Do Individual Characteristics and Social Support Increase Children's Use of an MHealth Intervention? Findings from the Evaluation of a Behavior Change MHealth App,Aim2Be

Abstract

Purpose:

Mobile health (mHealth) apps may support improved health behavior practice among youth living in larger bodies. However, long-term use is low, limiting effectiveness. This study evaluated whether youths' motivation, satisfaction, engagement with social features, or parent co-participation supported long-term use of an app named Aim2Be.

Methods:

A secondary analysis of two versions of Aim2Be (preteen and teen versions) using covariate-adjusted multivariable regression was conducted. We evaluated associations between social support features (a virtual coach, a social poll, or a social wall), parent co-participation (time spent in the parent app), and app satisfaction on use (time spent in Aim2Be). Models were stratified by age and satisfaction was explored as a moderator.

Results:

Preteens (n = 83) engagement with the social poll (β = 0.26, p < 0.001), virtual health coach (β = 0.24, p = 0.01), app satisfaction (β = 0.31, p = 0.01), and parent co-participation (β = 0.24, p = 0.01) predicted use. In teens (n = 90), engagement with the virtual coach (β = 0.31, p < 0.001) and full utilization of social wall features (β = 0.41, p < 0.001) predicted use. Furthermore, satisfaction moderated the effects of partial utilization of the social wall among teens (β = 0.32 p = 0.02).

Conclusion:

Social support in mHealth apps may impact users differently depending on age. Features that include health professionals or peers may be more advantageous across ages. App developers should consider age when designing interventions. Clinical Trial Registration NCT03651284

Introduction

Thirty percent of Canadian youth (6–18 years) self-report living in a larger body.¹ The gold standard to support these youth is through in-person family-based interventions that promote behavior changes.^2–4 These interventions elicit significant but modest effects on health outcomes.^5,6 This could be partially attributed to barriers participants face in accessing clinical sites,^5–8 stigma associated with attendance,⁸ or maintenance of behaviors after interventions end.^8,9 Mobile health (mHealth) interventions that take place on personal devices (i.e., apps) may help address these barriers as apps¹⁰ are highly accessible^10–12 and can be anonymous.^12,13

A recent meta-analysis of 10 randomized control trials found that apps had positive impacts on improving insufficient activity levels among teens (14–18 years).¹⁴ Additional meta-analyses have found small but significant impacts on health behavior change,¹⁵ and weight outcomes.¹⁶ However, reports with inconclusive results and no change in health behaviors have also been published.⁷ Typically, initial download and enrollment of apps is high,¹⁷ but continued use is low.^11,17–20 For example, in a trial evaluating teens' use of an mHealth intervention, only 21% continued logging on after 4 weeks.¹⁹ Lack of long-term use and engagement with active ingredients meant to promote behavior changes may contribute to null impacts. Therefore, understanding how to promote continued use of apps may improve their effectiveness.^13,20

The Player Experience and Need Satisfaction (PENS) model,²¹ an extension of self-determination theory (SDT),²² suggests motivation and satisfaction independently predict, in part, mHealth use.^11,13,23 Interestingly, in the context of mHealth, extrinsic motivation may be used to promote use and can be used to engage those who are less intrinsically motivated to change their health behaviors.^23–25 This opposes what behavior change theory posits as intrinsic motivation (a form of internally derived motivation) is expected to have a greater effect on behavior change as compared with extrinsic motivation.²⁶

Many apps include social support features (including peers,²⁵ parental,²⁵ or health professionals¹³ interactions) that rely on extrinsic motivation to support use. For example, peer-based features can promote peer norms to adopt new behaviors promoted in apps.²¹ Meta-analyses also suggests parent co-participation in mHealth interventions improve youth health outcomes.¹⁵ Finally, engagement with health professionals can encourage trust in app messaging,^23,25 which may in turn impact motivation to follow advice.¹³ Although most research agrees that app features need to be tailored to their audience,¹⁴ there is little agreement on how to do this among youth.^13,23,25,27

