Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

‘Personal Health Surveillance’: The Use of mHealth in Healthcare Responsibilisation

‘Personal Health Surveillance’: The Use of mHealth in Healthcare Responsibilisation PUBLIC HEALTH ETHICS VOLUME 14 ISSUE 3 2021 268–280 268 • • • ‘Personal Health Surveillance’: The Use of mHealth in Healthcare Responsibilisation Ben Davies *, Uehiro Centre for Practical Ethics, University of Oxford *Corresponding author: Ben Davies, Uehiro Centre for Practical Ethics, University of Oxford, Suite 8, Littlegate House, Oxford OX1 1PT, UK. Email: benjamin.davies@philosophy.ox.ac.uk There is an ongoing increase in the use of mobile health (mHealth) technologies that patients can use to monitor health-related outcomes and behaviours. While the dominant narrative around mHealth focuses on patient empowerment, there is potential for mHealth to fit into a growing push for patients to take personal responsibility for their health. I call the first of these uses ‘medical monitoring’, and the second ‘personal health surveillance’. After outlining two problems which the use of mHealth might seem to enable us to overcome— fairness of burdens and reliance on self-reporting—I note that these problems would only really be solved by unacceptably comprehensive forms of personal health surveillance which applies to all of us at all times. A more plausible model is to use personal health surveillance as a last resort for patients who would otherwise independently qualify for responsibility-based penalties. However, I note that there are still a number of ethical and practical problems that such a policy would need to overcome. The prospects of mHealth enabling a fair, genuinely cost-saving policy of patient responsibility are slim. and communications technologies for health (WHO j Mobile Health, Surveillance and eHealth, n.d.). mHealth includes applications on mobile Two Challenges for phones as well as more direct monitoring of patient health indicators such as wearable monitors and at- Responsibilisation home testing kits whose results can be transmitted by Technological advances are providing increasing abil- patients to medical professionals (DiStefano and ity to monitor health outcomes and health-related Schmidt, 2016). behaviours outside traditional clinical settings and mHealth has the potential to facilitate two functions. relationships. Patients can self-administer tests for First, it may allow us to monitor our bodily processes blood sugars (Cvrkel, 2018; Istepanian and Al-anzi, which, while affected by behaviour, are not under direct 2018); oxygen saturation (Pantelopoulos and control (Lupton, 2013, 2019). If a patient knows broadly Bourbakis, 2010); blood pressure (Weber et al., which behaviours affect the relevant processes, they can 2012); heart rate (Chow et al., 2016: 804); mood attempt to indirectly moderate their health. mHealth (Harrison et al., 2011); and neurological function may thus relocate routine health monitoring from expli- (Behar et al., 2019). We can monitor health-related citly medical settings to the home, workplace and wider behaviours more easily (Sharon, 2017): wearable tech- world, which Swan (2012) describes as an ‘institutional nologies can help monitor activity levels and diet recasting’ of healthcare (see also Carter et al., 2015). This (Connelly et al., 2013); alcohol consumption (Cohn can be seen positively as ‘shifting (health management) et al., 2011); and medication use (Cavoukian et al., into the hands of empowered patients’ (an ideal 2010; Martani and Starke, 2019: 251), as well as pro- reported, critically, by Ruckenstein and Schu ¨ ll, 2017: viding mental health services (Martinez-Martin and 262), liberating them from time-costly medical appoint- Kreitmair, 2018). Collectively, these technologies are ments (Topol, 2015), or more negatively as an over- known as mobile health (mHealth) (WHO Global medicalization of previously more carefree spaces. Observatory for eHealth and World Health Second, mHealth may facilitate monitoring of behav- Organization, 2011). mHealth is generally defined as iours that affect our health, but which are difficult to a subcategory of ‘e-health’ (Chatzipavlou et al., 2016: track unaided, and about which we are wont to self-de- 1), which encompasses the general use of information ceive. I tell my doctor that I stick to the UK government’s doi:10.1093/phe/phab013 Advance Access publication on 16 May 2021 V C The Author(s) 2021. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. PERSONAL HEALTH SURVEILLANCE 269 guideline of 14 units of alcohol per week. Perhaps I be- measures which would increase costs or affect access to lieve this to be true; judging my unit intake requires a care for those who are suitably responsible for their ill relatively complicated calculation. The relative complex- health (Schmidt, 2007, 2009; Hancock, n.d.; Ter Meulen ity of keeping tabs on an enjoyable activity facilitates my and Maarse, 2009), with Lupton (2016: 155) noting how reluctance to confront the truth about my alcohol con- Anglophone countries have retained a focus on personal sumption. An app that calculates the units for me might responsibility that developed in the mid-20th century, help me follow the guidelines. with a ‘renewed emphasis on lifestyle change’. The pos- We might therefore see mHealth as a way for patients sibility of responsibility-centred rationing is not an ab- to take back control over their health from institutional stract possibility, with Pillutla et al. (2018: 1) noting medicine, saving time and effort by reducing unneces- recent local proposals in the UK to ‘restrict elective sur- sary contact with medical professionals. Not coinciden- gery for patients who either smoke or are obese’. tally, optimists might see mHealth as promising public When health costs are borne partly or wholly by the spending savings at no cost to public health. Finally, the state, it is not a significant leap to think that where in- data generated by the use of mHealth has the potential to surance companies and private employers lead, public feed into public health research. health systems could be tempted to follow. At one ex- Yet mHealth could be integrated into healthcare in treme is the widespread use of data for citizen tracking another way, using technological monitoring to increase currently operating in China (Botsman, 2017). Yet even the role of individual responsibility not only as a method if such a comprehensive system of surveillance seems of empowering patients, but also to hold them account- unlikely in more democratic states, the use of mHealth able as users of public resources. For clarity between may seem to ‘close the loophole of practical enforceabil- these two uses of mHealth devices I use the term ‘medical ity’ when it comes to judgements of personal responsi- monitoring’ for more standard current uses of monitor- bility (Martani and Starke, 2019: 241). ing devices, e.g. allowing patients to access data about As mHealth grows in both private and public usage, themselves. When it comes to using mHealth to enforce particularly in the context of increased political and so- responsibility, I will use the term ‘personal health sur- cial focus on personal responsibility, it is important to veillance’. The same device or app can therefore be used consider the implications of using mHealth to enforce in either monitoring or surveillance. As far as I am aware, responsibilisation. In doing so, I remain neutral for the this distinction has not previously been explicitly dis- sake of this article about whether it is legitimate to hold cussed in work on mHealth. people responsible for their health by imposing add- The use of mHealth by some insurance companies itional burdens on their access to care when their poor (Shemkus, 2015; Lupton, 2016: 164; Henkel et al., health is suitably due to their own free choices. These 2018; O’Neill, 2018; Martani et al., 2019) offers insight burdens may range from the severe (denial of care) to the into the possible institutional uses of mHealth. While mild (some additional co-payments). most companies currently use mHealth technology by There are a number of arguments for and against the offering positive rewards to those who achieve particular idea of healthcare responsibilisation. Some ‘luck egali- targets, Raber et al. (2019: 1767–8) note that these sys- tarians’, for instance, consider inequalities that individ- tems ‘could be used by insurers in the future to penalize uals cannot avoid to be unjust, but do not necessarily users’. Similarly, where private employers are respon- condemn inequalities which reflect exercises of respon- sible for part of employees’ health insurance costs, there sibility (e.g. Arneson, 1999; Dworkin, 2000: 77–8; have been attempts by some to mandate employees’ Vallentyne, 2002; Lippert-Rasmussen, 2016); while health-related behaviours through the use of mHealth not all luck egalitarians apply this view directly to real- (Lupton, 2016; Nissenbaum and Patterson, 2016: 84, 88; world healthcare, some do (Roemer, 1993; Cappelen and Barlin, 2018). Writing just over a decade ago, Hendrix Norheim, 2005; Segall, 2010; Le Grand, 2013: 303; and Buck (2009: 466) describe how ‘employers have Albertsen and Knight, 2015; Albertsen, 2020), though begun to implement increasingly aggressive wellness generally not in anything like the simplistic manner programs that provide incentives to employees who imagined by critics. Others have argued from alternative meet certain health standards, while creating disincen- perspectives that responsibility may be a reasonable part tives for those who fail to meet the standards’. of any healthcare system (Buyx, 2008; Savulescu, 2018). My central focus in this article is on the potential for Others regard the luck egalitarian stance as excessively the state to follow suit. The idea of responsibility is a ‘harsh’ (Fleurbaey, 1995; Anderson, 1999; Voigt, 2007; familiar theme in publicly funded health, with many Venkatapuram, 2011, 198); see the practical aim of hold- jurisdictions either implementing or considering ing people responsible as inappropriately focused on a 270 DAVIES small section of our choices (Minkler, 1999; Wikler, Some of the issues raised by infectious disease surveil- 2002; Sharkey and Gillam, 2010; Friesen, 2018); criticise lance are ones which also affect the use of mHealth in the a reframing of social problems as individual ones (Ayo, context of personal responsibility. A particularly obvious 2012; Lupton, 2012); or doubt our ability to appropri- example is privacy, e.g. Lee (2019: 323). However, while ately take responsibility for our health (Levy, 2018). Yet there may be cases where a person is suitably responsible it is of independent value to demonstrate, as I hope to do, for having an infectious disease which is the subject of that even if we grant the legitimacy of holding people traditional infectious disease surveillance, the justifica- responsible in some cases for their poor health, it is tion for surveillance in this case—preventing the spread very difficult to justify the use of mHealth technologies of disease—is very different than the justification in cases for enforcing this. of responsibilisation. Indeed, the primary justification I will shortly outline an initial ‘optimistic’ case for for infectious disease surveillance offered in the litera- how mHealth might indeed ‘close the loop’ of en- ture is a broadly consequentialist one, presupposing a forceability for personal responsibility (e.g. Swan, specifically health-related benefit which could not be 2012; Wiederhold, 2012; Topol, 2015) before going achieved in other ways, and which outweighs potential on to raise a number of ethical and practical chal- harms (Fairchild et al., 2008; Lee et al., 2012: 38–42; lenges. Before doing so, however, it is worth com- Petrini and Ricciardi, 2015: 273; WHO, 2017). Even menting briefly on an area of public health policy this is not universally accepted—for instance, Rubel which I will not discuss in detail, but which has (2012: 2) rejects justifications that reveal to an aggregate tangential relevance to this issue. This is the issue good, arguing that surveillance can be justified only if it of traditional public health surveillance for the pur- protects ‘basic interests’—but in any case does not ob- poses of controlling infectious disease. This issue will viously apply to surveillance in the service of responsibi- be familiar to many because of the (at the time of lisation. Rather, the most obvious justification for writing) ongoing COVID-19 pandemic. Infectious responsibility-based surveillance would be desert-based, disease surveillance obviously predates this crisis i.e. that those who are suitably responsible for their ill and is primarily justified by the potential for expo- health ought to bear the burdens of it (financial or other- nential escalation and significant harm (Gilbert et al., wise). Of course, one might also hope that a focus on 2019: 176). personal responsibility will improve public health by Fairchild et al. (2008: 30) outline a traditional under- disincentivising certain behaviours. Yet the central jus- standing of infectious disease surveillance, as ‘the on- tification for infectious disease surveillance seems inimi- going, name-based reporting of cases of disease to state cal to the idea of responsibilisation, with the WHO and local health departments’. However, others (e.g. (2017: 46) reinforcing the idea that relevant data should Samerski, 2018:1; Mello and Wang, 2020: 951) note not be used, nor given to those who would use it, to ‘take the growing influence of mHealth in potentially more action against’ individuals. proactive—and invasive—surveillance, including in the Moreover, whereas infectious disease surveillance is context of COVID-19 (Ve ´ liz, 2020). A number of typically focused on aggregate effects and on guiding authors stress the centrality of surveillance to public public policy, personal health surveillance by necessity health efforts, as well as the potential risks of failure to will involve a focus on individuals. Thus, while there surveil (e.g. Fairchild et al., 2008: 30; Petrini, 2013; are some clear parallels between existing infectious WHO, 2017: 10, 17; Gilbert et al., 2019: 176; Lee, 2019: disease surveillance and ‘personal health surveillance’, 320; Wood et al., 2019) with Mello and Wang (2020: the latter is a clearly distinct (potential) phenomenon 951) suggesting that ‘the question is not whether to use that could not easily draw on the existence of the new data sources—such as cellphones, wearables, video former for justification. Nonetheless, both types of surveillance, social media, internet searches and news surveillance may fall under the broad sphere of ‘public and crowd-sourced symptom self-reports—but how’. health’. Whereas infectious disease surveillance is On the other hand, there are clear ethical issues involved more obviously concerned with public health, namely in infectious disease surveillance, especially when it is the targeting of public health policy, personal health opposed by many of those who are sufferers of the par- surveillance may be concerned with a number of ticular condition in question, as has been the case with issues that are related to public health, including pre- HIV/AIDS in some jurisdictions (Fairchild, 2003; vention of disease by disincentivising irresponsible be- Fairchild et al., 2008: 32–34; Klingler et al., 2017: 1–2; haviour, and the appropriate allocation of public Lee, 2019). health resources. PERSONAL HEALTH SURVEILLANCE 271 I return now to what I termed the ‘optimistic’ case for Surveillance might seem to mitigate both problems. Of course, only the most intrusive surveillance state the use of mHealth technologies in healthcare responsibilisation: could hope to fully eradicate the problem of detectability (and even this is doubtful). Responsibility for poor The state or an appropriate medical authority mon- health, and various factors that might justify unhealthy itors whether patients are behaving in appropriate behaviour, typically comes before any interaction with ways (e.g. taking moderate exercise) given their healthcare services. A comprehensively non-discrimin- health needs or achieving certain health targets atory system seemingly needs to surveil all individuals. (e.g. reductions in cholesterol) without patients needing frequent, direct medical contact. Since Unhealthy behaviours do not occur only in public, nor patients have direct, quantifiable