Data analytics for the sustainable use of resources in hospitals: Predicting the length of stay for patients with chronic diseases

Evaluating the effect of health insurance reform on health equity and financial protection for elderly in low- and middle-income countries: evidences from China

BACKGROUND

To achieve Universal Health Coverage (UHC), China have implemented health system reform to expend health coverage and improve health equity. Scholars have explored the implementing effect of this health reform, but gaps remained in health care received by elderly. This study aims to assess the effect of implementing health insurance payment reform on health care received by elderly, as well as to evaluate its effect on cost sharing to identify whether improve financial protection of elderly under this reform.

METHODS

We identified hospitalization of 46,714 elderly with cerebral infarction from 2013 to 2023. To examine the determinant role played by DRGs payment reform in healthcare for elderly and their financial protection, this study employs the OLS linear regression model for analysis. In the robustness checks, we validated the baseline results through several methods, including excluding the data from the initial implementation of the reform (2021), reducing the impact of the pandemic, and exploring the group effects of different demographic characteristics.

RESULTS

The findings proposed that implementing DRGs payment reduces drug expenses but increases treatment expense of chronic disease for elderly in China. This exacerbates healthcare costs for elderly patients and seems to be contrary to the original purpose of health care reform. Additionally, the implementation of DRGs payment reduced the spending of medical insurance fund, while increased the out-of-pocket of patients, revealing a shift in health care expenses from health insurance fund to out-of-pocket.

CONCLUSIONS

This study shares the lessons from China's health reform and provides enlightenment on how to effective implement health reform to improve health equity and achieve UHC in such low- and middle-income countries facing challenges in health financing.

Beyond the Bedside: Machine Learning-Guided Length of Stay (LOS) Prediction for Cardiac Patients in Tertiary Care

Efficient management of hospital resources is essential for providing high-quality healthcare while ensuring sustainability. Length of stay (LOS), measuring the duration from admission to discharge, directly impacts patient outcomes and resource utilization. Accurate LOS prediction offers numerous benefits, including reducing re-admissions, ensuring appropriate staffing, and facilitating informed discharge planning. While conventional methods rely on statistical models and clinical expertise, recent advances in machine learning (ML) present promising avenues for enhancing LOS prediction. This research focuses on developing an ML-based LOS prediction model trained on a comprehensive real-world dataset and discussing the important factors towards practical deployment of trained ML models in clinical settings. This research involves the development of a comprehensive adult cardiac patient dataset (SaudiCardioStay (SCS)) from the King Faisal Specialist Hospital & Research Centre (KFSH&RC) hospital in Saudi Arabia, comprising 4930 patient encounters for 3611 unique patients collected from 2019 to 2022 (excluding 2020). A diverse range of classical ML models (i.e., Random Forest (RF), Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LGBM), artificial neural networks (ANNs), Average Voting Regression (AvgVotReg)) are implemented for the SCS dataset to explore the potential of existing ML models in LOS prediction. In addition, this study introduces a novel approach for LOS prediction by incorporating a dedicated LOS classifier within a sophisticated ensemble methodology (i.e., Two-Level Sequential Cascade Generalization (2LSCG), Three-Level Sequential Cascade Generalization (3LSCG), Parallel Cascade Generalization (PCG)), aiming to enhance prediction accuracy and capture nuanced patterns in healthcare data. The experimental results indicate the best mean absolute error (MAE) of 0.1700 for the 3LSCG model. Relatively comparable performance was observed for the AvgVotReg model, with a MAE of 0.1703. In the end, a detailed analysis of the practical implications, limitations, and recommendations concerning the deployment of ML approaches in actual clinical settings is presented.

A Bibliometric Analysis of Diagnosis Related Groups from 2013 to 2022

Purpose

As an important management method of the global healthcare system, diagnosis related groups (DRGs) classify patients into different cost groups and pay more attention to the equitable distribution of medical resources and the quality of medical services. At present, most countries have used DRGs to help medical institutions and doctors to treat patients more accurately, avoid the waste of medical resources, and improve treatment efficiency.

