Ubiquitous health (u-health) aims to provide innovative healthcare services that are consumer-oriented, customized, and accessible anytime, anywhere by ubiquitous computing [1]. More than 10 years have passed since the concept of u-health was proposed in Korea [1]. Despite very high expectations, u-health still is in its very early stages, in part because its clinical effectiveness has not been clearly demonstrated [12]. It is also struggling in Korea in part because of the complex debates about telemedicine. However, electronic health (e-health), a concept that is similar to u-health, which utilizes information systems by applying them to existing healthcare practices to improve the productivity of the work of the health service provider [1], has been growing in Europe and the United States even though issues regarding its clinical effectiveness have not yet been resolved [34567]. The e-health market has grown rapidly, especially after the emergence of mobile health (m-health) in 2009 [34578910]. m-Health services using both wearable devices and smartphones are also expanding [111213].

m-Health provides innovative healthcare services by using the unique characteristics of the smartphone such as mobility, connectivity, accessibility, and the capacity to gather and transmit data [91011121314], so it is considered a subtype of u-health in Korea [12]. Diverse u-health devices, including wearable devices, can be connected to smartphones, and sometimes the smartphone itself can be used as a u-health device [1516]. m-Health services and trials using smartphone sensors, such as cameras, accelerometers, and pedometers, are frequent now [131718].

Despite this rapid growth, challenges remain which must be addressed if m-health is to result in the hoped-for changes [3812]. According to Dr. Joseph Kvedar, Director of the Center for Connected Health of Partners Healthcare, one in ten U.S. patients has used a health tracker, but the majority stop using it within six weeks, and for m-health applications (apps), 80% are abandoned after only two weeks [19]. Several studies have suggested key features that encourage users to be more likely to continuously use such apps, but most trackers and apps have little impact, and more research is needed [192021]. m-Health must also be connected to personal health records or Electronic Health Records to provide a continuous care model and to improve patient outcomes [8], but such connectivity and interoperability remain major issues. The Apple HealthKit and ResearchKit have the potential to be breakthrough technologies in this area [22]. Another big challenge is determining the clinical effectiveness of m-health. Whether the use of m-health can improve patient outcomes must be determined [2356823]. Like telemedicine cases, not all m-health services can achieve this goal [324]. The common features of m-health that impact its outcomes must be researched [8]. Therefore, welldesigned clinical trials and meta-analyses are required to definitively determine the effectiveness of m-health [368]. Healthcare organizations in Korea have also developed many m-health apps and have provided m-health services [925]. However, few outcomes research studies have been done in this area. The six aims for quality improvement of the Institute of Medicine (IOM) in the United States can be applied to evaluating the benefits of m-health apps [26]. In 2001 the IOM recommended that "health care should be supported by systems that are carefully and consciously designed to produce care that is safe, effective, patient-centered, timely, efficient, and equitable" and "health information technology (HIT) must play a critical role to achieve the six aims [26]." Some studies reviewed HIT applications with respect to the six aims to evaluate their benefits and impacts [2728].

Asan Medical Center in Seoul, Korea, has developed more than 20 m-health apps since 2010. Because no other hospital in Korea has developed as many m-health apps as our hospital has, we evaluated the m-health apps developed by our hospital with a particular focus on the IOM's six aims with the intention of sharing our findings.

The information about the launch date, the date of the final update, and installed data for the apps was collected from the mobile apps server or hospital information system server with cooperation of the Ubiquitous Health Center in the hospital. We collected the information about the apps from the day of distribution of the app until December 31, 2014. The Institutional Review Board approved our study protocol.

Table 1 presents the overall characteristics of the selected 23 apps. Eleven of the 23 apps were designed for use by patients and the general population, 12 targeted healthcare providers as users. Among the apps for patients, one app also had functions for healthcare providers. Figure 2 lists the number of users of each app. As expected, the public apps had more users.

The frequency that the 23 apps involved the six aims is described in Table 2. 'My chart in my hand' and 'My cancer diary' were matched for all the six aims. The specific functions or purposes of the apps correlate with the six aims as follows:

It was clear that many apps for healthcare address a number of the key quality aims. For example, Timeless was predominant due to the nature of smartphones, which offer accessibility and mobility. Desktop-based programs has a lot of limitations in terms of timeless. Patient safety was well implemented in both groups. This implies that m-health can be beneficial in the quality aims. Most patient-targeted apps covered Equity and Patient-centeredness. Most apps for healthcare providers satisfied Efficiency. Interestingly, there was no app for healthcare providers that satisfies all of the six aims, in contrast with the apps for patients, such as 'My cancer diary' and 'My chart in my hand.' In addition, Equity was not matched for any app for healthcare providers. This reminds us to be aware of vulnerable populations when developing an app, even for healthcare providers [29].

There were some limitations of this study. First, the questionnaires used to assess the apps were evaluated by two clinicians. Since one of the clinicians participated in the development of the apps, any matches between the apps and the six aims might be biased. Also, the other clinician had limited ability to evaluate some apps which ended their services. Therefore, further evaluation of apps is necessary, focusing on currently in-service apps. For more objective and precise evaluation of healthcare-related apps in the future, the assessment of the six aims should be performed by experts uninvolved in their development and should encompass patients who are currently using the apps. Second, we found a disparity in the number of users in accordance with the range of distribution and/or the date of launch. In addition, the number of users who installed an app does not necessarily correlate with the level of usage. Since the extent of usage or the activity of apps is difficult to evaluate precisely, more advanced measures should be developed.

In this study, most of the apps designed for patients provide health information or patient medical information and satisfy Patient-centeredness. These apps are also a good tool for gathering health information from patients. Unfortunately, the user data collected in this way have not been actively applied to clinical use until now. However, the patients' generated data in clinical apps, with the merits of Timeliness and Effectiveness, would promote user interaction. In addition, tailored treatments acquired through user interactions with these healthcare apps would contribute to better Patient-centeredness and Patient safety. Therefore, the evaluation of an individual app is essential for efficient management and continual improvement of the app.

In conclusion, although this study was performed using the apps developed by a single tertiary hospital in Korea, the evaluation results demonstrate the enormous potential of mhealth apps as important tools to achieve breakthrough of quality improvement. To achieve this breakthrough, there should be much more study to improve m-health apps.