Abstract
This study offers a description of factors that predict the adoption of mobile health technologies (mHealth) and their application for health self-management in emerging adults. Primary data collection occurred at three diverse postsecondary educational institutions (N= 1,329). The analysis used a logistic regression to identify predictors of mHealth adoption. Descriptive analyses are presented on health self-management applications and perceived ease of use and effectiveness. Use of mHealth was high in respondents (58.5%). Factors associated with increased likelihood of mHealth adoption included being female, overweight or obese, having a chronic condition, eating the recommended amount of daily fruit, and engaging in regular moderate exercise. Low household income was associated with being less likely to use mHealth. The most common self-management application for mHealth was for tracking physical activity. Findings related to ease of use and effectiveness ratings by applications may provide insight into designing more effective mHealth tools in this population.
Introduction
The development of digital technologies is spurring the expansion of mHealth, broadly defined as “health care and public health practices supported by mobile devices”.1 mHealth can be thought of as a subcategory of telehealth and includes health apps or mobile phone capabilities that enable immediate access to biometric data, health information, patient education, adherence reminders, or progress on a lifestyle modification goal.2 The prevalence of consumer mHealth apps has grown dramatically. In 2017 there were over 318,000 health apps available for download, doubling the number available in 2015.3 This growth is due, in part, to the increasing prevalence of smart phones and watches. As of 2018, the vast majority of individuals in the United States have a cell phone (95%) and smartphone ownership grew dramatically from 35% in 2011 to 77% in 2018.4
There is evidence of promising potential for mHealth to improve health behaviors and help individuals self-manage chronic conditions. A systematic review of such interventions based primarily on randomized controlled trials found that mHealth intervention participants were more successful in changing a variety of health behaviors and behavior-related outcomes, including physical activity, smoking cessation, consuming a healthy diet, weight loss, medication adherence, improved blood pressure control, and better glycemic control.1 Despite the potential benefits of mHealth, there are also significant barriers to widespread dissemination. Recent evidence suggests that as many as 45% of participants in a survey study reported that they downloaded health apps they no longer use.5 Furthermore, a 2017 study that explored sustained use for activity trackers found that, on average, participants used trackers for only 129 days.6 A separate study estimated that as many as one third of activity trackers were discontinued within 6 months of use.7 Successful adoption of mHealth will require a better understanding of the preferences of users and an emphasis on design to encourage sustained and effective use.8
With the increasing ubiquity of mHealth and health information technology, it is important to understand its use and impact on health, particularly for younger generations. Emerging adulthood is a distinct development period that occurs between 18 years and 25-30 years marked by an increase in autonomy of decision making and identity formation,9,10 including health behavior patterns that have long-term ramifications on the development or mitigation of chronic disease. Emerging adulthood is often associated with the height of health and well-being; however, recent trends indicate that the rate at which younger adults in the United States are being diagnosed with chronic medical and mental health conditions has grown dramatically.11 As a result, the periods of young and emerging adulthood provide opportunities for emphasizing primary prevention efforts to improve long term health outcomes.12 Enduring patterns of health behaviors and lifestyle choices are established during emerging adulthood, providing evidence that this period is uniquely relevant to health promotion and disease prevention efforts.13
mHealth technologies, whether consumer chosen or health provider directed, will likely play an increasingly important role in health care. To this end, health care improvement efforts have focused on implementing approaches to care that leverage mHealth to support health self-management.14 Health self-management consists of the broad set of skills, knowledge, and abilities to manage one’s health and health care decision making across the spectrum from health enhancement or prevention to disease management.15,16As a cost-effective strategy to improve health self-management,.14 understanding predictors of mHealth technology use and factors associated with their effectiveness is becoming increasingly important .14 To this end, the technology acceptance model sheds light on how users accept and use a technology like mHealth. In addition to important contextual factors, perceived ease of use and effectiveness of a technology influences how and whether the user will adopt a specific technology.17,18
Several distinct characteristics have been noted among users of mHealth. Users are younger,2,5,19 wealthier,5,20 more educated,2,19 and work full time.2 The increased use of health apps was associated with the prevalence of chronic conditions, including obesity.2,5 Fitness and nutrition have been identified as the most commonly reported categories of health apps used.5 Of respondents that use a fitness or nutrition app, most use them at least daily.5 These findings across the literature were based on samples of adults and did not include a sub-group analysis on younger emerging adults.
