The team, led by researchers at University Hospital Giessen and Marburg and the German Center for Lung Research, performed an umbrella review, searching PubMed, Scopus, Web of Science and the Cochrane Library for review publications from September 2019 through May 2026. Of 278 potentially relevant reviews identified, only nine survived a deliberately stringent two-step appraisal process. Four were systematic reviews, two were expert or narrative reviews, two were scoping reviews and one was a meta-analysis, spanning publication years 2022 to 2024. Each was appraised with the AMSTAR-2 instrument, and citation overlap among them was quantified using the Corrected Covered Area, coming out as slight for all three diseases: 3.6 percent for asthma, 2.9 percent for cystic fibrosis and 1.2 percent for COPD, which justified keeping all nine in the sample.

The two-step procedure is the methodological heart of the study. Rather than pooling summary sentences from the reviews, the investigators went a layer deeper, screening the 231 original publications that those reviews referenced. Only 18 met the full inclusion criteria. To qualify, an original publication had to report at least one barrier, facilitator or transferable concept in its Results section, had to allow clear identification of the stakeholder providing the evidence, and had to go beyond simple acceptability testing such as a System Usability Scale score. The rationale was to prevent a common failure of review-level synthesis: the loss of context when actual lived experiences are aggregated together with perceptions of those experiences, or when patient-reported barriers turn out to be healthcare professionals’ perceptions of what patients feel.

The exclusions were telling. The most frequent reason, affecting 83 publications, was the absence of a generalizable or extractable barrier or facilitator, typically because the study offered only a Likert-based usability score without qualitative content. Another 46 publications fell outside the ten-year cutoff, 29 reported perceptions rather than direct experience, and 26 were the wrong publication type altogether, such as conference proceedings or book chapters. Notably, no COVID-19 publication made it through. Although COVID-19 and Long COVID were the most frequently named diseases in the initial screening, appearing in 111 reviews, the COVID-focused literature predominantly described telemedicine delivered during the pandemic rather than barriers and facilitators stratified by the disease itself, and five recent COVID-related reviews were excluded because their cited studies did not meet the eligibility criteria.

The 18 included original publications broke down into eight on asthma, six on COPD and four on cystic fibrosis, and the pattern across them was revealing. Asthma and cystic fibrosis evidence operated at two levels: direct reports from individual patients or healthcare staff, and well-structured, framework-mapped experience from entire institutions or regional and national programs. Three asthma publications, for example, comprehensively described an established school-based pediatric asthma telehealth program. In cystic fibrosis, one publication stood out: a program-level study that applied the Consolidated Framework for Implementation Research (CFIR) to the results of a national survey administered by the US Cystic Fibrosis Foundation. COPD publications, by contrast, were all typical investigator-initiated studies with a single panel of participants, lacking this institutional tier.

To make the interconnections visible, the team built co-occurrence networks in which each barrier or facilitator was drawn as a node, with lines connecting codes reported together within the same publication. Nodes were color-coded green for facilitators, red for barriers, grey for factors reported as both, and navy-blue for neutral factors. The asthma networks, both patient-centered and healthcare-professional-centered, showed rich, well-structured clusters including equity dimensions such as place of residence and education. The cystic fibrosis patient network was sparse, with three small, isolated clusters each drawn from a single publication, and no professional-centered network could be constructed at all. The COPD networks were somewhat denser but still fragmented into small clusters with limited connections. Crucially, the cystic fibrosis and COPD networks were dominated by barriers, whereas the asthma networks were more balanced between barriers and facilitators.

The CFIR-mapped cystic fibrosis publication enabled a uniquely complete hierarchical analysis. That survey had separately collected data from two groups of program clinics: Group A, which rated telemedicine as similar to or better than in-person care and had already implemented it in routine care, and Group B, which rated telemedicine as worse and had slower uptake. Mapping the findings onto the 2009 version of CFIR revealed that four constructs were never reported, and several others appeared for only one clinic group, gaps the investigators deliberately kept visible. The exercise also demonstrated that the same factor can carry opposite polarities for different groups of professionals, appearing as a facilitator in one clinic and a barrier in another. Re-mapping onto the 2022 CFIR revision proved infeasible because of substantial structural differences between framework versions.

When all findings were mapped onto the simpler five-domain framework of Chaudoir and colleagues, covering technology, patient, provider, organizational and structural domains, asthma dominated every category. Asthma publications contributed 121 finding-mentions against 11 for cystic fibrosis and 37 for COPD, and even the combined CF and COPD counts could not match asthma’s richness. Among the most consistently reported barriers across all diseases were technology-related implementation challenges, time management difficulties and gaps in workflow coordination. The most consistently reported facilitators were improved communication, structured patient education and ease of use. Patient-level factors ranged from digital literacy, data privacy concerns and internet access to social support, reminders and prior experience with digital technology, with equity dimensions such as income, language and connection quality independently restricting uptake even in younger pediatric populations.

The review also surfaced transferable concepts with relevance across at least two diseases. Health and equity dimensions can be addressed even in resource-strapped settings by using public resources such as libraries and healthcare centers to equalize access. Quantitative metrics that distinguish technology adoption from sustained use can feed predictive models of utilization. Simultaneous evaluation in different end-user groups, such as utilizers versus non-utilizers, enables randomized controlled trial designs. And studies that prospectively anchor their assessments in established frameworks yield richer, more comprehensive evidence. The authors argue that a study appraising acceptance only through a usability score cannot later be mapped onto an implementation framework with sufficient veracity, whereas one including a qualitative or mixed-methods substudy built on a framework avoids that problem.

For policymakers and implementers, the authors distill three concrete priorities: investing in interoperable, low-maintenance technology with dedicated technical support; integrating digital medicine into existing clinical workflows rather than adding it as a parallel duty; and prioritizing structured onboarding and patient education, which links directly to the ease-of-use and communication facilitators. Digital literacy support, reported across more than one disease group, stands out as a transferable target independent of the specific condition. The authors acknowledge limitations, including a degree of subjectivity in classification, an exploratory scope that traded depth for breadth, and the inherent lag of umbrella reviews behind fast-moving fields such as AI-enabled chatbots. They call for a follow-up systematic review of original publications, and for future studies to be anchored in implementation frameworks so that knowledge can translate into implementability, particularly in COPD, where the evidence on barriers and facilitators has yet to become systematic.

Subject of Research: Barriers and facilitators to implementing digital medicine in chronic respiratory diseases

Article Title: Common barriers and facilitators to the implementation of digital medicine in chronic respiratory diseases: a systematic review of reviews

Article References: Korel, J., Kuhn, S., Gloeckl, R., Koczulla, R., Dreher, M., Müller, B. S., Hirschbrich, R., Junker, E., Hastenteufel, T., Llauradó, X., Haddi, Z., Vogelmeier, C. F., Rohde, G., & Dauletbaev, N. (2026). Common barriers and facilitators to the implementation of digital medicine in chronic respiratory diseases: a systematic review of reviews. Frontiers in Digital Health, 8, Article 1927835. https://doi.org/10.3389/fdgth.2026.1927835

Image Credits: AI Generated

DOI: 10.3389/fdgth.2026.1927835

Keywords: digital medicine, respiratory disease, asthma, COPD, cystic fibrosis, telemedicine, mHealth, implementation science, systematic review, CFIR, barriers, facilitators