Search

With the growing popularity of content-sharing platforms, patients are increasingly using the Internet as a critical source of health information. As one of the most popular video-sharing sites, YouTube provides easy access to health information seekers, but it is difficult and time-consuming to identify and retrieve high-quality videos that may serve as engaging patient education materials. This paper reports on an exploratory analysis of 317 YouTube videos on Obstructive Sleep Apnea (OSA) to better understand some key features of the videos and the relationships between them to facilitate subsequent video classification and recommendation. Features intrinsic to a video, such as video duration, and extrinsic, such as the number of views, are analyzed using unsupervised clustering methods and the Sankey diagram to discover the relationship between the clusters and their significance across different clusters, providing promising insights for the assessment of video quality.

The COVID-19 pandemic has highlighted the dire necessity to improve public health literacy for societal resilience. YouTube provides a vast repository of user-generated health information in a multi-media-rich format which may be easier for the public to understand and use if major concerns about content quality and accuracy are addressed. This study develops an automated solution to identify, retrieve and shortlist medically relevant and understandable YouTube videos that domain experts can subsequently review and recommend for disseminating and educating the public on the COVID-19 pandemic and similar public health outbreaks. Our approach leverages domain knowledge from human experts and machine learning and natural language processing methods to provide a scalable, replicable, and generalizable approach that can also be applied to enhance the management of many health conditions.

The Cascade-HF protocol is a Continuous Remote Patient Monitoring (CRPM) study at a major health system in the United States to reduce Heart Failure (HF)-related hospitalizations and readmissions using wearable biosensors to collect physiological data over a 30-day period to determine decompensation risk among HF patients. The alerts produced, coupled with electronic patient-reported outcomes, are utilized daily by the home health team, and escalated to the heart failure team as needed, for proactive actions. Limited research has examined anticipating the implementation and workflow challenges of such complex CRPM studies such as resource planning and staffing decisions that leverage the recorded data to drive clinical preparedness and operational efficiency. This preliminary analysis applies discrete event simulation modeling to the Cascade-HF protocol using pilot data from a soft launch to assess workload of the clinical team, evaluate escalation patterns and provide decision support recommendations to enable scale-up for all post-discharge patients.

Several indices exist to classify Congestive Heart Failure (CHF) patients' propensity for early mortality; however, they are primarily based on limited data and are not intuitive to use at the point of care. We investigate a novel, data-driven, risk assessment and visualization approach to investigate mortality prediction of CHF patients using data retrieved from an intensively digitized hospital's data repository. Combining well-known, computationally efficient, dimensionality reduction (DR) methods with 2-d information visualization, the method classifies and visualizes CHF patients into high and low risk groups, contextualized by the factors driving their classification. The DR method performed similar to logistic regression (LR), but visualized the classification and its significant factors at the population level, individual level and the potential impact of interventions for an individual patient. These are encouraging results in favor of the proposed visualization approach, and contributes to the current focus on advancing patient care via large-scale visual analytics.

Patients with multiple chronic conditions (MCC) pose an increasingly complex health management challenge worldwide, particularly due to the significant gap in our understanding of how to provide coordinated care. Drawing on our prior research on learning data-driven clinical pathways from actual practice data, this paper describes a prototype, interactive platform for visualizing the pathways of MCC to support shared decision making. Created using Python web framework, JavaScript library and our clinical pathway learning algorithm, the visualization platform allows clinicians and patients to learn the dominant patterns of co-progression of multiple clinical events from their own data, and interactively explore and interpret the pathways. We demonstrate functionalities of the platform using a cluster of 36 patients, identified from a dataset of 1,084 patients, who are diagnosed with at least chronic kidney disease, hypertension, and diabetes. Future evaluation studies will explore the use of this platform to better understand and manage MCC.

Pediatric obesity is a growing epidemic, with unhealthy eating habits and poor physical activity being major contributors. While video and mobile games have been shown to have a positive impact on behavior change in children, the mechanisms underlying game play that impact outcomes of interest are poorly understood. This research aims to examine the impact of a novel mobile gaming app on the design of behavioral interventions by learning from the rich and unique game telemetry generated from a randomized controlled trial of the app use by school children. In this exploratory analysis, we extract a partial dataset to build and analyze chronological sequences of game plays to understand key patterns in the game mechanics that players utilize as they navigate the game, and possible implications of the results.

In this paper, we describe the development of a unified framework and a digital workbench for the strategic, tactical and operational hospital management plan driven by information technology and analytics. The workbench can be used not only by multiple stakeholders in the healthcare delivery setting, but also for pedagogical purposes on topics such as healthcare analytics, services management, and information systems. This tool combines the three classical hierarchical decision-making levels in one integrated environment. At each level, several decision problems can be chosen. Extensions of mathematical models from the literature are presented and incorporated into the digital platform. In a case study using real-world data, we demonstrate how we used the workbench to inform strategic capacity planning decisions in a multi-hospital, multi-stakeholder setting in the United Kingdom.

