Engaging in Weight Loss Tasks of Mobile Health Applications: The Dual Influence of Social Support and Body Condition

Abstract

Introduction:

In the past decades, the average weight of residents has increased significantly. Weight management has become an important issue in our society. With the developments of mobile technology and applications, mobile health applications (m-health apps) provide a convenient platform for users to engage in weight loss tasks and control their body weight. However, due to the lack of proper motivators, engaging in weight loss tasks on m-health apps is stressful for the users. The aim of this study is to understand why users engage in weight loss tasks, and to be specific, we establish an empirical model to examine the effects of social motivators (social support) and personal motivators (body condition) and their interactive relationship on the level of user engagement using self-determination theory.

Materials and Methods:

We developed a JAVA software program and automatically downloaded 1,138 users' information from an m-health app. Following, we used these data to calculate variables of our research model, including body mass index (BMI), informational support, emotional support, and the level of engagement in weight loss tasks. Additionally, we used the Ordinary Least Squares to estimate our research model. We also checked the robustness of the results by dividing the data into different groups.

Results:

The empirical results of our research model indicate that informational and emotional supports are positively associated with the engagement levels of users in weight loss tasks. Additionally, we have found out that body condition (using BMI as a proxy) has a U-shaped relationship effect on users' engagement. Furthermore, our research proves that body condition and informational support have a substitutive relationship in affecting user engagement.

Conclusions

: These findings can contribute to the literature concerning online weight loss and to provide suggestions for users and practitioners of m-health apps, catering different incentive mechanisms to users with different body condition to help them control their body weight.

Introduction

The rise in overweight and obesity rate among residents has become a serious issue in our society today,^1,2 causing weight management to become an important issue to people of all ages. To alleviate the issue and aid in weight management, mobile health applications (m-health apps) provide a convenient platform for users to manage their health and weight. Users of m-health apps can engage in online weight loss tasks (such as planning exercise routines and healthy meals) and implement their weight management plan. Therefore, using m-health apps is a means of health management to encourage users to keep their body weight in control. However, due to the lack of proper motivators on the applications, engaging in online weight loss tasks is nevertheless stressful for users. Hence, it is important for creators and practitioners of m-health apps to determine the variables affecting users' engagement in weight loss tasks on m-health apps.

Self-determination theory describes and explains variables affecting the motivation of individuals' behaviors.³ This theory indicates that humans are capable of self-regulating and developing to improve and enhance themselves.⁴ Self-determination helps individuals pursue their interests and encourage them to experience things and interests even if out of their comfort zone.³ Self-determination theory and the previous studies demonstrate how relevant motivators impacting people's behaviors can generally be split into two aspects, that is, personal and social motivators.⁵

m-Health apps are regarded as user-driven healthcare and a type of self-care model.⁶ Users can devise and create their weight management plan based on their body conditions to effectively engage in weight loss tasks designed by m-health apps. Therefore, it is certain that one's physical body condition can productively motivate and stimulate one to engage in weight loss tasks. Generally, people have an ideal weight or physical condition,^3,7 thus, if any users exceed their ideal weight, they will develop a stronger motivation to lose weight. Hence, body condition is an important personal motivator influencing user engagement in weight loss tasks.

In addition, without the spatial and temporal limitation, m-health apps have also created a convenient platform for users to interact with each other.^8
–10 Users can expand their social network through the applications by concurrently sharing their experiences and knowledge about weight loss issues, and receiving support from other online users.¹¹ People who wish to lose weight and manage their health may find it difficult to maintain an ideal weight without proper support. Consequently, relevant research in recent decades points out that social supports offered by peers from the applications do significantly impact one's health behaviors, resulting in successful weight loss.^9,12 Therefore, social support is an essential social motivator influencing users to engage in weight loss tasks on m-health apps.

