Feasibility and behavioral impact of a wearable-supported digital health intervention on attitudes and behaviors toward sleep and physical activity: A proof-of-concept study in Kobe City

Abstract

Objectives

This proof-of-concept study examined whether wearable devices integrated with the Welloop mobile application and a light incentive feature lead to changes in awareness and behaviors related to sleep and physical activity. The intervention provided personalized recommendations based on biometric data and explored the potential of digital tools to support behavior change.

Methods

This single-arm interventional study was conducted from December 2023 to March 2024. Participants aged 20–65 years were recruited in Kobe, Japan. Eligible individuals owned a smartphone, registered as Healthcare Citizen Supporters, and agreed to wear a wearable device (Oura Ring or Fitbit) throughout the study period (January 10–March 31, 2024). Participation was voluntary, including mobile communication costs. Participation was voluntary. During the intervention phase, participants received personalized lifestyle recommendations based on biometric data via the mobile application, along with a light incentive feature to support engagement. The primary outcomes were changes in attitudes and self-reported and device-based behavioral indicators related to physical activity and sleep.

Results

This single-arm interventional study was conducted from December 2023 to March 2024. A total of 130 participants (median age 49; 83% women), with 60% holding a university degree or higher were enrolled. Participants aged 20–65 years were recruited in Kobe, Japan. Attitudinal outcomes showed significant or near-significant improvements, including perceived enjoyment and necessity of physical activity, and a significant shift in the stage of change for dietary habits. Objective and self-reported behavioral changes were modest, with small improvements in exercise duration, walking time, and reduced sleep procrastination, though most behavioral outcomes were not statistically significant. No meaningful differences were observed based on incentive preferences.

Conclusions

The findings suggest that the intervention is feasible in a community setting. While behavioral change was limited, attitudinal improvements highlight potential for digital self-monitoring and personalized feedback to support early stages of behavior change.

Keywords

mobile applications exercise physical activity health behavior wearable device

Introduction

Physical inactivity and insufficient sleep are major contributors to the global burden of chronic diseases such as cardiovascular disease, type 2 diabetes, and depression.^1,2

These two behaviors are not only independent risk factors but are also closely interconnected. Adequate physical activity has been shown to improve sleep quality and duration. In addition, a bidirectional relationship exists between exercise and sleep, such that individuals with poor sleep tend to be less physically active than those without sleep issues.^3,4 In this context, improving either behavior may have synergistic effects on the other.

In Japan, both behaviors are of particular concern. According to OECD (Organisation for Economic Co-operation and Development) data, Japan ranks the lowest among member countries in average sleep duration, with adults sleeping less than 7 hours per night on average.⁵ Additionally, data from the National Health and Nutrition Survey shows that only 36% of men and 29% of women aged 20 years and older report regular exercise habits, meaning that roughly 68% of adults fail to meet recommended physical activity levels.⁶ Moreover, the 2023 Physical Activity Guide reports that only 46% of adults (aged 20–64 years) achieve the recommended ≥60 minutes per day of moderate-to-vigorous activity, indicating that 54% remain below this target.⁷

These patterns reflect broader lifestyle challenges—such as long working hours, urban commuting, and limited time for recovery—and signal a critical need for public health interventions that support the formation of sustainable health habits. To our knowledge, few studies have examined the combined changes in both attitudes and behaviors in a general population context using wearable-device-based interventions.^8,9

Behavioral change is widely recognized as a central challenge in public health. While population-level determinants such as socioeconomic status and urban infrastructure are crucial, individual-level factors also play a significant role in the adoption and maintenance of healthy behaviors. Effective interventions often require a dual approach that targets both environmental and personal determinants. In this context, wearable devices have attracted growing interest as tools for addressing individual-level barriers—particularly by enabling self-monitoring and continuous feedback.^10,11 These features align with key constructs from established behavior change theories, including Self-Regulation Theory, Social Cognitive Theory (e.g., self-efficacy, observational learning), and the Health Belief Model (e.g., perceived benefits, cues to action). They are also compatible with the Transtheoretical Model, which outlines stages of change, and the COM-B model, which emphasizes the interplay of capability, opportunity, and motivation in shaping behavior.^12,13 However, despite their theoretical appeal, empirical findings on the effectiveness of wearable-based interventions remain inconsistent.^14–16 The variation in outcomes is often attributed to differences in user engagement, device usability, and the lack of personalized support.¹⁷

