Enhancing Psychological Health and Weight-Related Behaviors in Older Adults Through Digital Interventions: Findings from the Get FIT Randomized Pilot Trial

Abstract

Objectives:

To evaluate the effects of two digital health interventions on anxiety symptoms, depressive symptoms, and health-related quality of life (HRQOL) in older adults at cardiovascular risk.

Methods:

In this randomized pilot trial, older adults (n = 54; mean age 65.6 ± 5.8 years; 60% women; 66% Hispanic) with intermediate or high cardiovascular risk were assigned to Get FIT (n = 24) or Get FIT+ (n = 30). Get FIT included one counseling session, an activity tracker, and a nutrition app. Get FIT+ included these components, plus weekly personalized motivational text messages. Outcomes were assessed at baseline, three months, and six months.

Results:

At 3 months, both groups showed small reductions in anxiety and slight improvements in HRQOL, with no significant differences between the groups. Depressive symptoms remained stable. By six months, anxiety trajectories differed: reductions were maintained in Get FIT+ but returned to baseline in Get FIT (p < .001). Both groups increased physical activity and reduced caloric intake. Weight loss occurred in both groups but was greater in Get FIT+ (all p < .001).

Conclusions:

The improved intervention in this initial study was associated with sustained reductions in anxiety and with greater weight loss.

Clinical Implications:

Customized digital programs support older adults at risk of heart issues by encouraging healthier habits and enhancing mental health.

Introduction

Healthcare systems face both possibilities and challenges due to the aging population. Older individuals are more susceptible to chronic diseases, disabilities, cognitive deterioration, and functional limitations; therefore, emphasis should be placed on prevention (Lunenfeld & Stratton, 2013). The U.S. Centers for Disease Control and Prevention reports that 24.3% of individuals aged 65 and older living outside institutions rate their health as fair or poor. Over 38% of males and 39.6% of women are classified as obese. Healthcare systems must adapt to meet the increasing demands of older adults, projected to constitute 23% of the population by 2054 (Pfitzer et al., 2025).

Rapid, scalable, and long-term solutions are necessary to fulfill the needs of older adults, who are more prone to chronic diseases such as diabetes, cardiovascular disease, hypertension, and obesity. To prevent the onset of chronic diseases and facilitate effective treatments, digital health applications offer quick access to information, personalized health management, and continuous monitoring (Ahmed et al., 2025). User-friendly applications help older adults track their medication schedules, physical activity, and dietary practices, thereby promoting autonomy, independence, and informed decision-making (Wang et al., 2025). These capabilities may also support mental health by increasing perceived control over personal health behaviors (Gan et al., 2021; Racey et al., 2023).

However, despite the considerable potential of digital health interventions for the prevention of chronic diseases, the evidence supporting their efficacy across diverse populations remains limited. Ensuring equitable access to digital health innovations requires attention to structural inequities that influence engagement and sustained use (Racey et al., 2023). Involvement is frequently examined; but the impact on long-term adherence and health-related quality of life (HRQOL) remains insufficiently understood, resulting in mixed findings (Golbus et al., 2023; Pfaeffli Dale et al., 2015).

These disparities are consistent with frameworks of the Social Determinants of Health (SDOH), which emphasize that structural conditions, including economic stability, education, access to healthcare, neighborhood resources, and digital accessibility, shape health behaviors and outcomes (Eaton et al., 2024). Individuals with limited financial resources, caregiving responsibilities, or reduced digital familiarity may face barriers to sustained engagement with digital interventions, even when such programs are available. Integrating SDOH perspectives into digital health research underscores the need for adaptable, culturally responsive interventions that reduce disparities in access and participation.

An analysis of the literature revealed that only 13% of digital therapies for cardiovascular health were fully automated, functioning independently of human assistance (Wongvibulsin et al., 2021). While technology-enhanced programs can increase reach, interventions requiring intensive human support may be difficult to scale. More effective methods are needed to evaluate digital treatments across different health areas and determine their potential for long-term health benefits (Aguiar et al., 2022). Continued investigation within this domain holds the potential to enhance the efficacy of interventions, refine engagement methodologies, and promote broader implementation (Racey et al., 2023). A comprehensive understanding of sustained adherence, coupled with the integration of personalized strategies and digital coaching, is crucial for optimizing the outcomes of initiatives to prevent and control chronic illnesses. Consequently, these strategies will contribute to achieving health equity among older adults (Eaton et al., 2024).