access to their can they always be detected after the fact. So, individuals health outcomes on a daily basis, they take greater would need to be surveilled at all times for the most responsibility for their health. Behavioural targets comprehensive—and, thus, one might think, fairest— are more precise: for instance, rather than recom- information about responsibility. For instance, Martani mending that a patient take ‘regular, moderate ex- and Starke (2019: 252) consider the possibility of health ercise’, doctors can recommend more personalised providers forcing a choice to prospective patients be- targets, knowing that the patient can keep track. tween providing evidence that they are not relevantly Previously opaque health outcomes are now avail- able. A diabetic patient who might have sincerely responsible for their health needs, and rationing access. believed they were keeping their blood sugars in con- This picture is deeply unattractive. Even if the citizen- trol could only check whether this was accurate by ry of a country supports an increase in responsibilisation regularly attending a medical appointment, which in healthcare for this reason, they may be unwilling to cannot occur every day (nor is it desirable that it accept such comprehensive surveillance. Such a system should do so). The ability to self-monitor on a daily would involve excessive capacity of government to dom- basis means that the patient now has more regular inate individuals (e.g. Pettit, 1997); an unwelcome in- access to relevant information. This is both intrin- crease in the political power of the state and its agents sically desirable and removes one kind of excuse against responsibility for health outcomes, since (e.g. Stahl, 2016); and would be excessively intrusive on patients cannot appeal to reasonable ignorance. citizens’ private lives (e.g. Lupton, 2012: 232, 239). Holding people responsible for their health is not of For those who wish to use responsibility as a criterion such urgency or necessity that the lack of a democratic for the allocation of healthcare (e.g. using responsibility mandate can be overruled. Even in the more limited as a tie-break when patients unavoidably compete for context of employer surveillance of their employees via resources), the idea of mHealth may seem attractive. mHealth apps, significant concerns have been raised al- Our health is affected by choices we make in every aspect ready, with Nissenbaum and Patterson (2016: 87) citing of our lives yet is subject to arguably more significant Stone’s (2002) objection to the establishment of ‘boun- influences from our social and physical environment. daryless workspaces’, and Selmi’s (2006: 1046) concern Some unhealthy behaviours are thus either easier to de- that ‘it is one thing to give an employer broad dominion tect, or more susceptible to being noticed for other rea- over its own workplace but quite another to extend that sons (e.g. because they are socially unpopular), than dominion wherever the employee goes’. others. It is unfair if some people are penalised for While a democratic mandate is necessary for sanc- choices that impact their health, while others make tioning such a programme of mass surveillance, it is choices with similar impacts but face no penalty. not sufficient. While people disagree about the moral Additionally, merely detecting a behaviour does not in- and political criteria for a justified surveillance pro- dicate its causes, e.g. whether patients engage in ‘un- gramme, there is general agreement that widespread sur- healthy’ behaviour due to limited options. veillance of the sort that covers an entire population In the absence of other evidence, the judgement about must meet a standard of proportionality (Macnish, whether a patient is responsible for their ill health must 2014; Rønn and Lippert-Rasmussen, 2020). Since even depend on the patient’s own reports. Even without pen- a well-intentioned surveillance programme, supported alties, patients are sometimes reluctant to be open with by a democratic majority, has the potential for signifi- doctors (Levy et al., 2018). Penalties will presumably cant abuse, the good that is acquired has to be significant. increase this tendency. Aside from undermining the evi- While some good might come out of comprehensive dence base for holding patients responsible, this will like- personal health surveillance, it seems unlikely to be suf- ly have a wider negative effect on the efficacy of ficient to justify such sweeping oversight, even on an treatment. undemanding understanding of what proportionality 272 DAVIES requires (e.g. that the benefits incurred must only equal system that risks demeaning patients, and turning ill the costs, as opposed to significantly outweighing them). health, which can already be a source of shame for vari- ous reasons, into a status of subjugation. Moreover, different health issues will require different Purely ‘Health’ Surveillance? kinds of surveillance. For instance, if the behaviour for which the patient is to be held responsible is taking their Supporters of responsibilisation might object that the daily medication, we might set up a pillbox that both above discussion is fanciful: nobody wants complete ac- prompts and records opening but does not surveil fur- quiescence to a surveillance state. The problem, they ther activity. Such cases sit at one end of a spectrum of might argue, is that such a state goes beyond health sur- intrusiveness and may seem to be a reasonable level of veillance to the surveillance of every aspect of our lives. surveillance. However, other behaviours seem to require This invites the question of what surveillance that almost constant surveillance. Consider a patient who is focused solely on health would look like. Carving out a held responsible for engaging in a particular level of ac- distinctive sphere of ‘health’ is difficult (Segall, 2007; tivity each day. We might begin with a pedometer, again Wilson, 2009) and goods which do not seem to be pri- a relatively unobtrusive form of surveillance. However, marily health-related may have greater effects on health while taking a greater number of steps is probably better than those behaviours and services which are commonly than a more sedentary lifestyle, merely taking a particu- seen as belonging in the ‘health’ sphere (Marmot, 2005). lar number of steps may not have a significant effect on One possible meaning of personal health surveillance is health; for instance, if those steps fail to get one’s heart stipulative: surveillance is health-related when it moni- rate up. An effective surveillance system might therefore tors a health condition, or a behaviour that has been need to target patients’ vital signs. Finally, a widespread established in that patient to contribute to a health con- adoption of activity surveillance may well lead to some— dition. For example, as someone with no diagnosed perhaps many—individuals ‘gaming’ the system. Those health conditions I can eat what I want, and it would who currently have a step counter on their phone, for be an unacceptable intrusion to monitor my health. If I instance, may know that the counter goes up not only if were diagnosed with diabetes, it would be a legitimately you walk or run somewhere, but also if you simply shake health-related form of surveillance to monitor my diet the phone. Insurers and governments might therefore and blood sugar levels. On this view, personal health decide that actual movement needs to be tracked as surveillance is a reactive rather than preventive measure. well as number of steps, taking advantage of the GPS This response must accept a partial retreat on one of capabilities that many phones have. In a climate of dis- the two problems that personal health surveillance was trust, we have therefore quickly moved from a relatively supposed to solve. We can abandon the ambition to hold low-level intrusion to a significant level of data people responsible for health-affecting choices they collection. make prior to entering the healthcare system. Alternatively, we must accept that due to a lack of sur- veillance, our evidence base for whether people are re- Surveillance as a Last Resort sponsible for their ill health will often be based on self- reporting and easily observable behaviours. In either In this section, I consider an even narrower scope for case, the issue of fairness re-emerges. personal health surveillance, focusing on patients who It is also not clear that even this reduced scope for state repeatedly fail to meet minimal standards of responsi- surveillance is proportionate, given the expected bene- bility for their health despite being capable of doing so. fits. A personal health surveillance system backing up a However, I also raise several problems with this pro- policy of responsibility-based penalties would require posal, both in this section and in the next. that personal data were readily available to a much wider The case for more limited personal health surveillance set of individuals than is currently normal. For instance, relies on the assumption that we are sometimes justified it would need to be transferred if the patient changed in giving additional burdens to those who are appropri- primary care doctor; it might need to be available in all ately responsible for their care, e.g. by denying them care; national hospitals. Such a system, even restricted solely setting their treatment as a lower priority relative to to personal health surveillance, routinely mistrusts others; or imposing (additional) financial costs beyond patients, treating them as though they are either inten- what is standardly imposed. Recall that this article tionally misleading the healthcare system or incapable of remains neutral on whether any of these are independ- handling their own health adequately. It is therefore a ently justified. Rather, the narrower version of personal PERSONAL HEALTH SURVEILLANCE 273 health surveillance considered in this section involves surveillance only when patients have already reached a using surveillance not as standard practice for all patients point where they have been deemed sufficiently respon- but a ‘Last Resort’ for patients who will otherwise legit- sible to face penalties, this problem remains. A policy of imately incur one of the above-mentioned penalties due responsibilisation will need an alternative way of evi- to their responsibility for their health needs. dencing patient responsibility, reintroducing the prob- The basic case for imposing penalties in such circum- lem of detection. Importantly, we cannot simply rely on stances is that when patients could reasonably be patients’ doctors to relay whether they have been making expected to make choices that would improve their reasonable efforts to remain healthy. While doctors health (i.e. when it would not involve significant burdens clearly have some advantage in judging what is best for in other areas of their lives, and when such choices are a patient, such a policy leaves far too much space for clearly explained and made available to them), but do personal and systemic biases. For instance, various find- not do so, they impose additional costs on the health care ings suggest that many medical professionals show bias service, and hence on some of those who use and fund in their treatment recommendations on the basis of sex that service. and gender (Hamberg, 2008), ethnicity (Hoffman et al., This case is highly controversial. Some deny that peo- 2016) and whether a patient is perceived as ‘fat’ (Fruh ple can be responsible in a way that justifies such penal- et al., 2016; Nath, 2019: 580). If medical professionals ties (Sharkey and Gillam, 2010; Pereboom, 2014; show bias in their treatment recommendations, there is Caruso, 2017). Others argue that whether or not this is clearly a risk that they will also show bias in making the conceptually possible, we are not able to detect such re- (arguably vaguer) judgement about what steps it is ‘rea- sponsibility with sufficient accuracy (Shelton and Balint, sonable’ for a patient to take, including misidentifying 1997; Glantz, 2007; Friesen, 2018). I remind readers, the burdens particular activities will have on a patient. A however, that my approach in this article is to criticise reasonable process of nomination for Last Resort would the use of personal health surveillance to enforce respon- therefore need to be more formalised and transparent sibility even if proponents of responsibilisation can over- than relying on doctors’ recommendations. It would also come these and other criticisms. need to be open to a process of appeal that was not (fi- The policy of Last Resort might seem to have several nancially or otherwise) inaccessible to patients. Aside advantages over the policies considered above. It does from anything else, this challenges the thought that a not place patients routinely under surveillance, and so is surveillance programme would be a cost-saving exercise. better placed with respect to proportionality. Since ac- cess to healthcare is a basic entitlement, there is no jus- tification for placing conditions on access for patients Penalties and Fairness who behave responsibly. However—despite being a Challenges of fairness arise whenever we select only some basic entitlement—patients might plausibly be thought of a relevant class of individuals for benefit or penalty. to have responsibilities as well as rights when it comes to Part of the answer to this challenge must be an admission accessing healthcare. Since the policy of Last Resort pla- that the problem of fairness arises in almost all attempts ces conditions on access only for those who have already to hold large groups of people to standards of behaviour. failed their responsibilities, an advocate might say, there In any widespread system, there will be false positives is justification available for surveillance that is not avail- (people who are held responsible despite not being so) able for more general policies. Precisely what the struc- and false negatives (people whose responsible behaviour ture of this justification is depends on a more general goes undetected). Nonetheless, when the system in ques- argument about why it is legitimate to hold patients sub- tion allocates something of such importance as health- stantively responsible for their health. But in focusing on care, this answer is not enough: it must also be clear that patients whose responsibility has already been reason- incidences of these types of mistake are kept sufficiently ably established, Last Resort is better placed than similar low. policies with a wider scope to meet this justificatory bur- This challenge can be mitigated if we can show that den. Moreover, the default approach is to trust patients, although not all of the relevant class of individuals were and to treat them as though they are entitled to the ser- vice they are using. correctly selected, the most significant cases were. For instance, the degree of justification for penalising people Yet this in itself raises a challenge. Recall that one pu- tative attraction of personal health surveillance was to who are responsible for their own poor health seems to overcome epistemic barriers to determining patient re- increase when they are more reckless, more unreasonable, sponsibility. If we are justified in implementing or had greater opportunity to avoid the relevant 274 DAVIES behaviour (where this involves both the range of alter- account for such considerable differences in natives available to a person, and the ease with which circumstances. those options can be chosen). A mechanism that picked Personal health surveillance as a way of enforcing re- out the most reckless, unreasonable and easily avoidable sponsibility also introduces new issues. Consider two cases for penalty might thus be fair even if it did not pick types of surveillance technologies, which correspond to out every case. the two functions of mHealth introduced in the above What would it take to focus on the most reckless or section, ‘Mobile Health, Surveillance and Two unreasonable cases? All else being equal, I assume that it Challenges for Responsibilisation’. Behaviour-tracking is more unreasonable for someone to engage in a health- technology would track users’ activities, assuming that affecting behaviour if they have been offered support in particular behaviours increase the likelihood of desired avoiding that behaviour; if they have been warned of the health outcomes. To make such targets enforceable with health effects of the behaviour; and if avoiding the be- penalties, we would need excellent evidence that they are haviour would have relatively few costs (Savulescu, both achievable and effective not only on average, but for 2018). While these are not the only ways of being unrea- the particular patient in question. The use of ‘generic’ sonably irresponsible, this does suggest that if a health- targets that fail to take account of a patient’s personal care system provided such support and information, it circumstances and health needs raise issues of fairness might then be acceptable to hold patients substantively where this leads to a patient being forced to adopt be- responsible. havioural targets that are not appropriate for them, or Importantly, however, such a policy must take ac- not achievable in their personal circumstances. count of the personal circumstances of a patient. One Consider, for example, the claim that many mobile of the most compelling objections to calls for responsi- phone-based pedometers do not accurately count steps bilisation is that they will tend to target those who are when the user is pushing a pram, a complaint that many already vulnerable in society, and/or for whom adapting users have posted about online. The internet is full of mandated behaviour changes will be particularly bur- ‘hacks’ to get around this problem, such as strapping the densome. It is essential to the fairness of holding monitor to one’s ankle, and so it is not insurmountable. patients responsible that the difficulty of adhering to par- But there are more general issues raised by this example: ticular habits and behaviours is recognised, and that it is 1. The manufacturers of the products did not consider acknowledged to vary depending on one’s circumstan- a form of exercise that is common for many people, ces. In addition, the reasonableness of failing to adopt namely taking their child out for a walk. certain healthy behaviours also varies depending on 2. The solution was not immediately obvious for many one’s circumstances, since health is not the only thing users, because many simply did not realise what the of value in our lives. Sometimes we rightly sacrifice problem was. health for other benefits, either for ourselves or others. 