Healthcare Policy, Innovation and Digitalization

Table of Contents

1. Frontmatter

2. Introduction

   1. Frontmatter

   2. Chapter 1. Contemporary Strategy and Approaches in Healthcare Policy, Innovation and Digitalization

      Hilal Özen, Eyüp Çetin

      Abstract

      In today’s world, people are more aware and conscious about health issues than before. The main reason behind this can be both the developments in health issues and the process we have experienced during the COVID-19 pandemics. We all are more used to digital health applications and innovative health solutions. These developments trigger the changes and also arrangements in healthcare policy issues. All the topics have both theoretical and practical implications in the industry and academic world. This chapter tries to summarize the importance of the new strategies and approaches in healthcare policy, innovation, and digitalization by making an extensive introduction to those three concepts.

3. Healthcare Policy

   1. Frontmatter

   2. Chapter 2. Prioritization in Health Care: The Influence of Frames on Accepting Prioritization Criteria

      Adele Diederich, Marc Wyszynski

      Abstract

      Global pandemics, social and scientific developments such as growing and aging populations, novel and expensive health technologies, and improved medical treatments increase the demand for healthcare services and challenge healthcare systems worldwide. Prioritizing healthcare services according to some pre-defined criteria such as age, health behavior, and social responsibility has been proposed for distributing limited resources. A major concern in healthcare policies is establishing fair and legitimized procedures for distributing scarce healthcare resources. Ways of legitimizing prioritization criteria in terms of fairness and justice are, for instance, drawing on theoretical considerations such as (normative) distributive justice principles, like equity, equality, and need, mostly relying on experts’ opinions; or using more practical approaches such as direct-democracy-like elements, i.e., including the general public in decisions on prioritizing scarce healthcare resources. For the latter, opinions and preferences of those are typically elicited by offering them survey questionnaires. However, past research has shown that participants’ preferences may be influenced by how the questionnaire is constructed and items are phrased and framed. This chapter discusses the impact of framing on attitudes toward prioritization criteria in healthcare service. In particular, questions are framed in terms of providing and withholding services. By combining psychological theory on judgment and decision-making with recent empirical findings and previous research, we show that the effect of framing on people’s preferences for healthcare prioritization criteria is an essential factor that healthcare policymakers may want to consider when using elements of direct democracy.

   3. Chapter 3. Using Pharmacoepidemiologic Studies to Inform Drug Policy and Spending: A Health Economics Perspective

      Konstantinos Zisis, Kostas Athanasakis, Kyriakos Souliotis

      Abstract

      Healthcare resource allocation today is more difficult than ever. The asymmetry between available resources and health needs requires methodologically sound and transparent decision-making processes for resource allocation. Clinical effectiveness and cost-effectiveness of health technologies are major determinants of pricing and reimbursement decisions—however, the above must be demonstrated both in the controlled and in the real-world setting. Pharmacoepidemiology can play a major role in evidence-informed pharmaceutical policy decisions, determining what “actually works” in real-life conditions. We discuss the role of pharmacoepidemiology in the decision-making context of pharmaceutical policy and, in specific, its role in the resource allocation framework, focusing on its impact on Health Technology Assessment, the efficiency of spending, and handling of uncertainty in coverage decisions, the latter achieved through risk-sharing schemes.

   4. Chapter 4. Accelerating Personalized Medicine Adoption in Oncology: Challenges and Opportunities

      Fredrick D. Ashbury, Keith Thompson

      Abstract

      Cancer therapy has been shifting away from “one-size-fits-all” approaches to treatment decisions that are predicated on the molecular profile of the patient’s cancer. Personalized medicine, precision medicine, precision oncology, or “omics-guided” therapy are expressions used for this paradigm shift often interchangeably. The uptake of precision medicine, while advancing in cancer care, has faced several adoption challenges, including education, policy, and practical factors. Facilitating the transformation toward personalized medicine that will improve patient outcomes in the oncology setting requires a coordinated effort among policymakers, cancer agencies, health systems, and industry. To implement precision medicine effectively in cancer practice requires an informatics solution beyond the legacy electronic medical record platforms currently available to clinical teams.

