- 1 We would like to thank Tereza Stöckelová, Hana Porkertová, anonymous reviewers, and the editor of t (...)
1When we arrived at the weekend educational retreat for people diagnosed with obesity and/or diabetes organised in a regional town in Central Bohemia, we were immediately surrounded by wearable technological devices1. In the hotel lobby, a line had formed as a nurse applied continuous glucose monitoring sensors to the arms of participants with diabetes, some of whom were already calibrating their readings with glucometers. Nearby, a woman sitting on a couch pulled out an insulin pump in an elegant pink case from her cleavage to check her glucose and insulin levels. As we walked toward the canteen, we passed a group of women logging the lunch menu into their self-tracking apps. Later in the afternoon, self-tracking technologies were again the topic. During a seminar on nutrition in obesity treatment, a group of about twenty participants listened attentively to Professor Topol, the head bariatrician, who has been organising the educational retreats annually with the goal of creating a space where people with obesity and type 1 diabetes (hereafter T1D) can gain expert insights and share their own experiences. Professor Topol explained that using a calorie-tracking app can help participants better understand their intake and achieve a «caloric deficit», where energy consumed is less than energy expended. Pausing briefly, he added, «We can calculate and tell you what to eat and drink, but we have no way of knowing what your gut will actually take from it». He acknowledged that a human body is more complex than a simple input-output equation and that metabolic processes are far from predictable. And yet, maintaining a caloric deficit remains fundamental to weight management. Later that day, during the after-lunch walk, Ms. Stříbrská, a woman in her fifties living with T1D, shared her experience using a hybrid closed-loop system – an advanced technology in diabetes care that combines a continuous glucose monitor (CGM) with an algorithm-driven insulin pump which adjusts delivery in real-time. She could have calculated her carbohydrate intake and dosed insulin, but her volatile insulin needs, especially during physical activity, make this challenging. Exercise generally lowers blood glucose, so insulin taken for a meal might later cause hypoglycaemia during sports. Instead, she logs her activities, such as eating or cycling, into the system without pre-emptively deciding on insulin dosage: «I don’t know if I’ll burn as much as I ate, so I tell [the loop] I’ve had 30g of carbs but don’t bolus any insulin – I just log it». In this way, she uses the loop as a communication tool for her metabolism.
2The clinical approach to nutrition in obesity and T1D management has evolved under the model of metabolism as a «human factory» that transforms food into energy. The mechanistic notion of input and output that this model implies is inscribed into treatment procedures such as the «regime of a diabetic», which involves the meticulous counting of carbohydrate intake converted into amounts of injected insulin, or the «reduction diet» for losing «excess» weight based on maintaining a caloric deficit. The American historian and sociologist Hannah Landecker (2013) argues that the scientific understanding of metabolism developed in the 19th century under the influence of industrialisation, resulting in what she defines as «industrial metabolism». Within this framework, self-tracking devices and enumeration practices appear as tools of accounting used to calculate, equate, or convert food into numbers. However, recent biomedical research highlights the complexity and dynamism of metabolism indicating that the industrial model of metabolism is overly simplistic. Rather than functioning as a factory, metabolism is a «regulatory zone» that responds not only to what passes through the digestive system, but also to environmental stimuli that may get «under the skin» in ways other than through the digestive system (Landecker 2024). This post-industrial understanding of metabolism, which is «constituted by a dynamic web of cellular signals, built by and responding to environmental information» (Landecker 2013: 496), underscores its complexity and malleability. Landecker further argues that food itself should not be approached as a passive object merely consumed by the eating body but rather as an active «metabolic partner» that «speaks, cues, and signals» (2020). Eating can thus be analysed as an interlocution – or, as she puts it, as a «chemical chatter between food, microbes, intestines, brains and nerves» (ibidem).
