This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR mhealth and uhealth, is properly cited. The complete bibliographic information, a link to the original publication on http://mhealth.jmir.org/, as well as this copyright and license information must be included.
Mobile health (mHealth) apps that support individuals pursuing health and wellness goals, such as weight management, stress management, smoking cessation, and self-management of chronic conditions have been on the rise. Despite their potential benefits, the use of these tools has been limited, as most users stop using them just after a few times of use. Under this circumstance, achieving the positive outcomes of mHealth apps is less likely.
The objective of this study was to understand continued use of mHealth apps and individuals’ decisions related to this behavior.
We conducted a qualitative longitudinal study on continued use of mHealth apps. We collected data through 34 pre- and postuse interviews and 193 diaries from 17 participants over two weeks.
We identified 2 dimensions that help explain continued use decisions of users of mHealth apps: users’ assessment of mHealth app and its capabilities (user experience) and their persistence at their health goals (intent). We present the key factors that influence users’ assessment of an mHealth app (interface design, navigation, notifications, data collection methods and tools, goal management, depth of knowledge, system rules, actionable recommendations, and user system fit) and relate these factors to previous literature on behavior change technology design. Using these 2 dimensions, we developed a framework that illustrated 4 decisions users might make after initial interaction with mHealth apps (to abandon use, limit use, switch app, and continue use). We put forth propositions to be explored in future research on mHealth app use.
This study provides insight into the factors that shape users’ decisions to continue using mHealth apps, as well as other likely decision scenarios after the initial use experience. The findings contribute to extant knowledge of mHealth use and provide important implications for design of mHealth apps to increase long-term engagement of the users.
The use of smartphones to deliver health care services has been consistently on the rise for over a decade [
Studies that looked at individuals’ adoption and use of mHealth have shown the importance of factors, such as users’ motivation, existing health conditions, individual differences [
As such, we conducted a qualitative, longitudinal, and exploratory study on continued use of health and wellness apps, a set of apps that are not disease-specific and aim to promote general wellness. The analysis revealed 2 important dimensions related to users’ assessment of an mHealth app and its capabilities (user experience) and the users’ persistence at achieving their health goals (intent). On the basis of these 2 dimensions, we proposed a 2×2 matrix to depict 4 type of users’ decisions after adopting mHealth, which are to
The issue of adoption and use of technology has long been pursued by IS scholars. Earlier studies provided an overview of the basic predictors of adoption (see Venkatesh et al [
mHealth technologies are designed to motivate and persuade behavior change to help users achieve their health and wellness goals. A number of studies have reported on the importance of design in persuasive technologies [
Participants were recruited through an open call in 2 universities (1 public and 1 private) located in the Northeastern United States. We used purposive sampling [
We used semistructured, face to face interviews and daily use diaries to collect data. The interview questions were developed based on a review of existing literature and further refined via discussions with 3 academic experts in health informatics.
During the first round of interviews (preuse interviews), participants answered questions about their approach to health and wellness, motivation to follow a healthy lifestyle, and level of confidence in making lifestyle changes to improve health. At the end of the preuse interviews, we asked participants to (1) identify a health and/or wellness goal toward which they wanted to work during the upcoming 14 days, (2) select a free mHealth app to use and download on their phones, and (3) describe how they are planning to use the new mHealth app.
Given that most users tend to withdraw from mHealth apps before the end of first week [
After the 14-day use period was over, participants were invited for the second round of interviews (postuse interviews). The participants were asked to describe their experience with the app, reflect back on their goals and motivation, assess if the app helped them achieve their goals, and discuss the reasons behind their positive or negative decisions to continue or withdraw use. Finally, the participants were asked to provide design suggestions for app developers that would improve their experience with mHealth apps and result in continued use. The pre- and postuse interviews took, on average, 21 min (13-47 min). The final dataset, collected from May to August 2017, included a total of 34 interviews and 193 daily usage diaries (some participants did not complete between 1 and 3 days of diary keeping). The participants received gift cards at the end of the closing interview.
Participants were aged between 18 and 51 years, 70% (12/17) females, and 70% (12/17) iPhone users. Among 17 participants, 10 continued using the apps they picked during the initial interviews. However, 7 participants decided to try a new mHealth app because their original choices did not satisfy their needs. This is illustrated in
Study participant characteristics and goals.
