A Comprehensive Profile of Those Who Have Health-Related Apps

Abstract

This study investigated sociodemographic, health-related, technological, and motivational factors associated with having health-related apps. Focusing on motivational factors, this study chose five general healthy intentions (about fruit, vegetable, and soda intake, weight control, and amount of exercise) and examined whether those with intention to change their current state (change group) differ from those who want to maintain (maintain group) or pay no attention to their current state (no attention group). A secondary analysis of data from the Health Information National Trends Survey 4 (Cycle 4), collected from a representative sample of U.S. adults aged 18 years or older, was conducted. Only responses from Internet users were analyzed (N = 2,802). Regarding sociodemographics, younger individuals and those with higher income were more likely to have health apps. Hispanics and the less educated were less likely to have health apps. Also, technological factors, such as smartphone and Wi-Fi use, were associated with having apps. Regarding motivational factors, Model 1 (no attention group as a reference group) showed differences in having apps between those who wanted to change their weight and those who paid no attention. Model 2 (maintain group as a reference group) revealed a difference between those who tried to change the amount of exercise and those who maintained it. The findings provided a comprehensive profile of those with health apps in the United States: non-Hispanic young people with higher income, higher education, a smartphone, and a Wi-Fi connection who want to change (but not maintain) their weight and amount of exercise.

Keywords

exercise health motivation health-related apps mobile technology weight control

Mobile health (mHealth) refers to public or medical health practices in the age of mobile technology, such as the use of mobile devices to seek health information, communicate with healthcare professionals, and monitor individual health (Becker et al., 2014). In addition to mobile devices with wireless Internet, mHealth requires software (Becker et al., 2014), represented by health-related apps (health apps). Apps are software applications on smartphones or tablets (van Velsen, Beaujean, & van Gemert-Pijnen, 2013). In the health context, apps can encourage preventive behaviors such as weight loss (Turner-McGrievy et al., 2013) and exercise (Glynn et al., 2014) and promote adherence to prescribed treatment (Becker et al., 2014).

Given the positive outcomes, many studies have explored correlates of health app use and they can be classified into three types. First, studies have documented the sociodemographic characteristics of those who have health apps, including age, gender, race/ethnicity, education, and income (Fox & Duggan, 2012; Krebs & Duncan, 2015). Second, scholars have identified theory-based predictors of health apps. For example, some studies, using the technological acceptance model (TAM; Davis, 1989), have explored the effect of perceived usefulness (PU) and perceived ease of use (PEOU) on intention to use health apps or actual use of health apps (Cho, Lee, Kim, & Park, 2015; Mackert, Mabry-Flynn, Champlin, Donovan, & Pounders, 2016). Lee and Cho (2017), based on uses and gratifications (Katz, Blumler, & Gurevitch, 1973), employed media-oriented and user-oriented reasons to predict continued use of health apps. Third, several studies have investigated the relationship between health apps and health-related variables, for example, health consciousness, health app efficacy (Cho, Park, & Lee, 2014), and eHealth/health literacy (Cho et al., 2014; Mackert et al., 2016).

This study aimed to extend the previous findings. Most previous studies have not used a nationally representative sample. Also, they focused on either sociodemographic or health-related factors, but not both. To address these limitations, this study included various factors related to health apps that were chosen in light of TAM (Davis, 1989). TAM and its updated version (TAM2; Venkatesh & Davis, 2000) posit that external variables (e.g., education, experience) influence beliefs (PE and PEOU) about new technology, and that beliefs affect attitudes toward the technology, which finally lead to intention and actual technology use. Usually PE ad PEOU are believed to fully mediate the relationship between external variables and attitudes/intention, but studies have found that external factors can also directly influence technology use (Burton-Jones & Hubona, 2006; Pavlou, 2003). As a secondary analysis, this study did not directly test TAM, but sought to identify external factors that can potentially influence TAM variables (PE or PEOU) or technology use. Although this study concerns having apps rather than using apps, to use apps, individuals first need to download and install them. In summary, the goal of the current study was—inspired by TAM—to construct a comprehensive profile of individuals with health apps based on nationally representative data from the Health Information National Trends Survey 4 (HINTS 4, Cycle 4).