Methods

The Web of Science database was searched to collect all relevant literature on DRGs from 2013 to 2022. The literature information was imported into CiteSpace, Vosviewer, and Histcite for data analysis and visualization of the results. Analyze the cooperative relationship among the countries, institutions, journals, and authors. The usage trend of keywords; Highlight the content of the cited articles.

Results

The number of articles published in this decade was stable, and the number of citations in 2014 was the highest. The United States and Germany, as the first countries to use the DRGs system, are ahead of other countries in terms of the number and quality of articles. We have carried out content research on the articles with high citations, and summarized the application range of DRGs; classification method; advantages and disadvantages of the application. In general, the development trend of DRGs in foreign countries is to continuously optimize the classification method, expand the scope of application, and improve the application effect. These provide support and reference for the improvement of medical services and the perfection of the medical insurance system.

Conclusion

The application of DRGs can improve the quality and efficiency of medical services, and reduce the waste of medical expenses. It can also promote the rational allocation of medical resources and the equity of medical services. In the future, DRGs will pay more attention to the personalized diagnosis and treatment and fine management of patients, and the sharing and standardization of medical data, to promote the development of medical informatization.

Predicting and Mitigating Freshmen Student Attrition: A Local-Explainable Machine Learning Framework

With the emergence of novel methods for improving machine learning (ML) transparency, traditional decision-support-focused information systems seem to need an upgrade in their approach toward providing more actionable insights for practitioners. Particularly, given the complex decision-making process of humans, using insights obtained from group-level interpretation of ML models for designing individual interventions may lead to mixed results. The present study proposes a hybrid ML framework by integrating established predictive and explainable ML approaches for decision support systems involving the prediction of human decisions and designing individualized interventions accordingly. The proposed framework is aimed at providing actionable insights for designing individualized interventions. It was showcased in the context of college students' attrition problem using a large and feature-rich integrated data set of freshman college students containing information about their demographics, educational, financial, and socioeconomic factors. A comparison of feature importance scores at the group- vs. individual-level revealed that while group-level insights might be useful for adjusting long-term strategies, using them as a one-size-fits-all strategy to design and implement individual interventions is subject to suboptimal outcomes.

Prescriptive analytics applications in sustainable operations research: conceptual framework and future research challenges

In the broad sphere of Analytics, prescriptive analytics is one of the emerging areas of interest for both academicians and practitioners. As prescriptive analytics has transitioned from its inception to an emerging topic, there is a need to review existing literature in order to ascertain development in this area. There are a very few reviews in the related field but not specifically on the applications of prescriptive analytics in sustainable operations research using content analysis. To address this gap, we performed a review of 147 articles published in peer-reviewed academic journals from 2010 to August 2021. Using content analysis, we have identified the five emerging research themes. Through this study, we aim to contribute to the literature on prescriptive analytics by identifying and proposing emerging research themes and future research directions. Based on our literature review, we propose a conceptual framework for studying the impacts of the adoption of prescriptive analytics and its impact on sustainable supply chain resilience, sustainable supply chain performance and competitive advantage. Finally, the paper acknowledges the managerial implications, theoretical contribution and the limitations of this study.

A Machine Learning Framework for Assessing the Risk of Venous Thromboembolism in Patients Undergoing Hip or Knee Replacement