Digital health technologies, particularly wearables, hold potential to bridge the gap between intention and action.^18,19 By passively collecting behavioral data and delivering feedback in real-time, such tools can strengthen intrinsic motivation and facilitate the self-regulation necessary for long-term habit formation²⁰ These technologies may also promote reflective processes, helping individuals understand the links between their behavior and health outcomes²⁰ By facilitating timely adjustments and reinforcing behavioral intentions—consistent with constructs in the Theory of Planned Behavior—wearable-based feedback can serve as a bridge between short-term motivation and longer-term habit formation, as framed by Social Cognitive Theory and Self-Regulation Theory^21,22 Their potential utility spans multiple stages of change—from precontemplation to maintenance—as conceptualized in the Transtheoretical Model²³ particularly when integrated into supportive environments that also address capability and opportunity, as described in the COM-B model¹²

To address these gaps—particularly the limited empirical evidence on the relationship between attitudinal changes and actual behavioral changes, as well as the effectiveness of wearable-based interventions with personalized feedback in real-world settings— we conducted a proof-of-concept (PoC) study in Kobe City, Japan, using wearable devices linked to a mobile application that delivered personalized behavioral recommendations based on biometric data. This study targeted the general population through a municipal health initiative. By leveraging an existing civic health platform, the study examined the feasibility of deploying digital behavior change interventions at the community level. In addition to evaluating behavioral outcomes, this PoC study contributes to the growing evidence base on how municipalities can integrate digital tools into citizen-centered health promotion strategies. As local governments increasingly seek cost-effective and scalable interventions, understanding the practical implementation and acceptability of such technologies is vital for future program design and public policy development.

This study aimed to evaluate a mobile-based, wearable-supported intervention by addressing the following research questions: 1) Do participants’ health-related attitudes and behaviors (e.g., physical activity, sleep) change between baseline and follow-up?; 2) what are the characteristics of individuals who respond to the intervention compared to those who do not? By integrating self-reported survey data with biometric data collected from wearable devices, this study offers early insights into how digital self-monitoring technologies may support health behavior attitudes and behaviors in everyday community settings.

Methods

Study design

This study was a single-arm, pre–post trial conducted to assess whether a wearable-supported, app-based intervention could promote positive changes in sleep and physical activity behaviors, foster shifts in health-related attitudes, and differentiate participants based on behavioral responsiveness.

Participants and recruitment

Participants were recruited through Kobe City’s official website and related outreach. Eligible individuals were aged 20–65 years, resided in Kobe, owned a smartphone with internet access, and were registered as “Healthcare Citizen Supporters,” a local public health initiative.²⁴ A total of 130 individuals provided written informed consent (via digital platform) and were enrolled in the study. No formal a priori sample size calculation was performed because this was a proof-of-concept study. The target sample size was determined pragmatically based on recruitment feasibility, study duration, and available operational resources.

Inclusion and exclusion criteria

Participants were eligible if they were aged 20–65 years, owned a smartphone, registered as Healthcare Citizen Supporters, and agreed to wear a wearable device throughout the study period. Individuals who did not meet these criteria, were unable to provide written informed consent, or had incomplete participation were excluded.

Intervention

All participants were provided with a commercially available wearable device—either an Oura Ring or a Fitbit—and instructed to wear the device continuously throughout the 3-month study period. During the first month (January, 2024), data were collected without feedback. During the following two months (February–March, 2024), participants received weekly personalized lifestyle recommendations via the Welloop mobile application, informed by their biometric and behavioral data, along with incentive systems (Wellcoins). The application included an incentive feature (Wellcoins), where participants earned points for wearing the device and completing recommended behaviors derived from baseline patterns (e.g., consistent sleep timing or achieving ≥90 METs of activity, etc). Because of the small sample size, incentive effects were not analyzed, and this feature was treated as exploratory. The intervention emphasized self-monitoring, feedback, and personalized behavioral prompts, informed by established behavior change theories including Self-Regulation Theory, Social Cognitive Theory, and the COM-B model.

Data collection

Data were collected at two time points: baseline (January 2024) and post-intervention (March–April 2024), using two sources: a) Self-reported data were collected via online questionnaires and included demographic characteristics, physical activity and sleep habits, and attitudes toward health behaviors; b) wearable-derived biometric data were passively collected from the Oura Ring or Fitbit through the Welloop platform. Daily summaries were retrieved via API and averaged over four-week baseline and follow-up periods.

