Ladders to better health? Identifying the cognitive pathways and barriers to value attainment in wearable health device usage

Abstract

Objective

Wearable health devices (WHDs) are increasingly important in supporting personal health goals, yet their effectiveness is mediated by the user experience. This study employs means-end chain (MEC) analysis to map the cognitive pathways through which users transition from perceiving device attributes to attaining core health values.

Methods

We applied a soft-laddering technique to interview 63 adult WHD users. The interview data were analyzed using content analysis to identify key attributes, consequences, and values. A summary implication matrix was constructed to illustrate the linkages between these factors, culminating in the generation of a hierarchical value map.

Results

Results indicate that concrete WHD attributes (e.g., wearability, health tracking, haptic reminders) facilitate positive user consequences, such as a sense of companionship, risk awareness, and self-regulation. These pathways lead to the attainment of high-level health values including mental wellness, safety, self-optimization, and lifestyle improvement. Conversely, we identified significant “broken ladders” in which persistent design flaws (e.g., inaccurate/inconsistent data, black-box algorithms, mistimed prompts, unrealistic goals), lead to negative consequences such as decision paralysis, distress, and motivation crowding. These barriers actively hinder value attainment in WHD usage.

Conclusions

This study provides a comprehensive framework of the WHD user journey, highlighting that user value is not derived solely from the technology but also from psychological and cognitive mediators. Findings suggest that developers must move beyond usability to prioritize data transparency and literacy education, while implementers should monitor for compulsive usage patterns to reduce psychological burdens and support long-term digital health engagement.

Keywords

wearable health devices health self-management user experience means-end chain laddering technique

Introduction

Wearable health devices (WHDs) are sensor-integrated technologies worn on the body that facilitate continuous tracking of physiological and behavioral data—such as heart rate, blood oxygen saturation, physical activity, and sleep—for health-related purposes.^1,2 Beyond passive monitoring, these devices typically utilize mobile health (mHealth) applications that leverage persuasive design—including gamified goals, haptic reminders and social comparisons—to educate and nudge users toward better habits and better health.^3,4

The commercial and social significance of this sector is undeniable; the global wearable market is projected to reach $114.36 billion by 2028.⁵ This proliferation has catalyzed a shift toward proactive health management, empowering the public to engage with their personal health data, thereby enhancing health literacy and autonomy.^6,7 Consequently, the integration of WHDs into public health programs has become increasingly common due to their relative affordability and scalability.⁸

Despite their technical potential, the deployment of WHDs is currently hindered by a “utilitarian paradox.” Many WHD-based interventions report significant challenges, including low engagement levels, high attrition rates, and a failure to translate the technology affordances into sustained clinical or behavioral benefits.^9,10 Emerging evidence suggests that the effectiveness of WHDs is not merely a function of technical affordances but is fundamentally mediated by the user experience.⁸

The existing literature characterizes the WHD user experience as inherently complex and ambivalent.¹¹ Users often reported a spectrum of contrasting psychological states, ranging from reassurance to data-driven anxiety, and from perceived utility to skepticism.^12–15 Although existing studies have identified various fragmented elements of this experience, they often lack a cohesive theoretical framework to explain how these divergent outcomes emerge from the same technological affordances. Specifically, the cognitive mechanisms that link low-level device attributes to high-level personal health values remain underexplored.

Our investigation fills in this gap by applying the Means-End Chain (MEC) theory. MEC posits that consumer perceptions are structured as a hierarchical cognitive linkage between perceived product attributes (the “means”), the consequences of use, and the ultimate personal values (the “ends”) that these consequences fulfill.¹⁶ By adopting this framework, we shift the focus from descriptive user experience elements to the relational pathways that define the user journey.

Therefore, this study proposes the following three research questions.

RQ1: How do the multifaceted factors of WHD user experience interweave into a hierarchical means-end structure?

RQ2: What are the predominant cognitive pathways through which users derive personal health values from specific WHD attributes?

RQ3: What “broken ladders” or barriers do users encounter that disrupt the process of value attainment?

This paper holds both theoretical and practical importance. Theoretically, this study contributes to digital health user experience research by applying MEC theory to wearable health devices, moving beyond superficial usability metrics to uncover the deep cognitive pathways that link device features to users’ core health values. On the practical front, the study provides a clear, actionable roadmap for developers and clinicians to design and deploy WHD-based digital health interventions that are not just technically robust but cognitively aligned with users’ core health values.

In the remainder of this paper, we first introduce the theoretical foundation of this study and then describe data collection and analytical procedures in the methodology section. The subsequent results section details and interprets the factors and linkages derived from the data, followed by a discussion of the whole structure of WHD user c experience ladders. We then critically examine the strengths and limitations of this study and outline promising avenues for future research. Finally, we present a concise conclusion that summarizes the study.

Theoretical foundation

The means-end chain (MEC) framework

The MEC theory, originally conceptualized by Gutman (1982),¹⁶ provides a systematic methodology for modeling the cognitive structures that underpin user preferences and behaviors. The core tenet of MEC is that users do not perceive products as isolated entities; rather, they view them as a “means” to achieve desired “ends”.¹⁷ This relationship is modeled as a hierarchical structure comprising three primary levels: attributes, consequences, and values.

Under the assumption that every interaction with a product yields specific outcomes, users learn to associate attributes of the product with subjective consequences, which in turn satisfy deeply held personal values. These cognitive linkages—often referred to as “ladders”—represent the internal logic through which a user justifies their engagement with a product.¹⁸ To capture more nuanced layers of abstraction, as illustrated in Figure 1, Walker and Olson (1991)¹⁹ expanded this framework into a six-level hierarchy.

Figure 1.

The structure of the means-end chain framework, from Gutman (1982)¹⁶ and Walker and Olson (1991)¹⁹.

In terms of attributes, these represent the fundamental characteristics of a product that are recognized and perceived by users. Concrete attributes refer to the physical, tangible characteristics of a product that can be objectively measured (e.g., screen size, battery life). Abstract attributes are defined as subjective characteristics of a product that normally reside in the minds of users (e.g., ease of use, perceived reliability).²⁰

In terms of consequences, they are the physiological or psychological effects arising directly from product use. The extended MEC framework distinguishes between functional consequences and psychological consequences. Functional consequences are the practical, immediate user outcomes resulting from product consumption (e.g., convenience, cost savings).^21,22 Psychological consequences reflect internalized feelings, emotions, experiences or social perceptions that users connect to product attributes (e.g., a sense of accomplishment, ease of anxiety).²²

In terms of values, they refer to the ultimate “end states” or enduring beliefs that guide a user’s behavior.¹⁶ Instrumental values refer to preferred modes of conduct (e.g., being disciplined), whereas terminal values represent final goals of existence (e.g., longevity). As one moves upward through the means-end hierarchy, the level of abstraction increases, and the focus moves from products to users. This progression allows researchers to move beyond what a technology is (technical specs) to uncover the fundamental psychological drivers that make it matter to the user.