3. The activity in question is one that, while certainly Finally, recent work on the capacities required for moral undertaken by men, is still more likely to be under- responsibility has stressed the importance of seeing such taken by women (given, for instance, the common capacities—e.g. the capacity to respond to moral rea- disparities in social expectations about care, and legal sons—as ‘relational .. . partly constituted by both agent allowance of parental leave). There is therefore an and circumstance’ (Vargas, 2013: 206). A policy of unintentional gender bias in the way these products responsibilisation must recognise the considerable role track fitness. An uncritical adoption of similar tech- of social circumstances in determining people’s health, nology in personal health surveillance would trans- and the limits such circumstances place on a person’s late this bias to enforcement of responsibility. ability to pursue ‘reasonable’ behaviours. It is possible to theoretically imagine a healthcare system that held peo- On the other hand, outcome-tracking technologies ple substantively responsible in this sensitive way, and we offer more direct access to patients’ biological processes. thus cannot rule out the idea of personal health surveil- Outcome-tracking might enforce responsibility by get- lance on these grounds absolutely. Yet as a pragmatic ting patients to self-monitor their health and take appro- objection, worries about insufficient differentiation of priate action when readings hit particular levels. Such circumstance are significant. Particularly where personal technologies could be used for surveillance by reporting health surveillance is pursued as a primarily cost-cutting both the outcome-related data and whether the patient exercise, we have reason to be sceptical about whether it responds appropriately to readings which fall outside of is realistic to expect healthcare systems to properly their targets. PERSONAL HEALTH SURVEILLANCE 275 We should note the distinction between holding a under a scheme of Last Resort could have the typical patient responsible for responding appropriately to an right to decide precisely who has access to their health off-target reading and making them responsible for data. Yet there is still an onus on those who manage the bringing their biological readings back on target. While relevant data to ensure that it is stored securely: wider the former is still ultimately a form of behaviour that access is not universal access. Patients under personal patients can adopt, the latter will often be out of the health surveillance cannot be treated as if their privacy patient’s control; they might do everything they ought, does not matter. In addition, patients still have the right and yet still fail to achieve their target. Even with this to know who has access to their data, under what cir- distinction in place, there is a risk that holding patients cumstances, and why. As Mendelson and Wolf (2017:5) responsible for their biomedical states places too much note, there is even in non-punitive cases of the use of burden on them. Although such data can be translated mHealth an ‘asymmetry of power’ between those whose for the patient (‘If the reading is below 80, you need to data are accessed and those who access it. This asym- take your medication’), holding patients responsible this metry seems bound to be exacerbated when (the terms way may increase reluctance to seek medical help be- of) a patient’s access to care is on the line. cause they may feel expected to ‘fix’ problems them- One way to think about this is in terms of ownership. selves; this may be particularly acute where patients Cvrkel (2018: 517) raises the question of who owns the know they will be penalised for failing to behave data that is generated by users of mHealth apps. ‘appropriately’. Assuming that our default answer is that the user should Personal health surveillance faces further ethical have at least partial ownership rights, there is no reason issues. One such issue, which has been central to aca- to think that patients who are covered by the Last Resort demic and popular discussion of the ethics of mHealth approach should completely forego ownership of their quite generally, is privacy. Privacy can be understood in own data; rather, they simply have it limited in one various ways, though Avancha et al. (2012) suggest that way. Thus, even if patients who face the option of Last ‘control .. . is fundamental to privacy’, an idea echoed by Resort have a reduced claim of control over particular Kotz (2011: 1), who says that ‘health information privacy forms of data (i.e. the data directly relevant to the health is an individual’s right to control the acquisition, uses or condition for which Last Resort is imposed), this does disclosures of his or her identifiable data’. Privacy is an not mean that the treatment of patient data is straight- under-regulated element of mHealth, which as forward. Consider a case where the relevant mHealth Martinez-Martin and Kreitmair (2018) suggest, is ‘a data involves tracking a patient’s movement (e.g. to en- major concern when it comes to protecting the interests sure that they have done enough exercise). The most of users’, with ‘behavioural information .. . shared, straightforward way to do this would be through an stored and potentially sold to third parties’. Avancha app on the patient’s mobile phone. Since mobile phones et al. suggest several issues which are central to the regu- typically have one or more geolocation technologies, lation of privacy, including individual control over data; such tracking also raises the possibility of finding out openness and transparency of those accessing and con- other facts about the patient. As Carter et al. (2015) trolling data; and accountability for misuses of data, note, this information may include ‘where you live, while also outlining various privacy-protection frame- where your children go to school, whether you visit a works. Other, similar accounts can be found in therapist and if so how often, how often you visit drink- Mendelson and Wolf (2017); Jusob et al. (2017); and ing or gambling establishments, whether you arrive early Iwaya et al. (2018), while privacy as a concern for or late to work, whether you have participated in a pro- mHealth or health surveillance more generally is raised test or are associated with outlawed or terrorist organ- by Hendrix and Buck (2009: 482–499); Nissenbaum and izations and other habits or routines’. This particular Patterson (2016); Kreitmair et al. (2017); the WHO form of mHealth generates the possibility of patients (2017: 37); Cvrkel (2018: 517); Kreitmair (2019: 158); being pressured into providing information they are Wood et al. (2019: 471); Lee (2019: 324–6); and Ve ´ liz not happy to share, and which has no direct relevance (2020). to the justification for surveilling them in the first place. These broader concerns apply only partially to the case The justification for placing someone under personal of Last Resort. Patients who are subject to Last Resort health surveillance on the Last Resort model is not that surveillance must by necessity have less control over who they have behaved in a way that undermines any right to can access their data, and thus there is an inherent limit privacy or autonomy, but that their actions have specific to their privacy rights compared with the typical implications in one area of their life alone (see Sax, 2017, mHealth user. Thus, it is not true that a patient operating cited in Martani and Starke, 2019: 242). That someone 276 DAVIES has been placed under personal health surveillance as a degree to which personal health surveillance repre- Last Resort cannot be used to justify further, unrelated sents a cost-saving exercise, while the latter option is incursions on their rights. clearly unjust, since it excludes people from a pro- A further issue with the use of some forms of mHealth gram of public healthcare provision based solely on for surveillance is the question of whether patients are wealth. adequately equipped to respond appropriately to data. We can imagine, for instance, a patient who is tasked with increasing exercise in order to reduce their percent- Conclusions age of body fat. The patient duly completes the required My aim in this article has been to critically examine a amount of exercise, but for whatever reason sees very view which is conditional on the moral acceptability of little change in body fat percentage. If a patient has sim- sometimes holding patients responsible for their health- ply been left to deal with this information on their own, related behaviours. Without endorsing such a view, I they may easily become demoralised, reducing their suggested that an under-explored issue with this ap- short-term motivation to continue exercising proach is the problem of enforceability and detection, (Castelnuovo et al., 2014; Lucivero and Jongsma, 2018: i.e. how we know when a patient has been behaving in the 687). Patients under personal health surveillance may relevant ways. I suggested that, in the context of increas- therefore need to be provided with access to regular check-ins with doctors, medical counsellors or peer sup- ing use of mHealth technologies by employers and in- port networks (in person or through other forms of e- surance companies to engage in ‘personal health health), in order to put data into context, and to remind surveillance’ against employees and clients, there is real them that their targets are behavioural rather than out- potential for political states to begin exploring this op- come-focused. tion too. It should be clear, then, that personal health surveil- However, I argued that while the increasing use of lance, even as a last resort, cannot simply involve hand- mHealth technologies may appear to present a solution ing patients a device and some instructions. As Lucivero to several problems facing those who wish to use respon- and Jongsma (2018: 686) put it, ‘despite the hype around sibility as a rationing tool in healthcare, any plausible mHealth, there are still many uncertainties around the attempt to realise this faces significant ethical and prac- safety, reliability and accuracy of mHealth systems’ (see tical problems of its own. The problem of fairness, also Martani et al., 2019:5; Martani and Starke, 2019: related to detection, could only be solved by an un- 256). Patients’ ability to meet targets and to interpret acceptably broad scope for personal health surveillance. results, and their understanding of precisely what they Offering personal health surveillance as a last resort to have responsibility for, need to be carefully considered. patients who have already been judged suitably respon- In addition, we must be realistically confident that meas- sible for their health needs is a more plausible proposal, urements provided by personal health surveillance tech- but still faces a range of ethical challenges. Thus, while nologies are accurate (DiStefano and Schmidt, 2016: mHealth technologies may appear to promise to ‘close 215). And even if patients allow their data to be accessed the loop’ of enforceability when it comes to the respon- by a wider range of individuals than normal, the process sibilisation of healthcare, in practice it faces considerable of data storage and sharing must be both secure and challenges. transparent. A final, practical problem with using mHealth tech- nologies for surveillance is the ‘digital divide’ (Wood Notes et al., 2019: 472; Mello and Wang, 2020: 951). While 1. As far as I am aware, this is the only other work to some forms of mHealth involve giving patients speci- consider this possibility directly; Martani and Starke alised devices, others make use of existing devices offer distinct criticisms of this proposal, which are such as smart phones. Yet some patients (Paldan complementary to mine and with which I concur. et al., 2018; Raber et al., 2019) do not have access 2. See Stemplowska (2009) for discussion. to these technologies. If some personal health surveil- 3. My thanks to an anonymous referee for pointing out lance relies on existing device ownership, we would this gap in the original draft. face a choice between providing patients with the rele- vant technologies or excluding them from the oppor- 4. See Nath (2019) on the burdens some face in trying to tunity to opt for personal health surveillance instead lose weight. of exclusion. The former option reduces further the 5. See also McGeer (2015). PERSONAL HEALTH SURVEILLANCE 277 Buyx, A. M. (2008). Personal Responsibility for Health as a Acknowledgements Rationing Criterion: Why We Don’t Like It and Why Thanks to an anonymous reviewer for some extremely Maybe We Should. Journal of Medical Ethics, 34, 871–874. helpful comments which improved the focus of this art- Cappelen, A. W. and Norheim, O. F. (2005). icle considerably. Responsibility in Health Care: A Liberal Egalitarian Approach. Journal of Medical Ethics, 31, 476–480. Carter, A., Liddle, J., Hall, W., and Chenery, H. (2015). Funding Mobile Phones in Research and Treatment: Ethical Guidelines and Future Directions. JMIR Mhealth This research was funded by the Wellcome Trust, grant Uhealth, 3, e95. WT104848/Z/14/Z. Carter, P., Laurie, G. T., and Dixon-Woods, M. (2015). The Social Licence for Research: Why Care.data Ran into Trouble. Journal of Medical Ethics, 41, 404–409. Conflict of Interest Caruso, G. D. (2017). Public Health and Safety: The Social None declared. Determinants of Health and Criminal Behavior. London, UK: ResearchLinks Books. Castelnuovo, G., Manzoni, G. M., Pietrabissa, G., Corti, References S., Giusti, E. M., Molinari, E., and Simpson, S. (2014). Albertsen, A. (2020). Personal Responsibility in Health Obesity and Outpatient Rehabilitation Using Mobile and Health Care: Luck Egalitarianism as a Plausible Technologies: The Potential mHealth Approach. and Flexible Approach to Health. Political Research Frontiers in Psychology, 5, 559. Quarterly, 73, 583–595. Cavoukian, A., Fisher, A., Killen, S., and Hoffman, D. A. Albertsen, A. and Knight, C. (2015). A Framework for (2010). Remote Home Health Care Technologies: Luck Egalitarianism in Health and Healthcare. How to Ensure Privacy? Build It in: Privacy by Journal of Medical Ethics, 41, 165–169. Design. Identity in the Information Society, 3, 363–378. Anderson, E. (1999). What Is the Point of Equality? Chatzipavlou, I. A., Christoforidou, S. A., and Ethics, 109, 287–337. Vlachopoulou, M. (2016). A Recommended Arneson, R. (1999). Egalitarianism and Responsibility. Guideline for the Development of mHealth Apps. Journal of Ethics, 3, 225–247. mHealth, 2, 21–28. Avancha, S., Baxi, A., and Kotz, D. (2012). Privacy in Chow, C. K., Ariyarathna, N., Islam, S. M. S., Mobile Technology for Personal Healthcare. ACM Thiagalingam, A., and Redfern, J. (2016). mHealth Computing Surveys, 45, 1–54. in Cardiovascular Health Care. Heart, Lung and Ayo, N. (2012). Understanding Health Promotion in a Circulation, 25, 802–807. Neoliberal Climate and the Making of Health Cohn, A. M., Hunter-Reel, D., Hagman, B. T., and Conscious Citizens. Critical Public Health,22, 99–105. Mitchell, J. (2011). Promoting Behavior Change Barlin, S. (2018). Strap on the Fitbit: John Hancock to Sell from Alcohol Use through Mobile Technology: The Only Interactive Life Insurance. Reuters (2018, Future of Ecological Momentary Assessment. September 19), available from: https://www.reuters. Alcoholism: Clinical and Experimental Research, 35, com/article/us-manulife-financi-john-hancock-life 2209–2215. ins/strap-on-the-fitbit-john-hancock-to-sell-only- Connelly, J., Kirk, A., Masthoff, J., and MacRury, S. interactive-life-insurance-idUSKCN1LZ1WL (2013). The Use of Technology to Promote Physical [accessed 27 August 2020]. Activity in Type 2 Diabetes Management: A Behar, J. A., Oster, J., Vos, M. D., and Clifford, G. D. (2019). Systematic Review. Diabetic Medicine, 30, 1420–1432. Wearables and mHealth in Mental Health and Cvrkel, T. (2018). The Ethics of mHealth: Moving Neurological Disorders. Physiological Measurement, 40, Forward. Journal of Dentistry, Digital Technologies in Oral & Dental Research, 74, S15–S20. Botsman, R. (2017). Big Data meets Big Brother as China DiStefano, M. and Schmidt, H. (2016). mHealth for Moves to Rate Its Citizens. Wired (2017, October 21), Tuberculosis Treatment Adherence: A Framework available from: https://www.wired.co.uk/article/chin to Guide Ethical Planning, Implementation, and ese-government-social-credit-score-privacy-inva Evaluation. Global Health: Science and Practice, 4, sion [accessed 27 August 2020]. 