Self-tracking and the quantified self are not new labels. As early as the 1970s, wearable computers for self-tracking were used as forms of personal surveillance [25]. In 2001, a small number of media practitioners began to use newly available digital technology to track their daily life for designing web 1.0 interfaces [26]. According to Lupton [1], self-tracking “involves practices in which people knowingly and purposively collect information about themselves, which they then review and consider applying in their lives.”

As further pointed out by Lupton [1], in addition to self-tracking, there are several other terms—such as lifelogging, personal informatics, and the quantified self—used to describe the practices by which people may seek to monitor their everyday life. Lifelogging can be viewed as the practice of recording information about one’s life using digital tools. Personal informatics is a term used mostly in the academic human-computer interaction community [1]. The concept of the “quantified self,” originally coined by Wolf and Kelly in 2007, refers to “self-knowledge through numbers” [1]. They used quantitative data as a means or an embodiment of monitoring the elements of everyday life. The term can be further viewed as a collaboration between users and toolmakers [27], a cultural phenomenon involving technology [8], and an outcome of self-tracking [28]. In fact, the original intention of Wolf and Kelly was to use the term “the quantified self” to describe this digital self-tracking phenomenon [1].

According to the World Health Organization, health is explicitly linked with well-being, which is “a state of complete physical, mental, and social well-being and not merely the absence of disease or infirmity” [29]. With the development and diffusion of wearable, wireless communication, and cloud computing technologies, self-tracking devices can not only provide data, but also refine them into key performance indicators and produce visualizations of these data. People use self-tracking technologies to self-collect personal data including biological, physical, behavioral, or environmental information [30,31]. Moreover, according to Lupton (2016), the domain of self-tracking today also includes relationships and work productivity. In the scope of self-tracking or the quantified self, health and well-being are the main tracking domains. People who knowingly and purposively track their health and well-being information (such as heart rate, sleep, physical activity, calories, clinical symptoms, stress, and recovery) to review and modify their lives widely use self-tracking devices.

Self-tracking and the quantified self have been examined in various disciplines and from various theoretical and methodological premises. For example, the computer science research field has examined wearable augmented reality systems based on walking locomotion analysis [32] and developed algorithms for monitoring sleep [33]. The relevant literature also indicates certain design considerations, such as presenting negative data in a way that does not demotivate users [34]. Communication research has discussed topics such as self-tracking as a communicative phenomenon with social media, the self, and social networks of peers [35]. Furthermore, sociological studies have discussed ways of attributing meaning to data-gathering practices in terms of the quantified self [36]. Prior literature also includes research focusing on the societal and ethical concerns regarding self-tracking [37], including the value of personalized health care. In medicine, prior studies have explored the clinical experience of self-tracking technologies in the context of chronic diseases [38,39] and use of self-tracking devices in rehabilitation [40,41]. Finally, research on information systems has examined themes such as user acceptance of self-tracking, user motivation, and goal attainment related to self-tracking and the quantified self [42-44].

To keep the scope of the study manageable, we deliberately focused on self-tracking and the quantified self, executed via devices, apps, and platforms, and excluded digital diaries and video recordings to focus solely on the role of self-tracking in health and well-being. Moreover, we focused on empirical research where the primary research subjects are humans. Thus, purely technical papers and articles focusing on products, services, and markets were excluded.

Articles were obtained from five academic databases: Scopus, Web of Science, PubMed, Taylor & Francis, and the Association for Information Systems eLibrary. The search strategy included keywords such as “self-track*,” “self track*,” “quantified self,” “quantified-self,” “self quantif*,” and “self-quantif*.” We limited the search to journal and conference papers published in English. After conducting the literature search in the 5 databases, we eliminated duplicates. The details of the search strategy are shown in Multimedia Appendix 1.

Two authors (SF and MM) conducted the inclusion/exclusion procedure, prepared notes, and exchanged information throughout the process. Citation chaining was conducted to further reinforce the comprehensiveness of the article search. Textbox 1 presents the inclusion and exclusion criteria.

With respect to the studies that adopted the end users’ perspectives, user motivation constitutes an important stream of research. Most of the research examined the motivation for self-tracking usage [10-12,62,72]. In their research, Attig and Franke [10] divided the motivation for tracking physical activity into intrinsic and extrinsic motivations. The results revealed that people with strong extrinsic motivation may be highly dependent on the use of the tracking devices. Suh [62] classified motivation into hedonic, utilitarian, and eudaemonic. The results illustrated that unlike hedonic motivation, utilitarian and eudaemonic motivations positively impact continuance intention use of self-tracking technologies. Further, Pingo and Narayan [12] suggested that users may experience gratification when they can use tracking devices to control their health status. To clarify the reason for the usage of tracking devices, Gimpel et al [44] employed a survey to develop a five-factor framework of self-tracking motivations. The factors were self-entertainment, self-association, self-design, self-discipline, and self-healing. Gimpel et al [11] also illustrated that only the motivation for self-entertainment increases the wearable self-tracking device usage, which in turn influences the fulfillment of the self-entertainment motivation. Baudier et al [59] also emphasized that self-entertainment and self-design have a positive effect on the behavioral intention to use the health care Internet of Things in the workplace.