3In this article, we offer an analysis of self-tracking devices as part of post-industrial metabolism. In this view, we argue, the understanding of devices such as nutritional apps, glucose monitors, and algorithm-driven hybrid closed-loops as simple accounting tools is insufficient. Rather, they become tools of communication, which amplify and facilitate a more-than-human conversation involving humans, foods, the gut microbiome, drugs, and the larger metabolic environment (cf. Kolářová et al., 2022). Self-tracking devices can then be analysed as interlocutors and partners in metabolic sensing, negotiations, and experimentations. The numbers derived from these self-monitoring practices are employed as a language of communication within the regulatory network of metabolic processes. Our research is situated within the ongoing «metabolic turn» in both life and the social sciences, which focuses on the social, political, and environmental dimensions of «metabolic living» (Solomon 2016; see also Hardon, Smith-Morris 2017; Landecker 2024; Barua 2024). Following Landecker, we employ the notion of metabolism as an analytical tool in order to examine the more-than-human assemblages that constitute (metabolic) health, which has been widely datafied and digitalised in recent years (Candell et al., 2020). We explore how the shift from the industrial to the post-industrial model of metabolism redefines practices of quantifying metabolism. To trace how more-than-human metabolic relations are integrated into self-care routines in obesity and T1D management, we zoom in on two topics: the interplay between high-tech and low-tech measuring devices and the imperatives of control. Our analysis is guided by the following questions: How do digital and analogue devices complement or challenge each other in health monitoring and what models of metabolism are enacted in such practices? How is the notion of control reinforced through the use of self-tracking technologies and how does this notion change when people use them otherwise? And how do diabetes and obesity treatments transform with the integration of more-than-human metabolic relations (beyond the individual factory-like model of metabolism) into care for one’s metabolism?
4After situating our study in the context of the existing literature on the datafication and digitalisation of health and briefly describing our research design and methods, we present our findings in two sections that reflect the interwoven ways in which bodies, health, and illness are lived and experienced through numbers and measurements. In the first section, we explore how and why healthcare professionals and individuals with the condition(s) we are studying here alternate between high-tech and low-tech devices, prioritise specific measurement practices, and engage with the produced data. We focus on how these practices include or disregard more-than-human metabolic relations and what model of metabolism they enact. The second section examines the enactment of control through the continuous monitoring of bodily processes and the transformation of control when self-tracking devices are used as tools of communication within the metabolic «regulatory zone». The closing discussion addresses how ethnographic methods that trace more-than-human metabolic assemblages can offer a more nuanced understanding of the role of self-monitoring, self-control, and self-tracking technologies in enactments of bodies in health and disease.
5In recent years, there has been a growing body of social science research on datafication in healthcare. In her pioneering studies, the Australian sociologist Deborah Lupton (2014, 2016) examines how self-tracking practices align with the concept of self-governance (Foucault 1986), which is rooted in continuous observation and in the imperatives of self-optimisation. Her analysis emphasises how the knowledge generated by self-tracking technologies reinforces individual responsibility for achieving bodily states that fit into societal expectations of health, normalcy, and selfhood. This dynamic, Lupton argues (2013a, 2013b), is contributing to the emergence of the «digitally engaged patient» – a figure moulded by these technologies and their promise of control. Self-monitoring devices and apps can reinforce the idea of the «good neoliberal citizen» who takes personal responsibility for their health by actively engaging with technologies to enhance self-optimisation and self-knowledge. In this context, a person with diabetes who is not «well-compensated» – meaning that they have failed to maintain target glycaemic levels – might be perceived as irresponsible or neglectful. This pressure on patients to control their metabolic conditions frames disease management as a matter of personal discipline rather than acknowledging biosocial complexities. As we explore in this paper, the pursuit of control through the quantification of health often competes with other values in life, revealing the limits of control and predictability.
6In her study on technological devices in diabetes care, Annemarie Mol (2008) analysed how glucometer advertisements framed self-monitoring through the logic of choice, portraying patients as informed consumers responsible for selecting the right device to control their bodies. This contrasts with the logic of care which Mol observed in clinical practice, where managing diabetes is an ongoing, adaptive process that acknowledges the unpredictability of chronic illness. Revisiting her work two decades later, with glucometers largely replaced by continuous glucose monitors (CGMs) and algorithmically driven hybrid closed-loop systems, and with the shift in biomedical models away from a factory-like notion of metabolism to a post-industrial, networked understanding (Landecker 2013), we choose to focus on the notion of control rather than following the defined logics. How is control enacted in more-than-human metabolic assemblages, where bodies, sensors, and algorithms communicate rather than merely calculate? And how does metabolism, as an analytical tool, reframe the interplay between Mol’s logics of choice and care in the technologised landscape of diabetes management?