Even though, prior related studies had been done to explore many different motivators affecting users' participation and usage of m-health apps,^11,13 empirical studies are still lacking in two important aspects. First, there is scant research on the role of users' physical body condition affecting their health status. m-Health apps are self-care platforms for users to manage their weight. Therefore, users' physical body condition plays an important role for users to engage in m-health apps. Second, there are not many literatures analyzing or interpreting the relationship between social motivators and personal motivators. Substitute motivators in affecting users' behaviors can be due to the varying needs, social supports, and body condition of different individuals. Understanding the relationship between the two motivators can help us better understand users' behaviors and motivations regarding engagement in weight loss tasks. Thus, it is important to investigate the interactions between social and personal motivators; this research article aims to analyze the effects of social supports and body condition on the level of users' engagement in weight loss tasks.

The main research questions of this article are:

What is the effect of social support on the level of users' engagement in weight loss tasks?

What types of users' body conditions tend to engage in weight loss tasks?

What is the interactive relationship between social support and body condition?

Materials and Methods

Research Hypotheses

m-Health apps launched many health tasks (such as exercise and diet) to assist users to lose weight. However, engaging in those weight loss tasks can be stressful for users. According to a related theory, social support is essential in helping users to cope with the stress¹¹ as it relates to the exchange and delivery of information and emotions among users of m-health apps.¹² Informational support obtained by other people offers more personal opinions and experiences which can change users' behaviors to lose weight.¹⁴ The knowledge acquired from informational support helps users to better understand their problem and get a more definite solution. Emotional support refers to the sharing of concerns and encouragement among users,¹² sending a message to users that they are being taken care of in m-health apps. Due to constraints in space and time, users are having difficulty obtaining useful support.¹⁵ However, m-health apps make it possible for users to interact with other users and receive emotional support without any restrictions in time and space. Therefore, informational and emotional support are especially important for users to engage in weight loss tasks. Based on what was discussed above, we hypothesize:

H1. Informational support is positively associated with the level of user engagement in weight loss tasks.

H2. Emotional support is positively associated with the level of user engagement in weight loss tasks.

Users' physical body condition is also an important motivator in affecting the level of engagement in weight loss tasks. Generally, motivators relate to individuals' primary needs, which specify the necessary conditions for psychological well-being. To fulfill those needs, individuals would come up with their own goals¹⁶ and different goal pursuits have different impacts on individuals' behaviors. For users of m-health apps, the primary goal is to control their body weight and enhance their body condition. There are two main reasons as to why one's body condition can impact the level of engagement in online weight loss tasks, that is, physical and physiological causes.¹⁰ First, individuals who are obese generally suffer a poor status and are more vulnerable to other diseases (such as hypertension and diabetes). Thus, users want to lose weight and change their body condition. Second, individuals who are obese often feel distressed and anxious; users tend to lose weight to cope with those physiological pressures. As a result, individuals who are obese are more likely to engage in weight loss tasks due to physical and physiological causes. Hence, we hypothesized:

H3. The body condition is positively associated with users' level of engagement in weight loss tasks.

Self-determination theory indicates that the impacting strength of motivators depends on individual's primary needs. When the user's body condition is good, the personal health motivator is not the main factor influencing the user to engage in weight loss tasks. Then, obtaining social support and interacting with other users become the main reason why users use m-health apps to engage in weight loss tasks. In contrast, if the user is obese, enhancing one's body condition will be his\her main goal to engage in weight loss tasks. Therefore, there is a substitutive relationship between social support and body condition in impacting the level of engagement in weight loss tasks. Hence, we hypothesized:

H4a. The impact of body condition and informational support on the level of engagement in weight loss tasks is substitutable.

H4b. The impact of body condition and emotional support on the level of engagement in weight loss tasks is substitutable.

Study Design

To test our research hypotheses, we collected data from an application, “Bohe.” It is one of the more popular m-health apps in China and users of this application can upload their data online to manage their body condition and control their body weight. Moreover, users of “Bohe” can exchange information, obtain social support, and develop online friendships. Hence, this application offers ample information for the empirical research of our article. Data utilized include users' age, gender, height, initial body weight, current weight lost, and goal weight.

The study design and research process of this article are as follow: First, we used the data collected to calculate the variables of our research model, including body mass index (BMI), informational support, emotional support, and the level of engagement in weight loss tasks. Second, we used Ordinary Least Squares (OLS) to estimate our research model and test the different hypotheses. Third, we divided data into different groups to do robustness check of research results.