The role of satisfaction on app use is poorly understood in the literature. According to the Technology Acceptance Model (TAM), app satisfaction is determined by a users' evaluation of how easy or useful an app is at fulfilling initial motivation and/or personal goals.^28,29 Satisfaction could have a direct association on use or act as a moderator.^13,23 For example, presence of peer-based features increased teens' satisfaction with an app in one study.²³ However, another study suggests that peer-based features hinder use because other channels such as social media already fill this niche.¹³ It is possible that peer-based features do have a positive impact on behavior change, but this likely depends on a user's level of satisfaction in using certain features (i.e., perceived ease of use or usefulness). Furthermore, age may also impact app effectiveness¹⁴ as the appeal of different features may vary with the age of the user.^13,23

Despite the fact that youths' use of apps rapidly decreases with time,^11,17–20 few studies have examined the factors that impact use of an app.^30,31 Understanding how youth can be encouraged to use apps is critical as previous studies of Aim2Be have linked change in behaviors with overall app use³² and specific features of the app.³³ To address this gap, this study evaluated if preteens’^10–12 and teens’^13–17 initial motivation, engagement with different social features (peer-based features and a virtual health coach) and parent co-participation (time spent in a companion app) predicted use of an mHealth app (namely Aim2Be). Furthermore, the potential role of satisfaction as a moderator was explored to help clarify its role in app use.

Methods

Study Design

This study is a secondary analysis of data from a randomized control trial that evaluated the effectiveness of Aim2Be.³² The study protocol was approved by the Children's and Women's Research Ethics Board at the University of British Columbia (H16-03090/H17-02032), the Health Research Ethics Board at the University of Alberta (Pro00076869), the Hospital for Sick Children Research Ethics Board (REB1000059362), the Hamilton Integrated Research Ethics Board (Project #4250), and the Children's Hospital of Eastern Ontario Research Ethics Board (18/01E). The trial was registered in August 2018 and data were collected from February 2019 to June 2020.

Study Participants

Participants (n = 214 parent–child dyads) were recruited from six clinical sites across Canada and through targeted Facebook ads. To be eligible, youth had to be 10–17 years old, living in a larger body based on international cutoffs,³⁴ and read at a minimum grade five level. Parents had to be the primary caregiver of their participating child and proficient in English. Additional details surrounding study eligibility can be found in the published protocol.²⁰ To be included in this analysis, youth had to be enrolled into Aim2Be (n = 173 dyads).

Study Protocol

Recruited participants who provided telephone consent were e-mailed a baseline questionnaire. Once complete, families were randomized into the intervention (immediate app access) or the waitlist control (access after 3 months) groups. Follow-up assessments were conducted at 3 and 6 months. This study evaluated the first 3 months of app use from either group (intervention group from baseline to 3 months, control group from 3 to 6 months). Both groups received an identical version of Aim2Be except that the intervention group received access to a “live health coach” feature.

Aim2Be Intervention

Details of the Aim2Be app are described elsewhere.²⁰ In brief, Aim2Be is a gamified lifestyle intervention rooted in theory to encourage youth's adoption of health behaviors (improved eating, physical activity, screen time, and sleep habits). The app integrated elements from Social Cognitive Theory (SCT),³⁵ the PENS model (a gamification-based extension of SDT)^36,37 and the ACUDO (Agency, Discovery, Uncertainty, Challenges, Outcomes) framework.^32,38 Using these, Aim2Be developed features to encourage behavior change (i.e., setting aims, self-monitoring, completing tasks, and health literacy resources), including support through social interactions (i.e., peer-based social wall/poll, automated coach, and live coach) or gamified elements (i.e., quizzes, story progression, and currency collection).³⁹ These features were intended to inform users about how to make incremental changes to their behaviors (e.g., walking for 10 minutes after dinner).

Social features differed for preteens (10–12 years) and teens (13–17 years). The preteen version included a voting feature called the “social poll.” Preteens answered “yes or no” to app prompts (e.g., “Do you prefer running or walking?”). Once answered, preteens could view other preteens' votes. The teen version contained a “social wall” where, similar to the preteen version, teens could vote and view the responses of others. However, teens could also post full responses to the prompts, add reactions (e.g., emojis) or comments. Posting full responses, reactions, or comments was unique to the teen version and facilitated by the presence of open-ended prompts (e.g., “What do you experience more often? Peer expectations or adult expectations?”). As part of the teen intervention, “mature teens” were hired in the background of Aim2Be to post responses on the social wall every week.

Both child app versions were accompanied by a companion parent app. The parent app also included behavioral, social, and gamified components. It further encouraged parents to make changes in the home environment to support their child's adoption of behaviors promoted by Aim2Be.

