We identified 1533 articles, removed 751 duplicates, and obtained 782 unique articles. After reviewing the titles and abstracts, we selected 110 articles to read in full. Following a review of the complete articles and application of our exclusion criteria, we finally selected 41 articles for analysis. Our selection process is illustrated in a flow chart (Fig. 1 - Flow chart of article selection process).
1- Description
Table 2 - General characteristics of articles shows the results of our analysis of 41 articles, with 6 published before 2017 and 36 published between 2017 and 2022. Of the articles reviewed, 33 focused on digital interventions for a specific target audience, with 12 of these specifically aimed at adolescents and young adults. The 11 articles that are not about a digital intervention for a target audience are thematic articles that don’t address a specific digital health intervention. In terms of health topics, 35 articles focused on a specific health issue, with prevention and/or health promotion being the most common topic in 6 articles, followed by AIDS in 5 articles, obesity in 4 articles and tobacco in 3 articles
Table 2
General characteristics of articles
First author
|
Year of publication
|
Study type
|
Target Public
|
Health topic or only on participation
|
Kwann P
|
2017
|
formative research
|
adult smokers of Polynesian, Melanesian and Micronesian origin
|
tobacco
|
Cole-Lewis H
|
2019
|
qualitative (presents a framework)
|
|
on participation
|
DeSmet A.
|
2016
|
meta analysis
|
|
promotion of healthy living
|
Da Silva C.C.
|
2018
|
book chapter
|
children, adolescents and young adults
|
nutrition
|
De Fatima Marin H.
|
2017
|
book chapter
|
|
on participation
|
Craig Lefebvre R.
|
2010
|
randomized controlled trial (RCT)
|
internet users aged 18–65
|
on participation
|
Chen Z.
|
2015
|
exploratory study
|
cancer survivors
|
cancer
|
Kim M.
|
2021
|
RCT
|
obese people
|
obesity
|
Corbett T.
|
2018
|
qualitative (synthesis article)
|
cancer survivors
|
cancer
|
Baltierra N.B.
|
2016
|
RCT
|
YBMSM/TW aged 18–30
|
health promotion
|
Fulton EA
|
2018
|
qualitative (explores a co-creation method)
|
smokers in the UK
|
tobacco
|
Alshurafa N
|
2018
|
qualitative (provides a framework)
|
students
|
health promotion
|
Hightow-Weidman L.B.
|
2021
|
qualitative (compares 8 strategies)
|
young people with HIV
|
AIDS
|
Graffigna G.
|
2020
|
descriptive article
|
Italian citizens
|
covid-19
|
Levine D.
|
2011
|
formative research
|
young adults with HIV
|
AIDS
|
Ronen K
|
2020
|
literature review
|
among affected peers
|
AIDS
|
Livingood W.C.
|
2017
|
qualitative CPR (community-based participatory research)
|
African-American adolescents
|
obesity
|
Van Bruinessen I
|
2017
|
descriptive article on co-creation
|
Older people
|
physical activity
|
Pernencar C
|
2018
|
qualitative (presents the results of a design study)
|
adolescents
|
obesity
|
Short CE
|
2018
|
descriptive article
|
|
on participation
|
Lawrence K
|
2021
|
RCT
|
pre-diabetic or at risk for diabetes in the US
|
diabetes
|
Wagner B
|
2017
|
qualitative (describes a framework and tests it)
|
people who want to lose weight
|
dietary intake and physical activity monitoring
|
Skinner H
|
1997
|
action research
|
young people
|
health promotion
|
van Hierden Y
|
2021
|
qualitative (presents the participatory design)
|
young adults
|
Well-being
|
Schroeer C
|
2021
|
scoping review
|
|
prevention and health promotion
|
Rai T
|
2021
|
3 linked qualitative studies
|
stroke patients
|
blood pressure management
|
Nurmi J
|
2020
|
descriptive article
|
adults
|
physical activity
|
Mustafa A.S.
|
2022
|
cross-Sectional Survey (cross-sectional analysis)
|
|
on participation
|
Morisson J
|
2014
|
mixed method
|
young people
|
mild intestinal disorders
|
Partridge S.R.