Sociodemographic Factors

Studies have shown that sociodemographic variables can influence PEOU or PU regarding health apps based on TAM (Mackert et al., 2016). Among various factors, younger age, higher income, and higher education are directly associated with having health apps (Fox & Duggan, 2012; Krebs & Duncan, 2015). However, the role of race/ethnicity or gender has been inconsistent. Krebs and Duncan (2015) reported that Hispanics are more likely to download apps compared with Whites. Mackert et al. (2016) found that Hispanics and Blacks have a higher level of PEOU and PU regarding health apps. Fox and Duggan (2012) did not find the effect of race/ethnicity but reported that females were more likely to have health apps. Krebs and Duncan (2015) found no gender effect. Moreover, marital status and employment status are all associated with health or health information seeking (Kelly et al., 2010; Ross & Mirowsky, 1995; Umberson, 1992), but their relationships with health apps have not been studied. Therefore, the current study tested the association between aforementioned sociodemographic factors and having health apps.

Technological Factors

With health apps, users can monitor their health status and receive tailored feedback. For such activities, convenient access and use is important (Breton, Fuemmeler, & Abroms, 2011). Then, those with better devices and higher quality of access may think that using health apps is easy (i.e., influencing PEOU). This study examined two mobile technologies: mobile devices and mobile Internet access. Although both tablets and smartphones are personal media and function similarly, tablets are often shared with family members for games or education (Müller et al., 2012). A focus group study (Dennison, Morrison, Conway, & Yardley, 2013) showed that young people are concerned with the privacy of information that health apps collect. Smartphones may be easier to carry and provide a more convenient environment for the private use of health apps.

In addition, apps need either Wi-Fi or a cellular network. Wi-Fi refers to a local wireless network that connects electronic devices to the Internet (Wi-Fi, 2016). Mobile carriers provide cellular networks, such as 4G. Although many health apps can operate without an Internet connection (Martínez-Pérez, de la Torre-Díez, & López-Coronado, 2013), searching and installing apps requires a mobile connection. To use some app functions, such as interacting with other users (e.g., receiving feedback from others, finding friends with similar interests), mobile devices require a wireless connection. Although many people use both Wi-Fi and a cellular network, some (approximately 7% in the United States) do not have a broadband connection at home and their Internet access is limited within a cellular data plan (Smith, 2015). Thus, this study tested the difference between those who use such mobile technology and those who do not.

Health-Related Factors

Based on TAM, intrinsic involvement, defined as personal relevance of new technology, is positively associated with attitude toward technology as well as PU (Jackson, Chow, & Leitch, 1997). Thus, individuals with health problems will adopt health apps if apps are relevant for their problems. Krebs and Duncan (2015) reported that obese people are more likely to have apps. Thus, a higher body mass index may be associated with having health apps. Next, this study added self-reported health status, a subjective but important predictor of health behavior. People who perceive themselves as healthy are less likely to visit a physician and more likely to think that their mortality risk is low (Miilunpalo, Vuori, Oja, Pasanen, & Urponen, 1997). Thus, lower health status may be related to higher PU, which finally leads to the use of new technology. In addition, this study also considered psychological variable related to health. Self-efficacy is a strong predictor of preventive health behaviors, such as weight control, exercise, and healthy diet (Cerin & Leslie, 2008; Dehghan, Akhtar-Danesh, & Merchant, 2011) for which apps are used. Studies based on TAM have reported that self-efficacy not only influences PU and PEOU but also directly affects intention to use technology (Park, 2009). Similarly, those with high health efficacy (self-efficacy in the health context) who believe they can control their health may feel that new technology is personally relevant for them to use.

Motivational Factors

Another possible source of intrinsic involvement is motivation to use health apps. In this study, motivation refers to reasons that promote a willingness to have health apps. Those who have reason to use health apps have a greater level of intrinsic involvement (personal relevance) regarding health apps. Therefore, motivation was operationalized as the degree to which individuals intend to get involved in popular healthy activities that can be promoted by using health apps. Each app has its own specific purpose, but generally, health apps for exercise, diet, and weight control are most popular in the United States (Fox & Duggan, 2012; Krebs & Duncan, 2015). These topics can be seen as the main reasons or motivations to use health apps.