The goal of this study was to define factors associated with mHealth use, to describe the intended health self-management applications, and determine their ease of use and effectiveness among emerging adults attending three different institutions of post-secondary education in the southeastern United States. The three primary research questions in this study were: 1) What contextual factors, behaviors, and characteristics of emerging adults are predictive of the likelihood of using mHealth? 2) For emerging adults that use mHealth, what health self-management activities are they used for? 3) What is the perceived ease of use and effectiveness of the mHealth for each health self-management activity?
There are notable gaps in the literature surrounding mHealth use in emerging adulthood. To address them, this study is the first to examine predictors of mHealth utilization among emerging adulthood including describing health self-management applications of mHealth and evaluating perceived effectiveness and ease of use across mHealth self-management applications. The findings will help inform the design and implementation of mHealth enabled prevention efforts in this critical population.
Materials and Methods
Study design and subjects
This data was collected from three different institutions of post-secondary education in a medium sized city located in the southeastern United States. The post-secondary educational institutions surveyed include a historically black college (HBCU), a R1 private university, and a technical community college. All three institutions collaborated on a community engaged and participatory research initiative to inform and improve population health management efforts. Each institution provided the research team with school e-mail addresses from either a random sample of their student body or the entire student body. Students received three e-mail prompts over two weeks from the research team to participate in the study and complete the survey through the Qualtrics web platform.21 After the survey was completed, students could enter a lottery to win a gift card to a popular online retailer. The study team had a response rate of 18%, which is typical of online surveys sent to students.22 Inclusion criteria required the respondents be currently enrolled at one of the participating institutions and have an active institutionally affiliated e-mail address. Respondents over the age of 30 or under the age of 18 were excluded from the analysis to focus on the emerging adulthood period of development. This study was approved by the Duke University Institutional Review Board.
Data collection
The survey captured a breadth of information from respondents including demographics, health behaviors, social determinants of health, and information on consumer/commercial oriented mHealth use. mHealth use was collected by asking respondents, “In the past year, how many health-related smartphone apps and/or wearable devices (example: dieting application, FitBit, Apple Watch, sleep tracking) have you used related to your health?”. Respondents provided information on age, gender, race, self-reported family income, and the degree they were pursuing. Data on daily fruit and vegetable in-take and moderate and vigorous exercise were collected using adapted items from the International Physical Activity Questionnaire.23 Health status was measured using a validated 1-item measure of self-reported health24, Body Mass Index (BMI), a validated screening tool for clinical depression25, and a self-reported item that asked respondents to indicate any current chronic condition diagnoses. Finally, primary care utilization was measured by asking respondents the frequency of getting routine outpatient care (“How long has it been since you last visited a healthcare provider for a routine checkup or yearly physical?”).
Respondents who indicated they used one or more apps/devices were asked to report what mHealth was used for by selecting a self-management category. Response options for self-management activities included: (1) to track how much activity/exercise I get (e.g. walking, running, cycling); (2) to help me watch what I eat (e.g. food intake, nutritional values); (3) to show or teach me exercises or workouts; (4) to track a health measure (e.g. heart rate, calories burned, blood pressure); (5) to help me manage a specific condition or behavior (e.g. diabetes, smoking cessation); (6) sleep tracking; (7) reproductive health (e.g. ovulation tracking, family planning, pregnancy progression); (8) stress/anxiety management, mindfulness, guided meditation. Respondents were asked additional questions about their experience using mHealth for each selected self-management activity. Of specific interest was self-reported ratings of ease of use and usefulness/effectiveness which have been shown to be predictive of intention to use, self-reported usage, and attitude toward technology use across a broad array of technologies.26,27
Statistical analysis
A logit model was used to estimate the effect of included variables on the probability of using mHealth. The results are estimates of the average marginal effect of each variable on the likelihood of self-reported mHealth adoption. This approach was chosen given that the outcome (mHealth use) is binary and the strength of this dataset is the richness of covariates and diversity of the respondents. The primary hypothesis was higher BMIs would be associated with an increased likelihood of using mHealth as existing literature suggests that this association is present in the adult population.2,5,19
The analysis also explored what self-management applications were supported by the use of mHealth. The descriptive analysis of the subgroup of emerging adults that reported using mHealth technology illustrates how these technologies were being used to manage health and promote specific health behaviors. Given existing literature examining how mHealth is used,19,5 the secondary hypothesis is that the two most common uses for mHealth technologies would be to improve nutrition and physical fitness.
Finally, descriptive statistics were calculated of perceived effectiveness and ease of use of mHealth technologies for each self-management activity to predict intention to use and sustained use17,28 Ease of use and effectiveness ratings were analyzed to provide insights into the categories of mHealth enabled self-management activities that were the most likely to have an impact on long-term lifestyle modification.
