Privacy is the ability to control the recording and sharing of personal information with others. This requires knowledge of what will be recorded, how it will be used and for how long, who will have access to this information, and what the risks are of discovery and misuse by third parties.

Before using mHealth technologies, clinicians and researchers need to seek and obtain the informed consent from participants. Participants need to be informed of the risks and benefits of using mHealth technologies and must have the capacity to understand these risks and make a free and uncoerced decision about whether to participate [28]. A challenge in mHealth is communicating the complex nature of the risks raised by this technology and negotiating the risks that individuals are willing to face. How much input should people have over what they wish to have recorded and shared? How will the data be used, where will it be stored and in what form, how will it be shared, and for how long? Participants must also be made aware of what will happen to their data once their study participation is complete.

Researchers and clinicians using mobile phone technologies need to develop consent processes that actively engage individuals in their own privacy decision making as much as possible. Understanding the risks posed by third-party access to their personal health data can be difficult to communicate given the complexity of mHealth technologies. Meeting these ethical challenges will require technology developers to create apps that assist participants in understanding their participation and give them as much say as possible in what is shared and with whom. This may involve trade-offs between, on the one hand, maximizing the data obtained and what may be done with it, and, on the other hand, enabling participants to control what they consent to and how their data may be used and stored.

As part of the consent process, researchers and clinicians need to inform participants about the circumstances in which they are obliged to disclose the participants’ personal information. This potentially includes information that can pose an immediate and likely threat to themselves or to others (eg, suicide, homicide). Patients must also be informed of the potential for identification of “incidental findings,” specifically clinically significant information related to their health, such as cognitive impairment and dementia, depression and other psychiatric disorders, or medication abuse or other compulsive behaviors that may benefit from additional treatment. Patients should be informed about the processes in place to deal with potential incidental findings and provide them with further information and access to clinical treatment should any emerge. Participants should also be given the opportunity to indicate whether they would like to be told about the presence or absence of incidental findings.

The potential for mobile phones to record conversations about third parties or bystanders raises additional ethical and legal concerns. For example, mobile phones may record conversations to examine the impact of PD on patients’ speech and socialization [11,12]. Researchers may obtain consent from the research participant to have their conversations recorded but they cannot easily obtain consent from bystanders (eg, friends and family) who may also be recorded via the mobile phone. The recording and/or communicating of third-party conversations (even to a secure portal) is illegal in some jurisdictions. Researchers may therefore need to develop methods for ensuring that bystanders are able to consent to having their conversations recorded. One option would be to provide an alert on the mobile phone indicating to the participant that a recent conversation has been audio recorded so that they can ask third parties to consent to the recording. An opt-in approach could be applied where the conversation is only sent to researchers or clinicians if the third-party individuals agree. This would further burden the participant who would have to disclose their mHealth participation and, therefore, possibly their clinical condition to obtain the third party’s consent. The opt-in approach also obtains consent after the fact, and therefore, may be inadequate in some jurisdictions.

How mobile phone data are stored and shared can have important implications for privacy risk. Developers may choose to create a secure “vault” on the device that is physically transferred to researchers or clinicians at predetermined intervals. This allows greater control of data and protection for participants, but at the cost of reduced flexibility and timeliness. It could limit the benefits that a patient or participant may derive from the ability to self-manage their condition.

The most clinically and scientifically useful approach is to periodically upload the data electronically (eg, via 3G or wireless networks) to researchers or clinicians. This allows for instantaneous transfer of data, but provides less privacy protections as it involves transmitting data over third-party networks. Developers will need to decide where and when data are uploaded (eg, only at certain locations or times, such as from a home Wi-Fi network) as opposed to routinely uploading via public Wi-Fi or mobile networks. Developers will need to reach a balance between maximizing clinical or scientific utility while minimizing the risk to privacy. Solutions will vary depending on how the technology is being employed.

Developers will also need to consider what data to collect and transmit from the mobile phone. The ability to collect as much raw data as possible maximizes the information that a research and clinical team can extract. It also increases the potential risk to the participant. A balance needs to be struck between maximizing scientific and clinical utility and protecting privacy. It is preferable to only collect data sufficient for the purpose of the study or clinical intervention (eg, postcode or distance from home), rather than collecting all data routinely. By collecting only the minimum amount of information, app developers can help participants maintain control over their raw data [20].

Developers also need to make data comprehensible to the participant before they are sent or before consent is given to share the data. Developers should provide easily understandable visual information to participants about where their information will go, who will have access to it, and for how long. This will greatly increase participants’ understanding of the risks involved with this technology and enable them to make more informed decisions. Such processes will help participants to see what their sharing policies are and what the results of these policies will be [20].

A critical decision in the clinical or research use of mobile phones is when and what to tell research participants or patients about their data. The communication of clinically meaningful feedback to participants about their data maximizes the benefits for participants and mitigates the harm from potential privacy violations. To be clinically meaningful, the findings must be scientifically robust. The ethical obligation to share an individual’s data with them is more acute in the clinical setting where the explicit aim is to facilitate better clinical management of the disease, and there is strong evidence to support the clinical claims being made. Feedback is also provided by a qualified individual responsible for the patient’s clinical management. In research, the clinical relevance of data is, by definition, less certain because the research aim is to establish an evidence base for clinical intervention. In research, providing individualized clinical feedback from the data is inappropriate, and may lead to additional harm if the clinical advice is misleading or provided by nonprofessionals.

