Data for Life: Wearable Technology and The Design of Self-Care?

Original Article

Data for life: Wearable technology and the design of self-care

Natasha Dow Schüll

Department of Media, Culture and Communication, New York University, 239 Greene Street, New York, NY 10003, USA.

Abstract Over the last 5 years, wearable technology – comprising devices whose embedded

sensors and analytic algorithms can track, analyze and guide wearers’ behavior – has increasingly captured the attention of venture capitalists, technology startups, established electronics companies and consumers. Drawing on ethnographic fieldwork conducted 2 years running at the Consumer

Electronics Show and its Digital Health Summit, this article explores the vision of technologically assisted self-regulation that drives the design of wearable tracking technology. As key artifacts in a

new cultural convergence of sensor technology and self-care that I call ‘data for life’, wearables are marketed as digital compasses whose continuous tracking capacities and big-data analytics can help

consumers navigate the field of everyday choice making and better control how their bites, sips, steps and minutes of sleep add up to affect their health. By offering consumers a way to simulta-

neously embrace and outsource the task of lifestyle management, I argue, such products at once exemplify and short-circuit cultural ideals for individual responsibility and self-regulation.

BioSocieties advance online publication, 7 March 2016; doi:10.1057/biosoc.2015.47

Keywords: digital health; wearable technology; self-tracking; self-care; big data

In January, 2014, exactly 1 week after President Obama’s Affordable Care Act’s mandate for universal coverage went into effect, I found myself standing in the back of a crowded conference room on the second floor of the Las Vegas Convention Center. Downstairs, the annual Consumer Electronics Show (CES) – the leading international venue for showcasing new consumer technologies – was in full swing. Over the next 2 days, 750 of the roughly 150 000 attendees would stream in and out of the room to hear the proceedings of the Digital Health Summit, a carefully curated lineup of keynotes, panels and super-sessions on how the tech sector might “capitalize on the new opportunities” brought by health reform.1

“Take a deep dive into the giant umbrella that is Digital Health”, invited the press release, explaining that the term encompassed “telehealth systems, mobile health applications and devices, sensor-based technologies, big data and predictive analytics, chronic care

1 The seed for this article was a short piece in the MIT Technology Review titled “Obamacare meets wearable technology” (Schüll, 2014).

© 2016 Macmillan Publishers Ltd. 1745-8552 BioSocieties 1–17 www.palgrave-journals.com/biosoc/

management, genomics, wearables, and wellness and fitness devices”. Throughout the proceedings, speakers described digital health technologies and the Affordable Care Act as a “dynamic duo”, to take the title of one session, that would “work together to leverage each other to achieve mutual success”. The new legislation would trigger a cascade of “incentivized compliance”, compelling insurers, health care providers and consumers to cut costs. The audience learned how big data is transforming epidemiology and public health, how advances in electronic health record systems are streamlining the practice of medicine, and how continuous monitoring devices are shifting the management of chronic conditions like diabetes and heart disease away from hospitals and doctors and into the hands of patients themselves.2

Most of the discussion at the Digital Health Summit, however, focused on the well, not the sick. As speakers frequently reminded attendees, 50–75 per cent of the monstrous US $2.7 trillion expended annually in the United States on health care is spent on preventable conditions linked to everyday behavior such as overeating, under-exercising and smoking – “lifestyle diseases”, or “diseases where you have a choice”, as one presenter put it.3 Consensus had it that the physical and financial costs of such conditions could be controlled by managing one’s day-to-day decisions: what to eat and drink, how much to move one’s body, how to avoid stress. The task of so-called lifestyle management was to keep oneself well by keeping one’s quotidian choices in check.4 “Health is not a side event, discrete and separate from everyday life”, commented Tom Paul, Chief Consumer Officer for UnitedHealth and a panelist at the Digital Health Summit. “Now it can be part of day-to-day living, everyday life”. Such a claim corresponds with a recent mutation in the notion of health that the

anthropologist Dumit (2012) has identified: once understood as a baseline state temporarily interrupted by anomalous moments of illness, health has been recast as a perpetually insecure state that depends on constant vigilance, assessment and intervention. “Health is no longer the silence of the organs”, writes Dumit; “it is the illness that is silent, often with no symptoms” (p. 15).5 The sociologist Rose (2007) makes a similar observation when he writes of a

2 The US government’s Obama administration has taken a keen interest in the power of big data to transform health care. The US Department of Health and Human Services, the US National Institutes of Health’s Office of Behavioral and Social Sciences Research, and government-funded entities such as the National Science Foundation and the Robert Wood Johnson Foundation have invested in mHealth (or “mobile health”) initiatives as a way to address wide-scale population health problems. Projects include smoker cessation apps, health text messaging, digital tools for the management of diabetes or for medication compliance, and the like. Market research shows that over one third of doctors recommend health or medical apps for their patients (MobiHealthNews, 2014). (See also Goetz, 2010; Topol, 2012, 2015).

