Skip to main navigation Skip to main content
  • KSCN
  • E-Submission

CNR : Clinical Nutrition Research

OPEN ACCESS
Search
Advanced Search
ABOUT
BROWSE ARTICLES
EDITORIAL POLICIES
FOR CONTRIBUTORS

Articles

Page Path

Review Article

Personalized nutrition interventions using digital technologies for older adults: a scoping review

Soyoung Jungorcid, Hae Jin Kangorcid, Yoo Kyoung Parkorcid
Clinical Nutrition Research 2026;15(3):228-241.
Published online: July 31, 2026

AgeTech-Service Convergence Major, Department of Medical Nutrition, Graduate School of East-West Medical Science, Kyung Hee University, Yongin, Korea

Correspondence to: Yoo Kyoung Park Department of Medical Nutrition, Graduate School of East-West Medical Science, Kyung Hee University, 1732 Deogyeong-daero, Giheung-gu, Yongin 17104, Korea Email: ypark@khu.ac.kr
• Received: July 13, 2026   • Revised: July 21, 2026   • Accepted: July 21, 2026

© 2026 The Korean Society of Clinical Nutrition

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 24 Views
  • 2 Download
prev

Full Article

Download PDF Download PDF
  • Objective
    As global populations age, nutrition management is important for healthy aging. Digital technologies, including mobile applications, web-based platforms, messaging tools, and artificial intelligence (AI)-enabled systems, are used in personalized nutrition interventions. However, evidence on their characteristics, effectiveness, and user experiences remains limited. This scoping review examined technology-based personalized nutrition interventions for older adults.
  • Methods
    This review followed the Joanna Briggs Institute methodology and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews guidelines. PubMed, Embase, and Web of Science were searched for studies published from January 1, 2010, to June 30, 2026. Among 2,727 records, 2,044 were screened after duplicate removal, 28 full-text articles were assessed, and 13 articles representing 12 studies were included.
  • Results
    Technologies included mobile, messaging, web-based, information and communications technology, tablet-based, and AI-supported systems. Interventions involved dietary recording, personalized feedback, remote counseling, coaching, and self-monitoring. Outcomes included dietary intake and quality, nutritional status, physical function, frailty, cognition, cardiometabolic indicators, engagement, usability, and acceptability. Findings were more consistent for dietary behaviors and individualized nutrition management. Evidence for physical function, frailty, cognition, cardiometabolic health, and quality of life was limited and often based on small or multidomain studies. Familiar interfaces and professional support were accepted, whereas low digital literacy, complex navigation, and limited food databases were barriers.
  • Conclusion
    Digital personalized nutrition interventions may support dietary behavior change and individualized nutrition management in older adults. Larger studies and user-centered systems addressing digital literacy, usability, cultural context, and AI-feedback validation are needed.
  • Keywords: Aged; Digital health; Nutrition therapy; Scoping review; Artificial intelligence
Globally, population aging is accelerating due to longer life expectancy and declining fertility rates. The World Health Organization estimated that 1 billion of the global population aged ≥60 years in 2019 and that this number is expected to reach 1.4 and 2.1 billion by 2030 and 2050, respectively [1]. These demographic changes underscore the importance of healthy aging, which extends beyond longevity to maintaining physical, cognitive, and social functions in later life [2,3]. Nutrition plays a pivotal role in maintaining physical function, preserving cognitive function, and preventing and managing chronic diseases during aging. Inadequate nutritional status is closely associated with frailty, sarcopenia, cognitive decline, and reduced quality of life [4-7]. Therefore, intervention strategies to enhance nutritional status should be developed and assessed in geriatric healthcare [8,9].
Recent advances in information and communications technology (ICT) have rapidly expanded digital health interventions using mobile applications, web-based platforms, wearable devices, remote monitoring systems, and artificial intelligence (AI)-based tools [10-13]. These digital technologies facilitate automated dietary recording, continuous collection of health and dietary data, real-time feedback, remote counseling, and monitoring, addressing the limitations of traditional face-to-face interventions [14,15]. Particularly, personalization strategies that provide tailored information and feedback based on individual nutritional status, dietary intake, health risks, physical function, food preferences, and stages of behavioral change have become increasingly well known for their potential to improve engagement and adherence and sustain behavioral change [16,17].
However, older adults may be hesitant to accept and use digital technologies due to differences in previous experience with digital devices and digital literacy levels, as well as age-related declines in sensory and cognitive functions and physical limitations [18,19]. Therefore, technological efficiency, usability, accessibility, acceptability, and the potential for sustained engagement should be considered in designing digital technology-based nutrition interventions for older adults [20,21]. Furthermore, an integrated review should be conducted to assess the criteria used to develop personalization strategies, the digital technologies through which these strategies are delivered, and their effects on nutritional status, health outcomes, and user experiences [14].
Various countries and settings have implemented digital technology-based nutrition interventions for older adults. However, only a few studies have comprehensively synthesized the types of technologies employed, personalization criteria, intervention delivery modes, outcome measures, and user experiences [14,15]. Although emerging technologies, such as AI-based dietary recognition, algorithm-based recommendations, and wearable-linked feedback, have been introduced into nutrition interventions, a systematic review is needed to determine how these technologies are applied in digital nutrition interventions for older adults.
Accordingly, this study used a scoping review approach to explore digital technology-based personalized nutrition interventions for older adults and systematically summarize their characteristics and current applications. Specifically, this study aimed to (1) determine the types of digital technologies and delivery modes adopted in these interventions; (2) examine the criteria and methods used for personalization strategies; (3) synthesize the main outcome measures and findings, including nutritional status, physical function, cognitive function, cardiovascular health, usability, and acceptability; and (4) identify implications and research gaps for the future design of digital nutrition interventions for older adults.
Ethics statement
This scoping review used previously published studies and did not involve human participants or primary data collection. Hence, Institutional Review Board approval and informed consent were not required.
Study design
This scoping review systematically explores digital technology-based personalized nutrition interventions for older adults in accordance with the Joanna Briggs Institute methodology for scoping reviews [22]. Study selection and reporting followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) guidelines. The population, concept, and context (PCC) framework was used to develop the research question and eligibility criteria. The review process consisted of developing a search strategy, conducting the literature search, selecting studies, extracting data, mapping evidence, and synthesizing findings. No review protocol was registered. Due to the absence of a registered protocol, the PCC framework was used to predefine the research question and eligibility criteria, and the search, study selection, and data extraction procedures were established before screening and used consistently throughout the review.
Search strategy
The PubMed, Embase, and Web of Science databases were searched on June 30, 2026, covering articles published from January 1, 2010, through June 30, 2026. Search terms were developed according to the PCC framework and combined terms: older adults (“older adults,” “elder,” “elderly,” and “senior”), nutrition interventions (“nutrition intervention,” “dietary intervention,” “nutrition education,” “nutrition counseling,” “dietary advice,” “dietary modification,” “nutrition program,” “nutrition therapy,” and “nutritional intervention”), digital and AI technologies (“digital health,” “eHealth,” “mHealth,” “mobile health,” “mobile app,” “mobile application,” “telehealth,” “telemedicine,” “artificial intelligence,” “AI,” “machine learning,” “algorithm,” “chatbot,” “wearable,” and “remote”), and personalization (“personalized,” “tailored,” “individualized,” “customized,” “adaptive,” “targeted,” “feedback,” and “recommendation”). The search strategy was adapted to the search structure of each database. Title/abstract fields in PubMed, title/abstract/keyword fields in Embase, and topic fields in Web of Science were searched for original peer-reviewed journal articles published in English between 2010 and 2026. Theses, conference abstracts, commentaries, editorials, letters, news articles, and other nonoriginal publications were excluded.
Inclusion and exclusion criteria
The inclusion criteria were as follows: studies involving adults aged ≥60 years; studies evaluating nutrition or dietary improvement interventions delivered through digital technologies (e.g., mobile applications, web-based programs, remote counseling, AI, chatbots, text messaging, wearable devices, or ICT); studies with intervention content or feedback personalized according to individual nutritional and health statuses, dietary intake, food preferences, physical function, disease-related information, or behavioral data; original intervention studies (e.g., randomized and nonrandomized controlled trials [RCTs] as well as quasi-experimental, pilot, feasibility, mixed-methods, and observational intervention studies); and peer-reviewed journal articles published in English. The exclusion criteria were as follows: studies involving children, adolescents, younger adults, or mixed-age populations without separate outcomes for older adults; evaluating only exercise, cognitive, or disease management interventions without a nutrition intervention component; studies involving face-to-face interventions without digital technology; including interventions that provided only standardized general nutrition information or dietary recording without personalized feedback; usability, perception, technology development, or algorithm development studies; examining only the validity of nutrition assessment applications or tools without implementing an actual nutrition intervention; and protocol papers, theses, conference abstracts, commentaries, editorials, letters, news articles, animal studies, and cell studies.
Study selection process
After removing duplicates, the retrieved articles were screened using the predefined inclusion and exclusion criteria. First, titles and abstracts were screened based on the target population, intervention components, use of digital technologies, inclusion of personalization strategies, and study type. Subsequently, full-text articles were reviewed to identify final eligibility. During the full-text review, studies were reviewed for eligibility by considering the involvement of older adults, inclusion of a nutrition intervention, use of digital technologies, incorporation of personalized feedback or intervention tailoring, implementation of an actual intervention, and meeting the publication type and language criteria. The final study selection process was documented using a PRISMA flow diagram, with results presented in Fig. 1. Study screening and full-text review were performed based on the predefined criteria, and the research teams evaluated studies with unclear eligibility until a final decision was made. Articles reporting on the same participant cohort were retained but treated as a single study.