Statistical analyses were performed by using Stata statistical software (version 15.1, StataCorp LLC, College Station, TX, USA).29
Results
1,329 students were included in the analysis. A strength of this dataset is the diversity of respondents. 21.2% were African American, 40.5% White, 16.9% Asian, and 10.9% Hispanic (Table 1). Across the three post-secondary education institutions, 70.1% of the sample was female. The proportion of students that were overweight or obese was highest at the historically black college. The private university had the highest mean age and self-reported household income. mHealth use was prevalent and the difference in mHealth use was not statistically significant between the three institutions: private college (60.6%), technical community college (55.6%), and historically black college (56.4%).
Table 1.
Description of population/analytic sample
| Private University |
Technical Community College |
Historically Black College | Total | |
| N | 782 | 248 | 299 | 1,329 |
| Demographics | ||||
| Female | 505 (64.7%) | 180 (73.2%) | 246 (82.6%) | 931 (70.1%) |
| > 24 years | 373 (47.7%) | 96 (38.7%) | 88 (29.4%) | 557 (41.9%) |
| Married or in a relationship | 352 (45.5%) | 101 (40.7%) | 115 (38.9%) | 568 (42.7%) |
| Race | ||||
| White | 395 (51.4%) | 99 (39.9%) | 44 (15.0%) | 538 (40.5%) |
| African American | 34 (4.4%) | 54 (21.8%) | 194 (66.0%) | 282 (21.2%) |
| Asian | 200 (26.0%) | 17 (6.9%) | 7 (2.4%) | 224 (16.9%) |
| Hispanic | 73 (9.5%) | 55 (22.2%) | 17 (5.8%) | 145 (10.9%) |
| Other | 66 (8.6%) | 23 (9.3%) | 32 (10.9%) | 121 (9.1%) |
| Social Determinants | ||||
| Education | ||||
| Vocational or Associate’s degree | 6 (0.8%) | 206 (83.1%) | 1 (0.3%) | 213 (16.0%) |
| Bachelor’s degree | 242 (30.9%) | 18 (7.3%) | 225 (75.3%) | 485 (36.5%) |
| Graduate degree | 520 (66.5%) | 0 (0.0%) | 66 (22.1%) | 586 (44.1%) |
| Annual Income | ||||
| < $20,000 | 58 (7.8%) | 42 (17.9%) | 73 (25.6%) | 173 (13.0%) |
| $20,000 –$49,999 | 106 (14.2%) | 90 (38.3%) | 98 (34.4%) | 294 (22.1%) |
| $50,000 –$99,999 | 186 (24.9%) | 64 (27.2%) | 80 (28.1%) | 330 (24.8%) |
| > $100,000 | 397 (53.1%) | 39 (16.6%) | 34 (11.9%) | 470 (35.4%) |
| Health Status | ||||
| Overweight or obese | 196 (25.1%) | 110 (44.4%) | 158 (52.8%) | 464 (34.9%) |
| Any chronic illness | 330 (42.2%) | 139 (56.0%) | 139 (46.5%) | 608 (45.7%) |
| Depression | 95 (12.1%) | 58 (23.4%) | 67 (22.4%) | 220 (16.6%) |
| Excellent/Very Good Self- Reported Health | 507 (64.8%) | 124 (50.2%) | 150 (50.3%) | 781 (58.8%) |
| Health Behaviors | ||||
| mHealth use | 472 (60.6%) | 138 (55.6%) | 168 (56.4%) | 778 (58.5%) |
| PCP Visit within past year | 572 (73.9%) | 177 (71.7%) | 233 (78.5%) | 982 (73.9%) |
| Vegetable, Recommended daily intake | 276 (35.5%) | 78 (31.5%) | 58 (19.4%) | 412 (31.0%) |
| Fruit, Recommended daily intake | 422 (54.0%) | 126 (50.8%) | 156 (52.2%) | 704 (53.0%) |
| Moderate Exercise | 202 (25.8%) | 43 (17.3%) | 56 (18.7%) | 301 (22.6%) |
| Vigorous Exercise | 95 (12.1%) | 11 (4.4%) | 21 (7.0%) | 127 (9.6%) |
A logit model was used to predict the likelihood of mHealth adoption based on the variables included in the model (Table 2). These results indicated that being overweight or obese is associated with a 10.5 percentage point increase in the predicted probability of using mHealth technology controlling for the other covariates included in the model. This finding offered support for the hypothesis that being overweight would be associated with a higher likelihood of using mHealth technologies.
Table 2.