211–221. 278 DAVIES Dworkin, R. (2000). Sovereign Virtue. Cambridge, MA: Sciences of the United States of America, 113, Harvard University Press. 4296–4301. Fairchild, A. L. (2003). Dealing with Humpty Dumpty: Istepanian, R. S. H. and Al-anzi, T. M. (2018). m-Health Research, Practice, and the Ethics of Public Health Interventions for Diabetes Remote Monitoring and Surveillance. Journal of Law, Medicine & Ethics, 31, Self Management: Clinical and Compliance Issues. 615–623. mHealth, 4, 4–3. Fairchild, A. L., Bayer, R., and Colgrove, J. (2008). Iwaya, L. H., Fischer-Hu ¨ bner, S., Ahlfeldt, R., and Privacy, Democracy and the Politics of Disease Martucci, L. A. (2018). mHealth: A Privacy Threat Analysis for Public Health Surveillance Systems. In Surveillance. Public Health Ethics, 1, 30–38. Fleurbaey, M. (1995). Equal Opportunity or Equal Social 2018 IEEE 31st International Symposium on Outcome. Economics and Philosophy, 11, 25–55. Computer-Based Medical Systems. pp. 42–47. Friesen, P. (2018). Personal Responsibility within Health https://ieeexplore.ieee.org/xpl/conhome/8410976/ Policy: Unethical and Ineffective. Journal of Medical proceeding (accessed 4 May 2021). Ethics, 44, 53–58. Jusob, F. R., George, C., and Mapp, G. (2017). Exploring Fruh, S. M., Nadglowski, J., Hall, H. R., Davis, S. L., the Need for a Suitable Privacy Framework for Crook, E. D., and Zlomke, K. (2016). Obesity mHealth When Managing Chronic Diseases. Journal Stigma and Bias. The Journal for Nurse Practitioners, of Reliable Intelligent Environments, 3, 243–256. 12, 425–432. Klingler, C., Silva, D. S., Schuermann, C., Reis, A. A., Gilbert, G. L., Degeling, C., and Johnson, J. (2019). Saxena, A., and Strech, D. (2017). Ethical Issues in Communicable Disease Surveillance Ethics in the Public Health Surveillance, a Systematic Qualitative Age of Big Data and New Technology. Asian Review. BMC Public Health, 17, 295–307. Bioethics Review, 11, 173–187. Kotz, D. 2011. A threat taxonomy for mHealth privacy. Glantz, L. (2007). Should Smokers Be Refused Surgery? 2011 Third International Conference on BMJ, 334, 21–21. Communication Systems and Networks Hamberg, K. (2008). Gender Bias in Medicine. Women’s (COMSNETS 2011). 1–6, doi: 10.1109/ Health (Lond), 4, 237–243. COMSNETS.2011.5716518. Hancock, M. (n.d.). My Vision for Prevention—Public Kreitmair, K. V., Cho, M. K., and Magnus, D. C. (2017). Health Matters, available from: https://publichealth Consent and Engagement, Security, and Authentic matters.blog.gov.uk/2018/11/05/matt-hancock-my- Living Using Wearable and Mobile Health vision-for-prevention/ [accessed 29 January 2020]. Technology. Nature Biotechnology, 35, 617–620. Harrison, V., Proudfoot, J., Wee, P. P., Parker, G., Kreitmair, K. V. (2019). Dimensions of Ethical Pavlovic, D. H., and Manicavasagar, V. (2011). Direct-to-Consumer Neurotechnologies. AJOB Mobile Mental Health: Review of the Emerging Neuroscience, 10, 152–166. Field and Proof of Concept Study. Journal of Mental Le Grand, J. (2013). Individual Responsibility, Health, Health, 20, 509–524. and Health Care. In N., Eyal, S. A., Hurst, O. F., Hendrix, A. and Buck, J. (2009). Employer-Sponsored Norheim, and D., Wikler (eds), Inequalities in Wellness Programs: Should Your Employer Be the Health: Concepts, Measures and Ethics. Oxford: Boss of More than Your Work. Southwestern Law OUP, pp. 299–306. Review, 38, 465–502. Lee, L. M. (2019). Public Health Surveillance: Ethical Henkel, M., Heck, T., and Go ¨ retz, J. (2018). Rewarding Considerations. In A. C., Mastroianni, J. P., Kahn, Fitness Tracking—The Communication and and N. E., Kass (eds), The Oxford Handbook of Promotion of Health Insurers’ Bonus Programs and Public Health Ethics. Oxford: OUP, pp. 319–330. the Use of Self-Tracking Data. In G., Meiselwitz (ed.), Lee, L. M., Heilig, C. M., and White, A. (2012). Ethical Social Computing and Social Media. Technologies and Justification for Conducting Public Health Analytics, Lecture Notes in Computer Science. Cham: Surveillance without Patient Consent. American Springer International Publishing, pp. 28–49. Journal of Public Health, 102, 38–44. Hoffman, K. M., Trawalter, S., Axt, J. R., and Oliver, M. Levy, A. G., Scherer, A. M., Zikmund-Fisher, B. J., Larkin, N. (2016). Racial Bias in Pain Assessment and K., Barnes, G. D., and Fagerlin, A. (2018). Prevalence Treatment Recommendations, and False Beliefs of and Factors Associated with Patient Nondisclosure about Biological Differences between Blacks and of Medically Relevant Information to Clinicians. Whites. Proceedings of the National Academy of JAMA Network Open, 1, e185293. PERSONAL HEALTH SURVEILLANCE 279 Levy, N. (2018). Taking Responsibility for Health in an Nath, R. (2019). The Injustice of Fat Stigma. Bioethics, 33, 577–590. Epistemically Polluted Environment. Theoretical Medicine and Bioethics, 39, 123–141. Nissenbaum, H. and Patterson, H. (2016). Biosensing in Context: Health Privacy in a Connected World. In D., Lippert-Rasmussen, K. (2016). Luck Egalitarianism. Nafus (ed.), Quantified: Biosensing Technologies in London: Bloomsbury. Everyday Life. Cambridge, MA: MIT Press, Lucivero, F. and Jongsma, K. R. (2018). A Mobile pp.79–100. Revolution for Healthcare? Setting the Agenda for O’Neill, S. (2018). As Insurers Offer Discounts for Fitness Bioethics. Journal of Medical Ethics, 44, 685–689. Trackers, Wearers Should Step with Caution [WWW Lupton, D. (2012). m-Health and Health Promotion: Document]. NPR.org, available from: https://www. The Digital Cyborg and Surveillance Society. Social npr.org/sections/health-shots/2018/11/19/ Theory & Health, 10, 229–244. 668266197/as-insurers-offer-discounts-for-fitness- Lupton, D. (2013). Quantifying the Body: Monitoring trackers-wearers-should-step-with-cautio [accessed and Measuring Health in the Age of mHealth 29 January 2020]. Technologies. Critical Public Health, 23, 393–403. Paldan, K., Sauer, H., and Wagner, N.-F. (2018). Lupton, D. (2016). Digitized Health Promotion: Risk Promoting Inequality? Self-Monitoring Applications and Responsibility for Health and Illness in the Web and the Problem of Social Justice. AI & Society,doi: 2.0 Era. In J. E., Davies and A. M., Gonzalez (eds), To 10.1007/s00146-018-0835-7. Fix or to Heal: Patient Care, Public Health and the Pantelopoulos, A. and Bourbakis, N. G. (2010). Limits of Biomedicine. New York: NYU Press, pp. Prognosis—A Wearable Health-Monitoring System 52–76. for People at Risk: Methodology and Modeling. Lupton, D. (2019). ‘It’s Made Me a Lot More Aware’: A IEEE Transactions on Information Technology in New Materialist Analysis of Health Self-Tracking. Biomedicine, 14, 613–621. Media International Australia, 171, 66–14. Pereboom, D. (2014). Free Will, Agency, and Meaning in Macnish, K. (2014). Just Surveillance? Towards a Life. Oxford: Oxford University Press. Normative Theory of Surveillance. Surveillance and Petrini, C. (2013). Ethics in Public Health Surveillance. Society, 12, 142–153. Annali Dell’Instituto Superiore di Sanita `, 49, 347–353. Marmot, M. (2005). Social Determinants of Health Petrini, C. and Ricciardi, G. (2015). Ethical Issues in Inequalities. The Lancet, 365, 1099–1104. Public Health Surveillance: Drawing Inspiration Martani, A., Shaw, D., and Simone, E. B. (2019). Stay Fit from Ethical Frameworks. Annali Dell’Instituto or Get Bit—Ethical Issues in Sharing Health Data with Superiore di Sanita `, 51, 270–276. Insurers’ Apps. Swiss Medical Weekly, 149, 1–8. Pettit, P. (1997). Republicanism: A Theory of Freedom and Martani, A. and Starke, G. (2019). Personal Responsibility Government. Oxford: Oxford University Press. for Health: The Impact of Digitisation. Journal of Pillutla, V., Maslen, H., and Savulescu, J. (2018). Medical Law and Ethics, 7, 241–258. Rationing Elective Surgery for Smokers and Obese Martinez-Martin, N. and Kreitmair, K. (2018). Ethical Patients: Responsibility or Prognosis? BMC Medical Issues for Direct-to-Consumer Digital Psychotherapy Ethics, 19, 1–10. Apps: Addressing Accountability, Data Protection, and Raber, I., McCarthy, C. P., and Yeh, R. W. (2019). Health Consent. JMIR Mental Health, 5, e32. Insurance and Mobile Health Devices: Opportunities McGeer, V. (2015). Building a Better Theory of and Concerns. JAMA, 321, 1767–1768. Responsibility. Philosophical Studies, 172, 2635–2649. Roemer, J. (1993). A Pragmatic Theory of Responsibility Mello, M. M. and Wang, C. J. (2020). Ethics and for the Egalitarian Planner. Philosophy & Public Governance for Digital Disease Surveillance. Science, Affairs, 22, 146–166. 368, 951–954. Rønn, K. V. and Lippert-Rasmussen, K. (2020). Out of Mendelson, D. and Wolf, G. (2017). Health Privacy and Proportion? On Surveillance and the Proportionality Confidentiality. In I., Freckelton and K., Peterson Requirement. Ethical Theory and Moral Practice, 23, (eds), Tensions and Traumas in Health Law. Sydney: 181–199. Federation Press, pp. 266–282. Rubel, A. (2012). Justifying Public Health Surveillance: Minkler, M. (1999). Personal Responsibility for Health? Basic Interests, Unreasonable Exercise, and Privacy. A Review of the Arguments and the Evidence at Kennedy Institute of Ethics Journal, 22, 1–33. Century’s End. Health Education & Behavior, 26, Ruckenstein, M. and Schu ¨ ll, N. D. (2017). The Datafication 121–140. of Health. Annual Review of Anthropology, 46, 261–278. 280 DAVIES Samerski, S. (2018). Individuals on Alert: Digital Quantified Self, and the Participatory Biocitizen. Epidemiology and the Individualization of Journal of Personalized Medicine, 2, 93–118. Surveillance. Life Sciences, Society and Policy, 14, 1–11. Ter Meulen, R. and Maarse, H. (2009). Increasing Savulescu, J. (2018). Golden Opportunity, Reasonable Individual Responsibility in Dutch Health Care: Is Risk and Personal Responsibility for Health. Journal Solidarity Losing Ground? (SSRN Scholarly Paper of Medical Ethics, 44, 59–61. No. ID 1447168). Rochester, NY: Social Science Sax, A. (2017). Eigenverantwortung. Schweiz Arzteztg, Research Network. 98, 174. Topol, E. J. (2015). The Future of Medicine Is in Schmidt, H. (2007). Personal Responsibility for Your Smartphone. Wall Street Journal. https://on Health—Developments under the German line.wsj.com/articles/the-future-of-medicine-is-in- Healthcare Reform 2007. European Journal of Health your-smartphone-1420828632?reflink¼desktopweb Law, 14, 241–250. share_permalink (accessed 4 May 2021). Schmidt, H. (2009). Personal Responsibility in the NHS Vallentyne, P. (2002). Brute Luck, Option Luck, and Constitution and the Social Determinants of Health Equalityof Initial Opportunities. Ethics, 112, Approach: Competitive or Complementary? Health 529–557. 10.1086/339275. Econ Policy Law, 4, 129–138. Vargas, M. (2013). Building Better Beings: A Theory of Segall, S. (2007). Is Health Care (Still) Special? Journal of Moral Responsibility. Oxford: OUP. Political Philosophy, 15, 342–361. Ve ´ liz, C. (2020). Privacy during the Pandemic and Segall, S. (2010). Health, Luck and Justice. Princeton, NJ: Beyond. The Philosophers’ Magazine, 90, 111–113. Princeton University Press. Venkatapuram, S. (2011). Health Justice: An Argument Selmi, M. (2006). Privacy for the Working Class: Public from the Capabilities Approach. Cambridge: Polity. Work and Private Lives. Louisiana Law Review, 66, Voigt, K. (2007). The Harshness Objection: Is Luck 1046–1056. Egalitarianism Too Harsh on the Victims of Option Sharkey, K. and Gillam, L. (2010). Should Patients with Luck? Ethical Theory and Moral Practice, 10, 389–407. Self-Inflicted Illness Receive Lower Priority in Access Weber, S., Strommenger, D., Kertzscher, U., and Affeld, to Healthcare Resources? Mapping out the Debate. K. (2012). Continuous Blood Pressure Measurement with Journal of Medical Ethics, 36, 661–665. Ultrasound. Biomedical Engineering/Biomedizinische Sharon, T. (2017). Self-Tracking for Health and the Technik, 57, 407–410. doi:10.1515/bmt-2012-4108. Quantified Self: Re-Articulating Autonomy, WHO (2017). WHO Guidelines on Ethical Issues in Public Solidarity, and Authenticity in an Age of Health Surveillance. Geneva: WHO. Personalized Healthcare. Philosophy & Technology, WHO j eHealth. (n.d.) WHO, available from: http:// 30, 93–121. www.who.int/ehealth/en/ [accessed 29 January 2020]. Shelton, W. and Balint, J. A. (1997). Fair Treatment of WHO Global Observatory for eHealth and World Health Alcoholic Patients in the Context of Liver Organization (2011). MHealth: New Horizons for Transplantation. Alcoholism: Clinical and Health through Mobile Technologies. Geneva: World Experimental Research, 21, 93–100. Health Organization. Shemkus, S. (2015). Fitness Trackers Are Popular among Wiederhold, B. K. (2012). Self-Tracking: Better Medicine Insurers and Employers—But Is Your Data Safe? The through Pattern Recognition. Cyberpsychology, Guardian. theguardian.com, available from: https:// Behavior, and Social Networking, 15, 235–236. www.theguardian.com/lifeandstyle/2015/apr/17/fit Wikler, D. (2002). Personal and Social Responsibility ness-trackers-wearables-insurance-employees-jobs- for Health. Ethics & International Affairs, 16, health-data [accessed 29 January 2020]. 47–55. Stahl, T. (2016). Indiscriminate Mass Surveillance and Wilson, J. (2009). Not So Special after All? Daniels and the Public Sphere. Ethics and Information the Social Determinants of Health. Journal of Medical Technology, 18, 33–39. Ethics, 35, 3–6. Stemplowska, Z. (2009). Making Justice Sensitive to Wood, C. S., Thomas, M. R., Budd, J., Mashamba- Responsibility. Political Studies, 57, 237–259. Thompson, T. P., Herbst, K., Pillay, D., Peeling, R. Stone, K. V. W. (2002). Employee Representation in the W., Johnson, A. M., McKendry, R. A. and Stevens, Boundaryless Workplace. Chicago-Kent Law Review, 77, 773–819. M. M. (2019). Taking Connected Mobile-Health Diagnostics of Infectious Diseases to the Field. Swan, M. (2012). Health 2050: The Realization of Personalized Medicine through Crowdsourcing, the Nature Review, 566, 467–474. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Public Health Ethics Oxford University Press

‘Personal Health Surveillance’: The Use of mHealth in Healthcare Responsibilisation

Davies, Ben
Public Health Ethics , Volume 14 (3): 13 – May 16, 2021
Download PDF
13 pages

Loading next page...