Data were collected using commercially available wearable devices (Oura Ring and Fitbit), each employing proprietary algorithms to estimate sleep and physical activity metrics. Although the specific algorithms are not publicly disclosed, both devices have been validated against reference measures such as polysomnography or research-grade accelerometers in prior studies. Previous research has shown that these devices provide reasonably accurate estimates of sleep and physical activity in free-living conditions.^25–29

Outcome measures

Outcome measures were categorized into three domains: biometric, self-reported behavioral, and attitudinal.

Biometric outcomes (objectively measured):

• Sleep efficiency (% of time in bed spent asleep)

• Total sleep duration (minutes per night)

• Activity burn (kilocalories per day from physical activity)

• Steps (average daily step count)

• Inactive time (minutes/day spent sedentary)

• Low, medium, and high activity time (minutes/day of light, moderate, and vigorous physical activity, respectively)

Days with physiologically implausible values (e.g., <100 steps or <2 hours of sleep) were excluded from analysis.

Self-reported behavioral outcomes:

• Sleep duration on weekdays and weekends (hours)

• Frequency of bedtime procrastination

• Exercise duration (categorized as none, <1 hour, ≥1 hour/day)

• Walking or standing time (categorized as <1 hour, 1–3 hours, ≥3 hours/day)

• Vigorous physical activity (binary: any vs. none)

• Estimated daily steps on weekdays and weekends (self-estimated)

Attitudinal outcomes:

• Perceived importance, enjoyment, ease, and necessity of physical activity

• Confidence in managing sleep (self-efficacy)

• Change in intention toward healthier eating habits (e.g., “intend to improve” to “currently working on it”)

• Self-reported changes in attitudes toward exercise and physical activity

All attitudinal items were rated using Likert-type scales. Some items were phrased as retrospective evaluations of change.

Statistical analysis

Descriptive statistics were used to characterize the sample. Within-subject changes in continuous variables were assessed using paired t-tests or Wilcoxon signed-rank tests, depending on distribution. Ordinal outcomes were analyzed using ordered logistic regression. Subgroup comparisons (e.g., responders vs. non-responders) were examined using Fisher’s exact tests and non-parametric methods. All analyses were conducted using R (version 4.3.1). A two-tailed p-value of <0.05 was considered statistically significant.

Ethical considerations

This study was approved by the ethics review committee of Kyoto University of the Arts in December 2023. All participants provided written informed consent (via digital platform) prior to enrollment, in accordance with the Declaration of Helsinki.

Results

Sample characteristics

Of the 130 individuals who were enrolled in the study, 70 participants completed both baseline and follow-up questionnaires and were included in the final analysis. Table 1 summarizes the demographic and behavioral characteristics of these participants at baseline. The average age was 46.8 years (SD = 11.6), and 70.0% were female. The majority were employed either full-time or part-time (77.1%), and 74.3% lived with family members. In terms of socioeconomic status, 40.0% of participants reported an annual household income below ¥5 million. The majority (58.6%) had attained a university degree or higher. At baseline, 15.7% of participants engaged in any form of vigorous physical activity, and 50.0% reported exercising less than one hour per day. Walking or standing for three hours or more per day was reported by 64.3% of participants. Bedtime procrastination—defined as delaying sleep three or more times per week—was reported by 20.0%. Wearable devices were nearly evenly distributed between Oura Ring (n = 33) and Fitbit (n = 37).

Table 1.

Baseline characteristics of participants (n = 70).

Characteristic	n (%) or Mean ± SD
Sex (Female)	49 (70.0%)
Age (years)	46.8 ± 11.6
Employment status
Employed (full-/part-time)	54 (77.1%)
Unemployed/homemaker	10 (14.3%)
Living arrangement
Living with family	52 (74.3%)
Living alone	16 (22.9%)
Annual household income
< ¥5 million	28 (40.0%)
≥ ¥5 million	35 (50.0%)
Educational attainment
University degree or higher	41 (58.6%)
Vigorous physical activity	11 (15.7%)
Exercise duration <1 hour/day	35 (50.0%)
Walking/standing ≥3 hours/day	45 (64.3%)
Bedtime procrastination ≥3–4 times/week	14 (20.0%)
Device type
Oura Ring	33
Fitbit	37

Quantitative findings

Between baseline and follow-up, several positive changes were observed across health-related attitudes and behaviors (Table 2). Participants showed non-significantly increased enjoyment of physical activity (OR = 2.57, 95% CI: 0.98–6.76, p = 0.056) and higher perceptions of its importance (OR = 1.83, 95% CI: 0.94–3.57, p = 0.07) and necessity (OR = 1.96, 95% CI: 0.88–4.44, p = 0.10). A significant improvement was observed in the stage of dietary behavior change (OR = 3.10, 95% CI: 1.39–12.0, p < 0.001), while confidence in sleep management decreased (OR = 0.43, 95% CI: 0.22–0.85, p = 0.01).

