Search strategies were developed with a medical information specialist from Leiden University Medical Center. Searches were conducted in PubMed, Web of Science, Embase, PsycINFO, and The Cochrane Library combining key search terms “chronic lung disease,” “digital technology,” “health education,” and related terms. To ensure a comprehensive scope by covering all relevant studies on DHE for CLD and capturing a full breadth of the topic, studies were searched from inception to the day we conducted the search (February 26, 2023) without starting time limits. The complete search strategy is available in Multimedia Appendix 1. Literature screening was performed with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) [20] checklist (Multimedia Appendix 2). Inclusion and exclusion were performed according to the Participants, Intervention, Comparison, and Outcomes (PICO) strategy [21], shown in Textbox 1.

After removing duplicates, one reviewer (CS) used the open-source machine learning tool ASReview Lab19 [22] to screen all the titles and abstracts. Another reviewer (RL) used Endnote X9 to double-screen a random proportion of all articles independently. The choice of 15% was made by the reviewer team to ensure the efficiency and reliability of screening based on ASReview developers’ recommendations [23] and previously published articles that screened the same way [24]. If the agreement rate between the 2 reviewers was less than 80%, an additional random 15% of all titles and abstracts were screened by RL. Title and abstract screening were stopped only when the agreement rate exceeded 80%. Then, the included articles were selected by screening the full text by CS and HD. CS then screened all references of included articles to identify if extra articles could be included. Any discrepancies during the screening process were solved by achieving consensus or discussing with the review team.

Reviewer teams (CS, HD, and HW) extracted data with a predesigned standardized data extraction form (Multimedia Appendix 3). Study characteristics (author, publication year, country, profession, study design, sample size, and population levels) were extracted to understand the specific context of each study and assess the relevance, research development, and quality. DHE characteristics (type of DHE medium, content, intervention mechanisms, duration, and frequency) were extracted to understand what aspect of educational platform, content, and behavioral theories were used and how. Outcomes, main findings, and conclusions of included studies were extracted to analyze the impact of DHE on patients with CLD.

Study characteristics and DHE characteristics were summarized using means and percentages where appropriate. Outcome indicators were summarized in 6 categories. CLD awareness evaluated patients’ ability to perceive or to be conscious of CLD or health states (eg, health literacy). Clinical outcomes included physiological and biochemical indicators measured by clinical testing and laboratory equipment (eg, blood pressure, forced expiratory volume [FEV]). Feasibility, acceptability, and satisfaction [25] focused on the design or the development of DHE, evaluating the state or degree of being easily or conveniently used, being subject to acceptance, and the expectations, needs, and pleasure of patients. Lifestyle outcomes included diets, sleep, medication adherence, and physical activity. Cost-effectiveness evaluated the cost expenditure of patients or health care systems on the whole process of disease prevention, treatment, and management. Psychosocial outcomes were observed effects on patients’ emotions, well-being, quality of life (QoL), behavior or social interactions [26]. The impacts of DHE were categorized as positive, negative, or no significant difference based on the extracted results and conclusions. Descriptive analysis was used to summarize all the coded data.

The Critical Appraisal Skills Program (CASP) checklist was used and adapted for appraising the evidence of different study designs [27]. Following CASP tool recommendations, we classified the evidence quality as high (more than 2/3 of questions are answered by “yes”), moderate (more than 1/3 of questions are answered by “yes”), or low (lower than 1/3 of questions are answered by “yes”).

A total of 22 articles were included in this scoping review (Figure 1), consisting of 19 interventional studies (9 randomized controlled trials [RCTs], 7 quasi-experimental studies, and 3 prospective interventional studies) and 3 mixed methods studies. These studies were published between 2000 and 2022 in health care or related professions, showing a distinct increasing trend with the highest number after 2020 (n=11). Details of included articles and the trend of publication year were shown in Multimedia Appendix 3. Among these studies, only 27% (6/22) were conducted in LMICs, with China, India, Indonesia, Lebanon, and Iran 1 or 2, respectively (Figure 2). United States accounted for 32% (7/22) of these studies among developed countries. The sample size of the studies varied from 15 to 9452 participants, of which the population level ranged from hospitals, and clinics to general practices, communities, and recruited volunteers. By following the CASP checklist, all studies were medium (n=4) to high quality (n=18). Further details can be found in Table 1 [28-49] and the quality assessment process was described in Multimedia Appendix 4.

Regarding DHE mediums among the 22 studies, 10 (45%) of them were websites or web-based platforms developed by researchers, along with 6 (27%) studies delivered by mobile applications or tablets, 5 (23%) by video or text messaging, and 1 (5%) by MP3 (Table 2 [28-49]). No more than half (9/22, 41%) of studies conducted DHE for over 3 months. Only 7 studies reported intervention frequency and 4 provided DHE based on the “teach-to-go” strategy, Theory of Planned Behavior, and self-regulation.