3 For more on the rise of chronic disease, see the accounts of historians of medicine Weisz (2014), Armstrong (2014) and Greene (2007) who writes of “a shift in the basic conception of chronic disease from a model of inexorable degeneration to a model of surveillance and early detection” (p. 84). For analyses of the idea of “lifestyle” see Friedman (1994), Giddens (1991) and Dumit (2012).

4 In 1980, the sociologist Robert Crawford described an early version of lifestyle management linked to the simultaneous depoliticization and privatization of health then taking place in America: collective struggles for wellbeing were being replaced by an emphasis on individual self-care in the form of lifestyle modification (1980, p. 365). Solutions to bad diet, for instance, were located “within the realm of individual choice”, in the ability to resist advertising and overcome bad habits (p. 368).

5 Unlike acute diseases that arise suddenly, lifestyle diseases pose “a more sinister threat, another type of mortal hazard with slower effects that go stealthily into the blood one cancerous bacon sandwich or poisonous drink at a time, potential killers by degrees that might catch up with us later in life” (Blastand and Speigelhafter, 2014).

Schüll

2 © 2016 Macmillan Publishers Ltd. 1745-8552 BioSocieties 1–17

pervasive sense of susceptibility or “the sense that some, perhaps all, persons, though existentially healthy are actually asymptomatically or pre-symptomatically ill” (p. 20). In the era of lifestyle disease, everyone is potentially sick and must take measures to keep well. Dumit is specifically concerned with the characterization of health as lifelong pharmaceutical

treatment, what he calls “drugs for life”. The culturally valorized mode of living that corresponds to this notion of wellness entails closely watching one’s bio-levels and adjusting pharma-cocktails at any sign of slipping above or below ever-shifting thresholds of normality. At stake in this article is what I call ‘data for life’, a related, complimentary response to the notion that we are all potentially sick in which wellness depends on the continuous collection, analysis and management of personal data through digital sensor technologies (see Swan, 2012). “We discovered that people didn’t necessarily need more data about their medical lives”, reads a report on the value of personal data by the Robert Wood Johnson Foundation health care think tank; “instead, they needed more information about how their everyday actions influence their health”. Following this logic, “constant informational body monitoring is imperative”, as Viseu and Suchman (2010, p. 173) have described the mandate of wearable computing. Although people have long used simple, analog devices to record, reflect upon and regulate

their bodily states and processes (for example, diaries, scales, wristwatches, thermometers), the present historical moment is witnessing a dramatic efflorescence in the use of digital technology to self-track (Crawford et al, 2015). As mobile technology spreads, as electronic sensors become more accurate, portable and affordable, and as analytical software becomes more powerful and nuanced, consumers are offered an ever-expanding array of gadgets equipped to gather real-time information from their bodies and lives, convert this information into electrical signals, and run it through algorithms programmed to discern patterns and inform interventions into future behavior.6

As recently as 5 years ago, individuals who embraced such technology were likely to identify as members of Quantified Self, a community of avid self-trackers whose tagline is “self- knowledge through numbers”.7 Yet wearable technology has increasingly captured the attention of venture capitalists, technology startups, established electronics companies and

6 Scholars of the “Internet of things” (Halpern et al, 2013) and “sensor society” (Andrejevic and Burdon, 2014) have called attention to the importance of sensor technology to contemporary life. Dramatic increases in the sensitivity and sophistication of sensors along with decreases in their size means they can be loaded into clothing, pillboxes, toothbrushes and smartphones – which are becoming wearable tracking devices in themselves. Algorithms operating on the tracked data “can analyze data along multiple lines – time, frequency, episode, cycle and systemic variables”, writes Swan (2013) , a science and technology innovator and philosopher, and in this way detect “elements that are not clear in traditional time-linear data”: patterns, cycles, exceptions, the emergence of new trends, episodic triggers, variability, correlations and early warning signs (p. 90).

