It is now well known that positive lifestyle habits, such as regular exercise and adequate sleep, profoundly affect one’s mental and physical health. A sedentary lifestyle is one of the prominent modifiable contributors of mortality and disease globally. It undoubtedly plays a central role in increasing the risk of developing chronic health conditions, such as cardiovascular disease, hypertension, metabolic syndrome, obesity, cancer, and depression (Richardson et al., 2004; Charansonney and Després, 2010; Teychenne et al., 2010; Tremblay et al., 2010; León-Latre et al., 2014; Mainous et al., 2019), and serves as a major predictor of hospitalization and mortality (Biswas et al., 2015). Despite a widespread understanding of the beneficial aspects of physical activity, roughly half of American adults fail to meet the minimum guidelines for recommended exercise (Piercy et al., 2018; Du et al., 2019; Singh et al., 2020). Similarly, short sleep duration, coupled with deficits in sleep quality or restfulness, predisposes individuals to obesity, mental health conditions, neuroendocrine dysfunctions, and cardiovascular disease (Durmer and Dinges, 2005; Vorona et al., 2005; Mullington et al., 2009; Medic et al., 2017). Recent literature indicates that more than 1 in 3U.S. adults do not regularly get enough restful sleep, defined as at least seven hours per night (Centers for Disease Control and Prevention (CDC), 2009; Gamble et al., 2017). Strategies that assist individuals in adopting and maintaining healthier lifestyles may be an important catalyst in addressing public health concerns worldwide.

A growing body of research indicates that the use of digital health technology, notably “wearables,” facilitates the adoption of healthy behaviors with the potential of playing a salubrious role in disease prevention. Wearable technologies are smart microelectronic devices worn on or close to the body that utilize sensors or micro-controllers to detect, transmit, and analyze physiological parameters in real time. Refined sensory technology is also rapidly advancing within the medical device field (Presti et al., 2020). At times, these devices have demonstrated a profound positive impact on health behaviors, such as exercise, sleep, and more (Dunn et al., 2018; Massaroni et al., 2019). Wearables are uniquely capable in that they leverage the users data to provide motivation and accountability (Bianchi, 2018; Nanda et al., 2019; Düking et al., 2020). During sleep deprivation, wearables can encourage the user to get adequate sleep. By corollary, during extended periods of inactivity, they can remind you to increase your activity level. Commonly, wearables include a screen interface or mobile application to augment intrinsic behavior-regulating strategies, such as goal setting, associations, and self-monitoring. Extrinsic strategies enhanced by digital technology include education, reinforcement, encouragement, and actionable feedback (Bandura, 1991; Asbjørnsen et al., 2019).

While many studies support the efficacy of wearables to positively impact behavior in the short term, there is a dearth of evidence on long-term adherence and adaptation. The use of many devices tends to result in only short-term adoption and temporary changes in motivation and behaviors (Klasnja et al., 2011). In fact, a study by Lee et al. found that one-third of consumers who have owned a wearable device stopped using it within six months (Lee et al., 2016). If wearable devices are to be part of the solution of sustained behavior change, a major challenge indeed, they need to leverage principles from theories of health behavior. Current consumer-grade wearables align well with the behavioral change technique of self-monitoring, but without an action planning and commitment step, they may have less of an impact on actual behavioral change.

Adherence seems to be influenced by the individual’s recognition of long-term benefits, social support, and internal motivation (Kononova et al., 2019). User surveys have shown that low levels of compliance are associated with discomfort, the inconvenience of wearable devices, and a lack of activity specificity that is important for goal setting and reinforcement (McNamara et al., 2016; Faust et al., 2017; Ridgers et al., 2018). Thus, perhaps smaller, lighter, and more inconspicuous wearables that require less frequent charging may mitigate some compliance issues. Additionally, research indicates that integrating wearable devices into more comprehensive, personally tailored behavioral interventions improves compliance and behavior change (Dolezal et al., 2015, 2019).

Behavioral modification interventions can empower positive lifestyle and behavioral changes by improving health education, increasing motivation, and providing actionable goals. The present study was conducted to examine the efficacy of a personalized behavioral modification protocol informed by data from a biometric tracking ring over a one-year period. We hypothesized that the behavioral intervention augmented by wearable technology would have improved adherence and subsequent sleep and exercise outcomes compared to a non-augmented intervention.

For the first three months of the study, fifty-six adult volunteers ages 28–47years old (28 females) were enrolled and randomized to the control group (26 participants, 13 female) or intervention group (30 participants, 15 female; Table 2). Following the partitioning of the intervention group after month 3, 15 participants (7 female) in the LT-GTF group and 15 participants (8 female) in the ST-GTF group completed the entire 12-month study. Daily Oura ring data was successfully retrieved from 55 of the 56 study participants. One control participant had an unknown technical issue precluding any data from after the 3rd week. Thus, 99.44% (19,962) of the possible 20,075 daily Oura measures were successfully captured during the 12-month period. During the 30-day baseline collection prior to study start, there were eleven total days of missing data cases among the 56 participants: five due to initially low battery charge status prior to recording, and six short-term memory issues from not opening the associated Oura App for data download within two days (memory limitation).