|
2018
|
descriptive article
|
adolescents
|
obesity
|
Mauka W
|
2021
|
qualitative (participatory design)
|
female Sex Workers (FSW) and Men who have Sex with Men (MSM) in Tanzania
|
AIDS
|
Solomon M
|
2012
|
RCT
|
adults with a chronic illness
|
health education\disease management
|
Nitsch M
|
2016
|
feasibility study based on a mixed method
|
people with body image concerns or eating disorder symptoms
|
eating disorders
|
Sucala M
|
2020
|
descriptive (describes a framework)
|
|
on participation
|
Laidlaw R
|
2017
|
action research
|
low income country population
|
on participation
|
Toefy Y
|
2016
|
qualitative
|
recent recipients of VMMC (voluntary medical male circumcision)
|
sexuality
|
Njie-Carr VPS
|
2018
|
qualitative with a phenomenological approach
|
older women with HIV infection
|
AIDS
|
Myneni S
|
2018
|
mixed method
|
smokers
|
tobacco
|
Musso M
|
2020
|
case study
|
|
health promotion
|
Milward J
|
2018
|
qualitative
|
young adults
|
alcohol consumption
|
Saleem M
|
2021
|
scoping review
|
|
mental health
|
Depending on the article, we were able to obtain some or all of the information regarding definitions and attributes of participation, scales and methods of assessment, techniques used to involve users, and facilitators and barriers to implementing participation (Table 3 - Information contained in the articles).
Table 3
Information contained in the articles
Reference and first author
|
Contains a definition or attributes of participation
|
Contains a scale of participation
|
Contains evaluation for participation
|
Contains methods for engaging
|
Contains facilitators for participation
|
Contains barriers to paticipation
|
Kwann P
|
|
|
|
X
|
X
|
|
Cole-Lewis H
|
X
|
X
|
|
|
X
|
|
DeSmet A.
|
|
|
|
X
|
|
|
Da Silva C.C.
|
|
|
|
X
|
X
|
|
De Fatima Marin H.
|
X
|
X
|
|
X
|
|
X
|
Craig Lefebvre R.
|
X
|
|
|
X
|
|
|
Chen Z.
|
|
|
|
X
|
|
|
Kim M.
|
X
|
|
|
X
|
|
|
Corbett T.
|
|
|
|
X
|
|
X
|
Baltierra N.B.
|
|
X
|
|
X
|
|
X
|
Fulton EA
|
X
|
|
|
X
|
|
|
Alshurafa N
|
X
|
X
|
X
|
|
|
|
Hightow-Weidman L.B.
|
X
|
X
|
|
X
|
|
X
|
Graffigna G.
|
|
X
|
|
X
|
X
|
|
Levine D.
|
|
|
|
X
|
|
|
Ronen K
|
|
|
|
X
|
|
|
Livingood W.C.
|
|
|
|
X
|
|
|
Van Bruinessen I
|
|
|
|
X
|
X
|
X
|
Pernencar C
|
|
|
|
X
|
|
|
Short CE
|
X
|
|
X
|
|
|
|
Lawrence K
|
|
X
|
|
X
|
|
X
|
Wagner B
|
X
|
|
X
|
X
|
|
|
Skinner H
|
|
|
|
X
|
|
|
van Hierden Y
|
X
|
|
|
X
|
|
|
Schroeer C
|
|
X
|
|
X
|
|
X
|
Rai T
|
X
|
|
|
X
|
X
|
X
|
Nurmi J
|
X
|
|
|
X
|
X
|
X
|
Mustafa A.S.
|
|
X
|
|
X
|
|
X
|
Morisson J
|
|
|
X
|
X
|
|
X
|
Partridge S.R.
|
|
|
|
X
|
|
X
|
Mauka W
|
|
|
|
X
|
|
|
Solomon M
|
X
|
X
|
X
|
X
|
|
X
|
Nitsch M
|
X
|
|
|
X
|
|
X
|
Sucala M
|
|
|
|
X
|
|
X
|
Laidlaw R
|
|
|
|
X
|
|
|
Toefy Y
|
|
|
|
X
|
|
X
|
Njie-Carr VPS
|
|
|
|
X
|
|
X
|
Myneni S
|
X
|
X
|
X
|
X
|
X
|
|
Musso M
|
X
|
|
|
X
|
|
|
Milward J
|
X
|
X
|
|
X
|
|
X
|
Saleem M
|
X
|
|
X
|
X
|
|
X
|
2- Engagement attributes
Currently, there is no widely accepted definition of participation (24), and there is a variability (28, 29) and a lack of clarity around the question of how to conceptualize engagement (30).
