Introduction
Chronic kidney disease (CKD) is a progressive condition characterized by the gradual loss of kidney function, ultimately leading to end-stage kidney disease (ESKD) in a substantial proportion of patients. Globally, CKD affects approximately 10%–15% of adults and represents a major contributor to morbidity, mortality, and healthcare resource utilization [1–3].
The CKD continuum encompasses a wide spectrum of disease stages—from early renal impairment to patients requiring dialysis or kidney transplantation—each associated with distinct clinical challenges, comorbidities, and psychosocial burdens. Effective management across this continuum demands rigorous adherence to complex therapeutic regimens, lifestyle modifications (including diet and physical activity), and routine monitoring of biochemical, physiological, and psychosocial parameters [4,5]. However, traditional care models have often faced significant limitations in providing continuous patient support, individualized guidance, and timely intervention, particularly given the increasing prevalence of CKD and constraints on healthcare resources [6–8].
Of note, digital health interventions, including mobile health (mHealth) applications, telehealth platforms, wearable devices, web-based programs, and multicomponent eHealth systems, have emerged as transformative tools in chronic disease management. These interventions are designed to enhance patient engagement, support self-management, and facilitate real-time monitoring of clinical and behavioral outcomes [9,10]. In the context of CKD, digital tools can provide individualized feedback, reminders for medication or dialysis adherence, dietary guidance, symptom tracking, educational content, and virtual consultations, all of which may help mitigate disease progression and improve patient quality of life (QoL) [11–13].
Over the past decade, an increasing number of systematic reviews and meta-analyses have evaluated digital interventions in CKD populations, targeting a range of outcomes including clinical indicators (e.g., blood pressure, serum creatinine, dialysis adherence), behavioral outcomes (e.g., medication adherence, dietary management), psychosocial outcomes (e.g., self-efficacy, QoL, depression, anxiety), and patient-reported experiences [12,14–16]. While many reviews demonstrated positive effects of digital interventions—particularly on proximal outcomes such as treatment adherence, blood pressure control, and engagement with self-management activities—findings are heterogeneous, with inconsistent or modest effects reported for distal outcomes including QoL, long-term renal function, and hospitalization rates [12,14–18]. This variability reflects differences in intervention design (e.g., mobile apps vs. telemonitoring vs. multicomponent platforms), intensity, duration, patient populations (predialysis, dialysis, or transplant), and methodological rigor of primary studies.
Most importantly, the expanding yet heterogeneous body of literature on digital health interventions for CKD underscores the need for an umbrella review to synthesize and critically evaluate existing evidence. The proliferation of reviews has also highlighted critical methodological concerns, including small sample sizes, short intervention durations, reliance on self-reported outcomes, high heterogeneity across studies, and limited use of standardized outcome measures. Moreover, some reviews focused primarily on narrow patient subgroups or single intervention types, limiting the generalizability of findings across the broader CKD population [14,15]. These challenges underscore the need for a high-level synthesis that systematically evaluates and integrates evidence from multiple reviews while assessing methodological quality and risk of bias.
Umbrella reviews provide a robust framework for addressing these challenges by synthesizing findings from existing systematic reviews and meta-analyses. This approach allows for comprehensive mapping of available evidence, identification of consistent patterns of effectiveness, critical appraisal of methodological quality, and detection of research gaps. Indeed, by consolidating evidence across studies, umbrella reviews offer clinicians, policymakers, and researchers a clear overview of the efficacy and limitations of interventions, supporting evidence-informed decision-making and guiding the design of future studies [19,20].
Despite the growing evidence base, no prior umbrella review has comprehensively evaluated the effects of digital health interventions across the full CKD continuum—including predialysis, dialysis, and kidney transplant populations—while integrating clinical, behavioral, and psychosocial outcomes. Furthermore, systematic appraisal of methodological rigor and risk of bias in these reviews is lacking, which limits confidence in the current evidence and the translation of findings into practice.
The present umbrella review addresses these gaps by synthesizing high-quality systematic reviews and meta-analyses examining digital health interventions in CKD. Specifically, it seeks to (1) summarize the types of digital interventions studied, (2) evaluate their effectiveness across clinical, behavioral, and psychosocial outcomes, (3) assess methodological quality and risk of bias of included reviews, and (4) identify areas for future research to optimize digital health strategies in CKD management.
Materials and methods
This umbrella review was conducted following the PRIOR (Preferred Reporting Items for Overviews of Reviews) guidelines [21]. The complete PRIOR checklist is provided in Supplementary Table 1 (available online).
