Cardiovascular diseases are the cause of approximately 20% of deaths worldwide [1]. Numerous studies have confirmed that not only the treatment of obesity, diabetes, hypertension, and hyperlipidemia with drugs, but also pre-treatment management is essential for preventing such diseases [2]. Practicing self-management and maintaining a healthy lifestyle, such as healthy dietary habits and increased physical activity, are critical for managing cardiometabolic risk factors [345]; furthermore, such management can lower the socioeconomic costs incurred by cardiometabolic diseases [1].

One of the most important roles of health-care services is helping health-care consumers manage chronic disease and cardiometabolic risk factors by themselves, which is achieved through encouraging these consumers to practice healthy behaviors. Thus far, such healthcare services have been provided through face-to-face consultations with healthcare providers (e.g., physicians) [6]; however, with limits to the hours of care and educational materials health-care providers can dispense, providing customized education and monitoring is practically difficult [7]. Further, healthcare consumers also find it difficult to practice and maintain the lifestyle-modification behaviors that they have been taught, such as maintaining a healthy diet and physical activity, quitting smoking, drinking in moderation, and managing their weight [8]. Therefore, there is a need for a new method that can continuously provide consumers with information and feedback concerning their health behaviors.

Over the past 10 years, the field of computerized health-behavior interventions has been rapidly advancing and disseminating, primarily as a result of high levels of interest in and demand for individualized medical and health management; this interest and demand has been caused by recent technical advances, aging population, and improved quality of life [9]. In addition to web-based health-behavior interventions, the rapid and wide penetration of cellphones over the past 10 years [10] has led to the introduction of mobile health (mHealth), and its effectiveness in terms of various health-behavior interventions has been investigated in previous studies [1011]. Specifically, mHealth is defined as a “medical and public health practice supported by mobile devices, such as mobile phones, patient-monitoring devices, personal digital assistants (PDAs), and other wireless devices” [12]. Mobile-based interventions are more beneficial than conventional face-to-face consultations, as they can provide individualized health information and management methods more frequently and continuously [13]. Interventions that involve sending patient-directed short-message-service (SMS) through mobile applications (apps) help consumers achieve their behavioral goals by providing them with information about lifestyle changes, such as healthy diets or exercise, and self-management, as well as motivating educational materials or supportive messages [14]. Innovation in mHealth services may help prevent and manage cardiovascular diseases [15]. While many studies have proven that mHealth is effective for reinforcing health behaviors [16] and self-managing obesity [17], diabetes [18], hypertension [19], and cardiovascular diseases [20], there have also been contrary reports suggesting that it is ineffective in this regard [212223]. Despite the fact that smartphone penetration in South Korea (88%) is the highest worldwide (with the US at 75% and Europe at 80%) [24], few studies have investigated the effects cellphone-based apps have on health management; most have only investigated ubiquitous health, such as the effects phone calls and web services [25], hospital environments [926], and obesity management have on health [26]. Considering this, the present clinical trial aimed to evaluate whether an mHealth service app can instigate healthy behavior changes and improvements in metabolic disorders in people with metabolic abnormalities. Consequently, a 24-week study was conducted to determine whether mHealth is more effective for improving healthy behaviors and managing metabolic risk than the standard care received in public health centers.

This study aimed to determine whether the use of mHealth services through a mobile app is more effective than face-to-face consultations for improving health behaviors and health status in Koreans with metabolic risks. Consequently, the mHealth service was indeed found to be more effective than conventional face-to-face consultations in this regard. Performing moderate physical activity, eating at least two types of vegetables at every meal, not adding salt or soy sauce to food, and removing fat from meat before eating were particularly helpful factors of improving metabolic abnormalities.

Several studies have confirmed that mHealth services effectively change behaviors, such as those relating to physical activity and dietary habits [31]. According to Payne and colleague's [11] systematic review of mobile apps used in health-behavior interventions, mobile apps are useful for changing health behaviors and are acceptable tools for consumers. In the present study, an app was used for not only recording food and/or exercise activities, but also investigating changes in knowledge and behaviors.