Table 2.

Changes in health-related attitudes, self-reported behaviors, and biometric data from baseline to follow-up.

Domain	Outcome	Estimate (95% CI)	p-value
Attitudes	Change in attitude toward exercise level	OR 1.36 (0.72–2.56)	0.34
	Change in attitude toward physical activity level	OR 1.83 (0.94–3.57)	0.07
	Liking physical activity	OR 2.57 (0.98–6.76)	0.056
	Perceived necessity of physical activity	OR 1.96 (0.88–4.44)	0.10
	Confidence in sleep management	OR 0.43 (0.22–0.85)	0.01
	Stage of intention to improve eating habits	OR 3.10 (3.99–12.0)	<0.001
Self-reported behavior	Sleep duration (weekdays)	+8 min (−18 to +34)	0.537
	Sleep duration (weekends)	−4 min (−32 to +24)	0.764
	Bedtime procrastination (frequency)	OR 0.56 (0.56–0.57)	<0.001
	Exercise duration (0–1 hr → ≥1 hr)	OR 8.24 (3.4–20.2)	<0.001
	Walking/standing ≥3 hrs/day	OR 3.19 (1.48–6.85)	<0.001
	Vigorous activity duration	+8 min (−9 to +25)	0.35
	Steps (self-estimated, weekdays)	OR 1.08 (0.62–1.91)	0.77
	Steps (self-estimated, weekends)	OR 1.16 (0.66–2.03)	0.60
Biometric data	Sleep efficiency (%)	−2.0 (−4.0 to −0.1)	0.04
	Total sleep (mins/night)	−0.89 (−7.3 to 5.6)	0.79
	Activity burn (kcal/day)	+8.9 (−12.9 to 30.8)	0.42
	Steps/day	+296 (−72 to 663)	0.12
	Inactive time (mins/day)	+6.9 (−8.0 to 21.4)	0.36
	High activity time (mins/day)	−0.48 (−1.67 to 0.71)	0.43
	Medium activity time (mins/day)	+2.26 (0.42 to 4.09)	0.02
	Low activity time (mins/day)	+3.57 (−4.79 to 12.0)	0.40

Self-reported behaviors also improved. The proportion of participants engaging in ≥ 1 hour of daily exercise increased markedly (OR = 8.24, 95% CI: 3.40–20.2, p < 0.001), as did those walking or standing ≥ 3 hours per day (OR = 3.19, 95% CI: 1.48–6.85, p < 0.001). Bedtime procrastination frequency decreased significantly (OR = 0.56, 95% CI: 0.56–0.57, p < 0.001). In contrast, self-reported weekday and weekend sleep durations showed no significant change (+8 min and −4 min, respectively), nor did estimated step counts.

Biometric data from wearable devices indicated a small but significant increase in moderate-intensity physical activity (+2.26 min/day; 95% CI: 0.42–4.09; p = 0.02). No significant changes were found in total steps (+296 steps/day; p = 0.12), activity burn (+8.9 kcal/day; p = 0.42), or high-/low-intensity activity time. Sleep metrics remained stable overall, with total sleep time unchanged (−0.89 min/night; p = 0.79) and a slight reduction in sleep efficiency (−2.0%; 95% CI: −4.0 to −0.1; p = 0.04).

Together, these findings indicate favorable attitudinal and self-reported behavioral shifts following the intervention, with more limited changes observed in objective activity and sleep outcomes.

Baseline characteristics of responders (n = 16) and non-responders (n = 54) are shown in Table 3. Responders were defined as participants who reported improved daily exercise duration after the intervention. No significant differences were observed between the two groups in age (median 51 vs 49 years, p = 0.45), gender (81% vs 83% female, p = 0.99), education (62% vs 59% with a university degree or higher, p = 0.87), financial strain (38% vs 28%, p = 0.23), or desk-based occupation (69% vs 52%, p = 0.29). However, responders were more likely to report lower household income (<¥5 million: 31% vs 17%, p = 0.02) and a higher prevalence of no physical labor or daily activity at baseline (81% vs 26%, p < 0.001), suggesting that those with lower baseline activity levels benefited more from the intervention.

Table 3.

Baseline characteristics of responders and non-responders.