Eligibility criteria
This umbrella review included systematic reviews and meta-analyses evaluating the effects of digital health interventions on outcomes across the CKD continuum, including predialysis, dialysis, and kidney transplant populations. Eligible reviews were required to follow a systematic methodology, incorporating a clearly defined research question, a comprehensive search across at least two electronic databases, and a structured synthesis of findings. Only peer-reviewed reviews adhering to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [22] guidelines were included, ensuring methodological transparency and reproducibility.
Reviews were excluded if they did not explicitly report patient-centered, behavioral, clinical, or psychosocial outcomes, or if relevant outcomes were secondary within broader studies not focused on CKD management. Studies that exclusively examined provider-level or system-level interventions without patient-related outcomes, or that included preclinical, animal, or in vitro data without separate human analyses, were excluded to maintain clinical relevance. Studies that assessed self-management but did not specifically focus on or report digital self-management were excluded.
To ensure methodological rigor, narrative reviews, opinion pieces, single-author reviews, and reviews based on a single database were excluded. Reviews lacking defined inclusion criteria, systematic data synthesis, or quality assessment of included studies were not considered. These criteria align with established umbrella review standards, ensuring the reliability, applicability, and robustness of the synthesized evidence regarding digital interventions in CKD populations. A summary of the inclusion and exclusion criteria for this umbrella review is provided in Table 1, outlining the characteristics of eligible reviews, populations, interventions, outcomes, and quality thresholds, as well as reasons for exclusion of ineligible studies.
Information sources and search strategy
To comprehensively capture the current state of evidence, we systematically searched six major scientific databases: PubMed (MEDLINE), Embase, Web of Science, Cochrane Library, Epistemonikos, and Google Scholar. The search included peer-reviewed, English-language systematic reviews and/or meta-analyses published up to December 18, 2025, with no lower date restriction.
The search strategy encompassed three core domains: (1) CKD populations and treatment modalities, including terms such as “chronic kidney disease,” “dialysis,” “hemodialysis,” “peritoneal dialysis,” and “kidney transplantation”; (2) digital health interventions, including “eHealth,” “mHealth,” “telehealth,” “mobile application,” “digital self-management,” and “computer-assisted intervention”; and (3) review-type descriptors, including “systematic review” and “meta-analysis.” Boolean operators were applied to generate comprehensive query combinations, and controlled vocabulary (e.g., MeSH terms in PubMed) was used when applicable. An example of the PubMed search string was: (“chronic kidney disease” OR “dialysis” OR “kidney transplant”) AND (“digital health” OR “eHealth” OR “mHealth” OR “telehealth” OR “mobile application”) AND (“systematic review” OR “meta-analysis”).
Database searches were adapted to account for platform-specific functionality and indexing. Full search strategies are detailed in Supplementary Table 2 (available online). Reference lists of included studies were manually screened to identify any additional relevant reviews. Duplicates were removed using EndNote 2025 (Clarivate), and the remaining records were imported into a systematic review management tool for screening and eligibility evaluation.
Selection process
The selection of eligible systematic reviews and meta-analyses was conducted following a structured and transparent protocol. Each study was assessed using the PICOS (Population, Intervention, Comparator, Outcomes, and Study Design) framework and evaluated for adherence to the PRISMA [22] and PRISMA-ScR (PRISMA for Scoping Reviews) [23] guidelines.
To minimize redundancy, when multiple reviews addressed overlapping research questions (e.g., similar populations or interventions), preference was given to the most recent, comprehensive, and methodologically rigorous reviews. Reviews that conducted sensitivity analyses, clearly defined inclusion criteria, and provided detailed descriptions of populations and interventions were prioritized. The selection process was carried out independently by three reviewers (TD, YX, and NM). Titles and abstracts were initially screened, followed by full-text review to determine final inclusion. Any discrepancies during screening or eligibility assessment were resolved through group discussion until consensus was achieved.
The methodological quality of included reviews was evaluated using AMSTAR-2 (A Measurement Tool to Assess Systematic Reviews 2) [24] (Supplementary Table 3, available online). Two reviewers independently rated each review, with any disagreements resolved through consultation with the full reviewer team. Reviews rated as critically low in quality according to AMSTAR-2 were excluded from the final analysis. This rigorous, multistep approach ensured that only high-quality, non-redundant evidence was synthesized in this umbrella review.
