Purpose: Predictive models for medical applications have shown promise, yet their clinical adoption remains limited due to concerns about prediction uncertainty, lack of interpretability, and limited reproducibility. Goal of this study was to design a deployable ML development framework that operationalizes trustworthy AI for clinical prediction. Furthermore, the study aimed to demonstrate applicability and validity of the framework in real world clinical practice by using it to build a patient no-show predictor in collaboration with a large pediatric hospital in the United States. Methods and Data: This study is guided by Design Science Research Methodology (DSRM) that is well-suited for research where the primary contribution is an artifact intended for real-world use. An ambulatory clinic patient no-show prediction application was chosen as the framework validation use case and was constructed using five years of electronic health record data (7.9 million appointments) from the collaborating hospital. Result: We designed and evaluated a novel socio-technical artifact—a trustworthy ML design framework (UML-Med) for clinical prediction—that addresses the trust gap in medical ML deployment. The resulting framework artifact (UML-Med) comprised of integrated uncertainly quantification module, multi-level explainability layer (local and global), reproducibility pipeline, structured evaluation governance, and an uncertainty-aware model selection method involving competing ML models based on performance-uncertainty tradeoff – not accuracy alone. All code and data are made publicly available to ensure full reproducibility. The UML-Med framework contributes as a generalizable development pipeline and a reusable blueprint for delivering trustworthy clinical prediction applications.

The pervasive integration of digital devices has fundamentally reshaped daily life, delivering numerous benefits while simultaneously posing challenges, including increased digital dependency and diminished well-being. Motivated by the principles of cyberpsychology, this study reframes the traditional view of digital detox by exploring "unplugging" as an emerging phenomenon aimed at mitigating the harmful effects of constant digital engagement. Despite its growing relevance, limited research exists on how unplugging can be intentionally practiced and what conditions foster its success. Using a two-phase methodology, we first apply Grounded Theory coding techniques to explore the problem space, analyzing 15 in-depth interviews to uncover underlying triggers, contextual factors, and behavioral patterns that motivate or act as barriers to digital disengagement. These findings inform a redefinition of unplugging as a purposeful, sustainable practice embedded in daily life and oriented toward long-term well-being. In the second phase, we employ Design Science Research (DSR) to develop and iteratively refine a conceptual framework intended to support strategies that reduce digital dependency and promote well-being among adults. This framework lays the groundwork for future research and the development of intervention artifacts that promote sustainable digital well-being.

There is considerable debate in IS-DSR community about what constitutes design knowledge and whether they are best represented as design principles or design theories. While designing an IT artifact to solve a problem addresses the “how” question, we agree that the “why” question (why did this work?) is important to answer. This essay attempts to shed light on two fundamental questions in this domain: (1) What are the various ways that knowledge contribution in DSR can be stated? (2) What pathways exists to create and re(use) knowledge in DSR projects? We present a simple 2 × 2 framework that explains with concrete examples four knowledge types - design principles, design attribute postulates, design theories, and good design practices from DSR projects. We further present a dual-use of design knowledge framework that shows pathways and re(use) of knowledge in DSR. We think that this typology is a useful theoretical contribution that can benefit both academics and practitioners conducting DSR projects.

