This systematic review was conducted and reported in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 statement (Checklist 1) and its expanded checklist (Checklist 2) [24], the PRISMA 2020 for Abstracts checklist (Checklist 3) [24], the PRISMA-S (PRISMA Extension for Reporting Literature Searches in Systematic Reviews; Checklist 4) [25], and the ENTREQ (Enhancing Transparency in Reporting the Synthesis of Qualitative Research; Checklist 5) statement [26]. Because this review employed a qualitative synthesis approach without meta-analysis, the SWiM (Synthesis Without Meta-Analysis; Checklist 6) reporting guideline was also followed [27]. This review deviated from the protocol registered in PROSPERO (registration number CRD420251002601) in 2 respects. First, the age criterion for participant inclusion was broadened from “60 years or older” to “50 years or older” during screening to capture the heterogeneous experiences spanning the early-old to the oldest-old. Second, 2 additional databases (PsycINFO and Scopus) were added to the original search strategy to enhance the comprehensiveness of the literature retrieval. These amendments were implemented during the search and screening stages of the review and did not alter the overall research questions or synthesis approach.

Inclusion criteria for the study were (1) research topic: exploring older people’s experiences participating in community digital health communication; (2) participants were older adults (aged ≥50 y) residing in the community for over 6 months or members of a subgroup, regardless of health status, socioeconomic background, or digital technology experience; (3) qualitative or mixed methods research designs were used, and relevant qualitative data could be extracted. Exclusion criteria were (1) purely quantitative studies, case reports, and reviews; (2) studies where full-text access was unavailable or data were incomplete. If a study evaluated a broader or multicomponent program (eg, a comprehensive health education intervention) that combined communicative and intervention elements, it was not excluded on that basis; in such cases, only the participant accounts concerning digital health communication were extracted and synthesized at the data-extraction stage.

Systematic searches were conducted in PubMed (NLM), CINAHL (EBSCOhost), Embase (Elsevier), PsycINFO (EBSCOhost), Scopus (Elsevier), ProQuest Health & Medical Collection (ProQuest), Web of Science Core Collection databases (Clarivate), CNKI (China National Knowledge Infrastructure), and Wanfang Data. No study registries were searched. No additional studies or data were sought by contacting authors, experts, or manufacturers. The literature search period spanned from the establishment of each database to June 2026. Researchers manually searched the references, research protocols, and gray literature of relevant systematic reviews. Screening records that cited the included studies was also performed. Each database was searched separately, and no databases were searched simultaneously on a shared platform.

A combination of subject headings and expanded terms was employed, including search terms such as “AI, digital, e-health, online, health knowledge, medical advice, wellness facts, popular*, disseminat*, public participation,” among others. The search strategy was partly informed by a previous meta-synthesis of qualitative research [28], but we did not use published search filters. The full search strategies for each database are provided in Multimedia Appendix 1. Studies with abstracts published in Chinese or English were searched due to the language competencies of the review team. The initial search was conducted in March 2025. The search was rerun in May 2026 and June 2026 to capture any new studies published since the initial search. No other date or study-design restrictions were applied. The search strategy was not formally peer-reviewed using the PRESS (Peer Review of Electronic Search Strategies) checklist. However, it was iteratively refined through internal team discussions involving a research librarian and a senior researcher experienced in systematic reviews.

EndNote X9 (Clarivate Analytics) software was used for deduplication. Two researchers independently screened study titles, abstracts, and full texts of eligible studies. In cases of disagreement, a third researcher made the final decision.

Two researchers independently conducted detailed readings of the included studies, identifying their metaphors, concepts, and themes. Content extraction was performed using Microsoft Excel 2019. Disputes were resolved through discussions between the 2 researchers. No automation tools were used in the data collection process. The template data extraction form is provided in Multimedia Appendix 2.

The following study-level characteristics were extracted from each included study: first author, year of publication, country or region, study aim, study design, participant characteristics (sample size, age range, gender composition, health status, and digital technology experience), data collection method, analytical approach, and key qualitative findings (Multimedia Appendix 3).

As this is a qualitative meta-ethnography, no prespecified quantitative outcome measures or effect sizes were sought. The qualitative data items comprised 2 levels: first-order constructs, defined as direct quotations from study participants reflecting their lived experiences; and second-order constructs, defined as the interpretive themes and concepts developed by the original study authors based on participant data [29,30]. For studies that evaluated broader or multicomponent programs, only data concerning participants’ experiences of digital health communication—that is, their reception, exchange, understanding, and the use of health information—were extracted. Data relating exclusively to the noncommunicative components of an intervention were not extracted or synthesized. Where information was missing or unclear in the original reports, this was recorded in the data extraction form.