Study Measures

Social support was measured using three components: (1) parent co-participation (total minutes spent in the companion app), (2) engagement with peer features (wall/poll), and (3) health professionals (“live coach”). Among preteens, peer support was dichotomized as “did not vote” or “voted” in the social poll. Owing to high multicollinearity in social wall variables among teens, an additive categorical variable was created. These categories were as follows: “no engagement,” “viewed only,” “partial utilization” (users viewed and either posted or reacted), and “full utilization” (users viewed, commented, and reacted). Interactions with the live coach were dichotomized into “did not use” (including participants from the waitlist control group without access, and those from the intervention group with access who did not use) and “used” (those with access who used). This was done due to the small number of participants who engaged with the live coach in the intervention group. All data were collected objectively using app-analytics.

Children's motivation to change their health behaviors (diet, physical activity, and screen time) was evaluated at baseline with 12 Likert style items (1 = strongly disagree to 5 = strongly agree) adapted from the Self-Regulation Questionnaires.⁴⁰ Two items assessed intrinsic motivation (i.e., “I simply enjoy participating in…” and “It is personally important to me…”) and two items assessed extrinsic motivation (i.e., “Others would be mad at me if I didn't…” and “I would feel bad about myself if I didn't…”) for each health behavior. A summary variable for each type of motivation was created with higher mean scores indicating greater identification with that type of motivation. In our sample, the scale for intrinsic motivation across all three health behaviors had a Cronbach alpha of 0.82 in preteen and 0.75 in teens (extrinsic motivation alpha values of 0.64 and 0.72, respectively).

Children's app satisfaction was assessed after 3 months of app use using five items developed for this study (“I enjoy using Aim2Be,” “I feel Aim2Be provides a safe environment for me to share information,” “I find it easy to navigate through Aim2Be,” “I find the information in Aim2Be is relevant to me” and “I look forward to the next time I can use Aim2Be”). In our sample, the Cronbach alpha value of this scale was 0.80. Responses were measured using a range (1 = strongly disagree to 5 = strongly agree) and split at the median into low and high satisfaction. Participants who used the app and did not report their satisfaction were coded as missing.

Children's app use was defined as total minutes spent in the app for 3 months. This variable was measured using app-analytics and log transformed to address skew.

Data Analysis

Univariable regression models were used to build a multivariable model.⁴¹ Baseline motivation for health behaviors, app satisfaction, use of social features, or parent time spent in the companion app on app use were all explored. To be retained for the final multivariable model, a p-value <0.10 was needed. Next, app satisfaction was explored as a moderator by adding an interaction term with each independent predictor (p < 0.10 required for retention). In the final models, p-value >0.05 indicated significance. All models controlled for child sex, parent sex, age, and educational attainment. Models were stratified by app version. All analyses were conducted in Stata.⁴²

Results

Baseline demographic differences and study variables between preteen and teens can be found in Tables 1 and 2, respectively. Parents of preteens were younger on average than teens (p < 0.001; Table 1). Preteens also reported higher intrinsic (p = 0.02) and extrinsic (p = 0.03) motivation (Table 2).

Table 1.

Demographic Details of Study Participants

	Preteens (n = 82)	Teens (n = 91)	Difference^a
	n (%) or mean (median)		p
Parents
Age (years)	42.4 (42.0)	45.99 (46.5)	<0.001
Sex			0.65
Male	5 (6.0)	7 (7.8)
Female	78 (94.0)	83 (92.2)
Income			0.36
<$100,000	38 (45.8)	47 (52.5)
≥$100,000	36 (43.4)	33 (36.7)
Education			0.68
<Bachelors	44 (53.0)	50 (55.6)
≥Bachelors	39 (47.0)	39 (43.3)
Ethnicity			0.38
White	56 (67.5)	54 (60.0)
Other^b	10 (12.1)	7 (7.8)
Mixed^c	8 (9.6)	11 (12.2)
Black	1 (1.2)	5 (5.6)
Chinese	4 (4.8)	2 (2.2)
Aboriginal^d	2 (2.4)	3 (3.3)
Youth
Age (years)	10.9 (11)	14.6 (15)	<0.001
Sex			0.41
Male	43 (51.8)	41 (45.6)
Female	40 (48.2)	49 (54.4)
BMI (z-scores)	2.7 (2.8)	2.9 (2.8)	0.19

Difference calculated using t-tests, chi-squared, or Fisher's exact tests. Not all values add to 100 due to missing data.