BackgroundThe World Health Organization (WHO) reported that cardiovascular diseases (CVDs) are the leading cause of death worldwide. CVDs are chronic, with complex progression patterns involving episodes of comorbidities and multimorbidities. When dealing with chronic diseases, physicians often adopt a “watchful waiting” strategy, and actions are postponed until information is available. Population-level transition probabilities and progression patterns can be revealed by applying time-variant stochastic modeling methods to longitudinal patient data from cohort studies. Inputs from CVD practitioners indicate that tools to generate and visualize cohort transition patterns have many impactful clinical applications. The resultant computational model can be embedded in digital decision support tools for clinicians. However, to date, no study has attempted to accomplish this for CVDs. ObjectiveThis study aims to apply advanced stochastic modeling methods to uncover the transition probabilities and progression patterns from longitudinal episodic data of patient cohorts with CVD and thereafter use the computational model to build a digital clinical cohort analytics artifact demonstrating the actionability of such models. MethodsOur data were sourced from 9 epidemiological cohort studies by the National Heart Lung and Blood Institute and comprised chronological records of 1274 patients associated with 4839 CVD episodes across 16 years. We then used the continuous-time Markov chain method to develop our model, which offers a robust approach to time-variant transitions between disease states in chronic diseases. ResultsOur study presents time-variant transition probabilities of CVD state changes, revealing patterns of CVD progression against time. We found that the transition from myocardial infarction (MI) to stroke has the fastest transition rate (mean transition time 3, SD 0 days, because only 1 patient had a MI-to-stroke transition in the dataset), and the transition from MI to angina is the slowest (mean transition time 1457, SD 1449 days). Congestive heart failure is the most probable first episode (371/840, 44.2%), followed by stroke (216/840, 25.7%). The resultant artifact is actionable as it can act as an eHealth cohort analytics tool, helping physicians gain insights into treatment and intervention strategies. Through expert panel interviews and surveys, we found 9 application use cases of our model. ConclusionsPast research does not provide actionable cohort-level decision support tools based on a comprehensive, 10-state, continuous-time Markov chain model to unveil complex CVD progression patterns from real-world patient data and support clinical decision-making. This paper aims to address this crucial limitation. Our stochastic model–embedded artifact can help clinicians in efficient disease monitoring and intervention decisions, guided by objective data-driven insights from real patient data. Furthermore, the proposed model can unveil progression patterns of any chronic disease of interest by inputting only 3 data elements: a synthetic patient identifier, episode name, and episode time in days from a baseline date.

Cancer caregivers are often informal family members who may not be prepared to adequately meet the needs of patients and often experience high stress along with significant physical, emotional, and financial burdens. Accurate prediction of caregiver's burden level is highly valuable for early intervention and support. In this study, we used several machine learning approaches to build prediction models from the National Alliance for Caregiving/AARP dataset. We performed data cleansing and imputation on the raw data to give us a working dataset of cancer caregivers. Then a series of feature selection methods were used to identify predictive risk factors for burden level. Using supervised machine learning classifiers, we achieved reasonably good prediction performance (Accuracy ∼ 0.94; AUC ∼ 0.97; F1∼ 0.93). We identify a small set of 15 features that are strong predictors of burden and can be used to build Clinical Decision Support Systems.

Sensors embedded in smart objects, smart machines, and smart buildings produce ever-growing streams of contextual data that convey information of interest about their operating environment. Although an increasing number of industries have embraced the utilization of sensors in routine operations, no clear framework is available to guide designers who aim to leverage contextual data collected from these sensors to develop predictive systems. In this paper, we applied design science research methodology to develop and evaluate a general framework that can help designers build predictive systems utilizing sensor data. Specifically, we developed a framework for designing context-aware predictive systems (CAPS). We then evaluated the framework through its application in MAN Diesel & Turbo, which served as a case company. The framework can be generalized into a class of demand-forecasting problems that rely on sensor-generated contextual data. The CAPS framework is unique and can help practitioners make better-informed decisions when designing context-aware predictive systems.

The papers in this special section focus on design science research in information systems and technology. The rapid digital transformation of businesses and society creates new challenges and opportunities for information systems (IS) research, with the designing of system becoming a strong focus. Technologies, such as blockchain, cloud computing, and wearable devices, have spawned a paradigmatic shift in the design of IS. Thus, design science research in IS has evolved to become a mature yet thriving area of research. Design science seeks to extend the boundaries of human and organizational capabilities by designing new and innovative constructs, models, methods, processes, and systems. Scholars having diverse backgrounds in fields, such as IS, computer science, software engineering, energy informatics, and medical informatics, are actively engaged in generating novel solutions to interesting design problems in IS.