Baseline characteristic	Non-responders (n = 54)	Responders (n = 16)	P-value
Age, median [IQR] (years)	49 [40–56]	51 [47–57]	0.45
Female (%)	45 (83%)	13 (81%)	0.99
University education or higher (%)	32 (59%)	7 (62%)	0.87
Household income < ¥5 million (%)	9 (17%)	5 (31%)	0.02
Financial strain in living conditions (%)	15 (28%)	7 (38%)	0.23
Desk-based work (%)	28 (52%)	11 (69%)	0.29
Weekday sleep duration, median [IQR] (mins)	390 [360–420]	360 [345–390]	0.08
Stage of change for diet improvement (%)	30 (56%)	11 (69%)	0.88
No physical labor or activity (0 min/day) (%)	14 (26%)	13 (81%)	<0.001
No vigorous activity (0 min/week) (%)	44 (82%)	12 (75%)	0.24
Attitude toward physical activity, median [IQR]	37 [33–41]	35 [30–37]	0.08

Several baseline tendencies showed non-significant differences: responders had shorter weekday sleep duration (median 360 vs 390 minutes, p = 0.08) and slightly lower attitudes toward physical activity (median 35 vs 37, p = 0.08). No group differences were observed in stage of change for dietary improvement (69% vs 56%, p = 0.88) or weekly vigorous exercise participation (75% vs 82%, p = 0.24).

Overall, these findings suggest that participants with lower socioeconomic status and lower baseline physical activity levels were more likely to respond favorably to the intervention, while demographic and attitudinal factors showed limited influence.

Discussion

Summary of key findings

In this interventional study, although changes in several health-related attitudes and general attitudes were not statistically significant, they shifted in a favorable direction. The stage of behavioral change related to dietary habits showed a significant improvement, whereas confidence in maintaining adequate sleep significantly declined. As for self-reported health-related behaviors, the proportion of participants engaging in at least one hour of physical activity per day and those walking more than three hours per day significantly increased. In contrast, the frequency of bedtime procrastination significantly decreased. There were no significant changes in self-reported sleep duration. Device-based data showed a significant increase in moderate physical activity as measured by the wearable device, although the magnitude of change may not be clinically meaningful. For sleep-related indicators, the biometric data revealed trends in the opposite direction to the self-reported questionnaire results.

Regarding the comparison between responders and non-responders, there were no significant differences in most demographic factors. However, responders had significantly lower household income and tended to be less engaged in physically demanding occupations. At baseline, although not statistically significant, responders also tended to report shorter sleep duration and lower attitude scores toward physical activity compared to non-responders.

The observed discrepancy between self-reported improvements and more modest changes in device-based metrics aligns with findings from previous studies which uses digital devices. Many wearable-based interventions have reported that while users often perceive themselves to be more active or health-conscious, these perceptions do not always translate into measurable changes in behavior.^14,30–32 This may be due in part to increased awareness or social desirability bias in self-reporting, as well as the complex, often incremental nature of behavior change. Notably, our study found that participants’ health-related attitudes improved despite the absence of any explicit incentives, suggesting that self-monitoring alone may be sufficient to trigger reflective processes and enhance motivation. This is consistent with research demonstrating that self-tracking can enhance cognitive engagement and foster goal formation, particularly during early phases of change. These attitudinal gains, however, did not consistently lead to observable improvements in objective behaviors such as physical activity or sleep. This reflects a broader critique in the digital health field: while passive data collection and feedback systems can effectively increase awareness, they are often insufficient to produce lasting behavioral outcomes without additional motivational or environmental support.^33,34 Interestingly, individuals with lower baseline activity levels exhibited greater responsiveness to the intervention. This suggests that individuals with lower baseline physical activity levels were more likely to benefit from the intervention. This may reflect a room for improvement effect, where feedback from wearables is more novel or salient to inactive users, prompting greater engagement and behavioral reflection.

The greater responsiveness observed among individuals with lower baseline activity and income suggests that individuals with lower baseline physical activity levels were more likely to complete the intervention. It may be explained by several mechanisms. First, participants with lower baseline physical activity levels may have had greater room for improvement, avoiding the ceiling effects observed in more active populations.^35,36 Second, prior research has suggested that individuals from lower socioeconomic backgrounds often face greater psychological and structural barriers to adopting healthy behaviors, including lower self-efficacy and limited access to social and informational resources.^37,38 In this context, the use of wearable devices may have acted as a compensatory mechanism—providing timely feedback, raising awareness, and enhancing perceived behavioral control. Studies have shown that objective self-monitoring tools can be particularly impactful for individuals who may otherwise lack access to structured health education or coaching.³⁹ These mechanisms may help explain why participants from lower activity and income backgrounds demonstrated greater behavioral shifts in this intervention.

