Using relevant electronic databases (PubMed and Web of Science), a systematic search was performed on the literature. Key concepts such as perception and experience of mHealth were used to search the databases. The search was completed using a Medical Subject Heading keyword combination (eg, “telemedicine” AND “qualitative study”) and other relevant keywords (Textbox 1).

The studies included in this review were related to mHealth or similar concepts (ie, telehealth apps, eHealth, or digital devices). Other terms and keywords used for this purpose were mobile health application(s) (or apps), eHealth app(s), telehealth devices, telehealth systems, and digital devices. Other keywords related to perception included in the search were experience, views, perspective, perception, feasibility, usability, review, utility, acceptability, evaluation, quantified self, and self-assessment. Keywords used to describe these apps were mHealth, mobile health, eHealth, telecare technologies, apps, mobile health, health technology, mobile applications, smartphones, digital health, telemedicine, and mobile apps. We restricted our focus to one main population of users, consisting of patients and potential patients. In order to determine what patients believe and how they perceive mHealth apps, we were particularly interested in original studies using a qualitative approach. Qualitative methods are potentially useful for understanding the individual needs of patients, their experiences, and their perception of mHealth apps [18]. Combinations of keywords including the term “qualitative study” were also used. In addition to the results obtained by searching databases and journals, other references to relevant articles were retrieved by performing a manual search. The studies included had to be in English and had to have been published within the last 5 years (from January 2013 to June 2018). The use of smartphones, especially iPhones and Androids, increased sharply to over 55% in 2013, along with the use of smartphone apps [19]. We therefore decided to focus on articles published within the last 5 years.

Since this research project focused on how mHealth apps can be used to improve patients’ health care, any apps designed for the sole purpose of surveillance, location tracking, consultation, changing health behavior or health styles, or monitoring patient activity were excluded from the study. Since the focus was also restricted to patients, any studies focusing only on caregivers and other members of patients’ social networks (such as spouses and parents) were excluded. Lastly, papers only reviewing an app’s user interface and usability were not included unless they contributed to understanding patients’ perception of mHealth app usage.

In view of the fast progress of technology, papers related solely to the use of short message service were not included because information, reminders, and other data are communicated via mHealth apps themselves. Studies on the wide range of digital and other technological tools that are in development, such as FitBits, eHealth, telemonitoring devices, and telehealth systems, were also excluded in order to focus on mobile phone apps alone.

Since this review examines studies with qualitative designs, all relevant articles were finally double checked to make sure that they were in line with the Consolidated Criteria for Reporting Qualitative Research, after which no further studies were excluded.

The titles and abstracts were scanned to retrieve the keywords and combinations of keywords mentioned above in order to identify relevant articles and exclude those that were not within the scope of this review. The Methods section of each article was reviewed extensively to ensure that each study was based on a qualitative design. Full texts of relevant articles were retrieved and further reviewed to ensure that they matched the objectives of this study, and any incomplete studies and studies in progress were excluded. To list the information presented in each paper, a table was drawn up on an Excel file, in which the following details were recorded: author, country involved in the study, study population and demographics, methods, disease/condition of interest, type and purpose of the app, results, themes, and time scheme.

Data were extracted from the articles included in this review using the meta-ethnographic framework and the corresponding interpretative method, which was designed for developing new interpretations via comparisons rather than aggregate findings [20]. The first step consisted of arranging these studies in chronological order and extracting the themes from the Results and Discussion sections. While continuing to analyze other articles, we continued to keep an eye out for any emerging themes and include them in the ongoing analysis.

The search conducted on the literature yielded 356 articles, 43 of which met the inclusion criteria (Figure 1). Three of these articles were included as the result of a manual search, while the other 40 articles were found by consulting databases. The reference numbers, the authors’ names, the year of publication, the countries in which the studies were conducted, data on participants and their disease/condition, the methods used, the purpose of the app/device, and the most significant findings obtained in the studies selected are all presented in Table 1.

