Rigorous and rapid evidence assessment in digital health with the evidence DEFINED framework

Seeing the random forest through the decision trees. Supporting learning health systems from histopathology with machine learning models: Challenges and opportunities

This paper discusses some overlooked challenges faced when working with machine learning models for histopathology and presents a novel opportunity to support "Learning Health Systems" with them. Initially, the authors elaborate on these challenges after separating them according to their mitigation strategies: those that need innovative approaches, time, or future technological capabilities and those that require a conceptual reappraisal from a critical perspective. Then, a novel opportunity to support "Learning Health Systems" by integrating hidden information extracted by ML models from digitalized histopathology slides with other healthcare big data is presented.

The PICOTS-ComTeC Framework for Defining Digital Health Interventions: An ISPOR Special Interest Group Report

OBJECTIVES

Digital health definitions are abundant, but often lack clarity and precision. We aimed to develop a minimum information framework to define patient-facing digital health interventions (DHIs) for outcomes research.

METHODS

Definitions of digital-health-related terms (DHTs) were systematically reviewed, followed by a content analysis using frameworks, including PICOTS (population, intervention, comparator, outcome, timing, and setting), Shannon-Weaver Model of Communication, Agency for Healthcare Research and Quality Measures, and the World Health Organization's Classification of Digital Health Interventions. Subsequently, we conducted an online Delphi study to establish a minimum information framework, which was pilot tested by 5 experts using hypothetical examples.

RESULTS

After screening 2610 records and 545 full-text articles, we identified 101 unique definitions of 67 secondary DHTs in 76 articles, resulting in 95 different patterns of concepts among the definitions. World Health Organization system (84.5%), message (75.7%), intervention (58.3%), and technology (52.4%) were the most frequently covered concepts. For the Delphi survey, we invited 47 members of the ISPOR Digital Health Special Interest Group, 18 of whom became the Delphi panel. The first, second, and third survey rounds were completed by 18, 11, and 10 respondents, respectively. After consolidating results, the PICOTS-ComTeC acronym emerged, involving 9 domains (population, intervention, comparator, outcome, timing, setting, communication, technology, and context) and 32 optional subcategories.

CONCLUSIONS

Patient-facing DHIs can be specified using PICOTS-ComTeC that facilitates identification of appropriate interventions and comparators for a given decision. PICOTS-ComTeC is a flexible and versatile tool, intended to assist authors in designing and reporting primary studies and evidence syntheses, yielding actionable results for clinicians and other decision makers.

Technology and health inequities in diabetes care: How do we widen access to underserved populations and utilize technology to improve outcomes for all?

Digital health technologies are being utilized increasingly in the modern management of diabetes. These include tools such as continuous glucose monitoring systems, connected blood glucose monitoring devices, hybrid closed-loop systems, smart insulin pens, telehealth, and smartphone applications (apps). Although many of these technologies have a solid evidence base, from the perspective of a person living with diabetes, there remain multiple barriers preventing their optimal use, creating a digital divide. In this article, we describe many of the origins of these barriers and offer recommendations on widening access to digital health technologies for underserved populations living with diabetes to improve their health outcomes.

A sociotechnical framework to assess patient-facing eHealth tools: results of a modified Delphi process

Among the thousands of eHealth tools available, the vast majority do not get past pilot phases because they cannot prove value, and only a few have been systematically assessed. Although multiple eHealth assessment frameworks have been developed, these efforts face multiple challenges. This study aimed to address some of these challenges by validating and refining an initial list of 55 assessment criteria based on previous frameworks through a two-round modified Delphi process with in-between rounds of interviews. The expert panel (n = 57) included participants from 18 countries and 9 concerned parties. A consensus was reached on 46 criteria that were classified into foundational and contextual criteria. The 36 foundational criteria focus on evaluating the eHealth tool itself and were grouped into nine clusters: technical aspects, clinical utility and safety, usability and human centricity, functionality, content, data management, endorsement, maintenance, and developer. The 10 contextual criteria focus on evaluating the factors that vary depending on the context the tool is being evaluated for and were grouped into seven clusters: data-protection compliance, safety regulatory compliance, interoperability and data integration, cultural requirements, affordability, cost-benefit, and implementability. The classification of criteria into foundational and contextual helps us assess not only the quality of an isolated tool, but also its potential fit in a specific setting. Criteria subscales may be particularly relevant when determining the strengths and weaknesses of the tool being evaluated. This granularity enables different concerned parties to make informed decisions about which tools to consider according to their specific needs and priorities.

Exploring the Use of a Digital Platform for Cancer Patients to Report Their Demographics, Disease and Therapy Characteristics, Age, and Educational Disparities: An Early-Stage Feasibility Study

INTRODUCTION

The increasing burden of cancer, the development of novel therapies, and the COVID-19 pandemic have made cancer care more complex. Digital innovation was then pushed toward developing platforms to facilitate access to cancer care. Age, education, and other disparities were, however, shown to limit the use of the digital health innovation. The aim of this early-stage feasibility study was to assess whether Greek cancer patients would register at CureCancer and self-report their demographics, disease and therapy characteristics, and socioeconomic issues. The study was organized by the Hellenic Society of Medical Oncology.

METHODS

Patients from nine cancer centers were invited to register on the CureCancer platform and complete an anonymous questionnaire on demographics, disease and therapy characteristics, and socioeconomic issues. Patients were also encouraged to upload, in a secure area for them, their medical files and share them with their physicians. They were then asked to comment on their experience of registration and how easy it was to upload their medical files.

RESULTS

Of the 159 patients enrolled, 144 (90.56%) registered, and 114 of those (79.16%) completed the questionnaire, suggesting that the study is feasible. Users' median age was 54.5 years, and 86.8% of them were university and high school graduates. Most patients (79.8%) reported their specific type of cancer diagnosis, and all reported their therapy characteristics. Breast and lung cancers were the most common. A total of 87 patients (76.3%) reported being on active cancer therapy, 46 (40.4%) had metastatic disease, and 51 (44.7%) received supportive care medications. Eighty-one (71.05%) patients received prior cancer therapies, and twenty-seven recalled prior supportive care medications. All patients reported visiting non-oncology Health Care Professionals during the study. Nineteen of 72 (26.39%) patients who worked prior to cancer diagnosis changed work status; 49 (42.98) patients had children under 24 years; and 16 (14%) patients lived alone. Nine (7.9%) patients were members of patient associations. Registration was "much/very much" easy for 98 (86.0%) patients, while 67 (58.8%) had difficulties uploading their files. Patients commented on the well-organized data access, improved communication, feeling safe, medication adherence, interventions from a distance, and saving time and money. Over 80% of patients "preferred the digital way".

DISCUSSION

A total of 114 patients succeeded in registering on the digital platform and reporting their demographics, disease and therapy characteristics, and socioeconomic issues. Age and educational disparities were disclosed and highlighted the need for educational programs to help older people and people of lower education use digital innovation. Health care policy measures would support patients' financial burden associated with work changes, living alone, and children under 24 years old at school or college. Policy actions would motivate patients to increase their participation in patient associations. According to the evidence DEFINED framework, the number of patients, and the focus on enrollment, engagement, and user experience, the study fulfills actionability level criterion 1.