Efficiency and prevention

mHealth is argued to offer efficient and cost-effective solutions for disease prevention, monitoring and management. Take for example an app that uses smartphone cameras to check moles for skin cancer risk (eg, https://skinvision.com/): the user holds the device over a spot on her skin and takes a picture, and the app’s algorithm immediately analyses the spot and recommends an action to take (eg, whether the user should go to a specialist) or provides relevant information on skin cancer. This system allows users to archive their skin pictures, keep track of changes over time and share them with their doctor. These types of apps claim to prevent risks of skin cancer by offering a cheap tool for early self-assessment.

To live up to these promises, mHealth systems, like other medical interventions, should not harm and should (preferably) benefit users. However, despite the hype around mHealth, there are still many uncertainties around the safety, reliability and accuracy of mHealth systems.9 10 Risks include security issues and harms that may derive from potential disclosure of sensitive information to third parties or identity theft,11 and risks of false results. For example, a skin screening app may not be accurate and might fail to recognise an early-stage melanoma, falsely reassuring rather than alerting the user.12 Other concerns have been raised concerning the efficiency of these devices: it has been argued that most sensors are unreliable in their signal detections, that it is unclear whether these devices can induce people to change their behaviour towards healthier lifestyles,13 and that clinical professionals are unaware of how to integrate these systems in their workflow.14

Moreover, when weighting the benefits and harms of mHealth, one needs to think of which benefits and harms are involved, and to consider who will actually benefit and who will experience the harms from such technologies. Some apps are primarily helpful for patients and may simultaneously help physicians, healthcare personnel or commercial parties to monitor patients, do research or sell products. The benefits and risks may not be evenly distributed among these stakeholders. It is important to ensure that the risks of harm will not predominantly lie at the user/patient side, because they are the more vulnerable actors and may have no understanding of or protection against such harms.

In general, despite the promise of providing an effective and efficient tool for disease prevention, the actual benefits of mHealth are still unclear. Although mHealth systems are sometimes presented as a replacement of traditional care, it is unlikely that they will completely replace traditional healthcare. Empirical research has indicated that patients and physicians are happy about the quality of web consultations,15 but little is known about their safety, feasibility, cost-effectiveness and efficiency. It is not clear in what contexts and situations these tools are the best possible option to meet patients’ care needs: for example, it may not always be desirable, as face-to-face contact can be preferred, more efficient or simply irreplaceable, because of the need for physical examination, sensitivity of the problems, validation of mHealth data or for maintenance of the doctor–patient relationship. Also, more research is needed to understand which types of interventions and monitoring can be left to apps and wearables, and when personal, synchronous care relationship will need to be kept to ensure good care.

Self-management and empowerment

Another promise is that mHealth will empower patients.2 Many apps and wearable sensors provide self-management functions for patients or healthy citizens who want to engage in healthy lifestyles. Having access to one’s own health data, without the interference of health professionals, is argued to be a prerequisite to understanding, controlling and managing one’s behaviour, and therefore regarded as empowering patients and supporting their autonomy.16–18 An example is offered by the apps provided mymhealth.com for patients with asthma, chronic obstructive pulmonary disease, diabetes and heart disease that enable patients to ‘achieve their goals and self-management’ while being followed by a clinical team (see https://mymhealth.com/).