MEC in the digital health context

The MEC framework has gained traction in digital health research due to its ability to map complex user-technology interactions. Recent applications include investigations into user experiences with location-based augmented reality games for physical and mental health,²³ older adults’ perceived values of using smart bracelets,²⁴ and female fitness app users’ motivations and perceptions.²⁵

Importantly, MEC offers a distinct advantage over traditional acceptance models (such as TAM or UTAUT), which often focus primarily on “ease of use” and “usefulness” as predictors of adoption. Although these models are effective for measuring intent, they often fail to capture the ambivalence inherent in complex real-world user practices, such as long-term health monitoring in this case.²¹ Because MEC allows modeling both desirable and undesirable consequences, it is uniquely suited to investigate the “broken ladders,” in which technical attributes lead to negative psychological states such as decision paralysis or distress. This capacity for neutral, non-binary modeling makes MEC the ideal lens for exploring the multifaceted mechanisms that drive or disrupt value attainment in the wearable ecosystem.

Methods

Study design

Given the exploratory nature of this study, a qualitatively dominated approach was adopted. Data were collected using laddering interviews and were analyzed following the methodology proposed by Reynolds and Gutman.²⁶ The study design was approved by the Northeast Normal University Ethics Review Committee on March 18, 2023 (R-2023-03-001). This study is reported according to the COREQ (Consolidated Criteria for Reporting Qualitative Research) guidelines.²⁷

Participants and recruitment

This study targeted individuals with experience using WHDs. As acquiring qualitative data is constrained by time, place, and circumstance, sample relevance is considered more critical than sample statistical representation in qualitative-dominated research.²⁸ Accordingly, we employed purposive sampling, a form of nonprobability sampling in which sample selection is based on the researcher’s judgment of which elements best fit the study’s criteria. Considering that social media has a high penetration rate in China (80% by 2025, according to CNNIC) and that users of WHDs significantly overlap with social media users,²⁹ we implemented a low-cost recruitment campaign through leading Chinese social media platforms, including WeChat, QQ and Baidu Tieba, which ran from October 6, 2024 to February 11, 2025. The recruitment campaign advertised the researchers’ roles and the aims of the study. Interested individuals completed an online questionnaire, which collected information on whether they currently used a WHD, the last time they interacted with their device(s), and their age. Inclusion criteria required that participants be 18 years of age or older and have experience using at least one WHD. Exclusion criteria included: a washout period of more than three months for former users (to minimize recall bias) and inability to comprehend written or spoken Mandarin. A total of 71 respondents were initially recruited, of whom 63 were deemed eligible according to the inclusion and exclusion criteria.

Interviews were conducted and coded concurrently with the recruitment process, with the final sample size determined by the achievement of data saturation. By the time the interview reached the 51st respondent, no new themes had emerged. Following established practices in prior research,^22,26 we conducted an additional 12 interviews (about one-fourth of the sample) to test for saturation. Data saturation was considered achieved in this study since no new themes emerged from these additional interviews, resulting in a final sample size of 63 participants.

Laddering interview

A laddering interview is an in-depth, one-on-one interviewing technique used to develop an understanding of how users translate the attributes of products into meaningful associations with respect to the self.²⁶ The laddering method supports two approaches: hard and soft laddering. Hard laddering requires interviewees to generate ladders one-by-one, whereas soft laddering follows their natural flow of speech and allows forks, loops and blind alleys.¹⁷ As this study is exploratory, we employed soft laddering to elicit rich information from participants.

A semi-structured interview guide was prepared, as shown in Table 1. The interview procedure included three stages of Opening, Laddering, and Closing. In the opening stage, the interviewer started with an introduction of the research purpose. As recommended, we proactively disclosed the laddering method vulnerability to the interviewee to facilitate their cooperation in the following stage. Prior to the commencement of the laddering interview, we obtained written informed consent for interview recordings and data usage from each participant and reminded them of their right to withdraw at any point during the research. The laddering interview stage started with eliciting attributes by asking the interviewee to walk the interviewer through their WHDs, followed by a series of directed probes to move up the ladder to consequences and value. This procedure was iterated until no additional information emerged from the respondent. In the closing stage, the interviewer confirmed the interview and collected the interviewee’s demographic information.

Table 1.

The outline of a semi-structured interview.

Procedure	Outline
Opening	• Introduction of the research purpose. • Proactive disclosure of the vulnerability of the laddering method to facilitate the cooperation of interviewees when facing predicaments. Specifically, it is stated that “many of the interview questions may seem somewhat obvious and possibly even stupid, but are designed to acquire deeper reflections from the respondent” • Written informed consent for interview recordings and data usage.
Laddering interview	At start, the interviewee was asked to introduce his/her devices to the interviewer for attributes elicitation, e.g., “Could you please introduce the wearable device you are currently use?”, followed by a series of directed probes typified as “How do you feel about this feature? Why is that important to you (the answer in the last question)?” questions. The following six techniques were applied in the interview process. • Recalling memories, e.g., “Tell me the story of when you started using the device”, “What made you decide to start using one?” • Predicating exclusion of certain attributes, e.g., “What would you think if the device is lost or broken?” • Negative laddering, e.g., “What are the features that you do not like? Why?” • Comparing with a past situation, e.g., “How is the Freestyle Libre glucose monitor different from the device that you used in the past?” • Pretending to be a third person, e.g., “Why do you think other people would want to use wearables?” • Repeating to confirm. Sometimes when respondents gave an obscure answer, the interviewer repeated the same answer to encourage clarification.
Closing	• Concluding questions, e.g., “Is there anything else you wish to tell me in relation to the things we have discussed today?” • Demographic information questionnaire.

The interviews were conducted by CL (a research fellow specializing in Health Informatics), who has expertise in interviewing in general and the specific laddering approach. No prior relationship was established between the researcher and the participants before the commencement of the interviews, and participants were informed of the researcher’s role and purpose through the recruitment and consent process. The interviews took place on the campus of Northeast Normal University, China, from October 2024 to February 2025. Participants were provided with light refreshments during the interview, and no other compensation was offered. All interviews were conducted face-to-face, audio recorded, and transcribed verbatim. The transcripts were not returned to participants for review; however, the researcher checked them against the original recordings to ensure accuracy. Non-verbal cues and the researcher’s initial reflections were documented in field notes. On average, the interviews lasted 25 minutes, with 4 elicited attributes and 3 rungs on a ladder.

Data analysis

Content analysis and intercoder reliability. The present study employed a three-step content analysis procedure comprising open coding, consolidation, and categorization. The MAXQDA2022 software was used for content analysis, as recommended in the literature.^20,21 In open coding, the researchers read the transcripts word-by-word and assigned codes. Next, codes that resemble each other semantically or had different phrases, but clear connections were mapped onto unique factors. For instance, “goal achievement” and “goals and feedback” are closely related and were therefore combined into the construct of “gamified goals”. The constant comparison technique is practiced in the consolidation process to ensure the best fit between the conceptualized factors and the data.³⁰ In the third step, the inductive factors were deductively categorized into attributes, consequences and values, according to the means-end chain framework. Samples of the content coding process are shown in Table 2.

Table 2.

Samples of content coding.