7Based on a study conducted among young people with T1D, Scheldeman (2010) argues that personal medical devices, such as insulin pumps and glucose sensors, not only distribute knowledge across technologies, objects, and bodies but also extend senses and feelings through sensors and algorithms that mediate bodily processes (see also Mialet 2022). By thinking with the notion of metabolism, we can focus on metabolic mediators – or rather metabolic partners – beyond technologically extended bodies into their environments. Viewed through the paradigm of the post-industrial metabolism, the devices studied by Scheldeman can be analysed as part of a broader network of metabolic communication. We also draw inspiration from Hockin-Boyers, Jamie, and Pope (2024), who proposed the notion of «intuitive tracking» to account for how highly rationalised and enumerated practices of health are combined with a deeply embodied awareness of the body. In line with the authors, we focus on how people living with T1D and obesity manage to integrate and calibrate between data derived from self-tracking and bodily self-awareness to maintain their well-being.
8This paper emerged within the ongoing research project «Technocultures of extended metabolism», which explores how metabolism is shaped in people living with type 1 diabetes (study 1), obesity (study 2), and home parenteral nutrition (study 3). The research started in January 2024 and is conducted by a research team based at the Institute of Sociology of the Czech Academy of Sciences in collaboration with the Centre for Research on Nutrition, Metabolism, and Diabetes at the Third Faculty of Medicine, Charles University.
9In this paper, we draw on case studies concerned with T1D and obesity. Employing ethnographic research methods, which, according to Whatmore (2003: 93), «come closest to the notion of “generating materials”, as opposed to “collecting data”», we follow how people living with obesity and T1D attend to their metabolisms, nutrition, and health while engaging with various technologies and means of digitalisation and datafication. We aim to flesh out the complexities and intricacies of self-monitoring practices that are often reduced to the narratives of self-optimisation and self-control promoted in biomedical discourse (cf. Clarke et al. 2003) and technology design. We build on approximately 40 hours of semi-structured interviews conducted with people living with obesity or T1D and with healthcare professionals. In the interviews with healthcare professionals, we focused on treatment protocols and procedures, the use of new drugs and technologies, and the specificities of the Czech healthcare system that they consider relevant for the treatment of obesity and diabetes. The questions in the interviews with people living with obesity and T1D focused on their experiences with the Czech healthcare system, their everyday practices of cooking and eating, and their relationship to their embodiment and the technologies they use. For the purpose of this paper, we concentrate on codes relating to the use of self-tracking technologies, measuring devices, and quantification practices that both doctors and patients use in their daily lives. These codes include «measuring food», «weight measurement», «counting calories», «measuring glycaemic levels», «counting carbs», «counting insulin boluses», and the «constant visibility of numbers».
10The analysis also builds on participant observation at events dedicated to the treatment and management of obesity and T1D, such as the educational retreat mentioned in the opening of this papers, a certified medical conference on advances in glucose monitoring, and several round tables on obesity treatment and prevention. In total, we attended 13 events.
11All the interviews and all the conversations that occurred as part of participant observation in both studies took place in Czech, and the excerpts used in the paper were translated into English by the authors. All the names used in the paper are pseudonyms. The research was approved by the Research Ethics Board of the Institute of Sociology of the Czech Academy of Science (approval no. SOÚ-384_3/2024)
12In this paper, the majority of the ethnographic vignettes are drawn from the educational weekend retreat for people living with obesity and T1D organised by Professor Topol in April 2024 and online webinars organised by his team throughout the summer of 2024. One of the authors approached Professor Topol with a request for a research interview within the ongoing research project. Intrigued by the project’s research questions, he invited us to attend the retreat and participate in other events within the network of educational groups and patient support meetings that he coordinates. He also introduced us to the chair of the patient organisation, whom we continue to collaborate with, focusing on users’ experiences with the technologies used in T1D management. At the education retreat, he granted us access to all the workshops and events that were part of the retreat and introduced us to the retreat participants as «sociologists conducting ethnographic research» so that they were aware of our presence.