Data extraction and analysis
Data, including author and publication year, country, study design, participant characteristics and age, sample size, intervention duration, intervention setting, type of digital technology used, delivery mode, study objective, nutrition intervention components, personalization criteria and methods, AI use, main outcome measures, effectiveness outcomes, and user perception and acceptability-related findings, were extracted from included studies and descriptively summarized based on study characteristics, types of digital technologies, intervention delivery modes, and personalization strategies. The main outcomes were categorized and synthesized across the following domains: dietary intake and diet quality, nutritional status, physical function and frailty, cognitive function, cardiovascular and health-related indicators, behavioral change and engagement, usability, and acceptability. Qualitative findings associated with user experience, acceptability, usability, and feasibility were thematically summarized based on recurring patterns across the included studies.
Study selection process
The database search identified 2,727 records. After removing 683 duplicate records, 2,044 were screened based on their titles and abstracts using the predefined inclusion and exclusion criteria (Table 1). Eligible studies involved adults aged ≥60 years, included nutrition intervention components, incorporated digital or AI-enabled technologies, and used personalization strategies. During this process, 2,016 records that did not meet the inclusion criteria were excluded, and 28 articles were subsequently subjected to a full-text review. All full-text articles were retrieved, and none were excluded due to unavailability. After the full-text review, 15 articles were excluded: non-intervention studies (n=4), lack of personalization strategy (n=4), inclusion of populations other than adults aged ≥60 years or mixed-age populations without separately reported data for this age group (n=5), algorithm or simulation studies without implementation of an actual intervention (n=1), and insufficient information to assess eligibility (n=1). Finally, 13 articles representing 12 unique studies were included in the analysis. The study selection process is shown in Fig. 1.
Review scope according to the PCC framework
This scoping review was defined using the PCC framework (Table 2). The study population comprised older adults aged ≥60 years, including community-dwelling older adults, frail or prefrail individuals, those at risk of malnutrition, and those residing in nursing homes or long-term care facilities. The concept was defined as personalization strategies incorporated into digital or AI-enabled nutrition interventions, such as personalized dietary feedback, individualized nutrition counseling, meal planning, self-monitoring, and algorithm- or AI-based recommendations. The context included community, home-based, clinical, institutional, long-term care, remote, and telehealth-based settings.
General characteristics of the included studies
The 13 included articles, representing 12 unique studies, were conducted in various countries and regions, including the Netherlands, Thailand, the United States, the United Kingdom, Chile, France, Japan, Taiwan, Germany, and Singapore, as well as multinational studies conducted in Europe and Japan (Table 3). The study designs included RCTs, pilot RCTs, quasi-experimental studies, feasibility or pilot studies, cluster RCTs, iterative design and pilot studies, as well as usability and development evaluation studies. Most studies were designed as pilot or feasibility studies to examine intervention feasibility before conducting a full-scale effectiveness evaluation. Study participants were mainly community-dwelling older adults aged ≥60 or 65 years. Some studies targeted specific at-risk populations, such as older adults at risk of frailty, cognitive decline, or dementia, overweight older adults at risk of diabetes, and rehabilitation inpatients. Sample sizes ranged from 15 to 224 participants. Although many interventions were short-term (approximately 3 months), some implemented longer interventions up to 12 months. Mobile applications were frequently used to deliver interventions. Other technologies included messaging platforms, such as LINE, web-based platforms, tablet-based applications, wearable devices, and AI-based image recognition or recommendation systems. The intervention delivery modes ranged from fully automated approaches offering recommendations or feedback via digital technologies alone to those that combined digital tools with support from dietitians, coaches, researchers, or care staff.
Intervention characteristics and main findings
The intervention objectives and contents varied across the included studies (Table 4). Some studies evaluated the development and field applicability of personalized dietary recording, assessment, or feedback tools. Conversely, others aimed to enhance diet quality, manage nutritional status, prevent frailty, improve physical function, prevent cognitive decline or diabetes, or maintain quality of life in older adults. The intervention components included personalized dietary recommendations, dietary recording, nutrition counseling, dietary self-monitoring, education on the Mediterranean diet or healthy dietary patterns, protein intake counseling, multidomain lifestyle interventions, and AI-based dietary analysis and feedback. Several studies reported positive changes in diet quality, dietary adherence, protein intake, cardiovascular health indicators, physical function, frailty-related indicators, and cognitive function. Specifically, studies providing personalized feedback based on individual dietary intake or health status using digital tools reported positive findings in terms of dietary behavior change and sustained engagement. Users generally considered mobile applications, messaging platforms, web-based platforms, tablet-based e-coaches, and AI-based image recognition functions acceptable. However, some studies also reported limitations, such as small sample sizes, limited food databases, usability challenges, and the burden of digital technology use.
Mapping of the reported effectiveness outcomes
To enhance evidence mapping, effectiveness outcomes reported across the included studies were organized into broad outcome categories (Table 5) used for descriptive purposes and not mutually exclusive, as several studies reported outcomes across more than one area. Regarding dietary intake, diet quality, and dietary adherence, personalized nutrition counseling, mobile app-based feedback, AI-based dietary analysis, and web-based dietary management interventions improved diet quality or adherence to healthy dietary patterns. Reported dietary changes included improved adherence to the Mediterranean diet, increased legume intake, and improved meal balance scores. Interventions that included protein counseling reported increased protein intake. For nutritional status, risk of malnutrition, and anthropometric measures, digital assessment tools utilized in long-term care settings supported the identification of malnutrition risk and food preferences. Some studies reported reductions in waist circumference, improvements in cardiovascular health indicators, and changes in mid-upper arm or calf circumference. Regarding physical function, frailty, and activities of daily living, positive changes were observed in indicators such as the short physical performance battery (SPPB), gait speed, Timed Up and Go test, muscle strength, physical activity levels, frailty-related indicators, and activities of daily living. Cognitive outcomes were assessed mainly in digital multidomain interventions targeting older adults at risk of cognitive decline, although findings were inconsistent across studies. Cardiovascular and other health-related outcomes included cardiovascular health scores, body weight, blood pressure, and selected cardiovascular indicators. Several studies reported high participant retention rates, strong adherence to self-monitoring, progression to the maintenance stage of behavioral change, and sustained app use regarding behavioral change, engagement, and retention. User-related outcomes, such as usability, acceptability, and user perceptions, positively perceived simple and familiar interfaces, image-based dietary recording, AI-based automated recognition, familiar platforms such as LINE, and professional support.
User-related findings
Findings associated with user experience, acceptability, usability, and feasibility were grouped by recurring considerations in all included studies (Table 6). Older adults perceived digital interventions more positively when tools were simple, familiar, and directly associated with their daily dietary practices. Automated or simplified dietary recording methods, including image-based dietary recording and AI-based food recognition, reduced the burden of manual dietary recording and facilitate more efficient nutrition monitoring. Tailored feedback based on dietary intake, nutritional status, health risks, stages of behavioral change, food preferences, or activity data supported dietary behavioral change and intervention engagement. Human support was also commonly reported as an important component of digital interventions. Several studies reported that dietitians, counselors, coaches, or care staff interpreted information collected using digital tools and provided individualized guidance. Combining digital tool use and professional guidance may be particularly useful for older adults who experience difficulties using digital technologies. Several challenges were also identified, such as low digital literacy, limited familiarity with web-based platforms, complex navigation structures, and difficulties using digital devices. Continued participation was supported by goal setting, self-monitoring, regular contact, tailored tasks, and strategies based on stages of behavioral change. AI-based image recognition and algorithm-based recommendations can potentially support real-time personalized dietary guidance. AI-enabled approaches may facilitate dietary assessment and personalized feedback. However, current evidence is limited; therefore, the clinical validity of algorithm-based nutrition recommendations remains unclear. Furthermore, intervention designs that accounted for care environments, user preferences, and available resources enhanced the relevance and usability of digital nutrition interventions for older adults.
This scoping review analyzed 13 articles representing 12 unique studies on digital technology-based personalized nutrition interventions for older adults and synthesized evidence on the types of digital technologies used, personalization strategies, intervention effectiveness, and user perceptions across the included studies [23-35]. The included studies used various digital technologies: mobile applications, web-based platforms, tablet-based e-coaches, ICT platforms, messaging applications, wearable devices, and AI-based food recognition and recommendation systems. Moreover, interventions considerably varied across studies, comprising personalized dietary feedback to improve diet quality, nutritional status management, protein intake promotion, combined exercise and nutrition interventions for frailty prevention, programs for physical function improvement, multidomain interventions for cognitive decline prevention, cardiovascular health management, diabetes prevention programs, and strategies for maintaining quality of life [25,26,28,29,32-35]. These findings reveal that digital technology-based personalized nutrition interventions have evolved beyond the simple delivery of nutrition information or facilitation of dietary recording. Instead, they are increasingly used as integrated tools for behavioral change that continuously collect dietary and health-related data, offer individualized feedback, and support self-management in everyday life. This shift reflects that sustained dietary behavioral change in older adults requires ongoing monitoring and personalized support rather than one-time nutrition education [23,25,28,30,31,35].
