Cognitive Informatics in Biomedicine and Healthcare Pei-Yun Sabrina Hsueh Thomas Wetter Xinxin Zhu Editors Personal Health Informatics Patient Participation in Precision Health Cognitive Informatics in Biomedicine and Healthcare Series Editor Vimla L. Patel, Ctr Cognitive Studies in Med & PH New York Academy of Med, Suite 454, New York, NY, USA Enormous advances in information technology have permeated essentially all facets of life. Although these technologies are transforming the workplace as well as leisure time, formidable challenges remain in fostering tools that enhance productivity, are sensitive to work practices, and are intuitive to learn and to use effectively. Informatics is a discipline concerned with applied and basic science of information, the practices involved in information processing, and the engineering of information systems. Cognitive Informatics (CI), a term that has been adopted and applied particularly in the fields of biomedicine and health care, is the multidisciplinary study of cogni- tion, information, and computational sciences. It investigates all facets of computer applications in biomedicine and health care, including system design and computer- mediated intelligent action. The basic scientific discipline of CI is strongly grounded in methods and theories derived from cognitive science. The discipline provides a framework for the analysis and modeling of complex human performance in tech- nology-mediated settings and contributes to the design and development of better information systems for biomedicine and health care. Despite the significant growth of this discipline, there have been few systematic published volumes for reference or instruction, intended for working professionals, scientists, or graduate students in cognitive science and biomedical informatics, beyond those published in this series. Although information technologies are now in widespread use globally for promoting increased self-reliance in patients, there is often a disparity between the scientific and technological knowledge underlying healthcare practices and the lay beliefs, mental models, and cognitive representa- tions of illness and disease. The topics covered in this book series address the key research gaps in biomedical informatics related to the applicability of theories, models, and evaluation frameworks of HCI and human factors as they apply to clini- cians as well as to the lay public. Pei-Yun Sabrina Hsueh • Thomas Wetter Xinxin Zhu Editors Personal Health Informatics Patient Participation in Precision Health Editors Pei-Yun Sabrina Hsueh Thomas Wetter Bayesian Health Heidelberg University Hospital New York, NY, USA Institute of Medical Informatics Heidelberg, Germany Xinxin Zhu Biomedical Informatics and Medical Center for Biomedical Data Science Education Yale University University of Washington New Haven, CT, USA Seattle, WA, USA ISSN 2662-7280 ISSN 2662-7299 (electronic) Cognitive Informatics in Biomedicine and Healthcare ISBN 978-3-031-07695-4 ISBN 978-3-031-07696-1 (eBook) https://doi.org/10.1007/978-3-031-07696-1 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Foreword We live in turbulent times. The pace of technological innovation around us is relent- less. Its impacts on society, some expected some not, are reshaping the way we engage with each other and with once trusted monoliths like the healthcare system. Washing over all this tumult are the far reaching and still unrevealed consequences of climate change and what may well be an age of pandemics. Society will not be the same, no matter how we struggle to engineer that it will. It is no surprise then that as individuals we all want more control, and more understanding of the direction of our health care. Many expect to be equal partners with clinicians in healthcare decisions. We want to know how vaccines work, and the relative risk of one vaccine strategy over another, before we decide to follow public health recommendations. We want to know that our disease treatments are safe, and for major illness many want to leave no option unexplored. Some of us go one step further and want to be the decision-maker, relegating trained professionals to advisers. The challenges we face as engaged patients however are formidable. If it is truly not possible for a clinician to be up to date with medical science because the pace of innovation is so frenetic, what chance is there for a patient? When in doubt, our human response is to speak to others and understand their own journey and deci- sions—but that strategy can be risky if we use social media to ask the questions. The promised bounty of the internet, connection to all people and all things, is increas- ingly tarnished by community polarization, lack of trust, and skepticism in science. Truth itself is often victim to disinformation campaigns created for unclear motive. It is with this challenging context that we must try not just to understand the sweeping changes that are happening in health care, but to get ahead of them. We are tasked to imagine ways of rebalancing the new dynamic of the world by giving patients and clinicians new options to succeed. What skills and tools must we invent so that patients can both find the information they need and understand the underly- ing science behind it—itself ever mutating because of technological innovation? What new conceptions of communication and relationships will help patients and their professional carers work together with trust when the old