Data items, synthesis methods, and risk of bias assessment
This umbrella review synthesizes findings from 16 systematic reviews and meta-analyses [12,14–17,25–35] examining the effects of digital health interventions across CKD populations, including predialysis CKD, dialysis, and kidney transplant recipients. The included reviews focused on clinical and patient-centered outcomes such as treatment adherence, lifestyle and dietary management, psychological well-being, QoL, self-efficacy, and selected physiological or biochemical indicators associated with intervention use.
Screening, data extraction, and verification were performed independently by three researchers (TD, YX, and NM) to enhance reliability. Duplicate records were managed using EPPI-Reviewer (version 6.15.0.2, UCL Social Research Institute) and Microsoft Excel 2021 (Microsoft Corp.). Any discrepancies arising during selection or extraction were resolved through discussion until consensus was achieved. To evaluate the methodological rigor and potential bias of the included reviews, the ROBIS tool [36] was employed, with emphasis on clear eligibility criteria, comprehensive search strategies, and robust data collection and synthesis. For reviews warranting further examination, the Cochrane Risk of Bias [37] tool was applied to their primary studies to assess study-level bias (Supplementary Table 4, available online). This comprehensive approach ensures the reliability and trustworthiness of the evidence synthesized in this umbrella review.
Results
Database searches initially identified 28 studies, with three more retrieved through reference list screening. Of the 31 systematic reviews and meta-analyses considered, 16 met the inclusion criteria and were included for in-depth evaluation [12,14–17,25–35]. The selection process is depicted in Fig. 1.
Table 2 summarizes the methodological features of the included studies, organized by the PICOS framework.
Broad electronic health and digital self-management interventions across chronic kidney disease populations
Several systematic reviews have examined heterogeneous eHealth and information technology-based self-management interventions across the CKD continuum, encompassing predialysis CKD, dialysis-dependent populations, and kidney transplant recipients. Collectively, these reviews illustrate a recurring pattern: digital interventions tend to exert their most robust and consistent effects on proximal, behaviorally mediated outcomes, whereas effects on distal clinical endpoints remain variable and often inconclusive [12,25,26,31].
Shen et al. [25] offered one of the earliest and most comprehensive syntheses, integrating evidence from randomized trials and mixed-methods studies. Their review demonstrated consistent improvements in blood pressure control, medication adherence, self-monitoring behaviors, and patient knowledge, alongside high levels of feasibility, usability, and patient satisfaction. Importantly, these benefits were most evident in interventions that combined automated feedback, remote monitoring, and patient education within multicomponent digital platforms. In contrast, outcomes that require longer biological latency—such as health-related QoL, hospitalization rates, and renal function decline—showed inconsistent or null effects, underscoring the temporal mismatch between short intervention durations and the slow progression of kidney disease [25].
Similarly, Jeddi et al. [26] focused specifically on automated IT-based self-management systems, including web-based platforms, telemonitoring tools, and algorithm-driven feedback systems. Across predominantly randomized controlled trials, the authors reported largely favorable effects on clinical indicators (e.g., blood pressure, biochemical parameters) and process-of-care outcomes (e.g., treatment adherence, appointment attendance, and self-efficacy). Nevertheless, the strength of inference was constrained by substantial intervention heterogeneity, variability in outcome measurement, and relatively short follow-up periods, which collectively limited conclusions regarding sustainability and long-term clinical impact [26].
More recent reviews by Yang et al. [31] and Zheng et al. [12] reinforced and extended these observations within the context of mHealth and digitally mediated symptom-management tools. These reviews reported neutral-to-positive effects on self-monitoring behaviors, patient engagement, symptom awareness, and selected QoL domains, particularly when interventions leveraged mobile applications, short message service reminders, and interactive feedback mechanisms. Notably, interventions targeting patient-reported outcomes—such as symptom burden, perceived control, and self-efficacy—tended to demonstrate greater consistency than those targeting hard clinical endpoints.
Despite these encouraging trends, both reviews emphasized persistent methodological and conceptual gaps. These included limited use of explicit behavioral or self-management theories to guide intervention design, underrepresentation of long-term and post-intervention follow-up, and insufficient attention to safety, data governance, and ethical considerations. Moreover, sociocultural tailoring, health literacy adaptation, and equity-focused design were rarely addressed, raising concerns about scalability and real-world implementation across diverse CKD populations. Taken together, the evidence suggests that while digital self-management interventions are feasible and effective for enhancing engagement and intermediate outcomes in CKD, their translation into durable clinical benefit remains contingent on improved theoretical grounding, longer evaluation horizons, and more context-sensitive implementation strategies.