Participants and Data

We designed a JAVA program to download data from “Bohe” application automatically and input those data into an established database. After excluding incomplete data, such as the lack of demographics and social information, statistics from a total of 1,138 users were collected.

The dependent variable of our research model is the level of engagement in weight loss tasks. We used task scores as a proxy for the level of engagement and they were dependable on whether or not users accomplish their weight loss tasks (exercises and calorie intake). The greater the task score, the more weight loss tasks an individual has accomplished.

One of the independent variables in our research is body condition. BMI is an important tool for people to assess and measure the level of obesity and body condition. World Health Organization (WHO) pointed out that the ideal body condition has a BMI of 18.5–24.9. BMI value greater than 25 indicates that an individual is overweight and if the BMI value is greater than 27, it indicates obesity. Hence, BMI is an effective tool to measure one's body condition. The specific method to calculate BMI is as follows: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}{ \rm{BMI}} \; = \;{ \rm{Weight}} \; \left( {{ \rm{kg}}} \right) / { \rm{Heigh}}{{ \rm{t}}^2} \left( { \rm{m}} \right). \end{align*} \end{document}

The other independent variable is social support received by users. We divided social support into two aspects, that is, informational and emotional support. In our research model, we used the number of postings as a proxy of informational support received by users. In m-health apps, users can send postings to obtain related information from other users. Therefore, the number of postings effectively reflects informational support. Moreover, this article uses the number of online friends as a proxy for emotional support. Online friends provide relevant emotional support by making each other feel like they are taken care of.

In addition, we need to control the effects of individual differences in our research. This article uses age and gender to control the role of users' demographic characteristics in the research model. Prior studies have shown that users' online behaviors could be impacted by their age.^17
–19 It means that there are significant differences between the young and old users. Moreover, users' gender also plays an important role in the level of engagement in weight loss tasks. This article uses a dummy variable for gender difference, with female and male expressed as 0 and 1, respectively.^18,20 Furthermore, we take into consideration the number of days on which users use this application. Table 1 shows the description of the variables.

Table 1.

Demographic Information (N = 1,138)

	FREQUENCY	PERCENTAGE
Gender
Female	532	46.7
Male	606	53.3
Age
≤19	34	2.9
20–29	478	42
30–39	432	37.9
40–50	194	17
BMI
18–25	326	28.6
25–27	341	30
27–30	327	28.7
≥30	144	12.7

BMI, body mass index.

Research Model

To test the hypothesis on the effectiveness of social support and body condition, this article builds the following empirical model. The users' body condition is measured using the BMI. However, as too large or small values of BMI reflect unhealthy weight conditions, this article uses square term (U-shape) to investigate the effects of BMI in our research model. Furthermore, due to the significant variance in independent variables (which mainly focuses on informational and emotional support), the distribution of variables is not normal. Hence, the research model was adjusted to linear-logarithmic regression: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}{ \rm{Log}} \left( {{ \rm{engagemen}}{{ \rm{t}}_{ \rm{i}}}} \right) \; = \;{{ \rm{a}}_0} \; + \;{{ \rm{a}}_1}{ \rm{Ag}}{{ \rm{e}}_{ \rm{i}}} \; + \;{{ \rm{a}}_2}{ \rm{Gende}}{{ \rm{r}}_{ \rm{i}}} \; + \;{{ \rm{a}}_3}{ \rm{Log}} \left( {{ \rm{da}}{{ \rm{y}}_{ \rm{i}}}} \right) \; \\+ \;{{ \rm{a}}_4}{ \rm{Log}} \left( {{ \rm{informationa}}{{ \rm{l}}_{ \rm{i}}}} \right) \; + \;{{ \rm{a}}_5}{ \rm{Log}} \left( {{ \rm{emotiona}}{{ \rm{l}}_{ \rm{i}}}} \right) \; \\+ \;{{ \rm{a}}_6}{ \rm{BM}}{{ \rm{I}}_{ \rm{i}}} \; + \;{{ \rm{a}}_7}{ \left( {{ \rm{BM}}{{ \rm{I}}_{ \rm{i}}}} \right) ^2} \; \\+ \;{{ \rm{a}}_8}{ \rm{BM}}{{ \rm{I}}_{ \rm{i}}}^*{ \rm{Log}} \left( {{ \rm{informationa}}{{ \rm{l}}_{ \rm{i}}}} \right) \; \\+ { \rm{ \;}}{{ \rm{a}}_9}{ \rm{BM}}{{ \rm{I}}_{ \rm{i}}}^*{ \rm{Log}} \left( {{ \rm{emotiona}}{{ \rm{l}}_{ \rm{i}}}} \right) \; + \;{{ \rm{u}}_{ \rm{i}}}. \end{align*} \end{document}