Venous thromboembolism (VTE) is a well-recognized complication that is prevalent in patients undergoing major orthopedic surgery (e.g., total hip arthroplasty and total knee arthroplasty). For years, to identify patients at high risk of developing VTE, physicians have relied on traditional risk scoring systems, which are too simplistic to capture the risk level accurately. In this paper, we propose a data-driven machine learning framework to identify such high-risk patients before they undergo a major hip or knee surgery. Using electronic health records of more than 392,000 patients who undergone a major orthopedic surgery, and following a guided feature selection using the genetic algorithm, we trained a fully connected deep neural network model to predict high-risk patients for developing VTE. We identified several risk factors for VTE that were not previously recognized. The best FCDNN model trained using the selected features yielded an area under the ROC curve (AUC) of 0.873, which was remarkably higher than the best AUC obtained by including only risk factors previously known in the medical literature. Our findings suggest several interesting and important insights. The traditional risk scoring tables that are being widely used by physicians to identify high-risk patients are not considering a comprehensive set of risk factors, nor are they as powerful as cutting-edge machine learning methods in distinguishing low- from high-risk patients.

Preadmission assessment of extended length of hospital stay with RFECV-ETC and hospital-specific data

Background

Patients who exceed their expected length of stay in the hospital come at a cost to stakeholders in the healthcare sector as bed spaces are limited for new patients, nosocomial infections increase and the outcome for many patients is hampered due to multimorbidity after hospitalization.

Objectives

This paper develops a technique for predicting Extended Length of Hospital Stay (ELOHS) at preadmission and their risk factors using hospital data.

Methods

A total of 91,468 records of patient’s hospital information from a private acute teaching hospital were used for developing a machine learning algorithm relaying on Recursive Feature Elimination with Cross-Validation and Extra Tree Classifier (RFECV-ETC). The study implemented Synthetic Minority Oversampling Technique (SMOTE) and tenfold cross-validation to determine the optimal features for predicting ELOHS while relying on multivariate Logistic Regression (LR) for computing the risk factors and the Relative Risk (RR) of ELOHS at a 95% confidence level.

Results

An estimated 11.54% of the patients have ELOHS, which increases with patient age as patients < 18 years, 18–40 years, 40–65 years and ≥ 65 years, respectively, have 2.57%, 4.33%, 8.1%, and 15.18% ELOHS rates. The RFECV-ETC algorithm predicted preadmission ELOHS to an accuracy of 89.3%. Age is a predominant risk factors of ELOHS with patients who are > 90 years—PAG (> 90) {RR: 1.85 (1.34–2.56), P:  < 0.001} having 6.23% and 23.3%, respectively, higher likelihood of ELOHS than patient 80–90 years old—PAG (80–90) {RR: 1.74 (1.34–2.38), P:  < 0.001} and those 70–80 years old—PAG (70–80) {RR: 1.5 (1.1–2.05), P: 0.011}. Those from admission category—ADC (US1) {RR: 3.64 (3.09–4.28, P:  < 0.001} are 14.8% and 70.5%, respectively, more prone to ELOHS compared to ADC (UC1) {RR: 3.17 (2.82–3.55), P:  < 0.001} and ADC (EMG) {RR: 2.11 (1.93–2.31), P:  < 0.001}. Patients from SES (low) {RR: 1.45 (1.24–1.71), P:  < 0.001)} are 13.3% and 45% more susceptible to those from SES (middle) and SES (high). Admission type (ADT) such as AS2, M2, NEWS, S2 and others {RR: 1.37–2.77 (1.25–6.19), P:  < 0.001} also have a high likelihood of contributing to ELOHS while the distance to hospital (DTH) {RR: 0.64–0.75 (0.56–0.82), P:  < 0.001}, Charlson Score (CCI) {RR: 0.31–0.68 (0.22–0.99), P:  < 0.001–0.043} and some VMO specialties {RR: 0.08–0.69 (0.03–0.98), P:  < 0.001–0.035} have limited influence on ELOHS.

Conclusions

Relying on the preadmission assessment of ELOHS helps identify those patients who are susceptible to exceeding their expected length of stay on admission, thus, making it possible to improve patients’ management and outcomes.