There exist individual differences in such intentions. Some people want to change their current state about the healthy activities (change group), but others want to maintain the state (maintain group). Also, there may be people who pay no attention to such activities (no attention group). Health apps are usually viewed as tools for behavior change (Dennison et al., 2013). Those who want to lose weight or increase the amount of exercise may need health apps to make a change in their lifestyle (i.e., change group). Such people may have a higher level of intrinsic involvement or PU regarding health apps. The no attention group may not need health apps because no action is required for them. The question becomes whether the maintain group needs health apps. Possibly, the maintain group would not be interested in health apps because they can simply continue their current behaviors. However, to maintain a good health state, individuals also need to monitor and check their current health, which is easier when using apps. Therefore, this study investigated whether any difference exists between the three groups.

Method

Participants and Procedures

This study used the data from HINTS 4 (Cycle 4), collected from a nationally representative sample of U.S. adults aged 18 years or older. Cycle 4 was conducted from August through November 2014, and 3,677 people completed the mailed questionnaire with an overall response rate of 34.44% based on the American Association for Public Opinion Research response rate 2 formula (National Cancer Institute, 2014). Among 3,677 people, this study included only those who identified themselves as Internet users, which left 2,802 cases. It makes little sense that a person never uses the Internet but has health apps. HINTS uses two types of weights, a full-sample weight and a set of 50 replicate weights. All values reported in this study are weighted. For descriptive statistics, see Table 1.

Table 1.

Descriptive Statistics (N = 2,802).

Predictors (Nonmissing n)	M (SD) or %
Sociodemographic factors
Age, years (2,698)	44.07 (16.10)
18-34	35.2
35-49	28.2
50-64	24.94
≥65	11.67
Gender, female (2,766)	50.81
Race/ethnicity (2,601)
White	69.12
Hispanic	13.11
Black	10.85
Asian	4.56
Others	2.36
Employed (2,683)	65.35
Married (2,708)	55.43
Education (2,723)
Less than high school	6.50
High school	15.91
Some college	32.08
College degree	28.49
Postgraduate	17.02
Income, $ (2,770)
0-14,999	10.88
15,000-34,999	16.01
35,000-74,999	31.89
≥75,000	41.23
Technological factors
Wi-Fi (2,738)	78.90
Cellular network (2,731)	59.72
Smartphone (2,764)	76.64
Tablet (2,764)	57.62
Health-related factors
Health status^a (2,730)	3.56 (0.88)
Health efficacy^b (2,739)	3.89 (0.82)
Body mass index, kg/m² (2,695)	27.99 (6.42)
Motivational factors
Fruit intake (2,730)
Change (increase)	34.97
Maintain	24.02
No attention	41.01
Vegetable intake (2,745)
Change (increase)	45.28
Maintain	22.35
No attention	32.38
Soda intake (2,593)
Change (decrease)	64.02
Maintain	12.84
No attention	23.14
Weight control (2,732)
Change (lose or gain)	64.17
Maintain	20.21
No attention	15.62
Exercise amount (2,735)
Change (increase)	57.51
Maintain	23.97
No attention	18.53
Having health-related apps (2,209)	35.48

Note. Only the Internet users were included in the analysis. All values are weighted.

Health status was measured on a 5-point scale (1 = poor to 5 = excellent). ^bHealth efficacy was measure on a 5-point scale (1 = not confident at all to 5 = completely confident).

Measures

Sociodemographic Factors

Participants reported their age, gender, race/ethnicity, employment status, marital status, income, and education. See Table 1 for response options.

Technological Factors

Participants indicated whether they used (1) a smartphone, (2) a tablet, (3) a cellular network, and (4) Wi-Fi. These were four separate items and participants could choose “yes” for all.

Health-Related Factors

Body mass index, calculated from reported height and weight, and general health status (1 = poor to 5 = excellent) were adopted. Health efficacy was measured by asking how confident participants were in their ability to take good care of their own health (1 = not confident at all to 5 = completely confident).

Motivational Factors

HINTS provides five items about healthy intentions. For example, participants were asked whether they had intentionally tried to (1) increase, (2) maintain, or (3) pay no attention to the amount of exercise. They had to choose one of three answers. The same questions were asked regarding intake of fruit/vegetables/soda and weight control, respectively. For weight, four options were given (lose, gain, maintain, and no attention). Two aspects should be noted regarding these items. First, the items do not actually measure behavior although they ask about a specific behavior. For example, Jayanti and Burns (1998) measured preventive behavior, by asking participants how often they engaged in behaviors like exercise or diet. In contrast, HINTS items ask whether participants, at any time in the previous year, intentionally tried to increase or decrease a certain behavior without assessing frequency. Thus, HINTS items are about intention rather than actual behavior. Second, the items represent intentions to engage in those behaviors, but with regard to health apps, they are the motivation, the reason to have apps. Intention to use health apps should be directly measured by the deliberateness of using health apps (e.g., I plan to use health apps). Therefore, this study used the five items to assess the reasons for having health apps and divided answers for each item into three groups: change, maintain, no attention. The change group consisted of people who wanted to make a positive change. For fruit intake, vegetable intake, and exercise, the increase group is the change group. For soda intake, the decrease group is the change group. For weight, both lose and gain groups are the change group. All variables were dummy coded.