Firstly, there are different attributes depending on the discipline: according to scientists, participation is adherence and satisfaction, while for computer scientists, it corresponds to increased attention and enjoyment (30); in behavioral literature, participation is frequency or duration, while for the gamification industry, it is subjective experience with the service including affect, interest, attention and flow (21).
One recurring way of conceptualizing engagement is the duration and frequency with which the participant uses the eHealth intervention (30–32). In the literature, engagement is generally considered to be ease of use, measuring interactions with the features and functions of the digital solution (24). Engagement with digital health interventions can be defined by using quantity (e.g., total number of times the intervention was accessed), duration (e.g., total duration of use), frequency (e.g., patterns of use) and depth (e.g., content consumed during the intervention) (33).
However, when it comes to the frequency or duration of time spent using a digital service, some studies shown that low frequency of use can be just as effective as higher frequency (21) and an increase in usage time can signal low ease of use rather than attractive content (29). In the context of mobile health, engagement behavior is frequently assessed through passive measures of application usage (34), which fail to provide a comprehensive understanding of engagement with a digital platform (28). Additionally, it is important to focus on "effective engagement" rather than simply "increased engagement" (35). In two articles, the authors differentiate between engagement (meaning interactions with the tool) and effective engagement (sufficient participation in the intervention to achieve the desired effects) (29, 35). To achieve these desired effects, there is the need to determine an optimal duration and frequency of engagement (29). However, the concept of optimal engagement dose is not yet clear in the field of digital health interventions (35). Also, the concept of participation is often confused with compliance which refers to an individual following an instruction and using the intervention as intended by the developers (29, 30, 36).
In addition, to characterize participation, some studies focus on the user's experience, but they do not define experience in the same way (30). For some, user experience is about interaction (24, 28) or the desire of a user to use the application for longer and at frequent intervals. User experience quality is characterized by attributes of challenge, positive affect, endurance, aesthetic and sensory appeal, attention, feedback, variety/novelty, interactivity, and perceived user control (37) and includes sensual (aesthetic or novel elements that promote attention and interest), emotional (positive or negative effect elicited), and spatio-temporal (perceptions and awareness of time and the physical environment that the user experiences during use) categories (31).
Although authors assert that the subjective user experience can provide information about the degree of immersion in an application, it does not generally explain the cognitive engagement in the behavior change process. A health application can be entertaining and have good usability, but it does not necessarily lead the user to reflect on their behavior or to take action towards achieving their goals (21).
In one study, to define participation the aforementioned elements of usage and experience are combined. Engagement in digital interventions is defined through the measures of usage extent (e.g. quantity, frequency, duration, depth) and subjective experience (e.g. attention, interest, and affect) (35). In the field of digital health behavior change interventions, engagement should therefore integrate objective measures (defined based on usage habits measured using various tools to track the number of logins, time spent online, and amount of content used during the intervention period and also physiological measures using wearable sensors such as heart rate and electrodermal activity, as well as eye-tracking to determine psychophysiological measures) and subjective measures (defined based on self-evaluation questionnaires measuring levels of engagement with digital games and the intervention) (29, 35).
Finally, the literature includes some definitions focused on the objective to be achieved: health benefit (36, 38), changes for benefit (39) or behavior change (40). One source defines participation in eHealth and health behavior change as “the process of involving users in health content in a way that motivates and drives health behavior change” while specifying that this is influenced by a number of variables (35, 39–41) and is a complex process because it is evolving (41). Engagement itself is influenced by intervention characteristics such as content, delivery mode, and contextual characteristics such as physical environment and individual characteristics; there are also potential moderators and mediators of the engagement process (30). This multidimensional aspect is also mentioned in several articles, where engagement is described as a multidimensional process that involves, among other things, cognitive, emotional, and behavioral dimensions (32, 34, 37, 38). Thus, for evaluation, engagement is too complex to be evaluated by a single method, and a combination of several methods should be used (30, 42).
We have grouped together all of the information on engagement attributes identified in the articles in Table 4.