The study’s objectives were to investigate: (1) changes in anxiety symptoms, depressive symptoms, and HRQOL among participants receiving Get FIT and Get FIT+ at baseline, three months, and six months; and (2) associations among sociodemographic factors (age, gender, race/ethnicity, marital status, education), engagement in healthy behaviors, mental health symptoms, and HRQOL. This study also aimed to examined whether the enhanced Get FIT+ program provides additional benefits beyond standard digital support. Understanding how sociodemographic factors and ongoing engagement affect outcomes may help improve intervention targeting and strengthen digital strategies to boost mental health and quality of life in aging populations.

Methods

Study design

This double-arm randomized pilot trial examined the intermediate (3-month) and longer-term (6-month) effects of two digital health interventions—Get FIT and Get FIT+—on mental health symptoms and HRQOL among older adults at elevated cardiovascular risk. The detailed trial methods has been published elsewhere (Cacciata et al., 2025) and are briefly outlined below. The trial was prospectively registered at ClinicalTrials.gov (Identifier: NCT03720327) and approved by the University of California, Irvine Institutional Review Board (HS#2016–2713). Data collection occurred from June 2019 to December 2022. Participants were randomly assigned to either the Get FIT group (n = 24) or the Get FIT+ group (n = 30) using a computer-generated allocation sequence. Participant flow through the trial is presented in a diagram (Figure 1) and reporting adhered to the CONSORT (Consolidated Standards of Reporting Trials) standards for pilot and feasibility randomized controlled trials (Hopewell et al., 2025).

Figure 1.

CONSORT flow diagram of participant recruitment, allocation, follow-up, and analysis.

Interventions

Those in the Get FIT group received a single, in-person counseling session. This session focused on healthy eating and physical activity. In addition, participants received a handbook, a wearable activity tracker, and access to the MyFitnessPal nutrition tracking app. These digital tools were provided during the first three months of the intervention.

Participants in the Get FIT+ group received all parts of the Get FIT program, along with personalized motivational text messages. These messages were designed to reinforce behavioral goals, encourage accountability, and support ongoing participation in healthy habits. The messaging continued throughout the six-month study, providing behavioral support that went beyond the initial use of the digital tool. The message bank and tailoring procedures were developed using a user-centered co-design approach (Ardo et al., 2021).

Participants

The sample consisted of community-dwelling adults aged 60 years and older who were classified as having intermediate or high cardiovascular risk based on Framingham Risk Scores (D’Agostino et al., 2008). We recruited participants through community outreach, primary care clinics, media advertisements, and partnerships with organizations that help older adults.

Eligibility requires fluency in English or Spanish and at least a sixth-grade reading proficiency. Individuals were excluded if they had a history of chronic substance abuse; end-stage renal, hepatic, or pulmonary disease; active cancer other than localized skin cancer; physical limitations preventing moderate-to-vigorous physical activity; gastrointestinal conditions requiring specialized diets; or cognitive impairment (assessed using the Mini-Cog screening tool) (Borson et al., 2005).

Data collection instruments

Mental health was assessed using the Hospital Anxiety and Depression Scale (HADS), a 14-item self-report instrument frequently utilized with older populations, with an anxiety and a depression subscale (Zigmond & Snaith, 1983). Each subscale score ranges from 0 to 21, where higher scores signify more severe symptoms. A score of 8 or above is regarded as clinically significant. The HADS has demonstrated reliability and validity in older adults residing in the community (Djukanovic et al., 2017).

The 12-Item Short Form Health Survey (SF-12) was used to evaluate health-related quality of life, producing both physical and mental composite scores (Ware et al., 1996). Higher scores indicate better self-reported health. The SF-12 has demonstrated strong psychometric properties across diverse groups of older adults (Sansom et al., 2020).

Wearable activity trackers that continuously monitor movement were used to measure moderate-to-vigorous physical activity (MVPA). The weekly average minutes of MVPA were calculated from the data collected by these devices, a standard method in digital health monitoring (Irwin & Gary, 2022; Li et al., 2025).

Participants used a nutrition tracking app to keep three-day food diaries, which were used to estimate their dietary intake. They were informed of the recording periods in advance. The average daily caloric intake was calculated from recorded data using established dietary self-monitoring methods (Byrne, 2015; Chang et al., 2020). During in-person study visits, which occurred at the start, three months, and six months, body weight was measured using calibrated digital scales.