13A few years ago, Ms Šedá underwent gastric sleeve surgery – a procedure in which a part of the stomach is removed in order to help patients with obesity lose weight. One of the most important rules of postoperative care is that a serving of food must not exceed 150 ml, as to repeatedly break this rule would destroy the weight loss effect of the surgery. Since undergoing the surgery, Ms Šedá always carries a 150 ml food container in her purse to measure her food: «When I am not sure about the size of a portion, I just put it in the container. See, [it’s] not that complicated». To help patients measure portion sizes, dietitians recommend translating the numbers into a material form, as Ms. Šedá does with her food container. The number of millilitres will be indicative, not exact. But, as Ms Šedá told us, when the situation is uncertain the body steps in and does the rest: «Your body just won’t let you eat more than [it can]. So, you just listen to your body». In this case, the stomach becomes a measuring device that supports the work of the food container. But feeding the body is about more than just the amount of food, as the nutritional composition of the food matters as well. «This app shows me how many nutrients I need to consume. I have a premium account, so it also shows me the amount of salt and sugar», Ms Šedá explained, pointing to the colourful diagrams in her app for tracking calorie intake. She complained that while some nutrients are «easy to get», others require more attention: «Proteins are hard to obtain, and fats ... I am done with fats before noon!» The app visualises the recommendations as colourful diagrams. Every time Ms Šedá logs the food she has eaten in the app, the diagrams change slightly and show how much of the particular nutrient she has consumed. Ms Šedá’s «goal», in simple terms, is to literally complete all the diagrams in order to provide her body with what it requires and to avoid eating more than indicated. However, in contrast to the 150 ml rule, where, if she pays enough attention, her body can let her know when to stop eating, in the case of nutrient tracking, Ms Šedá has to rely on the diagrams so she knows when she is «done with fats» and instead needs to get more proteins.
14The diagrams in the app visualise something that is not easily articulated or felt. While the body might eventually recognise deficiencies or imbalances over a longer time period, the goal here is to maintain a consistently balanced intake. The app provides real-time feedback, which enables Ms Šedá to make immediate adjustments, rather than relying on the body’s slower, cumulative signals. Her use of the nutrient-tracking app reflects an industrial model of metabolism, where bodily needs are understood as precise material inputs that must be supplied in correct proportions. Constant monitoring and external control are required to ensure that the body receives the necessary «building blocks» and avoids excess. The colourful diagrams reinforce this mechanistic logic, representing food intake as a quantifiable and goal-oriented process, where nutritional balance is achieved by completing visual charts rather than by responding to internal bodily cues. The body’s regulatory capacities are thus overshadowed in favour of data derived from the app. To better manage her nutrition, Ms Šedá employs digital tools such as an electronic kitchen scale at home or a calorie-tracking app on her smartphone. These technologies help her to stay properly nourished despite the restrictions imposed by her limited stomach capacity. Measuring portions helps her to maintain the weight-loss effects of her sleeve surgery. However, when away from home and without her digital scale, Ms Šedá turns to an analogue alternative to measure 150 ml: a food container. While it lacks the precision of a digital scale, it is portable and lightweight, doesn’t require charging, and fits in a handbag. It might not provide the most accurate results, but it is good enough to do the job. If she measures incorrectly, Ms Šedá can rely on her stomach to signal when it is full.
15In obesity treatment, the use of low-tech tools is not a sign of backwardness. In fact, these tools can often deliver practical and sufficiently accurate results. During a roundtable discussion on the national obesity prevention plan, a medical professor remarked: «Body Mass Index won’t help you to diagnose obesity. You just need to measure your patient’s waist with a tape measure». With this statement, he highlighted the limitations of BMI – a screening tool widely criticised for its diagnostic irrelevance (e.g., Gutin 2021) – and pointed to the practicality of using a tape measure instead. If a patient’s waist circumference exceeds 80 cm in women or 94 cm in men, the doctor should consider additional criteria to assess potential overweight or obesity. Unlike BMI, which fails to account for body fat distribution or distinguish between fat and lean tissue, a tape measure can immediately reveal fat concentrated in the abdominal area – a significant indicator of obesity-related health risks.
16In T1D management, measurement precision and accuracy seem to be more important than they are in obesity management. In this case, the golden number is «15 g», which refers to the 15 grams of carbohydrates that a person should consume to help them get through an episode of hypoglycaemia, a serious condition in which a person’s blood sugar level falls below 3.9 mmol/l, potentially leading to unconsciousness. However, a person experiencing hypoglycaemia should not consume more than 15 g, as doing so can quickly lead to hyperglycaemia, another dangerous state in which blood sugar levels rise to over 10 mmol/l. The question is how to get the desired 15 g in the everyday practice of T1D management?