4. Innovation

   1. Frontmatter

   2. Chapter 5. Framework for Epidemic Risk Analysis

      Maryna Zharikova, Stefan Pickl

      Abstract

      Global urbanization presents numerous health challenges and necessitates a deeper comprehension of how diseases spread in urban settings. The chapter proposes a novel framework for epidemic risk analysis by incorporating human mobility. The proposed framework combines evidence-based risk analysis driven by facts with the advantages of flexible precautionary models taking into account causation-and-effect connections and spatial reference. The precautionary approach is implemented in the form of a set of interrelated models, which, as a result, give estimates of a potential risk, obtained in advance and distributed in space. To obtain these estimates, the following models are built: a spatial model, a model of human mobility, as well as a model of disease spread. The spatial model comprises a set of confined locations of people’s concentration. These locations are subdivided into several levels, such as residential buildings (base locations where people live), and workplace locations / locations of study (locations that people attend regularly). Over the spatial model, we build the model of human mobility represented as a network with the nodes being locations of different levels. Each node at every time moment is characterized by a certain number of people. The arcs of the network show human flows within the spatial model, which influence disease spread. The disease spreading model reflects disease movement through the nodes of the network. Each node of the network has its status regarding a certain disease, such as susceptible, infected, or recovered. A transition of the node from one status to another is represented as an event. A spread of the disease through the nodes of the network is represented by an event-tree network. We also consider external factors (meta-factors) that can influence the rate of spread of the disease such as weather conditions (temperature, humidity) and air quality. A distinctive feature of the proposed approach is a spatial reference of network nodes. This chapter addresses the framework for epidemic risk analysis. The proposed framework is based on two main principles such as an evidence-based approach and a precautionary principle. These principles are used in the following way. Using the precautionary principle, scenarios of disease spread and risk assessments are built in advance and can be used for disease prevention and mitigation. With the help of the evidence-based approach, the risk assessments obtained using the precautionary approach are adjusted. Justified risk assessments can be used for making decisions to counter the disease. The proposed framework makes it possible to diagnose the situation within the area of interest and sheds new light on disease-spreading analysis. The most dangerous areas can be identified, and measures can be taken to stop the disease from spreading. The work contributes to the creation and putting into practice the policies that slow down or prevent the spread of illnesses.

   3. Chapter 6. Transmissibility and Death Index from Pandemic COVID-19 Among Nations Across Continents

      Ramalingam Shanmugam, Karan P. Singh

      Abstract

      In this article, stochastic models are introduced to address, capture, and illustrate how much of the pandemic COVID-19 transmissibility occurred in regular as well as anomalous patterns in nations within continents: Africa, America, Asia, Europe, and Oceania. The results are interesting, not only drastic differences in the neighboring nations but also the existence of spiral pulling up of the COVID-19 death rates. In this process, a new death index is developed to portray the uniqueness to fight against the COVID-19 cases versus deaths. The index offers an insight for the health administrators to formulate preventive strategies for a successful tackling of this unprecedented uneven pandemic.

   4. Chapter 7. Assessing Health Inequalities of Diabetes Care Through the Application of the Bio-ecology Theory

      Alan Shaw

      Abstract

      The National Institute for Health and Care Excellence (NICE) guidelines for the management of diabetes state that structured diabetes education should be offered to every person and their carer(s) at or around the time of diagnosis, with annual reinforcement and review. In 2016, the UK’s Health and Social Care Information Centre’s National Diabetes Audit for England identified only 6% of newly diagnosed Type 2 diabetics attended a course. Diabetes UK has called for radical improvements to the provision. This study attempts to determine why the uptake has been so poor and then offer possible solutions. The study utilised Bronfenbrenner’s bio-ecology theory and was made up of four phases: phase one, a pilot study of health educators to identify why patients were not attending the courses. Phase two a qualitative review, using thematic analysis, of patients on their views of structured education. Phase three a census investigating the provision of structured education. It compared the 152 Primary Care Trusts (PCTs) with the new 194 Clinical Commissioning Groups (CCGs) in England. Phase four is a qualitative review using thematic analysis of healthcare professionals (HCPs) on their reasons for providing the care they did. NHS England has a decentralisation approach to managing diabetes structured education in England. There is a lack of awareness of these programmes amongst patients. This is driven by the proliferation of courses provided by NHS England and the budget restrictions to promote them. The quality of diabetes structured education and the ability of patients to attend varied by PCT/CCG, creating a non-inclusive service. In this example, it was established that centralising elements of the diabetes structured education programmes like branding, marketing, course development and programme management could alleviate many of the problems that NHS England currently faces and increase patient engagement. Such a move would also reduce costs and help bridge the current budget deficit. This chapter demonstrates how researchers can utilise Bronfenbrenner’s bio-ecology theory to investigate healthcare management processes. More specifically, it is an example of investigating patients, their careers, healthcare professional and policy all in one study. It also addresses a common debate amongst healthcare managers whether systems should be centralised or decentralised.