The quality of the included studies was assessed using the CASP (Critical Appraisal Skills Programme) qualitative research checklist (Multimedia Appendix 4) [31]. Two researchers independently assessed the quality of each included study, with disagreements resolved through discussion. All studies were incorporated into the data synthesis to enhance the depth and breadth of the meta-ethnography synthesis regardless of their quality scores [30].

This review employed a qualitative meta-ethnographic synthesis approach rather than a quantitative meta-analysis. Accordingly, the present synthesis does not report effect measures in the conventional sense; the “effects” reported here take the form of interpretive themes and higher-order conceptual constructs derived from the synthesis of qualitative findings across the included studies, rather than quantifiable effect estimates.

Our study adopted the Noblit and Hare [29] meta-ethnography as a qualitative synthesis method. Meta-ethnography is an interpretive approach to synthesizing qualitative research, originally developed by Noblit and Hare in 1988. Unlike aggregative approaches that merely summarize findings, meta-ethnography seeks to translate studies into one another by identifying and comparing key concepts, metaphors, and themes across studies, thereby generating new interpretive understandings that go beyond the findings of individual studies [29]. This inductive and interpretive approach allows for the development of conceptual understanding [30] and has been extensively used in synthesizing qualitative health care research [32]. The eMERGe (Meta-Ethnography Reporting Guidance) framework, which provides a set of structured reporting criteria specifically designed to improve the completeness and clarity of meta-ethnography reports, was used to guide the reporting of this review [33].

The data synthesis followed the phases of meta-ethnography: reciprocal translation, refutation, and line-of-argument syntheses [29]. Two researchers independently and repeatedly reviewed each included study and extracted its first-order and second-order constructs. Through line-by-line comparison across studies, conceptually similar constructs were grouped. Subsequently, using a cross-textual comparative approach, the researchers identified recurring concepts and metaphors while examining similarities and differences in research contexts to establish logical connections between themes and build third-order interpretations. Discrepancies at every stage were resolved through discussion, and the developing themes were checked against the original study data.

To mitigate potential reporting bias, we comprehensively searched 9 electronic databases and supplemented this with manual reference searching and gray literature screening. The risk of selective nonreporting of qualitative findings within individual studies was addressed as part of the CASP quality appraisal and the GRADE-CERQual (Grading of Recommendations Assessment, Development, and Evaluation—Confidence in the Evidence from Reviews of Qualitative Research) adequacy of data assessment.

We employed the GRADE-CERQual framework to assess the confidence in evidence [34]. Two researchers independently assessed the confidence in evidence for each review finding. The GRADE-CERQual assessment considered: (1) methodological limitations, (2) relevance, (3) coherence, and (4) adequacy of data. Based on these 4 components, an overall confidence assessment was made for each finding, categorized as high, moderate, low, or very low confidence. Each finding started with high confidence and was downgraded according to concerns identified across the 4 components, with the final rating agreed upon through discussion.

This systematic review utilized the PRISMA flow diagram to illustrate the literature screening process (Figure 1). The initial search yielded 2282 records. After removing duplicates, 1777 studies were selected for title and abstract screening. Ultimately, 14 studies meeting the inclusion criteria were incorporated into this systematic review.

All included studies incorporated qualitative research designs, with 3 using mixed methods approaches. Seven studies used thematic analysis, 3 applied content analysis, 1 adopted inductive analysis, 1 adopted deductive analysis, 1 adopted codebook thematic analysis, and another followed grounded theory methodology. The majority of the research was conducted in North America (n=7), followed by Asia (n=5) and Europe (n=2; Multimedia Appendix 3).

The average CASP score for studies included in this systematic review was 17.7 (SD 1.3) points, ranging from 15 to 20 points. Twelve studies were rated as moderate quality, while only 2 were classified as high quality (total scores of 19 or 20; Multimedia Appendix 4). Among these, 8 studies did not adequately address the relationship between researchers and participants, specifically, the researchers’ roles, potential biases, and influences during subject selection; how they managed events occurring during the study period; and whether they considered the impact of any changes to the study design.