17In reference to this problem, Dr Buk advised attendees not to drink juice – which is widely used as a source of quick carbs in cases of hypoglycaemia – straight out of the carton but to pour it into a glass to roughly estimate 15 g. Indignant laughter was heard from the audience as Mr Bílý took the floor: «With all due respect, have you ever tried pouring juice from a carton into a glass during hypo?» The audience murmured and nodded in agreement. A glass is not a device well-suited for measuring 15 g of carbs. This process requires thinking, taking a glass off a shelf, opening the fridge, opening the lid, and pouring the juice out of the carton: a series of simple movements that become extremely difficult to perform for a rather unsteady hypoglycaemic body. It has to be possible to achieve the precise number of 15 g with much less effort. That is why Mr Bílý and other participants prefer to have little pouches of fruit purée or grape-sugar candies ready at hand.
18The choice of measuring tool thus depends not only on the nature of the technology involved and the concrete situation in which it is used but also on the individual’s ability to perform the measurement. Endocrinologist Dr Olšanská argued that teaching a seventy-year-old diabetic patient to use a CGM sensor with tiny screens they can barely read is somewhat impractical. Increasing technologisation does not necessarily translate into an improved quality of life or treatment process. In some cases, a simpler device, such as a glucometer, may not be just good enough but may be the best tool for the purpose.
19There are also other cases where glucometers are still highly valued. As we chatted during a Nordic walking workshop, Ms Stříbrská noted that «the loop is dependent on the accuracy of the data from the CGM sensor […] You do not want your loop to work based on the wrong information». That is why she always makes sure to calibrate the data from the sensor with how she herself feels and eventually by pricking her finger with a glucometer. When using a high-tech device such as a hybrid-closed loop system for insulin delivery, the precision of the input data matters, and such precision is achieved through a combination of low-tech (glucometer, a person’s own feelings) and high-tech (CGM sensor) measurement tools.
20Other input data where precision matters are the amount of carbohydrates and the insulin-to-carb ratio that people using insulin pens have to calculate to determine their insulin dose. During an educational workshop on counting carbohydrates, we observed a striking degree of simplicity in the approach to achieving a sufficiently precise measure. After introducing a widely known app that monitors the daily intake of macronutrients, but that among people living with T1D is used mostly to determine the carbohydrate values of various foods, the nutritional therapist spread out tiny papers on the table. Some of them contained numbers such as 0, 10, 15, 20, 30, 45, and 60, others were pictures of basic food items: an egg, tomatoes, strawberries, an apple, a glass of milk, pasta, couscous, oatmeal, dark chocolate, dates, and cheese. Participants immediately started animatedly discussing which pictures matched which number. They mostly got it right, and in one case even corrected the nutritional therapist. «A glass of milk cannot have 20 [g of carbohydrates]», Mr Černý exclaimed, throwing his arms around while glucose monitoring sensors peeked out from under his sleeve on each arm. The nutritional therapist double-checked and said, «Oh, you are right». «I can see that you like milk», she smiled, while moving the picture of milk to the paper with the number 10. This logic of counting carbohydrates is imbued with the industrial notion of metabolism where the functioning of the «human factory» depends on calculating the insulin-to-carbs ratio.
21«It was a good practice», one of the participants, Ms Fialová, evaluated the strikingly simple workshop, «it helps you to get a sense of an estimation, so you don’t have to count it or weigh it on a scale». The playful practice of mixing and matching pictures materialises numbers in the more easily imagined form of a handful of couscous or a cluster of tomatoes. This «inner calculator», as Ms Fialová put it, built over months of experience, which included engaging in these playful exercises with pieces of paper, works well to replace the counting apps that may yield more accurate numbers but are not always ready at hand or practical to use.
22Here, counting associated with the factory-like model of metabolism persists, because insulin is rather a dangerous substance and even a playful approach to estimating a dosage has to be precise to avoid fatal consequences. However, as we show below, calculating the insulin-to-carb ratio is far from the only practice that people living with diabetes use technological devices for. We show how, beyond accounting for meals eaten or insulin injected, people use these devices to communicate stress or physical activity with their hybrid closed-loop systems. This practice enables them to better care for their metabolic health, by moving away from the factory mode of metabolism towards the communication network model.