   5. Chapter 8. The Computational Perspective on Internalized and Simplex-Structured Motivation

      Ali Ünlü

      Abstract

      Self-determination theory (SDT), introduced by Deci and Ryan, is a popular theory of motivation. Applications of SDT are numerous and include areas like health care, health professions education, and digital health. Over the recent years, the author published a series of quantitative papers on SDT. To the author’s knowledge, these contributions have remained relatively unrecognized in the SDT community. However, the methodology developed therein can be useful to the field. With the present work, the author reviews, exemplifies with data, and mathematically describes that methodology, in a coherent manner. The focus of this chapter is on computational as well as mathematical aspects. For the investigation of motivation internalization and simplex structure, the author recapitulates the convex decomposition, or constrained regression, model and assembles the computation steps of the convex quadratic program. The author also contributes to the mathematical foundations of the methodology. In particular, mathematical definitions are proposed for the in SDT important concepts of theoretical closeness of regulations, cumulative internalities of regulations along the motivation continuum, and simplex structure of motivation. The idea is to consider a linear order on the set of regulations, take the induced geodesic distance, form a linear motivational structure, and posit that these distances, as measure of closeness, are compatible with the shares of the convex decomposition model. By examples, the author shows how the method can be used for the exploratory data analysis of simplex structure. In particular, for a given intermediate regulation, the author employs the method to estimate the theoretically closer regulation of the two neighbor regulations, contiguous to it. In addition, the technique was applied in a systematic empirical study. The study compared science teaching in a classical school class versus an expeditionary outdoor program. Succinctly, the main results of this study are recapped. In the internal and external shares of identified regulation, the science teaching formats did not differ. The teaching formats differed in the internalization of introjected regulation, which was more strongly external motivation in the outdoor program. The simplex structure of SDT could basically be supported in the study data. The statistical computing and graphics environment R is powerful. Throughout this chapter, computations were made in R, with the package SDT. The functions internalization and simplex of the package SDT were used for computations of the internalization shares and simplex structure shares, respectively. Finally, this chapter concludes with a few general ideas about the motivation theory and with personal suggestions for modifications of it.

   6. Chapter 9. Recent Developments of Multiple Myeloma: Analysis and Analytical Modeling Using Real Data

      Chris P. Tsokos, Lohuwa Mamudu

      Abstract

      Multiple myeloma cancer (MMC), also known as Kahler disease, myelomatosis, and plasma cell myeloma, is a devastating type of cancer that still remains incurable. In this chapter, we reviewed and identified some of the very recent research developments and findings from real data of patients diagnosed with MMC which we believe are significant to the subject matter of MMC survivorship. Our recent findings involved developing a parametric approach to survival analysis. We conducted parametric and nonparametric analyses of the survival times of patients diagnosed with MMC. We included in our analysis the applicability of the Cox-proportional hazard analytical model that is being driven by significant risk factors that significantly contribute to the survival of MMC and compared it with our parametric findings and other research findings on survival analysis of MMC. We concluded our studies with a very recent and effective analytical model that was developed using real data that very accurately predicts the survival time of MMC patients. This parametric analytical modeling approach is driven by the significant risk factors and interactions that are associated with the survival time of MMC patients. In addition, we identified new risk factors associated with the survival of MMC. Our recent findings about the survival time of MMC and the new approach to survival analysis provide a novel strategy and ways to investigate the survival of MMC and other types of cancers. The findings from our studies can help improve the therapeutic/treatment strategy of MMC disease, hence, improvement in MMC survival.