Psychological health, quality of life, and patient-reported outcomes in advanced chronic kidney disease and end-stage kidney disease
A distinct subset of systematic reviews has concentrated primarily on psychological and patient-reported outcomes, reflecting a growing recognition that emotional well-being, perceived control, and self-efficacy are central determinants of long-term self-management in advanced kidney disease. Within this cluster, digital health interventions appear particularly well-suited to address domains that are inherently subjective, dynamic, and responsive to behavioral support.
Zhu et al. [27] provided the most comprehensive and up-to-date synthesis in this domain, evaluating randomized trials of digital interventions targeting psychological health, self-efficacy, and QoL in adults with ESKD across hemodialysis, peritoneal dialysis, transplantation, and conservative management. The review demonstrated statistically significant improvements in depressive symptoms and overall QoL, alongside meaningful reductions in anxiety. These effects were most pronounced in interventions delivered via mobile applications and telemedicine platforms, particularly among hemodialysis patients, suggesting that frequent, structured digital contact may partially mitigate the psychological burden associated with treatment intensity and lifestyle restriction. Effects on self-efficacy, however, were more heterogeneous and modality-dependent, with benefits emerging primarily in interventions incorporating interactive feedback, goal setting, and real-time monitoring rather than passive educational content alone [27].
Despite the overall favorable direction of effect, Zhu et al. [27] highlighted substantial between-study heterogeneity related to intervention duration, theoretical underpinning, and outcome measurement. Most trials relied on self-reported instruments to assess psychological outcomes, and few incorporated objective behavioral correlates or clinician-rated endpoints. Consequently, the certainty of evidence was rated as low to moderate, reflecting concerns regarding measurement bias, variability in intervention fidelity, and limited follow-up beyond the active intervention phase [27].
Complementing these findings, Ng et al. [15] focused specifically on telehealth self-management interventions in hemodialysis populations and conducted a quantitative synthesis of self-efficacy outcomes. Their meta-analysis identified a statistically significant, medium-sized improvement in self-efficacy, underscoring the capacity of telehealth modalities to enhance patients’ confidence in managing complex treatment regimens, symptom monitoring, and lifestyle adjustments. Notably, self-efficacy improvements did not consistently translate into significant changes in downstream clinical outcomes, such as interdialytic weight gain, serum electrolyte control, or health-related QoL, although trends generally favored intervention groups [15]. This apparent dissociation between psychological gains and hard clinical endpoints likely reflects a combination of short intervention durations, limited statistical power, and the multifactorial nature of clinical outcomes in ESKD, which are influenced by comorbidity burden, dialysis adequacy, and structural care constraints. Moreover, Ng et al. emphasized the very low certainty of evidence across most outcomes, driven by small sample sizes, cultural and contextual heterogeneity, and inconsistency in intervention components and outcome definitions [15].
Taken together, these reviews suggest that digital health interventions may exert their strongest and most reproducible effects on psychological constructs—particularly depression, anxiety, and self-efficacy—that are proximal to patient engagement and behavioral change. However, the translation of these psychological benefits into sustained behavioral modification and clinically meaningful outcomes remains insufficiently established. Future trials would benefit from explicitly theorized intervention frameworks, integration of psychological and behavioral endpoints with objective clinical measures, and longer-term follow-up to clarify whether early improvements in patient-reported outcomes function as true mediators of downstream health benefit in ESKD populations.
Digital interventions targeting treatment and medication adherence
Several systematic reviews converged on treatment adherence as a central and clinically actionable outcome, reflecting the pivotal role of adherence in determining morbidity and healthcare utilization across dialysis and broader CKD populations. Within this body of evidence, digital health interventions consistently demonstrated their greatest strengths in modifying discrete, behaviorally proximal targets, particularly medication-taking behaviors and attendance or compliance with dialysis regimens.
Zhang et al. [28] provided the most rigorous synthesis of adherence-focused digital interventions in dialysis populations, demonstrating robust and statistically significant improvements in overall treatment adherence. The largest effect sizes were observed for medication adherence and dialysis adherence, outcomes that are highly structured, repetitive, and amenable to digital reinforcement through reminders, self-monitoring, and feedback loops. In contrast, dietary and fluid adherence exhibited more modest but still meaningful improvements, likely reflecting the greater cognitive, environmental, and social complexity inherent in sustained lifestyle modification. Although the direction of effect was uniformly favorable across included trials, substantial heterogeneity in intervention modalities, adherence definitions, and measurement tools, alongside predominantly short follow-up periods, limited inference regarding the durability of effects beyond the intervention window [28].