Having Health-Related Apps (Dependent Variable)

Participants reported whether they had health apps on their mobile devices (1 = yes). The “don’t know” option was treated as “no.” Importantly, 458 people who did not have a mobile device responded to the question (coded as “inapplicable” in the original data). They were treated as missing cases, because the question clearly asked about the existence of health app on mobile devices. This resulted in the loss of many cases, but the inclusion of such cases would have made the interpretation of the results difficult because it means predicting the adoption of apps on the mobile devices in the absence of mobile devices.

Statistical Analyses

Poisson regression with robust variance was conducted with STATA 14. Public health research often uses logistic regression which produces odds ratio (OR), but OR tends to overestimate the prevalence ratio (see Barros & Hirakata, 2003; Viera, 2008). All analyses incorporated the two types of weights and the jackknife variance estimation method (National Cancer Institute, 2014). Two models were constructed; sociodemographic, health-related, and technological factors were the same in both models. For motivational factors, Model 1 used the no attention group as a reference group and Model 2 used the maintain group as a reference group to see the difference. All predictors were entered into the models simultaneously.

Results

For both models, the results for sociodemographic, health-related, and technological factors are the same because they had the same reference group across the models. Only the results for motivational factors differed across models. Table 2 shows the results for Model 1, and Table 3 shows only the results for motivational factors in Model 2.

Table 2.

The Association Between Various Factors and Having Health-Related Apps When no Attention Group Is Used as a Reference Group for Motivational Factors.

Factor	Bivariate correlation with the DV	Model 1 (using no attention group as a reference)
Factor	Bivariate correlation with the DV	IRR [95% CI]	Coefficients (p value)
Sociodemographic
Age, years
18-34	0.11**	1.71 [1.17, 2.50]**	0.54 (.006)
35-49	0.03	1.51 [1.03, 2.20]*	0.41 (.034)
50-64	−0.09**	1.28 [0.86, 1.91]	0.25 (.216)
≥65	−0.10***	Ref.	Ref.
Gender (female)	0.02	0.96 [0.76, 1.20]	−0.04 (.706)
Race/ethnicity
White	−0.04	Ref.	Ref.
Hispanic	−0.07**	0.69 [0.49, 0.98]*	−0.36 (.036)
Black	0.04	1.25 [0.97, 1.63]	0.23 (.086)
Asian	0.11**	1.39 [0.87, 2.20]	0.33 (.160)
Others	0.04	1.24 [0.78, 1.96]	0.22 (.350)
Employed	0.04	0.97 [0.77, 1.23]	−0.03 (.794)
Married	0.00	0.97 [0.76, 1.23]	−0.03 (.783)
Education
Less than high school	−0.13***	0.26 [0.09, 0.75]*	−1.33 (.014)
High school	−0.11***	0.69 [0.52, 0.93]*	−0.37 (.016)
Some college	−0.02	0.89 [0.70, 1.14]	−0.11 (.354)
College degree	0.09*	Ref.	Ref.
Postgraduate	0.10**	1.10 [0.88, 1.37]	0.09 (.405)
Income, $
0-14,999	−0.01	1.32 [0.81, 2.13]	0.27 (.256)
15,000-34,999	−0.04	1.26 [0.87, 1.83]	0.23 (.223)
35,000-74,999	−0.11**	Ref.	Ref.
≥75,000	0.14***	1.37 [1.09, 1.73]**	0.32 (.009)
Technological
Smartphone	0.12**	1.64 [1.001, 2.70]*	0.50 (.049)
Tablet	0.07	1.11 [0.92, 1.37]	0.11 (.262)
Wi-Fi	0.15***	1.74 [1.12, 2.70]*	0.55 (.014)
Cellular	0.13***	1.09 [0.87, 1.38]	0.09 (.435)
Health-related
Health status	0.01	0.88 [0.77, 1.01]	−0.13 (.068)
Health efficacy	0.03	1.13 [0.96, 1.32]	0.12 (.140)
Body mass index	0.06	1.02 [1.00, 1.03]	0.02 (.057)
Motivational
Fruits
Increase	0.08*	1.02 [0.78, 1.33]	0.02 (.895)
Maintain	0.01	1.12 [0.81, 1.54]	0.11 (.495)
No attention	−0.09**	Ref.	Ref.
Vegetable
Increase	0.13***	1.03 [0.73, 1.47]	0.03 (.859)
Maintain	−0.04	1.03 [0.68, 1.58]	0.03 (.882)
No attention	−0.11**	Ref.	Ref.
Soda
Decrease	0.06	0.87 [0.63,1.20]	−0.14 (.377)
Maintain	−0.01	0.89 [0.58, 1.37]	−0.11 (.604)
No attention	−0.06	Ref.	Ref.
Weight
Change	0.18***	1.89 [1.23, 2.91]**	0.64 (.004)
Maintain	−0.10**	1.49 [0.86, 2.57]	0.40 (.150)
No attention	−0.13**	Ref.	Ref.
Exercise
Increase	0.20***	1.12 [0.77, 1.61]	0.11 (.548)
Maintain	−0.15***	.71 [0.43, 1.16]	−0.35 (.166)
No attention	−0.10**	Ref.	Ref.
Total N		1,684	1,684