Identification number | Age (years) | Sex | Phone | Selected apps (second choice, if changed) | Area of focus |
1 | 23 | Fa | Android | MyFitnessPal | Diet |
2 | 24 | F | iPhone | ARise | Diet/physical activity |
3 | 18 | F | Android | Nike+ Training | Physical activity |
4 | 50 | Mb | iPhone | Calorie Counter (Food Diary) | Diet |
5 | 26 | M | iPhone | Strong | Physical activity |
6 | 35 | F | iPhone | MapMyWalk | Physical activity |
7 | 28 | F | iPhone | Sleep Better (TracknShare LITE) | Sleep |
8 | 51 | F | iPhone | Weight Watchers | Diet |
9 | 47 | F | iPhone | Relax Lite | Mindfulness |
10 | 33 | F | iPhone | Aura and Headspace | Mindfulness |
11 | 28 | F | iPhone | 5 Minute Home Workouts | Physical activity |
12 | 29 | F | iPhone | Female Fitness (Fitbit) | Physical activity |
13 | 29 | F | iPhone | Plant Nanny and Garmin | Diet (water)/physical activity |
14 | 40 | F | Android | Headspace (Aura) | Mindfulness |
15 | 32 | F | iPhone | Map my run (HabitBull) | Physical activity/habit building |
16 | 24 | M | Android | Samsung Health (Aura) | Physical activity/mindfulness |
17 | 29 | M | Android | Charity Miles (ASICS) | Physical activity |
aFemale.
bMale.
All interviews and individual diaries (over 300 pages) were transcribed and added to QSR’s NVivo application, which was used to code the data and conduct content analysis. To ensure anonymity, each participant was assigned a study identification number (ID). The analysis was performed using the grounded theory approach suggested by Strauss and Corbin (1998). During the open coding phase, we identified a total of 48 codes (eg, app customization, effort needed, reminder/alerts, motivation, activeness, content quality, context access, and continuance/discontinuance) that related to how participants described their use, assessed mHealth apps, and evaluated their willingness to keep using the app. During the axial coding phase, we discussed in multiple rounds how the 48 themes were related or distinct to determine overarching themes; the result was identification of 9 key dimensions that determined continued use of the app. Finally, focusing on the nature of decisions regarding continued use, we proposed a framework (2×2 matrix) based on users’
Among the 19 individuals who were screened, 18 were eligible to participate in this study. A participant was excluded because of existing chronic conditions. Of the 18 eligible participants, 1 dropped out of the study during the 14-day use period and was not included in the analysis (final sample=17).
The literature reports that when testing the influence of behavior change technologies on users’ behavior, characteristics of users should be considered to make sense of the study results [
The first dimension represents the initial user experience with mHealth apps and whether users have a positive or negative assessment of the capabilities of these apps. This dimension resembles previous study findings that highlight the role of satisfaction with IT use as an important step for users to extend the use of technology [
Interface design related comments reflected participants’ preference for
Navigation (how users move through the menus and different features to accomplish their tasks) is another important factor that shapes users’ opinions. The participants articulated their preference for an easy-to-understand
The participants also expressed the need for
Factors influencing decision to continue use.
Factors | Evidence from data |
Interface: clean and simple design; appearance of advertisements | “I liked the way the dashboard looked. It was just so clean and so I said ‘Alright well I’ll download that and give it a try!’ and I’ve been really happy with [using] it.” [P12]; “I think the app had a lot of ads, and I know they have to make their money...when I was trying to add something, an ad keeps trying to pop up, it was frustrating.” [P13] |
Navigation: navigation menu and flow of pages; training and wizards | “I don’t have to find all these different buttons and how to navigate through it. It’s very simple to use. So, whereas the other app when there’s so many different features...I don’t have time to go through all of them.” [P13]; “Having some sort of quick tutorial orientation...you have to have that...I find it helpful for most apps, so I understand what it does.” [P7] |
Notifications: alerts and reminders; control over alerts | “I would expect it to give me text updates or notifications, so I don’t have to go into the app. Because if I have to go into the app [to check my progress] then I would be less likely to check...if it alerts me that would be wonderful.” [P2]; “I mean it actually had a feature that you could set reminders. But...I don’t like any notifications turned on...To me, it is always a distraction...it may work for others.” [P5]; “I think it made me more active, especially because this [referring to a wearable device] has little red lights that pop up. So that kind of forced me hey I haven’t walked for a while or maybe I’ll do a walk around the building!” [P13] |
Data collection: data entry convenience; need for extra device | “I mean part of the reason why the step app worked so well was that you literally turn it on it does everything. There isn’t really a lot I need to do to interact with it further.” [P6]; “Yes, that’s the only thing I don’t like right now is that I don’t generally have pockets to carry my phone with me. So, I don’t think it’s accurately reflecting my step count. But if you carry it around it definitely would.” [P12] |
Goal management: setting up goals; notifications about progress | “I thought that was one of their big positives. For this app, I think the customizable side of it and being able to track exactly what I wanted is probably its biggest feature and something that I’ve been missing in other apps.” [P7]; “You can click this and then you can go look at your trends over the past several days where here it’s giving you the hourly trend or weekly ones.” [P16]; “You can click this and then you can go look at your trends over the past several days where here it’s giving you the hourly trend or weekly ones.” [P16]; “The app, for instance, sent me emails saying that ‘You have recorded your nutrition for seven days!’ which I found pretty motivating. Kept me going!” [P8] |