Logit model of marginal effects on mHealth adoption
| Variables | Logit Coefficients (SE) | Logit Marginal Effects (SE) | 95% Confidence Interval |
| Demographics | |||
| Female | 0.655*** (0.138) | 0.148*** (0.031) | [0.087, 0.208] |
| > 24 years | 0.044 (0.148) | 0.010 (0.033) | [-0.054, 0.074] |
| Married or in a relationship | 0.151 (0.126) | 0.033 (0.028) | [-0.021, 0.088] |
| Race | |||
| African American | -0.163 (0.197) | -0.036 (0.044) | [-0.123, 0.050] |
| Hispanic | 0.041 (0.223) | 0.009 (0.049) | [-0.087, 0.105] |
| Asian | -0.049 (0.186) | -0.011 (0.041) | [-0.092, 0.070] |
| Other | -0.074 (0.223) | -0.016 (0.050) | [-0.115, 0.082] |
| Social Determinants | |||
| Education | |||
| Vocational or Associate’s degree | -0.054 (0.189) | -0.012 (0.042) | [-0.094, 0.070] |
| Graduate degree | 0.118 (0.166) | 0.026 (0.037) | [-0.046, 0.098] |
| Annual Income | |||
| < $20,000 | -0.585** (0.205) | -0.133* (0.047) | [-0.225, -0.041] |
| $20,000 – $49,999 | -0.031 (0.176) | -0.007 (0.039) | [-0.083, 0.069] |
| $50,000 – $99,999 | ref | ref | ref |
| > $100,000 | 0.176 (0.164) | 0.039 (0.036) | [-0.032, 0.110] |
| Health Status | |||
| Overweight or obese | 0.484*** (0.138) | 0.105*** (0.029) | [0.048, 0.163] |
| Any chronic illness | 0.451** (0.131) | 0.100*** (0.029) | [0.044, 0.156] |
| Depression | -0.295 (0.168) | -0.066 (0.038) | [-0.140, 0.008] |
| Excellent/Very Good Self-Reported Health | 0.491*** (0.133) | 0.110*** (0.030) | [0.052, 0.168] |
| Health Behaviors | |||
| PCP Visit within past year | -0.018 (0.142) | -0.004 (0.031) | [-0.065, 0.057] |
| Vegetable, Recommended daily intake | 0.172 (0.144) | 0.038 (0.031) | [-0.024, 0.100] |
| Fruit, Recommended daily intake | 0.278* (0.128) | 0.062* (0.028) | [0.006, 0.117] |
| Moderate Exercise | 0.471** (0.163) | 0.102** (0.034) | [0.035, 0.169] |
| Vigorous Exercise | -0.244 (0.234) | -0.054 (0.052) | [-0.157, 0.048] |
| Constant | -1.006*** (.278) | - | - |
| Standard errors are presented in parentheses; Asterisks denote 5% (*) 1% (**) and .1% (***) significance; Pseudo R2 = 0.0683; Log likelihood = -783.177 | |||
Being female was associated with an over 14 percentage point increase in the predicted probability of using mHealth controlling for other variables in the model. Respondents from households under $20,000 were less likely to use mHealth when compared to respondents from households with incomes between $50,000-$99,999. Other income levels were not associated with a higher or lower probability of using mHealth. The endorsement of healthy behaviors also predicted mHealth use. Respondents that reported engaging in the recommended amount of moderate physical exercise according to the American Heart Association30 were more likely to use mHealth. In addition, eating the recommended daily intake of fruits is associated with a 6.2 percentage point increase in the likelihood of using mHealth.
Finally, there were associations between using mHealth across multiple measures of health status. Having a chronic illness was associated with using mHealth as was self-reported health of ‘Very Good’, or ‘Excellent’ which was associated with an 11-percentage point increase in the predicted probability of using mHealth. There was some preliminary evidence that suggests having screened positive for clinical depression is associated with a 6.6 percentage point decrease in the likelihood that a respondent used mHealth; however, this result was significant at a 90% confidence level. The other variables included in the model (vegetable consumption, relationship status, age, race, recommended vigorous exercise, annual primary care visit, or degree type) did not affect the likelihood of mHealth technology adoption.
Respondents that reported using mHealth were asked follow up survey questions to capture how mHealth technologies were used and their perceived ease of use and effectiveness for each health self-management application (Tables 3 and 4).
Table 3.