An explanatory machine learning framework for studying pandemics: The case of COVID-19 emergency department readmissions

One of the major challenges that confront medical experts during a pandemic is the time required to identify and validate the risk factors of the novel disease and to develop an effective treatment protocol. Traditionally, this process involves numerous clinical trials that may take up to several years, during which strict preventive measures must be in place to control the outbreak and reduce the deaths. Advanced data analytics techniques, however, can be leveraged to guide and speed up this process. In this study, we combine evolutionary search algorithms, deep learning, and advanced model interpretation methods to develop a holistic exploratory-predictive-explanatory machine learning framework that can assist clinical decision-makers in reacting to the challenges of a pandemic in a timely manner. The proposed framework is showcased in studying emergency department (ED) readmissions of COVID-19 patients using ED visits from a real-world electronic health records database. After an exploratory feature selection phase using genetic algorithm, we develop and train a deep artificial neural network to predict early (i.e., 7-day) readmissions (AUC = 0.883). Lastly, a SHAP model is formulated to estimate additive Shapley values (i.e., importance scores) of the features and to interpret the magnitude and direction of their effects. The findings are mostly in line with those reported by lengthy and expensive clinical trial studies.

The Impact of Digital Transformation on Inpatient Care: A Mixed Design Study (Preprint)

BACKGROUND

In the context of the digital transformation of all areas of society, health care providers are also under pressure to change. New technologies and a change in patients' self-perception and health awareness require rethinking the provision of health care services. New technologies and the extensive use of data can change provision processes, optimize them, or replace them with new services. The inpatient sector, which accounts for a particularly large share of health care spending, plays a major role in this regard.

OBJECTIVE

This study examined the influences of current trends in digitization on inpatient service delivery.

METHODS

We conducted a scoping review. This was applied to identify the international trends in digital transformation as they relate to hospitals. Future trends were considered from different perspectives. Using the defined inclusion criteria, international peer-reviewed articles published between 2016 and 2021 were selected. The extracted core trends were then contextualized for the German hospital sector with 12 experts.

RESULTS

We included 44 articles in the literature analysis. From these, 8 core trends could be deduced. A heuristic impact model of the trends was derived from the data obtained and the experts' assessments. This model provides a development corridor for the interaction of the trends with regard to technological intensity and supply quality. Trend accelerators and barriers were identified.

CONCLUSIONS

The impact analysis showed the dependencies of a successful digital transformation in the hospital sector. Although data interoperability is of particular importance for technological intensity, the changed self-image of patients was shown to be decisive with regard to the quality of care. We show that hospitals must find their role in new digitally driven ecosystems, adapt their business models to customer expectations, and use up-to-date information and communications technologies.

Network analytics and machine learning for predicting length of stay in elderly patients with chronic diseases at point of admission

Background

An aging population with a burden of chronic diseases puts increasing pressure on health care systems. Early prediction of the hospital length of stay (LOS) can be useful in optimizing the allocation of medical resources, and improving healthcare quality. However, the data available at the point of admission (PoA) are limited, making it difficult to forecast the LOS accurately.

Methods

In this study, we proposed a novel approach combining network analytics and machine learning to predict the LOS in elderly patients with chronic diseases at the PoA. Two networks, including multimorbidity network (MN) and patient similarity network (PSN), were constructed and novel network features were created. Five machine learning models (eXtreme Gradient Boosting, Gradient Boosting Decision Tree, Random Forest, Linear Support Vector Machine, and Deep Neural Network) with different input feature sets were developed to compare their performance.

Results

The experimental results indicated that the network features can bring significant improvements to the performances of the prediction models, suggesting that the MN and PSN are useful for LOS predictions.

Conclusion

Our predictive framework which integrates network science with data mining can forecast the LOS effectively at the PoA and provide decision support for hospital managers, which highlights the potential value of network-based machine learning in healthcare field.

Evaluating the Effective Factors of Hospital Rooms on Patients' Recovery Using the Data Mining Method

OBJECTIVES

This study aims to investigate the effective environmental factors of hospital rooms in patients' recovery through data mining techniques.