Table 4
Summary of information about engagement attributes
Attributes of participation
|
Elements
|
variety and disparities
|
no widely accepted definition, lack of clarity on how to conceptualize, attributes depend on the different disciplines
|
usage measures
|
duration, frequency, ease of use, quantity, dose (meaning optimal duration and frequency)
|
user experience
|
interaction, desire of user to use, experience quality characterized by different factors (see below)
|
usage and user experience combination
|
element of usage extent (for example quantity, frequency, duration, depth) + subjective experience (for example attention, affect, interest)
|
purpose to achieve
|
benefit for health, change to it benefit, for behavior change
|
multidimensional
|
evolving, complex, cognitive-emotional- behavioral dimensions
|
3- Scale and gradation of participation
The gradation of participation in articles is present in two different ways: either within the various stages of the process, or through the establishment of a gradient that characterizes a level of engagement, ranging from low to high.
The conceptual model in Yardley et al. distinguishes between the micro and macro levels of engagement. Micro engagement refers to the user's immediate interactions with the technology's features, including the extent of intervention use (e.g. number of completed activities) and the user's experience (e.g. level of interest and attention during activity completion specifically; linked to attention, interest, and affect). Macro engagement, on the other hand, refers to how the user engages with the overall goal of behavior change. After a period of effective micro engagement, the user may disengage from the platform while still being immersed in the behavior change process (30, 31, 33). Another similar method is that defining Big E as engagement in the targeted health behavior, and little e as engagement in the digital intervention for behavior change (24, 43).
Regarding scales of participation ranging from least engaged to most engaged, scales are identified in some articles, but they are not specific to digital health interventions. Firstly, the most common participation scale is Arnstein's 8-rung ladder of citizen participation, ranging from non-participation in citizen power through the stages of manipulation, information, consultation, advocacy, partnership, delegation, control, and self-management (44). Additionally, there are classifications in terms of usage quantity: an active user may post on a forum, for example, whereas a passive user may read an informative article (45). Alternatively, a heavy user is one who uses the intervention several times a day or 1–2 days per week, and a light user is one who uses the intervention twice per month or once per month, or who has not used it in the past month (46). There is also a 4-position psychological scale on a continuum ranging from minimum to maximum engagement (47), patient participation levels at 3 levels in the healthcare sector (36), 4-stage scale representing the developmental nature of patient activation (48) and others specifically designed questionnaires for a study (48, 49). Taki et al., on the other hand, suggest a mathematical indicator of engagement that uses five different indices to assess user engagement based on usage (29).
User typologies are also found. One study classifies users based on their alcohol consumption: "Trackers" monitor and track their alcohol consumption, "Cut-Downers" intend to reduce their alcohol consumption (similar to trackers, they not only used the monitoring features of the application, but also goal-setting and feedback functions), and "Non-committers" lack the motivation to use health applications (31). Another study categorizes users into three groups based on their frequency of engagement in peer interactions: "Conversation Starters" who initiated the highest number of threads, "Frequent Posters" who posted the highest number of messages on forums, and "Conversation Attractors" whose messages received the most responses (28).
We have grouped together all of the information that we found in the articles about engagement gradation in Table 5.
Table 5
Summary of information about engagement gradation
Scale and gradation of participation
|
Elements
|
within various stage of process
|
- big E or macro engagement:
interaction with technology’s features (usage and experience)
- little e or micro engagement:
engagement with overall goal of behavior (30,33)
|
level of engagement from low to high
|
not specific to digital health interventions, different scale depends on quantity or quality of usage
|
user typologies
|
specific of some interventions based on specific usage
|
4- Methods/techniques used to engage users
Many authors emphasize the need for theory-based interventions in order to engage users. A theory-based intervention is a systematic approach that uses theoretical models to understand how behaviors are influenced by factors such as beliefs, attitudes, and social norms. By applying these theories to the design of digital health interventions, designers can create programs that specifically target these factors, ensuring that they are based on solid data and effectively encourage user participation (33, 47, 50). These theories can be specific to health behavior change models or specific to digital interventions (33): derivatives of behavioral and neurocognitive theories and models such as the Theory of Planned Behavior, Social Cognitive Theory, Self-Regulation Theory, and Social Learning Theory (50).
Similarly, the use of appropriate behavior-change techniques can also increase user engagement and help them maintain healthy behaviors in the long term. This can improve the overall user experience and make digital health interventions more effective. "Behavior-change techniques" (BCTs) are the smallest active components of an intervention that are capable of changing behavior. However, the author claims that they depend on their design, dose and duration, as well as their content, to optimize their effectiveness. The most advanced and complete BCT would be useless if it did not generate interest through an attractive design, or if it was not user-friendly and intuitive to use in terms of digital functionalities (37).