Personalized interventions in this review were implemented in different ways across studies. Some studies offered tailored feedback based on dietary intake, food preferences, nutritional requirements, health status, physical function, frailty level, and stage of behavioral change [24-26,29,31,33,34]. Conversely, other studies conducted algorithm- or AI-based analyses to assess dietary intake and provide personalized interventions [23,28,35]. These findings reveal that personalized nutrition interventions are evolving from simple rule-based tailored messages to more comprehensive approaches that consider individual nutritional requirements, health conditions, functional status, dietary habits, living environments, and digital literacy. Such personalization may be particularly important for older adults, as their nutritional needs can vary depending on frailty, chronic diseases, appetite, and physical function. However, due to varied criteria and methods for tailoring across studies, future studies should clearly describe how interventions are tailored to individual participants and determine whether these personalized approaches improve dietary behavioral change and health outcomes. The findings present the broader transition from standardized nutrition education toward personalized nutrition approaches. Future interventions should include interindividual differences in nutritional status, chronic disease burden, functional capacity, dietary preferences, and behavioral readiness. As digital technologies evolve, personalized nutrition interventions may increasingly integrate precision nutrition concepts to provide more adaptive and individualized recommendations. Digital nutrition interventions for older adults should assess malnutrition risk, appetite, unintentional weight loss, muscle status, swallowing difficulties, and disease-specific dietary needs. Nutrition goals should be personalized due to varied priorities between older adults at risk of malnutrition or sarcopenia and those with obesity, diabetes, hypertension, or renal disease. Qualified nutrition professionals should review automated recommendations, particularly when identifying clinically significant nutrition risks. Another important finding of this review is that several studies not only relied on digital tools but also included support from dietitians, coaches, researchers, care staff, or healthcare professionals [24,28-31,33,34]. These professionals reviewed dietary records, health status, and behavioral data collected using digital tools and offered individualized counseling or tailored feedback [24,28-31,33,34]. These findings reveal that professional support and ongoing communication may be as critical as the technology itself in designing digital nutrition interventions for older adults. Differences in digital device experience and digital literacy level among older adults may make fully automated systems burdensome. Therefore, digital nutrition interventions for older adults should be designed to combine digital tools with professional support, rather than relying solely on technology.
Using familiar digital platforms was also identified as an important factor influencing the acceptability and use of interventions among older adults. Messaging platforms such as LINE may be useful as they are already familiar to many older adults and can promote intervention participation and ongoing communication with healthcare professionals [24,29,34]. In contrast, newly developed applications can offer various functions; however, processes such as installation, login, screen navigation, and notification management may be burdensome for older users [25,26,28]. Therefore, ease of use in daily life should be prioritized over the number of new functions when designing digital interventions for older adults.
AI-based technologies represent one of the most rapidly developing areas of personalized nutrition. However, most applications in the reviewed studies were largely limited to food image recognition or dietary assessment rather than AI-driven personalized decision support. Further advances will require more sophisticated algorithms that can incorporate dietary intake, nutritional status, chronic diseases, physical function, and behavioral data to generate adaptive nutrition recommendations [23,28,35]. Food photo records and AI-based food recognition can address the burden of written dietary records and allow faster dietary assessment and feedback [23,28]. However, AI-based dietary assessment remains limited in accurately recognizing complex mixed dishes, estimating portion sizes, and identifying hidden ingredients. Limitations in food database coverage may also reduce the accuracy of nutrient estimation. These limitations are especially relevant for older adults with malnutrition, sarcopenia, or chronic diseases. Therefore, algorithm-based nutrition recommendations should be clinically validated and reviewed by qualified nutrition professionals before routine use.
Most studies remain at the development or feasibility evaluation stage. Further research is needed to expand food database coverage, improve the accuracy of food and mixed-dish recognition, and clinically validate AI-generated assessments and feedback [23,28,35].
Regarding effectiveness, several studies have reported positive findings, despite the varied areas of improvement. Reported dietary outcomes included improved adherence to healthy dietary patterns, meal balance scores, diet quality, and protein intake [25,28,33]. Other studies showed positive changes in physical function, frailty-related indicators, cardiovascular health indicators, cognitive function, and quality of life [26,29,32,34,35]. Although multiple outcome domains improved, evidence was relatively stronger for dietary behaviors and diet quality. Conversely, evidence for cognitive function, cardiovascular health, and quality of life was more limited and often derived from multidomain interventions. Therefore, identifying independent contribution of nutritional components remains difficult. However, these effects should be interpreted as the outcome of tailored feedback, professional support, behavioral change strategies, social interaction, and continuous monitoring, rather than the effect of digital technology alone.
Many included studies incorporated nutrition interventions into broader multidomain interventions that also included physical activity, cognitive training, social activities, sleep management, or psychosocial support [26,27,29,32,34,35]. Several included studies combined personalized nutrition with physical activity, cognitive training, or behavioral support. Therefore, observed effects cannot be attributed to nutrition alone. Future studies should compare nutrition-only interventions with combined interventions and assess outcomes such as dietary intake and nutritional status to clarify specific effects of nutrition.
Usability and acceptability also provided important insights. Included studies reported that simple and familiar screen layouts, image-based dietary recording, automated feedback, and professional support may help improve participation and acceptability among older adults [23,24,26,28-31,34,35]. Conversely, some studies reported challenges associated with low digital literacy, limited familiarity with web-based platforms, complex screen designs, and difficulty using devices [26,27,31,35]. Therefore, digital nutrition interventions for older adults should prioritize simplicity and usability rather than additional features. Future intervention development should consider older adult-friendly design features, such as large font sizes, clear instructions, simple input procedures, intuitive screen layouts, repeated user training, and the support of caregivers or care staff.
Limitations
This review has several limitations. Although the review protocol was not prospectively registered, predefined eligibility, screening, and data extraction procedures were consistently used to enhance methodological transparency. In addition, several included studies were small-scale pilot, feasibility, or short-term intervention [23-26,28-31,34,35], and some had no control group or small sample sizes [23,26,31]. Furthermore, participants were often relatively healthy or familiar with digital devices, which limits finding generalizability among older adults who experience difficulty using digital technologies [26,28,30,31,35]. Therefore, positive findings in this review should be interpreted as preliminary evidence supporting the potential of digital personalized nutrition interventions. Future studies should include RCTs with sufficient sample sizes and long-term follow-up and assess applicability in real-world community and long-term care settings. Collectively, dietary intake, nutritional status, physical function, cognitive function, quality of life, usability, and cost-effectiveness should be evaluated. This scoping review identified digital personalized nutrition interventions for older adults using mobile applications, messaging platforms, web-based systems, and AI-supported tools. The evidence revealed potential benefits for dietary self-monitoring, diet quality, adherence, and individualized nutrition management. However, findings for physical function, frailty, cognition, cardiometabolic health, and quality of life were less consistent and were frequently based on small-scale or multidomain studies. Further rigorous studies and clinical validation are needed before AI-assisted assessment and recommendations can be routinely adopted in geriatric nutrition care.
Overall, this review reveals that digital personalized nutrition interventions are a promising strategy for promoting dietary behavioral change and individualized nutrition management among older adults. Future interventions should shift from technology-centered approaches to person-centered models that integrate personalized nutrition, age-friendly digital design, and continuous professional support to achieve sustainable benefits in older adults.
Conclusion
This scoping review investigated 13 articles representing 12 unique studies on digital technology-based personalized nutrition interventions for older adults and found that mobile applications, messaging platforms, web-based systems, and AI-supported tools have been used to support dietary behavioral change, nutritional status management, and improvements in physical and cognitive function. In particular, blended approaches integrating digital technologies with healthcare professional support, combined with platforms familiar to older adults, emerged as important strategies for improving intervention acceptability and sustaining engagement. However, current evidence is largely based on small-scale pilot or feasibility studies. Therefore, future studies should include sufficient sample sizes, rigorous study designs, and long-term follow-ups. Furthermore, user-friendly designs should reflect older adults’ digital literacy and usability needs, alongside validation of AI-based dietary assessment and feedback. These approaches may help develop sustainable digital nutrition management strategies to support healthy aging among older adults.