Includes South Asian, Japanese, and free text response.

Includes two or more of the aforementioned groups listed.

Includes North American Indian, Métis, and Inuit.

SE, standard error.

Table 2.

Differences in Study Variables Between Preteens (n = 82) and Teens (n = 91)

	Preteens (n = 82)	Teens (n = 91)	Difference^a
	n (%) or mean (median)		p
Children's app use (minutes)	92.4 (10.2)	72.2 (14.2)	0.26
Child motivation (1 = low; 5 = high)
Intrinsic	3.5 (3.5)	3.2 (3.3)	0.02
Extrinsic	3.3 (3.3)	3.1 (3.2)	0.03
Social support
Parent co-participation (minutes)	72.4 (44.9)	72.4 (45.0)	1.00
Social app feature engagement^b			—
Did not use	30 (36.1)	29 (32.2)
Social poll: Voted	53 (63.9)	—
Social wall: Viewed only	—	25 (27.8)
Social wall: Partial utilization	—	20 (22.2)
Social wall: Full utilization	—	16 (17.8)
Child live coach engagement			0.18
Did not use or have access	54 (65.1)	67 (74.4)
Used	29 (34.9)	23 (25.6)
Child app satisfaction			0.86
Low	11 (13.3)	14 (15.6)
High	40 (48.2)	47 (52.2)

Difference calculated using t-tests, chi-squared, or Fisher's exact tests. Not all values add to 100 due to missing data.

Classification into each respective group for social poll/wall engagement was determined as used vs. did not use.

Univariable and multivariable models exploring the associations between predictors and app use are displayed in Table 3. In univariable models, higher satisfaction predicted greater use (preteens: $β$ = 0.31, p = 0.01 and teens: $β$ = 0.45, p < 0.001). Engagement with social features, including the live health coach (preteens: $β$ = 0.53, p < 0.001 and teens: $β$ = 0.44, p < 0.001), social poll (preteens: $β$ = 0.54, p < 0.001), and social wall (teens: partial utilization: $β$ = 0.31, p < 0.001; full utilization: $β$ = 0.63, p < 0.001) also predicted use. Parent co-participation was associated with greater use in preteens only ( $β$ = 0.47, p < 0.001). Finally, app satisfaction was significantly associated with teens' partial utilization of the social wall features ( $β$ = 0.33, p = 0.03) and had a borderline association with full utilization ( $β$ = 0.29, p = 0.09) of the social wall features.

Table 3.

Associations Between Motivation, Social Support, Parent Co-Participation, and App Satisfaction on Preteens' (n = 82) and Teens' (n = 91) App Use

	Preteens (app use)		Teens (app use)
	Univariable model	Multivariable model	Univariable model	Multivariable model
	β (p-value)	β (p-value)	β (p-value)	β (p-value)
Covariates
Child sex^a (Female)	0.05 (0.68)	−0.02 (0.76)	−0.07 (0.54)	−0.02 (0.74)
Parent sex^a (Female)	0.11 (0.33)	0.09 (0.26)	0.16 (0.12)	−0.04 (0.63)
Parent age	−0.14 (0.23)	−0.08 (0.34)	−0.25 (0.02)	−0.03 (0.67)
Parent education^b (Bachelors or more)	0.09 (0.46)	0.09 (0.29)	0.08 (0.43)	0.12 (0.13)
Child intrapersonal factors
Intrinsic motivation	0.02 (0.87)	—	0.09 (0.41)	—
Extrinsic motivations	−0.10 (0.40)	—	0.07 (0.49)	—
App satisfaction^c (high)	0.31 (0.01)	0.35 (<0.001)	0.45 (<0.001)	−0.04 (0.81)
Social support factors
Live coach engagement	0.53 (<0.001)	0.24 (0.01)	0.44 (<0.001)	0.31 (<0.001)
Parent co-participation	0.47 (<0.001)	0.24 (0.01)	0.21 (0.05)	0.06 (0.47)
Social poll: voted	0.54 (<0.001)	0.26 (<0.001)	—	—
Social wall: Viewed only	—	—	0.16 (0.11)	0.06 (0.66)
Social wall: Partial utilization	—	—	0.31 (<0.001)	0.02 (0.87)
Social wall: full utilization	—	—	0.63 (<0.001)	0.41 (<0.001)
Moderators
App satisfaction X Intrinsic motivation	0.70 (0.33)	—	0.13 (0.82)	—
App satisfaction X Extrinsic motivation	−0.37 (0.54)	—	0.27 (0.59)	—
App satisfaction X Live coach engagement	0.21 (0.23)	—	0.09 (0.56)	—
App satisfaction X Parent co-participation	0.09 (0.58)	—	−0.09 (0.58)	—
App satisfaction X Social poll: Voted	0.20 (0.44)	—	—	—
App satisfaction X Social wall: Viewed only	—	—	0.24 (0.12)	0.21 (0.12)
App satisfaction X Social wall: Partial utilization	—	—	0.33 (0.03)	0.32 (0.02)
App satisfaction X Social wall: Full utilization	—	—	0.29 (0.09)	0.25 (0.11)
Total adjusted R²	—	0.54	—	0.57
Incremental adjusted R^{2 d}		0.54		0.51