In prior research, goal setting and motivation always appeared at the same time. Gordon et al [13] found that users of weight loss–tracking apps preferred to choose achieving goals as their motivation. Self-monitoring motivation and attitude toward weight loss goals are vital for predicting goal achievement. For fitness tracking apps and exercise encouragement apps, Rockmann and Gewald [14], and Hamari et al [60] noted individual differences and found that different motivational affordances, such as self-quantification, gamification, and social networking capabilities, help users achieve their goals.

The second cluster under the end users’ perspective focused on the different uses of self-tracking and the outcomes of usage. With respect to the different uses and use cases, prior research has explored the use of self-tracking for physical activities [8,9,35,57,58,73-77], sleep [57,58,74,78], diet [57,58], fertility [55], caloric intake [79], and alcohol-harm reduction [80].

Prior literature also provides classifications of users and self-tracking tools. Makkonen et al [81] identified 4 distinct consumer segments of self-tracking based on their technology adoption patterns: pro-trackers, semitrackers, interested trackers, and nontrackers. In the context of app usage for alcohol consumption reduction, Milward et al [80] identified three types of users: the trackers, cut-downers, and non-committers. Spotswood et al [9] explored the role of self-tracking in supporting healthy behavior and found three mechanisms for teleoaffective shaping: labeling, rewarding, and materializing effort. Lyall and Robards [82] identified three roles of self-tracking devices for the user, namely as a tool, toy, and tutor.

Moreover, previous research has explored the outcomes of self-tracking. Shin and Biocca [63] noted that health consciousness is the most significant factor for staying healthy. Stiglbauer, Weber, and Batinic [67] proved that wearing a fitness tracking device can make users more conscious regarding their physical health but did not have a significant effect on their mental health. Furthermore, according to Ravichandran [78], sleep-tracking devices can help users better understand and improve their overall sleep habits. Rönkkö [74] proposed that certain design features such as graphical feedback, information sharing, and social communities in self-tracking devices may be particularly important in contributing toward positive lifestyle changes, with the most important factor being the presence of personal long-term goals.

The third cluster of studies from the end users’ perspectives focused on sustained usage of self-tracking and the quantified self. For example, Chuah [61] found that the perceived benefits (utilitarian, hedonic, social, and symbolic) indirectly affect users’ continuance intentions regarding smartwatch use through inspiration and well-being. Shin and Biocca [63] investigated the relative effects of hedonic and utilitarian motives as determinants of confirmation, satisfaction, and continuance intention regarding wearable devices. In contrast, Suh [62] observed that hedonic motivation has a negative influence on the continuance intention when using self-tracking devices, but utilitarian and eudaemonic motivations have a positive effect. As for web-based fitness community usage, Stragier et al [68] proposed that self-regulatory and social motives directly predict sustained web-based fitness community usage. Rockmann et al [83] also theorized an emotional “carry-over effect” in activity continuance decisions on activity tracking.

With respect to the long-term use of self-tracking and the quantified self, Hardey [84] employed a mixed-methods approach and found that visualization and a long-term healthy state motivate and support long-term tracking. In addition, Meyer et al [85] built 5 use cases for personal health devices that can be used as a long- and short-term comparison list. The five use cases are supporting health behavior, improved self-understanding, identification of trends and relations, decision-making, and data collection for future use.

The fourth cluster of studies from the end users’ perspectives emphasized managing personal data. This comprises two subtopics: coping with personal data and privacy concerns. Feng et al [86] showed that people preferred to use health or fitness apps as their personal health information management tools, and the most popular app was Fitbit. When reflecting on the personal data stored on self-tracking devices, Maltseva and Lutz [28] proposed that individuals who habitually use self-tracking apps and devices are more likely to self-disclose their personal data. When self-tracking devices are to social networks, the proportion of lower-performing friends may positively influence users’ physical activities [56]. Moreover, Yli-Kauhaluoma and Pantzar [87] examined the gap between individual experiences and self-tracking data. They found that individuals always feel upset and confused when comparing invisible or inaccurate personal health data with their daily experiences and may eventually refuse to use self-tracking devices. In another words, individuals may experience an emotional reaction, which can then stop changing their behavior [43].