Two complementary reviews—Paneerselvam et al. [34] and Ghozali et al. [16]—narrowed the analytic lens to mobile app-based interventions, providing greater granularity on which digital features may drive adherence-related benefit. Across these reviews, modest-to-moderate improvements were observed in self-management behaviors, medication adherence, systolic blood pressure, and patient satisfaction. Interventions that integrated multiple behavior-support mechanisms—such as automated reminders, gamification elements to sustain engagement, educational content tailored to CKD-specific regimens, and in some cases electronic medication monitoring—consistently outperformed simpler, single-component apps. These findings align with behavioral theory, suggesting that adherence is more effectively supported when cognitive reinforcement, motivation, and feedback are addressed simultaneously rather than in isolation.
However, both reviews [16,34] underscored important limitations in the underlying evidence base. Improvements in biochemical and physiological outcomes, including serum electrolytes, body weight, and disease burden indices, were inconsistent or nonsignificant, raising questions about the extent to which improved adherence metrics translate into measurable clinical benefit. Evidence quality was generally rated as low, driven by small sample sizes, short intervention durations, heterogeneity in app design, and reliance on self-reported adherence measures. Moreover, few studies evaluated long-term engagement, scalability, or integration into routine clinical workflows.
Collectively, these adherence-focused reviews suggest that digital interventions—particularly mobile applications—are effective tools for enhancing structured, routine-dependent behaviors in CKD and dialysis populations. Yet, the sustainability of adherence gains and their downstream impact on biochemical control and hard clinical outcomes remain incompletely defined. Future research should prioritize longer-term trials, standardized adherence metrics, and explicit linkage between digital adherence support, intermediate physiological markers, and patient-centered outcomes to better delineate the clinical significance of these interventions.
Telehealth and digital interventions for lifestyle and dietary management
Ellis et al. [29] and Kosa et al. [35] synthesized evidence specifically addressing telehealth and mobile nutritional applications designed to modify lifestyle and dietary behaviors in CKD populations, spanning both non-dialysis and dialysis settings. Across these reviews, digitally delivered nutritional interventions—often incorporating dietitian support, dietary self-monitoring, feedback, and educational components—were consistently associated with improvements in dietary quality and dietary adherence. In parallel, modest but favorable effects were observed for blood pressure control and patient-reported constructs such as perceived control and self-management capacity, suggesting that digital tools may enhance patients’ sense of agency in managing complex dietary regimens.
In contrast, effects on downstream clinical and biochemical outcomes, including estimated glomerular filtration rate (eGFR), serum electrolytes, and other laboratory markers, were mixed or inconsistent. This pattern likely reflects both the relatively short duration of most interventions and the multifactorial determinants of renal and metabolic biomarkers, which may be less responsive to isolated dietary change over limited time horizons. Additionally, variability in baseline disease severity, dialysis modality, and nutritional prescriptions further complicates the detection of consistent biochemical effects across studies.
Methodologically, both reviews [29,35] were characterized by small, predominantly single-center studies, substantial heterogeneity in intervention design and outcome measurement, and follow-up periods ranging from weeks to a few months. Despite these limitations, a notable and consistent finding across both syntheses was high usability, feasibility, and acceptability of nutritional digital tools, even among older and dialysis-dependent patients. This suggests that dietary self-management represents a particularly suitable target for digital intervention, given its reliance on frequent self-monitoring and real-time decision-making.
Digital health in kidney transplant recipients
Eslami et al. [32] and Kuang et al. [33] examined the role of digital health interventions in kidney transplant recipients, a population characterized by complex, lifelong self-management demands and narrow therapeutic windows for immunosuppressive therapy. Across both reviews, digital interventions—most commonly smartphone applications and multicomponent systems integrating reminders, telemonitoring, feedback, and clinician communication—were associated with predominantly positive effects on medication adherence, self-management behaviors, and patient-reported QoL. Improvements were also observed in selected clinical and care-process outcomes, including reductions in medication errors, unplanned admissions, and indicators of health-related anxiety or self-burden [32,33].
Notably, Kuang et al. [33] highlighted favorable economic signals, with several studies reporting reductions in unplanned hospitalizations, nursing costs, and overall medical expenditure, suggesting that digital follow-up may enhance the efficiency of posttransplant care. Smartphone-based and multimodal interventions consistently outperformed single-component approaches, underscoring the importance of combining behavioral support with real-time monitoring and feedback in this high-risk population.