The provision of immediate and useful information by appropriately qualified clinicians can potentially better enable individuals to manage their health and improve the clinician/patient relationship by empowering patients to take more control over their health and well-being [29,30]. However, to serve this purpose the information collected by the mobile phone needs to be presented in ways that are meaningful, accurate, and easily understood. Well-designed platforms that provide such feedback to patients will maximize the potential for individuals to self-manage their health and well-being. Participants may also be assisted to access effective clinical services for the treatment of any symptoms identified in the data collected via their mobile phone. Developers and researchers may also wish to identify opportunities in the app in which participants can receive information that will enable them to make healthier choices. Research is needed to establish what impact the provision of clinical information recorded on mobile phones may have on a person’s behavior. Communication of deteriorating motor symptoms, for example, could adversely impact on well-being and health-related behaviors. See Eonta et al [31] for practical guidance on the ethical communication of clinical information to patients using mobile phones and other mHealth technologies.

While there has been a rapid growth in mobile phone coverage in recent years, some segments of the population lack access. People from lower socioeconomic groups may not be able to afford a phone capable of supporting the app or connecting with mobile or Internet networks required to transmit potentially large volumes of data. There is an ethical imperative not to exclude these patients from benefiting from mHealth monitoring.

Disease-related impairments may also interfere with the effective use of these technologies. PD may impair both fine motor and speech skills, therefore creating difficulty in accessing touch screens or voice-based interaction with mobile phones. Designers of mHealth apps must consider ways of reducing the effects of these cognitive, motor, or other impairments. Having a choice on how to interact with an app is one approach (eg, voice-activated, image- or text-based interface). People with PD who experience fine motor difficulties may be assisted by reducing the sensitivity of the screen to repeated touch, appropriate spacing and sizing of buttons, use of swipe on-screen keyboard, and external devices such as a stylus. People with voice-related difficulties may be assisted by a plug-in microphone and software that can be trained to an individual’s voice (eg, Dragon Dictate).

Given the complex array of symptoms and differences in comfort and literacy with technology, it is likely that these solutions will need to be individualized. It is therefore critical that developers of mHealth apps engage with patients throughout the development process to ensure that the technology meets their needs. This will assist developers, clinicians, and research teams to target the symptoms that are of greatest concern to patients. Developers should use reference groups (eg, consumer, family, industry, health care professional) to anticipate challenges in developing technology that meets these challenges [11]. As one patient reported in regard to the use of mobile phones for diabetes, “It’s not just about blood glucose results and HbA1c results, it’s about how people feel and, perhaps, how their mood may affect their glucose levels” [29]. A summary of the recommendations for managing these ethical challenges is presented in Table 1.

An explosion in mHealth research [32] has led to a rapid proliferation of small pilot or seed programs, many of which lack scientific evidence of efficacy on which to base clinical use. There is limited evidence on the effectiveness, or safety, of mHealth apps as a self-management tool for improving health [29]. Hence, there is an urgent need to evaluate their effectiveness via randomized controlled trials before the widespread rollout of mHealth apps and associated products (eg, brain or heart monitors) [33]. This need was recognized by the World Health Organization and other leading agencies responsible for implementing medical products in the Bellagio call to action on global eHealth evaluation that called for rigorous evaluation “to generate evidence and promote the appropriate integration and use of technologies...to improve health and reduce health inequalities” [34].

It is imperative that the use of mHealth apps by researchers, clinicians, universities, and hospitals is based on rigorous evaluations of their effectiveness and safety [29], using the CONSORT-EHEALTH checklist [35] and GRADE framework [32]. Policy makers should not be seduced by the hype and promise of mHealth technology. The prima facie simplicity and cost effectiveness of mHealth solutions may blind decision makers to the lack of robust empirical evidence that is needed to justify their routine use [29]. Premature implementation of untested mHealth interventions may result in failed projects, wasted resources, and poorer health outcomes for patients.

There is currently no regulation of mHealth devices or apps and no guarantee that they provide clinically accurate information. The Food and Drug Administration (FDA) recently released guidelines for how they intend to regulate the marketing of mHealth apps that meet the definition of medical devices (ie, those “whose functionality could pose a risk to a patient’s safety if the mobile app were to not function as intended”) [36], although these recommendations are currently nonbinding and do not prevent apps from being made available. The FDA has recently approved the marketing of an mHealth app to continuously monitor glucose levels [36].

However, most apps are made available to patients directly via publicly available app stores, without passing through regulatory gatekeepers to ensure their safety and effectiveness. The clinical use of these devices and apps needs to be regulated in the same way as any other medical or psychotherapeutic intervention. This is particularly important given the influence of vested commercial interests that may push for quick rollout and be more concerned with growing markets than improving global health [37]. The regulation of mHealth products would also help to minimize privacy violations through malware or other computer-based viruses. The development of a Web-based approval system for verifying the quality of mHealth apps, such as the Health on the Net Foundation Code of Conduct, or National Health Service app stores, such as the United Kingdom Health Apps Library, may be useful for ensuring the quality of mHealth apps.

Regulators will also need to deal with the increasingly globalized nature of mHealth research. Traditional processes of institutional ethics approval and recruitment provide significant barriers to conducting mHealth research that crosses national borders and involves numerous agencies. Ethical oversight of mobile phone research will need to evolve in order to allow it to happen, and to prevent it from being done with insufficient ethical oversight by commercial entities (eg, Apple, Facebook, Google). Failure to do so may leave these companies with a vastly superior understanding of health and behavior than researchers, governments, and policy makers. Unlike publicly funded research, these findings will be protected by commercial-in-confidence and trade secrets laws.