Quotes from the transcripts (direct Chinese-English translation)	Codes	Factors	Categories
“Three activity rings… they gave me goals and feedback about activity… that’s quite nice” [R1].*	Goals and feedback about activity	Gamified goals	Attributes
“A spark-like celebration interface will appear when I make my exercise goals. It is very entertaining” [R1].	Entertaining	Entertainment	Consequences
“It simply works…I definitely walk more in my daily life” [R1].	Walk more	Lifestyle improvement	Values

*[R1] denotes the first interview respondent.

All procedures were conducted independently by CL and XC, and disagreements between them were reconciled through discussion together with LW. In brief, a total of 1056 codes emerged from open coding. Through consolidation, the total number of first-level codes was reduced to 668, represented by 42 distinct factors. These distinct factors were then categorized into 20 attributes, 18 consequences, and 4 values. The intercoder reliability was assessed by calculating Cohen’s Kappa, which is a coefficient that compares the observed agreement between two coders to the agreement expected by chance. A Cohen’s Kappa score above 0.60 is considered substantial agreement. For codes, factors, and categories, the Cohen’s Kappa scores were 0.73, 0.80, and 0.78, respectively, demonstrating substantial reliability.

Construction of the Summary Implication matrix (SIM). The SIM summarizes the linkages between factors identified in the content analysis, effectively transforming qualitative data from the interviews into quantitative data. The linkages were identified from the coded responses to the laddering questions. Two types of linkages existed between factors: direct and indirect. A direct linkage exists when one factor is mentioned directly after another factor in the same ladder. An indirect linkage between two factors exists when they are mentioned in the same ladder, separated by one or more intermediary factors. For example, if a respondent mentioned that the goal setting and feedback features helped him/her schedule and control daily activities, thereby supporting an energetic lifestyle, then a hierarchical linkage is generated: from gamified goals to self-regulation, then to lifestyle improvement. Direct linkages existed between gamified goals and self-regulation and between self-regulation and lifestyle improvement, while indirect linkages existed between gamified goals and lifestyle improvement.

All linkages among factors were summarized in a matrix, with direct linkages appearing on the left of the semicolon and indirect linkages on the right. CL first coded linkages for each respondent and counted them into the SIM. Thereafter, XC performed a reexamination. Based on the SIM, we calculated the abstractness and centrality of each factor. Abstractness of an element is the ratio of in-degree (i.e., the column sum of the element in the SIM) divided by in-degree plus out-degree (i.e., the row sum of the element in the SIM). Factors with higher abstractness scores occupy higher layers in a means-end hierarchy, thus providing an objective method to validate the validity of categorization in content analysis.³¹ Centrality of a factor is calculated by dividing total degree (i.e., in-degree plus out-degree) by the sum of all active cells in the SIM, which indicates the relative importance of each factor.

Generating the hierarchical value map (HVM). The HVM visually illustrates the relationships between factors and is built up by considering the linkages in the SIM. By inspecting the HVM, the dominant value attainment paths can be identified.²⁶ Usually, the HVM does not display all the factors and linkages in the SIM because a trade-off must be made between retaining sufficient information and creating a clear concise structure. This study follows a common practice proposed by Bagozzi and Dabholkar (1994)³² to determine the cut-off value through sensitivity analysis. In this approach, only factors and links above the cut-off value are represented in the structure. The cut-off value is determined by comparing the proportion of selected linkages to the total number of linkages in the SIM. As a guideline, the HVM should cover more than two-thirds of all linkages in the SIM.²⁶

Results

Participant characteristics

The interview sample consisted of 63 WHD users (see Table 3). The gender and age distribution of the sample largely matches that of WHD consumers. In addition, the sample covers diverse educational backgrounds. Regarding brands, multiple mainstream products, including Xiaomi, Apple, Huawei, Libre and others were involved. The sample also includes users with varying levels of experience, from novices (<3 months) to long-term users (>1 year). This diversity enabled the collection of rich qualitative data for this study.

Table 3.

Participant characteristics (N = 63).

Traits	Category	n	Traits	Category	n
Gender	Male	31	Brands of WHDs	Xiaomi	27
Gender	Female	32		Apple	15
Age	18-45	42		Huawei	12
	45-65	19		Libre	3
	>65	2		others	9
Education level	Less than a bachelor’s degree	18	Usage period	<3 months	16
	Bachelor’s degree	24		3-12 months	18
	Master’s degree or higher	21		> 1 year	29

Factors of WHD user experience

From the content analysis, we obtain 20 attributes, 18 consequences, and 4 values, as shown in Table 4. A detailed list of factors with definitions and the number of respondents who mentioned each is presented in Appendix A. In terms of attributes, health tracking (n = 63), gamified goals (n = 23), wearability (n = 20), and social interaction (n = 17) are most frequently mentioned by respondents. Notably, respondents expressed dissatisfaction with attributes related to health data quality—citing issues such as inaccurate data (n = 15), inconsistent data (n = 14), insufficient statistics (n = 7), and black-box algorithms (n = 7); as well as with interaction design attributes, such as mistimed prompts (n = 10), unrealistic goals (n = 10), and upward comparison (n = 8) in social interaction.

Table 4.

Factors of WHD user experience.

Attributes	Consequences	Values
A1: health tracking	C1: self-regulation	V1: lifestyle improvement
A2: gamified goals	C2: risk awareness	V2: mental wellness
A3: wearability	C3: decision paralysis	V3: safety
A4: social interaction	C4: confusion	V4: self-optimization
A5: inaccurate data	C5: motivation crowding
A6: inconsistent data	C6: self - knowledge
A7: haptic reminders	C7: a sense of control
A8: mistimed prompts	C8: a sense of companionship
A9: unrealistic goals	C9: distress
A10: notifications	C10: frustration
A11: upward comparison	C11: entertainment
A12: insufficient statistics	C12: self- reflection
A13: black-box algorithms	C13: annoyance
A14: personalized design	C14: compulsive checking
A15: manual data acquisition	C15: self-experimentation
A16: battery life	C16: inconvenience
A17: price	C17: social burden
A18: avatars	C18: privacy concern
A19: tiny screen
A20: breath coaching

The Summary Implication Matrix and statistics.

In terms of consequences, users reported both positive effects and unintended negative effects. On one hand, WHDs fostered positive effects including enhanced self-regulation (n = 36), self-experimentation (n = 9), self-reflection (n = 12), improved risk awareness (n = 30), self-knowledge (n = 18), sense of control (n = 17), sense of companionship (n = 15), and increased entertainment (n = 12). On the other hand, users experienced side effects like confusion (n = 26), decision paralysis (n = 29), compulsive checking (n = 10), annoyance (n = 10), motivation crowding (n = 23), frustration (n = 13), and distress (n = 15).

There are mainly four main types of health-related values that users achieve through the appropriation of WHDs, including lifestyle improvement (n = 49), mental wellness (n = 40), safety (n = 29), and self-optimization (n = 23).