Effectiveness of mHealth technology by self-management activity
| Effectiveness | |||||||
| mHealth Function | Prevalence |
Not effective at all (5) |
Slightly effective (4) |
Moderately effective (3) |
Very effective (2) |
Extremely effective (1) | Average (SD) |
| Activity and Exercise Tracking | 645 (82.9%) | 25 (3.9%) | 85 (13.2%) | 233 (36.1%) | 185 (28.7%) | 117 (18.1%) | 2.56 (1.05) |
| Nutrition Tracking | 311 (39.9%) | 18 (5.8%) | 46 (14.8%) | 122 (39.2%) | 73 (23.5%) | 52 (16.7%) | 2.69 (1.09) |
| Show Exercise and Workouts | 276 (35.5%) | 9 (3.3%) | 26 (9.4%) | 99 (35.9%) | 85 (30.8%) | 57 (20.7%) | 2.43 (1.02) |
| Health Measure Tracking | 273 (35.1%) | 5 (1.8%) | 15 (5.5%) | 105 (38.5%) | 84 (30.8%) | 64 (23.4%) | 2.32 (0.95) |
| Sleep Tracking | 252 (32.4%) | 10 (4.0%) | 36 (14.3%) | 93 (36.9%) | 73 (29.0%) | 40 (15.9%) | 2.62 (1.04) |
| Manage Reproductive Health | 151 (19.4%) | 3 (2.0%) | 5 (3.3%) | 30 (19.9%) | 50 (33.1%) | 63 (41.7%) | 1.91 (0.96) |
| Manage Stress or Anxiety | 105 (13.5%) | 11 (10.5%) | 21 (20.0%) | 36 (34.3%) | 27 (25.7%) | 10 (9.5%) | 2.96 (1.13) |
| Manage Condition or Behavior | 24 (3.1%) | 2 (8.3%) | 2 (8.3%) | 10 (41.7%) | 4 (16.7%) | 6 (25.0%) | 2.58 (1.21) |
Table 4.
Ease of use of mHealth technology by self-management activity
| Ease of Use | |||||||
| mHealth Function | Prevalence | Extremely difficult (5) | Moderately difficult (4) |
Neither easy nor difficult (3) |
Moderately easy (2) | Extremely easy (1) | Average (SD) |
| Activity and Exercise Tracking | 646 (83.0%) | 3 (0.5%) | 8 (1.2%) | 60 (9.3%) | 234 (36.2%) | 341 (52.8%) | 1.60 (0.75) |
| Nutrition Tracking | 313 (40.2%) | 6 (1.9%) | 24 (7.7%) | 60 (19.2%) | 135 (43.1%) | 88 (28.1%) | 2.12 (0.96) |
| Show Exercise and Workouts | 275 (35.3%) | 1 (0.4%) | 6 (2.2%) | 24 (8.7%) | 116 (42.2%) | 128 (46.5%) | 1.68 (0.76) |
| Health Measure Tracking | 274 (35.2%) | 0 (0.0%) | 1 (0.4%) | 20 (7.3%) | 92 (33.6%) | 161 (58.8%) | 1.49 (0.65) |
| Sleep Tracking | 252 (32.4%) | 1 (0.4%) | 6 (2.4%) | 20 (7.9%) | 85 (33.7%) | 140 (55.6%) | 1.58 (0.77) |
| Manage Reproductive Health |
152 (19.5%) | 2 (1.3%) | 2 (1.3%) | 5 (3.3%) | 42 (27.6%) | 101 (66.5%) | 1.43 (0.74) |
| Manage Stress or Anxiety | 106 (13.6%) | 1 (0.9%) | 3 (2.8%) | 14 (13.2%) | 40 (37.7%) | 48 (45.3%) | 1.76 (0.86) |
| Manage Condition or Behavior | 24 (3.1%) | 0 (0.0%) | 0 (0.0%) | 5 (20.8%) | 6 (25.0%) | 13 (54.2%) | 1.67 (0.82) |
Physical activity and nutrition tracking were common self-management applications, lending support for the hypothesis that exercise and nutrition would be common uses of mHealth. However, despite the prevalence of mHealth technologies to assist with nutrition tracking, respondents indicated that the effectiveness for this function was lower than other self-management functions with 40.2% indicating nutrition tracking apps were extremely or very effective and only 28.1% reporting that they were extremely easy to use. Respondents, however, found mHealth was especially useful for managing reproductive health with over 74.8% of respondents reporting that it was extremely or very effective and over 66.5% reporting that it was extremely easy to use. Respondents were less likely to find mHealth useful for managing stress or anxiety with only 35.2% reporting that mHealth was extremely or very effective. However, mHealth for managing stress or anxiety had relatively high ratings for ease of use with 83% of respondents indicating moderately or extremely easy to use.
