Mobile medical applications’ state quo and underlying barriers: review of literature^†

Studies abroad have shown that there are currently 165,000 (free and paid) health apps in 46 major apps stores around the world, categorized into Health Management apps (fitness, nutrition, weight loss, lifestyle adjustments) and Disease Management apps (diabetes, cardiovascular disease, mental health); the latter functions include self-diagnosis, medication reminder, medication therapy, and rehabilitation exercise.¹ Among them, the most popular apps are medication reminder apps, followed by medical contact apps and symptom tracking apps.⁷ Domestic scholars are also focused on the emerging fields; for example, Peng et al.⁸ has invented the Smart Medication Management System (including smart pill box and mini program of WeChat) based on the Internet and big data. It has been used in long-term care facilities and has shown a significant consequence on elevating the medication adherence (Z = 2.085, P < 0.05). Chen⁹ designed a smart medical box based on co-design (or collaborative design), which respectively include designers, users, and other stakeholders involved in the medical box. These apps convey medical solutions based on wireless mobile networks, providing educational support, medication tracking, and remote consultation. Intelligent apps at home and abroad both empower patients and facilitate their self-management. As an auxiliary resort to doctors’ intervention, they encourage more patients to participate in medical decision-making, thereby improving the disease control capability.¹⁰ On the other side, apps increase the communication between patients and their doctors or care team, and are a relatively cheap, affordable, and apply convenient technology that relies on existing mobile networks to remotely monitor patients who are difficult to contact or require strict monitoring. Apps also have the potential to improve the control of risk factors and health conditions, especially for patients with chronic diseases such as arterial hypertension,⁶ diabetes,¹¹ heart failure,¹² etc.

Medication noncompliance has been a common health care challenge. Poor adherence causes approximately 33%–69% of medication-related hospitalization and accounts for US$100 billion in annual health care cost.¹³ The average adherence rate to chronic medication therapy is approximately 50%.¹⁴ One of the functions of mmAPP is medication reminder; it includes 2 other functions: (1) Daily reminder, which can be paused, rescheduled, and (or) marked as “taken” or “not taken” doses; (2) Default setting, that is, periodic reminder every 10 min, up to three times; and (3) Other functions, such as medication supplement reminders, adherence statistics, export functions, and the ability to share information with others, as well as the ability to be reminded by others (e.g. family members) when participants forget to take medication. It can voice “Ding” to remind patients to take drugs by a settled down planned time. It allows patients to enter their medication list manually or by scanning the barcode on the prescription bottles, provides a visualized medication calendar, and automatically sends notification to patients when they are due for medication doses. For example, it can remind you to take medicine before lunchtime, if you take gastric mucosa protectant aluminum hydroxide. Especially for low-adherence chronic patients, it means a lot. Low adherence causes aggravated financial burden due to unsafe behavior of taking medication.¹³ On average, 1 in 6 people need to be reminded of taking medication.³ Therefore, medication reminder is in great demand. A majority of the users (68%) affirmed with the medication reminder,¹⁵ whereas some complained that it did not work as they wished, and that the efficiency and utility needed more research. Considering users’ requirement and personal preference, designers should meet their needs as closely as possible.

By the end of 2020, there was a retrieval of 3.25 million mmAPPs in common Applications Stores (Android and Apple); it had increased to a full 50% more than 10 years ago.⁸ Data show that the customer pays more attention to acquiring and exchanging health information. Unfortunately, less than 3% of mmAPPs are invented by professional medical institutions,⁷ and medical data that refer to patient safety lacks authorization and approval.¹ What is more, although mmAPP is quickly emerging, there are still some barriers including (1) Low quality of mmAPP:¹⁶ to be more specific, low quality refers to font size that is too small to read, voice is too small to hear, and the user interface is too complicated. Morawski et al.'s⁶ research represented that 65.6% patients hoped to enlarge the font, bold the font, and increase the volume. If mmAPP makes an adaptive change, it would be preferable for degenerative hearing and vision patients; (2) Slow update rate: it represented insufficient scientific data, and was deficient in rigorous calculation, especially lack of high-quality best practice based on evidence.¹⁷ So far, the most significant and active evidence are focused only on type 2 diabetes, cardiovascular system, and obesity. Nowadays, we have not retrieved any clinical practical study above intelligent mmAPP, it does not produce any available clinical data yet.¹⁷ Therefore, we suggest adding more high-quality evidence on the application to enrich the content. However, there is a giant challenge that the high-quality evidence needs at least 3–5 years’ clinical randomized control trial, which is far behind the APP's update rate. How to make high-quality evidence at the same pace as the APP update rate is a critical problem waiting to be solved urgently; (3) Lack of a standard evaluation tool or criterion:¹⁸ early in 2016, the U.S. Food Drug Administration (FDA)¹⁹ authorized a company named Social Wealth to proceed with certification approval and a quality control of 328 mmAPPs. Even today, a number of countries and regions still have not build a general, standard criterion for mmAPPs. Thus, we hope that the government departments, third-party entrepreneurs, and professional groups should construct a standard and systematic method to evaluate the APP's utility and effectiveness.

Sunyaev et al.²⁰ have shown that among the 600 most used applications, only 183 (30.5%) made a privacy protect policy, while two-thirds of the applications did not specially design a privacy protect policy. Bitglass²¹ (a technology company to provide safety service) medical safety report claims that in 2016, in the United States, over 111 million of personal data were lost due to hacker attacks or other reasons. Each lost or stolen medical record resulted in an average loss of US$ 363. The National Health Service (NHS)²² randomly examined 79 mmAPPs and found that none of them stored personal information on mobile devices in an encrypted way. As high as 66% of applications failed to protect and encrypt users’ data. In my perspective, it is a tremendous potential threat for digital data safety and privacy based on the Web and Internet methods. Several studies and reports²⁰ have shown that the majority of mmAPP developers currently fail to provide privacy policies for their applications. Users have expressed concern and worry about the safety and effectiveness of mmAPP. More than 50% of users choose to unload or uninstall applications for fear security of their personal information,⁷ and call for stricter scrutiny on the privacy and security of mmAPP. Consequently, we suggest strengthening the protection of users’ personal information and privacy guidance. We call for relevant departments to implement the privacy protection policies of mmAPP for users’ personal information, and impose certain economic penalties on applications that illegally embezzle personal information. Application developers should guarantee the security of applications, and ensure rigorous testing prior to releasing, in order to reduce the underlying system vulnerability such as data storage, information encryption, and identification verification.