The parameters from a₀ to a₉ are estimated in research model. The variables BMI_i*Log(informational_i) and BMI_i*Log(emotional_i) are interactive terms to test the relationship between body condition and social support.

Results

Data Analysis

Table 2 shows the demographic information of our collected users' data. Table 3 presents the descriptive statistics, whereas Table 4 shows the correlations of independent and dependent variables in our research model. The variables include age, gender, social support, BMI and the level of engagement in weight loss tasks.

Table 2.

Descriptive Statistics of Variables (N = 1,138)

	MIN	MAX	MEAN	DEVIATION
Age	18.000	50.000	32.574	2.453
Gender	0.000	1.000	0.532	0.499
Day	3.000	271.00	138.330	51.684
Goal pursuit	5.000	65.000	18.2036	11.847
Emotional support	3.00	273.00	105.00	49.58
Informational support	0.00	9,107.00	37.587	343.597
BMI	19.010	69.200	28.258	4.799
Engagement	100.000	42,162.00	599.750	2,029.356

Table 3.

Correlations of Variables (N = 1,138)

VARIABLE	1	2	3	4	5	6	7	8
1. Age	1
2. Gender	0.013	1
3. Day	−0.21	0.243^a	1
4. Goal pursuit	0.084^a	−0.043	0.063	1
5. Informational support	0.053	−0.092^a	0.054	−0.015	1
6. Emotional support	0.065^b	−0.146^a	0.012	−0.003	0.347^a	1
7. BMI	0.111^a	0.175^a	0.064	0.495^a	−0.012	−0.003	1
8. Engagement	0.110^a	−0.182^a	0.037	−0.035	0.549^a	0.568^a	−0.041	1

Correlation is significant at the 0.01 level (two-tailed).

Correlation is significant at the 0.05 level (two-tailed).

Table 4.

Parameter Estimates

VARIABLE	1	2	3	4
Constant	2.899^*** (7.990)	4.063^*** (17.095)	4.818^*** (12.046)	4.160^*** (9.564)
Age	0.123^*** (11.890)	0.047^*** (6.722)	0.049^*** (6.943)	0.048^*** (6.914)
Gender	−0.961^*** (−17.762)	−0.320^*** (−8.193)	−0.279^*** (−6.795)	−0.276^*** (−6.760)
Log(day)	0.086^** (2.365)	0.026(1.086)	0.023(0.932)	0.022(0.896)
Log(goal)	0.051^** (2.254)	0.075^** (2.580)	0.005(0.121)	0.003(0.070)
Log(informational support)		0.114^*** (5.126)	0.120^*** (5.373)	0.367^*** (2.881)
Log(emotional support)		0.419^*** (21.046)	0.415^*** (20.845)	0.491^*** (4.633)
BMI			−0.053^** (−2.410)	−0.025(−1.088)
BMI²			0.001^** (1.970)	0.001^** (1.948)
BMI^*Log(informational support)				−0.009^** (−2.014)
BMI^*Log(emotional support)				−0.002(−0.667)
Adjust-R ²	0.289	0.699	0.701	0.705
F	116.286^***	442.109^***	334.984^***	272.780^***
N	1,138	1,138	1,138	1,138

t Statistics in parentheses.

p < 0.05, ^*** p < 0.01.

Model Estimation

Table 5 shows the empirical results of our research model estimated by OLS and the empirical model is presented hierarchically. It shows a model with control variables in columns 1, and the independent variables and interaction terms shown in column 2, 3, and 4. Moreover, the adjusted R ² and F values of regression were reasonable and statistically significant.

Table 5.