23The coexistence and interaction of high-tech and low-tech devices underscore the complexity of technological advancement in medical and self-care contexts. Older tools and methods remain relevant even as newer technologies emerge. Rather than rendering previous methods obsolete, technological innovations expand the repertoire of self-care methods. Advocating traditional practices, such as the use of a tape measure for obesity or a package of apple purée in the case of hypoglycaemia, does not mean rejecting high-tech devices. It means rather that people should not feel compelled to adopt highly technologised practices at any cost. Innovative tools should be used when appropriate and in situations where they are beneficial and practical.
24Marketing campaigns for wearable technologies and mobile health apps promise users control by being provided with a steady stream of health-related data. This notion of control rests on the assumption that numerical knowledge allows people to be in charge of their bodily processes. However, our observations suggest that for people living with T1D or obesity, control is enacted in more nuanced and variable ways. For some, control means stepping on the bathroom scale each morning to track weight loss progress through a visual graph in the app. For others, control means abandoning the scale altogether, as some healthcare professionals advise, recognising that constant exposure to numbers can lead to frustration, demotivation, and, paradoxically, a loss of control. Rather than a straightforward path to empowerment, self-tracking can sometimes create psychological burdens that challenge the very notion of control that these technologies claim to offer.
25This is why some clinical practitioners – as the opening vignette with Professor Topol illustrates – are sometimes critical of precise tracking and «bookkeeping». Even while meticulously counting her food intake, Ms Šedá can’t control how her more-than-human body, including the gut microbiome, metabolises food. According to recent research on the gut microbiome (ZOE 2024), certain compounds, such as artificial sweeteners, are not metabolised by human cells. This has led to the perception that they offer the benefit of enjoying sweet foods without the metabolic risks associated with sugar consumption. «If the label on a sweet drink says zero sugar, I still wouldn’t drink gallons of it thinking my blood-glucose levels would stay intact», said Ms Kaštanová, one of the nutritional therapists working at the retreat, during an educational workshop about carbohydrates. This statement surprised the participants. Evidence suggests that artificial sweeteners can interact with the microbes in our gut, potentially triggering changes in their composition that may eventually contribute to metabolic complications. While these compounds don’t directly affect human cells, the gut microbiome may still react to them. In this way, the microbiome acts as a kind of «responsiveness hub», transmitting the effects of sweeteners to the human body and potentially causing metabolic complications (Suez et al. 2022). These complications, however, are not «mechanistic malfunctions» but rather a «regulatory crisis» that involves signalling and cueing from the microbiome to human cells.
26Stress is another metabolic signal that can cause regulatory crises in balancing blood-glucose levels. Dr Olšanská said that in her clinical practice, she had been observing how the continuity of data provided by CGM sensors and the predictability the device enables made patients living with diabetes less afraid of hypoglycaemia and thus helped to reduce the overall stress and anxiety they experience in their daily management of T1D. In this regard, Hockin-Boyers, Jamie, and Pope note that for people recovering from eating disorders who engage in self-tracking practices around eating and exercise, calculability and maintaining control paradoxically allow for flexibility and intuition (2024: 1838). That echoes the experience of Dr Platan, who talked about the «distribution of control»: By leaving the closed-loop hybrid system in control to a certain degree, people are able to prevent diabetes from disrupting their sleep. But during the daytime, a continuous stream of data may yield the promise of such total control that it could easily become a burden when the slightest fluctuations of glucose levels cause worries and excessive concern.
27That is what Ms Olivová described while we were walking her dog Cookie. She has been living with diabetes since she was a child and has tried every kind of device to manage her diabetes, from insulin syringes and glucometers to CGM sensors and AI-driven insulin pumps in hybrid closed-loop systems:
I don’t feel anything when I read 11 [mmol/l] on my pump. It’s only at 15 [mmol/l] or more that I start feeling thirsty, tired, and I have a weird aftertaste in my mouth. When you’re [at] 11 [mmol/l], you’re still okay. But I don’t like 11 [mmol/l], it’s too close to 12 [mmol/l], and I’m used to being around 7 [mmol/l]. So, when it’s already been two hours that I’ve been out of range at 11 [mmol/l], I don’t trust the pump if [it] doesn’t offer me any bolus [to compensate the high glucose], and I add one or two units of insulin. Of course, afterward, I end up in hypo.