The claim that these digital tools empower patients and support their autonomy through self-management should be critically assessed with regard to the degree and types of autonomy that these tools promote. The underlying idea of self-management of such apps induces patients to comply to a strict (medical) regimen rather than enhancing their self-determination.19 As these devices influence or direct the behaviour of the users, concerns have been uttered that these technologies may be disciplining the users to perform specific medical tasks, rather than truly empowering them.17 18 Moreover, as described in the context of governmental surveillance of internet use, the so-called ‘chilling effects’20 can make users of mHealth adapt their behaviours when using wearables and apps because of the feeling they are being monitored, rather than because they are themselves motivated to behave in a certain way. This means that mHealth technologies indirectly hinder them to act in the way they would have otherwise, and raise questions concerning the true meaning of the promised patient empowerment. Not only expectations that apps and self-monitoring devices empower patients are deceptive because they promote a specific medical quantified regimen rather than supporting a plurality of self-determined patients’ valueⁱ, they also seem to assign extra responsibilities to patients. In fact, these apps and wearables delegate some tasks to patients that are traditionally carried out by healthcare professionals (eg, monitoring vital signals, updating symptoms) and continuously demand patients to perform tasks within a certain time frame. This shift of tasks is the flip side of having direct access to health data and falls into the general neoliberal trend of shifting responsibilities from the state (or healthcare providers/experts) to citizens (or individual patients), a trend often referred to as ‘responsibilisation’ (see http://nuffieldbioethics.org/report/personalised-healthcare-2/what-is-personalised-healthcare). Several concerns have been raised towards such trend and are likely to apply to the field of mHealth18 ,ⁱⁱ: first of all, it may charge patients with additional practical burdens and labour which would otherwise be taken up by healthcare professionals.21 Also, such burden may cause unduly emotional stress for patients who may feel left alone to their own care by the system. Furthermore, as several empirical studies indicate, merely having access to data does not imply that people have an increased control over their behaviour and health outcomes, as for behavioural change to happen the intervention of a health professional or counsellor is often necessary. 22

Whether the expectations of empowering patients are indeed met or not, an aspect that needs close consideration is how these devices and practices influence the doctor–patient relationship. It is not only a matter of introducing more distance in the doctor–patient interaction via online tools, but also a matter of giving patients tools to control their health and rethinking the roles that doctors have in healthcare.23 By shifting tasks and responsibilities, proposing specific roles and offering tools to understand health and its correlation with lifestyle, these systems introduce new interactions between care providers and patients. This raises questions on how these relationships will take shape, whether valuable features of traditional relationships will get lost, whether medical training needs to be rethought or patient expectations redirected.

Yet another critical aspect related to promises to support patients’ autonomy concerns the confidentiality and control of data. Data collected and processed by wearable sensors and apps may contain sensitive information that is made accessible to several actors, as once such data are obtained they can be reproduced and used endlessly. For example, data may be analysed by the manufacturers for improving the app performance or sold to third parties for research or marketing. Users of such wearables and apps often have limited control over who has access to their data behind the scenes. Third parties’ access to health-related data can be harmful for app users who could, for example, be profiled by marketing agencies and receive targeted unwanted or upsetting advertisements,24 or incur in discriminatory policies by insurance companies or employers. Although safeguards are in place and users have the right to know what data will be collected, how these will be processed and to whom these will be distributed, this appear to be problematic and complicated in practice in the age of increasing data collection and big data analyticsⁱⁱⁱ.25 Just to mention two issues: first, in a context wherein credit card purchases reveal individuals’ lifestyle choices (eg, eating or working out habits) and enable correlations with health conditions through linkages with medical or health data (collected via apps, for example), the distinction between sensitive and non-sensitive data is ambiguous, and the definition of health-related data becomes too broad to be meaningful.26 Second, traditional safeguards in the clinical context, like practices of informed consent, are inapplicable in a consumer-focused domain where the mediation of healthcare professionals and researchers is shrinking. In an age of ambiguity between commercial and medical domain, confidentiality of health data is constantly challenged, and there is a need to discuss the moral basis of governance models that foster autonomous choice and encourage mHealth users to make an informed decision and control which data remain protected.^iv

Wide accessibility to health and social justice

Finally, mHealth comes with the promise to make healthcare widely accessible.2 mHealth is argued to have the potential to bring healthcare to areas that are difficult to reach, or to people who have mobility problems and cannot easily go to the hospital to receive care. An example can be found in a campaign raising health awareness in Bangladesh via SMS (text messages) to mobilise citizens for National Immunisation Day. 1 In the context of this campaign, messages were sent to encourage parents to bring their children to get vaccinated, along with the event’s date. A similar campaign involved pregnant women in remote villages who could register their mobile numbers to receive prenatal advice related to their gestation stage. With the spreading of mobile phone use in developing and rural areas, mHealth is considered a tremendous opportunity to improve healthcare in resource-poor countries in affordable ways.