$β$ = Standardized regression coefficients. All models included the covariates. In preteen models, parent education was significant ( $β$ = 0.20, p = 0.05) in the univariable model for parent engagement only. In the teen models, parent age was a significant covariate in the extrinsic motivation ( $β$ = −0.25, p = 0.02) and intrinsic motivation ( $β$ = −0.23, p = 0.03) univariable models. X, interaction term entered in the model.

Reference group: Male.

Reference group: less than a bachelor's degree.

Classification into each respective group for social poll/wall engagement were determined as used vs. did not use (reference group).

Incremental adjusted R² = total adjusted R² − covariates adjusted.

In the multivariable models, higher satisfaction ( $β$ = 0.31, p = 0.01), live coach engagement ( $β$ = 0.24, p = 0.01), parent co-participation ( $β$ = 0.24, p = 0.01), and voting in the social poll ( $β$ = 0.26, p < 0.001) remained predictive of preteens' use. Among teens, interactions with the live coach ( $β$ = 0.31, p < 0.001) and full utilization of the social wall ( $β$ = 0.41, p < 0.001) remained significant. App satisfaction remained a moderator of the association between partial utilization of social wall features and app use ( $β$ = 0.32, p = 0.02). The explanatory variables in the final models explained 54% and 51% of the variance of preteens' and teens' use, respectively.

Discussion

This study investigated factors that predicted preteens' and teens' use of Aim2Be, an mHealth app. We found that use of social features, including interactions with a health professional (live coach) or peers and parent co-participation increased use. Although greater use of social features does relate to greater time spent in an app, relationships differed by users' age (i.e., parent co-participation significant in preteens only and satisfaction as a moderator in teens only). These results imply that providing social support within apps likely impacts overall app use but could have different effects based on age or satisfaction.

Similar to other studies,^25,31,43,44 we found that users' engagement with a live coach or peer-based social features promoted use. In a 2017 qualitative study, adolescent girls said they preferred physical activity apps that included a peer component.⁴⁴ In a recent meta-analysis, face-to-face counseling with health experts was advantageous in internet-based self-monitoring interventions.¹⁶ However, not all apps incorporate peers or health professionals in the same way. This diversity makes comparisons across studies difficult⁷ and instead, may point toward individual tailoring as a more effective route to promote use.¹⁴

Parent co-participation within Aim2Be had positive effects only on preteens' app use. This is not surprising as teens often seek greater autonomy from their parents, relying on them less than preteens.^45,46 However, parents do still play a central role in teens' behaviors,^5,8,47,48 and other literature has found merit in a whole family approach⁸ or parent co-participation^7,16 among teens. Discrepancies may have to do with differences in how parents are involved within different interventions. As Aim2Be had separate parent and child app, future studies should explore whether an interactive parent–child app may promote teens' continued use.