This study demonstrates the efficacy of utilizing wearable technology in remote delivery of an app-based intervention to improve sleep, activity levels, fitness, and health. In general, a trend emerged across several key outcome measures, affirming the importance of including guidance and encouragement, provided in this study through GTF, as key components of a successful behavior modification intervention. The CON group showed no difference in SOL, Step Count, % Time Jogging, VO₂max, BF%, and HRV, while significant improvement occurred among the INT group over the first three months. During the following nine months, those who continued to receive GTF (LT-GTF group) continued to improve in each of these outcomes except BF%, while scores for those who received no further GTF (ST-GTF group) remained relatively stagnant or regressed slightly. Figure 6 illustrated trends for SOL, Step Count, BF%, and HRV over the course of the study.

SOL is commonly used to aid in the diagnosis of insomnia and to monitor the efficacy of treatment, yet there is no widespread agreement on a clinical threshold. Many researchers and clinicians employ a diagnostic cutoff of >30min (Martin and Ancoli-Israel, 2002; Lichstein et al., 2003); however, Lineberger et al. (Lineberger et al., 2006) demonstrated that a cutoff value of >20min best-discriminated individuals with insomnia from normal sleepers. This threshold has been endorsed in more recent publications (Edinger, 2016; Exelmans et al., 2018). With baseline latencies of 20.6min for the CON group and 25.0min for the INT group, participants in the current study could be considered borderline to meet these criteria for insomnia. Thus, it is meaningful that SOL in the INT group improved to 14.0min at the 3-month follow-up, dropping below the 20-min insomnia threshold and exceeding the published criteria of 10-min improvements for clinical importance (Sateia et al., 2017). In fact, with SOL improvements of −11min at 3months and−15min at 12months among the LT-GTF group, the GTF-enabled behavioral intervention rivaled or outperformed 13 of 14 commonly used pharmacologic treatments analyzed in a recent systematic review and Clinical Practice Guideline commissioned by the American Academy of Sleep Medicine (Sateia et al., 2017). Many considerations factor into treatment decisions for insomnia. Still, these results indicate that behavioral interventions similar to the one presented in this study should be considered, especially among those with mild or sub-clinical insomnia or for whom the pharmacologic intervention is contraindicated.

The continued improvement in SOL among the LT-GTF group is noteworthy. Most interesting, however, is the sustained healthy values of the ST-GTF group, which received no further external engagement after the short-term 3-month intervention, with both groups remaining under the 20-min insomnia threshold at the 12-month follow-up. A similar sustained effect was found in a recent study by Nguyen et al., which demonstrated maintained improvement in sleep disturbances and PSQI score 12weeks after stopping the behavioral feedback and health coaching intervention. Unlike the present study, however, subjects continued using the wearable fitness and sleep tracker (Nguyen et al., 2021). Similar to the present, Nguyen et al. included a control group that wore a fitness tracker without feedback for a three-month period. Although the control group recorded small improvements in wake after sleep onset (WASO) and sleep efficiency, only the feedback group improved number of awakenings (NWAKE) and Pittsburgh sleep quality index (PSQI) score. Taken together, these studies suggest that simply wearing a fitness tracker in isolation is insufficient to realize a meaningful change in most people. Perhaps more importantly, these studies also establish that a relatively brief period of guided feedback may be enough to enact long-term change. While continued feedback appears to produce continued improvement, a short-term period of personalized coaching at the outset of using a fitness tracker may be sufficient to achieve meaningful results and create long-lasting, healthy habits.

This pattern of initial and continued improvement with GTF, and maintenance of improved values upon removal of GTF, was also observed in Step Count, % Time Jogging, and VO₂max. Previous research demonstrates similar improvements in physical activity and cardiovascular fitness resulting from interventions incorporating activity trackers and personalized feedback (Braakhuis et al., 2019; Brickwood et al., 2019, 2021; Ellingson et al., 2019). Furthermore, there is a dose-response relationship between physical activity and major health outcomes, where simply increasing daily step count has a positive impact on mortality and morbidity risk. A widely reported problem with physical activity interventions is low compliance in maintaining the new lifestyle (van der Bij et al., 2002; Müller-Riemenschneider et al., 2008); however, better long-term adaptations have been shown when interventions incorporate feedback and personalized behavioral change techniques compared to generic activity recommendations or activity trackers alone (Dolezal et al., 2015; Ellingson et al., 2019; Brickwood et al., 2021). A study by Ellingson et al. recruited participants with a wide range of physical activity levels and found that among those entering their study with higher activity levels, combining activity tracker use with personalized coaching techniques such as motivational interviewing and habit education was more effective at maintaining high activity levels than receiving an activity tracker alone (Ellingson et al., 2019). This reinforces the previously discussed potential for short-term feedback to facilitate long-term changes.