The promise of increasing accessibility to healthcare meets several challenges. For example, mHealth interventions like the ones described above, using SMS (text messages), may not reach those who are most in need of care, either because they are illiterate or local languages are not supported by mobile phones,27 or because apps require mobile phones with a fast internet connection and some abilities to interact with these phones, thereby excluding certain groups (eg, low-income groups or elderly people with less dexterity with smartphones). These issues raise the question whether the people who actually are in need of better care or improved access to care are currently being served by mHealth technologies. Furthermore, it also indicates a certain hidden normativity in the ways in which services are offered and the expected users. Accessibility-related concerns also play a role in Western countries as users of some of these technologies (eg, iPhones, Fitbit) are on average younger, more educated and wealthier than average. These aspects threaten the promises of accessibility to health and raise questions whether mHealth exacerbates rather than mitigates issues of social justice both at a global and local level by excluding groups from using these services.

From promises to practices

When exploring potential implications of mHealth systems, we need to be aware that developers’ websites, explanations or demos are only one side of the story describing the intended use. The way in which these technologies will actually be used in specific contexts by real people is a different matter, but both may be relevant for the ethical assessment of mHealth technologies. Literature on the history of technology and user studies show that there is often a gap between expected technology-driven scenarios and their actual realisation in specific societies.28 29 When used in practice technologies often ‘bite back’, having the exact opposite effect in practice to what they were supposed to do.30 For example, although computers were often introduced in working spaces with the promise of creating a ‘paperless office’, they had the opposite effect as people tended to print the same or even more than usual to store and distribute documents. Mismatches between designers’ expectations of users’ preferences and habits, on the one hand, and actual uses of technologies are very frequent as technologies introduce new burdens, labour and responsibilities that remain invisible at a first sight.21 31 An investigation of actual uses is not always feasible (eg, the system is at an early stage of development or only used in few pilot studies, or logistical constraints prevent ethicists from engaging in such fieldwork). However, bioethicists can play a role, already during the development of devices, by uncovering the hidden normativity in the intended use and users, and by exploring possibilities for abuse and unforeseen effects. This type of early analysis can result in an alteration of the design of the device, app or service. To this end, not only the end-product but the criteria that inform different design stages should be explored and critically assessed. Bioethicists can ask questions like: is this particular system configuration going to exclude some users? Shall users be enabled to choose how they want their clinicians to monitor them through an app? These questions can then be discussed with manufacturers and prospective users to design more desirable products.

It is therefore important for bioethicists to go beyond the abstract promises and rhetoric of mHealth and situate the ethical analysis in real-world practices and specific contexts. As mHealth is often used as an ‘umbrella term’ to refer to a broad variety of devices with diverse functionalities, it is crucial to establish relevant moral differences among these objects and raise specific normative questions concerning the types of users, places where systems are used, the type of device and practices that develop around it.

Beyond safety and effectiveness

As the problem of securing sensitive health information and protecting individual privacy becomes crucial in the context of commercial gadgets collecting health data and the question of uploading these data on medical records is explored, most policy discussions around mHealth focus on issues of effectiveness and safety of these devices^v. While these topics are indeed crucial for the ethical assessment of mHealth as they refer to principles of beneficence and non-maleficence, these are not the only values and principles that are relevant from an ethical perspective. Issues of social justice, for example, should receive more attention. As discussed above, while mHealth has the potential to serve a wider range of people, it might as well exclude a large range of proportion of the population from care altogether. Questions with regard to who has access, who is excluded and whether those in need are actually served and get adequate care require careful analysis and justification, already during the development of such technologies. Some of these questions are not new as mHealth is perpetuating existing injustices of problems with new means. However, digital health, including mHealth, also brings about new ways of exploitation and commodification of data. Many mHealth services have an ambivalent status between lifestyle/well-being gadgets and health-related products,32 making it opaque how these services should be governed. Data shared on social media, online patient platforms or sold to third parties require ethical scrutiny, for privacy reasons and to protect vulnerable groups from harm and discrimination. Finally, as mentioned above while changing the ways in which healthcare is offered, mHealth affects the relationship between healthcare experts and patients, as well as people’s understandings of self-care and management. The meanings of mHealth services for individuals and collectives should be explored both empirically and conceptually. Bioethicists are exceptionally placed to bring these more conceptual and empirical issues into the public debate and extend the current discourse beyond questions of effectiveness and safety of mHealth.