In 31 studies, the participants were all patients, whereas in 12 studies, the participants included both patients and providers. All 43 articles discussed patients’ opinions about including patient-centered mHealth apps or eHealth apps in their mobile phones. The aim of 37 of the 43 articles was to report patients’ assessment of an app, while 5 articles included either patients’ expectations or their perception of an app prior to its use, and one study included both their expectations and assessments.

Most of the articles reviewed included populations inhabiting developed countries, apart from one that focused on a developing country (South Africa). Most of these qualitative studies focused on the United States [21,23-25,27,31,34,36,41,43,45,47,48,54] and the United Kingdom [30,32,35,42,53,61]. Other countries included were the Netherlands [22,33,44], Denmark [26,55,58], Sweden [38,40,49], Australia [28,29,46], Canada [50,56], South Africa [62], Singapore [59], China [63], Germany [37], Ireland [57], Belgium [60], Spain [39], and Switzerland [52]. One paper included populations originating from seven European countries [51]. Although all the articles selected had been published during the last 5 years, about 70% of them were published more recently, during the last 2 years (2018: n=17; 2017: n=12). In addition, 8 articles were published in 2016, 4 articles were published in 2015, and 1 article was published in 2014.

Most of the apps under consideration were tailored to deal with chronic diseases: cancer [22,30,33,38,40,56,63]; HIV [23,24,31,45,48]; diabetes [28,50,55]; hypertension and cardiovascular diseases [43,54,57]; chronic kidney disease [27,42]; cystic fibrosis [21,51]; chronic pain [26,58]; juvenile arthritis [35,61]; brain and spinal cord anomalies [34]; multiple sclerosis [52]; chronic illnesses, in general [25,62]; and mental health disorders [32,36,39,41,46,47,53,60]. Other apps were specific to pregnancy/obstetric care [29,37,44,59], and one was for postoperative care [49]. Since almost all the studies’ objectives focused on chronic diseases/conditions, the views and expectations of patient users were related to apps designed for the purpose of providing support, including giving access to information, promoting interventions, promoting compliance with treatment, assisting with the management of treatment or disease, and facilitating discussions.

This article discusses the main strengths and weaknesses of using mHealth apps from the patients’ point of view. The strengths mentioned can be categorized into two main aspects: patients’ engagement and their empowerment. The four main weaknesses to which the subjects objected were the apps’ lack of trustworthiness, appropriateness, personalization, and accessibility (Textbox 2).

One of the most critical factors stressed upon by many participants in these studies was the need to personalize the content of mHealth apps to a greater extent. Patients believed that since mHealth apps were created for their use, they should be able to personalize the apps to meet their own needs. From the patients’ point of view, using personalized apps could be as simple as changing the text size. It can be more complex to, for instance, make a drug regimen notification more flexible, give users the choice of adding personal information and receiving advice, or program a self-management app acting like a virtual personal coach [22,32,44,60]. Users wanted to be able to adjust the timing of their prompts, the number of symptoms reported, the frequency of health tips, and so forth [47]. Users also suggested that personalization of the app could also include the language used on the app: Some of them felt that the language used was rather patronizing, and they wanted to be able to change the type of language used to suit their tastes [47].

Greater personalization of apps could also make users feel more interactive with them. One user suggested, for example, that in order to meet health-related goals, they could have avatars that change, mature, and develop as the users learn more, achieve more goals, and become more independent by improving their self-management skills [34]. On the whole, patients value apps that are customized and tailored to meet their needs [25]. With more complex diseases such as cystic fibrosis, personalization is an essential requirement for users. With some diseases and conditions, management or treatment of the disease can vary from one patient to another, and patients may therefore have different routines to follow [51]. It would be best if these apps included “diverse functions” that enable patients to personalize and tailor them to meet their needs