 
/lp/oxford-university-press/personal-health-surveillance-the-use-of-mhealth-in-healthcare-abqhq4EOSP

References (121)

Publisher
Oxford University Press
Copyright
© The Author(s) 2021. Published by Oxford University Press.
ISSN
1754-9973
eISSN
1754-9981
DOI
10.1093/phe/phab013
Publisher site
See Article on Publisher Site

Abstract

PUBLIC HEALTH ETHICS VOLUME 14 ISSUE 3 2021 268–280 268 • • • ‘Personal Health Surveillance’: The Use of mHealth in Healthcare Responsibilisation Ben Davies *, Uehiro Centre for Practical Ethics, University of Oxford *Corresponding author: Ben Davies, Uehiro Centre for Practical Ethics, University of Oxford, Suite 8, Littlegate House, Oxford OX1 1PT, UK. Email: benjamin.davies@philosophy.ox.ac.uk There is an ongoing increase in the use of mobile health (mHealth) technologies that patients can use to monitor health-related outcomes and behaviours. While the dominant narrative around mHealth focuses on patient empowerment, there is potential for mHealth to fit into a growing push for patients to take personal responsibility for their health. I call the first of these uses ‘medical monitoring’, and the second ‘personal health surveillance’. After outlining two problems which the use of mHealth might seem to enable us to overcome— fairness of burdens and reliance on self-reporting—I note that these problems would only really be solved by unacceptably comprehensive forms of personal health surveillance which applies to all of us at all times. A more plausible model is to use personal health surveillance as a last resort for patients who would otherwise independently qualify for responsibility-based penalties. However, I note that there are still a number of ethical and practical problems that such a policy would need to overcome. The prospects of mHealth enabling a fair, genuinely cost-saving policy of patient responsibility are slim. and communications technologies for health (WHO j Mobile Health, Surveillance and eHealth, n.d.). mHealth includes applications on mobile Two Challenges for phones as well as more direct monitoring of patient health indicators such as wearable monitors and at- Responsibilisation home testing kits whose results can be transmitted by Technological advances are providing increasing abil- patients to medical professionals (DiStefano and ity to monitor health outcomes and health-related Schmidt, 2016). behaviours outside traditional clinical settings and mHealth has the potential to facilitate two functions. relationships. Patients can self-administer tests for First, it may allow us to monitor our bodily processes blood sugars (Cvrkel, 2018; Istepanian and Al-anzi, which, while affected by behaviour, are not under direct 2018); oxygen saturation (Pantelopoulos and control (Lupton, 2013, 2019). If a patient knows broadly Bourbakis, 2010); blood pressure (Weber et al., which behaviours affect the relevant processes, they can 2012); heart rate (Chow et al., 2016: 804); mood attempt to indirectly moderate their health. mHealth (Harrison et al., 2011); and neurological function may thus relocate routine health monitoring from expli- (Behar et al., 2019). We can monitor health-related citly medical settings to the home, workplace and wider behaviours more easily (Sharon, 2017): wearable tech- world, which Swan (2012) describes as an ‘institutional nologies can help monitor activity levels and diet recasting’ of healthcare (see also Carter et al., 2015). This (Connelly et al., 2013); alcohol consumption (Cohn can be seen positively as ‘shifting (health management) et al., 2011); and medication use (Cavoukian et al., into the hands of empowered patients’ (an ideal 2010; Martani and Starke, 2019: 251), as well as pro- reported, critically, by Ruckenstein and Schu ¨ ll, 2017: viding mental health services (Martinez-Martin and 262), liberating them from time-costly medical appoint- Kreitmair, 2018). Collectively, these technologies are ments (Topol, 2015), or more negatively as an over- known as mobile health (mHealth) (WHO Global medicalization of previously more carefree spaces. Observatory for eHealth and World Health Second, mHealth may facilitate monitoring of behav- Organization, 2011). mHealth is generally defined as iours that affect our health, but which are difficult to a subcategory of ‘e-health’ (Chatzipavlou et al., 2016: track unaided, and about which we are wont to self-de- 1), which encompasses the general use of information ceive. I tell my doctor that I stick to the UK government’s doi:10.1093/phe/phab013 Advance Access publication on 16 May 2021 V C The Author(s) 2021. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. PERSONAL HEALTH SURVEILLANCE 269 guideline of 14 units of alcohol per week. Perhaps I be- measures which would increase costs or affect access to lieve this to be true; judging my unit intake requires a care for those who are suitably responsible for their ill relatively complicated calculation. The relative complex- health (Schmidt, 2007, 2009; Hancock, n.d.; Ter Meulen ity of keeping tabs on an enjoyable activity facilitates my and Maarse, 2009), with Lupton (2016: 155) noting how reluctance to confront the truth about my alcohol con- Anglophone countries have retained a focus on personal sumption. An app that calculates the units for me might responsibility that developed in the mid-20th century, help me follow the guidelines. with a ‘renewed emphasis on lifestyle change’. The pos- We might therefore see mHealth as a way for patients sibility of responsibility-centred rationing is not an ab- to take back control over their health from institutional stract possibility, with Pillutla et al. (2018: 1) noting medicine, saving time and effort by reducing unneces- recent local proposals in the UK to ‘restrict elective sur- sary contact with medical professionals. Not coinciden- gery for patients who either smoke or are obese’. tally, optimists might see mHealth as promising public When health costs are borne partly or wholly by the spending savings at no cost to public health. Finally, the state, it is not a significant leap to think that where in- data generated by the use of mHealth has the potential to surance companies and private employers lead, public feed into public health research. health systems could be tempted to follow. At one ex- Yet mHealth could be integrated into healthcare in treme is the widespread use of data for citizen tracking another way, using technological monitoring to increase currently operating in China (Botsman, 2017). Yet even the role of individual responsibility not only as a method if such a comprehensive system of surveillance seems of empowering patients, but also to hold them account- unlikely in more democratic states, the use of mHealth able as users of public resources. For clarity between may seem to ‘close the loophole of practical enforceabil- these two uses of mHealth devices I use the term ‘medical ity’ when it comes to judgements of personal responsi- monitoring’ for more standard current uses of monitor- bility (Martani and Starke, 2019: 241). ing devices, e.g. allowing patients to access data about As mHealth grows in both private and public usage, themselves. When it comes to using mHealth to enforce particularly in the context of increased political and so- responsibility, I will use the term ‘personal health sur- cial focus on personal responsibility, it is important to veillance’. The same device or app can therefore be used consider the implications of using mHealth to enforce in either monitoring or surveillance. As far as I am aware, responsibilisation. In doing so, I remain neutral for the this distinction has not previously been explicitly dis- sake of this article about whether it is legitimate to hold cussed in work on mHealth. people responsible for their health by imposing add- The use of mHealth by some insurance companies itional burdens on their access to care when their poor (Shemkus, 2015; Lupton, 2016: 164; Henkel et al., health is suitably due to their own free choices. These 2018; O’Neill, 2018; Martani et al., 2019) offers insight burdens may range from the severe (denial of care) to the into the possible institutional uses of mHealth. While mild (some additional co-payments). most companies currently use mHealth technology by There are a number of arguments for and against the offering positive rewards to those who achieve particular idea of healthcare responsibilisation. Some ‘luck egali- targets, Raber et al. (2019: 1767–8) note that these sys- tarians’, for instance, consider inequalities that individ- tems ‘could be used by insurers in the future to penalize uals cannot avoid to be unjust, but do not necessarily users’. Similarly, where private employers are respon- condemn inequalities which reflect exercises of respon- sible for part of employees’ health insurance costs, there sibility (e.g. Arneson, 1999; Dworkin, 2000: 77–8; have been attempts by some to mandate employees’ Vallentyne, 2002; Lippert-Rasmussen, 2016); while health-related behaviours through the use of mHealth not all luck egalitarians apply this view directly to real- (Lupton, 2016; Nissenbaum and Patterson, 2016: 84, 88; world healthcare, some do (Roemer, 1993; Cappelen and Barlin, 2018). Writing just over a decade ago, Hendrix Norheim, 2005; Segall, 2010; Le Grand, 2013: 303; and Buck (2009: 466) describe how ‘employers have Albertsen and Knight, 2015; Albertsen, 2020), though begun to implement increasingly aggressive wellness generally not in anything like the simplistic manner programs that provide incentives to employees who imagined by critics. Others have argued from alternative meet certain health standards, while creating disincen- perspectives that responsibility may be a reasonable part tives for those who fail to meet the standards’. of any healthcare system (Buyx, 2008; Savulescu, 2018). My central focus in this article is on the potential for Others regard the luck egalitarian stance as excessively the state to follow suit. The idea of responsibility is a ‘harsh’ (Fleurbaey, 1995; Anderson, 1999; Voigt, 2007; familiar theme in publicly funded health, with many Venkatapuram, 2011, 198); see the practical aim of hold- jurisdictions either implementing or considering ing people responsible as inappropriately focused on a 270 DAVIES small section of our choices (Minkler, 1999; Wikler, Some of the issues raised by infectious disease surveil- 2002; Sharkey and Gillam, 2010; Friesen, 2018); criticise lance are ones which also affect the use of mHealth in the a reframing of social problems as individual ones (Ayo, context of personal responsibility. A particularly obvious 2012; Lupton, 2012); or doubt our ability to appropri- example is privacy, e.g. Lee (2019: 323). However, while ately take responsibility for our health (Levy, 2018). Yet there may be cases where a person is suitably responsible it is of independent value to demonstrate, as I hope to do, for having an infectious disease which is the subject of that even if we grant the legitimacy of holding people traditional infectious disease surveillance, the justifica- responsible in some cases for their poor health, it is tion for surveillance in this case—preventing the spread very difficult to justify the use of mHealth technologies of disease—is very different than the justification in cases for enforcing this. of responsibilisation. Indeed, the primary justification I will shortly outline an initial ‘optimistic’ case for for infectious disease surveillance offered in the litera- how mHealth might indeed ‘close the loop’ of en- ture is a broadly consequentialist one, presupposing a forceability for personal responsibility (e.g. Swan, specifically health-related benefit which could not be 2012; Wiederhold, 2012; Topol, 2015) before going achieved in other ways, and which outweighs potential on to raise a number of ethical and practical chal- harms (Fairchild et al., 2008; Lee et al., 2012: 38–42; lenges. Before doing so, however, it is worth com- Petrini and Ricciardi, 2015: 273; WHO, 2017). Even menting briefly on an area of public health policy this is not universally accepted—for instance, Rubel which I will not discuss in detail, but which has (2012: 2) rejects justifications that reveal to an aggregate tangential relevance to this issue. This is the issue good, arguing that surveillance can be justified only if it of traditional public health surveillance for the pur- protects ‘basic interests’—but in any case does not ob- poses of controlling infectious disease. This issue will viously apply to surveillance in the service of responsibi- be familiar to many because of the (at the time of lisation. Rather, the most obvious justification for writing) ongoing COVID-19 pandemic. Infectious responsibility-based surveillance would be desert-based, disease surveillance obviously predates this crisis i.e. that those who are suitably responsible for their ill and is primarily justified by the potential for expo- health ought to bear the burdens of it (financial or other- nential escalation and significant harm (Gilbert et al., wise). Of course, one might also hope that a focus on 2019: 176). personal responsibility will improve public health by Fairchild et al. (2008: 30) outline a traditional under- disincentivising certain behaviours. Yet the central jus- standing of infectious disease surveillance, as ‘the on- tification for infectious disease surveillance seems inimi- going, name-based reporting of cases of disease to state cal to the idea of responsibilisation, with the WHO and local health departments’. However, others (e.g. (2017: 46) reinforcing the idea that relevant data should Samerski, 2018:1; Mello and Wang, 2020: 951) note not be used, nor given to those who would use it, to ‘take the growing influence of mHealth in potentially more action against’ individuals. proactive—and invasive—surveillance, including in the Moreover, whereas infectious disease surveillance is context of COVID-19 (Ve ´ liz, 2020). A number of typically focused on aggregate effects and on guiding authors stress the centrality of surveillance to public public policy, personal health surveillance by necessity health efforts, as well as the potential risks of failure to will involve a focus on individuals. Thus, while there surveil (e.g. Fairchild et al., 2008: 30; Petrini, 2013; are some clear parallels between existing infectious WHO, 2017: 10, 17; Gilbert et al., 2019: 176; Lee, 2019: disease surveillance and ‘personal health surveillance’, 320; Wood et al., 2019) with Mello and Wang (2020: the latter is a clearly distinct (potential) phenomenon 951) suggesting that ‘the question is not whether to use that could not easily draw on the existence of the new data sources—such as cellphones, wearables, video former for justification. Nonetheless, both types of surveillance, social media, internet searches and news surveillance may fall under the broad sphere of ‘public and crowd-sourced symptom self-reports—but how’. health’. Whereas infectious disease surveillance is On the other hand, there are clear ethical issues involved more obviously concerned with public health, namely in infectious disease surveillance, especially when it is the targeting of public health policy, personal health opposed by many of those who are sufferers of the par- surveillance may be concerned with a number of ticular condition in question, as has been the case with issues that are related to public health, including pre- HIV/AIDS in some jurisdictions (Fairchild, 2003; vention of disease by disincentivising irresponsible be- Fairchild et al., 2008: 32–34; Klingler et al., 2017: 1–2; haviour, and the appropriate allocation of public Lee, 2019). health resources. PERSONAL HEALTH SURVEILLANCE 271 I return now to what I termed the ‘optimistic’ case for Surveillance might seem to mitigate both problems. Of course, only the most intrusive surveillance state the use of mHealth technologies in healthcare responsibilisation: could hope to fully eradicate the problem of detectability (and even this is doubtful). Responsibility for poor The state or an appropriate medical authority mon- health, and various factors that might justify unhealthy itors whether patients are behaving in appropriate behaviour, typically comes before any interaction with ways (e.g. taking moderate exercise) given their healthcare services. A comprehensively non-discrimin- health needs or achieving certain health targets atory system seemingly needs to surveil all individuals. (e.g. reductions in cholesterol) without patients needing