Statistical analysis

Employing intention-to-treat principles (Gupta, 2011), all analyses used a two-tailed significance threshold of α = .05. Linear mixed-effects models were used to assess the intervention’s impact on study outcomes. This method is well-suited for analyzing data from clinical trials that involve repeated measurements (Wiley & Rapp, 2019).

The intervention groups (Get FIT and Get FIT+) and the time points (baseline, 3 months, and 6 months) were analyzed using fixed effects, including the interaction between the groups and time points. No additional covariates were included in the primary models, given the randomized design and pilot sample size. To account for the correlation between repeated measures within each individual, random intercepts were included. An appropriate covariance structure was specified (Yang, 2013).

The repeated-measures analysis integrated the baseline values rather than considering them as separate variables. Maximum likelihood estimation was used to handle missing data, assuming they were missing at random (Little, 2024). At three-months, there were no dropouts, and by six months, the dropout rate was 4%, with one participant from each group dropping out. An examination of the initial characteristics showed no significant differences between the participants who completed the study and those who did not.

Sensitivity analyses were performed by re-running the models using only complete-case data and by comparing results with and without participants with missing follow-up data. The data yielded consistent results with the primary analyses, supporting their reliability. The compound symmetry covariance structure fit the data well and was easy to use.

Pearson or Spearman correlation coefficients, selected based on the distributions of the variables (Hazra & Gogtay, 2016), were employed to investigate the relationships among behavioral engagement, weight change, and mental health symptoms. A post hoc power analysis demonstrated that the pilot sample size was sufficient to detect medium effect sizes in group-by-time interactions, aligning with the feasibility study’s objectives (Faul et al., 2009).

Results

Baseline sociodemographic characteristics were comparable between groups (Table 1). Women comprised 60% of participants. The sample was racially diverse, with 49% identifying as Asian and 28% as White; the remaining participants identified as multiracial. About 65% of the participants identified as Hispanic, and 60% were married. In addition, 52% of the group had a high school education or less.

Table 1.

Baseline sociodemographic characteristics of participants (N = 54).

Characteristic	All Participants (n = 54)	Get FIT (n = 24)	Get FIT+ (n = 30)	p-value
Age, years (Mean ± SD)	65.6 ± 5.8	66.5 ± 7.1	64.7 ± 4.7	.167
Female, n (%)	33 (60.0)	11 (45.8)	22 (73.3)	.052
Race, n (%)				.257
Black	1 (2.8)	0 (0.0)	1 (2.8)
White	21 (38.2)	10 (41.7)	11 (36.7)
Asian	27 (49.1)	10 (41.7)	17 (56.7)
Other	5 (9.1)	4 (16.7)	1 (3.3)
Hispanic ethnicity, n (%)	36 (65.5)	13 (54.2)	23 (76.7)	.081
Married, n (%)	33 (60.0)	13 (54.2)	20 (66.7)	.336
Education, n (%)				.526
≤High school graduate	28 (52.7)	11 (45.8)	17 (56.7)
College graduate	16 (29.1)	7 (29.2)	9 (30.0)
>College	10 (18.2)	6 (25.0)	4 (13.3)

Table 2 summarizes changes in health behaviors, weight, mental health symptoms, and HRQOL across study time points. At the three-month mark, both cohorts showed slight decreases in anxiety symptoms and minor improvements in HRQOL metrics; however, the differences between the groups did not reach statistical significance. Depressive symptom levels remained largely stable during this period.

Table 2.

Baseline, three-month, and six-month healthy behaviors, psychological, and HRQOL outcomes.