7 Founded by two former editors of Wired magazine in 2007, Quantified Self currently claims 45 000 members in 40 countries. In online forums and in meetings around the world, quantified selfers share their attempts to experiment with diet and meditation, monitor drug side effects, correlate hormone levels with mood fluctuations and relationship dynamics, or even evaluate semantic content in daily email correspon- dence for clues to stress and unhappiness. Social studies of quantified self include Lupton, 2015, 2013a, 2016; Albrechtslund, 2013; Boesel (see her blog, http://www.thesocietypages.org/cyborgology/author/ whitneyerinboesel/), Mackenzie, 2008; Nafus and Neff, 2016; Nafus and Sherman, 2014; Oxlund, 2012; Pantzar and Ruckenstein, 2015; Ruckenstein, 2014; Potts, 2010; Schüll, forthcoming; Till, 2014; Berson, 2015; Watson, 2013. While journalists typically cast those who live by numbers as narcissistic and obsessive in their zeal for personal data, digital health pundits hold them up as beacons of a sensible tracking future. At the same time, they recognize that mass-market users are not as responsive to quantification as the typical QS member and that technology must be designed in a way that makes it “automated, easy, inexpensive, and comfortable” (Swan, 2013, p. 93).

Data for life

3© 2016 Macmillan Publishers Ltd. 1745-8552 BioSocieties 1–17

mass-market consumers. Revenue from digital fitness devices such as the popular FitbitTM

wristband totaled $330 million in 2013 and is expected to reach $1.8 billion in 2015 – and $5.8 billion by 2018 (GovLab, 2014; Consumer Electronics Association, 2015). Attesting to the robustness of this new market, the aisles of Best Buy and Walmart are abundantly stocked with gadgets designed to record personal metrics, the Internet rife with downloadable smartphone apps that can monitor and help adjust behavior. The online marketplace Amazon has launched a specialty shop for “Wearable Technology” featuring approximately 800 products, the vast majority categorized under “fitness and wellness”; the shop includes a buyer’s guide for understanding what wearables are and how to incorporate them into one’s lifestyle. Personal electronics and smartphone giants Apple and Samsung, as well as software and Internet leaders Google and Microsoft, have all recently introduced health and fitness tracking systems. Consumer tech pundits have forecast that by the close of 2015 well over 500 million people will use mobile health applications and devices (Rahns et al, 2013). “Dashboards 24–7”, said Michael Yang of Comcast Ventures when pressed to give his own

prediction at the 2014 Digital Health Summit. That year, summiteers frequently invoked the 2013 Pew report “Tracking for Health” (Fox and Duggan, 2013) and its finding that nearly 70 per cent of American adults already tracked weight, diet, or exercise while one-third tracked health indicators such as blood pressure, blood sugar, headaches, or sleep patterns.8

What made these trackers an especially promising market was the fact that only 20 per cent tracked with technology; the rest, who used pen and paper or simply tracked “in their heads”, could be targeted for conversion to the use of digital tools. Although it is too early to tell whether wide-scale digital tracking will come to pass (indeed, studies are beginning to show that mainstream consumers do not use wearables consistently or as intended),9 people are purchasing self-tracking gadgets and downloading self-tracking apps in rising numbers; tech companies are dedicating significant resources to the development of new gadgets and apps; health-care policymakers and insurance companies are optimistic that these technologies will help mitigate lifestyle diseases; and drug companies hope they will help solve the problem of medication compliance.10 Whatever the future holds, the present moment is one of heavy investment in tracking technology by multiple stakeholders. Social scientists who have begun to explore this moment most often focus on the experience

and practices of self-trackers, particularly their ambivalent embrace, creative repurposing, or outright rejection of tracking technology and the project of “living by numbers” (Mackenzie, 2008; Mol, 2009; Potts, 2010; Oudshoorn, 2011; Boesel, 2012a, b; Oxlund, 2012; Schüll, forthcoming; Albrechtslund, 2013; Watson, 2013; Nafus and Sherman, 2014; Ruckenstein,

8 “People living with chronic conditions”, the authors of the report write, “are significantly more likely to track a health indicator or symptom” (Fox and Duggan, 2013, p. 2). They go on to note that two-thirds of US adults are considered overweight or obese and half are living with at least one chronic condition – most often high blood pressure and diabetes (ibid., p. 6).

9 According to recent reports by industry analysts, a third of people discontinue tracking within the first 6 months (Ledger, 2014; Ledger andMcCaffrey, 2014). Nafus and Sherman (2014) have shown how trackers frequently switch between devices, interrupting data streams and amounting to a form of “soft resistance”.