frequent, direct medical contact. Since Unhealthy behaviours do not occur only in public, nor patients have direct, quantifiable access to their can they always be detected after the fact. So, individuals health outcomes on a daily basis, they take greater would need to be surveilled at all times for the most responsibility for their health. Behavioural targets comprehensive—and, thus, one might think, fairest— are more precise: for instance, rather than recom- information about responsibility. For instance, Martani mending that a patient take ‘regular, moderate ex- and Starke (2019: 252) consider the possibility of health ercise’, doctors can recommend more personalised providers forcing a choice to prospective patients be- targets, knowing that the patient can keep track. tween providing evidence that they are not relevantly Previously opaque health outcomes are now avail- able. A diabetic patient who might have sincerely responsible for their health needs, and rationing access. believed they were keeping their blood sugars in con- This picture is deeply unattractive. Even if the citizen- trol could only check whether this was accurate by ry of a country supports an increase in responsibilisation regularly attending a medical appointment, which in healthcare for this reason, they may be unwilling to cannot occur every day (nor is it desirable that it accept such comprehensive surveillance. Such a system should do so). The ability to self-monitor on a daily would involve excessive capacity of government to dom- basis means that the patient now has more regular inate individuals (e.g. Pettit, 1997); an unwelcome in- access to relevant information. This is both intrin- crease in the political power of the state and its agents sically desirable and removes one kind of excuse against responsibility for health outcomes, since (e.g. Stahl, 2016); and would be excessively intrusive on patients cannot appeal to reasonable ignorance. citizens’ private lives (e.g. Lupton, 2012: 232, 239). Holding people responsible for their health is not of For those who wish to use responsibility as a criterion such urgency or necessity that the lack of a democratic for the allocation of healthcare (e.g. using responsibility mandate can be overruled. Even in the more limited as a tie-break when patients unavoidably compete for context of employer surveillance of their employees via resources), the idea of mHealth may seem attractive. mHealth apps, significant concerns have been raised al- Our health is affected by choices we make in every aspect ready, with Nissenbaum and Patterson (2016: 87) citing of our lives yet is subject to arguably more significant Stone’s (2002) objection to the establishment of ‘boun- influences from our social and physical environment. daryless workspaces’, and Selmi’s (2006: 1046) concern Some unhealthy behaviours are thus either easier to de- that ‘it is one thing to give an employer broad dominion tect, or more susceptible to being noticed for other rea- over its own workplace but quite another to extend that sons (e.g. because they are socially unpopular), than dominion wherever the employee goes’. others. It is unfair if some people are penalised for While a democratic mandate is necessary for sanc- choices that impact their health, while others make tioning such a programme of mass surveillance, it is choices with similar impacts but face no penalty. not sufficient. While people disagree about the moral Additionally, merely detecting a behaviour does not in- and political criteria for a justified surveillance pro- dicate its causes, e.g. whether patients engage in ‘un- gramme, there is general agreement that widespread sur- healthy’ behaviour due to limited options. veillance of the sort that covers an entire population In the absence of other evidence, the judgement about must meet a standard of proportionality (Macnish, whether a patient is responsible for their ill health must 2014; Rønn and Lippert-Rasmussen, 2020). Since even depend on the patient’s own reports. Even without pen- a well-intentioned surveillance programme, supported alties, patients are sometimes reluctant to be open with by a democratic majority, has the potential for signifi- doctors (Levy et al., 2018). Penalties will presumably cant abuse, the good that is acquired has to be significant. increase this tendency. Aside from undermining the evi- While some good might come out of comprehensive dence base for holding patients responsible, this will like- personal health surveillance, it seems unlikely to be suf- ly have a wider negative effect on the efficacy of ficient to justify such sweeping oversight, even on an treatment. undemanding understanding of what proportionality 272 DAVIES requires (e.g. that the benefits incurred must only equal system that risks demeaning patients, and turning ill the costs, as opposed to significantly outweighing them). health, which can already be a source of shame for vari- ous reasons, into a status of subjugation. Moreover, different health issues will require different Purely ‘Health’ Surveillance? kinds of surveillance. For instance, if the behaviour for which the patient is to be held responsible is taking their Supporters of responsibilisation might object that the daily medication, we might set up a pillbox that both above discussion is fanciful: nobody wants complete ac- prompts and records opening but does not surveil fur- quiescence to a surveillance state. The problem, they ther activity. Such cases sit at one end of a spectrum of might argue, is that such a state goes beyond health sur- intrusiveness and may seem to be a reasonable level of veillance to the surveillance of every aspect of our lives. surveillance. However, other behaviours seem to require This invites the question of what surveillance that almost constant surveillance. Consider a patient who is focused solely on health would look like. Carving out a held responsible for engaging in a particular level of ac- distinctive sphere of ‘health’ is difficult (Segall, 2007; tivity each day. We might begin with a pedometer, again Wilson, 2009) and goods which do not seem to be pri- a relatively unobtrusive form of surveillance. However, marily health-related may have greater effects on health while taking a greater number of steps is probably better than those behaviours and services which are commonly than a more sedentary lifestyle, merely taking a particu- seen as belonging in the ‘health’ sphere (Marmot, 2005). lar number of steps may not have a significant effect on One possible meaning of personal health surveillance is health; for instance, if those steps fail to get one’s heart stipulative: surveillance is health-related when it moni- rate up. An effective surveillance system might therefore tors a health condition, or a behaviour that has been need to target patients’ vital signs. Finally, a widespread established in that patient to contribute to a health con- adoption of activity surveillance may well lead to some— dition. For example, as someone with no diagnosed perhaps many—individuals ‘gaming’ the system. Those health conditions I can eat what I want, and it would who currently have a step counter on their phone, for be an unacceptable intrusion to monitor my health. If I instance, may know that the counter goes up not only if were diagnosed with diabetes, it would be a legitimately you walk or run somewhere, but also if you simply shake health-related form of surveillance to monitor my diet the phone. Insurers and governments might therefore and blood sugar levels. On this view, personal health decide that actual movement needs to be tracked as surveillance is a reactive rather than preventive measure. well as number of steps, taking advantage of the GPS This response must accept a partial retreat on one of capabilities that many phones have. In a climate of dis- the two problems that personal health surveillance was trust, we have therefore quickly moved from a relatively supposed to solve. We can abandon the ambition to hold low-level intrusion to a significant level of data people responsible for health-affecting choices they collection. make prior to entering the healthcare system. Alternatively, we must accept that due to a lack of sur- veillance, our evidence base for whether people are re- Surveillance as a Last Resort sponsible for their ill health will often be based on self- reporting and easily observable behaviours. In either In this section, I consider an even narrower scope for case, the issue of fairness re-emerges. personal health surveillance, focusing on patients who It is also not clear that even this reduced scope for state repeatedly fail to meet minimal standards of responsi- surveillance is proportionate, given the expected bene- bility for their health despite being capable of doing so. fits. A personal health surveillance system backing up a However, I also raise several problems with this pro- policy of responsibility-based penalties would require posal, both in this section and in the next. that personal data were readily available to a much wider The case for more limited personal health surveillance set of individuals than is currently normal. For instance, relies on the assumption that we are sometimes justified it would need to be transferred if the patient changed in giving additional burdens to those who are appropri- primary care doctor; it might need to be available in all ately responsible for their care, e.g. by denying them care; national hospitals. Such a system, even restricted solely setting their treatment as a lower priority relative to to personal health surveillance, routinely mistrusts others; or imposing (additional) financial costs beyond patients, treating them as though they are either inten- what is standardly imposed. Recall that this article tionally misleading the healthcare system or incapable of remains neutral on whether any of these are independ- handling their own health adequately. It is therefore a ently justified. Rather, the narrower version of personal PERSONAL HEALTH SURVEILLANCE 273 health surveillance considered in this section involves surveillance only when patients have already reached a using surveillance not as standard practice for all patients point where they have been deemed sufficiently respon- but a ‘Last Resort’ for patients who will otherwise legit- sible to face penalties, this problem remains. A policy of imately incur one of the above-mentioned penalties due responsibilisation will need an alternative way of evi- to their responsibility for their health needs. dencing patient responsibility, reintroducing the prob- The basic case for imposing penalties in such circum- lem of detection. Importantly, we cannot simply rely on stances is that when patients could reasonably be patients’ doctors to relay whether they have been making expected to make choices that would improve their reasonable efforts to remain healthy. While doctors health (i.e. when it would not involve significant burdens clearly have some advantage in judging what is best for in other areas of their lives, and when such choices are a patient, such a policy leaves far too much space for clearly explained and made available to them), but do personal and systemic biases. For instance, various find- not do so, they impose additional costs on the health care ings suggest that many medical professionals show bias service, and hence on some of those who use and fund in their treatment recommendations on the basis of sex that service. and gender (Hamberg, 2008), ethnicity (Hoffman et al., This case is highly controversial. Some deny that peo- 2016) and whether a patient is perceived as ‘fat’ (Fruh ple can be responsible in a way that justifies such penal- et al., 2016; Nath, 2019: 580). If medical professionals ties (Sharkey and Gillam, 2010; Pereboom, 2014; show bias in their treatment recommendations, there is Caruso, 2017). Others argue that whether or not this is clearly a risk that they will also show bias in making the conceptually possible, we are not able to detect such re- (arguably vaguer) judgement about what steps it is ‘rea- sponsibility with sufficient accuracy (Shelton and Balint, sonable’ for a patient to take, including misidentifying 1997; Glantz, 2007; Friesen, 2018). I remind readers, the burdens particular activities will have on a patient. A however, that my approach in this article is to criticise reasonable process of nomination for Last Resort would the use of personal health surveillance to enforce respon- therefore need to be more formalised and transparent sibility even if proponents of responsibilisation can over- than relying on doctors’ recommendations. It would also come these and other criticisms. need to be open to a process of appeal that was not (fi- The policy of Last Resort might seem to have several nancially or otherwise) inaccessible to patients. Aside advantages over the policies considered above. It does from anything else, this challenges the thought that a not place patients routinely under surveillance, and so is surveillance programme would be a cost-saving exercise. better placed with respect to proportionality. Since ac- cess to healthcare is a basic entitlement, there is no jus- tification for placing conditions on access for patients Penalties and Fairness who behave responsibly. However—despite being a Challenges of fairness arise whenever we select only some basic entitlement—patients might plausibly be thought of a relevant class of individuals for benefit or penalty. to have responsibilities as well as rights when it comes to Part of the answer to this challenge must be an admission accessing healthcare. Since the policy of Last Resort pla- that the problem of fairness arises in almost all attempts ces conditions on access only for those who have already to hold large groups of people to standards of behaviour. failed their responsibilities, an advocate might say, there In any widespread system, there will be false positives is justification available for surveillance that is not avail- (people who are held responsible despite not being so) able for more general policies. Precisely what the struc- and false negatives (people whose responsible behaviour ture of this justification is depends on a more general goes undetected). Nonetheless, when the system in ques- argument about why it is legitimate to hold patients sub- tion allocates something of such importance as health- stantively responsible for their health. But in focusing on care, this answer is not enough: it must also be clear that patients whose responsibility has already been reason- incidences of these types of mistake are kept sufficiently ably established, Last Resort is better placed than similar low. policies with a wider scope to meet this justificatory bur- This challenge can be mitigated if we can show that den. Moreover, the default approach is to trust patients, although not all of the relevant class of individuals were and to treat them as though they are entitled to the ser- vice they are using. correctly selected, the most significant cases were. For instance, the degree of justification for penalising people Yet this in itself raises a challenge. Recall that one pu- tative attraction of personal health surveillance was to who are responsible for their own poor health seems to overcome epistemic barriers to determining patient re- increase when they are more reckless, more unreasonable, sponsibility. If we are justified in implementing or had greater opportunity to avoid the relevant 274 DAVIES behaviour (where this involves both the range of alter- account for such considerable differences in natives available to a person, and the ease with which circumstances. those options can be chosen). A mechanism that picked Personal health surveillance as a way of enforcing re- out the most reckless, unreasonable and easily avoidable sponsibility also introduces new issues. Consider two cases for penalty might thus be fair even if it did not pick types of surveillance technologies, which correspond to out every case. the two functions of mHealth introduced in the above What would it take to focus on the most reckless or section, ‘Mobile Health, Surveillance and Two unreasonable cases? All else being equal, I assume that it Challenges for Responsibilisation’. Behaviour-tracking is more unreasonable for someone to engage in a health- technology would track users’ activities, assuming that affecting behaviour if they have been offered support in particular behaviours increase the likelihood of desired avoiding that behaviour; if they have been warned of the health outcomes. To make such targets enforceable with health effects of the behaviour; and if avoiding the be- penalties, we would need excellent evidence that they are haviour would have relatively few costs (Savulescu, both achievable and effective not only on average, but for 2018). While these are not the only ways of being unrea- the particular patient in question. The use of ‘generic’ sonably irresponsible, this does suggest that if a health- targets that fail to take account of a patient’s personal care system provided such support and information, it circumstances and health needs raise issues of fairness might then be acceptable to hold patients substantively where this leads to a patient being forced to adopt be- responsible. havioural targets that are not appropriate for them, or Importantly, however, such a policy must take ac- not achievable in their personal circumstances. count of the personal circumstances of a patient. One Consider, for example, the claim that many mobile of the most compelling objections to calls for responsi- phone-based pedometers do not accurately count steps bilisation is that they will tend to target those who are when the user is pushing a pram, a complaint that many already vulnerable in society, and/or for whom adapting users have posted about online. The internet is full of mandated behaviour changes will be particularly bur- ‘hacks’ to get around this problem, such as strapping the densome. It is essential to the fairness of holding monitor to one’s ankle, and so it is not insurmountable. patients responsible that the difficulty of adhering to par- But there are more general issues raised by this example: ticular habits and behaviours is recognised, and that it is 1. The manufacturers of the products did not consider acknowledged to vary depending on one’s circumstan- a form of exercise that is common for many people, ces. In addition, the reasonableness of failing to adopt namely taking their child out for a walk. certain healthy behaviours also varies depending on 2. The solution was not immediately obvious for many one’s circumstances, since health is not the only thing users, because many simply did not realise what the of value in our lives. Sometimes we rightly sacrifice problem was. health for other benefits, either for ourselves or others. 