Different contents of DHE in CLD were summarized and divided into several categories (Multimedia Appendix 5). Self-management skills of CLD (21/22, 95%) were the most common educational content, including managing general CLD conditions (20/22, 91%), and exacerbations or asthma attacks (7/22, 32%). Treatment of CLD, disease characteristics, motivational messages, or stories were imparted in 15, 15, and 5 studies, respectively. Specifically, medicine or inhaler education (14/22, 64%) was the most frequently mentioned topic about CLD treatment, followed by breathing exercises or physical activity (5/22, 23%) and psychosocial therapy (3/22, 14%).

After classification and synthesizing, we found that 16/22 (73%) studies explored participants’ CLD awareness. Clinical outcomes (6/22, 27%), DHE feasibility, acceptability, and satisfaction (8/22, 36%), lifestyle (3/22, 14%), cost-effectiveness (2/22, 9%), and psychosocial outcomes (8/22, 36%) were also reported in different studies.

Current evidence presents comprehensive characteristics and the potentially positive impact of DHE on CLD. A total of 22 published studies with medium to high quality from 2000 to 2022 were analyzed in this review, including 9 (41%) RCTs, 7 (32%) quasi-experimental studies, 3 (14%) mixed methods studies, and 3 (14%) prospective interventional studies. Most studies (16/22, 73%) were conducted in developed countries. Websites and mobile applications (10/22, 45%) were the most widely used medium to conduct DHE among patients with CLD, followed by mobile apps or tablets, video or text messaging, and MP3. Participants were mainly patients with CLD recruited from hospitals and general practices. No more than half of these studies (9/22, 41%) implemented DHE interventions for more than 3 months, and few (3/22, 14%) were based on theoretical mechanisms when educating the target population. The DHE content varied from a general introduction of CLD characteristics (definition and etiology), and treatment to self-management skills of CLD, motivational messages, and stories for patients and their relatives. Positive impacts were reported among most of these studies, including increased CLD awareness (14/16, 88%), clinical (3/6, 50%), lifestyle (3/3, 100%), psychosocial outcomes (7/8, 88%) and impressive feasibility, satisfaction, and acceptability (6/8, 75%). Only 2 studies reported cost-effectiveness outcomes.

The growing number of publications pertaining to DHE in CLD and the widespread use of websites and mobile apps shed light on an increasingly important role of digital technology in health and health care. As of January 2023, there were 5.16 billion internet users worldwide, over 60% of the global population currently connected to the world wide web with an estimated 192 minutes per day per user spent on the web [50]. In addition, integrating digital technology in health care is revolutionary by providing easy access to health services, improving patient engagement, supporting remote supervision of physical activity regimes at home, delivering timely detection of deteriorating health status, and serving as a cost-saving alternative health service [51]. Although not mentioned in included studies, the integration of artificial technology (AI) in health care, which is another promising DHE medium, holds a significant potential to deliver more efficient and personalized educational content through advanced algorithms and predictive analytics [52,53]. Therefore, users could access a diverse range of health information with better accessibility, efficiency, and interactivity. The increasing trend of exploring the impacts of DHE intervention will continue in future research.

More attention should be paid to implementing and scaling DHE to LMICs and the general population with limited access to digital technology. As reported, the internet penetration rate of China was relatively low at 76% compared with South Korea and Japan, which ranged over the 90% mark, although it is home to the largest digital community in the world, with about 1.07 billion netizens as of 2022 [54]. The same pattern is apparent in Southeast Asian countries with little internet accessibility at home and low self-perceived digital literacy among Southeast Asian youth, thereby reducing health care engagement and hindering health information dissemination for the local people [55]. Larger gains in health can be achieved in these areas due to the huge population underserved, which could also contribute to global health in the future. Furthermore, new opportunities and challenges are presented when scaling up DHE with the increasing aging population and worldwide health problems. Digital technologies play an increasingly important role not only in health education and promotion but also in addressing issues of health equality [56].