Table 1 presents the confidence assessment of research findings based on the GRADE-CERQual methodology. As shown in Table 1, the overall confidence in the evidence was rated as high for all 4 themes, with all included studies demonstrating methodological quality at moderate or higher levels. Furthermore, the included studies feature rich interview data from diverse countries, alleviating concerns regarding adequacy.

Through the systematic integration of the included studies, 4 core themes were identified: core objectives, behavioral outcomes, trust and responsibility, and adapting to digital technologies. These themes comprehensively reveal the motivations, practical behavioral impacts, trust relationship dynamics, and the facilitating and hindering forces in the adoption of digital health technologies among community-dwelling older adults. Following the Noblit and Hare [29] synthesis process, the analysis revealed that the relationships across studies were predominantly reciprocal rather than refutational. No direct contradictions between studies were identified; instead, the included studies consistently reinforced and extended one another’s findings within each theme. Specifically, the unifying concept of “core objectives” centers on older adults’ proactive pursuit of health information and technological competence; “behavioral outcomes” converges on multidimensional empowerment spanning health management, social participation, and skill acquisition; “trust and responsibility” is unified by the interplay between authority-based trust and multistakeholder accountability; and “adapting to digital technology” shares a common dual-factor structure of promoting and barrier forces. Table 2 illustrates the cross-study construction process for each theme.

This qualitative systematic review and meta-ethnography synthesized qualitative evidence on the experiences of older adults engaging with community-based digital health communication. By integrating 14 studies conducted across 8 countries and involving community-dwelling older adults aged 50 years and older with diverse health conditions and levels of digital technology experience, 4 overarching themes were identified: core objectives, behavioral outcomes, trust and responsibility, and adapting to digital technology. The synthesis was characterized predominantly by reciprocal translations, suggesting substantial commonality in older adults’ experiences across cultural and geographic contexts; no fundamentally refutational findings were identified. Nevertheless, inherent tensions emerged within shared phenomena, most notably in the dual nature of family assistance and authority-based trust.

Our synthesis demonstrates that, within community-based digital health communication, older people are transitioning from recipients of medical instructions to proactive health managers [49]. This shift from passive treatment to active intervention has expanded health improvement beyond clinical indicators to include behavioral empowerment, highlighting the internalization of responsibility and awareness [50-52]. The findings are consistent with previous research on health communication [53,54].

In particular, the findings of our study indicate that digital technologies appear to have substantially reshaped the patterns of health information dissemination within communities. They have also prompted some older people to learn new technologies to avoid falling behind, reflecting both technological needs and the demand for social participation [55]. These outcomes illustrate that digital health communication supports networked interactions within communities, expanding the communication subjects from a unidirectional, provider-to-patient model to a multidirectional exchange [56]. Older people are not only information recipients but also become nodes for disseminating health information [57-59]. In this process, the older adult population builds networks of experiential empathy, fostering social connections and interactions that effectively alleviate feelings of loneliness [60].

Our study reveals that digital health technologies have profoundly reshaped the trust structures and responsibility allocation among older adults. Older participants consistently exhibit authority dependence, placing high trust in technology platforms recommended by health care professionals while valuing credibility endorsements from official institutions [61-64]. Regarding responsibility allocation, older people perceive their health responsibilities as shared among themselves, their families, health care providers, and the community [65]. We interpret this ideal multistakeholder collaboration as a potential closed-loop support system; in practice, however, this ideal is rarely fully realized [52]. Indeed, the introduction of digital media has led to notable vulnerabilities in responsibility allocation [66]. Some studies indicate that reliance on family technical assistance and medical authority can inhibit the autonomy of older people [43,44].

The synthesis also identified inherent complexities within the trust and responsibility dynamic. Our findings suggest that older adults face a paradoxical situation: the same digital environment that broadens their access to health information simultaneously overwhelms their capacity to evaluate its credibility [67]. This is consistent with previous research indicating that information overload and ambiguous sourcing pose particular challenges for populations with limited digital health literacy [68]. When older adults lack the skills to independently verify information, they may either withdraw from digital engagement or uncritically accept nonauthoritative content, both ultimately compromising health decision-making [67]. Privacy concerns further compound this complexity. Some older adults avoid data sharing due to fears of unauthorized use of health information, while insufficient transparency in technological design exacerbates their distrust of digital platforms [66]. Meanwhile, while family members’ technical assistance addresses immediate operational needs, it may simultaneously create a dependency that displaces opportunities for independent skill acquisition [69]. Digital media extends trust to intangible technological entities and shifts responsibility allocation from a person-to-person contract to a triangular relationship involving people, technology, and institutions [70-72]. This restructuring requires the community to move beyond its foundational technical support role and instead assume the critical function of coordinating responsibility allocation [73].