Author Contributions

Conceptualization: all authors. Methodology: all authors. Data curation: all authors. Investigation: all authors. Data extraction: SJ. Data synthesis and interpretation: all authors. Writing - original draft: SJ. Writing - review & editing: all authors. All authors read and approved the final manuscript.

Conflict of Interest

None.

Funding

None.

Data availability

Datasets are not publicly available but are available from the corresponding author upon reasonable request.

Fig. 1.
Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) flow diagram of study selection.
cnr-2026-0027f1.jpg
Table 1.
Inclusion and exclusion criteria
Table 1.
Category Inclusion criteria Exclusion criteria
Population Studies targeting older adults aged ≥60 years and reviewing community-dwelling, institutionalized, frail, prefrail, nutritionally at risk, or sarcopenia-risk older adults, as well as older adults with chronic conditions Studies targeting children, adolescents, younger adults, or mixed-age populations that did not separately identify data for older adults
Concept: nutrition intervention Studies focusing on nutrition education, counseling, dietary feedback, meal planning, dietary behavior change, nutrition monitoring, or malnutrition prevention/management Studies without nutrition or dietary intervention components
Concept: digital or AI-enabled component Studies using apps, web platforms, telehealth, messaging tools, ICT platforms, wearable devices, algorithms, AI, or image-based dietary assessment to implement or support the intervention Face-to-face-only interventions without digital, remote, ICT, or AI-enabled components
Concept: personalization strategy Studies designing the intervention based on dietary intake, nutritional status, health status, preferences, physical function, frailty, behavior, user profile, app data, wearable data, or AI-generated recommendations Studies providing only standardized general information without individualized feedback, tailoring, goal setting, or recommendations
Study type RCTs, nonrandomized trials, quasi-experimental studies, pilot/feasibility studies, mixed-methods studies, and observational intervention studies Reviews, editorials, commentaries, letters, conference abstracts, theses, animal studies, cell studies, and non-intervention studies
Publication type Peer-reviewed journal articles Gray literature and non-peer-reviewed documents
Language English Non-English articles
Publication year Articles published from January 1, 2010, through June 30, 2026 Articles published before January 1, 2010, or after the final search date

AI, artificial intelligence; ICT, information and communications technology; RCT, randomized controlled trial.

Table 2.
PCC framework for scoping review
Table 2.
PCC element Definition in this review
Population (P) Older adults aged ≥60 years, including community-dwelling, institutionalized, frail, prefrail, and nutritionally at risk older adults
Concept (C) Personalization strategies in digital or AI-enabled nutrition interventions, such as individualized dietary feedback, tailored nutrition counseling, meal planning, self-monitoring, and algorithm- or AI-supported recommendations
Context (C) Community, home-based, clinical, institutional, long-term care, and remote or telehealth-based settings

AI, artificial intelligence.

Table 3.
Characteristics of the included studies
Table 3.
Study Country Design Population Age (yr) No. of participants Duration Setting Digital technology Delivery mode
Connelly et al. [23] 2025 United Kingdom Co-design/facilitation test Care home older adults and staff Older adults; not specified 154 participants 2-wk test Care homes Mobile app; AI system AI-assisted assessment
Gaewkhiew et al. [24] 2025 Thailand Pilot RCT Older dental patients ≥60 y 30 3 mon Community LINE app Blended, dietitian-led
Su et al. [25 2023 United States Pilot RCT Frail older adults 66–77 y 15 3 mon Home-based Mobile app (Olitor) App-based feedback
Gutierrez et al.[26] 2025 Chile Iterative design + pilot Community-dwelling older adults Mean 71 y 60 3 mon Home Mobile app (NeoMayor) App-based self-monitoring
de Souto Barreto et al. [27] 2021 France Pilot RCT Older adults with subjective memory complaints ≥65 y 120 6 mon Community Web platform/tablet Web-based multidomain
Kurotani et al. [28] 2025 Japan Quasi-experimental Community-dwelling frail older adults ≥65 y 29 3 mon Community AI-assisted smartphone app AI feedback
Li et al. [29] 2025 Taiwan Quasi-experimental Community-dwelling frail older adults Mean 74 y 81 6 mon Community center/home LINE app Hybrid coaching
Gerber et al. [30] 2024 United States Pilot RCT Overweight older adults at risk of diabetes ≥60 y 20 Approximately ~6 months, 22 sessions Online Web/video system Remote coach-led
Happe et al. [31] 2023 Germany Pilot study Older rehabilitation inpatients ≥70 y 21 9 wk Rehabilitation/home Tablet app; ECA E-coach
Rainero et al. [32] 2021 Europe/Japan RCT Community-dwelling older adults at risk of frailty Mean 70 y 201 12 mon Home ICT platform, sensors, and app Algorithm-based package
van den Helder et al. [33] 2020 Netherlands Cluster RCT Community-dwelling older adults Mean 72 y 224 6 mon Community/home Tablet app (VITAMIN) Blended app + expert coaching
Pei-Shan Li et al. [34] 2025 Taiwan Quasi-experimental Community-dwelling frail/pre-frail older adults ≥65 y 81 6 mon Community care center/home LINE app Hybrid coaching
Chew et al. [35] 2025 Singapore Quasi-experimental Community-dwelling older adults at risk of cognitive decline ≥60 y 150 6 mon Senior center/app AI toolkit, mobile app, Kinect AI-supported multidomain toolkit

References 29 and 34 are companion reports from the same FANS study cohort and were counted as one unique study.

AI, artificial intelligence; ECA, embodied conversational agent; ICT, information and communication technology; RCT, randomized controlled trial.

Table 4.
Intervention characteristics and key findings of the included studies
Table 4.
Study Objectives Intervention description Outcome: efficacy Outcome: users’ perceptions
Connelly et al. [23] (2025) To design and evaluate a digital nutrition assessment tool for identifying food preferences and malnutrition risk among older adults in care homes. A mobile app and AI-supported system were used for meal photo recording, preference documentation, symptom monitoring, and malnutrition risk alerts using care-home menu data. The system demonstrated potential to alleviate documentation burden and support malnutrition risk identification through AI-assisted food image recognition. Older adults and care staff perceived the tool as useful, particularly because image-based input reduced manual recording burden. Limited food database coverage was noted.
Gaewkhiew et al. [24] (2025) To examine the impact of nutrition counseling delivered via an online application on dietary intake among older adults. The participants received individualized nutrition counseling through LINE based on 3-day dietary records, health conditions, and eating habits. Retention was 100%. Energy and protein intake exhibited improvement trends, and nutritional status assessed by MNA-SF was monitored. LINE was considered familiar and accessible for older adults, supporting the feasibility of remote expert-led nutrition counseling.
Su et al. [25] (2023) To evaluate a mobile intervention designed to support healthy eating among frail older adults. The Olitor mobile app supported Mediterranean diet recording and provided visual feedback by comparing individual MedD scores with recommended levels. Adherence to the Mediterranean diet significantly improved in the intervention group. Physical function indicators, such as SPPB, were also assessed. The small pilot study suggested that visual feedback was acceptable and could motivate dietary behavior change among frail older adults.
Gutierrez et al. [26] (2025) To develop and pilot-test the NeoMayor mHealth app for promoting healthy lifestyles among older adults in Chile. The app provided lifestyle recommendations based on Life’s Essential 8, including nutrition-related feedback, self-monitoring, and personalized images/messages based on baseline health scores. Cardiovascular health scores significantly improved, as well as diet quality indicators, including the MIND score, following the intervention. The participants generally accepted the app-based self-monitoring approach, although the study sample was predominantly female and relied on self-reported data.
de Souto Barreto et al. [27] (2021) To evaluate the feasibility of a web-based multidomain lifestyle intervention for older adults with subjective memory complaints. A web platform or tablet delivered multidomain lifestyle content, including cognitive training, physical activity, and nutrition education, with regular online contact. The intervention demonstrated good feasibility and acceptability. Cognitive function and nutritional status indicators, including MNA, were evaluated. Regular online contact and structured web content were perceived as feasible, although imbalance in participant characteristics limited the interpretation.
Kurotani et al. [28] (2025) To determine whether an AI-assisted smartphone application could improve diet quality among community-dwelling frail older adults. The app enabled recording of 10 food groups and provided AI-assisted personalized feedback using image analysis and text-based recommendations. The intervention group demonstrated a significant improvement in the dietary balance score compared with the control group. The AI-assisted feedback was considered useful for real-time dietary self-management, although the study was small and nonrandomized.
Li et al. [29] (2025) To evaluate the effects of the Fitness and Nutrition Program for Seniors on frailty prevention and physical function among older adults. A hybrid intervention combined face-to-face sessions with LINE-based remote coaching, nutrition education, and monitoring of dietary behavior. The SPPB score and 4-m gait speed significantly improved. BMI and calf circumference were maintained during the intervention. The hybrid model was feasible during the COVID-19 pandemic and supported continued engagement through familiar mobile communication.
Gerber et al. [30] (2024) To evaluate the feasibility of an eHealth diabetes prevention program adapted for older adults at risk of diabetes. A web-based/video-delivered lifestyle intervention based on the Diabetes Prevention Program employed remote coaching, behavioral strategies, and individualized progress reports. Participation was high, and body weight exhibited a favorable decreasing trend. Adherence and session completion were also evaluated. The participants demonstrated high engagement with the remote program, supporting the feasibility of counselor-supported eHealth coaching for older adults.
Happe et al. [31] (2023) To evaluate the usability and feasibility of a tablet-based e-coach for older adults in rehabilitation. A tablet app with an embodied conversational agent supported nutrition and exercise diary, stage-based goal setting, and tailored guidance based on the transtheoretical model. The average SUS score was favorable, and adherence to nutrition-related recommendations improved during the intervention. The e-coach format was perceived as usable and supportive. However, the study had no control group and experienced attrition.
Rainero et al. [32] (2021) To determine whether the My-AHA ICT platform could prevent declines in quality of life among older adults at risk of frailty. An ICT platform integrating apps and sensors provided multidomain interventions, including nutrition, cognition, physical activity, and sleep management, matched to individual needs. The intervention helped prevent declines in quality of life compared with the control group. Nutritional status indicators, such as MNA, were also monitored. The platform demonstrated feasibility for integrating multidomain risk assessment and individualized intervention delivery in older adults.
van den Helder et al. [33] (2020) To evaluate the effects of blended home-based exercise and dietary protein intervention on physical function in community-dwelling older adults. A tablet-supported program combined resistance exercise with dietary protein counseling. App-recorded dietary data were utilized to facilitate expert nutrition coaching. Protein intake significantly increased in the exercise-plus-protein group. Physical performance outcomes, including m-PPT, were evaluated. The blended approach supported home-based coaching, although the high dropout rate and nonblinded design were noted as limitations.
Li et al. [34] (2025) To determine whether the FANS program improved frailty status and health-related quality of life among frail or pre-frail older adults. The program used community center sessions and LINE-based home care. Monthly individualized home and group tasks were adjusted according to behavioral stage and cognitive-behavioral strategies. The frailty scores and instrumental activities of daily living significantly improved compared with the control group. More than 80% of the participants reached the maintenance stage for dietary behavioral change. The hybrid intervention was considered practical for frail older adults and supported sustained behavioral change through stage-based personalization.
Chew et al. [35] (2025) To assess the ADL+ digital toolkit for multidomain cognitive, physical, and nutritional interventions among older adults at risk of cognitive decline. The AI-based mobile toolkit utilized image-based dietary records, AI nutrition analysis, Kinect-based activities, and cognitive prediction models to adjust intervention intensity. Global cognitive scores improved in the intervention group, particularly in processing speed. Activity levels were maintained in the intervention group compared with the control group. Gamified and AI-supported features promoted engagement and social interaction; however, differences in digital literacy may have influenced cognitive burden.