There is agreement in the literature that app satisfaction in adults relates to intention to use mHealth apps.^28,29,49 However, the mechanisms are not well understood in youth. In our study, preteens' and teens' use was positively influenced by satisfaction, but in different ways. Among preteens, higher satisfaction remained an independent predictor of use. In contrast, teens' satisfaction moderated the association between overall use and partial use of the social wall. This suggests that preteens and teens may enjoy using different app features, influencing overall satisfaction in different ways. A quantitative study considering barriers and facilitators to teens' (12–18 years) mHealth adherence found that 17- to 18-year-olds wanted personalized e-mail feedback, and awards/rewards for achieving goals increased their adherence. These features were not suggested by 12- to 16-year-olds.¹³ Older age groups or teens may prioritize social conformity more than preteens.⁴⁹ This may explain why satisfaction acted as a moderator on the relationship between the social wall and overall app use only among teens. Understanding how different age groups view app satisfaction is key for designing more engaging apps.⁵⁰ As the literature surrounding app satisfaction and how it impacts use is understudied, future study should expand on our findings to better understand how preteens and teens determine app satisfaction.

In this study, intrinsic and extrinsic motivation for health behaviors did not predict use. This aligns with other studies where motivation related to intentions for app use, but not actual use.^20,28,29,49 For example, among 165 Portuguese teens (12–18 years) the main reason for stopping use was lack of interest regardless of many initial motivators being reported.¹³ It is possible that app satisfaction plays a larger role in adherence than initial motivation.^11,13,23,51 Other app elements such as design or gamification^23,31 may also be more advantageous in fostering app use and should be investigated.

This study is not without limitations. First, our use of a clinical population limits generalizability to other groups. Second, we evaluated satisfaction and motivation with self-reported measures that are subject to social desirability. Third, our sample size was modest, and overall app use was low. At 4 weeks, ∼50% of preteens and 30% of teens were still using the app. Although this is not surprising, it did contribute to the categorization of some variables and could have reduced our analytical power. Finally, our findings can only be applied to the discussion surrounding the promotion of app use and not behavior change.

Conclusion

Our findings suggest that social features including peers or health professional had positive effects on app use regardless of age, whereas parent co-participation may benefit preteens more. This highlights the need to consider how older teens can be supported by their parents while using mHealth apps. To maximize app use, and potentially promote behavior change,^8,21,32,37 factors that affect engagement should be considered during app design¹³ and tailored by age.⁴³

Footnotes

Funding Information

Funding for the Aim2Be Trial was provided by a contract from the Childhood Obesity Foundation to LCM, which received funds from the Public Health Agency of Canada (Project No. 1617-HQ-000046, Contact No. 604-251-2229). This was matched financially and in-kind by funds from Ayogo Health, Inc., Merck Canada, Inc., Heart & Stroke, Obesity Canada, Diabetes Canada, Dietitians of Canada, Craving Change, Canadian Society for Exercise Physiology, David Suzuki Foundation, and Pacific Blue Cross Foundation. The Childhood Obesity Foundation owns the content of Aim2Be, which was developed with expert input. Ayogo Health, Inc. contributed financial and technical resources to develop the Aim2Be app, which is powered by Ayogo's Empower platform.

Author Disclosure Statement

E.J.B. received a postdoctoral award from BC Children's Hospital Research Institute (BCCHR). G.D.C.B. and K.M.M. have consulted for NovoNordisk Canada. G.D.C.B. was supported by an Alberta Health Services Chair in Obesity Research. L.C.M. received salary support from the BCCHR. A.L.D., O.D.-J.G., J.H., and A.B. have no conflicts of interest to declare.

References

Statistics Canada. Children's Body Mass Index—World Health Organization Classification System, Inactive [Internet]. Available from: https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1310032201&pickMembers%5B0%5D=3.3 [Last accessed December 21, 2019].

Al-Khudairy

, Loveman

, Colquitt

, et al. Diet, physical activity and behavioural interventions for the treatment of overweight or obese adolescents aged 12 to 17 years. Cochrane Database Syst Rev, 2017; 6(6):CD012691; doi: 10.1002/14651858.CD012691.

Kobes

, Kretschmer

, Timmerman

, et al. Interventions aimed at preventing and reducing overweight/obesity among children and adolescents: A meta-synthesis. Obes Rev, 2018; 19(8):1065–1079.

Wang

, Cai

, Wu

, et al. What childhood obesity prevention programmes work? A systematic review and meta-analysis. Obes Rev, 2015; 16(7):547–565.

Skelton

, Beech

. Attrition in paediatric weight management: A review of the literature and new directions. Obes Rev, 2011; 12(5):e273–e281.

Barlow

, Ohlemeyer

. Parent reasons for nonreturn to a pediatric weight management program. Clin Pediatr, 2006; 45:355–360.