Limitation and future directions

Several limitations should be mewledged when interpreting these findings. First, the sample size was relatively small and not randomized in this PoC study, which limits the precision of inference and the ability to infer causality. Second, the participants were recruited on a voluntary basis through a municipal health program, potentially introducing selection bias toward individuals already interested in health. Third, while the study incorporated both subjective and objective data sources, the duration of follow-up was limited, and the sustainability of observed changes over time merits further investigations. Fourth, there was variability in device adherence and data completeness; not all participants wore the device consistently or synchronized their data regularly, which may have affected the accuracy of outcome measurements. Furthermore, differences in the type of device used (e.g., Oura vs. Fitbit) may have introduced measurement heterogeneity. Lastly, while this study provided a PoC for municipal-level implementation, it did not include a control group or account for seasonal or external behavioral influences. Future research should consider rigorous designs such as randomized controlled trials, longer follow-up periods, and stratified analyses by device type or participant demographics. Integrating qualitative methods to explore user experiences and barriers to engagement would also enrich the understanding of how and why such interventions may—or may not—lead to sustained behavior change. Additionally, future programs may further enhance behavioral outcomes by incorporating strategies that address environmental opportunity, such as workplace policy changes or access to local exercise resources, consistent with COM-B model principles.

Conclusion

This PoC study supports the feasibility and potential utility of wearable-integrated, mobile-based health interventions at the community level. By demonstrating attitudinal and behavioral shifts without the use of financial incentives or intensive coaching, the findings underscore the power of self-monitoring and real-time feedback to activate internal motivation—particularly among individuals with lower baseline engagement in health behaviors. The municipal implementation model also highlights a scalable and inclusive approach to digital health promotion that can reach populations who may be underserved by traditional healthcare systems. Moreover, while this intervention primarily targeted individual-level factors, its integration into a public-sector platform provides a foundation for addressing broader environmental opportunities in future implementations, in line with the COM-B framework. As a PoC study, this project successfully tested both the technical and operational components of a digital intervention in a real-world setting, providing valuable insights for the design of future community-based programs. Continued refinement and expansion of such initiatives, informed by user-centered design and behavioral science, hold promise for building healthier and more proactive populations through everyday technologies.

Footnotes

Acknowledgement

We are grateful to the Foundation for Biomedical Research and Innovation at Kobe (FBRI)—Tadashi Matsubara, Kaori Ishida, Shuji Senda, and Yutaka Kobayashi—for their invaluable support. We also thank the City of Kobe’s Medical Industry City Promotion Division and the Kobe Research Complex Council Secretariat, particularly Shoko Asada; the Planning and Coordination Bureau (Coordination Division), City of Kobe—Kohei Miyoshi and Tsuyoshi Itoh; Food Bank Kansai (certified NPO), especially Maki Nakajima; and W Co., Ltd., including Daisuke Hirooka and Takafumi Nakagawa. Their contributions made this study possible. Furthermore, we would like to thank the Suntory Global Innovation Center Ltd. for providing the platform for this proof-of-concept study, as well as the wearable devices and mobile application. Their contributions made this study possible.

ORCID iDs

Hana Hayashi

Yukiko Takagi

Ethical considerations

This study was approved by the Ethics Committee of Kyoto Institute of Technology (Approval number:2022-68, 2023-91). Informed consent was obtained from all subjects involved in the study.

Author Contributions

HH made substantial contributions to the conception and design of the study, collected and interpreted the data, and drafted the manuscript. KH conducted the statistical analyses and contributed to writing the analysis section of the manuscript. HM led the development of the digital application used in the intervention, coordinated collaboration with the City of Kobe, and provided expert input on the study design and interpretation of findings. YT supported manuscript preparation, coordinated document revisions, and assisted with project-related tasks. All authors reviewed and approved the final version of the manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the City of Kobe through the FY2023 Be Smart KOBE Project (proposal title: “Health promotion of citizens and community revitalization using ‘health coins’”; proposing organization: Suntory Global Innovation Center, Ltd.). Down to Earth Inc. received consulting fees from Suntory Global Innovation Center, Ltd. for research design and data analysis related to this project. The funders had no role in the design of the study; in the collection, analysis, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Declaration of Conflicting Interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The authors declare that the study design, data collection, analysis, interpretation, and manuscript preparation were conducted independently of the funding sources. The content of this manuscript was prepared objectively and without influence from the sponsors. KH is employed by Sanofi at the time of publication.

