ABSTRACT Introduction Intensive-Care-Units (ICUs) play a vital role in healthcare systems to deliver life-saving care to critical-patients, emergency-patients, and elective-surgical-patients who require specialized post-surgery facilities. Arrivals to this unit and Length-of-Stays(LOS) are both stochastic and affect the occupied capacity of the unit. Because of limited ICU resources, capacity management has paramount importance, involving optimal utilization of available resources. Objectives This paper proposes a stochastic model for capacity management aimed at reducing congestion in ICU through a two-step approach. Methodology A Multilayer-Perceptron Artificial Neural Network(MLP-ANN) model is employed based on time-series data to address the uncertainty associated with the admission of various patient types and the daily-occupied capacity of the ICU. Then, a stochastic model is developed to determine the ICU occupancy level over the study period. The first stage predictive models’ outputs are utilized along with appropriate probability distribution functions for LOSs to account for the uncertainty inherent in serving patients. Results We recommend policies to control the arrival of the elective-surgical-patients who require admission to ICU, to manage bed capacity. The results indicate a reduction (at least 1.6% per day) in congestion within the unit. Practical Implications We provide guidelines (e.g. protocol-driven OR-ICU-coordination) for decision-makers to more efficiently manage the ICU.

ABSTRACT The rapid evolution of artificial intelligence (AI) is reshaping healthcare by improving diagnostics, patient outcomes, and operational efficiency. Yet, many frameworks for AI adoption overlook the complex and iterative nature of healthcare systems. This study introduces the AI Healthcare Symbiosis Cycle (AI-HSC), a novel framework based on Dynamic Capabilities Theory, Systems Theory, and Kotter’s 8-Step Change Model, conceptualising AI adoption as a continuous and adaptive process. Dynamic Capabilities Theory highlights the need for organisations to sense opportunities, seize resources, and change processes in response to AI advancements. Systems Theory focuses on optimising interdependencies within healthcare organisations, while Kotter’s model ensures a structured approach to managing change. The AI-HSC aligns phases of AI integration – initiation, integration, evolution, and revolution – with Kotter’s steps, promoting a systematic and scalable adoption strategy. Key recommendations include implementing pilot programs, fostering interdisciplinary coalitions, embedding AI literacy into organisational culture, and developing robust ethics and compliance frameworks. By bridging theory with practice, the AI-HSC provides actionable strategies for sustainable AI integration, addressing critical barriers and fostering continuous innovation. This research contributes to the digital change discourse, offering valuable insights for academia and healthcare practitioners.

ABSTRACT Outpatient chemotherapy is experiencing increased demand while available resources are limited and costly. It is important to have efficient planning for outpatient chemotherapy to optimize the utilization of resources while maintaining the optimal quality of care services. This paper provides a comprehensive review of research studies on outpatient chemotherapy planning (OCP) and decision making. A total of 1,262 articles were retrieved from the most prominent databases (Scopus, PubMed, Web of Science, and Science Direct) and other resources, 99 of which passed the inclusion criteria and were included in this study for analysis. A bibliometric analysis was subsequently conducted with a comprehensive classification of studies based on several predefined criteria that included decision level, problem scope, performance metrics and objectives, complexity factors, methodology, and solution methods. The papers are reviewed in structured categories to provide a detailed overview of this field and highlight the areas that need to be focused on, and the gaps and limitations are summarized. Finally, future research trends on OCP and several promising lines of research that are worthy of study in the future are identified and discussed.

ABSTRACT Modelling and simulation studies have been used to inform the choices and development of quality improvement (QI) initiatives in health care, for example, by helping refine the intervention to be implemented or support decisions around the management of demand and capacity. We do not know whether a modelling study can itself be informed by a QI project and what are the associated benefits and challenges. In this research, we sought to investigate the opportunities and challenges associated with an ongoing health service-led QI project in informing the development of a stochastic simulation-based decision support tool to inform decisions around the commissioning of anticoagulation services for patients with atrial fibrillation. We found that the positive synergies offered by the QI project included good access to stakeholders and envisaged end users, co-producing relevant and impactful scenarios for experimentation, as well as access to good quality individual patient level data. On the other hand, substantial effort was required to populate input parameters with values that pertain to the natural history of the disease and the effectiveness of the different treatments. Our findings indicate that, if stakeholders require modelling results to inform aspects of a QI project, upfront investment is needed to ensure timely interaction between the two studies.

ABSTRACT We examine the information acquisition process regarding a patient’s status under emergency department (ED) congestion conditions. We focus on two key information channels: 1) Electronic Health Record (EHR) that provide the patient’s medical history and 2) Medical tests conducted in real-time. Whereas the EHR provides the physician with easily accessible information with little delay, real-time medical tests can provide more current information, but are time-consuming. We examine physicians’ decisions in cases of ED congestion, using a dataset that includes more than 1.4 million visits. When congestion is low, the information channels are complementary – acquiring information from the EHR is positively correlated with information acquisition from the medical tests channel, representing an incentive for the physician to acquire all possible information before providing diagnosis. However, as the congestion increases, there is less reliance on medical tests; this effect is amplified when EHR information is used. To avoid excessive congestion, physicians apparently refrain from sending patients for medical tests, and compensate for loss of information using EHR information. The impact of high system workload on the quality of medical service is an essential concern for managers; we show the indirect benefit of investment in EHRs through reduced blood-tests without increasing revisit rates.