3. The activity in question is one that, while certainly Finally, recent work on the capacities required for moral undertaken by men, is still more likely to be under- responsibility has stressed the importance of seeing such taken by women (given, for instance, the common capacities—e.g. the capacity to respond to moral rea- disparities in social expectations about care, and legal sons—as ‘relational .. . partly constituted by both agent allowance of parental leave). There is therefore an and circumstance’ (Vargas, 2013: 206). A policy of unintentional gender bias in the way these products responsibilisation must recognise the considerable role track fitness. An uncritical adoption of similar tech- of social circumstances in determining people’s health, nology in personal health surveillance would trans- and the limits such circumstances place on a person’s late this bias to enforcement of responsibility. ability to pursue ‘reasonable’ behaviours. It is possible to theoretically imagine a healthcare system that held peo- On the other hand, outcome-tracking technologies ple substantively responsible in this sensitive way, and we offer more direct access to patients’ biological processes. thus cannot rule out the idea of personal health surveil- Outcome-tracking might enforce responsibility by get- lance on these grounds absolutely. Yet as a pragmatic ting patients to self-monitor their health and take appro- objection, worries about insufficient differentiation of priate action when readings hit particular levels. Such circumstance are significant. Particularly where personal technologies could be used for surveillance by reporting health surveillance is pursued as a primarily cost-cutting both the outcome-related data and whether the patient exercise, we have reason to be sceptical about whether it responds appropriately to readings which fall outside of is realistic to expect healthcare systems to properly their targets. PERSONAL HEALTH SURVEILLANCE 275 We should note the distinction between holding a under a scheme of Last Resort could have the typical patient responsible for responding appropriately to an right to decide precisely who has access to their health off-target reading and making them responsible for data. Yet there is still an onus on those who manage the bringing their biological readings back on target. While relevant data to ensure that it is stored securely: wider the former is still ultimately a form of behaviour that access is not universal access. Patients under personal patients can adopt, the latter will often be out of the health surveillance cannot be treated as if their privacy patient’s control; they might do everything they ought, does not matter. In addition, patients still have the right and yet still fail to achieve their target. Even with this to know who has access to their data, under what cir- distinction in place, there is a risk that holding patients cumstances, and why. As Mendelson and Wolf (2017:5) responsible for their biomedical states places too much note, there is even in non-punitive cases of the use of burden on them. Although such data can be translated mHealth an ‘asymmetry of power’ between those whose for the patient (‘If the reading is below 80, you need to data are accessed and those who access it. This asym- take your medication’), holding patients responsible this metry seems bound to be exacerbated when (the terms way may increase reluctance to seek medical help be- of) a patient’s access to care is on the line. cause they may feel expected to ‘fix’ problems them- One way to think about this is in terms of ownership. selves; this may be particularly acute where patients Cvrkel (2018: 517) raises the question of who owns the know they will be penalised for failing to behave data that is generated by users of mHealth apps. ‘appropriately’. Assuming that our default answer is that the user should Personal health surveillance faces further ethical have at least partial ownership rights, there is no reason issues. One such issue, which has been central to aca- to think that patients who are covered by the Last Resort demic and popular discussion of the ethics of mHealth approach should completely forego ownership of their quite generally, is privacy. Privacy can be understood in own data; rather, they simply have it limited in one various ways, though Avancha et al. (2012) suggest that way. Thus, even if patients who face the option of Last ‘control .. . is fundamental to privacy’, an idea echoed by Resort have a reduced claim of control over particular Kotz (2011: 1), who says that ‘health information privacy forms of data (i.e. the data directly relevant to the health is an individual’s right to control the acquisition, uses or condition for which Last Resort is imposed), this does disclosures of his or her identifiable data’. Privacy is an not mean that the treatment of patient data is straight- under-regulated element of mHealth, which as forward. Consider a case where the relevant mHealth Martinez-Martin and Kreitmair (2018) suggest, is ‘a data involves tracking a patient’s movement (e.g. to en- major concern when it comes to protecting the interests sure that they have done enough exercise). The most of users’, with ‘behavioural information .. . shared, straightforward way to do this would be through an stored and potentially sold to third parties’. Avancha app on the patient’s mobile phone. Since mobile phones et al. suggest several issues which are central to the regu- typically have one or more geolocation technologies, lation of privacy, including individual control over data; such tracking also raises the possibility of finding out openness and transparency of those accessing and con- other facts about the patient. As Carter et al. (2015) trolling data; and accountability for misuses of data, note, this information may include ‘where you live, while also outlining various privacy-protection frame- where your children go to school, whether you visit a works. Other, similar accounts can be found in therapist and if so how often, how often you visit drink- Mendelson and Wolf (2017); Jusob et al. (2017); and ing or gambling establishments, whether you arrive early Iwaya et al. (2018), while privacy as a concern for or late to work, whether you have participated in a pro- mHealth or health surveillance more generally is raised test or are associated with outlawed or terrorist organ- by Hendrix and Buck (2009: 482–499); Nissenbaum and izations and other habits or routines’. This particular Patterson (2016); Kreitmair et al. (2017); the WHO form of mHealth generates the possibility of patients (2017: 37); Cvrkel (2018: 517); Kreitmair (2019: 158); being pressured into providing information they are Wood et al. (2019: 471); Lee (2019: 324–6); and Ve ´ liz not happy to share, and which has no direct relevance (2020). to the justification for surveilling them in the first place. These broader concerns apply only partially to the case The justification for placing someone under personal of Last Resort. Patients who are subject to Last Resort health surveillance on the Last Resort model is not that surveillance must by necessity have less control over who they have behaved in a way that undermines any right to can access their data, and thus there is an inherent limit privacy or autonomy, but that their actions have specific to their privacy rights compared with the typical implications in one area of their life alone (see Sax, 2017, mHealth user. Thus, it is not true that a patient operating cited in Martani and Starke, 2019: 242). That someone 276 DAVIES has been placed under personal health surveillance as a degree to which personal health surveillance repre- Last Resort cannot be used to justify further, unrelated sents a cost-saving exercise, while the latter option is incursions on their rights. clearly unjust, since it excludes people from a pro- A further issue with the use of some forms of mHealth gram of public healthcare provision based solely on for surveillance is the question of whether patients are wealth. adequately equipped to respond appropriately to data. We can imagine, for instance, a patient who is tasked with increasing exercise in order to reduce their percent- Conclusions age of body fat. The patient duly completes the required My aim in this article has been to critically examine a amount of exercise, but for whatever reason sees very view which is conditional on the moral acceptability of little change in body fat percentage. If a patient has sim- sometimes holding patients responsible for their health- ply been left to deal with this information on their own, related behaviours. Without endorsing such a view, I they may easily become demoralised, reducing their suggested that an under-explored issue with this ap- short-term motivation to continue exercising proach is the problem of enforceability and detection, (Castelnuovo et al., 2014; Lucivero and Jongsma, 2018: i.e. how we know when a patient has been behaving in the 687). Patients under personal health surveillance may relevant ways. I suggested that, in the context of increas- therefore need to be provided with access to regular check-ins with doctors, medical counsellors or peer sup- ing use of mHealth technologies by employers and in- port networks (in person or through other forms of e- surance companies to engage in ‘personal health health), in order to put data into context, and to remind surveillance’ against employees and clients, there is real them that their targets are behavioural rather than out- potential for political states to begin exploring this op- come-focused. tion too. It should be clear, then, that personal health surveil- However, I argued that while the increasing use of lance, even as a last resort, cannot simply involve hand- mHealth technologies may appear to present a solution ing patients a device and some instructions. As Lucivero to several problems facing those who wish to use respon- and Jongsma (2018: 686) put it, ‘despite the hype around sibility as a rationing tool in healthcare, any plausible mHealth, there are still many uncertainties around the attempt to realise this faces significant ethical and prac- safety, reliability and accuracy of mHealth systems’ (see tical problems of its own. The problem of fairness, also Martani et al., 2019:5; Martani and Starke, 2019: related to detection, could only be solved by an un- 256). Patients’ ability to meet targets and to interpret acceptably broad scope for personal health surveillance. results, and their understanding of precisely what they Offering personal health surveillance as a last resort to have responsibility for, need to be carefully considered. patients who have already been judged suitably respon- In addition, we must be realistically confident that meas- sible for their health needs is a more plausible proposal, urements provided by personal health surveillance tech- but still faces a range of ethical challenges. Thus, while nologies are accurate (DiStefano and Schmidt, 2016: mHealth technologies may appear to promise to ‘close 215). And even if patients allow their data to be accessed the loop’ of enforceability when it comes to the respon- by a wider range of individuals than normal, the process sibilisation of healthcare, in practice it faces considerable of data storage and sharing must be both secure and challenges. transparent. A final, practical problem with using mHealth tech- nologies for surveillance is the ‘digital divide’ (Wood Notes et al., 2019: 472; Mello and Wang, 2020: 951). While 1. As far as I am aware, this is the only other work to some forms of mHealth involve giving patients speci- consider this possibility directly; Martani and Starke alised devices, others make use of existing devices offer distinct criticisms of this proposal, which are such as smart phones. Yet some patients (Paldan complementary to mine and with which I concur. et al., 2018; Raber et al., 2019) do not have access 2. See Stemplowska (2009) for discussion. to these technologies. If some personal health surveil- 3. My thanks to an anonymous referee for pointing out lance relies on existing device ownership, we would this gap in the original draft. face a choice between providing patients with the rele- vant technologies or excluding them from the oppor- 4. See Nath (2019) on the burdens some face in trying to tunity to opt for personal health surveillance instead lose weight. of exclusion. The former option reduces further the 5. See also McGeer (2015). PERSONAL HEALTH SURVEILLANCE 277 Buyx, A. M. (2008). Personal Responsibility for Health as a Acknowledgements Rationing Criterion: Why We Don’t Like It and Why Thanks to an anonymous reviewer for some extremely Maybe We Should. Journal of Medical Ethics, 34, 871–874. helpful comments which improved the focus of this art- Cappelen, A. W. and Norheim, O. F. (2005). icle considerably. Responsibility in Health Care: A Liberal Egalitarian Approach. Journal of Medical Ethics, 31, 476–480. Carter, A., Liddle, J., Hall, W., and Chenery, H. (2015). Funding Mobile Phones in Research and Treatment: Ethical Guidelines and Future Directions. JMIR Mhealth This research was funded by the Wellcome Trust, grant Uhealth, 3, e95. WT104848/Z/14/Z. Carter, P., Laurie, G. T., and Dixon-Woods, M. (2015). The Social Licence for Research: Why Care.data Ran into Trouble. Journal of Medical Ethics, 41, 404–409. Conflict of Interest Caruso, G. D. (2017). Public Health and Safety: The Social None declared. Determinants of Health and Criminal Behavior. London, UK: ResearchLinks Books. Castelnuovo, G., Manzoni, G. M., Pietrabissa, G., Corti, References S., Giusti, E. M., Molinari, E., and Simpson, S. (2014). Albertsen, A. (2020). Personal Responsibility in Health Obesity and Outpatient Rehabilitation Using Mobile and Health Care: Luck Egalitarianism as a Plausible Technologies: The Potential mHealth Approach. and Flexible Approach to Health. Political Research Frontiers in Psychology, 5, 559. Quarterly, 73, 583–595. Cavoukian, A., Fisher, A., Killen, S., and Hoffman, D. A. Albertsen, A. and Knight, C. (2015). A Framework for (2010). Remote Home