Data Availability Statement

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.*

References

Lee

Shiroma

Lobelo

, et al. Eff ect of physical inactivity on major non-communicable diseases worldwide: an analysis of burden of disease and life expectancy. The Lancet (British edition) 2012; 380: 219-229.

Itani

Jike

Watanabe

, et al. Short sleep duration and health outcomes: a systematic review, meta-analysis, and meta-regression. Sleep Medicine 2017; 32: 246–256. https://doi.org/10.1016/j.sleep.2016.08.006

Kline

. The Bidirectional Relationship Between Exercise and Sleep. American journal of lifestyle medicine 2014; 8: 375–379. https://doi.org/10.1177/1559827614544437

Martin

Boland

Neufeld

, et al. Interrelationship between Sleep and Exercise: A Systematic Review. Advances in Preventive Medicine 2017; 2017: 1364387. https://doi.org/10.1155/2017/1364387

Anonymous OECD . Health Statistics, 2021.

Ministry of Health LaW . The National Health and Nutrition Survey in Japan, 2023.

Sasakawa Sports Foundation & Meiji Yasuda Life Foundation of Health and Welfare . Survey on the physical activity levels of Japanese people (accessed July 31 2025.

Rowland

Bach

Simon

, et al. Effectiveness of digital health interventions to increase cardiorespiratory fitness: A systematic review and meta-analysis. Digital health 2024; 10: 20552076241282381. https://doi.org/10.1177/20552076241282381

Brickwood

Watson

O'Brien

, et al. Consumer-Based Wearable Activity Trackers Increase Physical Activity Participation: Systematic Review and Meta-Analysis. JMIR mHealth and uHealth 2019; 7: e11819. https://doi.org/10.2196/11819

10.

Kang

Exworthy

. Wearing the Future—Wearables to Empower Users to Take Greater Responsibility for Their Health and Care: Scoping Review. JMIR mHealth and uHealth 2022; 10: e35684. https://doi.org/10.2196/35684

11.

Natalucci

Marmondi

Biraghi

, et al.

The Effectiveness of Wearable Devices in Non-Communicable Diseases to Manage Physical Activity and Nutrition: Where We Are?

Nutrients 2023; 15: 913. https://doi.org/10.3390/nu15040913

12.

Michie

van Stralen

West

. The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation science : IS 2011; 6: 42. https://doi.org/10.1186/1748-5908-6-42

13.

Glanz

Health behavior. 5 ed. Jossey-Bass, Wiley, 2015.

14.

Chen

. Wearable activity trackers for promoting physical activity: A systematic meta-analytic review. International journal of medical informatics (Shannon, Ireland) 2021; 152: 104487. https://doi.org/10.1016/j.ijmedinf.2021.104487

15.

Ferguson

Olds

Curtis

, et al. Effectiveness of wearable activity trackers to increase physical activity and improve health: a systematic review of systematic reviews and meta-analyses. The Lancet Digital Health 2022; 4: e615–e626. https://doi.org/10.1016/s2589-7500(22)00111-x

16.

Tang

MSS

Moore

McGavigan

, et al. Effectiveness of Wearable Trackers on Physical Activity in Healthy Adults: Systematic Review and Meta-Analysis of Randomized Controlled Trials. JMIR mHealth and uHealth 2020; 8: e15576. https://doi.org/10.2196/15576

17.

Berry

ECJ

Sculthorpe

Warner

, et al. A scoping review of the feasibility, usability, and efficacy of digital interventions in older adults concerning physical activity and/or exercise. Frontiers in aging 2025; 6: 1516481. https://doi.org/10.3389/fragi.2025.1516481

18.

Patel

Asch

Volpp

. Wearable Devices as Facilitators, Not Drivers, of Health Behavior Change. JAMA : the journal of the American Medical Association 2015; 313: 459–460. https://doi.org/10.1001/jama.2014.14781

19.

Xie

Hong

. Electronic wearable device and physical activity among US adults: An analysis of 2019 HINTS data. International journal of medical informatics (Shannon, Ireland) 2020; 144: 104297. https://doi.org/10.1016/j.ijmedinf.2020.104297

20.

Zhu

Long

Wang

, et al. Digital Behavior Change Intervention Designs for Habit Formation: Systematic Review. Journal of medical Internet research 2024; 26: e54375. https://doi.org/10.2196/54375

21.