Depth of knowledge: available content; accuracy of data and content; completeness | “If an [nutrition] app had links to websites that explains how to ferment vegetables, or...links to helpful resources or articles recipes would help me more to get there.” [P4]; “They have a lot of information and you can see kind of like during the night if it spikes when you woke up and it was pretty accurate that way and you could feel like a dream journal and put in you know if you had caffeine late and things like that to kind of track if that affects your sleep.” [P7]; “So, the one app I had initially downloaded I thought had too much locked content, and I felt like I didn’t have enough options. So, I deleted that, and I did download the other app.” [P12]; “But there seem to be no consistent rules. It was overly complicated. I’m like I don’t know how kids would play this. And there was no help document for me to read...and it was vastly inconsistent in terms of the content.” [P6]; “I don’t know if it’s a bug or if it’s supposed to be that way but if you have to pause it doesn't work and it's like you didn’t even do it.” [P14] |
System rules: process of the app; clarity of rules and functions | “But there seem to be no consistent rules. It was overly complicated. I’m like I don’t know how kids would play this. And there was no help document for me to read...and it was vastly inconsistent in terms of the content.” [P6]; “I don’t know if it’s a bug or if it’s supposed to be that way but if you have to pause it doesn’t work and it’s like you didn’t even do it.” [P14] |
Actionable recommendations: personalized progress analysis; amount of usage time needed | “It did help me become more conscientious about getting some food in me three times a day at least. Becoming more aware of how many calories I was taking in. So, I could meet my goals.” [P4]; “The only thing that [was needed] is to send me related notifications like ‘2000 steps from your goal for the day’.” [P12] |
Fit between user and system: match between features and user needs | “I think this app would work for a lot of people. For me what they provide value in, like in their add-ons, does not work. I get [the value] in other places already. So, if I didn't do podcasts, that would be a really nice way to introduce you to walking and running.” [P6]; “From the notification, I knew how I was doing. So, it was nice because I wasn’t doing anything extra to get this information. I didn’t have to go in and really use the app where, with Aura [another app], I had to actively go in and open it up and make the three four minutes for each meditation.” [P17] |
Notifications (
Although notifications could help motivate more use, the participants prefer some
Data collection methods and tools utilized by the mHealth apps are expected to be
The analysis showed that efficiency can be achieved when users can quickly interact with a system to perform the intended task. In this study’s context, when an app allows for automatic or quick data entry, it facilitates efficiency in use. This is consistent with the reduction principle in PSD framework [
In cases where manual data entry is unavoidable (eg, tracking diet), users seek features that provide convenient data entry, such as nutrition apps with comprehensive databases that provide nutritional values for a variety of options. Moreover, wearing and
Goal management is a necessary functionality that enables users to reach their goals [
In addition, participants expressed that to be able to follow up with their goals, they needed the app to send regular
Depth of knowledge provided, which refers to the freely
The clarity of the system rules embedded in the technology, in other words the way the system is designed to work, emerged as a major theme in the analysis. Although the specifics may vary in each app and context, users expect to easily understand the process underlying the design of the app. When the
Although the participants used apps with different focuses and features, they stated that to realize the benefits, the apps needed to provide actionable recommendations for improving the current conditions and specify what needed to be done to reach the goals in the intended timeline. This refers to the system’s ability to offer clear next steps that users can follow to reach their goals and is relevant to the tunneling principle in PSD framework, which suggests guiding users during the change process by providing means for action that helps them get closer to their goals [
The
Finally, although every app embeds a different set of features and provides various functionalities, there should be a
The second dimension related to continued use that emerged from the analysis was the intent of the users. In addition to the factors related to the technology being used, motivation of users played a vital role in continuing to use an mHealth app. Having persistence at intended health goals and being able to pursue them despite the likely challenges appeared to play a key role toward continued use of mHealth apps. Behavior change is difficult to achieve, even with the use of persuasive technologies [
The results revealed 2 dimensions that were related to the continued use of mHealth apps. The first dimension considers an overall assessment of user experience and how an mHealth app provides opportunities for reaching health goals through the factors identified in the results section. This subjective assessment made by users can vary from high to low, depending on the extent to which technology is enabling them to achieve the intended goals. For instance, a user may believe an app is a high enabler because it provides notifications, has a simple interface, and allows for automatic data entry techniques, whereas another user may believe the same app is a low enabler because of the insufficient health information the app provides. The second dimension considers the intent of a user by assessing their level of commitment to their health goals. This is usually demonstrated by assessing the extent to which the users exhibit undivided attention and persistent efforts toward achieving goals and could range from low to high.