As our intervention targeted multiple lifestyle factors simultaneously, it is possible the effects observed in certain outcome measures are at least partially the result of improvements made in other related outcomes. Indeed, there is ample evidence indicating that exercise positively affects sleep quality (Yang et al., 2012), and since we did not include groups receiving only partial intervention (e.g., sleep coaching, but no physical activity encouragement), we are unable to conclusively determine to what extent observed improvements in SOL were due solely to an effective sleep intervention. However, prior research has demonstrated that the effects of exercise on sleep are highly variable depending on the age of the participant, and the intensity, and timing of exercise sessions. Additionally, not all sleep-related outcomes are equally affected by exercise (Dolezal et al., 2017). A recent systematic review by Dolezal et al. found good evidence that exercise promoted increased sleep efficiency and duration, yet 8 of 13 studies reporting SOL found no change resulting from exercise (Dolezal et al., 2017). Similarly, a meta-analysis by Banno et al. (2018) found benefits from exercise in subjective measures, such as the PSQI and insomnia severity index (ISI), but found no effect on directly measured sleep latency. Based on these findings, we believe it reasonable to state that improvements in SOL observed in the current study are evidence of an effective sleep intervention, although parsing this out further in future studies involving wearable technology and remotely delivered interventions would be warranted. Regardless, the observed improvement in the current study underscores the importance of a whole-person and whole-lifestyle approach to improved health.

Measuring rMSSD HRV provided a valuable measure of parasympathetic activity and cardiovascular homeostasis restoration. This outcome measure followed a similar pattern to the others discussed, with continued improvement observed in the LT-GTF group and no meaningful change for better or worse during periods when GTF was withheld. To our knowledge, there is no reported minimal clinically important change (MCID) for HRV metrics; however, a recent study reported Minimal Detectable Change scores between 6.92–9.81ms for rMSSD (Bassi et al., 2018), indicating the 12.1ms improvement observed among the LT-GTF group at 12months may be noteworthy. Multiple studies have shown that physical activity positively modulates HRV (Dixon et al., 1992; Malfatto et al., 1996; Sandercock et al., 2007), and HRV may positively influence sleep (Zhuang et al., 2005). Higher HF HRV metrics found during a pre-bedtime resting period have been associated with shorter sleep latency and fewer arousal events (Werner et al., 2015), while another study found higher rMSSD values to be predictive of shorter WASO times (Fantozzi et al., 2019). Therefore, the results of the current study support previous research suggesting a positive relationship between increased HRV and improved sleep metrics.

Wearables have the potential to be at the forefront of preventative medicine by allowing patients to take control of their own healthcare and well-being. For many people, it may be difficult to gauge one’s activity level and quality of sleep (Ferrari et al., 2007; Lauderdale et al., 2008). Providing users with regular biometric feedback eliminates subjectivity and allows the user to obtain a more accurate representation of their current health status. The improvements seen in SOL, Step Count, % Time Jogging, VO2max, BF%, and HR show wearables to be a promising means to generate lasting positive impacts on users’ health and overall quality of life.

The utilization of biometric trackers has become increasingly widespread in our data-driven society. By providing data regarding physical activity, sleep, and other fitness metrics, research has previously shown wearables have the potential to promote improved health and lifestyle. However, despite the extensive data available from a fitness tracker, many users do not consistently comply with or utilize the advantages of fitness tracking in the long term to promote healthy lifestyle habits. Moreover, visually appealing graphs from “big” data misses the point; that is, wearable data should be used as a medium that empowers positive changes in users lives not merely as a collection device with the expectation the user finds it interesting and actionable. That rarely happens. This study underscores the importance of integrating these devices in tailored, personalized behavioral interventions that provide dynamic educational content, real-time encouragement and feedback with progressive goal setting. In the present study, this combination resulted in improved sleep quality and increased physical activity and fitness. Perhaps most notably, our data show that improvements achieved from a short duration of personalized coaching are largely maintained with the continued use of a wearable biometric device.

The original contributions presented in the study are included in the article/supplementary material, and further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by UCLA Institutional Review Board. The patients/participants provided their written informed consent to participate in this study.

JB and BD conceived and coordinated the study and wrote the first manuscript draft. DB, JB, and EN participated in the conception of the study, writing of the first manuscript draft, and overall coordination of the study. BD and EN contributed to data collection and analysis. PG contributed to data collection. KI, QH, DG, and MP contributed to the writing of all subsequent manuscript drafts and a review of the literature. All authors read and provided significant input to all manuscript drafts, agreed to be accountable for all aspects of the work, and approved the final manuscript draft for publication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.