28In this case, unnecessary stress is caused by the prominence of one very specific number that does not correspond to any particular bodily sensations but is very much associated with embodied emotions of dislike. This stress then leads to a counter-productive action that eventually results in hypoglycaemia. In this case, having continuous access to numbers is at odds with a more balanced form of care that involves feelings of calm and being just okay. Instead, it creates a data burden (Ancker et al. 2015) that can exacerbate «diabetes distress» (Poole et al. 2024) – the emotional burden, anxieties, stress, worries, and frustrations related to the daily management of T1D. Ms Olivová’s ethnographic situation illustrates that the factory-like model falls short in accounting for stress as a significant part of metabolic health. Instead, by centring the active calculation of (in this case) carbohydrates input and output, this model amplifies the metabolic disbalance by establishing a feedback-loop of stress.
29A similar increase in stress caused by the promise of total control through constant numerical monitoring can be observed in the case of obesity. At a workshop dedicated to showing how to maintain a regimen during summer holidays, Professor Topol stated the following: «If a grilled sausage with some salad [at a barbecue party] satisfies a craving, it’s okay [to have it]». This perspective acknowledges that food is deeply entangled with social and experiential dimensions – celebrations, meetings with friends, or travel. While healthcare professionals often see little harm in relaxing dietary restrictions for special occasions, some patients may experience stress when logging these «extra food items» into self-tracking apps, which they regularly discuss with their dietitian.
30Such stress stems from the notion of control informed by the logic of choice as proposed by Mol (2008) – as the individual endeavour of using a product (self-tracking app) in a fixed manner based on the goal of losing weight, isolated from experiences such as social gatherings, which are also a part of health and well-being. Such an enactment of control can furthermore establish distrust between patients and their dietitians – rather than «disappointing» their dietitian, patients hide those extra food items. «If we see that our clients struggle to lose weight and yet their apps contain only rice, lean meat, and cottage cheese, it is obvious that they are keeping something from us», Ms Břízová said, in a sad tone of voice, to fellow nutritional therapists during an educational workshop. Nevertheless, if a patient has «perfect» data but reports no reduction in weight, Ms Břízová sees this as a cue that the patient is struggling with the meal plan. She encourages her fellow nutritional therapists to «read between the lines» and use those cues to communicate with their clients openly about the role played by broader circumstances, such as stress, working night shifts, or the absence of family support, that could be preventing them from adhering to a recommended diet. Here, people living with obesity and their nutritional therapist enact self-tracking apps as tools of communication of larger metabolic relations rather than instruments of accounting for calories within a factory-like metabolism.
31Control is a contested issue in the relationship between doctors and patients. While CGM promotion campaigns highlight the control that doctors can have over their patients thanks to real-time data sharing through applications such as Librelink or Glooko, the doctors with whom we spoke do not use these devices to expand control beyond clinical consultations. Besides not having the time to do so, they do not want to oversee their patients’ every step, as doing so could break the bond of trust established between them. Furthermore, psychologist Dr Vrbová, who specialises in diabetes, pointed out during our interview that such control can often come at the price of mental well-being: «Having “perfect numbers” on your CGM monthly report may not indicate good diabetes compensation, but can in actuality conceal an obsession with and anxiety about numbers that don’t translate into numerical results visible to the doctor». In other words, it is not possible to tell from the graphs in the monthly report whether good numbers are the result of balanced care or of an obsession about numbers.
32Achieving «good numbers» within range may be the result of obsessive control. That is why Dr Vrbová recommends that doctors always ask questions «beyond the data» and let the patients talk, for instance, about their day. Such advice illustrates that even though graphs with curves and daily patterns are a reliable source that can be used to evaluate the quality of diabetes compensation – the source that doctors often draw on during educational courses or negotiations with insurance companies – doctors do not read numbers as all-encompassing indicators of diabetes care during clinical consultations. Instead, they use them as guidance for asking about the circumstances behind the numbers, to see if people are stressed, sad, in a rush, or sick, or if they have their period. Rather than control, doctors strive to establish trust.