Literature has also documented problems related to tracking, managing, visualizing, and using personal data [88]. Privacy issues are obviously a central concern related to sharing self-tracking data. Chen et al [89] examined the sharing of health data among college students. Their results implied that users are generally willing to share personal health data for research purposes, and the reasons for not sharing are related to privacy concerns. Gui et al [53] pointed out that when fitness devices are connected to social networks, users are encouraged to share fitness data. However, the challenges of balancing awareness and privacy issues were still prevalent. To solve this problem, Zimmer [90] employed the communication privacy management theory to better understand the privacy boundaries related to personal fitness information and found that the advantages outweigh the disadvantages.

The fifth cluster of studies from the end users’ perspectives relates to rejection and discontinuance. In their study of experiences during the implementation of self-tracking technology, Kari et al [42] found that the experience during the initial phase of implementation will influence the decision on adoption or rejection. Moreover, previous studies have found that the abandonment of wearable fitness tracking devices can be owing to a loss of motivation, low acceptance levels for such devices, or inaccuracy and uselessness [91]. Harrison et al [92] cited tracking accuracy and device aesthetics as barriers to activity tracking. Esmonde [93] identified the following 4 strategies of resistance used by people to datafication in fitness tracking practices: excessive labeling of some forms of data rather than tracking everything, choosing not to track every day, acknowledging that they cannot be perfect and track without interruption like a machine, and accepting that people’s feelings are more important than data.

The sixth subset of studies under the end users’ perspectives focused on user characteristics. McKinney et al [94] showed that in self-tracking of food and activity levels, high health literacy supports health goals, which comprises proper understanding of a record and how to use data to support a health goal and the awareness of privacy and ownership. The digital divide may also influence the adoption of self-tracking and the quantified self [95]. According to Régnier and Chauvel, better-off individuals use self-tracking more than the socially disadvantaged [95]. In addition, Baumgart conducted a series of studies to investigate the interaction of self-tracking with users’ cognitions, behaviors, and emotions [64-66].

In the clinical context, patients, and doctors need to track health records for symptoms or contemporary sensor data. The feasibility and acceptability of using tracking technology may differ from that of healthy individuals. Beukenhorst et al [96] focused on the feasibility and acceptability of smartwatches in patients with knee osteoarthritis; people expressed enthusiasm for self-tracking of health data, but there were barriers to full engagement, such as limited battery lives, technical issues, and unfulfilled expectations. Kim et al [54] conducted a food logger applicability test on patients, and their results demonstrated a high adherence rate. However, wrong and unreasonable usage of weight loss apps may contribute to and exacerbate eating disorders [97]. Self-tracking also demonstrated potential in managing chronic diseases and rehabilitation. Goal-directed self-tracking can help people be well prepared in all stages and support chronic condition management [98]. Mishra et al [99] researched how tracking apps and technologies helped patients cope with Parkinson disease. Furthermore, Vogel et al [40] provided evidence for the positive effects of self-tracking technology on a patient’s cardiovascular system.

Patient-generated data for clinical purposes have also attracted research attention. Ancker et al [39] examined personal data tracking for people with multiple chronic conditions, revealing that health-tracking data may come at an emotional cost, such as depression and anxiety, which could ultimately lead to low adoption levels for consumer health information technology. Ivanov et al [27] provided insights into the influencing factors for sharing health-tracking data. They found that health motivation, the severity of the health problem or condition, and age positively influenced patients to share data with professors. People who had a certain perceived health status preferred to share data with an acquaintance.

Literature contains specialized research on the advantages and disadvantages of self-tracking in the clinical context. With the development of tracking technology, the clinical context now provides more agency, control, and information for patients, which can help them establish a relationship with professionals [100]. Piras and Miele [38] supported this idea, stating that self-tracking mediates the patient–doctor relationship.

As for the disadvantages of self-tracking, collaborations between patients and professionals involving personal data can easily cause misunderstandings regarding the patient-generated data usage, duration, and type of tracked data. There are also patient privacy issues to address [101]. In terms of special disorders, such as Parkinson disease, Riggare et al [102] indicated that self-tracking gives people a deeper understanding of their motor or nonmotor symptoms and contributes to decision-making regarding their self-care. However, the tracking workload is heavy. It is difficult to know what and how to track, and it is also difficult to ignore the risks of obsessive tracking. Therefore, it is necessary to find a proper balance between burdens and benefits [102].