References 29 and 34 are companion reports derived from the same FANS study cohort and were counted as one unique study.

AI, artificial intelligence; MNA-SF, Mini Nutritional Assessment–Short Form; MedD, Mediterranean diet; SPPB, short physical performance battery; COVID-19, coronavirus disease 2019; SUS, System Usability Scale; ICT, information and communication technology; m-PPT, modified Physical Performance Test; FANS, fitness and nutrition program for seniors; ADL, activity of daily living.

Table 5.
Objectives and findings of the included studies
Table 5.
Outcome variable Study Results
Dietary intake, diet quality, and dietary adherence Gaewkhiew et al. [24] (2025) No between-group differences in nutrient intake; both groups showed decreased energy/nutrient intake over time (P<0.05)
Su et al. [25] (2023) Mediterranean diet adherence improved (adjusted P=0.04); legume intake increased (adjusted P<0.01)
Kurotani et al. [28] (2025) Dietary quality improved (+9.5 points; 95% CI, 2.3–16.7; P=0.01)
van den Helder et al. [33] (2020) Protein intake increased (+0.32 g/kg/day, P<0.001)
Nutritional status, malnutrition risk, and anthropometric outcomes Connelly et al. [23] (2025) Malnutrition risk and food preference identification supported. P-value not reported
Gaewkhiew et al. [24] (2025) No between-group difference in body measurements
Gutierrez et al. [26] (2025) Waist circumference decreased, and CVH index improved (P<0.001)
Li et al. [29] (2025) Mid-upper arm circumference increased (P<0.001); calf circumference improved (P<0.01)
Physical function, frailty, and daily functioning Li et al. [29] (2025) SPPB improved (P<0.05); Five Times Sit-to-Stand test performance improved (P<0.05); TUG test performance improved (P<0.001); physical activity level increased (P<0.001)
van den Helder et al. [33] (2020) m-PPT, no difference (HBex, P=0.933; HBex-Pro, P=0.730); gait speed, improved (P=0.001); muscle strength, improved (P=0.001); muscle mass, improved (P=0.017)
Li et al. [34] (2025) Frailty and daily activities improved; exact P-value not reported in the abstract
Cognitive and brain-related outcomes Chew et al. [35] (2024) NTB composite score improved (between-group difference=0.17; 95% CI, 0.071–0.27; P=0.001)
Rainero et al. [32] (2021) Stroop C incongruent time improved (P=0.012); stroop C incongruent errors improved (P=0.015)
de Souto Barreto et al. [27] (2021) No significant effect for most cognitive outcomes
Cardiometabolic and health-related indicators Gutierrez et al. [26] (2025) CVH index improved from 64 to 68 (P<0.001)
Gerber et al. [30] (2024) Intervention group lost 9.5% body weight; control group gained 2.4%; 90% lost ≥5% body weight
Su et al. [25] (2023) HOMA-IR, no difference (adjusted P=0.85)
Behavioral change, engagement, and maintenance Gaewkhiew et al. [24] (2025) Retention, 100%
Su et al. [25] (2023) Retention, 100%; average app use, about 12 min/wk
Gerber et al. [30] (2024) Attendance and self-monitoring adherence, 100%; greater weight loss among eating behavioral change participants (−12.0% vs. −5.0%, P<0.005)
Happe et al. [31] (2023) Nutrition recommendation achievement increased from 24% to 66%; physical activity recommendation achievement increased from 33% to 71%. P-value not reported
Li et al. [29] (2025) Dietary behavioral action/maintenance stage reached by 81.4%
Usability, acceptability, and users’ perceptions Connelly et al. [23] (2025) AI-assisted food image recording reduced documentation burden. P-value not reported
Happe et al. [31] (2023) SUS, 78.6
Li et al. [29] (2025) Satisfaction, 9.25/10
Chew et al. [35] (2024) High adherence; average usability

CI, confidence interval; CVH, cardiovascular health; SPPB, Short Physical Performance Battery; TUG, Timed Up and Go; m-PPT, modified Physical Performance Test; HBex, home-based exercise; HBex-Pro, home-based exercise plus protein; HOMA-IR, homeostatic model assessment of insulin resistance; SUS, System Usability Scale.

Table 6.
Qualitative outcomes of the included studies
Table 6.
Identified theme Description
Acceptability of digital nutrition interventions Several studies showed that digital nutrition interventions were generally acceptable to older adults when the tools were simple, familiar, and directly related to their daily dietary practices. Mobile apps, web-based platforms, and messaging applications, such as LINE, were considered feasible delivery methods, particularly when supporting dietary self-monitoring, nutrition education, or remote counseling.
Reduced burden through automated or simplified dietary recording Image-based dietary recording, AI-assisted food recognition, and structured app-based food logs alleviated the burden of manual dietary recording. These approaches were specifically relevant for older adults and care staff as they reduced the effort required to document meals and supported more efficient nutrition monitoring.
Importance of personalization and feedback Personalized feedback based on dietary intake, nutritional status, health risk, behavioral stage, food preference, or activity data was considered useful for supporting behavioral changes. Studies using visual feedback, individualized goals, expert counseling, or algorithm-based recommendations reported that personalization may enhance engagement and adherence to nutrition-related behaviors.
Role of human support in digital interventions Although digital tools were widely used, several interventions relied on nutritionists, counselors, coaches, or care staff to interpret data and offer individualized guidance. Blended models combining digital platforms with professional support appeared particularly useful for older adults, who may require additional encouragement, clarification, or motivational support.
Digital literacy and usability challenges Some studies showed barriers associated with low digital literacy, unfamiliarity with digital platforms, complex navigation, or difficulties using digital devices. Older adults preferred clear instructions, simple interfaces, readable visual materials, and direct messages. These findings underscore the need for age-friendly design in digital nutrition interventions.
Engagement and behavior maintenance Interventions including goal setting, self-monitoring, regular contact, tailored tasks, or stage-based behavioral change strategies appeared to support continued engagement. Structured digital follow-up and ongoing professional contact may help older adults sustain nutrition-related behaviors.
Feasibility of AI-enabled approaches AI-enabled approaches, such as image recognition, algorithm-based recommendations, and generative AI feedback, potentially supported real-time and individualized dietary guidance. However, AI use remained limited across studies, and issues such as food database coverage, food recognition accuracy, small sample sizes, and validation of AI-generated recommendations should be considered.
Need for culturally and contextually appropriate design Included studies revealed that digital nutrition interventions should consider care settings, user preferences, digital literacy, and available resources. Familiar communication tools and practical support from healthcare professionals or care staff may improve the relevance and usability of interventions for older adults.