options fall away? v vi Foreword Some of the answers are social and cultural, some are to be found in the embrace of new models of care, where one face-to-face interaction is replaced by a multitude of small digital touches. Some technological answers bring both revolution and their own burden of problems. Artificial intelligence will likely transform the way many patients manage their care. Personal AIs will be our first port of call, noticing health problems emerge before we do, suggesting pathways for care shaped by our past preferences, and coordinating that journey for us. AIs will not just be digital front-doors for us into the health system, they will be networked into it, creating what is known as a cyber-social system. These AIs will be connected into our clothes, our wearable devices and jewelry, our home, and our cars. AIs will negoti- ate, search, recommend, optimize, and disagree—all on our behalf and not always transparently or fairly. A new age with new opportunity, and a new bag of chal- lenges and decisions for us to face. In the pages of this book, you will find many of the elements of this next stage in our journey towards a dynamic health system that better fits the challenges ahead. New ways of interacting, new ways of behaving, and new tools for thinking are all part of the solution. The chapters in this book also make clear how early we are in our journey to truly embrace what is to come. The healthcare systems of different countries are all unique, but in one way are all the same—they are monolithic in behavior and slow to change. It is often said that revolution cannot come from within old large organizational structures. Revolution comes from the boundaries of the old hegemony and the wild borderlands. So maybe you need to read these chap- ters not as stories from the old healthcare empire, but as reports from the wild bor- der country of patient-led change. Australian Institute of Health Innovation Enrico Coiera Macquarie University Sydney, NSW, Australia [email protected] Preface Overview The world of health informatics is constantly changing given the ever-increasing variety and volume of health data, care delivery models that shift from fee-for- service to value-based care, new entrants in the ecosystem, and the shifting regula- tory decision landscape. In the area of cognitive informatics, the changes have increased the importance of the role of patients in research studies for understand- ing work processes and activities within the context of human cognition, as well as the design and implementation of health information systems (Haldar et al. 2020; Trevor Rohm 2010). Therefore, personal health informatics, in recent years, has risen up to provide research tools and protocols to zoom into individual health- related contexts when developing engineering, computing, and service solutions that can improve clinical practice, patient engagement, and public health (Hsueh et al. 2017a; Hsueh et al. 2017b; Lai et al. 2017; Patel and Kannanmpallil 2015; Reading and Merrill 2018). This is particularly important to bridge the previous gaps in patient-provider information (Tang and Lansky 2005). The rise of personal health informatics is also in line with the emerging utiliza- tion of real-world evidence generated from real-world data. Here, “real-world data” (RWD) refers to data generated from the actual practice and delivery of health care (e.g., electronic health records, insurance claims, disease registries), while “real- world evidence” (RWE) refers to the inferences made from RWD. In some areas such as clinical trial design, RWE has been successfully applied to bring greater efficiency to the development of clinical programs (e.g., as its external control arm as indicated by the FDA’s Real-World Evidence Framework (FDA, n.d.) and received regulatory approval (Berger et al. 2016; Miksad and Abernethy 2018; Shah et al. 2019). In many other areas, the development of personal health informatics has opened up new opportunities to generate RWE through integrating data science with the science of care (Bica et al. 2020; Hsueh et al. 2018). Through this book, we intend to compile a collection of high-quality scholarly work that seeks to provide clarity, consistency, and reproducibility, with an updated vii viii Preface and shared view of the status quo of consumer and pervasive health informatics and its relevance to precision medicine and healthcare applications. The new term “Personal Health Informatics” is being proposed to cover a broader definition of this emerging field. In one way or another, individuals are not just consuming health; they are active participants, researchers, and designers in the healthcare ecosystem. The book will offer a snapshot of this emerging field, supported by the method- ological, practical, and ethical perspectives from researchers and practitioners. In addition to being a research reader, this book will provide pragmatic insights for practitioners in designing, implementing, and evaluating personal health informat- ics in healthcare settings. The volume will also be an excellent reader for students in all clinical disciplines as well as in biomedical and health informatics to learn from case studies in this emerging field. This is a starting point for us to show the direction where the whole field is going. The chapters include (1) case studies including reflections on implementation les- sons learned, (2) theoretical frameworks, (3) design methodologies, and (4) evalua- tion