but we also found evidence for the moderating role of equivocality. Specifically, as task analyzability decreases (i.e., increasing equivocality), clinical healthcare analytics becomes less effective in improving experiential quality. However, when equivocality is high because of differentiation, there is a positive relationship between clinical healthcare analytics and experiential quality but only in larger hospitals. From a managerial perspective, this study has implications for boundary conditions of data analytics in organizations. Collapse

Topics: healthcare analytics technology adoption electronic health record health information system artificial intelligence

Methods: theory development post-hoc analysis experimental group longitudinal research machine learning

Theories: organizational information processing theory

Human-machine collaboration for feature selection and integration to improve congestive Heart failure risk prediction

2023 | Decision Support Systems | Citations: 0

Authors: Ben-Assuli, Ofir; Heart, Tsipi; Klempfner, Robert; Padman, Rema

Abstract: The issue of harnessing machine learning (ML) algorithms for the prediction of a ... Expand

Abstract: The issue of harnessing machine learning (ML) algorithms for the prediction of adverse medical events is important considering the prevalence of vast repositories of patient-level medical data, and the quest to reduce healthcare costs while elevating the quality of care. Of particular interest is identifying a set of features that can better predict the risk of unplanned readmissions of chronically ill patients within 30 days post discharge, among them patients with Congestive Heart Failure (CHF). While numerous studies have employed ML algorithms to identify sets of significant features, only a handful have compared features extracted by various methods, and even fewer have compared the predictive power of feature sets selected by human experts to those selected by ML. Hence, this research aimed to elicit a comprehensive feature set selected by four ML algorithms and compare its predictive performance to that of a feature set selected by experts. We then evaluated a merged list of the two feature sets, constructing a human-machine collaborative set to predict the likelihood of 30-day unplanned readmission of CHF patients using data on 10,763 CHF patients hospitalized during the years 2010–2017. Our models achieved Area under the ROC Curve (AUC) above 0.8 and an Accuracy of ~0.89, comparable to the best models in the extant literature. Moreover, the Human-Machine Collaborative model significantly outperformed (best AUC 0.836) other human-only selected models as well as the machine-selected set. This study contributes to a better understanding of the power of using ML for patient risk stratification with special attention to the benefits of human-machine collaboration, even when these two entities work separately or in parallel, for improved clinical decision making. Collapse

Topics: healthcare data decision making big data missing data data warehouse

Methods: machine learning computational algorithm XGBoost descriptive statistic random forest classification

Two Sides of the Same Pillow: Unfolding the Relationship between Objective and Subjective Sleep Quality with Unsupervised Learning

2023 | International Conference on Information Systems | Citations: 0

Authors: Biedebach, Luka; Óskarsdóttir, María; Arnardottir, Erna; Islind, Anna Sigríður

Abstract: ... Expand

Abstract: Collapse

Topics: mobile system smart watch mobile application smart wearable device healthcare data

Methods: descriptive statistic cluster analysis parametric test ANOVA longitudinal research

Introduction to the Minitrack on Social Media and Healthcare Technology

2023 | HICSS | Citations: 0

Authors: Bock, Beth; Fraser, Hamish; Braciszewski, Jordan; Rosen, Rochelle

Abstract: Social media has changed the way individuals and institutions approach healthca ... Expand

Abstract: Social media has changed the way individuals and institutions approach healthcare and health information and offers opportunities to understand health-related interactions at all levels, from the micro to the macro. The Social Media and Healthcare Technology mini-track presents a series of research papers that address an array of social media and associated technology within healthcare and healthcare research including user perceptions and quality indicators used in online patient communities as well as the mining of online data to detect depressive symptoms, medical errors, and alcohol-related blackouts. Collapse

Topics: social media Twitter health information system electronic health online community

Methods: transformer model natural language processing logistic regression