The problem of readmission, wherein patients are readmitted for the same or a related condition shortly after discharge, has become a challenge worldwide from care quality and financial perspectives. In this study, we explore 30-day readmission data for predicting who is likely to be readmitted and understanding key factors contributing to preventable readmissions using the developmental trajectory of creatinine level as a key laboratory marker of serious illness and a potential predictor of future readmission. Using Electronic Health Record data on 928 patients over ten different visits to emergency departments across Israel, we apply a semi-parametric, statistical, group-based trajectory model to elicit three distinct creatinine-based trajectories over time with differing 30-day readmission rates for males and females. Analysis of readmission risk stratification of the patient population using other relevant factors is ongoing research.

Periodontal disease is a major cause for tooth loss and adversely affects individuals' oral health and quality of life. Research shows its potential association with systemic diseases like diabetes and cardiovascular disease, and social habits such as smoking. This study explores mining potential risk factors from dental electronic health records to predict and display patients' contextualized risk for periodontal disease. We retrieved relevant risk factors from structured and unstructured data on 2,370 patients who underwent comprehensive oral examinations at the Indiana University School of Dentistry, Indianapolis, IN, USA. Predicting overall risk and displaying relationships between risk factors and their influence on the patient's oral and general health can be a powerful educational and disease management tool for patients and clinicians at the point of care.

The objective of this paper is to evaluate the extent to which early determination of diagnosis-related groups (DRGs) can be used for better allocation of scarce hospital resources. When elective patients seek admission, the true DRG, currently determined only at discharge, is unknown. We approach the problem of early DRG determination in three stages: (1) test how much a Naïve Bayes classifier can improve classification accuracy as compared to a hospital's current approach; (2) develop a statistical program that makes admission and scheduling decisions based on the patients' clincial pathways and scarce hospital resources; and (3) feed the DRG as classified by the Naïve Bayes classifier and the hospitals' baseline approach into the model (which we evaluate in simulation). Our results reveal that the DRG grouper performs poorly in classifying the DRG correctly before admission while the Naïve Bayes approach substantially improves the classification task. The results from the connection of the classification method with the mathematical program also reveal that resource allocation decisions can be more effective and efficient with the hybrid approach.

Telephones are a ubiquitous and widely accepted technology worldwide. The low ownership cost, simple user interface, intuitive voice-based access and long history contribute to the wide-spread use and success of telephones, and more recently, that of mobile phones. This study reports on our preliminary efforts to leverage this technology to bridge disparities in the access to and delivery of personalized health and wellness care by developing and evaluating a Spoken Web enabled Care Management solution. Early results with two proxy evaluations and a few visually impaired users highlight both the potential and challenges associated with this novel, voice-enabled healthcare delivery solution.

Principles and theories of behavioral decision-making (BDM) are applicable to any field, but they have special import for health care, where low-impact but repeated decisions can significantly affect quality of life. Unfortunately, limited research has focused on effective behavioral design for personal health management mobile apps. This paper addresses patient engagement approaches using the National e-Health Collaborative's "Patient Engagement Framework" and outlines specific implementations of behavioral decision theory to create an effective system for self-care.

Higher cognitive workload due to poor usability is a significant, unanticipated consequence of healthcare information technology (IT), resulting in new types of medical errors. An important example of this can be observed in the use of order sets, which allow safe and efficient provider order entry guided by known best practices. This paper aims to improve IT-enabled order entry by re-designing order sets using data-driven approaches to develop new order sets that match current usage and workflow, while incurring minimum cognitive workload. Applying optimization models embedded with clustering techniques, our methods identify items for constituting order sets that are relevant based on historical ordering data wherein items for a single patient are often placed together or in close temporal proximity during hospital stay. Results indicate that the new approaches dominate current solutions, significantly reducing cognitive workload, and improving order set content. Data driven methods thus offer a promising approach for designing order sets that are generalizable, evidence-based and up-to-date with current best practices.

According to WHO, pediatric diabetes is a rising global public health problem, with increasing impact on developing nations. This study summarizes a multidimensional, scalable pilot evaluation of a diabetes self-management platform combining mobile technology with social networking to capture four key metrics of Type 1 diabetes self-management, associated social interactions, and gaming features providing targeted feedback to 8 pediatric users. Based on their 2-month interaction with the application, we analyze click-stream data from social interactions, key health metrics, text comments, and usability and satisfaction surveys to evaluate engagement with the platform and effectiveness in controlling blood glucose using a product-process-program framework. Our preliminary results indicate that this framework was successful in demonstrating the potential of the mobile health platform to effectively leverage the growing use of mobile applications and social media to present a unique benefit that engaged pediatric users and provided useful insights for self-health management.

Ultrasound is a low cost and efficient method of detecting diseases and abnormalities in the body. Yet there is a lack of precision and reliability associated with the technology, partly due to the operator dependent nature of ultrasound scanning. When scanning is performed to an agreed protocol, ultrasound has been shown to be highly reliable. This research aims to minimize these limitations that arise during ultrasound training, scanning and reporting by developing and evaluating an image analysis and decision support system that can aid the decision making process. We hypothesize that this intervention will likely increase the role of ultrasound in diagnosis when compared with other imaging technologies, particularly in low resource settings.