ABSTRACT Preeclampsia, a life-threatening condition in late pregnancy, has unclear causes and risk factors. Machine learning (ML) offers a promising approach for early prediction. This systematic review analyzes state-of-the-art studies on preeclampsia prediction using ML approaches. We reviewed articles published between January 1 2013 and December 31 2023, from Google Scholar and PubMed. Of 183 identified studies, 35 were selected based on inclusion and exclusion criteria. Our findings reveal that key predictive features commonly used in machine learning models include age, number of pregnancies, body mass index, diabetes, hypertension, and blood pressure. In contrast, factors such as medications, genetic data, and clinical imaging were considered less frequently. Random Forest, Support Vector Machine, Logistic Regression, Decision Tree, and Naïve Bayes were the most commonly used algorithms. Most studies were conducted in China and the USA, indicating geographic concentration. The field has seen a notable rise in research, especially in the past two years, though many studies rely on small datasets from single hospitals. This review highlights the need for more diverse and comprehensive research to enhance early detection and management of preeclampsia.

ABSTRACT This study identifies user-preferred features in apps for self-management of chronic headaches from the biopsychosocial perspective and examines the extent to which such features are represented in the current apps. We first conducted semi-structured interviews to identify user-preferred features that tap into the biopsychosocial domains of health. Then, we conducted a landscape analysis to review existing apps with respect to the identified features. Our findings revealed participants’ preferences for features in apps to self-manage chronic headaches and were categorised based on the biopsychosocial model. Further, our landscape analysis showed that several features (e.g. physical symptom tracker) are highly present in existing apps, while most of the preferred features (e.g. journaling) are scarce. The identified features appear to align with theory-based behaviour change techniques and, therefore, have implications for health behaviour change. In addition, our findings demonstrate that most of the user-preferred biopsychosocial features are not widely present in the existing headache apps. Overall, our study highlights the importance of incorporating user-preferred features that align with the biopsychosocial needs of headache app users. By acknowledging and addressing these needs, we can broaden the existing perspectives concerning app design and evaluation and cater to the holistic health experiences of users.

ABSTRACT This paper presents a two-stage approach for efficiently solving a weekly home healthcare scheduling and routing problem. Two new mixed-integer linear programming (MILP) models are proposed, where the first is used for making patient-therapist assignments over the week, and the second for deriving daily routes. In both MILPs, the objective function contains a hierarchically weighted set of goals. The major components of the full problem are continuity of care, downgrading, workload balance, time windows, overtime, and mileage costs. A new preprocessing procedure is developed to limit the service area of each therapist to a single group of overlapping patients. Once the groups are formed, weekly schedules are constructed with the MILPs. The overall objective is to minimize the number of unscheduled visits and total travel and service costs subject to the operational constraints mentioned above. Computational experiments are conducted with real data sets provided by a national home health agency. The results show that optimal solutions can be obtained quickly at both the assignment and routing stages and that they are comparable to the results obtained with a proposed integrated model. In either case, the corresponding schedules were better on all metrics when compared to the schedules used in practice.

ABSTRACT Pay-for-performance (P4P) reimbursement models were launched in 2013 to incentivise the value of healthcare delivered by including quality outcomes, such as mortality, readmission, and patient satisfaction, in hospital reimbursement in the U.S. Although a decade has passed, the efficacy of these P4P programs remains unclear. This research intends to evaluate their long-term performance implications along two critical dimensions – productivity and healthcare value. Drawing on a nationwide sample of U.S. hospitals collected from 2008 to 2019, we utilise data envelopment analysis to measure hospital performance and the Malmquist index to evaluate their longitudinal trends. Although average hospital productivity and value improved since the rollout of the P4P programs, we observe that a large proportion of laggard hospitals were unable to catch up with improvements to the performance frontier, raising concerns about disparities in the impact of future value-based programs. Our analyses also indicate that horizontal integration across hospitals is associated with greater productivity and value. While greater physician-hospital (vertical) integration is associated with higher hospital productivity, it does not have a positive impact on value. Our study provides new insights into the antecedents and performance consequences of implementing value-based healthcare initiatives and their implications for hospital managers and policymakers.

ABSTRACT Elderly falls are occurring at an alarming rate, with significant health risks for seniors. Current fall detection systems often lack accuracy, efficacy, and privacy considerations. This study examines three leading human pose estimation frameworks combined with transformer deep learning models to develop a lightweight, privacy-preserving fall detection system. Key features include: 1) It runs on low-power devices like Raspberry Pis; 2) It monitors seniors passively, without requiring active participation; 3) It can be deployed in any residential or senior care setting; 4) It does not rely on wearables; and 5) All processing occurs locally, ensuring privacy with only fall alerts transmitted to caregivers. In real-world tests, the model achieved 95.24% sensitivity, 89.80% specificity, 98.00% accuracy, a 90.91% F1 score, and 95.24% precision, highlighting its effectiveness in detecting falls among the elderly while maintaining privacy and security.