5. Digitalization

   1. Frontmatter

   2. Chapter 10. Refined Machine Learning Approaches for Mask Policy Analysis

      Lincy Y. Chen, John Tuhao Chen

      Abstract

      Worldwide, the Covid-19 pandemic has created an imminent need for new public policies focused on infectious disease intervention, management, and prevention. Mask debates take center stage amongst these policies, but disagreements abound regarding the efficacy, fairness, and health effects of mandatory versus voluntary masking. Properly assessing these policies with available data necessitates statistical models. One of the main problems with statistical models is the misusage and improper interpretations. This is particularly true in the context of COVID-19 because of the high degree of random variation and dynamic evolution processes aimed at restraining the transmission of airborne viruses. For example, public health officials frequently employ logistic regression models to analyze odds ratios associated with social and behavioral consequences, which determine public health and management policies. However, the implementation of statistical analysis usually involves hidden model assumptions. Violating these model assumptions often invalidates data analytical results, and consequently misleads the direction of policy reforms. In this chapter, we put forth new and refined statistical instruments with relaxing model assumptions for public health policy analysis. After a discussion on updated machine learning techniques including range regression, weighted classification with KNN (k-nearest-neighbors) algorithm, and enhanced decision trees, we apply the refined hybrid statistical methods to analyze a set of social and behavioral data collected during the Covid-19 pandemic. The dataset contains survey information on social and behavioral characteristics in conjunction with mask policies (mandatory vs voluntary). Results of the refined data analysis cast new light on the effects of mandatory versus voluntary mask policies. Although the former (mandatory) is more effective and better controls the frequency of airborne virus transmission, the latter appears to be more friendly and associated with higher community warmth indices.

   3. Chapter 11. Allocating Capacity for Office and Virtual Visits in Chronic Care Settings

      Xiao Yu, Armağan Bayram, Yuchi Guo, Gökçe Kahvecioğlu

      Abstract

      Access to care is an important measure, especially for chronic care patients who require regular doctor appointments to control their chronic condition. Due to limitations on the number of available office appointments and due to additional precautions during the pandemic, patients may have limited access to chronic care; consequently, they may not receive the preferred treatment. In an effort to improve access to care for all patients during the pandemic, most healthcare providers have introduced virtual appointments as an alternative to traditional office appointments. In this study, to investigate the utilization of virtual appointments in the context of chronic care, we take into account a cohort of patients receiving chronic care and develop a capacity allocation model. We use an open migration network model to analyze the patients’ flow between different types of appointments in chronic care and use the newsvendor model to investigate how the capacity is allocated between the virtual and office appointments. By considering patients’ disease progressions, we propose policies based on our findings to make more systematic capacity allocation decisions.

   4. Chapter 12. Collaborative Systems Analytics to Advance Clinical Care: Application to Congenital Cardiac Patients

      Eva K. Lee

      Abstract

      This chapter reports the operations research advances in the Edelman finalist work on “Collaborative Systems Analytics: Establishing Effective Clinical Practice Guidelines for Advancing Congenital Cardiac Care.” The clinical advances and results of this project have been reported elsewhere. This paper highlights the OR-analytic advances and briefly summarizes the clinical implementations, results, and impacts. Specifically, we devised a customizable model and decision support framework that combines systems modeling, simulation-optimization decision analytics, clustering, and machine learning within a collaborative learning paradigm to help hospitals pinpoint key factors on practice variation, and design clinical practice guidelines (CPGs) for rapid implementation to improve the outcomes of congenital heart defects surgeries. The OR-analytic collaborative learning framework described herein is generalizable and is applicable for numerous domains. Within healthcare, it enables systems redesign, quality improvement, resource allocation, and clinical support and decision advances. The computational engine facilitates systems and process optimization. The results improve efficiency of healthcare delivery, reduce costs and wastes while improving quality of life of patients. A critical contribution is that the system offers an effective, flexible way to study adequate numbers of patients across multiple sites with uncommon diseases through a common infrastructure for recruiting, monitoring, and following patients whose conditions will be characterized in a standard fashion. The modeling-computational framework facilitates the design of a common CPG, and its successful implementation with documented and measurable clinical outcomes. Such a framework permits a flexible clinical transformative environment that can accommodate practice variance while enabling care teams to identify critical system pathways for multiple-site clinical care and process improvement. The hypothesis testing and dissemination of findings allows for rapid learning and adoption at multiple sites with a shortened duration and only a fraction of the budget as opposed to the conventional randomized clinical trials. Hence, the OR-analytic collaborative learning framework can serve as a blueprint for other clinical and process-improvement initiatives.

6. Backmatter