The overall relationship of the ease of use and effectiveness ratings based on self-management is shown in Figure 1. mHealth self-management applications with higher effectiveness ratings also tended to have higher ease of use ratings. Ease of use and effectiveness ratings for each self-management category had a statistically significant positive correlation ranging from .29 to .47 with the exception of ‘Manage a Condition or Behavior’, likely due to the small sample size. For example, applications like reproductive health and health measure tracking had high levels of effectiveness and also had high levels of ease of use when compared to other applications. Similarly, nutrition tracking had lower levels of both effectiveness and ease of use. However, effectiveness tended to be rated much lower than ease of use. This was especially true for managing stress and anxiety, which had low levels of effectiveness but high levels of ease of use.
Figure 1.
mHealth effectiveness and ease of use ratings of self-management activity ordered by effectiveness
Discussion
Facilitating effective mHealth utilization within emerging adulthood is increasingly being recognized as an opportunity to promote health self-management and improve health outcomes later in life.13 As health care delivery models assume more responsibility for managing the health of defined populations and communities,31 novel ways that leverage available mHealth technologies to promote health self-management is a key consideration. This study confirms previous literature that indicates that younger adults are more likely to use mHealth technologies and provides insights into the determinants that predict mHealth adoption and the health self-management activities that these technologies support.
mHealth use was much higher in this sample of emerging adults (58.5%) than among a nationally representative sample of respondents found in relevant recent studies with mHealth usage rates of 22.8%19 in 2017 and 11.7% in 201220. The dramatic difference could be attributed to the focus on a younger population and the continuation of a general trend towards greater mHealth use. While mHealth use was ubiquitous among respondents, these findings suggest that a key barrier to mHealth adoption in emerging adulthood is related to socio-economic status. Participants from low income families (<$20,000) had a 13.3 percentage point lower likelihood of mHealth adoption when compared to middle income families ($50,000-$99,999). Previous research suggests that the burden of cost associated with mHealth and lower technology literacy levels are potential explanations for this finding.32
Being overweight or obese and having any chronic illness is associated with higher likelihood of mHealth adoption. This could be a result of these individuals seeking technologies to manage their condition. This is consistent with a previous finding that mHealth users were more likely to be obese.5 These results also reveal that being in excellent or very good self-reported health was associated with an increased likelihood of mHealth adoption. This is notable given that self-reported health is predictive of health care expenditures and mortality.24
Female emerging adult respondents were more likely to adopt mHealth technologies, a finding that has been replicated in older populations.19,20 A novel finding of this analysis suggests that being at risk of depression was associated with a lower likelihood of using mHealth, however, this result was only significant at a 90% confidence level, thus further research is required to confirm this relationship. Finally, this research provides evidence that there is a relationship between mHealth adoption and the endorsement of recommended beneficial health behaviors. Engaging in the recommended amount of moderate exercise and fruit intake were both associated with a higher likelihood of using mHealth. This is an area of particular promise and further research is required to evaluate the causal direction of this association and to explore the potential implications for designing more effective mHealth interventions.
This research is the first to describe the health self-management activities that mHealth technologies are used for and provide user feedback on perceived effectiveness and ease of use based on the technology acceptance model. Perceived effective and ease of use are predictive of intention to use health related technologies, a key consideration given that user acceptance is important to enable health informatics adoption in health service delivery.5,18 Effectiveness and ease of use have been shown to be influenced by individual and community contextual factors such as gender, culture, social determinants, and demographic characteristics.17 It will be important to also understand how these determinants are associated with continued use, and more importantly, positive health outcomes. Overall, effectiveness ratings tended to be strikingly lower than ease of use ratings across all self-management activities. This illustrates the challenge of designing mHealth to be effective for behavior change, it requires more than the technology being accessible and user-friendly. It is important to note that the most commonly used self-management activities were not the activities that respondents indicated to be the highest levels of effectiveness and ease of use. Nutrition tracking, for example, was the second most prevalent self-management activity but had lower effectiveness and ease of use ratings when compared to other self-management applications. Conversely, reproductive health self-management had lower rates of mHealth adoption but had the highest effectiveness and ease of use ratings. A possible explanation for this phenomenon is that respondents found that passive mHealth technologies that tracked activity or a health measure to be both effective and easy to use when compared to a self-management activity like stress management and nutrition tracking which requires greater user engagement.
Future directions for this work could create a more specific “user profile” and risk stratification to target mHealth interventions to users that are receptive to this medium of clinical or public health interventions. To do so, special consideration will be paid to the design of mHealth enabled self-management interventions to maximize effectiveness and the user experience. As opportunities to implement mHealth enabled prevention efforts that target emerging adults are explored, researchers and practitioners alike must continue to evaluate opportunities to personalize mHealth interventions and delivery models to realize the full potential health benefits.