AI, artificial intelligence.

  • 1. World Health Organization. Ageing and health [Internet]. World Health Organization; 2025 [cited 2026 May 13]. Available from: https://www.who.int/news-room/fact-sheets/detail/ageing-and-health
  • 2. Beard JR, Officer A, de Carvalho IA, et al. The World report on ageing and health: a policy framework for healthy ageing. Lancet 2016;387:2145-54.
  • 3. World Health Organization. WHO's work on the UN decade of healthy ageing (2021–2030) [Internet]. World Health Organization; 2022 [cited 2026 May 13]. Available from: https://www.who.int/initiatives/decade-of-healthy-ageing
  • 4. Arici H, Tor YB, Altinkaynak M, et al. Personalized diet with or without physical exercise improves nutritional status, muscle strength, physical performance, and quality of life in malnourished older adults: a prospective randomized controlled study. Nutrients 2025;17:675.
  • 5. Seo AR, Kim MJ, Kim B, et al. Associations between frailty in older adults and malnutrition in rural areas: 2019 Updated Version of the Asian Working Group for Sarcopenia. Yonsei Med J 2021;62:249-54.
  • 6. Landi F, Sieber C, Fielding RA, Rolland Y, Guralnik J. Nutritional intervention in sarcopenia: report from the International Conference on Frailty and Sarcopenia Research Task Force. J Frailty Aging 2018;7:247-52.
  • 7. Chou HH, Lai TJ, Yen CH, Chang PS, Pan JC, Lin PT. Sarcopenic obesity tendency and nutritional status is related to the risk of sarcopenia, frailty, depression and quality of life in patients with dementia. Int J Environ Res Public Health 2022;19:2492.
  • 8. Bernabei R, Landi F, Calvani R, et al. Multicomponent intervention to prevent mobility disability in frail older adults: randomised controlled trial (SPRINTT project). BMJ 2022;377:e068788.
  • 9. Norman K, Hass U, Pirlich M. Malnutrition in older adults: recent advances and remaining challenges. Nutrients 2021;13:2764.
  • 10. Tabassum N, Biediger-Friedman L, Johnson C, Lane M, Marceaux S. Digital and technology-based nutrition interventions, including Medically Tailored Meals (MTMs) for older adults in the U.S.: a scoping review. Nutrients 2026;18:385.
  • 11. De Santis KK, Mergenthal L, Christianson L, Busskamp A, Vonstein C, Zeeb H. Digital technologies for health promotion and disease prevention in older people: scoping review. J Med Internet Res 2023;25:e43542.
  • 12. Buja A, Grotto G. Digital health interventions for improving nutritional behaviors in older adults: a scoping review. Front Nutr 2026;13:1743189.
  • 13. Olmedo-Aguirre JO, Reyes-Campos J, Alor-Hernandez G, Machorro-Cano I, Rodriguez-Mazahua L, Sanchez-Cervantes JL. Remote healthcare for elderly people using wearables: a review. Biosensors (Basel) 2022;12:73.
  • 14. Brown RC, Cook MA, Berton Weitnauer R, et al. Digital nutrition intervention in Older Americans Act programs impacts knowledge and desire to participate in virtual programming. Front Public Health 2026;14:1695528.
  • 15. Tsai YI, Beh J, Ganderton C, Pranata A. Digital interventions for healthy ageing and cognitive health in older adults: a systematic review of mixed method studies and meta-analysis. BMC Geriatr 2024;24:217.
  • 16. Xu Z, Xu X, Sun L, et al. Effectiveness of personalized meal recommendation in improving dietary behaviors of Chinese community-dwelling elders: study protocol for a cluster randomized controlled trial. Trials 2024;25:252.
  • 17. Mooney P, Lam LT, Todd AR, Partridge SR, Raeside R. The inclusion and impact of digital determinants of health in digital nutrition interventions for adolescents: a systematic review. Health Promot Int 2025;40:daaf154.
  • 18. Khan S, Webster S, Puxty J, Robertson M. Key challenges and barriers to digital literacy for older adults: scoping review. JMIR Aging 2026;9:e80647.
  • 19. Hepburn J, Williams L, McCann L. Barriers to and facilitators of digital health technology adoption among older adults with chronic diseases: updated systematic review. JMIR Aging 2025;8:e80000.
  • 20. Spinello G, Gobbi E, Paoli A, Moro T. Older adults' acceptability and perceived barriers to digital health tools integrating nutrition and physical activity: a focus group study. Front Digit Health 2026;8:1803847.
  • 21. Munoz Esquivel K, Gillespie J, Kelly D, et al. Factors influencing continued wearable device use in older adult populations: quantitative study. JMIR Aging 2023;6:e36807.
  • 22. Peters MD, Marnie C, Tricco AC, et al. Updated methodological guidance for the conduct of scoping reviews. JBI Evid Implement 2021;19:3-10.
  • 23. Connelly J, Swingler K, Rodriguez-Sanchez N, Whittaker AC. Identifying food preferences and malnutrition in older adults in care homes: co-design study of a digital nutrition assessment tool. JMIR Aging 2025;8:e64661.
  • 24. Gaewkhiew P, Chatyanan S, Supavaree S, Chaiueai A, Vimolratchataporn A, Sridonpai P. The impact of nutrition counselling via online application on nutrient intake among Thai older adults during COVID-19: pilot study of randomised controlled trial. BMC Public Health 2025;25:3672.
  • 25. Su Y, Wu KC, Chien SY, Naik A, Zaslavsky O. A mobile intervention designed specifically for older adults with frailty to support healthy eating: pilot randomized controlled trial. JMIR Form Res 2023;7:e50870.
  • 26. Gutierrez M, Cabello V, SanMartin CD, et al. An mHealth app (NeoMayor) to promote healthy lifestyles and brain health in older adults: design and validation study. JMIR Aging 2025;8:e71936.
  • 27. de Souto Barreto P, Pothier K, Soriano G, et al. A web-based multidomain lifestyle intervention for older adults: the eMIND randomized controlled trial. J Prev Alzheimers Dis 2021;8:142-50.
  • 28. Kurotani K, Tanabe H, Yanai K, Sakamoto K, Ohkawara K. An artificial intelligence-assisted smartphone application for improving dietary quality among frail older adults: a quasi-experimental study. Geriatrics (Basel) 2025;10:160.
  • 29. Li PS, Hsieh CJ, Miao NF, et al. Application of the Fitness and Nutrition Program for Seniors (FANS) to improve physical activity and kinanthropometric measures among community-dwelling older adults living with frailty: a quasi-experimental study. BMC Geriatr 2025;25:543.
  • 30. Gerber S, Silver RE, Das SK, et al. Development and feasibility of an ehealth diabetes prevention program adapted for older adults: results from a randomized control pilot study. Nutrients 2024;16:930.
  • 31. Happe L, Sgraja M, Hein A, Quinten V, Diekmann R. Usability and feasibility of a tablet-based e-coach for older adults in rehabilitation units to improve nutrition and physical activity: a prospective pilot study. BMC Geriatr 2023;23:578.
  • 32. Rainero I, Summers MJ, Monter M, et al. The My Active and Healthy Aging ICT platform prevents quality of life decline in older adults: a randomised controlled study. Age Ageing 2021;50:1261-7.
  • 33. van den Helder J, Mehra S, van Dronkelaar C, et al. Blended home-based exercise and dietary protein in community-dwelling older adults: a cluster randomized controlled trial. J Cachexia Sarcopenia Muscle 2020;11:1590-602.
  • 34. Li PS, Hsieh CJ, Miao NF, et al. Enhancing frailty status and health-related quality of life in community-dwelling frail older adults. Gerontology 2025;71:273-91.
  • 35. Chew J, Zeng Z, Tan TH, et al. ADL+: a digital toolkit for multidomain cognitive, physical, and nutritional interventions to prevent cognitive decline in community-dwelling older adults. Int J Environ Res Public Health 2024;22:42.

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

Include:

Personalized nutrition interventions using digital technologies for older adults: a scoping review
Clin Nutr Res. 2026;15(3):228-241.   Published online July 31, 2026
DOI: https://doi.org/10.7762/cnr.2026.0027
Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.
Format:
Include:
Personalized nutrition interventions using digital technologies for older adults: a scoping review
Clin Nutr Res. 2026;15(3):228-241.   Published online July 31, 2026
DOI: https://doi.org/10.7762/cnr.2026.0027
Close

Figure

  • 0
Personalized nutrition interventions using digital technologies for older adults: a scoping review
Image
Fig. 1. Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) flow diagram of study selection.

Fig. 1.