Variable	Get FIT Baseline Mean ± SD	3 Months	6 Months	Get FIT+ Baseline Mean ± SD	3 Months	6 Months	p (time)	p (T × G)
Calorie intake (kcal/day)*	1721.0 ± 411.9	1535.9 ± 257.8	1563.7 ± 345.9	1679.6 ± 313.7	1378.9 ± 212.3	1356.5 ± 245.7	<.001	.162
MVPA (min/week)**	70.9 ± 60.5	203.5 ± 103.2	191.2 ± 89.8	62.1 ± 43.8	147.5 ± 83.5	130.2 ± 72.7	<.001	.011
Weight (lbs.)	180.9 ± 40.4	179.4 ± 39.8	179.7 ± 40.7	160.9 ± 23.7	152.1 ± 26.2	151.1 ± 26.4	<.001	<.001
Anxiety	12.8 ± 2.1	12.4 ± 1.6	13.1 ± 2.2	13.4 ± 1.8	12.2 ± 2.1	12.3 ± 2.4	<.001	.012
Depression	9.6 ± 1.9	9.3 ± 2.2	9.5 ± 2.1	8.9 ± 2.3	8.4 ± 2.0	7.9 ± 1.9	.171	.094
Physical HRQOL (SF-12 PCS)	40.9 ± 7.4	41.9 ± 6.7	43.3 ± 6.8	40.2 ± 6.6	41.5 ± 7.1	42.8 ± 6.2	.048	.995
Mental HRQOL (SF-12 MCS)	54.3 ± 6.9	54.6 ± 5.4	55.1 ± 5.3	52.5 ±9.1	53.3 ± 8.8	54.2 ± 6.9	.498	.899

^†One participant in each group lost to follow-up at 6 months.

^*Three-day food diary average.

^**Moderate-to-vigorous physical activity.

HRQOL = health-related quality of life; PCS = Physical Component Score; MCS = Mental Component Score.

At six months, the patterns of anxiety symptoms showed a clear divergence between the two groups. Participants in the Get FIT+ group maintained modest reductions in anxiety scores. In contrast, anxiety levels in the Get FIT group returned to baseline, yielding a statistically significant group-by-time interaction (p < .001). Despite statistical significance, mean anxiety scores in both groups remained within the clinical symptom range.

Both groups increased MVPA minutes and reduced caloric intake over time. These behavioral changes were associated with reductions in body weight in both groups; however, weight loss was greater in the Get FIT+ group, reflected in significant group-by-time interactions (all p < .001).

Correlational analyses at six months (Table 3) indicated that intervention group assignment was significantly associated with greater weight loss, higher physical activity levels, lower caloric intake, and lower depressive symptom scores. Higher physical activity levels were modestly associated with lower depressive symptom levels. No statistically significant associations were observed between behavioral engagement and anxiety symptoms or HRQOL outcomes.

Table 3.

Correlation matrix of behavioral, mental health, and sociodemographic variables at 6 Months (N = 52).

Variable	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. Intervention group	1
2. Sex	.280	1
3. Age	−.157	−.280*	1
4. Marital status	.140	−.051	.181	1
5. Race	−.112	.057	−.284*	−.179	1
6. Ethnicity	−.237	−.161	.374†	.063	−.530†	1
7. Education	−.145	.000	.314*	.077	−.298*	.663†	1
8. Physical activity	.350†	−.018	.069	.435†	−.007	.025	.045	1
9. Weight (6 months)	−.396†	−.215	−.085	.039	−.173	.350†	.123	−.167	1
10. Calorie intake (6 months)	−.336*	−.275*	−.023	.212	.019	.151	.250	.033	.259	1
11. Anxiety (6 months)	−.176	.049	−.122	−.115	.143	.017	.010	−.026	−.019	.106	1
12. Depression (6 months)	−.371†	−.054	−.197	−.170	.369†	−.447†	−.342*	−.280*	−.003	−.072	.110	1
13. Physical HRQOL (6 months)	−.032	−.018	−.114	.171	.080	.127	.133	.176	−.017	.245	.083	−.138	1
14. Mental HRQOL (6 months)	−.072	.011	−.044	−.048	.152	.091	.116	.230	−.095	.063	.342*	−.181	.053	1

^†p < .01

^*p < .05.

Behavioral outcomes were also linked to sociodemographic factors. Specifically, those who were divorced, widowed, or separated reported higher levels of moderate-to-vigorous physical activity compared to those who were married or single. Non-Hispanic White participants with higher education levels reported higher depressive symptom scores, whereas Hispanic participants tended to have higher body weight overall.

Although anxiety and depressive symptoms showed small reductions in both groups during the early study period, their long-term patterns differed. The Get FIT+ group demonstrated modest but statistically significant reductions in anxiety symptoms, with small, lasting improvements. In contrast, depressive symptoms did not significantly differ between the groups over time and remained relatively stable. Variations in depressive symptoms seemed more closely related to levels of physical activity than to intervention assignment.