Healthcare is primarily delivered in the ambulatory care setting worldwide. The high variability in service delivery encountered in this environment negatively impacts process efficiency and patient satisfaction. In this study, we analyze care delivery process in ambulatory care using time and location stamped data collected via Radio Frequency Identification (RFID)-enabled badges worn by patients, clinicians, and staff as they complete each clinic visit. With the objective of improving process visibility and minimizing patient waiting time, we examine this data to delineate the major components of waiting time and use simulation modeling to evaluate the impact of possible interventions. Results indicate that as a prevalent strategy, different appointment scheduling rules can only reduce patient waiting time in the waiting room. Surprisingly, waiting time in the exam room is unchanged, requiring new approaches to improve care coordination that address this delay. The results also highlight the value of RFID technology and the challenges in deploying them to improve service delivery.

Mounting evidence suggests a potential association of periodontal disease with systemic diseases such as diabetes, cardiovascular disease, cancer and stroke. The objective of this study is to develop an integrated risk assessment tool that displays a patients' risk for periodontal disease in the context of their systemic disease, social habits and oral health. Such a tool will be used by not just dental professionals but also by care providers who participate in the team-based care for chronic disease management. Displaying relationships between risk factors and its influence on the patient's general health could be a powerful educational and disease management tool for patients and clinicians. It may also improve the coordination of care provided by the provider-members of a chronic care team.

Patient online eVisits are gaining momentum due to increasing consumer demand for improved access to clinical services, availability of new technologies to deploy such services and development of reimbursement initiatives by major payers. The eVisit service provides patients with an online consultation through a series of structured, secure message exchanges with a physician, providing an alternative for onsite office visits and non-reimbursed phone-based care. In this study, we evaluate a pilot deployment of eVisits in a primary care clinic providing online consultation service for 7 simple health conditions at its three locations. We examine usage data over 3 months and survey and interview results for trends in adoption, demographic and temporal patterns of usage, clinician and patient expectations and experiences, and challenges to sustainability of the service. Based on our analysis, we conclude that the eVisit pilot was a success. Patients valued the new service being offered as demonstrated by a rapid increase in usage. The quality of service was good with fast turnaround times and few exchanges to resolve a request. These positive outcomes combined with a reimbursement model are promising indications of sustainability but several challenges remain.

Many information technology-enabled healthcare applications have failed because their interfaces are difficult to use. Unfortunately, little attention has been paid in the health informatics community to designing effective user interfaces that are acceptable to healthcare professionals. This paper illustrates a method for improving application interface usability by applying sequential pattern analysis to analyze temporal event sequences recorded in an electronic medical record system. Such event sequences, or clickstreams, reflect clinicians' navigation patterns in their everyday interactions with the computer system. The identified patterns have been used by software developers to calibrate the user interface of the system, so that the within-application workflow is better aligned with clinicians' mental model of medical problem-solving. Such inferred patterns may also help to modify clinicians' suboptimal practice behavior components, as manifested through their actual usage of this point-of-care electronic system.

Data sets with many discrete variables and relatively few cases arise in health care, commerce, information security, and many other domains. Learning effective and efficient prediction models from such data sets is a challenging task. In this paper, we propose a new approach that combines Metaheuristic search and Bayesian Networks to learn a graphical Markov Blanket-based classifier from data. The Tabu Search enhanced Markov Blanket (TS/MB) procedure is based on the use of restricted neighborhoods in a general Bayesian Network constrained by the Markov condition, called Markov Blanket Neighborhoods. Computational results from two real world healthcare data sets indicate that the TS/MB procedure converges fast and is able to find a parsimonious model with substantially fewer predictor variables than in the full data set. Furthermore, it has comparable or better prediction performance when compared against several machine learning methods, and provides insight into possible causal relations among the variables.

In this study, we assess 41 medical residents' acceptance and adoption of a clinical reminder system for chronic disease and preventive care management in the ambulatory care environment using a novel developmental trajectory approach. This group-based, semi-parametric statistical modeling method identifies distinct groups, following distinct usage trajectories, among those who recorded use of the reminder system within an evaluation period of 10 months. We trace system use within these groups over time using computer-generated logs and user satisfaction surveys. Our preliminary analysis of this small sample of users delineates three categories of users. Feedback from these categories of users is being used to re-engineer the application and adapt it better to their workflow and functionality requirements. Despite the small sample size in this particular study, we conclude that this methodology has considerable promise to provide new insights into system usability and adoption issues that may benefit clinical decision support systems as well as information systems more generally.

Regional healthcare initiatives seek to improve the quality of healthcare by collecting, analyzing, and disseminating information about chronic diseases such as diabetes. The data required to support such initiatives comes from several organizations such as insurers, physicians, hospitals, pharmacies and labs each of which gather and maintain data for the purpose of healthcare delivery. In this paper, we focus on mediator-based architectures and the privacy problems that arise in the healthcare context owing to the linkage of information about patients, physicians, and diseases enabled by the mediator. In particular, we examine privacy issues for the two separate steps of the actual data release. First, raw data is released to the (not necessarily trustworthy) mediator and second, the mediator creates and releases the health report. For both steps, we present a technical solution that permits the final report to be useful to the user while respecting the data owners' privacy.