By summarizing the number of linkages between factors, we obtained a

42 \times 42

SIM (see Appendix B). For illustrative purposes, a sample of the SIM is depicted in Table 5. Numbers on the left of the semi colon (;) represent direct linkages, and the numbers on the right are indirect linkages. For instance, the number (9;4) between A5 (inaccurate data) and C3 (decision paralysis) in the matrix means that A5 leads to C3 directly 9 times, while A5 leads to C3 indirectly 4 times through other intermediary factors. In summary, a total of 657 direct and 541 indirect linkages were identified.

Table 5.

A sample of SIM.

SIM	C1 self-regulation	C2 risk awareness	C3 decision paralysis	C4 confusion	C5 motivation crowding
A1 health tracking	33	30	0	0	1
A2 gamified goals	14	0	0	0	0;10
A3 wearability	0	0	0	0	0
A4 social interaction	0	0	0	0	0;11
A5 inaccurate data	0	0	9;4	6	0

Based on the SIM, we calculated the abstractness and centrality of each factor (see Appendix B). As shown in Figure 2, a scatter plot of factors’ abstractness scores by category, the abstractness level of factors tends to increase from attributes to consequences, then to values. Exceptions are A5, A8, A9, A12, A13 and A14, with abstractness scores of 0.5. These are discussed further with reference to relevant literature,^25,33,34 and are retained in their original category considering their inherent meaning (Figure 3).

Figure 2.

Abstractness scores of factors.

Figure 3.

Centrality scores of factors.

According to the centrality scores of the factors, A1 (health tracking), V1 (lifestyle improvement), C1 (self-regulation), C2 (risk awareness), C3 (decision paralysis), and C4 (confusion) play predominant roles in the structure, collectively accounting for 81% of all links in the implication matrix.

Mapping WHD users’ value attainment paths via HVM

As discussed in the Data Analysis section, we need to select a cutoff level for factor linkages before building the HVM.²⁶ A sensitivity analysis was constructed and presented in Table 6. Based on the analysis, a cut-off level of 6 was deemed most appropriate in the present case. The HVM was built using the cutoff level of 6, which accounts for 91.8% of all the linkages among factors (as shown in Figure 4). This decision was consistent with prior studies, which recommended a cut-off of 6 linkages as appropriate for a sample size of 60.

Table 6.

Sensitivity analysis to determine the cut-off level.

Cutoff value	Active linkages (%, N = 657)
4	639 (97.3%)
5	615 (93.3%)
6	603 (91.8%)
7	577 (87.8%)
8	514 (78.2%)

Figure 4.

The HVM of WHD user experience.

This study initially adopted a three-level HVM comprising attributes, consequences, and values. However, data analysis revealed significant hierarchical relationships within both attributes and consequences; therefore, the basic levels of attributes and consequences were further divided into two sublevels, drawing on concepts from previous studies.¹⁹ As a result, the final model comprised five structural level, labeled as “Concrete attributes,” “Abstract attributes,” “Functional consequences,” “Psychosocial consequences,” and “Values.” In total, 33 elements with 48 distinct linkages provide a graphical summary of the hierarchical structure pertinent to WHD user experience (Figure 4).

Discussion

The HVM of WHD user experience

In the HVM of WHD user experience, four critical values exist at the top: lifestyle improvement, mental wellness, safety and self-optimization. First, lifestyle improvement emerged as the most significant value across the whole HVM, with the highest number of linkages. In the existing literature, WHDs are frequently employed in interventions targeting key lifestyle elements, such as regular exercise, good sleep hygiene, and a balanced diet,³⁵ although their effectiveness remains controversial.³⁶ This study finds that lifestyle improvement is directly facilitated by self-regulation, entertainment, and risk awareness, but hindered by motivation crowding and decision paralysis, as unintended consequences of WHD usage. The HVM further reveals that the positive consequences supporting the ultimate value of lifestyle improvement stem from concrete technological attributes, notably the core feature of health tracking and interaction features such as notifications, haptic reminders, gamified goals, and social interaction. For instance, one respondent described the benefits of using WHDs for lifestyle improvement, “One fun feature is the (virtual) medals awarded for celebrating exercise goals. They make exercise more entertaining, and earning each one keeps me motivated to stay active in daily life” [R2]. The HVM further reveals critical breakdowns in the pathway to lifestyle improvement, rooted in persistent data quality issues and interaction design flaws. Data quality issues, including inaccurate/inconsistent data, insufficient statistics, and black-box algorithms, collectively lead to decision paralysis in lifestyle adjustment. Interaction design flaws—such as mistimed prompts, unrealistic goals, and upward comparison—crowd out motivation for lifestyle improvement. The path-specific discussion of the ambivalent effect of WHD usage on lifestyle improvement is conducted in subsequent sections.

Second, safety emerged as a key value sought by WHD users. The HVM indicates that one prerequisite for safety is risk awareness, which is enhanced through the continuous monitoring of critical health metrics (i.e., health tracking), paired with real-time notifications of abnormal events. Safety and health risk assessment is an important area in which wearable technology contributes. Previous studies have proposed wearable technologies as solutions for detecting and alerting users to falls,³⁷ extreme glucose levels,³⁸ fatigue³⁹ and heart disease.⁴⁰ However, the reliability of wearables in supporting safety and health risk management is often compromised by data quality issues. Specifically, health data quality issues, including inaccurate/inconsistent data, insufficient statistics, and black-box algorithms, first confuse users, then lead to decision paralysis, and ultimately undermine the value of safety. One respondent recounted an experience in which WHDs failed to provide adequate safety protection: “I had a severe low blood sugar episode. The monitor displayed 4.7 mmol/L, but a fingerstick test showed a much lower reading of 2.4 mmol/L. I cannot understand why the numbers don’t match… this uncertainty increases the risk” [R31].

Users also seek to improve mental wellness with the help of WHDs. The mental effects of WHDs have become an increasing focus in digital health literature.⁴¹ Some researchers have applied WHDs in the treatment of anxiety and depression, suggesting that WHDs can offer mental wellness-related values.^42,43 However, other studies indicate that WHDs may also provoke adverse psychological reactions.⁴⁴ Drawing on the mechanisms revealed by the HVM, mental wellness as a terminal value is positively influenced by a sense of companionship and a sense of control, while being negatively affected by the unintended consequence of distress. Notably, the attribute of wearability, which fosters a sense of companionship can also intensify distress by encouraging compulsive checking.

Studies on the quantified-self movement have noted that people are increasingly participate in digital monitoring of their health and fitness for self-optimization.⁴⁵ In the HVM analysis, self-optimization emerges as a significant value for WHD users. As one respondent explained, “I value the recorded data on my running activity; it helps me gauge my progress in increasing my speed” [R36]. From the vantage point of the whole HVM, the value of self-optimization is positively influenced by self-knowledge derived from practices of self-experimentation and self-reflection based on health tracking, while it is negatively affected by decision paralysis caused by data quality issues.

By analyzing the significant factors, linkages, ladders and chains, we divided the overall HVM structure into three sub-structures to facilitate a more in-depth discussion. These sub-structures are presented separately in Figures 5–7 and are discussed in-detail to provide deeper understanding and insights.

Figure 5.