Personalized nutrition interventions using digital technologies for older adults: a scoping review
Category Inclusion criteria Exclusion criteria
Population Studies targeting older adults aged ≥60 years and reviewing community-dwelling, institutionalized, frail, prefrail, nutritionally at risk, or sarcopenia-risk older adults, as well as older adults with chronic conditions Studies targeting children, adolescents, younger adults, or mixed-age populations that did not separately identify data for older adults
Concept: nutrition intervention Studies focusing on nutrition education, counseling, dietary feedback, meal planning, dietary behavior change, nutrition monitoring, or malnutrition prevention/management Studies without nutrition or dietary intervention components
Concept: digital or AI-enabled component Studies using apps, web platforms, telehealth, messaging tools, ICT platforms, wearable devices, algorithms, AI, or image-based dietary assessment to implement or support the intervention Face-to-face-only interventions without digital, remote, ICT, or AI-enabled components
Concept: personalization strategy Studies designing the intervention based on dietary intake, nutritional status, health status, preferences, physical function, frailty, behavior, user profile, app data, wearable data, or AI-generated recommendations Studies providing only standardized general information without individualized feedback, tailoring, goal setting, or recommendations
Study type RCTs, nonrandomized trials, quasi-experimental studies, pilot/feasibility studies, mixed-methods studies, and observational intervention studies Reviews, editorials, commentaries, letters, conference abstracts, theses, animal studies, cell studies, and non-intervention studies
Publication type Peer-reviewed journal articles Gray literature and non-peer-reviewed documents
Language English Non-English articles
Publication year Articles published from January 1, 2010, through June 30, 2026 Articles published before January 1, 2010, or after the final search date
PCC element Definition in this review
Population (P) Older adults aged ≥60 years, including community-dwelling, institutionalized, frail, prefrail, and nutritionally at risk older adults
Concept (C) Personalization strategies in digital or AI-enabled nutrition interventions, such as individualized dietary feedback, tailored nutrition counseling, meal planning, self-monitoring, and algorithm- or AI-supported recommendations
Context (C) Community, home-based, clinical, institutional, long-term care, and remote or telehealth-based settings
Study Country Design Population Age (yr) No. of participants Duration Setting Digital technology Delivery mode
Connelly et al. [23] 2025 United Kingdom Co-design/facilitation test Care home older adults and staff Older adults; not specified 154 participants 2-wk test Care homes Mobile app; AI system AI-assisted assessment
Gaewkhiew et al. [24] 2025 Thailand Pilot RCT Older dental patients ≥60 y 30 3 mon Community LINE app Blended, dietitian-led
Su et al. [25 2023 United States Pilot RCT Frail older adults 66–77 y 15 3 mon Home-based Mobile app (Olitor) App-based feedback
Gutierrez et al.[26] 2025 Chile Iterative design + pilot Community-dwelling older adults Mean 71 y 60 3 mon Home Mobile app (NeoMayor) App-based self-monitoring
de Souto Barreto et al. [27] 2021 France Pilot RCT Older adults with subjective memory complaints ≥65 y 120 6 mon Community Web platform/tablet Web-based multidomain
Kurotani et al. [28] 2025 Japan Quasi-experimental Community-dwelling frail older adults ≥65 y 29 3 mon Community AI-assisted smartphone app AI feedback
Li et al. [29] 2025 Taiwan Quasi-experimental Community-dwelling frail older adults Mean 74 y 81 6 mon Community center/home LINE app Hybrid coaching
Gerber et al. [30] 2024 United States Pilot RCT Overweight older adults at risk of diabetes ≥60 y 20 Approximately ~6 months, 22 sessions Online Web/video system Remote coach-led
Happe et al. [31] 2023 Germany Pilot study Older rehabilitation inpatients ≥70 y 21 9 wk Rehabilitation/home Tablet app; ECA E-coach
Rainero et al. [32] 2021 Europe/Japan RCT Community-dwelling older adults at risk of frailty Mean 70 y 201 12 mon Home ICT platform, sensors, and app Algorithm-based package
van den Helder et al. [33] 2020 Netherlands Cluster RCT Community-dwelling older adults Mean 72 y 224 6 mon Community/home Tablet app (VITAMIN) Blended app + expert coaching
Pei-Shan Li et al. [34] 2025 Taiwan Quasi-experimental Community-dwelling frail/pre-frail older adults ≥65 y 81 6 mon Community care center/home LINE app Hybrid coaching
Chew et al. [35] 2025 Singapore Quasi-experimental Community-dwelling older adults at risk of cognitive decline ≥60 y 150 6 mon Senior center/app AI toolkit, mobile app, Kinect AI-supported multidomain toolkit
Study Objectives Intervention description Outcome: efficacy Outcome: users’ perceptions
Connelly et al. [23] (2025) To design and evaluate a digital nutrition assessment tool for identifying food preferences and malnutrition risk among older adults in care homes. A mobile app and AI-supported system were used for meal photo recording, preference documentation, symptom monitoring, and malnutrition risk alerts using care-home menu data. The system demonstrated potential to alleviate documentation burden and support malnutrition risk identification through AI-assisted food image recognition. Older adults and care staff perceived the tool as useful, particularly because image-based input reduced manual recording burden. Limited food database coverage was noted.
Gaewkhiew et al. [24] (2025) To examine the impact of nutrition counseling delivered via an online application on dietary intake among older adults. The participants received individualized nutrition counseling through LINE based on 3-day dietary records, health conditions, and eating habits. Retention was 100%. Energy and protein intake exhibited improvement trends, and nutritional status assessed by MNA-SF was monitored. LINE was considered familiar and accessible for older adults, supporting the feasibility of remote expert-led nutrition counseling.
Su et al. [25] (2023) To evaluate a mobile intervention designed to support healthy eating among frail older adults. The Olitor mobile app supported Mediterranean diet recording and provided visual feedback by comparing individual MedD scores with recommended levels. Adherence to the Mediterranean diet significantly improved in the intervention group. Physical function indicators, such as SPPB, were also assessed. The small pilot study suggested that visual feedback was acceptable and could motivate dietary behavior change among frail older adults.
Gutierrez et al. [26] (2025) To develop and pilot-test the NeoMayor mHealth app for promoting healthy lifestyles among older adults in Chile. The app provided lifestyle recommendations based on Life’s Essential 8, including nutrition-related feedback, self-monitoring, and personalized images/messages based on baseline health scores. Cardiovascular health scores significantly improved, as well as diet quality indicators, including the MIND score, following the intervention. The participants generally accepted the app-based self-monitoring approach, although the study sample was predominantly female and relied on self-reported data.
de Souto Barreto et al. [27] (2021) To evaluate the feasibility of a web-based multidomain lifestyle intervention for older adults with subjective memory complaints. A web platform or tablet delivered multidomain lifestyle content, including cognitive training, physical activity, and nutrition education, with regular online contact. The intervention demonstrated good feasibility and acceptability. Cognitive function and nutritional status indicators, including MNA, were evaluated. Regular online contact and structured web content were perceived as feasible, although imbalance in participant characteristics limited the interpretation.
Kurotani et al. [28] (2025) To determine whether an AI-assisted smartphone application could improve diet quality among community-dwelling frail older adults. The app enabled recording of 10 food groups and provided AI-assisted personalized feedback using image analysis and text-based recommendations. The intervention group demonstrated a significant improvement in the dietary balance score compared with the control group. The AI-assisted feedback was considered useful for real-time dietary self-management, although the study was small and nonrandomized.
Li et al. [29] (2025) To evaluate the effects of the Fitness and Nutrition Program for Seniors on frailty prevention and physical function among older adults. A hybrid intervention combined face-to-face sessions with LINE-based remote coaching, nutrition education, and monitoring of dietary behavior. The SPPB score and 4-m gait speed significantly improved. BMI and calf circumference were maintained during the intervention. The hybrid model was feasible during the COVID-19 pandemic and supported continued engagement through familiar mobile communication.
Gerber et al. [30] (2024) To evaluate the feasibility of an eHealth diabetes prevention program adapted for older adults at risk of diabetes. A web-based/video-delivered lifestyle intervention based on the Diabetes Prevention Program employed remote coaching, behavioral strategies, and individualized progress reports. Participation was high, and body weight exhibited a favorable decreasing trend. Adherence and session completion were also evaluated. The participants demonstrated high engagement with the remote program, supporting the feasibility of counselor-supported eHealth coaching for older adults.
Happe et al. [31] (2023) To evaluate the usability and feasibility of a tablet-based e-coach for older adults in rehabilitation. A tablet app with an embodied conversational agent supported nutrition and exercise diary, stage-based goal setting, and tailored guidance based on the transtheoretical model. The average SUS score was favorable, and adherence to nutrition-related recommendations improved during the intervention. The e-coach format was perceived as usable and supportive. However, the study had no control group and experienced attrition.
Rainero et al. [32] (2021) To determine whether the My-AHA ICT platform could prevent declines in quality of life among older adults at risk of frailty. An ICT platform integrating apps and sensors provided multidomain interventions, including nutrition, cognition, physical activity, and sleep management, matched to individual needs. The intervention helped prevent declines in quality of life compared with the control group. Nutritional status indicators, such as MNA, were also monitored. The platform demonstrated feasibility for integrating multidomain risk assessment and individualized intervention delivery in older adults.
van den Helder et al. [33] (2020) To evaluate the effects of blended home-based exercise and dietary protein intervention on physical function in community-dwelling older adults. A tablet-supported program combined resistance exercise with dietary protein counseling. App-recorded dietary data were utilized to facilitate expert nutrition coaching. Protein intake significantly increased in the exercise-plus-protein group. Physical performance outcomes, including m-PPT, were evaluated. The blended approach supported home-based coaching, although the high dropout rate and nonblinded design were noted as limitations.
Li et al. [34] (2025) To determine whether the FANS program improved frailty status and health-related quality of life among frail or pre-frail older adults. The program used community center sessions and LINE-based home care. Monthly individualized home and group tasks were adjusted according to behavioral stage and cognitive-behavioral strategies. The frailty scores and instrumental activities of daily living significantly improved compared with the control group. More than 80% of the participants reached the maintenance stage for dietary behavioral change. The hybrid intervention was considered practical for frail older adults and supported sustained behavioral change through stage-based personalization.
Chew et al. [35] (2025) To assess the ADL+ digital toolkit for multidomain cognitive, physical, and nutritional interventions among older adults at risk of cognitive decline. The AI-based mobile toolkit utilized image-based dietary records, AI nutrition analysis, Kinect-based activities, and cognitive prediction models to adjust intervention intensity. Global cognitive scores improved in the intervention group, particularly in processing speed. Activity levels were maintained in the intervention group compared with the control group. Gamified and AI-supported features promoted engagement and social interaction; however, differences in digital literacy may have influenced cognitive burden.
Outcome variable Study Results
Dietary intake, diet quality, and dietary adherence Gaewkhiew et al. [24] (2025) No between-group differences in nutrient intake; both groups showed decreased energy/nutrient intake over time (P<0.05)
Su et al. [25] (2023) Mediterranean diet adherence improved (adjusted P=0.04); legume intake increased (adjusted P<0.01)
Kurotani et al. [28] (2025) Dietary quality improved (+9.5 points; 95% CI, 2.3–16.7; P=0.01)
van den Helder et al. [33] (2020) Protein intake increased (+0.32 g/kg/day, P<0.001)
Nutritional status, malnutrition risk, and anthropometric outcomes Connelly et al. [23] (2025) Malnutrition risk and food preference identification supported. P-value not reported
Gaewkhiew et al. [24] (2025) No between-group difference in body measurements
Gutierrez et al. [26] (2025) Waist circumference decreased, and CVH index improved (P<0.001)
Li et al. [29] (2025) Mid-upper arm circumference increased (P<0.001); calf circumference improved (P<0.01)
Physical function, frailty, and daily functioning Li et al. [29] (2025) SPPB improved (P<0.05); Five Times Sit-to-Stand test performance improved (P<0.05); TUG test performance improved (P<0.001); physical activity level increased (P<0.001)
van den Helder et al. [33] (2020) m-PPT, no difference (HBex, P=0.933; HBex-Pro, P=0.730); gait speed, improved (P=0.001); muscle strength, improved (P=0.001); muscle mass, improved (P=0.017)
Li et al. [34] (2025) Frailty and daily activities improved; exact P-value not reported in the abstract
Cognitive and brain-related outcomes Chew et al. [35] (2024) NTB composite score improved (between-group difference=0.17; 95% CI, 0.071–0.27; P=0.001)
Rainero et al. [32] (2021) Stroop C incongruent time improved (P=0.012); stroop C incongruent errors improved (P=0.015)
de Souto Barreto et al. [27] (2021) No significant effect for most cognitive outcomes
Cardiometabolic and health-related indicators Gutierrez et al. [26] (2025) CVH index improved from 64 to 68 (P<0.001)
Gerber et al. [30] (2024) Intervention group lost 9.5% body weight; control group gained 2.4%; 90% lost ≥5% body weight
Su et al. [25] (2023) HOMA-IR, no difference (adjusted P=0.85)
Behavioral change, engagement, and maintenance Gaewkhiew et al. [24] (2025) Retention, 100%
Su et al. [25] (2023) Retention, 100%; average app use, about 12 min/wk
Gerber et al. [30] (2024) Attendance and self-monitoring adherence, 100%; greater weight loss among eating behavioral change participants (−12.0% vs. −5.0%, P<0.005)
Happe et al. [31] (2023) Nutrition recommendation achievement increased from 24% to 66%; physical activity recommendation achievement increased from 33% to 71%. P-value not reported
Li et al. [29] (2025) Dietary behavioral action/maintenance stage reached by 81.4%
Usability, acceptability, and users’ perceptions Connelly et al. [23] (2025) AI-assisted food image recording reduced documentation burden. P-value not reported
Happe et al. [31] (2023) SUS, 78.6
Li et al. [29] (2025) Satisfaction, 9.25/10
Chew et al. [35] (2024) High adherence; average usability
Identified theme Description
Acceptability of digital nutrition interventions Several studies showed that digital nutrition interventions were generally acceptable to older adults when the tools were simple, familiar, and directly related to their daily dietary practices. Mobile apps, web-based platforms, and messaging applications, such as LINE, were considered feasible delivery methods, particularly when supporting dietary self-monitoring, nutrition education, or remote counseling.
Reduced burden through automated or simplified dietary recording Image-based dietary recording, AI-assisted food recognition, and structured app-based food logs alleviated the burden of manual dietary recording. These approaches were specifically relevant for older adults and care staff as they reduced the effort required to document meals and supported more efficient nutrition monitoring.
Importance of personalization and feedback Personalized feedback based on dietary intake, nutritional status, health risk, behavioral stage, food preference, or activity data was considered useful for supporting behavioral changes. Studies using visual feedback, individualized goals, expert counseling, or algorithm-based recommendations reported that personalization may enhance engagement and adherence to nutrition-related behaviors.
Role of human support in digital interventions Although digital tools were widely used, several interventions relied on nutritionists, counselors, coaches, or care staff to interpret data and offer individualized guidance. Blended models combining digital platforms with professional support appeared particularly useful for older adults, who may require additional encouragement, clarification, or motivational support.
Digital literacy and usability challenges Some studies showed barriers associated with low digital literacy, unfamiliarity with digital platforms, complex navigation, or difficulties using digital devices. Older adults preferred clear instructions, simple interfaces, readable visual materials, and direct messages. These findings underscore the need for age-friendly design in digital nutrition interventions.
Engagement and behavior maintenance Interventions including goal setting, self-monitoring, regular contact, tailored tasks, or stage-based behavioral change strategies appeared to support continued engagement. Structured digital follow-up and ongoing professional contact may help older adults sustain nutrition-related behaviors.
Feasibility of AI-enabled approaches AI-enabled approaches, such as image recognition, algorithm-based recommendations, and generative AI feedback, potentially supported real-time and individualized dietary guidance. However, AI use remained limited across studies, and issues such as food database coverage, food recognition accuracy, small sample sizes, and validation of AI-generated recommendations should be considered.
Need for culturally and contextually appropriate design Included studies revealed that digital nutrition interventions should consider care settings, user preferences, digital literacy, and available resources. Familiar communication tools and practical support from healthcare professionals or care staff may improve the relevance and usability of interventions for older adults.
Table 1. Inclusion and exclusion criteria