Discussion

The study evaluated the impacts of two digital health programs, Get FIT and Get FIT+, on mental health symptoms and HRQOL in older adults at higher cardiovascular risk. It also examined how sociodemographic factors, engagement levels, and these outcomes relate over time. Both intervention groups experienced slight reductions in anxiety symptoms and initial weight loss, indicating that even minimal digital support may promote short-term improvements in health behaviors and mental health measures. Participants in the Get FIT+ group maintained these improvements at the six-month follow-up, especially in weight and anxiety symptoms, and had larger reductions in anxiety scores compared to the Get FIT group.

While the decrease in anxiety scores reached statistical significance, the effect size was modest. The findings suggest that using communication strategies tailored to specific needs might help older adults keep up beneficial behaviors over time. Further research is needed better to understand the extent and duration of these observed effects.

Initially, the symptom patterns for anxiety and depression showed modest reductions in both groups. However, the patterns changed over the course of the study. Specifically, only the Get FIT+ group showed a statistically significant change in anxiety symptoms at six months. Throughout the study, symptoms of depression remained stable, and no significant differences were found between the groups. This finding indicates that factors influencing anxiety, possibly including ongoing motivational reinforcement, may differ from those affecting depression. The data showed a stronger correlation between mental health and the amount of exercise people did than with the specific type of intervention they received.

Participants who received personalized communication experienced consistent, although small, decreases in their anxiety symptoms (de Moel-Mandel et al., 2023; Xu et al., 2021). This pattern may reflect ongoing reinforcement, which encourages continued participation. Similar patterns have been observed in earlier digital programs designed to support heart health (Arnar et al., 2025; Yun et al., 2025). In these interventions, behavioral changes lasted only as long as support was provided. Subsequently, after assistance ended, the previously noted improvements generally declined. Similarly, mobile health initiatives for cardiac rehab showed positive effects on physical activity; however, these improvements did not always lead to lasting reductions in depressive symptoms or increased self-efficacy, highlighting the complex relationship between behavioral and psychological outcomes (Golbus et al., 2023; Molloy et al., 2023).

These findings highlight the importance of digital interventions designed with the user in mind and that are easily adjustable. The Get FIT+ program included personalized messaging, self-monitoring tools, and structured feedback – elements often linked to sustained engagement. Similar multi-component digital cardiovascular prevention tools that include coaching, tracking, and education have reported high user satisfaction and continued platform use, underscoring the value of interactive features over static content (Lockwood et al., 2024).

Our research adds to the ongoing discussion about passive digital monitoring. Previous studies indicate that while wearable devices and dietary tracking apps can help develop new health habits, keeping these habits going often requires structured and personalized support (Wongvibulsin et al., 2021). Evidence from groups with more complex health issues supports this idea. For example, remote monitoring programs for people with heart failure have shown they can be done, but they have not always improved treatment adherence or HRQOL (Rahimi et al., 2020). Similarly, participants in the Get FIT group who did not receive ongoing personalized support showed less ability to maintain behavioral changes over time. Therefore, these findings suggest that digital health interventions are more effective when they include personalized, ongoing support that considers individual needs and the specific situation.

Our findings showed that both groups increased MVPA over time; however, the changes in MVPA did not directly correspond to weight-loss patterns. This indicates that the intensity of physical activity and sustained engagement may have distinct effects on behavioral and psychological outcomes. Recent evidence indicates that higher volumes and intensities of MVPA are linked to better mental health outcomes in adults, including fewer depressive symptoms and improved mood and overall well-being (White et al., 2024). Objective measures indicate that substituting sedentary time with MVPA in older adults correlates with improved mental health outcomes, including reduced depressive symptoms and diminished feelings of loneliness. This highlights the psychological and social benefits of physical activity, which go beyond weight loss (Yao et al., 2025). Evidence indicates that various mechanisms, such as enhanced affect regulation, self-esteem, and the stress response, may mediate this relationship (White et al., 2024). Relationships between MVPA and weight outcomes are complex. Longitudinal data indicate that MVPA alone may not fully explain long-term weight changes, underscoring the need to consider broader behavioral and physiological factors when interpreting the effects of interventions on physical and mental health (Davies et al., 2025). These findings support our results, indicating that while increased MVPA positively impacts mental health and behavioral engagement, its effectiveness in achieving and maintaining weight loss may necessitate the incorporation of personalized support and contextual considerations.