p < .05. **p < .01. ***p < .001.

Table 3.

The Association Between Motivational Factors and Having Health-Related Apps When Using Maintain Group as a Reference Group.

Factor	Model 2 (using maintain group as a ref.)
Factor	IRR [95% CI]	Coefficients (p value)
Motivational
Fruits
Increase	0.91 [0.68, 1.22]	−0.09 (.525)
Maintain	Ref.	Ref.
No attention	0.90 [0.65, 1.24]	−0.11 (.495)
Vegetable
Increase	1.00 [0.78, 1.28]	0.00 (.998)
Maintain	Ref.	Ref.
No attention	0.97 [0.63, 1.48]	−0.03 (.882)
Soda
Decrease	0.97 [0.76, 1.23]	−0.03 (.796)
Maintain	Ref.	Ref.
No attention	1.12 [0.73, 1.72]	−0.11 (.604)
Weight
Change	1.27 [0.91, 1.78]	0.24 (.158)
Maintain	Ref.	Ref.
No attention	0.67 [0.39, 1.16]	−0.40 (.150)
Exercise
Increase	1.58 [1.16, 2.13]**	0.46 (.004)
Maintain	Ref.	Ref.
No attention	1.41 [0.86, 2.31]	0.35 (.166)
Total N	1,684	1,684

Note. IRR = incident rate ratio (same as the risk ratio, meaning how many times more likely the outcome occurs among people with certain characteristics); CI = confidence interval; Ref. = reference group. The results for other factors (sociodemographic, technological, and health-related factors) are the same as Model 1. Bivariate correlations are reported in Table 2. Significant results are in boldface.

p < .05. **p < .01. ***p < .001.

In both models, compared with those who were aged 65 years or older, people aged 18 to 34 and 35 to 49 years were more likely to have apps, but those aged 50 to 64 years were not different from the reference group. Hispanics (vs. non-Hispanic Whites) were less likely to have health apps. Those who completed high school or had less than a high school education showed a lower chance of having health apps than the college educated. As for income, those with an annual income of more than $75,000 were more likely to have apps than those whose income was $35,000 to $74,999. Gender, employment status, and marital status were not associated with the dependent variable. As for technological factors, smartphone users were more likely to have apps than nonusers. However, tablet users did not differ from nonusers. Wi-Fi users were 1.74 times more likely to have apps than nonusers, but the use of a cellular network did not make any difference. Health-related factors were not associated with having apps.

For motivational factors, Model 1 used the no attention group as a reference group. Those who want to change their weight were 1.89 times more likely to have apps than the no attention group, but the maintain group did not differ from the no attention group. For all other topics, no association was detected. Model 2 used the maintain group as a reference. Those who tried to increase the amount of exercise were 1.57 times more likely to have apps than the maintain group.