Limitations:
This work has several notable limitations. First, given the relatively modest, but not unexpected response rate, there should be caution when generalizing these results to all emerging adults in post-secondary institutions. As with many observational cross-sectional survey study designs, a self-selection bias could make the study sample systematically different from the population of interest. The respondents were all located in the same southeastern city which limits external validity to other geographic regions. The lack of longitudinal data limits the ability to determine the extent of sustained use of mHealth technologies and whether they fulfilled their intended outcome – a key consideration for mHealth to modify health behaviors.
Conclusion
This study provides important insights into mHealth use and their perceived ease of use and effectiveness by self-management application in emerging adulthood in diverse post-secondary institutions of learning. mHealth use is high among respondents when compared to the general population.5,20 Higher likelihood of mHealth adoption was associated with being overweight or obese, female, engaging in moderate exercise, eating the recommended amount of daily fruit, having a chronic condition, and higher levels of self-reported health. In contrast, low household income and screening positive for depression risk (90% confidence level) was associated with a decreased likelihood of mHealth adoption. Tracking exercise and nutrition were the most common applications of mHealth. Ease of use and effectiveness were the highest for mHealth self-management applications associated with reproductive health and health measure tracking and the lowest for managing stress and anxiety. A trend that emerged in the data is that passive self-management functions received higher ease of use and effectiveness ratings. Ease of use and effectiveness ratings were largely correlated with ease of use being consistently rated higher than effectiveness across self-management activities. This finding suggests that an mHealth technology that is easy to use is necessary but not necessarily sufficient to have an impact on health behaviors. Next steps for this work should focus on opportunities to design mHealth interventions that are tailored to the preferences and needs of emerging adults to support health self-management during this critical developmental period.
Funding
This research was supported by the Duke Center for Personalized Health Care and by the Duke Clinical and Translational Science Award, NIH Award number UL1TR001117.
Acknowledgements
We gratefully acknowledge Cindy Mitchell for assistance preparing and editing this manuscript.
Figures & Table
References
- 1.Hamine S, Gerth-Guyette E, Faulx D, Green BB, Ginsburg AS. Impact of mhealth chronic disease management on treatment adherence and patient outcomes: a systematic review. J Med Internet Res. 2015;17((2)) doi: 10.2196/jmir.3951. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ernsting C, Dombrowski SU, Oedekoven M, et al. Using smartphones and health apps to change and manage health behaviors: a population-based survey. J Med Internet Res. 2017;19((4)):101. doi: 10.2196/jmir.6838. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.The Growing Value of Digital Health Evidence and Impact on Human Health and the Healthcare System. IQVIA Institute; :2017. [Google Scholar]
- 4.Center PR. Mobile fact sheet. Pew Research Center: Internet, Science & Tech. 2017 [Google Scholar]
- 5.Krebs P, Duncan DT. Health app use among us mobile phone owners: a national survey. JMIR Mhealth Uhealth. 2015;3((4)) doi: 10.2196/mhealth.4924. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hermsen S, Moons J, Kerkhof P, Wiekens C, De Groot M. Determinants for sustained use of an activity tracker: observational study. JMIR Mhealth Uhealth. 2017;5((10)):e164. doi: 10.2196/mhealth.7311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Dan Ledger DM. Inside wearables: how the science of human behavior change offers the secret to longterm engagement. 2014 Available from: https://medium.com/@endeavourprtnrs/inside-wearable-how-the-science-of-human-behavior-change-offers-the-secret-to-long-term-engagement-a15b3c7d4cf3. [Google Scholar]
- 8.Steinhubl SR, Muse ED, Topol EJ. Can mobile health technologies transform health care? JAMA. 2013;310((22)):2395–2396. doi: 10.1001/jama.2013.281078. [DOI] [PubMed] [Google Scholar]
- 9.Arnett JJ. Emerging adulthood: a theory of development from the late teens through the twenties. Am Psychol. 2000;55((5)):469. [PubMed] [Google Scholar]