Socioeconomic factors and broader societal context strongly influence people’s involvement in digital health initiatives (Xu et al., 2021). Sustained engagement often requires financial resources, time, digital proficiency, and a supportive environment. In contrast, people facing caregiving responsibilities, financial constraints, or limited technological skills may encounter obstacles that structured behavioral interventions alone cannot overcome. These findings align with SDOH frameworks, which suggest that health behaviors are shaped by wider structural determinants, including financial security, educational attainment, community settings, and access to digital resources (Javed et al., 2022). This perspective elucidates the persistent disparities in digital engagement, even when exposure to interventions is comparable, and highlights the need to develop culturally sensitive, readily available programs.

Even with digital resources, the Get FIT+ group remained more engaged. The customized, interactive support provided might have played a crucial role. The ongoing weight loss in the Get FIT+ cohort emphasizes the importance of continuous support for sustaining new behaviors. Meanwhile, the partial weight regain seen in the Get FIT group indicates that counseling alone, without ongoing support, might not be enough for long-term behavioral maintenance.

Psychological reinforcement might be a key factor in maintaining positive behaviors. Furthermore, those who exhibited enduring behavioral improvements also reported enhanced mental health. This observation implies a potential correlation between active engagement and psychological welfare. Subsequent investigations should examine the influence of social support systems, effective goal-setting methodologies, and incentive-based programs on older adults’ sustained participation in diverse activities.

Strengths, limitations, and directions for future research

This study has several strengths, including its randomized design, a diverse group of older adults – especially Hispanic participants who are often underrepresented in digital health research – and a longitudinal assessment of psychological and behavioral outcomes at 3 and 6 months. Nonetheless, significant limitations must be acknowledged. The limited sample size restricts statistical power and may hinder the generalizability of the findings. Additionally, incomplete data on device and application usage constrain the evaluation of behavioral engagement and dose – response relationships. The study overlooked important contextual factors, including health literacy, perceived social support, and access to technology, which can influence adherence and mental health outcomes. Since anxiety and depression showed different response patterns, future research should avoid treating them as interchangeable measures of psychological distress and instead focus on condition-specific mechanisms of change. Future research should integrate more accurate metrics of digital engagement, employ mixed-methods strategies to discern barriers and facilitators of participation, and evaluate adaptive or AI-driven interventions that offer personalized support in real time. To confirm these findings and develop digital health solutions that are widely used and fair for older adults, more comprehensive research across diverse healthcare settings is essential.

Conclusion

The results of this study indicate that digital health interventions can promote healthier behaviors and improve mental well-being among older adults at higher risk of cardiovascular disease. While both programs produced early improvements, the Get FIT+ intervention – which included personalized weekly text messages – provided more lasting benefits, especially in reducing anxiety and helping with weight maintenance. These findings highlight the importance of offering personalized feedback and consistent behavioral reinforcement, rather than relying solely on passive digital monitoring.

The results also show that sociodemographic factors like education and ethnicity greatly influence mental health outcomes, highlighting the need for culturally responsive and adaptable digital programs. Anxiety and depression responded differently to the interventions, highlighting the need for tailored digital strategies that address specific psychological processes instead of assuming the same effects across all mental health issues. This study emphasizes the growing role of digital health strategies in sustaining engagement, boosting mental resilience, and reducing cardiovascular risk among older adults. Conversely, the restricted participant pool necessitates cautious interpretation of these findings; consequently, further research involving larger cohorts is crucial to validate the durability and broader applicability of these observations. Future investigations should prioritize ongoing support, fair access, and personalized services to improve digital health interventions.

Clinical implications.

Digital health initiatives, when implemented effectively, present a viable approach for incorporation into both geriatric and primary care settings. These programs are specifically developed to promote healthier lifestyle choices and enhance the psychological health of older adults, a demographic particularly vulnerable to cardiovascular problems. Interventions that prioritize continuous engagement and personalized assistance demonstrate greater capacity to produce enduring benefits than passive self-monitoring strategies. In particular, delivering personalized feedback alongside regular motivational encouragement – such as personalized text messaging – can strengthen long-term compliance with health behavior changes and support the persistence of improvements in both anxiety levels and weight control. Furthermore, the variation in mental health outcomes among different sociodemographic groups underscores the need for culturally sensitive and flexible digital interventions. Consequently, clinicians and program developers should tailor program content and delivery methods to address specific psychological needs, rather than assuming that interventions will have the same effect on all older adults.

Data availability statement

The data that support the findings of this study are available from the corresponding author, dc, upon reasonable request.