The assessment of health behaviors using a five-item questionnaire (three items for diet and two items for exercise) and of healthy dietary practice using an eight-item questionnaire showed that for improving behavior changes and dietary habits, the mobile app service was more effective than face-to-face consultations. Whereas the conventional IG only received in-person consultation, the mobile app featured behavior change techniques, such as motivational messages, goal-setting, tailored health education/advice, self-monitoring, performance feedback, gamification, and awards and rewards. In a meta-analysis by Kelly et al. [32], telehealth-delivered dietary interventions targeting whole food and/or dietary patterns were found to improve diet quality, fruit and vegetable intake, and sodium intake when compared to control groups. Further, a recent systematic review also reported that mobile apps are effective for improving diet, physical activity, and sedentary behavior and, consequently, preventing non-communicable diseases (NCDs) [14]. Maintaining good exercise and lifestyle habits is essential for managing metabolic risk [33]. Thus, this app can be considered useful for increasing knowledge about exercise and nutrition and for promoting behavioral changes for managing and preventing metabolic diseases.

Numerous longitudinal studies and RCTs have proven that mHealth services can lower metabolic risks by promoting lifestyle changes, such as improved exercise and diet [14]. Furthermore, previous studies have reported that using other intervention strategies in addition to mobile app interventions leads to a greater and longer effect than when using mobile app interventions alone [14]. Similarly, our study also found that health behaviors and parameters were more improved in the group of people who used the mobile app in conjunction with conventional face-to-face counseling. While the conventional IG only showed significant decreases in the proportion of people with TG abnormalities and the proportion of females with at-risk WC, the mHealth IG showed a significant reduction in all metabolic abnormalities, except FBG.

Several studies have suggested that mHealth is a good alternative to the conventional clinical management of cardiometabolic disease risk factors, mainly as a result of its accessibility and usefulness to healthcare providers and customers alike [3435]. Although participants, study periods, and the information and methods presented on apps have differed across studies, behavioral changes have nevertheless been found to lead to improvements in health outcomes, including body weight, BP, and cholesterol [14]. In a study that consisted of 99 Israeli adults, IG, in which they received, over a period of 14 weeks, real-time performance feedback concerning exercise and diet through a web-based app, diet quality score, and physical activity significantly increased; and body weight and body mass index (BMI) significantly decreased in IG compared with the control group [36]. Further, in Stuckey et al. [37], a small-scale study involving 26 Canadian adults with metabolic abnormalities, it was found that after eight weeks the IG's amount of exercise increased while BMI, BP, and cholesterol levels decreased The members of the IG performed self-monitoring using data from BP monitors, glucometers, and a pedometer featured on an app and, based on this information, received tailored counseling at 4-week intervals [37]. In another study involving young Austrian adults, an IG that received information once a week using four apps (one for each behavior), showed an increase in dietary habits and exercise and a decrease in weight and BMI after 12 weeks, compared to the control group [38]. Another study showed that when information is provided less frequently and over a period of six months, the IG maintains a reduced body weight, but does not present any significant changes in terms of health behaviors [39]. Therefore, it can be speculated that motivation to practice healthy behaviors increases with the frequency with which tailored information is provided, which consequently improves health outcomes.

On the other hand, some studies found mHealth interventions not to be effective [40]. For instance, a study that examined 51 young Australian adults aged 18-35 years over 12 weeks found that using a mobile app did not lead to significant differences in diet, exercise, BP, or weight in comparison to the control group [40]. Similarly, a study on 90 Irish individuals aged 16 years or older reported that the average number of steps increased, but weight and BP remained unchanged [41]. Personalization and adaptation in real time are important factors for engaging various users [42], and it could be speculated that studies that did not find significant effects might not have used individualized information, had low engagement [40], or had a study period that was too short to observe improvements in health outcomes [41].

Generally, app usage is known to markedly decrease within the first number of weeks [4344]; although our study could not analyze app usage, the service was launched for 1,000 people, and 911 people continued to use the app until week 24. This suggests a longer average period of use than that shown in other studies. The higher persistence of app utilization in our study might have resulted not only from allowing the participants to perform self-management through the mobile app throughout the 24 weeks, but also from the automated management featured in the app. Automated management included weekly reports concerning health information, monthly exercise, and nutritional consultations. Other factors that may have encouraged continued use of the app include: the monthly health practice reports, the compensation (incentive) system for promoting persistent use, and tailored analysis and consultations with experts at the public health center. Incentives were provided throughout the study with rewards (mobile gift cards) given monthly for excellent users (top 10%) and souvenirs given for people who completed the 24-week program. In the future, studies should be conducted to investigate whether the altered behaviors and health outcomes observed in this study are maintained over the long-term or are further improved. In addition, although the ideal protocol for mHealth monitoring is yet to be established, developing more sophisticated and diverse mobile app programs that can provide immediate feedback would be more effective for managing cardiometabolic risks for both medical professionals and patients.