Parameter Estimates

INDEPENDENT VARIABLE	1	2	3	4
Constant	2.799^*** (7.942)	3.917^*** (16.926)	4.825^*** (12.208)	4.165^*** (9.671)
Age	0.122^*** (11.833)	0.046^*** (6.523)	0.049^*** (6.945)	0.048^*** (6.917)
Gender	−0.954^*** (−17.741)	−0.312^*** (−7.993)	−0.278^*** (−7.013)	−0.276^*** (−6.989)
Log(day)	0.083^** (2.284)	0.021(0.847)	0.022(0.925)	0.021(0.894)
Log(informational support)		0.120^*** (5.373)	0.115^*** (5.167)	0.367^*** (2.881)
Log(emotional support)		0.415^*** (20.845)	0.423^*** (21.288)	0.492^*** (4.644)
BMI			−0.054^*** (−2.606)	−0.025(−1.173)
BMI²			0.001^** (2.049)	0.001^*** (2.550)
BMI^*Log(informational support)				−0.009^** (−2.014)
BMI^*Log(emotional support)				−0.002(−0.672)
Adjust-R ²	0.288	0.698	0.702	0.705
F	154.559^***	526.566^***	383.171^***	303.356^***
N	1,138	1,138	1,138	1,138

t Statistics in parentheses.

p < 0.05, ^*** p < 0.01.

H1 predicts that informational support is positively associated with the level of users' engagement in weight loss tasks. According to column 2 of Table 5, empirical results show evidence to support this hypothesis. The coefficient of informational support (a₄ = 0.120, t = 5.373, p < 0.01) is positive and statistically significant. This means that when the number of informational support increases, users' level of engagement in weight loss tasks will increase as well.

H2 predicts that emotional support is positively associated with the level of users' engagement in weight loss tasks. Column 2 of Table 5 provides support to this hypothesis as the coefficient of emotional support (a₅ = 0.451, t = 20.845, p < 0.01) is positive and statistically significant. This result proves that when the number of emotional support increases, the level of engagement in weight loss tasks will increase as well.

H3 predicts that body condition is positively associated with the level of users' engagement in weight loss tasks. However, in column 3 of Table 5, the coefficient of BMI (a₆ = −0.054, t = −2.606, p < 0.01) is negative and statistically significant; this includes the coefficient of the square term (a₇ = 0.001, t = 2.049, p < 0.05) being positive and statistically significant. It indicates that the BMI has a U-shape relationship effect on users' engagement.

H4a and 4b point out that there is a substitutive relationship between social support and body condition. In column 4 of Table 5, the coefficient of informational support (a₇ = −0.009, t = −2.014, p < 0.05) is negative and statistically significant. Empirical results show evidence to support hypothesis 4a, that is, the impact of body condition and informational support on the level of engagement in weight loss tasks is substitutable. However, we did not find any evidence to support hypothesis 4a because the coefficient of emotional support (a₈ = −0.002, t = −0.672, p > 0.1) is not statistically significant.

Figure 1 shows the results of our research model. In this figure, the symbol ns indicates that variable and hypothesis are not supported by research model. The imaginary line shows the square term.

Fig. 1.

Results of the research model. **p < 0.05, ***p < 0.01; ns indicates that the hypothesis is not supported.

Robustness Check

To test the effectiveness of the research model, this article divides the data into four groups based on the BMI values, that is, less than 25 (normal), 25–27 (overweight), 27–30 (obesity), and more than 30 (severe obesity). Thereafter, we run the research model for every group separately without BMI. The purpose of this step is to test the impact of informational and emotional support on users' engagement and the substitutive relationship between social support and body condition.

Table 6 shows the results of the robustness check. In every group, the coefficients of informational support and emotional support are positive and statistically significant. Furthermore, as the BMI value increases, the coefficient of informational support decreases, and the coefficient of emotional support does not change significantly. Those results are consistent with the prior model using total data with BMI. The informational and emotional support still positively impact users the level of engagement in weight loss tasks. Likewise, there is a substitutive relationship between emotional support and BMI.

Table 6.

