The review, led by Fanchang Wang of Zhejiang Chinese Medical University with colleagues from Wenzhou Medical University and the Second Affiliated Hospital of Zhejiang Chinese Medical University, synthesized 28 sources identified through PubMed/MEDLINE searches, backward citation tracking and targeted supplementary searches, following SANRA guidance for narrative reviews. The authors examined digital self-management across three overlapping delivery models: patient-centered digital support, remote patient monitoring, and integrated primary care or community coordination. Their organizing framework also specifies six implementation conditions that determine whether any of these models becomes chronic care infrastructure: equity and access, health and digital literacy, interoperability, privacy and governance, sustained engagement, and workforce and financing alignment.

The stakes are enormous. Global Burden of Disease 2021 projections cited in the review indicate that deaths and disability-adjusted life years attributable to non-communicable diseases—cardiovascular disease, cancer, chronic respiratory disease and diabetes—will keep rising toward 2050, with cardiovascular disease making a major contribution. The economic toll is equally daunting: non-communicable diseases are estimated to cost the global economy USD 47 trillion between 2011 and 2030. Yet in many health systems, self-management remains weakly supported, unevenly reimbursed and poorly linked to routine clinical workflows, which is precisely the gap digital tools claim to fill.

The first delivery model, patient-centered digital support, is the most familiar. One systematic scoping review identified 69 digital health platforms for self-management across roughly 20 chronic conditions, most of them mobile applications or combined web and mobile tools. From a management perspective, these tools are attractive because they can be scaled relatively easily and at low marginal cost. Their central weakness, however, is that they often function alongside routine care rather than within it, providing information and motivation without a clear pathway into clinical decision-making. Conversational agents expand the model with interactive education, reminders and coaching, and since 2024, large language model-based agents have enabled more flexible explanations, medication questions, visit preparation and behavioral support. They also introduce safety concerns—hallucinated advice, unsafe self-triage, bias, privacy leakage and unclear accountability—which the authors say require explicit clinical scope, escalation rules, human oversight and post-deployment monitoring.

The second model, remote patient monitoring, connects self-management with clinical oversight by transmitting blood pressure, glucose, weight, oxygen saturation or symptom data from the home to care teams. Programmes such as the Mayo Clinic nurse-based remote monitoring model illustrate how devices can be paired with structured nursing oversight and integration into electronic health records. Some studies have reported reductions in hospitalizations or emergency department visits among patients with chronic obstructive pulmonary disease or older adults with polypathology, but effects vary by population, workflow design, monitoring intensity and outcome definition. The review frames the central debate bluntly: does remote monitoring primarily prevent avoidable deterioration, or does it move clinical work into less visible digital channels? Alerts, asynchronous messages and trend reviews can add considerable workload when thresholds, escalation rules and documentation responsibilities are not clearly defined.

The third model is the least technologically glamorous and, the authors suggest, potentially the most important for equity. It places digital self-management within integrated care and local coordination, drawing on frameworks such as SELFIE for multimorbidity, which spans service delivery, leadership, workforce, financing, technologies and information across micro, meso and macro levels. Community health workers, health coaches and digital navigators can bring digital self-management to patients who might otherwise be left out, offering culturally adapted education and trusted human support. Analyses of digital transformation in primary health care identify interoperability, AI-enabled decision support and workforce training as key enablers, while infrastructure gaps and limited digital health literacy remain significant barriers.

Across all three models, the review identifies three recurring implementation tensions. The first pits empowerment against digital exclusion: uptake of digital health tools is often higher among urban, economically advantaged and digitally confident groups, whereas older adults, rural communities, minority groups and low-income patients are less consistently represented in the evidence base. The authors argue this gap cannot be fixed through better interface design alone; programmes should consider device lending or subsidies, low-bandwidth mobile-first designs, multilingual plain-language materials, telephone or in-person alternatives, and digital navigator support. The second tension concerns data visibility versus clinical accountability: more data do not necessarily make care safer, and without usable exchange with electronic health records, platforms may expand monitoring without giving clinicians actionable summaries. Interoperability standards, open APIs and shared data models should be treated as management requirements, not technical preferences. The third tension involves engagement versus durable outcomes and workload: use may decline within months even with reminders and gamification, and a randomized trial of an internet-based self-management programme for multiple chronic diseases did not significantly reduce all-cause hospitalizations over two years, while a meta-analysis of European primary care trials found no significant differences from usual care for most outcomes except a small improvement in systolic blood pressure.

For managers, commissioners and policymakers, the review distills the practical question into three priorities. First, redesign workflows around accountable action: define the pathway from patient-generated data to clinical action, including eligibility, onboarding, monitoring frequency, alert thresholds, escalation routes, documentation responsibilities and stopping rules, with staff responsibilities specified for nurses, pharmacists, community health workers, digital navigators and physicians. The authors offer a concrete illustration—a hypertension remote monitoring programme in which a nurse or pharmacist leads onboarding and device training, weekly readings feed a shared dashboard, pharmacist-led medication titration proceeds under an agreed protocol, physicians escalate for predefined safety thresholds, and the team reviews progress at three months before de-escalating once control is sustained. The device is visible, they note, but the workflow creates value. Second, procure for integration, safety and governance, prioritizing tools that exchange data with existing records, support open standards and provide transparent rules for data storage, vendor access, model updates and auditing. Third, evaluate value, equity and sustainability using frameworks such as RE-AIM and the Consolidated Framework for Implementation Research, stratifying outcomes by age, language, socioeconomic status, rurality and disability—and allowing programmes to fail in evaluation when they do not deliver value.

The review is candid about its limitations: it did not involve a fully systematic search, formal risk-of-bias assessment or quantitative synthesis, and its evidence base consists mainly of English-language sources drawn largely from high-income health systems. Published interventions may also differ from routine practice, where infrastructure, staffing, reimbursement and digital literacy vary considerably. Future work, the authors say, should prioritize evaluation of generative AI agents for safety, hallucination, bias and clinician workload; assessment of privacy-preserving, interoperable infrastructure for governance feasibility; reimbursement models that support team-based review, patient education and proactive outreach; and pragmatic evaluations of clinical outcomes, cost, alert fatigue and inequity.

The argument advanced here is not against digital tools but for service integration and accountability. Digital self-management has become a credible part of chronic disease care, the authors conclude, but its public health value remains conditional. It will contribute most when health systems treat it as service redesign rather than technology procurement, and its contribution will be weakest when adoption moves faster than the workflows, financing, governance and equity supports needed for chronic care infrastructure. The key question is no longer only whether these tools work under trial conditions, but whether systems can build the workflows, governance and equity supports needed for high-burden populations. For the millions of people managing diabetes, heart disease or COPD between visits, the difference between a promising app and genuine care may come down to who is accountable for the data it generates.

Subject of Research: Digital health

Article Title: Digital self-management for chronic disease care: a mini review of service integration, equity, and public health implementation priorities

Article References: Wang, F., Qiao, H., Wu, Y., He, X., & Ying, L. (2026). Digital self-management for chronic disease care: a mini review of service integration, equity, and public health implementation priorities. Frontiers in Public Health, 14, Article 1883085. https://doi.org/10.3389/fpubh.2026.1883085

Image Credits: AI Generated

DOI: 10.3389/fpubh.2026.1883085

Keywords: Digital, self-management, chronic, disease, care, mini, review, service, integration, equity, public, health