The sub-structure of psychological feelings in the HVM.

Figure 6.

The sub-structure of cognitive processing in the HVM.

Figure 7.

The sub-structure of motivational dynamics in the HVM.

The sub-structure of psychological feelings in the HVM

In the sub-structure of psychological feelings of WHD user experience (Figure 5), two pairs of conflicting user experience paths emerge from the same attributes of WHDs: wearability and health tracking. We provide a detailed description and discussion of ambivalent user experiences regarding these two attributes, as follows.

The wearability feature of WHDs, characterized by their proximity to the body and a ubiquitous presence,⁴⁶ can influence mental wellness in both positive and negative ways, depending on users’ psychological patterns. On the positive side, wearability, along with personalized design, can foster a sustained sense of companionship, which, in turn, enhance mental wellness. Jang and Kim⁴⁶ scrutinized companion-like experiences with WHDs through the lenses of relational and extended-self. Consistent with the present study, their findings demonstrated the significant effect of personalized design on perceived companionship. Conversely, wearability can also have negative mental effects: it may encourage compulsive checking behaviors, leading to increased distress and ultimately undermining mental wellness. This pattern is often observed among anxious patients, for whom the ubiquitous presence of WHDs serves as a constant reminder of their condition.⁴⁷ To cope with the fear of disease progression, these users may resort to compulsive checking, which in turn reinforces a cycle of distress. For instance, one respondent who was recently diagnosed with diabetes stated, “I feel the sensor all the time. It is implanted in my body and invades my mind. I can’t help but constantly think about my disease. Checking my glucose level hundreds of times a day… It’s just so exhausting” [R38]. While researchers and practitioners have primarily focused on issues of discontinuance and low engagement with WHDs, this study suggests that overuse and excessive dependence may also arise under certain conditions.

In contrast to the vague and implicit perception of one’s own health status, the precise, objective health tracking data provide users with a positive sense of control, which can ultimately foster mental health. A sense of control refers to the belief that one’s life is controllable by his/her own actions rather than being a consequence of chance.¹⁴ Numerous studies have demonstrated a strong relationship between perceived control and positive mental health outcomes. For example, Precht et al.⁴⁸ found that perceived control positively related to emotional well-being during the COVID-19 pandemic. In this study, a sense of control serves as an important psychological resource for individuals in coping with stressors arising from various health challenges. However, new stressors are introduced by health data quality issues (i.e., inconsistent/inaccurate data, insufficient statistics, and black-box algorithms), which first lead to confusion and then to distress, ultimately undermining mental wellness. One respondent who diagnosed with atrial fibrillation and experiencing anxiety described his interaction with health monitoring data as follows, “I need to reassure myself that my heart is doing okay. While the device is somewhat useful, it frequently detects AFib despite me feeling no discomfort. This confuses me and only makes me more anxious than before” [R43]. These findings indicate that although health tracking has significant potential to support mental wellness, their benefits are largely compromised by data quality issues. A detailed discussion of data quality issues is presented in the following section, which focuses on the cognitive processing structure of WHD user experiences.

The sub-structure of cognitive processing in the HVM

The sub-structure of cognitive processing in the HVM illustrates how WHD users’ information practices add value to health-related decision-making (Figure 6). First, by observing and assessing health tracking data, users can promptly identify potential health risks. In addition, advanced WHDs send notifications to report the automatically detected abnormalities. The resulting increase in risk awareness often encourages users to participate further in formal medical tests and interventions, thereby supporting health and safety. For example, one respondent stated: “The ECG provided by the Apple Watch showed that I had sinus rhythm and ventricular premature beats ... so I went to the hospital for further examination and was diagnosed with myocardial ischemia” [R30]. This process following paths summarize this process: from notifications/health tracking to risk awareness, and then to safety. In traditional contexts, health risks or subhealth conditions are often neglected by individual before obvious symptoms appear.⁴⁹ Consistent with Baumann and Weinberger,⁵⁰ and Charlton and Kyriaco,⁵¹ who argue that WHDs can foster preventive healthcare, this study’s findings suggest that WHDs can enhance risk awareness and support health and safety. Kirubakaran⁵² advocates a more direct approach: sharing health tracking data with healthcare providers for safety purposes. This approach is especially valuable for at-risk populations but less feasible for the general population due to the limited computing resources in healthcare systems.

Second, through self-experimentation and self-reflection, health tracking data are transformed into self-knowledge, ultimately contributing to self-optimization. This pathway illustrates how WHD users integrate information practices with concrete health practices. In self-experimentation, users aim to identify relationships between behavior and physiological responses.⁵³ For instance, one respondent noted, “I figured out how sensitive my body is to energy drinks through simple experiments. After drinking coffee, my resting heart rate went up from around 65 to 75. But with tea and Red Bull, the effect wasn’t as noticeable”[R5]. Through self-reflection, users review historical health records and develop rational explanations for data fluctuations. As one respondent explained, “After riding the spinning bike for an hour, I checked the exercise record and found that most of the effort went into endurance training. To improve calory-burning efficiency, I increased the resistance to the highest level I could sustain” [R52]. Self-experimentation and self-reflection represent two distinct modes of knowledge discovery—deduction and inference—both of which require a relatively high level of data literacy and scientific literacy.⁵⁴

Users often experience confusion when processing health data from WHDs, leading to decision paralysis across contexts. These negative consequences are linked to health data quality issues including inconsistent/inaccurate data, insufficient statistics and black-box algorithms. First, users compare different brands of WHDs when measuring the same health metric and often notice significant inconsistencies. As one respondent noted, “I use a Huawei and a Xiaomi band to track my sleep, and there is about an hour’s difference between the two … I’m not sure which to trust” [R62]. Such discrepancy stems in part from the lack of standardized data processing across the wearable industry.⁵⁵ Second, users frequently compare WHD outputs with “gold standards”, such as direct observations or formal medical tests, and identify clear measurement errors. Prior validation studies suggest that commercially available WHDs sacrifice accuracy in favor of convenience and low cost, and their estimation of sleep, heart rate and physical activity are susceptible to environment noise.⁵⁶ Third, some users expressed disappointment with the lack of potentially valuable statistical analysis, such as time-series or correlation analysis. For example, a respondent stated, “The sleep data hasn’t really helped with my insomnia, mainly because the wearables don’t connect it with my daily activities, which seems like a key factor” [R1]. Fourth, some users complain that sleep scores and physical activity index are difficult to interpret due to black-box algorithms. The complex synthetic process underlying these measures raise doubts about their meaning and reliability.² In summary, these negative experiences in the cognitive processing highlight the unmet information needs of WHD users in managing their personal health. As wearable technologies continue to evolve, emerging solutions in academia hold promise for addressing data quality issues raised by users. For instance, Weed et al.⁵⁷ and Kim et al.⁵⁸ have proposed different approaches for wearable-based circadian phase estimation using light exposure, activity, and body temperature data. These approaches not only have the potential to provide a new type of health indicator to general users, but also to enhance the accuracy and interpretability of downstream health metrics, such as sleep scores.