Discussion

The results indicated that a younger age, being non-Hispanic, having higher socioeconomic status, and using a smartphone and Wi-Fi were associated with having health apps. Those who want to change their weight and amount of exercise were more likely to have apps. There was no difference between the maintain group and the no attention group. The implications of finding are discussed below.

First, this study confirmed socio-demographic factors related to health apps. Younger age and higher socioeconomic status may help individuals perceive a new technology as easy to use and more useful as TAM suggests. However, gender was not associated with having apps in the current study. Females are generally more interested in health information (Kelly et al., 2010), but the adoption of new technology is related to the male gender as well as socioeconomic status (DiMaggio & Hargittai, 2001). Health apps are related to both health and technology, and the gender effect was not clear here. Also, Hispanics were less likely to have apps, contrary to Krebs and Duncan (2015). They used online opt-in panels and the ratio of Whites and Hispanics were the same (both 30% of the sample). In this study, Hispanics constituted 13.11% of the sample. According to the U.S. Census Bureau (2015), Hispanics comprise 17.6% of the U.S. population. Future research should reexamine the effect of race/ethnicity.

Second, this study assumed that a better technological environment can increase the PEOU, based on TAM, and the results potentially confirmed this assumption. Smartphone and Wi-Fi users were more likely to have health apps than non-users. In the current data set, 54.45% of the participants used both Wi-Fi and a cellular network. Only 5.28% of participants used a cellular network without Wi-Fi, and 24.64% used Wi-Fi without a cellular network (weighted percentage). In other words, cellular users in most cases also used Wi-Fi, and they can use any wireless network to search for, download, and use health apps. However, many Wi-Fi users had no cellular network, but this did not pose a problem for having apps. The lack of a cellular network was not a problem, but the lack of Wi-Fi was. As Wi-Fi requires a home broadband connection, the result implies the digital divide. Rains (2008) reported that the use of broadband is associated with diverse online health-related activities, which suggests that the quality of access can influence health inequality. Although Wi-Fi adoption will be saturated soon, with the development of another new communication technology, the digital divide may continue to exist (Viswanath, 2006).

Third, this study demonstrated the importance of intrinsic involvement, suggested as an external variable of TAM (Jackson et al., 1997). However, in the context of health apps, only specific and goal-oriented intrinsic involvement (i.e., motivational factors in this study) was associated with having health apps. More general intrinsic involvement related to individuals’ health was not associated with health apps. Even though individuals feel the personal relevance of health apps due to their general health problems (health status, body mass index) or their general attitudes toward health (health efficacy), they actually have health apps only when they are willing to engage in activities supported by health apps.

As for the degree of specific intrinsic involvement, this study showed that the change group differs from the maintain group and no attention group, but the maintain group does not differ from the no attention group. The maintain group (about weight and exercise) was even negatively associated with health apps in bivariate correlations (see Table 2). This study showed that only active motivation to make a visible change was related to engagement in mHealth. The intention to maintain the current state was not enough to perceive the usefulness of health apps, and people in the maintain group may refuse to adopt a new technology to keep consistency in their own lifestyle.

In addition, it is noteworthy that the change group (regarding exercise) did not differ from the no attention group in Model 1, but in Model 2 differed from the maintain group. This result is difficult to understand. However, if all other motivational factors are deleted from the analysis, the exercise increase group was more likely to have apps than the no attention group. Because all motivations are positively correlated, the effect of the change group in Model 1 was not detected beyond the other motivations. Another explanation is that those in the maintain group more strongly refuse to use health apps than members of the no attention group, because they want to maintain their way of exercise. Supporting this assumption, only regarding exercise, the maintain group was more strongly and negatively correlated with having health apps than the no attention group. In all other contexts, the no attention group was more strongly and negatively correlated with the dependent variable than the maintain group.

Fourth, among five health-related topics, only motivation regarding weight control and exercise were associated with having health apps. Although many apps for healthy diet are available, healthy diet intention was not associated with health apps. Three healthy diet variables were not associated with the dependent variable even when other motivations were deleted from the analysis. The results have practical implications. The desire to control weight and increase exercise seems to be a greater motivation for having health apps than the desire to have a healthy diet. Compared with other topics, weight control and exercise may be better promoted using mHealth. Otherwise, it may be that the real purpose of diet app use is weight control or exercise rather than change in diet.