Considering these 2 dimensions, we identified 4 possible scenarios as an outcome of users’ initial experiences with mHealth apps, which are the decisions to (1) abandon use, (2) limit use, (3) switch app, and (4) continue use.
The decision to abandon quadrant (low assessment and low persistence) represents a situation where users are skeptical about the capability of the mHealth apps they selected, but, at the same time, they do not show persistence at their health goals. In such conditions, we expect a user to abandon the mHealth app before having any meaningful interaction with it. An example is a user who shows willingness to
I feel like having only one means of communication or accountability is not good for me. I think if I'm serious about it, then I need to go to the meetings and be more engaged. Even though the system holds me accountable for it, it was not enough.
The decision to limit use quadrant (high assessment and low persistence) refers to a situation where an enabling mHealth app is available, yet users do not show persistence for pursuing their goals and stop use when they experience any difficulty. An example is a user who reports having a short and intermittent interaction with the selected app, although he or she found the app suitable for reaching the goals. In this situation, we expect users to have a limited use of the app, insufficient for significant improvement toward achieving goals.
The decision to switch quadrant (low assessment and high persistence) refers to a situation where users show commitment toward their goals but find the mHealth app to be a low enabler because of limited features of the app. In such situations, we expect a user to continue use but substitute that app for a better choice. For instance, a respondent admitted that “it [my use] depends on whether I find the app useful or not, because meditation is something that seems really helpful for me and I’d like the idea, but the [app] implementation isn't working so well…so, I'll go try something else and see.” (10). Although substituting can be distracting, it can still provide an opportunity for reaching goals if a better mHealth app is found and then used in the future.
The decision to continue use is indeed the ideal situation, where the app is enabling and users show persistence toward goals. Under this condition, we expect the users to continue engagement with the app for longer periods, which eventually helps them move toward their intended goals.
On the basis of these findings, we developed 4 propositions that describe circumstances associated with the decisions mHealth users make.
Use decision scenarios regarding mobile health app use.
Despite the exponential rate at which new mHealth apps are introduced to market, most users stop usage soon after initial use. The aim of this study was to further our understanding of continued use of mHealth apps. Through the analyses of qualitative data collected via interviews and daily use diaries, we identified key factors that influenced users’ decisions regarding continued use after the initial interaction with an app. Furthermore, based on the degree of users’ assessment of the app and their persistence toward their goals, we highlighted 4 decisions: to abandon use, to limit use, to switch app, and to continue use. We put forth propositions that can guide future research that aims to understand behaviors regarding the use of mHealth apps.
mHealth apps will continue to play a pivotal role in providing individual and customized health care services that can be reached anywhere, any time and at relatively low costs [
More importantly, the findings highlighted the importance of focusing on users’ goals and their commitment to these goals. Although motivation is shown to be sufficient for adoption of mHealth [
We acknowledge that this study has limitations. First, more than half of the participants were female, used an iPhone, and were highly motivated to take care of their health. These characteristics may have influenced the way they interacted with, and made decisions about, the mHealth app. Including larger dataset in a population (eg, balanced male/female; iPhone/Android; and motivated/unmotivated) will help improve the generalizability of the findings. Second, although we used 2 methods to collect longitudinal data, the provided information was self-reported and did not include objective measures. Using system log data in future studies may help provide additional insights on continued use of mHealth apps. Third, we focused on a limited range of mHealth apps (ie, health and wellness) as representative of mHealth apps. Future research is needed to replicate and extend the results to other contexts to have a more inclusive view of the continued use of mHealth. Fourth, participants were recruited in universities (although only 1 was a student) and received compensation for their participation.
In addition, we sent a daily reminder to participants to fill out their use diary, which could have influenced their interaction with the app and encouraged continued use. In the same vein, we acknowledge that other factors could influence continued use of mHealth app. For instance, previous research has shown the importance of privacy regarding adoption and use of mHealth app [
Finally, we note that the data focused on continued use as one way to successfully change the health behavior of individuals, yet we did not directly assess the behavior change of users per se. This presents a promising avenue for additional studies, for instance, using longer scope and more comprehensive data collection methods that pay specific attention to the relationship between continued mHealth use and health behavior change to assess how the former behavior instigates the latter.
A diary response example.
Coding structure in qualitative analysis.
identification number
information system
information technology
mobile health
persuasive system design
participant X
None declared.