The sub-structure of motivational dynamics in the HVM

Within the HVM, the motivational substructure includes two positive pathways that encourage WHD users to engage in lifestyle improvement, while two negative pathways hinder their participation. First, WHDs empower users to self-regulate key elements of their lifestyle, summarized as the chains: from health tracking/gamified goals/haptic reminders to self-regulation, and then to lifestyle improvement. For instance, a respondent said, “The calorie consumption data from my tracker is really helpful …I use it to manage my diet and avoid overeating” [R10]. Second, the interaction features of WHDs contribute to a more entertaining environment, which encourages sustained commitment to lifestyle improvement, captured by the chains: from gamified goals/social interaction to entertainment, and then to lifestyle improvement. As one respondent noted, “Apple Watch provides (virtual) medals for celebrating daily exercise goal attainment. It is fun. Getting medals is both the motivation and encouragement for exercise” [R11]. Similarly, existing studies show that WHD use is positively associated with enjoyment.⁵⁹ Lifestyle improvement requires long-term commitment to deliver health benefits, and the lack of immediate incentives makes it difficult for individuals to sustain motivation.⁶⁰ This study suggests that WHDs supplement this process by enabling timely self-regulation and offering immediate, entertainment-based incentives that support sustained lifestyle improve.

However, when external incentives from WHDs are not successfully internalized by users, they may lead to unintended negative consequences and ultimately undermine users’ self-determined motivation for lifestyle improvement. Specifically, two negative chains exist in the sub-structure of motivational dynamics. First, reminders are frequently perceived by users as mistimed prompts, triggering annoyance, and consequently producing a motivation crowding-out effect on lifestyle improvement. For instance, one respondent said, “It is really annoying getting the vibration from my wristband reminding me to stand up when I’m in an important meeting. I actually think it is intrusive and makes me feel like I’m losing autonomy over my own schedule” [R4]. Addressing this concern requires context-aware prompt timing. One promising avenue is to integrate circadian phase estimation into prompt-timing algorithms—an approach supported by Kim et al.,⁵⁸ who emphasize that accurate circadian phase estimation is a prerequisite for delivering timely interventions.

Second, some users experienced frustration arising from unrealistic goals and upward comparison in social interaction, which, in turn, threatened their self-determined motivations for lifestyle improvement. For example, one respondent lamented, “The 10,000-step goal feels unrealistic for me; it leaves me feeling frustrated and discouraged about trying to live an active life” [R11]. While WHDs provide users with other-determined motivation for lifestyle improvement,¹⁵ they may crowd out more self-determined forms of motivation, which are associated with better adherence to health behaviors.⁶¹ To mitigate those potential negative effects, we propose that WHDs should be further developed in three key areas: context-sensitive reminders that respond to users’ dynamic schedules, personalized goal setting, and a supportive, encouraging social interaction environment.

Strengths, limitations and future research

The main strength of the current study lies in the construction of a means-end chain structure of WHD user experience factors, which adequately captures the entire pathway from WHD attributes to the intermediate user consequences and ultimately to personal health values. By modeling both positively and negatively valence factors within the MEC framework, this study addresses the inherent complexity and ambivalence of WHD user experience. Moreover, while WHD user experience has been explored qualitatively in previous studies, the combined qualitative, quantitative and visual data analysis approach used in this study represents an additional strength.

There are however several limitations to the current study that could pave the way for future research. First, this study may be subject to sample bias given that the sample consisted solely of Chinese participants. The use of WHDs is profoundly shaped by contextual factors such as health beliefs, digital literacy, social norms, and trust in technology,⁶² which can differ significantly across cultures. Therefore, it is important to note that this study was conducted within the Chinese cultural context—a highly digitized society with a population of moderate health literacy levels. Future research that adopting a cross-cultural lens would be a valuable extension of this work. Second, the present study applied a qualitatively dominated approach, which does not allow for strict statistical generalization of the findings. While the sample size adhered to Reynolds and Gutman’s²⁶ guidelines for generating valuable insights into WHD user experiences, it does not permit a robust quantitative analysis. Future research can adopt a quantitative approach with a larger sample to validate the hierarchical structure of WHD user experience identified in this study and offer more generalizable implications regarding wearable technology development and application. Third, while our sample includes users with varying levels of experience, the group sizes limit our ability to conduct robust comparative analyses. To adequately capture how users’ perceptions of WHDs change over time, a longitudinal design should be considered in future research.

Conclusions

This study is among the first to investigate WHD user experience from a means-end chain perspective. Through systematic analysis, this study identifies not only positive user outcomes but also critical breakdowns in the technology-health chain. Specifically, our findings indicate that users generally undergo a complex multistage process to achieve personal health values, including lifestyle improvement, safety, mental wellness, and self-optimization, through their interactions with WHDs. This study reveals that the attributes and functions of WHDs contribute to health values by providing various intermediate benefits, such as fostering a sense of companionship and control, enhancing risk awareness, improving self-knowledge and self-regulation, and increasing entertainment. It also highlights the technological and user barriers that hinder the attainment of value. To cope with the negative user experiences, WHDs need to be further developed to address data quality issues, including inaccurate/inconsistent data, insufficient statistics, and black-box algorithms, as well as interaction design problems including mistimed prompts, unrealistic goals, and stressful upward social comparison. In addition, special attention should be given to compulsive use behaviors, which are likely to occur in vulnerable populations.

Supplemental material

Supplemental material - Ladders to better health? Identifying the cognitive pathways and barriers to value attainment in wearable health device usage

Supplemental material for Ladders to better health? Identifying the cognitive pathways and barriers to value attainment in wearable health device usage by Caining Li, Xiaoyu Chen and Lai Wei in Digital Health.

Supplemental material

Supplemental material - Ladders to better health? Identifying the cognitive pathways and barriers to value attainment in wearable health device usage

Footnotes

ORCID iD

Caining Li

Ethical considerations

The study received ethical approval from the Northeast Normal University Ethics Review Committee on March 18, 2023 (R-2023-03-001). Written informed consent was obtained from all participants prior to study initiation.

Author contributions

CL and LW designed the study framework. CL and XC conducted the interviews, the data analysis and interpretation. CL drafted the manuscript. All authors provided input into manuscript preparation and approved the final manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by National Natural Science Foundation of China (No. 72404053). CL is supported by the Social Science Research Project of Jilin Provincial Department of Education (No. JJKH20241438SK), the Fundamental Research Funds for the Central Universities (No. 2412024QD030), the Postdoctoral Fellowship Program of CPSF (No. GZC20230415).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Anonymized dataset supporting the conclusion of this article is available upon reasonable request for bona fide researchers with an established scientific record and bona fide organizations.

Guarantor

CL.

Supplemental material

Supplemental material for this article is available online.