10 Dumit (2012) has observed (personal communication) that what I call ‘data for life’ is becoming a part of the “drugs for life” agenda; although “changes in lifestyle such as exercising more and watching one’s diet are rendered secondary” to the administration of pharmaceuticals (p. 127), drug companies are increasingly looking to self-tracking technology to help solve the problem of medication compliance. As in the case of diabetes, it is suggested that ongoing glucose monitoring, exercise and diet be combined with a lifetime of drug-taking.

Schüll

4 © 2016 Macmillan Publishers Ltd. 1745-8552 BioSocieties 1–17

2014; Berson, 2015; Lupton, 2015, 2013a, 2016; Pantzar and Ruckenstein, 2015). Here I focus on an even less examined aspect of the contemporary tracking moment: the vision of technologically assisted self-care that drives the design of wearable tracking technology. “Little knowledge”, writes the sociologist Lupton (2014) in a recent article on tracking apps (p. 618), “is available on the practices and tacit assumptions of app developers and designers and the companies that commission apps”. While Lupton attempts to illuminate these practices and tacit assumptions by considering product logos, Websites and advertisements, my approach in this article is primarily ethnographic, looking ‘behind’ finished products to document the debates, challenges and emergent articulations of wearable technology stakeholders. What models of human behavior shape technologies of continuous tracking? How do the

technologies, in turn, shape new models for living? How does ‘data for life’ as a mode of self- regulation reflect and contribute to the broader field of contemporary liberal-democratic regulative rationalities? The Consumer Electronics Show, where I conducted two consecutive years of ethnographic research (in 2014 and 2015), affords a uniquely illuminating vantage on these questions. In the aisles, booths and meeting rooms of its sprawling exposition floor, new technologies are showcased and demonstrated, financed and bought, while on the panels of its carefully curated Digital Health Summit, new definitions of health and healthy behavior are formulated, promoted and debated.

Keeping Track: Designing Digital Self-care

Of the 3300 companies exhibiting at the 2014 Consumer Electronics Show, 300 ministered to digital health. Such exhibitors occupied 40 per cent more floor space than the previous year, with most clustered in the far lower wing of the Las Vegas Convention Center’s South Hall, in a 36 000 ft2 area that consisted of adjacent “Fitness Tech” and “Digital Health” zones. In these zones, tech companies showed off a dizzying array of devices for tracking personal data and promoting healthy behavior: heart-rate detecting ear buds, body-fat measuring bathroom scales, electronic skin patches that monitor blood flow, smart toothbrushes that help people brush their teeth correctly and long enough, and an impressive collection of wristbands packed with sensors to log footsteps, heart rate, sleep phases and more. By way of demonstration, gymnasts performed feats of physical prowess, showgirls danced and athletically clad young women speed-walked on treadmills while wearing the bands, their data displayed on large screens. A small podium at the intersection of the Fitness, Health, and Aging zones was busy throughout the conference with product demonstrations. A central booth in the Fitness Tech zone belonged to Fitbit, currently the undisputed market

leader in wearable fitness.11 The company makes a wearable movement-tracker that syncs with users’ personal computers, mobile phones, and now smart watches to continuously monitor steps taken, hours slept and other data they might choose to enter. The stated purpose is to bring about “a healthier you” (see Figure 1). Elaborating on this aspiration, a video advertisement for the device that played on a loop for the duration of the exposition featured close-up shots of different body parts in motion as the voiceover celebrated the arrival of a

11 One of Fitbit’s chief officers chairs the newly formed Health and Fitness Technology Division of the Consumer Electronics Association, which oversees the presence of digital health technology at each year’s Consumer Electronics Show.

Data for life

5© 2016 Macmillan Publishers Ltd. 1745-8552 BioSocieties 1–17

sensor that could keep track of routine activities. “Every step you take, every goal you set, every choice you make to be active” would be recorded – either by the band slipped over the wrist or the pendant clipped to the waist. A runner was shown in slow motion, concentric circles radiating out from his feet as they struck the pavement, illustrating the technical achievement of accurate, consumer-grade signal processing and prompting potential users to attend to the rippling consequences of their every movement. Next a woman was shown playing with her children in their backyard, segueing to her home office where crisp imaginary lines extended from her blurry form in the window to the smartphone and laptop on her desk. The sequence suggested that consumers could trust the device to capture the information they generated as they moved through their days – calories burned, distance covered, duration of activity, fluctuation in weight – and keep it synchronized in real-time, across all screens. Partway through the advertisement, a man paused at the turnstile of a subway entrance,