Finally, the current study has several limitations. As a secondary analysis, the current study could not use TAM variables such as PE and PEOU, but only used variables available in HINTS. The variables were assumed to be external variables that can directly or indirectly influence technology use. Future studies should reexamine the assumption in this study with TAM variables. Another limitation is, as with any cross-sectional survey study, the suggested relationships do not reveal causal effects. It is unlikely that health apps influence levels of education or ownership of mobile technology. The opposite makes more sense. However, it is possible that having health apps increases health motivation. Future research should more precisely demonstrate the relationship between health motivation and the use of mHealth technology with longitudinal data.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

References

Barros

A. J.

Hirakata

V. N.

(2003). Alternatives for logistic regression in cross-sectional studies: An empirical comparison of models that directly estimate the prevalence ratio. BMC Medical Research Methodology, 3(21). Retrieved from bmcmedresmethodol.biomedcentral.com/articles/10.1186/1471-2288-3-21

Becker

Miron-Shatz

Schumacher

Krocza

Diamantidis

Albrecht

U. V.

(2014). mHealth 2.0: Experiences, possibilities, and perspectives. JMIR mHealth and uHealth, 2(2), e24. doi:10.2196/mhealth.3328

Breton

Fuemmeler

Abroms

(2011). Weight loss—there is an app for that! But does it adhere to evidence-informed practices? Translational Behavioral Medicine, 1, 523-529. doi:10.1007/s13142-011-0076-5

Burton-Jones

Hubona

G. S.

(2006). The mediation of external variables in the technology acceptance model. Information & Management, 43, 706-717. doi:10.1016/j.im.2006.03.007

Cerin

Leslie

(2008). How socio-economic status contributes to participation in leisure-time physical activity. Social Science & Medicine, 66, 2596-2609. doi:10.1016/j.socscimed.2008.02.012

Cho

Lee

H. E.

Kim

S. J.

Park

(2015). Effects of body image on college students’ attitudes toward diet/fitness apps on smartphones. Cyberpsychology, Behavior, and Social Networking, 18, 41-45.

Cho

Park

Lee

H. E.

(2014). Cognitive factors of using health apps: Systematic analysis of relationships among health consciousness, health information orientation, eHealth literacy, and health app use efficacy. Journal of Medical Internet Research, 16(5), e125. doi:10.2196/jmir.3283

Davis

F. D.

(1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13, 319-340. doi:10.2307/249008

Dehghan

Akhtar-Danesh

Merchant

A. T.

(2011). Factors associated with fruit and vegetable consumption among adults. Journal of Human Nutrition and Dietetics, 24, 128-134. doi:10.1111/j.1365-277X.2010.01142.x

10.

Dennison

Morrison

Conway

Yardley

(2013). Opportunities and challenges for smartphone applications in supporting health behavior change: Qualitative study. Journal of Medical Internet Research, 15, e86. doi:10.2196/jmir.2583

11.

DiMaggio

Hargittai

(2001). From the ‘digital divide’ to ‘digital inequality’: Studying Internet use as penetration increases (Working Paper Series 15). Princeton, NJ: Center for Arts and Cultural Policy Studies, Woodrow Wilson School, Princeton University.

12.

Fox

Duggan

(2012). Mobile health 2012. Washington, DC: Pew Internet & American Life Project. Retrieved from http://www.pewinternet.org/2012/11/08/mobile-health-2012/

13.

Glynn

L. G.

Hayes

P. S.

Casey

Glynn

Alvarez-Iglesias

Newell

. . . Murphy

A. W.

(2014). Effectiveness of a smartphone application to promote physical activity in primary care: The SMART MOVE randomised controlled trial. British Journal of General Practice, 64, e384-e391. doi:10.3399/bjgp14X680461

14.

Jackson

C. M.

Chow

Leitch

R. A.

(1997). Toward an understanding of the behavioral intention to use an information system. Decision Sciences, 28, 357-389. doi:10.1111/j.1540-5915.1997.tb01315.x

15.

Jayanti

Burns

(1998). The antecedents of preventive health care behavior: An empirical study. Journal of the Academy of Marketing Science, 26, 6-15.

16.

Katz

Blumler

J. G.

Gurevitch

(1973). Uses and gratifications research. Public Opinion Quarterly, 37, 509-523. doi:10.1086/268109

17.