References

Mettler

Wulf

. Physiolytics at the workplace: Affordances and constraints of wearables use from an employee's perspective. Information Systems Journal 2019; 29: 245–273. https://doi.org/10.1111/isj.12205

Lin

Chib

. The state of wearable health technologies: a transdisciplinary literature review. Mobile Media & Communication 2021; 9: 353–376. https://doi.org/10.1177/2050157920966023

Villalobos-Zúñiga

Cherubini

. Apps that motivate: a taxonomy of app features based on self-determination theory. International Journal of Human-Computer Studies 2020; 140: 102449. https://doi.org/10.1016/j.ijhcs.2020.102449

Bartlett

Webb

Hawley

. Using persuasive technology to increase physical activity in people with chronic obstructive pulmonary disease by encouraging regular walking: a mixed-methods study exploring opinions and preferences. Journal of medical Internet research 2017; 19: e124. https://doi.org/10.2196/jmir.6616

Fortune . Fitness tracker market size. https://www.fortunebusinessinsights.com/fitness-tracker-market-103358 (2025, accessed June, 8 2023).

Dunn

Runge

Snyder

. Wearables and the medical revolution. Personalized Medicine 2018; 15: 429–448. https://doi.org/10.2217/pme-2018-0044

Liu

Yao

, et al. Proactive Health: An imperative to achieve the goal of healthy China. China CDC Wkly 2022; 4: 799–801, 2022/10/27. https://doi.org/10.46234/ccdcw2022.156

Tuominen

Koski

Axelin

, et al. Acceptable, useful, and ineffective? Recent retirees’ experiences of a 12-month activity tracker-based physical activity intervention. Digital Health 2023; 9: 20552076221147419. https://doi.org/10.1177/20552076221147419

Hermsen

Moons

Kerkhof

, et al. Determinants for sustained use of an activity tracker: Observational study. JMIR Mhealth Uhealth 2017; 5: e164. 20171030. DOI: https://doi.org/10.2196/mhealth.7311

10.

Chen

. Wearable activity trackers for promoting physical activity: A systematic meta-analytic review. International Journal of Medical Informatics 2021; 152: 104487, 2021/05/22. https://doi.org/10.1016/j.ijmedinf.2021.104487

11.

Shen

Sun

. Wearable health information systems intermittent discontinuance: A revised expectation-disconfirmation model. Industrial Management & Data Systems 2018; 118: 506–523. https://doi.org/10.1108/imds-05-2017-0222

12.

Andersen

Langstrup

Lomborg

. Experiences with wearable activity data during self-care by chronic heart patients. Journal of Medical Internet Research 2020; 22: e15873. https://doi.org/10.2196/15873

13.

Nelson

Sools

Vollenbroek-Hutten

MMR

, et al. A means-end chain model based on consumer categorization processes. Jmir Mhealth and Uhealth 2020; 8: e16973. https://doi.org/10.2196/16973

14.

Pingo

Narayan

. My smartwatch told me to see a sleep doctor”: a study of activity tracker use. Online Inf Rev 2020; 44: 503–519, Article. https://doi.org/10.1108/oir-04-2018-0115

15.

Burford

Golaszewski

Bartholomew

. I shy away from them because they are very identifiable”: A qualitative study exploring user and non-user's perceptions of wearable activity trackers. Digital Health 2021; 7: 20552076211054922. https://doi.org/10.1177/20552076211054922

16.

Gutman

. A Means-End Chain Model Based on Consumer Categorization Processes. Journal of Marketing 1982; 46: 60–72. https://doi.org/10.1177/002224298204600207

17.

Grunent

. Measuring subjective meaning structures by the laddering method: Theoretical considerations and methodological problems. International Journal of Research in Marketing 1995; 12: 209–225. https://doi.org/10.1016/0167-8116(95)00022-T

18.

Rietz

Maedche

. Ladderbot—A conversational agent for human-like online laddering interviews. International Journal of Human-Computer Studies 2023; 171: 102969. https://doi.org/10.1016/j.ijhcs.2022.102969

19.

Walker

Olson

. Means-end chains: Connecting products with self. Journal of Business Research 1991; 22: 111–118. https://doi.org/10.1016/0148-2963(91)90045-Y

20.

Heinze

Thomann

Fischer

. Ladders to m-commerce resistance: A qualitative means-end approach. Computers in Human Behavior 2017; 73: 362–374. https://doi.org/10.1016/j.chb.2017.03.059

21.

Kunst

Ringberg

Vatrapu

. Beyond popularity: A user perspective on observable behaviours in a digital platform. Information Systems Journal 2021; 32: 595–622. https://doi.org/10.1111/isj.12365

22.

X-Y

Tayyab

SMU

Chang

F-K

, et al. Hierarchical value-attainment paths of CBEC consumers: a means-end-chain perspective. Internet Research 2020; 31: 699–736. https://doi.org/10.1108/intr-10-2019-0397

23.

GCM

Shang

. How do location-based augmented reality games improve physical and mental health? Evaluating the meanings and values of Pokemon Go users' experiences through the means-end chain theory. Healthcare 2021; 9: 794. https://doi.org/10.3390/healthcare9070794

24.

Jeng

Yeh

Pai

. Analyzing Older Adults' Perceived Values of Using Smart Bracelets by Means-End Chain. Healthcare (Basel) 2020; 8: 494.

25.

Lyu

Abidin

NEZ

Zulnaidi

. Exploring female fitness app users' motivations and perceptions: A qualitative study. Medicine 2024; 103: e40076. https://doi.org/10.1097/MD.0000000000040076

26.

Reynolds

Gutman

. Laddering theory, method, analysis, and interpretation. Journal of Advertising Research 1988; 28: 11–31. https://doi.org/10.1080/00218499.1988.12467766

27.

Tong

Sainsbury

Craig

. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. International Journal for Quality in Health Care 2007; 19: 349–357. https://doi.org/10.1093/intqhc/mzm042

28.

Denzin

. Introduction: entering the field of qualitative research. Handbook of Qualitative Research. Sage Publications, 1994.

29.

Zheng

. Association between fitness technology use and physical activity mediated by communication behaviors on social media. Digital Health 2024; 10: 10. https://doi.org/10.1177/20552076241266367

30.

Manuel

Marlen

Yetton

, et al. Medicine grounded theory methodology in information systems research. Management Information Systems Quarterly 2017; 41: 685–701.

31.

Pieters

Baumgartner

Allen

. A means-end chain approach to consumer goal structures. International Journal of Research in Marketing 1995; 12: 227–244. https://doi.org/10.1016/0167-8116(95)00023-U

32.

Bagozzi

Dabholkar

. Consumer recycling goals and their effect on decisions to recycle: A means‐end chain analysis. Psychology & Marketing 1994; 11: 313–340. https://doi.org/10.1002/mar.4220110403

33.

Jeng

Yeh

Pai

. Analyzing older adults’ perceived values of using smart bracelets by means-end chain. Healthcare (Basel) 2020; 8: 20201118. https://doi.org/10.3390/healthcare8040494

34.

Jeng

Yeh

Pai

. A performance evaluation matrix for measuring the life satisfaction of older adults using ehealth wearables. Healthcare (Basel) 2022; 10: 20220323. https://doi.org/10.3390/healthcare10040605

35.

Khasentino

Belyaeva

Liu

, et al. A personal health large language model for sleep and fitness coaching. Nat Med 2025; 31: 3394–339420253403. 20250814. https://doi.org/10.1038/s41591-025-03888-0

36.