Health Care Technologies: Luck Egalitarianism in Health and Healthcare. How to Ensure Privacy? Build It in: Privacy by Journal of Medical Ethics, 41, 165–169. Design. Identity in the Information Society, 3, 363–378. Anderson, E. (1999). What Is the Point of Equality? Chatzipavlou, I. A., Christoforidou, S. A., and Ethics, 109, 287–337. Vlachopoulou, M. (2016). A Recommended Arneson, R. (1999). Egalitarianism and Responsibility. Guideline for the Development of mHealth Apps. Journal of Ethics, 3, 225–247. mHealth, 2, 21–28. Avancha, S., Baxi, A., and Kotz, D. (2012). Privacy in Chow, C. K., Ariyarathna, N., Islam, S. M. S., Mobile Technology for Personal Healthcare. ACM Thiagalingam, A., and Redfern, J. (2016). mHealth Computing Surveys, 45, 1–54. in Cardiovascular Health Care. Heart, Lung and Ayo, N. (2012). Understanding Health Promotion in a Circulation, 25, 802–807. Neoliberal Climate and the Making of Health Cohn, A. M., Hunter-Reel, D., Hagman, B. T., and Conscious Citizens. Critical Public Health,22, 99–105. Mitchell, J. (2011). Promoting Behavior Change Barlin, S. (2018). Strap on the Fitbit: John Hancock to Sell from Alcohol Use through Mobile Technology: The Only Interactive Life Insurance. Reuters (2018, Future of Ecological Momentary Assessment. September 19), available from: https://www.reuters. Alcoholism: Clinical and Experimental Research, 35, com/article/us-manulife-financi-john-hancock-life 2209–2215. ins/strap-on-the-fitbit-john-hancock-to-sell-only- Connelly, J., Kirk, A., Masthoff, J., and MacRury, S. interactive-life-insurance-idUSKCN1LZ1WL (2013). The Use of Technology to Promote Physical [accessed 27 August 2020]. Activity in Type 2 Diabetes Management: A Behar, J. A., Oster, J., Vos, M. D., and Clifford, G. D. (2019). Systematic Review. Diabetic Medicine, 30, 1420–1432. Wearables and mHealth in Mental Health and Cvrkel, T. (2018). The Ethics of mHealth: Moving Neurological Disorders. Physiological Measurement, 40, Forward. Journal of Dentistry, Digital Technologies in Oral & Dental Research, 74, S15–S20. Botsman, R. (2017). Big Data meets Big Brother as China DiStefano, M. and Schmidt, H. (2016). mHealth for Moves to Rate Its Citizens. Wired (2017, October 21), Tuberculosis Treatment Adherence: A Framework available from: https://www.wired.co.uk/article/chin to Guide Ethical Planning, Implementation, and ese-government-social-credit-score-privacy-inva Evaluation. Global Health: Science and Practice, 4, sion [accessed 27 August 2020]. 211–221. 278 DAVIES Dworkin, R. (2000). Sovereign Virtue. Cambridge, MA: Sciences of the United States of America, 113, Harvard University Press. 4296–4301. Fairchild, A. L. (2003). Dealing with Humpty Dumpty: Istepanian, R. S. H. and Al-anzi, T. M. (2018). m-Health Research, Practice, and the Ethics of Public Health Interventions for Diabetes Remote Monitoring and Surveillance. Journal of Law, Medicine & Ethics, 31, Self Management: Clinical and Compliance Issues. 615–623. mHealth, 4, 4–3. Fairchild, A. L., Bayer, R., and Colgrove, J. (2008). Iwaya, L. H., Fischer-Hu ¨ bner, S., Ahlfeldt, R., and Privacy, Democracy and the Politics of Disease Martucci, L. A. (2018). mHealth: A Privacy Threat Analysis for Public Health Surveillance Systems. In Surveillance. Public Health Ethics, 1, 30–38. Fleurbaey, M. (1995). Equal Opportunity or Equal Social 2018 IEEE 31st International Symposium on Outcome. Economics and Philosophy, 11, 25–55. Computer-Based Medical Systems. pp. 42–47. Friesen, P. (2018). Personal Responsibility within Health https://ieeexplore.ieee.org/xpl/conhome/8410976/ Policy: Unethical and Ineffective. Journal of Medical proceeding (accessed 4 May 2021). Ethics, 44, 53–58. Jusob, F. R., George, C., and Mapp, G. (2017). Exploring Fruh, S. M., Nadglowski, J., Hall, H. R., Davis, S. L., the Need for a Suitable Privacy Framework for Crook, E. D., and Zlomke, K. (2016). Obesity mHealth When Managing Chronic Diseases. Journal Stigma and Bias. The Journal for Nurse Practitioners, of Reliable Intelligent Environments, 3, 243–256. 12, 425–432. Klingler, C., Silva, D. S., Schuermann, C., Reis, A. A., Gilbert, G. L., Degeling, C., and Johnson, J. (2019). Saxena, A., and Strech, D. (2017). Ethical Issues in Communicable Disease Surveillance Ethics in the Public Health Surveillance, a Systematic Qualitative Age of Big Data and New Technology. Asian Review. BMC Public Health, 17, 295–307. Bioethics Review, 11, 173–187. Kotz, D. 2011. A threat taxonomy for mHealth privacy. Glantz, L. (2007). Should Smokers Be Refused Surgery? 2011 Third International Conference on BMJ, 334, 21–21. Communication Systems and Networks Hamberg, K. (2008). Gender Bias in Medicine. Women’s (COMSNETS 2011). 1–6, doi: 10.1109/ Health (Lond), 4, 237–243. COMSNETS.2011.5716518. Hancock, M. (n.d.). My Vision for Prevention—Public Kreitmair, K. V., Cho, M. K., and Magnus, D. C. (2017). Health Matters, available from: https://publichealth Consent and Engagement, Security, and Authentic matters.blog.gov.uk/2018/11/05/matt-hancock-my- Living Using Wearable and Mobile Health vision-for-prevention/ [accessed 29 January 2020]. Technology. Nature Biotechnology, 35, 617–620. Harrison, V., Proudfoot, J., Wee, P. P., Parker, G., Kreitmair, K. V. (2019). Dimensions of Ethical Pavlovic, D. H., and Manicavasagar, V. (2011). Direct-to-Consumer Neurotechnologies. AJOB Mobile Mental Health: Review of the Emerging Neuroscience, 10, 152–166. Field and Proof of Concept Study. Journal of Mental Le Grand, J. (2013). Individual Responsibility, Health, Health, 20, 509–524. and Health Care. In N., Eyal, S. A., Hurst, O. F., Hendrix, A. and Buck, J. (2009). Employer-Sponsored Norheim, and D., Wikler (eds), Inequalities in Wellness Programs: Should Your Employer Be the Health: Concepts, Measures and Ethics. Oxford: Boss of More than Your Work. Southwestern Law OUP, pp. 299–306. Review, 38, 465–502. Lee, L. M. (2019). Public Health Surveillance: Ethical Henkel, M., Heck, T., and Go ¨ retz, J. (2018). Rewarding Considerations. In A. C., Mastroianni, J. P., Kahn, Fitness Tracking—The Communication and and N. E., Kass (eds), The Oxford Handbook of Promotion of Health Insurers’ Bonus Programs and Public Health Ethics. Oxford: OUP, pp. 319–330. the Use of Self-Tracking Data. In G., Meiselwitz (ed.), Lee, L. M., Heilig, C. M., and White, A. (2012). Ethical Social Computing and Social Media. Technologies and Justification for Conducting Public Health Analytics, Lecture Notes in Computer Science. Cham: Surveillance without Patient Consent. American Springer International Publishing, pp. 28–49. Journal of Public Health, 102, 38–44. Hoffman, K. M., Trawalter, S., Axt, J. R., and Oliver, M. Levy, A. G., Scherer, A. M., Zikmund-Fisher, B. J., Larkin, N. (2016). Racial Bias in Pain Assessment and K., Barnes, G. D., and Fagerlin, A. (2018). Prevalence Treatment Recommendations, and False Beliefs of and Factors Associated with Patient Nondisclosure about Biological Differences between Blacks and of Medically Relevant Information to Clinicians. Whites. Proceedings of the National Academy of JAMA Network Open, 1, e185293. PERSONAL HEALTH SURVEILLANCE 279 Levy, N. (2018). Taking Responsibility for Health in an Nath, R. (2019). The Injustice of Fat Stigma. Bioethics, 33, 577–590. Epistemically Polluted Environment. Theoretical Medicine and Bioethics, 39, 123–141. Nissenbaum, H. and Patterson, H. (2016). Biosensing in Context: Health Privacy in a Connected World. In D., Lippert-Rasmussen, K. (2016). Luck Egalitarianism. Nafus (ed.), Quantified: Biosensing Technologies in London: Bloomsbury. Everyday Life. Cambridge, MA: MIT Press, Lucivero, F. and Jongsma, K. R. (2018). A Mobile pp.79–100. Revolution for Healthcare? Setting the Agenda for O’Neill, S. (2018). As Insurers Offer Discounts for Fitness Bioethics. Journal of Medical Ethics, 44, 685–689. Trackers, Wearers Should Step with Caution [WWW Lupton, D. (2012). m-Health and Health Promotion: Document]. NPR.org, available from: https://www. The Digital Cyborg and Surveillance Society. Social npr.org/sections/health-shots/2018/11/19/ Theory & Health, 10, 229–244. 668266197/as-insurers-offer-discounts-for-fitness- Lupton, D. (2013). Quantifying the Body: Monitoring trackers-wearers-should-step-with-cautio [accessed and Measuring Health in the Age of mHealth 29 January 2020]. Technologies. Critical Public Health, 23, 393–403. Paldan, K., Sauer, H., and Wagner, N.-F. (2018). Lupton, D. (2016). Digitized Health Promotion: Risk Promoting Inequality? Self-Monitoring Applications and Responsibility for Health and Illness in the Web and the Problem of Social Justice. AI & Society,doi: 2.0 Era. In J. E., Davies and A. M., Gonzalez (eds), To 10.1007/s00146-018-0835-7. Fix or to Heal: Patient Care, Public Health and the Pantelopoulos, A. and Bourbakis, N. G. (2010). Limits of Biomedicine. New York: NYU Press, pp. Prognosis—A Wearable Health-Monitoring System 52–76. for People at Risk: Methodology and Modeling. Lupton, D. (2019). ‘It’s Made Me a Lot More Aware’: A IEEE Transactions on Information Technology in New Materialist Analysis of Health Self-Tracking. Biomedicine, 14, 613–621. Media International Australia, 171, 66–14. Pereboom, D. (2014). Free Will, Agency, and Meaning in Macnish, K. (2014). Just Surveillance? Towards a Life. Oxford: Oxford University Press. Normative Theory of Surveillance. Surveillance and Petrini, C. (2013). Ethics in Public Health Surveillance. Society, 12, 142–153. Annali Dell’Instituto Superiore di Sanita `, 49, 347–353. Marmot, M. (2005). Social Determinants of Health Petrini, C. and Ricciardi, G. (2015). Ethical Issues in Inequalities. The Lancet, 365, 1099–1104. Public Health Surveillance: Drawing Inspiration Martani, A., Shaw, D., and Simone, E. B. (2019). Stay Fit from Ethical Frameworks. Annali Dell’Instituto or Get Bit—Ethical Issues in Sharing Health Data with Superiore di Sanita `, 51, 270–276. Insurers’ Apps. Swiss Medical Weekly, 149, 1–8. Pettit, P. (1997). Republicanism: A Theory of Freedom and Martani, A. and Starke, G. (2019). Personal Responsibility Government. Oxford: Oxford University Press. for Health: The Impact of Digitisation. Journal of Pillutla, V., Maslen, H., and Savulescu, J. (2018). Medical Law and Ethics, 7, 241–258. Rationing Elective Surgery for Smokers and Obese Martinez-Martin, N. and Kreitmair, K. (2018). Ethical Patients: Responsibility or Prognosis? BMC Medical Issues for Direct-to-Consumer Digital Psychotherapy Ethics, 19, 1–10. Apps: Addressing Accountability, Data Protection, and Raber, I., McCarthy, C. P., and Yeh, R. W. (2019). Health Consent. JMIR Mental Health, 5, e32. Insurance and Mobile Health Devices: Opportunities McGeer, V. (2015). Building a Better Theory of and Concerns. JAMA, 321, 1767–1768. Responsibility. Philosophical Studies, 172, 2635–2649. Roemer, J. (1993). A Pragmatic Theory of Responsibility Mello, M. M. and Wang, C. J. (2020). Ethics and for the Egalitarian Planner. Philosophy & Public Governance for Digital Disease Surveillance. Science, Affairs, 22, 146–166. 368, 951–954. Rønn, K. V. and Lippert-Rasmussen, K. (2020). Out of Mendelson, D. and Wolf, G. (2017). Health Privacy and Proportion? On Surveillance and the Proportionality Confidentiality. In I., Freckelton and K., Peterson Requirement. Ethical Theory and Moral Practice, 23, (eds), Tensions and Traumas in Health Law. Sydney: 181–199. Federation Press, pp. 266–282. Rubel, A. (2012). Justifying Public Health Surveillance: Minkler, M. (1999). Personal Responsibility for Health? Basic Interests, Unreasonable Exercise, and Privacy. A Review of the Arguments and the Evidence at Kennedy Institute of Ethics Journal, 22, 1–33. Century’s End. Health Education & Behavior, 26, Ruckenstein, M. and Schu ¨ ll, N. D. (2017). The Datafication 121–140. of Health. Annual Review of Anthropology, 46, 261–278. 280 DAVIES Samerski, S. (2018). Individuals on Alert: Digital Quantified Self, and the Participatory Biocitizen. Epidemiology and the Individualization of Journal of Personalized Medicine, 2, 93–118. Surveillance. Life Sciences, Society and Policy, 14, 1–11. Ter Meulen, R. and Maarse, H. (2009). Increasing Savulescu, J. (2018). Golden Opportunity, Reasonable Individual Responsibility in Dutch Health Care: Is Risk and Personal Responsibility for Health. Journal Solidarity Losing Ground? (SSRN Scholarly Paper of Medical Ethics, 44, 59–61. No. ID 1447168). Rochester, NY: Social Science Sax, A. (2017). Eigenverantwortung. Schweiz Arzteztg, Research Network. 98, 174. Topol, E. J. (2015). The Future of Medicine Is in Schmidt, H. (2007). Personal Responsibility for Your Smartphone. Wall Street Journal. https://on Health—Developments under the German line.wsj.com/articles/the-future-of-medicine-is-in- Healthcare Reform 2007. European Journal of Health your-smartphone-1420828632?reflink¼desktopweb Law, 14, 241–250. share_permalink (accessed 4 May 2021). Schmidt, H. (2009). Personal Responsibility in the NHS Vallentyne, P. (2002). Brute Luck, Option Luck, and Constitution and the Social Determinants of Health Equalityof Initial Opportunities. Ethics, 112, Approach: Competitive or Complementary? Health 529–557. 10.1086/339275. Econ Policy Law, 4, 129–138. Vargas, M. (2013). Building Better Beings: A Theory of Segall, S. (2007). Is Health Care (Still) Special? Journal of Moral Responsibility. Oxford: OUP. Political Philosophy, 15, 342–361. Ve ´ liz, C. (2020). Privacy during the Pandemic and Segall, S. (2010). Health, Luck and Justice. Princeton, NJ: Beyond. The Philosophers’ Magazine, 90, 111–113. Princeton University Press. Venkatapuram, S. (2011). Health Justice: An Argument Selmi, M. (2006). Privacy for the Working Class: Public from the Capabilities Approach. Cambridge: Polity. Work and Private Lives. Louisiana Law Review, 66, Voigt, K. (2007). The Harshness Objection: Is Luck 1046–1056. Egalitarianism Too Harsh on the Victims of Option Sharkey, K. and Gillam, L. (2010). Should Patients with Luck? Ethical Theory and Moral Practice, 10, 389–407. Self-Inflicted Illness Receive Lower Priority in Access Weber, S., Strommenger, D., Kertzscher, U., and Affeld, to Healthcare Resources? Mapping out the Debate. K. (2012). Continuous Blood Pressure Measurement with Journal of Medical Ethics, 36, 661–665. Ultrasound. Biomedical Engineering/Biomedizinische Sharon, T. (2017). Self-Tracking for Health and the Technik, 57, 407–410. doi:10.1515/bmt-2012-4108. Quantified Self: Re-Articulating Autonomy, WHO (2017). WHO Guidelines on Ethical Issues in Public Solidarity, and Authenticity in an Age of Health Surveillance. Geneva: WHO. Personalized Healthcare. Philosophy & Technology, WHO j eHealth. (n.d.) WHO, available from: http:// 30, 93–121. www.who.int/ehealth/en/ [accessed 29 January 2020]. Shelton, W. and Balint, J. A. (1997). Fair Treatment of WHO Global Observatory for eHealth and World Health Alcoholic Patients in the Context of Liver Organization (2011). MHealth: New Horizons for Transplantation. Alcoholism: Clinical and Health through Mobile Technologies. Geneva: World Experimental Research, 21, 93–100. Health Organization. Shemkus, S. (2015). Fitness Trackers Are Popular among Wiederhold, B. K. (2012). Self-Tracking: Better Medicine Insurers and Employers—But Is Your Data Safe? The through Pattern Recognition. Cyberpsychology, Guardian. theguardian.com, available from: https:// Behavior, and Social Networking, 15, 235–236. www.theguardian.com/lifeandstyle/2015/apr/17/fit Wikler, D. (2002). Personal and Social Responsibility ness-trackers-wearables-insurance-employees-jobs- for Health. Ethics & International Affairs, 16, health-data [accessed 29 January 2020]. 47–55. Stahl, T. (2016). Indiscriminate Mass Surveillance and Wilson, J. (2009). Not So Special after All? Daniels and the Public Sphere. Ethics and Information the Social Determinants of Health. Journal of Medical Technology, 18, 33–39. Ethics, 35, 3–6. Stemplowska, Z. (2009). Making Justice Sensitive to Wood, C. S., Thomas, M. R., Budd, J., Mashamba- Responsibility. Political Studies, 57, 237–259. Thompson, T. P., Herbst, K., Pillay, D., Peeling, R. Stone, K. V. W. (2002). Employee Representation in the W., Johnson, A. M., McKendry, R. A. and Stevens, Boundaryless Workplace. Chicago-Kent Law Review, 77, 773–819. M. M. (2019). Taking Connected Mobile-Health Diagnostics of Infectious Diseases to the Field. Swan, M. (2012). Health 2050: The Realization of Personalized Medicine through Crowdsourcing, the Nature Review, 566, 467–474.

Journal

Public Health Ethics – Oxford University Press

Published: May 16, 2021

There are no references for this article.