Future studies should further elucidate DHE characteristics and mechanisms to ensure long-standing effectiveness and verifiable impacts. First, the predominance of DHE contents focusing on self-management skills of CLD highlights the importance of empowering individuals to take an active role in their own health, which corresponds to the current research trends of chronic diseases [57,58]. Second, successful health education depends on repeatedly using a few messages of proven benefit in many forums [59]. As such, the DHE content and the authenticity of contents should be warranted, of which the latter was rarely mentioned. As an essential health education component, DHE frequency, duration, and intervention theories are closely connected with prospective health outcomes. Several existing models, such as the Unified Theory of Acceptance and Use of Technology, Technology Acceptance Model, Social Cognitive Theory, and Theory of Planned Behavior, have been identified as facilitators of adopting digital health interventions [60]. Specifically, interventions based on the theory of planned behavior tended to have more substantial effects on behavior compared to interventions incorporating fewer techniques [61]. In this review, however, few studies illuminated the frequency (7/22, 32%) and intervention theory (4/22, 18%), indicating a potential gap in understanding how DHE interventions work and the underlying mechanisms that impact patients with CLD and related outcomes.

The effects reported on awareness or knowledge are commonly acknowledged [62-64] because digital technologies can support a wide range of self-directed learning activities, providing learners with diverse resources, information, and network resources that meet their learning needs [9]. Improvement in clinical outcomes suggests the potential of DHE to empower patients to make informed decisions about their health, adhere to treatment regimens, and adopt healthier lifestyles after enhancing their understanding of disease and health conditions. Similar positive effects in QoL and health status were also reported in other reviews, but they mainly focus on digital health interventions [65,66]. Insignificant effects in several studies might be related to the intervention duration or inclusion of participants [37,38,44]. The reported impacts on DHE feasibility, acceptability, and satisfaction shed light on users’ active engagement and willingness with these interventions. These positive findings align with other studies [67,68], indicating that DHE is generally well-received by individuals with CLD and implies promising practicality and scalability in real-world settings.

A limited number of studies reported impacts on lifestyle (n=3) and cost-effectiveness outcomes (n=2) in this review, which needs more evidence to illustrate in the future. The lifestyle outcomes extracted from the included studies were limited to physical activity and medication adherence. According to previous studies, however, a sedentary and unhealthy lifestyle, including smoking, having an unhealthy diet, and insufficient exercise could fuel chronic disease progression by changing interstitial cell behavior [69]. In addition, we noticed that only one study applied behavioral theories but implemented DHE for only 2 months when exploring DHE impacts on lifestyle [32]. However, evidence has shown that a successful digital intervention can influence lifestyle management [70], and the theory of behavior change maintenance can guide the development and evaluation of interventions, thereby promoting sustained change in health behavior and lifestyles [71]. As such, a deeper exploration of specific lifestyles, such as the dietary pattern, types, frequencies, and intensities of exercise and other health-related outcomes should be mapped, with validated and effective theories. Currently, few studies in this review compare the cost or health care utilization after DHE intervention, which guided future cost-effectiveness studies with more accurate and complete data to fill in the gap.

Based on the current review, we recommend that future research and development in DHE for people with CLD should focus on elucidating the characteristics and mechanisms of the DHE intervention, prioritizing less developed areas, investigating more lifestyle or behavioral outcomes, evaluating cost-effectiveness among different DHE mediums and ensuring long-term sustainability, to ultimately improve the effectiveness, scalability of DHE, and the health status of patients with CLD. With the integration of AI in health care, we hope to see more personalized and efficient education content delivered and look forward to studies reporting its potential and effectiveness. Given that several studies were of medium quality based on the CASP tool, we recommend future researchers consider incorporating blinding among participants, investigators, and analysts, as well as addressing confounding factors in both study design and analysis, which will enhance the robustness and reliability.

This is the first review to explore the characteristics and impact of DHE on CLD. Diverse studies in this review were synthesized and analyzed under the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guideline, all showing medium or high-quality evidence with a practical quality assessment tool. The detailed descriptions of DHE and its impact on different outcomes provide an important theoretical basis and new thoughts for further research in DHE. However, this review included studies focusing more on mixed digital health intervention. Despite the potential benefits, whether it is the independent or compounding influence of DHE remains to be discussed further due to the inability of subgroup analysis. The incomplete screening of titles and abstracts by the second reviewer may have introduced potential bias into the review process. The exclusion of non-English articles and studies without full-text access could result in publication bias to some extent. In addition, it is worth mentioning the caution of extrapolating the impact to other DHE mediums, despite recognizing the reliability of health information in these studies. Variations in researchers’ profession might also contribute to heterogeneity in study implementation and reporting.

Despite the heterogeneity of the study situation, some aspects can be concluded. DHE can improve disease awareness and clinical outcomes among patients with chronic lung disease with good feasibility, acceptability, and satisfaction through different mediums and educational content. There is still relatively limited research among people in low- and middle-income countries. Future research should consider the impacts on cost-effectiveness outcomes, duration, frequency, and theoretical mechanisms of the DHE intervention to maximize the potential effects. It should also be conducted in the context of health services research to better reflect the real-world settings.