BACKGROUND

Previous studies have shown the positive effect of the interior environment of the hospitals on patients' recovery. The methods of these studies were mainly based on the evidence and patients' perception while hospital environments are associated with a large amount of data that make them an appropriate case for data mining studies. But data mining studies in hospitals mainly focused on medical and management purposes rather than evaluating the interior environment condition.

METHODS

We analyzed the hospital information system data of a hospital using Python programming language and some of its libraries. Preprocessing and eliminating the outliers, labeling and clustering of diseases, data visualization and analysis, final evaluation, and concluding were done using the knowledge discovery in databases process.

RESULTS

Pearson coefficient value for rooms' area was .5 and, respectively, for the distance from the ward entrance and nursing station were .75 and .70. The χ² values for the variables of room types, location, and occupation were 24.62, 18.98, and 21.53, respectively, and for the beds' location was 0.12.

CONCLUSIONS

The results confirmed the correlation of the length of stay with the room types, location, and occupation, distance from the nursing station and ward entrance and also showed a moderate correlation with the rooms' area. However, no evidence was found about the relationship between the beds' location in rooms and patients' length of hospital stay.

BIM for Smart Hospital Management during COVID-19 Using MCDM

In context of the recent COVID-19 pandemic, smart hospitals’ contributions to pre-medical, remote diagnosis, and social distancing has been further vetted. Smart hospital management evolves with new technology and knowledge management, which needs an evaluation system to prioritize its associated criteria and sub-criteria. The global effect of the COVID-19 pandemic further necessitates a comprehensive research of smart hospital management. This paper will utilize Analytical Hierarchy Process (AHP) within Multiple Criteria Decision Making (MCDM) to establish a smart hospital evaluation system with evaluation criteria and sub-criteria, which were then further prioritized and mapped to BIM-related alternatives to inform asset information management (AIM) practices. This context of this study included the expert opinions of six professionals in the smart hospital field and collected 113 responses from hospital-related personnel. The results indicated that functionalities connected to end users are critical, in particular IoT’s Network Core Functionalities, AI’s Deep Learning and CPS’s Special Network Technologies. Furthermore, BIM’s capability to contribute to the lifecycle management of assets can relate and contribute to the asset-intensive physical criteria of smart hospitals, in particular IoT, service technology innovations and their sub-criteria.

A framework for understanding artificial intelligence research: insights from practice

Purpose

The current evolution of artificial intelligence (AI) practices and applications is creating a disconnection between modern-day information system (IS) research and practices. The purpose of this study is to propose a classification framework that connects the IS discipline to contemporary AI practices.

Design/methodology/approach

We conducted a review of practitioner literature to derive our framework's key dimensions. We reviewed 103 documents on AI published by 25 leading technology companies ranked in the 2019 list of Fortune 500 companies. After that, we reviewed and classified 110 information system (IS) publications on AI using our proposed framework to demonstrate its ability to classify IS research on AI and reveal relevant research gaps.

Findings

Practitioners have adopted different definitional perspectives of AI (field of study, concept, ability, system), explaining the differences in the development, implementation and expectations from AI experienced today. All these perspectives suggest that perception, comprehension, action and learning are the four capabilities AI artifacts must possess. However, leading IS journals have mostly published research adopting the “AI as an ability” perspective of AI with limited theoretical and empirical studies on AI adoption, use and impact.

Research limitations/implications

First, the framework is based on the perceptions of AI by a limited number of companies, although it includes all the companies leading current AI practices. Secondly, the IS literature reviewed is limited to a handful of journals. Thus, the conclusions may not be generalizable. However, they remain true for the articles reviewed, and they all come from well-respected IS journals.

Originality/value

This is the first study to consider the practitioner's AI perspective in designing a conceptual framework for AI research classification. The proposed framework and research agenda are used to show how IS could become a reference discipline in contemporary AI research.