Despite these encouraging trends, substantial heterogeneity in intervention design, outcome definitions, and follow-up duration precluded quantitative synthesis in both reviews [32,33]. Moreover, effects on tacrolimus concentrations—a critical surrogate of immunosuppressive adherence—were inconsistent and inconclusive, likely reflecting variability in pharmacokinetics, monitoring frequency, and baseline adherence levels. Overall, while digital health interventions appear promising for supporting posttransplant self-management and reducing healthcare utilization, more standardized, theory-informed trials with longer follow-up are required to establish their impact on hard graft-related and pharmacological outcomes.
Digital health interventions in primary care and provider-focused contexts
In contrast to patient-centered interventions, Hui et al. [17] and Galbraith et al. [30] evaluated digital and computer-assisted interventions implemented at the primary care or provider level. Across both reviews, these interventions—typically comprising electronic reminders, computer-assisted decision support, or education-based chronic disease management tools—demonstrated minimal or nonsignificant effects on key care quality indicators, including angiotensin-converting enzyme inhibitor/angiotensin receptor blocker prescribing, blood pressure target achievement, proteinuria monitoring, documentation quality, and referral practices. Meta-analytic findings from Hui et al. [17] further indicated no statistically significant improvement in overall care quality, accompanied by substantial heterogeneity and limited intervention uptake.
Several structural and contextual factors likely underlie these null findings. Intervention reach and sustained use by clinicians were frequently low, reflecting competing clinical demands, alert fatigue, and limited integration into routine workflows. Moreover, most interventions emphasized provider behavior modification without parallel patient engagement or self-management support, potentially attenuating downstream clinical impact. Outcome selection also favored process measures rather than patient-centered or long-term clinical endpoints, further constraining observable effect sizes. Collectively, these findings suggest that digital interventions targeting providers alone may be insufficient to meaningfully improve CKD care quality unless embedded within broader, multilevel models that simultaneously address patient engagement, care coordination, and system-level incentives [17,30].
Qualitative evidence on user preferences and design principles
Finally, Trần et al. [14] uniquely contributed qualitative evidence elucidating patient preferences and priorities for digital lifestyle interventions across CKD stages and treatment modalities. Patients consistently highlighted the need for simplicity and intuitive design, reflecting concerns about cognitive load and technology literacy. Personalization—tailoring content, feedback, and goal-setting to individual clinical profiles, lifestyle habits, and motivational factors—was identified as a key driver of engagement and sustained use. Social support features, such as peer communities or caregiver involvement, were valued for encouragement and accountability, while educational content was most meaningful when directly linked to actionable recommendations, rather than abstract information. Timely reminders and automated monitoring were also emphasized as essential for maintaining adherence and fostering a sense of control over disease management [14].
These qualitative insights provide critical contextual evidence, complementing quantitative findings across adherence, self-management, and psychosocial outcomes. They underscore that intervention effectiveness depends not only on technological functionality but also on alignment with patient values, cognitive capacities, and behavioral motivations. Integrating such user-centered design principles across digital interventions—whether for self-management, dietary adherence, or lifestyle modification—may enhance engagement, acceptability, and ultimately clinical and psychosocial impact.
Overall synthesis
Across intervention categories, digital health interventions in CKD consistently demonstrated the strongest and most reliable effects on behavioral outcomes (e.g., medication adherence, dietary and fluid management), process-of-care measures (e.g., self-monitoring frequency, engagement with telehealth platforms), and patient-reported outcomes (e.g., QoL, self-efficacy, mood). Improvements were particularly notable in interventions that incorporated personalization, automated reminders, educational content linked to actionable behaviors, and social support features, reflecting alignment with patient preferences and engagement drivers identified in qualitative studies.
In contrast, effects on hard clinical endpoints—including kidney function markers (eGFR, serum creatinine), cardiovascular outcomes, hospitalization rates, and mortality—remained inconsistent, frequently limited by small sample sizes, short follow-up periods, and heterogeneous intervention designs. Even where positive trends were observed, effect sizes were generally modest and insufficiently powered to draw definitive conclusions (Table 3).
Discussion
This umbrella review synthesized evidence from 16 systematic reviews and meta-analyses [12,14–17,25–35] evaluating digital health interventions across the CKD continuum, including predialysis, dialysis, and kidney transplant populations. Overall, the findings suggest that digital interventions—ranging from mobile applications and telehealth platforms to web-based and wearable-supported systems—consistently improve patient-centered outcomes such as treatment adherence, self-management behaviors, dietary and lifestyle management, QoL, and, to a variable extent, psychological well-being and self-efficacy. Notably, interventions targeting medication adherence and dialysis-related behaviors showed the most robust and consistent positive effects, whereas improvements in distal clinical outcomes, such as kidney function or long-term biochemical markers, were less consistently observed [12,14–17,25–35].