- 10.Schwartz SJ, Cote JE, Arnett JJ. Identity and agency in emerging adulthood two developmental routes in the individualization process. Youth Soc. 2005;37((2)):201–229. [Google Scholar]
- 11.Blum RW. Transition to adult health care: setting the stage. J Adolesc Health. 1995;17((1)):3–5. doi: 10.1016/1054-139X(95)00073-2. [DOI] [PubMed] [Google Scholar]
- 12.Frech A. Pathways to adulthood and changes in health-promoting behaviors. Adv Life Course Res. 2014;19:40–49. doi: 10.1016/j.alcr.2013.12.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Nelson MC, Story M, Larson NI, Neumark-Sztainer D, Lytle LA. Emerging adulthood and college-aged youth: an overlooked age for weight-related behavior change. Obesity. 2008;16((10)):2205–2211. doi: 10.1038/oby.2008.365. [DOI] [PubMed] [Google Scholar]
- 14.Katz R, Mesfin T, Barr K. Lessons from a community-based mhealth diabetes self-management program:“It’s not just about the cell phone”. J Health Commun. 2012;17((sup1)):67–72. doi: 10.1080/10810730.2012.650613. [DOI] [PubMed] [Google Scholar]
- 15.Lorig KR, Holman HR. Self-management education: history, definition, outcomes, and mechanisms. Ann Behav Med. 2003;26((1)):1–7. doi: 10.1207/S15324796ABM2601_01. [DOI] [PubMed] [Google Scholar]
- 16.Grady PA, Gough LL. Self-management: a comprehensive approach to management of chronic conditions. Am J Public Health. 2014;104((8)):e25–e31. doi: 10.2105/AJPH.2014.302041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sun H, Zhang P. The role of moderating factors in user technology acceptance. Int J Hum Comput Stud. 2006;64((2)):53–78. [Google Scholar]
- 18.Rahimi B, Nadri H, Afshar HL, Timpka T. A systematic review of the technology acceptance model in health informatics. Appl Clin Inform. 2018;9((03)):604–634. doi: 10.1055/s-0038-1668091. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Carroll JK, Moorhead A, Bond R, LeBlanc WG, Petrella RJ, Fiscella K. Who uses mobile phone health apps and does use matter? A secondary data analytics approach. J Med Internet Res. 2017;19((4)) doi: 10.2196/jmir.5604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kontos E, Blake KD, Chou W-YS, Prestin A. Predictors of ehealth usage: insights on the digital divide from the health information national trends survey 2012. J Med Internet Res. 2014;16((7)) doi: 10.2196/jmir.3117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Qualtrics I. Qualtrics. Provo, UT, USA. 2013 [Google Scholar]
- 22.Cook C, Heath F, Thompson RL. A meta-analysis of response rates in web-or internet-based surveys. Educ Psychol Meas. 2000;60((6)):821–836. [Google Scholar]
- 23.Fogelholm M, Malmberg J, Suni J, et al. International physical activity questionnaire: validity against fitness. Med Sci Sports Exerc. 2006;38((4)):753–760. doi: 10.1249/01.mss.0000194075.16960.20. [DOI] [PubMed] [Google Scholar]
- 24.DeSalvo KB, Bloser N, Reynolds K, He J, Muntner P. Mortality prediction with a single general self-rated health question. J Gen Intern Med. 2006;21((3)):267. doi: 10.1111/j.1525-1497.2005.00291.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Arroll B, Goodyear-Smith F, Crengle S, et al. Validation of phq-2 and phq-9 to screen for major depression in the primary care population. Ann Fam Med. 2010;8((4)):348–353. doi: 10.1370/afm.1139. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Szajna B. Software evaluation and choice: Predictive validation of the technology acceptance instrument. MIS Q. 1994:319–324. [Google Scholar]
- 27.Sengars A, Grover V. Re-examining perceived ease of use and usefulness: a confirmatory factor analysis. MIS Q. 1993;17((4)):7. [Google Scholar]
- 28.Venkatesh V. Determinants of perceived ease of use: integrating control, intrinsic motivation, and emotion into the technology acceptance model. Inf Syst Res. 2000;11((4)):342–365. [Google Scholar]
- 29.StataCorp L. Stata statistical software. College Station TX. 2009 [Google Scholar]
- 30.Haskell WL, Lee I-M, Pate RR, et al. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Circulation. 2007;116((9)):1081. doi: 10.1161/CIRCULATIONAHA.107.185649. [DOI] [PubMed] [Google Scholar]
- 31.Alley DE, Asomugha CN, Conway PH, Sanghavi DM. Accountable health communities—addressing social needs through medicare and medicaid. N Engl J Med. 2016;374((1)):8–11. doi: 10.1056/NEJMp1512532. [DOI] [PubMed] [Google Scholar]
- 32.Martin T. Assessing mhealth: opportunities and barriers to patient engagement. J Health Care Poor Underserved. 2012;23((3)):935–941. doi: 10.1353/hpu.2012.0087. [DOI] [PubMed] [Google Scholar]