Robustness Check

INDEPENDENT VARIABLE	<25	25–27	27–30	>30
Constant	4.508^*** (10.215)	3.635^*** (8.872)	3.739(9.165)	3.117^*** (5.243)
Age	0.016(1.160)	0.062^*** (5.186)	0.060^*** (4.692)	0.063^*** (6.948)
Gender	−0.450^*** (−5.576)	−0.274^*** (−4.101)	−0.121^* (−1.701)	−0.177^* (−1.847)
Log(day)	0.028(2.284)	0.022(0.240)	−0.015(−0.355)	0.071(1.139)
Log(informational support)	0.229^*** (4.744)	0.080^** (2.144)	0.052^** (2.234)	0.078^* (1.750)
Log(emotional support)	0.372^*** (8.856)	0.445^*** (13.403)	0.460^*** (12.367)	0.388^*** (8.250)
Adjust-R ²	0.724	0.693	0.711	0.664
F	171.729^***	154.839^***	161.565^***	75.211^***
N	326	341	327	144

t Statistics in parentheses.

p < 0.1, ^** p < 0.05, ^*** p < 0.01.

Discussion

Summary of Findings

This article primarily explores the effects of social support and body condition on users' engagement in weight loss tasks. Based on the self-determination theory, we provide research hypotheses and build an empirical model. The results of our empirical research support most of these hypotheses. First, the empirical results prove that informational support and emotional support are positively associated with the level of users' engagement in weight loss tasks. Both supports are vital in helping users cope with the pressure of weight loss tasks. Thus, the greater the support, the greater the level of engagement users receive.

Second, apart from social support, body condition is also an important personal motivator influencing users' engagement in online weight loss tasks. We use a square term to investigate the role of BMI in our research model. The findings indicate that body condition has a U-shaped effect on the level of users' engagement in weight loss tasks. When BMI value is lower than 27 (indicating overweight), the level of engagement decreases as BMI increases. When the BMI value is larger than 27 (indicating obesity), the level of engagement increases as BMI increases. Hence, users who are overweight have the lowest level of engagement in weight loss tasks. The relationship between users' level of engagement and the value of BMI is presented in Figure 2. Two possible explanations for this occurrence are, first, many people prefer being slightly overweight because they feel that this body condition is healthier. Hence, users who have a BMI value of 27 think they have an ideal body condition and negatively engage in weight loss tasks. Second, users with a BMI value of 27 do not realize that they are overweight because the characteristics of obesity are not obvious. Thus, those users have low level of engagement in weight loss tasks.

Fig. 2.

The relationship between the level of engagement and BMI. BMI, body mass index.

In addition, we find that the impact of body condition and informational support on the level of engagement in weight loss tasks is substitutable. The empirical results of this article prove that when a user's BMI increases, the effect of informational support on the level of users' engagement decreases because improving body condition is a more important motivator impacting users' engagement in weight loss tasks. Users with high BMI desire to change their body condition. This shows that the impact of the personal motivator on users' engagement is stronger than the social motivator. However, we did not find any evidence on H4b regarding the impact of body condition and emotional support on the level of engagement in weight loss tasks substitutable. One possible reason could be, even though the primary purpose of users is to control their body weight and manage their body condition, they still need emotional support to motivate them to engage in weight loss tasks. Hence, there is no significant substitutable relationship between body condition and emotional support.

Implication

From the perspectives of users and practitioners, this article provides some insights into the weight management plan and the design of m-health apps. First, the results of this article prove that the body condition has a U-shaped effect on the level of users' engagement in weight loss tasks. Users with high BMI value (>27) are more likely to have the urge to improve their body condition and BMI is critical to users' weight loss performance. Therefore, users of m-health apps should create a goal for their ideal weight. Second, we had found a substitutive relationship between social support and body condition, that is, users who have good body condition are social orientated, whereas users who have poor body condition are health orientated. Therefore, the designer of m-health apps should design different incentive mechanisms for users with different body conditions (BMI) to help them engage in weight loss tasks.

Limitation

This article has its limitations. First, the research was conducted solely based on one specific m-health apps. In the future, data from several different applications can be collected to test the effects of social support and body condition on users' behaviors. Second, this article uses only cross-sectional data to carry out the regression analysis. This method limits the investigation of the causal relationship between related factors and users' behaviors. Therefore, empirical model based on panel data can be used in future researches to validate the results.