Based on analyzed studies [12,14–17,25–35], digital health interventions for CKD consistently improved proximal outcomes such as medication adherence, dialysis adherence, blood pressure control, dietary compliance, and self-management, with the strongest effects observed in interventions combining reminders, gamification, patient education, and monitoring tools. Notably, distal outcomes—including QoL, kidney function (eGFR), biochemical measures, and long-term clinical endpoints—showed mixed or inconsistent effects, often limited by short follow-up periods, heterogeneity, and study design. Mental health outcomes such as depression and anxiety improved in several studies, while self-efficacy showed modest but significant gains, particularly among hemodialysis patients using app- or telemedicine-based interventions. Process-of-care outcomes, including appointment adherence and treatment engagement, generally improved, and high patient satisfaction, usability, and acceptability were consistently reported. Nutritional and dietary apps produced reliable improvements in dietary behavior and adherence. Overall, effects varied by intervention type, study population, and modality, with some interventions showing neutral or nonsignificant impacts, particularly on overall care quality or specific clinical measures. In summary, digital health interventions for CKD are reliably beneficial for proximal outcomes and patient engagement, moderately effective for mental health and dietary management, but show variable impact on distal clinical outcomes.
Several interrelated mechanisms likely account for the beneficial effects observed across digital health interventions. First, automated reminders and alerts—delivered via mobile applications, text messaging, or telemonitoring systems—support medication adherence, appointment attendance, and dialysis-related routines by reducing forgetfulness and treatment burden [11,38]. Second, structured educational content embedded within digital platforms improves disease-related knowledge, health literacy, and patients’ understanding of complex CKD regimens, thereby strengthening self-management capacity. Interactive features, such as symptom tracking, bidirectional communication with healthcare providers, and remote monitoring of physiological parameters, enable early identification of clinical or behavioral deviations and foster a sense of continuous care rather than episodic clinic-based management [11,34,39,40].
Moreover, personalization algorithms that adapt content, feedback, or goal-setting to individual patient characteristics, disease stage, and behavioral patterns appear to enhance engagement and intervention relevance [41]. Gamification strategies—such as progress visualization, achievement badges, and motivational prompts—may further reinforce positive behaviors by leveraging reward-based learning and self-efficacy principles, contributing to more sustained behavioral change over time [42–44]. Despite these promising mechanisms, substantial heterogeneity in intervention design, technological complexity, target behaviors, outcome measures, and CKD populations limits direct cross-study comparisons and quantitative synthesis [11,12]. This variability underscores the need for standardized intervention reporting frameworks, harmonized digital health taxonomies, and the adoption of core outcome sets to improve comparability, reproducibility, and cumulative evidence generation in future digital nephrology research.
Across the included reviews, digital health interventions encompass a diverse set of named devices and platforms that collectively shift CKD management from episodic clinic-based care toward continuous, patient-centered support (Fig. 2).
Among mHealth applications, commercial platforms such as MyKidneyCare (Ardia Digital Health Ltd) and research- or institution-based applications such as Renal Health, KidneyAPPetite, My Dialysis Planner, and MyFluid are frequently cited. These applications typically function as integrated self-management hubs, combining medication reminders, dialysis scheduling, dietary and fluid intake tracking, symptom reporting, and stage-specific educational content. Many operate through rule-based decision engines that translate patient-entered data into immediate feedback, such as alerts for excessive sodium or fluid intake or missed medications. More advanced systems incorporate adaptive personalization, adjusting recommendations based on longitudinal behavior, laboratory trends, or clinician-defined targets [45–47]. Thus, by externalizing complex self-care tasks into structured digital workflows, these apps reduce cognitive burden, reinforce routine formation, and enhance self-efficacy.
Telemonitoring technologies are particularly prominent in advanced CKD and dialysis populations. Systems such as Nx2Me, Amwell (Amwell), Vivify Health (Vivify Health), and Health Buddy (Health Hero Network) integrate home-based physiological sensors—including Bluetooth-enabled blood pressure monitors, digital weight scales, and glucose meters—with secure cloud-based dashboards [48–50]. In peritoneal dialysis, cycler-integrated platforms such as Baxter Sharesource (Baxter Healthcare Corp.) automatically transmit treatment data to clinical teams. Embedded analytics identify deviations from patient-specific baselines, enabling early detection of fluid overload, blood pressure instability, or adherence lapses. This continuous monitoring model supports proactive clinical intervention while simultaneously strengthening patient accountability through real-time physiological feedback [51,52].