Self-tracking technology provides an easier way to collect patients’ health information. Health care professionals are also willing to accept self-tracking data to assess the health status of patients [39]. Therefore, patient-provider collaborations with respect to patient-generated data became one of the research topics for self-tracking usage in clinical setting. In a previous study, Chung et al [101] used a stage-based model of personal informatics and the theory of boundary negotiating artifacts to explain misunderstandings and privacy concerns in the collaboration stage. Prior research has also demonstrated that tracking apps and technologies helped patients cope with Parkinson disease. Based on the tracking data, health care partners who are friends and family members can significantly help Parkinson patients adopt positive strategies [99].

Tracking technology has also shifted the roles of patients and doctors. Owing to these technological changes, patients become health managers, and doctors are health organizers [103]. Schroeder et al [98] showed that goal-directed self-tracking can help people in setting goals, preparing knowledge, and contributing to the patient-doctor collaboration. Piras and Miele [38] supported this idea, stating that patients can negotiate a satisfactory relationship with their health care providers when using self-tracking. However, there is still a gap between users, experts, and self-tracking technologies [78]. To gain a professional view of self-tracking in clinical usage, prior research has collected data from doctors, such as in the study by Gabriels and Moerenhout [103]. They conducted an interview study to explore how medical doctors evaluate self-tracking methods and the changes after using those methods. Gabriels and Moerenhout found that regarding self-care, it is important to emphasize the contextual facets of self-tracking and the involvement of the health care professionals [103].

Changing habits typically takes time, and changes in people’s health often occur over time. Moreover, in different stages of usage, people may have different feelings or intentions, and engagement with their health data may change over time [106]. Prior research has extensively underscored the importance of this evolution over time and that of longitudinal research [28,59,91,99]. A longitudinal research design may also provide a better understanding of the causalities by observing and explaining changes over time [28,91]. Therefore, future studies should pay close attention to the evolution of the aforementioned aspects over time. A series of interesting topics were mentioned in prior research. First, the evolution of consumers’ acceptance level could be used to discover users’ attitudes and identify the barriers for continued usage [59]. Second, in clinical settings, changes in the roles of caregivers (friends and family members who manage the patients’ diseases) in helping patients’ self-tracking should be identified via longitudinal research [99].

As a second research direction, prior research has identified the need to shift focus from the mainstream users of self-tracking toward an increased emphasis on the modalities of use, namely intermittent users, nonusers, or prior users of self-tracking [14,53,68,95,102]. In extant literature, intermittent users comprise those who use self-tracking and self-quantification but not on a regular basis [95,99]. On the other hand, nonusers are those who are reluctant to adopt self-tracking technologies in all social milieus [95]. Former users are those who have used self-tracking or self-quantification in the past but have quit [68,95]. Focusing on the non-mainstream groups to identify and reduce the burden of self-tracking [102] as well as discovering ways to attract new users [14] are topics worthy of future research.

Issues related to sharing data generated by self-tracking, data privacy, and security have frequently been proposed as areas for future research [8,27,56,59,104,105]. Different experiences and concerns may promote different attitudes toward sharing personal data [104]. For example, Zhou et al [56] indicated that sharing self-tracking data may enable social comparisons. Furthermore, multiplatform and cross-platform data sharing has become a common phenomenon, which has consequently created challenges in terms of data privacy and security [27,105]. The tracking process of data collection, analysis, and storage also raises ethical issues [59].

Psychological and behavioral research encompasses themes such as users’ behaviors, attitudes, intentions, acceptance of technologies, and their respective motivations. The literature on motivation for personal health information management can provide new ideas to improve research design [86]. Existing research shows that future research will benefit from exploring the antecedents of the adoption decisions from a microlevel perspective [81]. Moreover, as the freemium business model is extremely popular among various web-based services and software firms [107,108], existing research highlights that future research could examine consumers’ willingness to make in-app purchases in the context of self-tracking and the quantified self [61].

In terms of independent variables, the subdivisions of intrinsic and extrinsic motivations should be considered in the research on usage [10]. In addition, the demographic characteristics, such as the educational level, socioeconomic status, and cultural background, could be examined as predictors of self-tracking usage [55]. Hamari et al [60] recommend that individual characteristics and personality types can be regarded as moderating variables of goal setting and the perception of motivational design.

With respect to the clinical apps of self-tracking, prior research has highlighted 2 specific areas for future research. The first direction involves health-tracking data and the patient-doctor relationship [38,100,101]. For example, future studies could explore the patient-doctor interaction with self-tracking in different clinical settings [101] and integrate the health-tracking data in the patient-doctor relationship [103]. Second, patients’ and doctors’ attitudes toward self-tracking also play important roles in clinical research. Patient usage experiments and the feasibility and acceptability of self-tracking should be investigated in the future [103]. As pointed out by Mishra et al [99], self-tracking can help patients communicate with clinicians. Thus, understanding clinicians’ communication needs and attitudes toward self-tracking represents another viable future research area.