Kelly

Hornik

Romantan

Sanford Schwartz

Armstrong

DeMichele

. . . Wong

(2010). Cancer information scanning and seeking in the general population. Journal of Health Communication, 15, 734-753. doi:10.1177/0092070398261002

18.

Krebs

Duncan

D. T.

(2015). Health app use among US mobile phone owners: A national survey. JMIR mHealth and uHealth, 3(4). doi:10.2196/mhealth.4924

19.

Lee

H. E.

Cho

(2017). What motivates users to continue using diet and fitness apps? Application of the uses and gratifications approach. Health Communication, 32, 1445-1453. doi:10.1080/10410236.2016.1167998

20.

Mackert

Mabry-Flynn

Champlin

Donovan

E. E.

Pounders

(2016). Health literacy and health information technology adoption: The potential for a new digital divide. Journal of Medical Internet Research, 18, e264. doi:10.2196/jmir.6349

21.

Martínez-Pérez

de La Torre-Díez

López-Coronado

(2013). Mobile health applications for the most prevalent conditions by the World Health Organization: Review and analysis. Journal of Medical Internet Research, 15, e120. doi:10.2196/jmir.2600

22.

Miilunpalo

Vuori

Oja

Pasanen

Urponen

(1997). Self-rated health status as a health measure: The predictive value of self-reported health status on the use of physician services and on mortality in the working-age population. Journal of Clinical Epidemiology, 50(5), 517-528. doi:10.1016/S0895-4356(97)00045-0

23.

Müller

Gove

J. L.

Webb

J. S.

(2012, September). Understanding tablet use: A multi-method exploration. Paper presented at the 14th International Conference on Human-Computer Interaction with Mobile Devices and Services. San Francisco, CA.

24.

National Cancer Institute. (2014). HINTS 4, Cycle 4, methodology report. Retrieved from http://hints.cancer.gov

25.

Park

S. Y.

(2009). An analysis of the technology acceptance model in understanding university Students’ behavioral intention to use e-learning. Educational Technology & Society, 12, 150-162.

26.

Pavlou

P. A.

(2003). Consumer acceptance of electronic commerce: Integrating trust and risk with the technology acceptance model, International Journal of Electronic Commerce, 7, 101-134. doi:10.1080/10864415.2003.11044275

27.

Rains

(2008). Health at high speed: Broadband Internet access, health communication, and the digital divide. Communication Research, 35, 283-297. doi:10.1177/0093650208315958

28.

Ross

Mirowsky

(1995). Does employment affect health? Journal of Health and Social Behavior, 36, 230-243.

29.

Smith

(2015). U.S. smartphone use in 2015. Retrieved from http://www.pewinternet.org/2015/04/01/us-smartphone-use-in-2015/

30.

Turner-McGrievy

Beets

Moore

Kaczynski

Barr-Anderson

Tate

(2013). Comparison of traditional versus mobile app self-monitoring of physical activity and dietary intake among overweight adults participating in an mHealth weight loss program. Journal of the American Medical Informatics Association, 20, 513-518. doi:10.1136/amiajnl-2012-001510

31.

Umberson

(1992). Gender, marital status and the social control of health behavior. Social Science & Medicine, 34, 907-917. doi:10.1016/0277-9536(92)90259-S

32.

U.S. Census Bureau (2015). Quick facts. Retrieved from http://www.census.gov/quickfacts/table/PST045215/00

33.

van Velsen

Beaujean

van Gemert-Pijnen

(2013). Why mobile health app overload drives us crazy, and how to restore the sanity. BMC Medical Informatics and Decision Making, 13, 23. doi:10.1186/1472-6947-13-23

34.

Venkatesh

Davis

F. D.

(2000). A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management Science, 46, 186-204. doi:10.1287/mnsc.46.2.186.11926

35.

Viera

A. J.

(2008). Odds ratios and risk ratios: What’s the difference and why does it matter? Southern Medical Journal, 101, 730-734. doi:10.1097/SMJ.0b013e31817a7ee4

36.

Viswanath

(2006). Public communications and its role in reducing and eliminating health disparities. In Thomson

G. E.

Mitchell

Williams

M. B.

(Eds.), Examining the health disparities research plan of the National Institutes of Health: Unfinished business (pp. 215-253). Washington, DC: Institute of Medicine.

37.

Wi-Fi. (2016). In Technology dictionary. Techopedia.com. Retrieved from https://www.techopedia.com/definition/10035/wireless-fidelity-wi-fi