33«“Oh, it’s the city council meeting, right?” My diabetologist always recognises this when she sees the high peaks on the graph every Wednesday», laughs Ms Růžová, as we chat after playing a game of table tennis. In this context, our observations of clinical practices echo the logic of care, which implies a «forgiving» (Mol 2008: 20) processual interaction during which technologies are not considered in isolation as a simple indicator of numerical results. Instead, these technologies are enacted as a tool of communication, which physicians can start a conversation with and can learn to read into the complex lived situations of their patients beyond just their carbohydrate intake. This ethnographic situation reveals how stress is being acknowledged as a metabolically significant factor that has considerable effects on blood-glucose variations, hinting at a post-industrial model of metabolism as a regulatory network. It is not only the amount of carbohydrates consumed that elevates glycaemic levels, as the factory-like model of metabolism suggests. Here, stress is a signal of broader metabolic disbalances.
34Both high-tech and low-tech tools enable people with obesity and T1D to gain better control over their conditions and improve self-care. Such control in practice, however, does not mean that one is fully in charge of one’s metabolic processes based on the available numerical data that can be checked anytime. Instead, as we have shown, such a notion of control can become a source of stress and a burden, interfering with other important life experiences and values. According to Ancker and colleagues (2015), the data burden involves emotions of frustration and distress associated with continuous access to (numerical) data on bodily processes and seeing them evolve in what may not be the desired direction. Nevertheless, our research shows that the data burden is also manifested as the elusive promise of «ultimate control», which might be at odds with the practices that are necessary to keep blood glucose levels within range or to follow a balanced diet.
35The control enacted through continuous data streams has its limits, which must be acknowledged. There is always room for unpredictability, whether it’s in the gut microbiome or the contingency of a social gathering. Control should therefore not be deemed the ultimate goal that will necessarily translate into better care or a better quality of life, which are context-specific. Improving one’s quality of life may sometimes mean ignoring a blood sugar number that is outside the desired range in order to avoid stress. In other words, regulate according to the different metabolic signals coming from outside and inside the body.
36Within a metabolic network of communication, control is not achieved by a central control unit – a wilful mind that strives to master the diseased body by using apps and devices as tools of accounting. Instead, it is persistently enacted as distributed, combining technological precision with the volatility of everyday life, and leaving room for being forgetful. Thus, revisiting Mol’s argument about the logic of choice and the logic of care two decades after her seminal work (2008), we zoomed in on the notion of control to explore how it is enacted in different practices informed by the two logics. With technological advancements, particularly in diabetes care, enabling continuous monitoring and steady access to a stream of data, pharmaceutical companies promote those products as tools for gaining control over one’s health and disease. However, in settings where the logic of care prevails, we observed that rather than being fully in the hands of the patient-customer, who would then have to account for every aspect of a chronic condition, control is distributed among people, numbers, measurements, self-tracking devices, physicians, and dietitians. This kind of context-specific and situated control – which is open-ended and ongoing, and which includes letting the loop control at night for better sleep or enjoying a grilled sausage at a social gathering without hiding it from the dietitian – allows for more complex metabolic care that better attends to the messy contingencies of life with a chronic disease.
37By analysing self-tracking practices through the lens of metabolism, it is possible to capture gaps in the paradigm of metabolic accounting, which rests on a factory-like, industrial model of metabolism built on the categories of input and output, and which is limited in its capacity to grasp «regulatory crises». As we showed in this paper, both healthcare professionals and people living with T1D and obesity see factors such as stress, sociality, or mental well-being as relevant for the management of a condition. While self-tracking devices and apps are evolving towards more multidimensional and multifactorial datafication, they are still scripted (Akrich 1992) as devices of calculation and control. However, as our ethnography illustrates, in the hands of patients living with diabetes and obesity, they can become tools of more-than-human communication that enable a truly holistic and «thick» (in the sense of ethnographic thick description) support for metabolic health in its situatedness.
38Recognising the complexity and unpredictability inherent to metabolic processes calls for an integrative approach to care – one that combines technological precision with the volatility of everyday life. Such an approach requires a careful exploration of how embodied and data-driven practices coexist and intra-act, and how they can be done otherwise. We believe that ethnographic data are invaluable in these efforts as they offer detailed insights into the dynamics and interplay between technologies, bodies, numbers, and measurements, ultimately contributing to more personalised care.
Attualità