Abstract According to the Centers for Disease Control and Prevention, in the United States, about 80% of older adults - individuals aged 65 years and older, have at least one chronic health condition. Chronic diseases require ongoing medical attention or limit activities of daily living, which in turn, affects quality of life. Heart diseases, cancer, and diabetes are the leading chronic diseases. Literature shows that timely and frequent follow-ups by providers significantly reduce readmissions of chronic patients and contribute to their well-being. However, health care provides do not have adequate bandwidth for regular patient follow-ups, which is vital for chronic patients. One solution for this growing problem, is to leverage smart home monitoring technologies facilitating self-care, which in turn can contribute to patients’ well-being. The goal of this research is to build and leverage an IoT-based smart home monitoring platform and assess its impact on the well-being of chronic patients. The platform included an mHealth mobile app with daily journaling capability, which allowed patients to blog about their symptoms, daily activities, and feelings. Analyzing patients’ daily blog data with Natural Language Processing (NLP) methods, this research revealed a strong correlation between a patients’ blog entries and their well-being score. The results also demonstrated that patients’ well-being scores can be predicted from such journal entries analyzed at real time, collected remotely from their home environment, leading to timely diagnosis and intervention opportunities. A field trial was conducted involving five heart disease and eight cancer patients for 180 days to validate the conclusions.

In the last decade, the need for smart-space design has been on the rise. Various data collected from Internet-of-Things (IoT) and sensors are used to optimize the operation of smart spaces, which, in urban areas, are evolving into smart cities. How can smart spaces provide value to citizens? There is a need to develop smart services that leverage emerging technologies while taking an inclusive and empowering approach to the inhabitants. To address this need, we present a framework for designing smart spaces and we use a bottom-up (inclusive) approach to instantiate a smart kiosk (SK). The SK prototype provides a practical approach for transforming a traditional building into a smart space utilizing IoT and artificial intelligence technologies. The design science research (DSR) methodology was followed for designing and evaluating the prototype. An iterative process that involves occupant feedback and brainstorming sessions coupled with a literature review was carried out to identify the issues and services related to a smart building. The SK prototype implements three intelligent services that were prioritized by the citizens of the building. The results show that the SK has a high usage and acceptance rate and it can transform a lobby into a highly engaged and smart building space. The prototyping process suggests important factors to ideate and assess smart services and shows that small-scale projects can be successful to enable smart buildings. The framework provides a theoretical contribution while the design and development process assists practitioners in identifying and developing intelligent services based on IoT technology.

BackgroundMost people with chronic conditions fail to adhere to self-management behavioral guidelines. In the last 2 decades, several mobile health apps and IT-based systems have been designed and developed to help patients change and sustain their healthy behaviors. However, these systems often lead to short-term behavior change or adherence while the goal is to engage the population toward long-term behavior change. ObjectiveThis study aims to contribute to the development of long-term health behavior changes or to help people sustain their healthy behavior. For this purpose, we built and tested a theoretical model that includes enablers of empowerment and an intention to sustain a healthy behavior when patients are assisted by information and communications technology. MethodsStructural equation modeling was used to analyze 427 survey returns collected from a diverse population of participants and patients. Notably, the model testing was performed for physical activity as a generally desirable healthy goal. ResultsMessage aligned with personal goals, familiarity with technology tools, high self-efficacy, social connection, and community support played a significant role (P<.001) in empowering individuals to maintain a healthy behavior. The feeling of being empowered exhibited a strong influence, with a path coefficient of 0.681 on an intention to sustain healthy behavior. ConclusionsThe uniqueness of this model is its recognition of needs (ie, social connection, community support, and self-efficacy) to sustain a healthy behavior. Individuals are empowered when they are assisted by family and community, specifically when they possess the knowledge, skills, and self-awareness to ascertain and achieve their goals. This nascent theory explains what might lead to more sustainable behavior change and is meant to help designers build better apps that enable people to conduct self-care routines and sustain their behavior.