Wearable devices, including Fitbit (Google LLC), Apple Watch (Apple Inc.), and Garmin trackers (Garmin International Inc.), contribute objective measures of physical activity, sleep patterns, and heart rate variability [53]. In CKD populations, these metrics provide clinically relevant proxies for fatigue, functional capacity, and cardiovascular health. When integrated with digital coaching platforms, wearable-derived data are transformed into personalized activity goals, progress visualizations, and motivational prompts, often employing gamification strategies grounded in behavioral science to sustain engagement and promote gradual lifestyle modification [53–56].
Telehealth platforms such as Teladoc Health platform (Teladoc Health, Inc.), Zoom for Healthcare (Zoom Communications, Inc.), Doxy.me (Doxy.me Inc.), and Epic MyChart (Epic Systems Corporation) further extend the reach of digital interventions by enabling structured video consultations, secure asynchronous messaging, and remote symptom assessments [57]. These systems facilitate timely education, psychosocial support, and treatment optimization, particularly for patients with limited access to specialized nephrology services. Asynchronous communication channels allow patients to report symptoms or concerns in real time, while clinicians respond within clinically appropriate timeframes, enhancing continuity of care and responsiveness [57–59].
Collectively, these named technologies operate through interconnected feedback loops that combine continuous data capture, algorithm-driven interpretation, and clinician oversight. Their observed benefits arise not merely from technological sophistication but from their ability to align real-time data with patient behavior, motivation, and clinical decision-making. Nonetheless, heterogeneity in platform design, data integration, and outcome reporting across studies underscores the need for standardized evaluation frameworks to improve comparability, scalability, and long-term implementation of digital health solutions in CKD care.
Limitations and future directions
Despite promising evidence supporting the use of digital health interventions in CKD, several important limitations constrain the strength, generalizability, and clinical translation of current findings. Foremost, substantial heterogeneity exists across interventions, populations, and outcome measures. Digital platforms range from simple reminder-based applications to complex, multicomponent systems integrating telemonitoring, wearables, and clinician dashboards. This variability, coupled with differences in CKD stage, treatment modality, age, and digital literacy, limits direct cross-study comparisons and precludes robust meta-analytic synthesis in many reviews.
Methodological limitations are also prominent. Many primary studies are small, short in duration, and underpowered to detect changes in distal clinical outcomes such as eGFR decline, hospitalization, or mortality. Reliance on self-reported measures for adherence, QoL, and behavioral change introduces reporting bias, while blinding is often infeasible due to the nature of digital interventions. Furthermore, several reviews report low to very low certainty of evidence, driven by risk of bias, inconsistency, and imprecision. Long-term sustainability, cost-effectiveness, and real-world implementation remain insufficiently explored.
Another critical gap lies in the limited theoretical grounding of many interventions. Behavioral change mechanisms are often implicit rather than explicitly designed or evaluated, and few studies systematically link intervention components—such as reminders, feedback, or gamification—to specific cognitive or motivational pathways. Equity considerations are similarly underrepresented; older adults, individuals with limited health literacy, and patients in low-resource settings are frequently excluded, raising concerns about digital divide–related disparities.
Future research should prioritize well-designed, adequately powered randomized trials with longer follow-up periods and standardized core outcome sets that encompass clinical, behavioral, and patient-reported endpoints. Greater transparency in reporting intervention components, data flows, and algorithms is essential to enable replication and mechanistic insight. Integration of implementation science frameworks will be critical to assess scalability, acceptability, and cost-effectiveness across diverse healthcare systems. Finally, next-generation digital interventions should emphasize adaptive personalization, interoperability with electronic health records, and co-design with patients and clinicians to ensure that technological innovation translates into meaningful, equitable, and durable improvements in CKD care.
Conclusion
The aim of this study was to synthesize and critically evaluate the existing high-level evidence on the effectiveness of digital health interventions across the continuum of CKD care. This study found that digital health interventions consistently demonstrate beneficial effects on behavioral outcomes, process-of-care indicators, and patient-reported measures such as adherence, self-management, self-efficacy, and QoL. Evidence for effects on hard clinical endpoints remains mixed and limited, largely due to short follow-up durations, heterogeneity of interventions, and insufficient statistical power. Overall, the findings support digital health tools as valuable adjuncts to conventional CKD management, while underscoring the need for theory-driven, standardized, and longitudinally evaluated interventions to strengthen clinical impact and inform routine implementation.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement table 1
Print