and critical appraisal. These chapters will be organized under four main sections: (1) the state-of-the- art novel care delivery models (using case study examples), (2) methods for trans- lating biomedical research and RWE into patient-centric precision health application, (3) methods for patient-centric design, and (4) ethics, bias, privacy, and fairness. Chapter authors have been invited based on their reputation and fit with the topic, and all chapters have been reviewed independently. This book is intended to appeal to a wide range of audiences including academic researchers, educators, professional informaticians, healthcare providers and administrators, healthcare consumers, and policymakers. Although this book is not considered a standard textbook, it will be of great value for graduate programs in which courses in applied informatics are relevant, such as courses that focus on behavioral, cognitive, and social aspects of health information technology. Section I: The State-of-the-Art Novel Care Delivery Models Since its inception—then by the name Consumer Health Informatics–about 25 years ago the field of Personal Health Informatics (PHI) has achieved worldwide reach across many health problems and clinical disciplines. The book presents ser- vices from approximately 20 countries and several methodological chapters with international reach. The authors provide us with a different lens to observe different target populations, learn about various methods of deployment, and how they fit with or transgress from present models of healthcare delivery. Kuziemsky et al. in their Chap. 1, “E-enabled Patient-Provider Communication in Context,” present examples of enhanced patient-provider communication from Denmark, Fiji, Columbia, and Canada. In Denmark, the emphasis is on overcom- ing the intermittent and fragmented practice of care for patients with Chronic Obstructive Pulmonary Disease. In the EHealth Care Model (ECM) patients Preface ix receive basic measurement and problem staging equipment for their home envi- ronment that connects them continuously, 24/7, to the same care team. Adaptations of the treatment regime can be initiated through the distance before emerging problems exacerbate. The project from Fiji addresses mental health problems for which specialists are extremely rare and care is mostly provided through nurses. To lift the level and guideline adherence of treatments, an educational effort was launched which compared the presentation of guideline knowledge through differ- ent media. A smartphone-b ased study arm showed the best results. In Columbia, similar to many low and middle-income countries, perinatal mortality is still high. In addition, the involvement of stakeholders hinders timely intervention during pregnancy. The presented project aims to technically support interventions sug- gested to the gynecologist through AI-driven clinical decision support. The respec- tive system needs to technically integrate with clinical workflows and is supposed to communicate with expecting mothers through their smartphones. The Canadian example studies collaboration through the macro, meso, and micro levels. It emphasizes the dynamic nature of collaboration which showed intensely in the transition of care with the onset of the COVID pandemic. New tools were sponta- neously adopted by individual providers and patients, posing organizational and data security challenges at the meso level. The challenges thus call for the reform at the national level on data standards, new legislation, and billing codes, to name just a few. In Chap. 2, “Direct Primary Care: A New Model for Patient-Centered Care,” Snowdon et al. lay out one specific novel model of Direct Primary Care (DPC). The DPC model aims to improve the quality of primary care through better patient- provider relations and communication. The payment model of DPC is based on a per capita fee per time period to a provider or its provider organization. It can be either paid by the patients or by their employers. Compared to the fee-for-service payment model, the value-based model can help incentivize providers to spend more quality time with the patient and promote prevention. For patients, it also removes financial and organizational barriers to healthcare access, which in turn leads to rational and sustainable utilization of primary care services. This chapter also helps the readers learn to distinguish the implementation needed at different scales, and how the implementation of DPC improves utilization of designated pre- ventive measures and decreases preventable emergency room visits. In Chap. 3, “Smart Homes for Personal Health and Safety,” Demiris et al. address how to make the home environment safer through controls, sensors, and algorithms that can interpret the clinical needs behind sensor signals. Smart home applications that can benefit from surveillance include physiological and functional indicators of health and their trajectory over time, protection against physical and intrusion hazards, and cognitive and social functioning. The authors illustrate the smart home- based delivery of care through the example of Mild Cognitive Impairment (MCI) and fall management in the Sense4Safety project. Fall risks increase slowly and are often unnoticeable, entailing tremendous cost and morbid- ity. Sense4Safety uses personalized configurations that include nursing assessment of the home situation and resident education and alerts to a trusted party to mitigate