The findings of this study indicate that older adults possess relatively abundant experience and insights regarding the application of digital technologies when participating in community-based digital health communication. The MATOA (Model for the Adoption of Technology by Older Adults) suggests that older adults’ willingness to use technology is influenced by multiple factors, including physiological aging limitations, anxiety, prerequisite knowledge, intrinsic motivation, and perceived usefulness, with education, age, and gender serving as moderating variables [74]. This study validates the applicability of MATOA in the field of health communication.

While technological affinity enhancements mitigate age-related operational barriers, the resulting privacy risks diminish the perceived benefits of digital health technologies for older adults [75]. This suggests that usability improvements alone are insufficient; they must be accompanied by transparent data governance to sustain older adults’ willingness to engage [66]. Frequent technological updates and iterations create barriers to practical applications for older adults, requiring them to enhance their digital technology skills and digital health literacy continually [76]. If relevant knowledge is not adequately disseminated, it may substantially undermine older adults’ user experience and willingness to adopt technology, deepening their negative attitudes toward technology and amplifying their anxiety about it [7,75].

Social support, as a key component of intrinsic motivation and perceived usefulness, positively drives technology adoption through community mutual aid networks and family assistance [69]. However, as discussed above, this positive role is not unconditional. Generational divides and ageism create unique barriers: the transfer of technological authority within family settings distorts the original intent of social support, while societal assumptions about capability exacerbate older adults’ technological exclusion [77].

Public health emergencies propel digital technologies from being “optional” to becoming “essential necessities,” markedly amplifying the imperative for technological adaptation beyond the intrinsic motivation inherent in the original framework of MATOA [19]. Additionally, when trust between physician and patients reaches a certain level, older people may perceive the adoption of digital technologies, such as those recommended by physicians through digital health platforms, as shifting from merely “potentially beneficial” to a “necessary safeguard” [78]. This transformative effect is not fully reflected in MATOA’s “usage expectations.” Therefore, future research may consider extending the existing model to accommodate the demands of complex scenarios.

Based on the findings of this review, several implications for different professional groups can be drawn. For clinicians and health care providers, measures should balance technological assistance with older people’s autonomy, avoiding excessive dependence that leads to skill stagnation and insufficient support that causes digital exclusion. Systematic approaches must be implemented to enhance older people’s digital health literacy, strengthening their ability to verify health information. For community program designers and policymakers, attention must be paid to the dynamic shifts in health needs among older people, encouraging their transition from basic health management to personalized health care. The social nature of health communication within community settings should also be considered, with the design of tiered social support programs prioritizing isolated seniors, those of advanced age, and childless older individuals. For methodological and theoretical researchers, future research may consider extending the MATOA framework to accommodate the demands of complex scenarios, particularly those involving crisis-driven adoption and trust-mediated technology acceptance.

This study has certain limitations. Most included studies (n=12) were assessed as having moderate methodological quality, with only 2 rated as high quality. Future qualitative research should strengthen its methodological rigor. Findings from studies with lower methodological quality scores should be interpreted with caution. During the screening phase, only studies with abstracts published in Chinese or English were included, which may have resulted in the exclusion of relevant studies published in other languages, thereby limiting the representativeness of the studies and the generalizability of the findings. Additionally, our search strategy was iteratively refined through internal team discussions but was not formally peer-reviewed using the PRESS checklist, which may have introduced search-strategy bias.

Drawing on 14 studies across 8 countries, this qualitative meta-ethnography takes community-based digital health communication itself as the unit of synthesis, departing from earlier reviews that have centered on the determinants of technology adoption or on clinical-pathway effectiveness. This review reframes the community as more than a setting in which digital health communication occurs; the community is itself a factor that shapes whether and how such communication works for older adults. Older adults’ engagement is best understood as an ongoing negotiation of motivation, trust, and responsibility, with that responsibility distributed unevenly across older adults, their families, peers, and health systems. These insights argue for a shift in practice and policy: policymakers, community program designers, and clinicians should move beyond efficiency-oriented service delivery toward building collective community capacity, tiered support for the most isolated older adults, and protection of older adults’ autonomy and capacity to verify health information.