Ajzen

. The theory of planned behavior. Organizational behavior and human decision processes 1991; 50: 179–211. https://doi.org/10.1016/0749-5978(91)90020-T

22.

Bandura

. Social cognitive theory of self-regulation. Organizational behavior and human decision processes 1991; 50: 248–287. https://doi.org/10.1016/0749-5978(91)90022-L

23.

Prochaska

Velicer

. The Transtheoretical Model of Health Behavior Change. American Journal of Health Promotion 1997; 12: 38–48. https://doi.org/10.4278/0890-1171-12.1.38

24.

Kobe city . Kobe Health Lab.

25.

Feehan

Geldman

Sayre

, et al. Accuracy of Fitbit Devices: Systematic Review and Narrative Syntheses of Quantitative Data. JMIR mHealth and uHealth 2018; 6: e10527. https://doi.org/10.2196/10527

26.

de Zambotti

Goldstone

Claudatos

, et al. A validation study of Fitbit Charge 2™ compared with polysomnography in adults. Chronobiology international 2018; 35: 465–476. https://doi.org/10.1080/07420528.2017.1413578

27.

Henriksen

Svartdal

Grimsgaard

, et al. Polar Vantage and Oura Physical Activity and Sleep Trackers: Validation and Comparison Study. JMIR formative research 2022; 6: e27248. https://doi.org/10.2196/27248

28.

Svensson

Madhawa

, et al. Validity and reliability of the Oura Ring Generation 3 (Gen3) with Oura sleep staging algorithm 2.0 (OSSA 2.0) when compared to multi-night ambulatory polysomnography: A validation study of 96 participants and 421,045 epochs. Sleep medicine 2024; 115: 251–263. https://doi.org/10.1016/j.sleep.2024.01.020

29.

Kristiansson

Fridolfsson

Arvidsson

, et al. Validation of Oura ring energy expenditure and steps in laboratory and free-living. BMC Med Res Methodol 2023; 23: 50. https://doi.org/10.1186/s12874-023-01868-x

30.

Coronel

Coakes

, et al. Is There a Benefit to Patients Using Wearable Devices Such as Fitbit or Health Apps on Mobiles? A Systematic Review. The American journal of medicine 2019; 132: 1394–1400. https://doi.org/10.1016/j.amjmed.2019.06.018

31.

Beckett

Curtis

Szeto

, et al. Changing User Experience of Wearable Activity Monitors Over 7 Years: Repeat Cross-Sectional Survey Study. Journal of medical Internet research 2025; 27: e56251. https://doi.org/10.2196/56251

32.

Tong

Quiroz

Kocaballi

, et al. Personalized mobile technologies for lifestyle behavior change: A systematic review, meta-analysis, and meta-regression. Preventive medicine 2021; 148: 106532. https://doi.org/10.1016/j.ypmed.2021.106532

33.

Yardley

Spring

Riper

, et al. Understanding and Promoting Effective Engagement With Digital Behavior Change Interventions. American Journal of Preventive Medicine 2016; 51: 833–842. https://doi.org/10.1016/j.amepre.2016.06.015

34.

Mercer

Giangregorio

Schneider

, et al. Acceptance of Commercially Available Wearable Activity Trackers Among Adults Aged Over 50 and With Chronic Illness: A Mixed-Methods Evaluation. JMIR mHealth and uHealth 2016; 4: e7. https://doi.org/10.2196/mhealth.4225

35.

Grydeland

Bergh

Bjelland

, et al. Intervention effects on physical activity: the HEIA study - a cluster randomized controlled trial. International Journal of Behavioral Nutrition and Physical Activity 2013; 10: 17. https://doi.org/10.1186/1479-5868-10-17

36.

Western

Standage

Peacock

, et al. Supporting Behavior Change in Sedentary Adults via Real-time Multidimensional Physical Activity Feedback: Mixed Methods Randomized Controlled Trial. JMIR formative research 2022; 6: e26525. https://doi.org/10.2196/26525

37.

Bandura

. Self-efficacy: the exercise of control. W. H. Freeman, 1997.

38.

Pampel

Krueger

Denney

. Socioeconomic Disparities in Health Behaviors. Annual Review of Sociology 2010; 36: 349–370. https://doi.org/10.1146/annurev.soc.012809.102529

39.

Viswanath

Kreuter

. Health Disparities, Communication Inequalities, and e-Health: A Commentary. American journal of preventive medicine 2007; 32: S131–S133. https://doi.org/10.1016/j.amepre.2007.02.012