then slowly turned to face the camera as a digital overlay indicated the choice before him: Monorail versus WALK. With a smile he rejected the train and set off on an illuminated footpath to realize the “potential 2000 steps” he would thereby add to his daily count. At the press of a button on his mobile smartphone, he broadcast his choice to a group of friends, his new step-count advancing him on the leaderboard of life. The advertisement closed on the figure of a woman sleeping, an expression of contentment on her face, her exposed arm revealing the band that promised to optimize “even inactivity”. Fitbit competitor Jawbone’s “idle alert” also addresses inactivity, vibrating when wearers

are still for too long. Some wearables focus entirely on bodily stillness – tracking it, preventing it, helping users maintain proper posture during it. A simple gadget called the Rise sits in one’s pocket and records time sat throughout the day. CES 2014, saw the debut of Lumo Lift posture technology, an update to the Lumo Back. In its latest iteration, the Lumo sensor fastens upon one’s lapel or brassiere strap, whence it records and corrects posture with subtle (or not so subtle) vibration. “Through the app, you can control when you’re buzzed, how you’re buzzed, and even how intensely it buzzes”, informed the plasma display in Lumo’s

Figure 1: “Small daily decisions=BIG results.” Still image from a promotional video for the Fitbit Zip from 2012, available online at youtube.com (https://youtu.be/fsKvNB0Fvb0).

Schüll

6 © 2016 Macmillan Publishers Ltd. 1745-8552 BioSocieties 1–17

booth. While the technology performs standard activity tracking, its primary purpose is to monitor and regulate the stationary states of sitting or standing: “All about your posture. All about your life. One move changes everything”. A promotional video for the Lift closes with images from a corporate meeting in which the wearer briefly pauses to lift her head and bring her shoulders back, seemingly unprovoked, and allows a triumphal smile to play across her face. “Small changes can be empowering”, the ad exults. Kitty corner from the Fitbit exposition booth was a smaller booth featuring a device designed to

entrain users to more mindful eating habits. The smart utensil HAPIfork intervenes in the habit of feeding by monitoring and recording the length of each meal, the number of fork-servings per meal and the time between each of these servings; if shorter than 10 seconds, the fork will oscillate so that the eater knows to slow down – an effect achieved via proprietary “slow control” technology. “You are advised to take about 10–20 chews. If you trigger the HAPIfork’s alarm [by eating too fast], don’t panic. Set the fork down at the side of the plate and wait until the light turns green again, signaling that it is safe to take another bite”. The device, which turns something as routine as a single bite of food into a matter of potential danger, is presented as an “everyday technology” that helps users “take control of their health”. The company recommends keeping smartphones in view so users can see their data as it is collected in real time; as they feed themselves, their data is fed back to them. “Every bite is a potential teaching experience”, noted a user in a TechCrunch review (Lawler, 2013.)12

Hydration, another mundane yet vital human action, is the focus of the H20-Pal monitoring device, a small wireless scale that one attaches to the bottom of any water bottle where it keeps track (using built-in flash memory and a weight measuring sensor and accelerometer) of how much liquid is consumed from it, conveying this information to a users’ smartphones where a corresponding application is programmed to alert those who have not hydrated enough. Similarly, BluFit’s bottle “passive hydration tracking” bottle uses built-in sensors to measure how much has been drunk, automatically adjusting daily goals based on temperature and humidity and providing feedback in the form of flashing lights. Yet another smart water bottle, the Ilumi, changes color from red, to yellow, to green throughout the day to signal users’ proximity or distance to their preset hydration goals. Like the other devices in the “Fitness Tech” and “Digital Health” zones at CES 2014, these devices transfer the burden of tracking – and, in some cases, behavior change – from selves to sensors and computational algorithms. Back upstairs at the Digital Health Summit, technology designers, doctors and government

representatives continued to brainstorm on how to get personal data technology onto the wrists and into the pockets of more consumers. The accuracy and feasibility of monitoring, they reported, was good and getting better, and data scientists were continuing to refine analytic algorithms; the challenge when it came to self-tracking devices and programs was consistent use – “getting people to use the damn thing, so that it becomes part of their lifestyle”, as the Executive Vice President and Chief Medical Officer of the UnitedHealth insurance company put it. To make data tracking a habitual facet of lifestyle, it was not enough, all agreed, to simply

provide the data. “Nine