Rose

, Barker

, Maria Jacob

, et al. A systematic review of digital interventions for improving the diet and physical activity behaviors of adolescents. J Adolesc Heal, 2017; 61(6):669–677.

Watson

, Dugdill

, Pickering

, et al. Distinguishing factors that influence attendance and behaviour change in family-based treatment of childhood obesity: A qualitative study. Br J Heal Psychol, 2021; 26(1):67–89.

Sease

, Rolke

, Forrester

, et al. Feedback following a family-focused pediatric weight management intervention: Experiences from the New Impact Program. J Patient Exp, 2021; 8:1–8.

10.

Park

. Emerging new era of mobile health technologies. Heal Inform Res, 2016; 22(4):253–254.

11.

Ahern

. Challenges and opportunities of eHealth research. Am J Prev Med, 2007; 32(5 Supp):S75–S82.

12.

Tercyak

, Abraham

, Graham

, et al. Association of multiple behavioral risk factors with adolescents' willingness to engage in eHealth promotion. J Pediatr Psychol, 2009; 34(5):457–469.

13.

Frontini

, Sousa

, Dixe

, et al. Designing a mobile app to promote healthy behaviors and prevent obesity: Analysis of adolescents' preferences. Informatics Heal Soc Care, 2020; 45(3):327–341.

14.

Baumann

, Fiedler

, Wunsch

, et al. mHealth interventions to reduce physical inactivity and sedentary behavior in children and adolescents: Systematic review and meta-analysis of randomized controlled trials. JMIR Mhealth Uhealth, 2022; 10(5):e35920-41.

15.

Fedele

, Cushing

, Fritz

, et al. Mobile health interventions for improving health outcomes in youth a meta-analysis. JAMA Pediatr, 2017; 171(5):461–469.

16.

TJH

, Lee

CCS

, Wong

, et al. Internet-based self-monitoring interventions for overweight and obese adolescents: A systematic review and meta-analysis. Int J Med Inform, 2018; 120:20–30.

17.

Piatkowski

, Faulkner

, Guhn

, et al. User characteristics and parenting practices associated with adolescents' initial use of a lifestyle behavior modification intervention. Child Obes, 2020; 16(6):367–378.

18.

Larsen

, Benitez

, Cano

, et al. Web-based physical activity intervention for latina adolescents: Feasibility, acceptability, and potential efficacy of the niñas saludables study. J Med Internet Res, 2018; 20(5):170–181.

19.

De Cock

, Van Lippevelde

, Vangeel

, et al. Feasibility and impact study of a reward-based mobile application to improve adolescents' snacking habits. Int J Middle East Stud, 2018; 21(12):2329–2344.

20.

Mâsse

, Watts

, Barr

, et al. Individual and household predictors of adolescents' adherence to a web-based intervention. Ann Behav Med, 2015; 49(3):371–383.

21.

Przybylski

, Rigby

, Ryan

. A motivational model of video game engagement. Rev Gen Psychol, 2010; 14(2):154–166.

22.

Deci

, Ryan

. The “what” and “why” of goal pursuits: Human needs and the self-determination of behavior. Psychol Inq, 2000; 11(4):227–268.

23.

Roberts

, Sage

, Geryk

, et al. Adolescent preferences and design recommendations for an asthma self-management app: Mixed-methods study. J Med Internet Res, 2018; 2(2):e10055–e10068.

24.

Martin

, Caon

, Adorni

, et al. A mobile phone intervention to improve obesity-related health behaviors of adolescents across Europe: Iterative co-design and feasibility study. JMIR Mhealth Uhealth, 2020; 8(3):1–16.

25.

Roberts

, Geryk

, Sage

, et al. Adolescent, caregiver, and friend preferences for integrating social support and communication features into an asthma self-management app. J Asthma, 2016; 53(9):948–954.

26.

Deci

, Ryan

. Self-Determination Theory: A Macrotheory of Human Motivation, Development, and Health. Canadian Psychology, 2008; 49(3):182–185.

27.

Rice

, Jumamil

, Jabour

, et al. Role of health coaches in pediatric weight management: Patient and parent perspectives. Clin Pediatr, 2017; 56(2):162–170.

28.

Cho

. The impact of post-adoption beliefs on the continued use of health apps. Int J Med Inform, 2016; 87(2016):75–83.

29.