After adjusting for factors that showed significant effects, it was found that the factors that significantly improved more than one health risk factor were: the female gender, moderate physical activity, vegetable intake at every meal, low sodium in food, and reducing grease or fats in food, respectively.

Recent findings on app technologies suggest that women and younger age groups use mobile apps more frequently [4546]; in our study, age was not a significant factor, but considering that risk decreased 1.37 times more in women than in men, we can speculate that women used and accepted the app more than the men.

Next, according to a meta-analysis of cohort studies, metabolic syndrome risk decreases with increased physical activity [4748]. This is believed to be because physical activity reduces abdominal obesity, which in turn reduces metabolic risks; this has been substantiated in some previous trials [4950]. Recent public health guidelines [5152] recommend individuals engage in at least 150 minutes of moderate physical activity or 75 minutes of vigorous physical activity per week in order to maintain good health. Despite evaluating in this study a lower amount of physical activity than public health guidelines recommend as sufficient, health risk factors decreased 1.61 times more for members of the mHealth IG than for those in the conventional IG. Nevertheless, to lower additional metabolic syndrome risks, strategies to increase moderate physical activity are needed.

Our findings concerning dietary practice are supported by the current evidence that suggests that daily intake of fruits and vegetables can lower cardiometabolic risks [53]. The rich reserve of potassium in fruits and vegetables lowers BP [54], while fiber helps to reduce risks of obesity and overweight by increasing feelings of fullness [55]; further, fiber also lowers BP [56] and regulates insulin [57]. In our study, eating fruit did not significantly lower risk factors; however, eating two types of vegetables at every meal lowered risks by 1.27 times. After the intervention, there were more people who began to eat an appropriate amount of vegetables than eat an appropriate amount of fruits, which may be attributable to the higher accessibility of vegetables due to their cheaper costs [58]. Research has suggested that a healthy diet that is low in fat and salt/sodium may help to prevent cardiovascular disease [5960] Due to the characteristics of Korean food culture, Koreans tend to consume high amounts of salt [5961], and the consumption of meats, which contain a relatively high amount of saturated fatty acids and are recommended to be consumed only in small quantities, is gradually increasing in the country [6061]. In our study, not adding salt or soy sauce to food during consumption and removing the fats in meats before eating lowered metabolic risks by 1.65 times and 1.77 times, respectively. These 2 lifestyle habits might have made the highest contributions to lowering metabolic risks in this study because they are relatively easier to practice at every meal, requiring no altering of cooking methods or amount of intake.

There were some limitations to our study. First, this was not a randomized controlled trial. Randomization was not possible because this study aimed to investigate the effects of mHealth services in regard to improving metabolic abnormalities, using data from three public health centers as part of a national project. Nevertheless, this feature of our study may be beneficial, as our study protocol can be used to expand the entire national project. Second, this study only presents results for adults who visited public health centers and who completed the intervention, so the findings cannot be generalized to less-motivated adults. Third, we could not analyze whether behaviors and metabolic risks differ in relation to the amount of app usage. Fourth, changes in dietary and exercise habits were assessed using questionnaires, but because respondents tend to lean toward better outcomes on self-reported questionnaires [62], the effects of the intervention in improving health behaviors may have been overestimated. Fifth, the proportions of sex that may have effect on the difference of other demographic factors were different between the two groups, but we did not stratify men and women in order to have enough statistical power.

Despite the above, to the best of our knowledge our study is the first to prove that mHealth services are effective for managing metabolic risks in public health center settings, as opposed to hospital settings, in Korea. In general, people who can monitor their health problems accept health information more readily because they feel more obliged, and are more willing, to alter their health behaviors [63]. This study was performed on individuals without disease but who had health risk factors, and we expect the effects of the intervention to be greater on people with diseases, such as diabetes, hypertension, hyperlipidemia, and cardiovascular diseases.

The findings of this study suggest that the additional use of community-based mHealth services via a mobile app is an effective method of improving health behaviors and lowering metabolic risks in Koreans. In particular, our finding that mHealth intervention improves metabolic risks by promoting changes in health knowledge and behaviors, such as dietary habits and exercise, in local communities is meaningful. Efforts to optimize the smartphone app as well as the conducting of randomized controlled clinical trials with a larger sample are needed in the future.