In this paper, we address the challenge of how social entrepreneurs can create a business model that can generate revenue while at the same time have an impact on society. We describe a novel methodology that we call Models of Impact (MOI) along with a toolkit. MOI has been used by thousands of entrepreneurs worldwide who have found it to be extremely useful. We present its usefulness and efficacy through a variety of use case studies.

There is an exceptionally high rate of readmissions and rehospitalizations for patients suffering from Heart Failure. Best efforts to address this alarming problem from the Caregiver community have fallen short due to a shortage of trained clinical staff, failure to perform necessary self-management, and money. Using a Design Science Research framework, this work designed and evaluated "DIL” (Sanskrit word for Heart), a Conversational Agent that complements the work of clinicians in achieving the desired behavioral and clinical outcomes. The aim is to provide the hospital with an information system that could bridge the current gap in care that occurs when the patient transitions from the hospital to the home environment. In a pilot study, we show that DIL was able to demonstrate the efficacy and utility as a tool to assist patients with heart failure in improving their self-care.

Firearm mortality has been a constant issue plaguing the United States for many decades. Unfortunately, with the accelerated spread of the novel virus COVID-19, those who live in America have to face the additional possibility of death. We investigated the impacts of these challenges at the state level. Specifically, we looked for potential trends in fatality rate from firearms and COVID-19 since the spread of the virus. Additionally, we carried out stepwise regression analysis to assess the state characteristics that most closely associated with the highest rates of deaths from gun violence, and what we found adds evidence and support for the argument that gun control laws save lives. We found that the number of gun laws, and the ideological leaning of states had statistically significant associations with the level of gun deaths. We also ranked states based on the severity of their level of firearm and COVID-19 fatality. © AMCIS 2021.

There is an exceptionally high rate of readmissions and rehospitalizations for patients suffering from Heart Failure. Best efforts to address this alarming problem from the Caregiver community have fallen short due to a shortage of trained clinical staff, failure to perform necessary self-management, and money. Using a Design Science Research framework, this work designed and evaluated "DIL” (Sanskrit word for Heart), a Conversational Agent that complements the work of clinicians in achieving the desired behavioral and clinical outcomes. The aim is to provide the hospital with an information system that could bridge the current gap in care that occurs when the patient transitions from the hospital to the home environment. In a pilot study, we show that DIL was able to demonstrate the efficacy and utility as a tool to assist patients with heart failure in improving their self-care.

Abstract The story of John Henry, the “steel-drivin’ man”, is well known to Black men in the United States. John Henry is considered a hero because he demonstrated tremendous strength and self-determination. The MANUP diabetes program used the John Henryism, defined as high-effort active coping in the face of adversity, as the basis of a diabetes intervention for Black men. MANUP conducted four community-based focus groups identifying topics of concern to Black men with type 2 diabetes (T2D). Interestingly, the men reported that high-effort active coping was crucial for successful diabetes self-management. MANUP then developed and implemented a longitudinal culturally targeted self-management program for 33 Black men with T2D in Flint, Michigan. MANUP included discussion groups, physical activity, and an app incorporating text-messaging, group-chat, and a blood glucose monitoring dashboard to improve glycemic control (A1c). This single-group, repeated measures intervention assessed A1c three times over a six-month period. Improvements in A1c were observed at: baseline – time 2: 8.9% vs 8.6%, p=0.14; time 2 – time 3: 8.6% vs 8.1%, p=0.21; and baseline – time 3: 8.9% vs 8.1%, p=0.005. After controlling for age and insulin use, the significant reduction in A1c over 6 months remained (p=0.01). These findings demonstrate that combining mobile health technology and moderate physical activity with culturally targeted discussion topics can improve T2D self-management and reduce A1c in Black men. More community-driven longitudinal intervention studies that improve diabetes self-management among Black men are needed to achieve gender and racial health equity..