In addition, one of the most widespread methods for engaging users is user involvement in the intervention design, a process that takes different names such as user-centered design (39, 43, 51, 52), co-creation (53–55), or participatory design (47, 56). It is important to involve end-users from the beginning of the process to ensure that the intervention is relevant and accepted by users in order to trigger effective participation afterwards (46, 57, 58). Users were also invited to implement and/or evaluate the intervention (43, 59, 60). Some authors mentioned that the process is spiral and works in a loop: ideas are generated by/with the community, implemented by the community, evaluated by the community, adjusted based on community feedback, then implemented and evaluated (42, 43, 52, 61).
Apart from being used for co-design, personalization is a technique used on its own to engage users (57, 62). Some articles mention the need to gather individuals' needs to tailor and personalize interventions in order to engage end users (32, 34, 35, 47, 57, 63) for example in adapting content by personalizing it according to variations in psychological, social, and behavioral profiles (33) or based on user typologies determined during application registration (for example, through a short questionnaire and targeting the application's content according to their typology) (31). The overall goal is to provide instant contextual support for targeted behaviors when the individual is most likely to be receptive. Just-in-time adaptive interventions could use sensory data, for example, a smartphone or smartwatch, and momentary information directly from participants (57).
Thus, autonomy is also one of the techniques employed to engage users (34). In a structured and self-directed behavior change intervention, all participants receive the same content but each individual is encouraged to choose their own behavioral goals and activities (48, 64). In order to provide specific recommendations based on individuals' needs, personalized feedback can be used (32, 33, 49).
Regarding users, some authors also suggest using online peer communities to engage users as they provide resources for sharing experiences and achieving better outcomes. These digital health communities may include discussion forums, online support groups, and social networks for patients (32, 34, 36, 42, 44, 62, 65, 66). This is believed to be particularly important for young people and marginalized populations as it provides an opportunity to address crucial behavior change issues (66) and allows social interaction, feeling supported, and understood by the community (28, 33, 63) to enhance participants' engagement in the program and foster their motivation (32), mentioned as a key to engaging users (21, 34, 37, 41). Regarding this, despite evidence of improved intervention engagement in physical interventions, motivational support such as motivational interviewing is rarely present in health apps for smartphones. One author suggest that the reason for the limited use of motivational interviewing may be its lack of a coherent theoretical framework and proposes that self-determination theory can be used as a theoretical basis because it shares fundamental principles with motivational interviewing (21). Self-determination theory, employed by several authors, emphasizes autonomy, competence, and social relatedness to encourage user participation (21, 42, 62).
While motivational interviewing provides tools for self-reflection and satisfaction, gamification can provide experiences of autonomy, competence, and relatedness by adding fun and excitement to activities (21, 33, 35, 44, 53, 56). Gamification can enhance intrinsic motivation by addressing the psychological and emotional requirements of individuals through the use of game-like components. Intrinsic motivation refers to performing an activity solely for pleasure, excitement, and interest. Authors assert that gamification of health applications is a promising approach to counter the often-decreasing long-term motivation of health application users (46).
We grouped all of the information that we found in the articles about methods used to engage users in Table 6.
Table 6
Summary of information about methods for engage users
Methods and techniques to engage users
|
Elements
|
based on theories and models
|
theories specific to health behavior change models or specific to digital interventions
|
BCT
|
behavior change techniques: design, dose, duration, content, interactions between them
|
user-centered
|
Interventions create and/or implement and/or evaluate by the target population
|
personalization
|
adaptation of content based on different factors (see below)
|
autonomy
|
users choose own behavioral goals and activities
|
using peers/ social support
|
discussion forums, online support groups, and social networks
|
motivation
|
motivational support, motivational interviewing
|
gamification
|
game-like components
|
5- Difficulties in engaging users
Authors encountered several barriers to user engagement in digital interventions. At the country level, in several countries and populations, technological adoption has been slow due to cost, infrastructure, design models, architecture, integration, usability and the implementation of public policies. Adoption and deployment are also dependent on the training and education of users provided by authorities (36).