AI, artificial intelligence; ICT, information and communications technology; RCT, randomized controlled trial.

Table 2. PCC framework for scoping review

AI, artificial intelligence.

Table 3. Characteristics of the included studies

References 29 and 34 are companion reports from the same FANS study cohort and were counted as one unique study.

AI, artificial intelligence; ECA, embodied conversational agent; ICT, information and communication technology; RCT, randomized controlled trial.

Table 4. Intervention characteristics and key findings of the included studies

References 29 and 34 are companion reports derived from the same FANS study cohort and were counted as one unique study.

AI, artificial intelligence; MNA-SF, Mini Nutritional Assessment–Short Form; MedD, Mediterranean diet; SPPB, short physical performance battery; COVID-19, coronavirus disease 2019; SUS, System Usability Scale; ICT, information and communication technology; m-PPT, modified Physical Performance Test; FANS, fitness and nutrition program for seniors; ADL, activity of daily living.

Table 5. Objectives and findings of the included studies

CI, confidence interval; CVH, cardiovascular health; SPPB, Short Physical Performance Battery; TUG, Timed Up and Go; m-PPT, modified Physical Performance Test; HBex, home-based exercise; HBex-Pro, home-based exercise plus protein; HOMA-IR, homeostatic model assessment of insulin resistance; SUS, System Usability Scale.

Table 6. Qualitative outcomes of the included studies

AI, artificial intelligence.

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

CNR : Clinical Nutrition Research

ABOUT

BROWSE ARTICLES

EDITORIAL POLICIES

FOR CONTRIBUTORS

© Korean Society of Clinical Nutrition.