In addition to the trustworthiness of the information and the appropriateness of the app, patients were also worried about a few issues related to their access to apps, such as the connectivity and cost. Some patients found that they had problems with their connection to these apps and other user interface issues. The frustration of not being able to connect with the apps was found to be particularly prevalent among older adults and the elderly, who frequently have poor eyesight and a lower level of digital literacy than other age groups [25,43,57,62]. The elderly patients included in the studies reviewed consisted of two groups: those wanting to acquire digital skills in order to be able to engage with these apps and those having no desire to improve their digital skills for this purpose [57]. Some expressed the feeling that placing greater reliance on technology meant that they were “admitting that the memory isn’t as good as it used to be,” which is distressing and prevented them from wanting to engage with their app [57]. Those with high-to-moderate levels of familiarity with computer technology were more likely to want to use apps, while those who were less familiar with computers and mobile phones were less likely to do so. People’s preferences for the use of apps varied, depending on their familiarity with this form of technology and their wish to become digitally competent. In addition to their digital incompetence, some patients have physical barriers that have to be to overcome to be able to use these apps. Elderly patients have described poor eyesight as one of the main barriers to using apps because they are not able to see their phone screens clearly [25,62].

Other accessibility challenges cited by patients are the extensive battery and memory requirements of smartphones [44]. For example, the mHealth app designed for pregnancy care used a lot of battery and memory space, which set patients problems. Patients had to ask themselves “what kind of apps do you delete?…[apps that use] lots of memory, lots of power, apps that are very active, [but] in that case your battery goes down…” [44]. Instead of being able to use their app to manage their care seamlessly, patients were bothered with having to think about whether an app was worth keeping.

Another issue patients faced was the cost of these apps [25], especially in developing countries where some patients cannot afford them [62]. In addition, some patients were worried about purchasing data to use the app when they were not living in an area with available Wi-Fi [62]. Not being able to use apps consistently because of data issues deterred some patients from engaging with them for long. An additional cost-related barrier was the fact that some apps charge patients a fee for obtaining full access to the app and for being advertisement-free [41,58]. Some patients would have to pay for using the full app, since some free versions provided only a few functions. The financial issues arising therefore prevented some potential app users from fully optimizing the use of an app and engaging with it.

To our knowledge, this is the first review of qualitative studies available in the literature that provides an overall picture of patients’ perceptions, beliefs, and experiences of mHealth apps. This study completes the model identifying the factors involved in the effective use of mHealth Apps for self-care purposes developed by Azhar and Dhillon [17]. In this model, the behavioral intention to use mHealth for self-care purposes is influenced by perceived usefulness, perceived ease of use, performance expectancy, social influence, self-efficacy, potential lack of privacy, and hedonic motives.

Our interpretation of qualitative findings shows how mHealth can strengthen patients’ engagement and sheds light on the dynamics of patients’ engagement and ways to make patients feel more empowered by using mHealth apps. As defined by Carman, the concept of patient engagement develops “as patients...and health professionals [work] in active partnership at various levels across the health care system - direct care, organizational design and governance, and policy making - to improve health and health care” [64]. Some authors have portrayed eHealth (internet and related technologies) as an important means of achieving patient engagement [65], especially in the case of isolated people and those who are hard to reach or have difficulty in remaining engaged in care [16]. Even when patients are not physically in a health care setting, health care advice and guidance are within easy reach at all times. From the patients’ point of view, mHealth could facilitate communication with health care providers and other patients, encourage them to be more participative during clinical encounters, and promote the use of coping techniques to manage their illness. However, some studies based on other methods have shown that some patients are reluctant to use mHealth apps, which is paradoxically the case with adolescent patients because they want to separate their feelings of being a patient from those of being a teenager and make their illnesses and diseases invisible [66]. In addition, they find apps, especially those with push notifications, annoying, intrusive, and time consuming [66]. Patients also stated that although these technical health innovations have supported them in many different respects, they still view their providers as the first point of contact to be consulted for discussing the options available. Health apps serve only as back-up consultations when they are really needed; apps are simply available to support physician-patient relationships and do not replace a physician in any way. mHealth is therefore not a substitute for care but simply a complementary tool [67].