At the intervention level, difficulty in involving all populations in participatory design is encountered due to knowledge asymmetry (60). Lack of user involvement in the design and a lack of desired features then leads to reluctance to adopt the intervention (46, 62, 63). In the long term, maintaining usage is also a recurring problem (35, 46, 61): there is evidence indicating that people often download an application and never use it again, with a rapid decline in usage after the first download (61) and there is limited research on the specific reasons for user abandonment and low engagement in online programs (32). Authors point out the lack of knowledge on the characteristics, or combinations of characteristics, that can optimize the effect of an intervention on user engagement and health-related outcomes (49). Several articles included in the review are either the first or among the few in their respective fields (31, 41, 59) for a specific health topic. Regarding the available data, studies highlight the lack of evidence-based information and the heterogeneity of data (32, 35, 44, 48, 49) for research purposes. This heterogeneity of data on engagement can render the results incomparable and hinder the understanding of the effectiveness of engagement strategies (35) and is linked to the difference in the definition of engagement, often defined as usage and assessed using quantitative methods (39, 45, 46). It is important to have a clear understanding of engagement before claiming that the engagement strategies used by these applications are effective (35).
Regarding the content of interventions, there is a lack of theory-based mobile health interventions, despite the fact that some behavior change theories have been well-validated and tested in evidence-based prevention, diagnostic, and care interventions (58). As for interfaces, some studies emphasize that interfaces are non-user-friendly (44) and that it is challenging to compete with social media and other entertainment-based applications (33, 57). As for all other online tools, data security is also a recurring concern (44). Once the results of an intervention are obtained, the literature mentions the difficulty of generalizing the results to other populations due to differences in context. What works in one context for a given subgroup of the population may be less effective or even harmful elsewhere for other subgroups (39).
At the user level, technical barriers and a lack of technological skills are cited as a hindrance to participation (43, 44, 63). Also, improving health literacy directly impacts the ability to act on health information and take greater control of health as individuals, families, and communities, and is considered an essential condition for patient participation (36, 38). Additionally, lack of support (44, 63) and lack of motivation were associated with low adoption of mobile health applications (46).
We have grouped together all of the information that we found in the articles about difficulties in engaging users in Table 7.
Table 7
Summary of information about difficulties in engaging users
Difficulties to engage users
|
Elements
|
accessibility
|
cost, infrastructure, design models, architecture, integration, usability, and the implementation of public policies
|
adoption
|
desired features of users when designing apps
|
literacy of technology and health
|
knowledge asymmetry in design group, training and education of users, technological skills of users
|
maintain participation
|
rapid decline in usage after the first download
|
non-theory-based conception
|
designing apps based on modus operandi but no based in validated theories
|
non user friendly
|
interfaces are non-user-friendly, difficulties to compete with social media and other entertainment-based applications
|
data security
|
users afraid of the security of their data
|
6- Facilitators and suggestions to engage users
At user level, engagement depends on several interdependent components such as social norms, positive and negative reinforcement factors, goal-setting, self-monitoring, self-evaluation (24, 50) social support (28, 50, 64), trust and adherence of the participants' family and peer modelling (39).
Regarding intervention features, gamification (21, 28, 37, 46, 56), accessibility, and ease of use (35), interactivity (64), attractiveness (53, 57), confidentiality and security of personal data (35), adapting the material for potential users from socially or educationally disadvantaged backgrounds (39) all help users to engage and are avenues to explore to increase user engagement within digital health interventions.
As regards the design of the interventions, results show that involving users at all stages (21, 42), addressing basic psychological needs (21) including psychosocial determinants anchored in relevant behavioral theories, such as the social ecological model, the COM-B (Capability, Opportunity, Motivation - Behavior) model, and social cognitive theory (24), using more systematic and progressive approaches for intervention development and evaluation (49), integrated skill acquisition, positive reinforcement, integration of cultural values and group norms (50) could be explored. One study suggests letting "people work on their own problems in their own way within a structured and supportive framework" (64).
We have grouped together all of the information found in the articles about facilitators of engagement in Table 8.
Table 8
Summary of information about facilitators of engagement
Facilitators to engage users
|
Elements
|
multifactor
|
social norms, positive and negative reinforcement factors, goal-setting, self-monitoring, self-evaluation, social support, trust and adherence of the participants' family and peer modelling
|
developing functionalities
|
gamification, accessibility, and ease of use, interactivity, attractiveness, confidentiality and security of personal data, adapting the material for potential users from socially or educationally disadvantaged backgrounds
|
focus on conception elements
|
involving users at all stages, addressing basic psychological needs, using more systematic and progressive approaches, integrated skill acquisition, positive reinforcement, integration of cultural values and group norms
|
All of our results are summarized in Fig. 2- All elements that affects users’ engagement in digital health interventions