Conclusions

m-Health apps have not only become a convenient platform for individuals to control their body weight but it has also attracted a considerable amount of attention in our society. Although extensive research has been done to discover the effect of social motivators on users' health behaviors, that is, social support, there were scant studies carried out to investigate the roles of personal factors. To fill those research gaps, an empirical model was developed in this research article through self-determination theory to examine the effects of social support and body condition (BMI) on the level of users' engagement in weight loss tasks. The empirical results indicated that online informational support and emotional support are fundamental for users to engage in m-health apps. Moreover, users' body condition has a U-shape effect on their engagement. Finally, we also found out that the impact of body condition and emotional support on the level of engagement in weight loss tasks is substitutable. These findings can contribute to the research on m-health apps and provide improvements for users and practitioners.

Footnotes

Acknowledgments

This study was partially funded by the China Postdoctoral Science Foundation Grant (2017M622647) and National Natural Science Foundation of China Grant (71572050). The authors highly appreciate the Editor and anonymous reviewers for their insightful comments and suggestions. All errors remain ours.

Disclosure Statement

No competing financial interests exist.

References

Fowler-Brown

, Kahwati

. Prevention and treatment of overweight in children and adolescents. Am Fam Physician, 2004; 69:2591–2598.

Franz

, Vanwormer

, Crain

, et al. Weight-loss outcomes: A systematic review and meta-analysis of weight-loss clinical trials with a minimum 1-year follow-up. J Am Diet Assoc, 2007; 107:1755–1767.

Ryan

, Deci

. Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. Am Psychol, 2000; 55:68.

Gagn

, Deci

. Self-determination theory and work motivation. J Organ Behav, 2005; 26:331–362.

Sun

, Fang

, Lim

. Understanding sustained participation in transactional virtual communities. Decis Support Syst, 2012; 53:12–22.

Lamberg

. Online empathy for mood disorders: Patients turn to internet support groups. JAMA, 2003; 289:3073–3077.

Ryan

, Deci

. Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemp Educ Psychol, 2000; 25:54–67.

Ziebland

, Chapple

, Dumelow

, et al. How the internet affects patients' experience of cancer: A qualitative study. BMJ, 2004; 328:564–569.

Yan

, Tan

. Feeling Blue? Go Online: An Empirical Study of Social Support Among Patients. Inform Syst Res, 2014; 25:690–709.

10.

Xiao

, Sharman

, Rao

, et al. Factors influencing online health information search: An empirical analysis of a national cancer-related survey. Decis Support Syst, 2012; 57:417–427.

11.

, Wang

. Digital health communities: The effect of their motivation mechanisms. Decis Support Syst, 2013; 55:941–947.

12.

Yan

, Tan

, Peng

. Network dynamics: How can we find patients like us?. Inform Sys Res, 2015; 26:496–512.

13.

Guo

, Guo

, Zhang

, et al. Doctor–patient relationship strength's impact in an online healthcare community. Inform Technol Dev, 2017; 24:279–300.

14.

Mcmullan

Patients using the Internet to obtain health information: How this affects the patient-health professional relationship. Patient Edu Couns, 2006; 63; 24–28.

15.

Bambina

. Online social support: The interplay of social networks and computer-mediated communication. Amherst, NY: Cambria Press, 2007.

16.

Deci

, Ryan

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

17.

Campbell

, Macauley

, Mccrum

, et al. Age differences in the motivating factors for exercise. J Sport Exercis Psychol, 2001; 23:191–199.

18.

Sun

, Lim

, Jiang

, et al. Do males and females think in the same way? An empirical investigation on the gender differences in Web advertising evaluation. Comput Hum Behav, 2010; 26:1614–1624.

19.

Zhou

, Jin

, Fang

. Moderating role of gender in the relationships between perceived benefits and satisfaction in social virtual world continuance. Decis Support Syst, 2014; 65:69–79.

20.

Caspersen

, Pereira

, Curran

. Changes in physical activity patterns in the United States, by sex and cross-sectional age. Med Sci Sports Exerc, 2000; 32:1601–1609.