Most of the studies included in this review suggest that patients feel empowered by the information provided by mHealth. WHO defines empowerment for health as: “…a process through which people gain greater control over decisions and actions affecting their life” [68]. According to the World Health Organization, adequate and understandable information is a necessary prerequisite for patient empowerment. Health care policymakers, politicians, and the media share the widespread idea that digital health technologies empower patients [9]. However, when faced with too much overwhelming information, users tend to feel more confused and possibly disempowered, which decreases the effectiveness of discussions with their physicians [67]. Users have also mentioned that the possible lack of validity of the information provided by apps makes it difficult to trust this information, which may make empowerment ambivalent. However, as mentioned above, if apps presented valid information more clearly and concisely, they could possibly incite users to engage more strongly with them as well as with users’ providers, caretakers, and support networks. Some authors have also reported that patients described themselves as reluctant, resistant, and anxious when using digital devices because they feel “disempowered” owing to the surveillance performed by some digital devices, which restricts their autonomy and reminds them that they are ill [69].

The results presented here suggest that despite the many advantages of mHealth apps, barriers to their successful adoption persist. Patients are still reluctant to rely solely on these tools for reasons related to privacy and security and the validity of the information provided. Other barriers to the optimal usage of apps are a lack of accessibility (the cost and absence of access to Wi-Fi) and issues concerning the technical and scientific validity of these tools. Many of the challenges could be met if there was more support on the part of health providers. In addition, standards could be developed and implemented to ensure that these apps provide patients with accurate evidence-based information. These standards could also address the security and privacy issues that many patients are concerned about as well as the compatibility of mobile apps with the technology with which existing health care systems are equipped. There is also a need for inexpensive quality apps and updates (possibly financed by health insurance funds or other agencies) that patients can easily afford.

The levels of engagement and empowerment resulting from the use of mobile phone apps and tools have been found to depend on the users. For example, one study showed that older adults were faced with barriers to adopting these tools because they were not as familiar with smartphones and tablets as younger people and had difficulty in using these technologies [70], whereas those who were digitally literate preferred to receive health information via tablets and electronic devices [70]. Patients’ requirements should be taken into account by those designing mobile health apps in order to alleviate some of the burden [71]. In addition, preference for the use of mobile devices can differ in some contexts. Female participants in Ghana spoke, for example, about how they did not comply voluntarily with health messages “because they didn’t see the point or because it went against their own experience or local knowledge” [72]. Users often expressed the feeling that automatic generic health messages seemed “depersonalized,” which may have resulted in “the opposite effects from those expected by the promoters of mHealth: Reduce caregiver-patient interactions and loosen the link with the health system” [72]. When developers create apps and tools of this kind for users, they should keep their users in mind and remember how their personal identity may influence users’ integration and engagement with the apps in order to facilitate adoption of the apps and the perpetuation of their use.

Although most authors focusing on developing countries discussed how mHealth apps help community health workers, few of them discussed the perspectives of patients using these apps. As the infrastructures with which the large cities in developing countries are improving, their inhabitants are gaining greater access to mobile data and Wi-Fi [73]. However, although the rates of penetration of mobile data and Wi-Fi have reached approximately 50% in these places, rural and underdeveloped regions still have no access at all to these services. It may be necessary to improve the infrastructure in developing countries’ rural and underdeveloped regions in order to be able to promote the use of mHealth apps [74].

Since only full-text articles available in English on PubMed were included in this review, many other studies have not been included because they were not written in English or were still in progress at the time of publication. The papers included here deal mainly with developed countries and less with developing countries, which limits the general validity of the results presented here. Since the rates of smartphone ownership in developing countries increased from 21% to 37% in 2015 and the active mobile phone subscription rates reached 53.6% in 2017, the uptake of mHealth apps may be different in these countries from the situation in more industrialized countries [3,75]. mHealth integration may have considerable implications in developing countries [76]. It is therefore worth noting the perceptions of individuals inhabiting developing countries in order to establish how this technological advancement is liable to improve or limit their access to health care [77].

In this review, a meta-ethnographic approach was used to summarize the data published in 43 qualitative studies on patients’ perceptions of mHealth. Although mHealth apps were considered a useful complementary tool by many of the patients studied, some major issues emerged with regard to the optimal use of mHealth technologies, such as the need for more highly tailored designs, their cost, the validity of the information they provide, and issues such as privacy and security. Lastly, there is definitely a need for apps to be more personalized in order to meet the needs of individual users and their particular disease or condition, by designing apps that are easier to use, for example, by those who are not as digitally literate as others.