Patel

Asch

Volpp

. Wearable devices as facilitators, not drivers, of health behavior change. Journal of American medical association 2015; 313: 459–460, 2015/01/09. https://doi.org/10.1001/jama.2014.14781

37.

Chai

Chen

, et al. Fall detection with wearable sensors: a hierarchical attention-based convolutional neural network approach. Journal of Management Information Systems 2021; 38: 1095–1121. https://doi.org/10.1080/07421222.2021.1990617

38.

Barnard-Kelly

Martínez-Brocca

Glatzer

, et al. Identifying the deficiencies of currently available CGM to improve uptake and benefit. Diabetic Med 2024; 41: e15338. https://doi.org/10.1111/dme.15338

39.

Horiuchi

Ogasawara

Miki

. Fatigue assessment by blink detected with attachable optical sensors of dye-sensitized photovoltaic cells. Micromachines 2018; 9. https://doi.org/10.3390/mi9060310

40.

Antoniou

Xanthopoulos

Giamouzis

, et al. Efficacy, efficiency and safety of a cardiac telerehabilitation programme using wearable sensors in patients with coronary heart disease: the TELEWEAR-CR study protocol. BMJ Open 2022; 12: 10, Article. https://doi.org/10.1136/bmjopen-2021-059945

41.

Alaboson

Coffey

Shrivastava

, et al. Impact of connected health on the psychological wellbeing and quality of life of people with multiple sclerosis and their caregivers: A systematic review. Digit Health 2025; 11: 20552076251326230. 20250331. https://doi.org/10.1177/20552076251326230

42.

Hunkin

King

Zajac

. Wearable devices as adjuncts in the treatment of anxiety-related symptoms: A narrative review of five device modalities and implications for clinical practice. Clin Psychol-Sci Pract 2019; 26: 12: Review. https://doi.org/10.1111/cpsp.12290

43.

Lekkas

Gyorda

Price

, et al. Depression deconstructed: Wearables and passive digital phenotyping for analyzing individual symptoms. Behav Res Ther 2023; 168: 104382. https://doi.org/10.1016/j.brat.2023.104382

44.

Gaffey

Hooker

Lampert

, et al. Psychological aspects of wearable health technologies in patients with cardiovascular disease: a literature review and call to action. Curr Cardiol Rep 2025; 27: 92: 20250606. https://doi.org/10.1007/s11886-025-02234-7

45.

Esmonde

. There's only so much data you can handle in your life': accommodating and resisting self-surveillance in women's running and fitness tracking practices. Qualitative Research in Sport Exercise and Health 2020; 12: 76–90. https://doi.org/10.1080/2159676x.2019.1617188

46.

Jang

Kim

. Healthier Life with Digital Companions: Effects of Reflection-Level and Statement-Type of Messages on Behavior Change via a Perceived Companion. International Journal of Human-Computer Interaction 2019; 36: 172–189. https://doi.org/10.1080/10447318.2019.1615722

47.

Dhir

Nijjer

Luo

, et al. Wearable discontinuance: Pathways in the volitional information systems discontinuance process. Information & Management 2024; 61: 18, Article. https://doi.org/10.1016/j.im.2024.104038

48.

Precht

Margraf

Stirnberg

, et al. It’s all about control: Sense of control mediates the relationship between physical activity and mental health during the COVID-19 pandemic in Germany. Curr Psychol 2023; 42: 8531–853120218539. 20211020. https://doi.org/10.1007/s12144-021-02303-4

49.

Balla

Hagger

. Protection motivation theory and health behaviour: conceptual review, discussion of limitations, and recommendations for best practice and future research. Health Psychol Rev 2025; 19: 145–171, Review. https://doi.org/10.1080/17437199.2024.2413011

50.

Baumann

Weinberger

Maia

, et al. User types, psycho-social effects and societal trends related to the use of consumer health technologies. Digit Health 2023; 9: 20552076231163996, 2023/04/11. https://doi.org/10.1177/20552076231163996

51.

Charlton

Kyriaco

Mant

, et al. Wearable photoplethysmography for cardiovascular monitoring. Proc IEEE Inst Electr Electron Eng 2022; 110: 355–381, 2022/04/01. https://doi.org/10.1109/JPROC.2022.3149785

52.

Kirubakaran

. Review on emerging frontiers in medical sensing: Innovations shaping the future of healthcare. Biomedical Materials & Devices 2025; 4: 1888–1908. https://doi.org/10.1007/s44174-025-00385-3

53.

Whelan

Denton

Bourne

CLA

, et al. A digital lifestyle behaviour change intervention for the prevention of type 2 diabetes: a qualitative study exploring intuitive engagement with real-time glucose and physical activity feedback. Bmc Public Health 2021; 21: 130. https://doi.org/10.1186/s12889-020-09740-z

54.

Chiauzzi

Hekler

Lee

, et al. In search of a daily physical activity “sweet spot”: Piloting a digital tracking intervention for people with multiple sclerosis. Digital Health 2019; 5: 11, Article. https://doi.org/10.1177/2055207619872077

55.

Ortiz

Gupta

Kumar

, et al. Data preprocessing techniques for AI and machine learning readiness: Scoping review of wearable sensor data in cancer care. JMIR Mhealth Uhealth 2024; 12: e59587. 20240927. https://doi.org/10.2196/59587

56.

Miller

Sargent

Roach

. A validation of six wearable devices for estimating sleep, heart rate and heart rate variability in healthy adults. Sensors (Basel) 2022; 22: 20220822. https://doi.org/10.3390/s22166317

57.

Weed

Jamgochian

Hilaire

, et al. Circadian phase estimation from ambulatory wearables with particle filtering: accuracy depends on initialization, recording duration, and light exposure. Journal of Biological Rhythms 2026; 41: 42–52. https://doi.org/10.1177/07487304251392289

58.

Kim

Lee

Forger

. Wearable data assimilation to estimate the circadian phase. SIAM Journal on Applied Mathematics 2023; 84: S452–S475. https://doi.org/10.1137/22m1509680

59.

Jung

Kang

. Self-determination in wearable fitness technology: The moderating effect of age. International Journal of Human-Computer Interaction 2021; 38: 1399–1409. https://doi.org/10.1080/10447318.2021.2002048

60.

Adjerid

Loewenstein

Purta

, et al. Gain-loss incentives and physical activity: The role of choice and wearable health tools. Management Science 2022; 68: 2642–2667. https://doi.org/10.1287/mnsc.2021.4004

61.

Kerner

Goodyear

. The motivational impact of wearable healthy lifestyle technologies: a self-determination perspective on Fitbits with adolescents. American journal of health education 2017; 48: 287–297. https://doi.org/10.1080/19325037.2017.1343161

62.

Kim

Huber

. Characteristics of personal health information management groups: findings from an online survey using Amazon's mTurk. Journal of the medical library association 2017; 105: 361–375, 2017/10/07. https://doi.org/10.5195/jmla.2017.312

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.14 MB

0.00 MB

0.02 MB