Lee

. Explaining and predicting users' continuance intention toward e-learning: An extension of the expectation-confirmation model. Comput Educ, 2010; 54(2):506–516.

30.

Timpel

, Cesena

FHY

, Costa C da

, et al. Efficacy of gamification-based smartphone application for weight loss in overweight and obese adolescents: Study protocol for a phase II randomized controlled trial. Ther Adv Endocrinol Metab, 2018; 9(6):167–176; doi: 10.1177/2042018818770938

31.

Soni

, Jain

, Jajodia

. Mobile health (mHealth) application loyalty in young consumers. Young Consumers, 2021; 22(23):429–455.

32.

Mâsse

, Vlaar

, Macdonald

, et al. Aim2Be mHealth intervention for children with overweight and obesity: Study protocol for a randomized controlled trial. Trials, 2020; 21(1):132–146.

33.

De-Jongh González

, Tugault-Lafleur

, Buckler

, et al. The Aim2Be mHealth intervention for children with overweight or obesity and their parents: Person-centered analyses to uncover digital phenotypes. J Med Internet Res, 2022; 24(6):e35285-104.

34.

de Onis

, Onyango

, Borghi

, et al. Development of a WHO growth reference for school-aged children and adolescents. Bull World Heal Organ, 2007; 85:660–667.

35.

Bandura

. Health promotion by social cognitive means. Heal Educ Behav, 2004; 31(2):143–164.

36.

Deterding

. The lens of intrinsic skill atoms: A method for gameful design. Hum Comput Interact, 2015; 30:294–335.

37.

Ryan

, Rigby

, Przybylski

. The motivational pull of video games: A self-determination theory approach. Motiv Emot, 2006; 30(4):347–363.

38.

Ayogo

Health

, Inc. Designing Digital Tools for Patient Engagement: What Games and Play Can Teach Us [Internet]; 2015. [Cited Feb 2022]. Available from: https://ayogo.com/wp-content/uploads/2018/07/designing-digital-tools-for-patient-engagement.pdf

39.

Michie

, Richardson

, Johnston

, et al. The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Ann Behav Med, 2013; 46(1):81–95.

40.

Self-Determination Theory. Self-Regulation Questionnaires (SRQ) [Internet]; 2022. [Cited Feb 2022]. Available from: http://selfdeterminationtheory.org/self-regulation-questionnaires/

41.

Bursac

, Gauss

, Williams

, et al. Purposeful selection of variables in logistic regression. Source Code Biol Med, 2008; 3:17–25.

42.

STATA Corp. Statistical Software: Release 16. StatCorp LP: College Station, TX; 2019.

43.

Schoeppe

, Alley

, Van Lippevelde

, et al. Efficacy of interventions that use apps to improve diet, physical activity and sedentary behaviour: A systematic review. Int J Behav Nutr Phys Act, 2016; 13(1):127.

44.

Depper

, Howe

. Are we fit yet? English adolescent girls' experiences of health and fitness apps. Heal Sociol Rev, 2017; 26(1):98–112.

45.

Story

, Neumark-Sztainer

, French

. Individual and environmental influences on adolescent eating behaviours. J Am Diet Assoc, 2002; Supp:s41–s51.

46.

Reicks

, Banna

, Cluskey

, et al. Influence of parenting practices on eating behaviors of early adolescents during independent eating occasions: Implications for obesity prevention. Nutrients, 2015; 7(10):8783–8801.

47.

Pedersen

, Grønhøj

, Thøgersen

. Following family or friends. Social norms in adolescent healthy eating. Appetite, 2015; 86:54–60.

48.

Draper

, Grobler

, Micklesfield

, et al. Impact of social norms and social support on diet, physical activity and sedentary behaviour of adolescents: A scoping review. Child Care Heal Dev, 2015; 41(5):654–667.

49.

Lazard

, Watkins

, Mackert

, et al. Design simplicity influences patient portal use: The role of aesthetic evaluations for technology acceptance. J Am Med Informatics Assoc, 2016; 23:e157–e161.

50.

Strömmer

, Shaw

, Jenner

, et al. How do we harness adolescent values in designing health behaviour change interventions? A qualitative study. Br J Heal Psychol, 2021; 26(4):1176–1193.

51.

Fitzgerald

, McClelland

. What makes a mobile app successful in supporting health behaviour change?. Heal Educ J, 2017; 76(3):373–3781.