Therapeutic value in the time of digital brainwaves

Abstract

This article examines the value of medical technology through the case of electroencephalograms (EEGs), devices used to visualize brain activity and diagnose seizures. Drawing on ethnographic fieldwork, the article shows that EEGs are valued differently by patients and medical practitioners. While practitioners value EEGs for their clinical utility, i.e., ability to inform clinical decisions, patients value EEGs even in the absence of clinical utility. Indeed, patients derive long-lasting therapeutic effects from this diagnostic technology. These findings intervene in the utilitarian calculus of therapeutic value—a mode of reasoning that equates value with clinical utility—commonly deployed in biomedicine and engineering and call for a recognition of alternative notions such as the therapeutic value of being witnessed and cared for by medical experts via EEGs and other technologies that require time to work. Expansive notions of therapeutic value are imperative for including marginalized patients—especially low-income, disabled, and women patients—in debates on automation and the future of healthcare. Studying how multiple stakeholders value a medical technology provides insight into valuation, objectification, expertise, and other concerns central to science and technology studies.

Proponents of automated technology operationalize therapeutic value as clinical utility, holding in high regard health data and practitioner effort with high clinical utility, and considering the rest wasted and inefficient. Clinical utility here refers to an entity’s capacity to inform clinical decision-making. Automated systems seek to optimize therapeutic value by focusing physician attention on data with high clinical utility. For example, neurologists use devices called electroencephalograms (EEGs) to record brain activity and treat seizures in people with epilepsy. EEG data indicative of seizures has the highest clinical utility. Hence, automated seizure detection algorithms present clinicians with short data snippets containing potential seizures instead of having physicians manually review all data, as is conventional (e.g. Boonyakitanont et al., 2020; Cherian & Kanaga, 2022; Fish, 2022). This emphasis on clinical utility is consistent with mainstream approaches in engineering and biomedicine. For example, health technology assessment guidelines state that diagnostic technologies have therapeutic value only when they change the course of treatment (e.g. Guyatt et al., 1986). I call this mode of valuation the utilitarian calculus through which therapeutic value is traditionally conceived.

This article troubles this utilitarian calculus by showing that patients find medical technologies therapeutic even in the absence of clinical utility. The utilitarian calculus fails to account for the therapeutic value of being seen and cared for by medical experts via extended interventions: technologies that, like EEGs, work over long periods of time, in contrast to speedy interventions such as blood draws and physical exams. With some stipulations, this insight on therapeutic value applies to many extended interventions, both diagnostic and therapeutic, including EEGs, fetal monitors, and intravenous infusions. Therefore, I propose that therapeutic value should be conceptualized as the healing potential of an intervention, and that debates on automation and efficiency should account for holistic notions of therapeutic value. This insight applies to both digital and non-digital interventions and requires interventions to have the potential to generate value for both physicians and patients. I define value expansively as the financial, social, moral, emotional, and epistemic evaluation of the worth, desirability, benefit, goodness, or importance of a given entity. Valuation refers to processes, decisions, and articulations around assigning value (Dussauge et al., 2015). I use the term care to refer to the work that medical practitioners perform in caring for patients, encompassing the broad range of physical, emotional, analytical, and administrative aspects of clinical work. ¹

Scholars of science, technology, and medicine have shown that biomedical experts value certain kinds of technologies and ways of knowing over others at a given time and place due to a range of social, financial, and legal considerations: Fetal monitors are prized for their ability to reduce uncertainty and liability (Cartwright, 2013; Davis-Floyd, 1987), prenatal testing spread in response to concerns over public expenditure and individual suffering (Löwy, 2015), institutional cost control efforts influence CT scan ordering (Saunders, 2008), and market-level manufacturer interests drive MRI volumes (Joyce, 2008). On the other hand, patients value medical technology because of physical separation, naturalization (taking disease outside the moral, spiritual, and psychological realms), and the sense of agency facilitated by technological objectification (Cussins, 1996; Davis-Floyd, 1994; Dumit, 2004). The question of how these varied notions of value interact remains unanswered because the patient perspective is underexamined in recent analyses of medical technology (Roberts, 2012). Without examining patient valuations, we risk accepting a future where technology experts limit practitioner involvement to situations with demonstrable clinical utility, ultimately limiting the value of medical intervention.

Drawing on fifteen months of ethnographic fieldwork at an academic medical center in the Midwestern United States, this article intervenes in the utilitarian calculus of therapeutic value by examining how patients and practitioners value EEGs, a common diagnostic test for epilepsy. Epilepsy is a condition characterized by recurrent and unpredictable seizures. EEGs are an infrastructural technology routine in, and integral to, epilepsy diagnosis and treatment. EEG patients and practitioners generate the data and knowledge underlying automated systems such as seizure detection algorithms. Focusing on EEGs thus engages in the infrastructuring of—that is, foregrounding the social and technological background of—automated systems, to develop long-term and relational insights for automation (Bowker, 1994; Khandekar et al., 2021; Marttila & Botero, 2017; Star & Ruhleder, 1996).

The fields of science and technology studies (STS) and valuation studies have long been interested in theorizing value and valuation, with a growing body of work focusing on valuation in biomedicine and technology. This scholarship has developed along three strands corresponding to distinct research orientations: political economy, moral economy, and pragmatism. Political economic approaches examine the production, exchange, consumption, and rhetoric surrounding biotechnologies using biocapital as the central analytical concept and capitalism as the organizing framework (Rose & Novas, 2004; Rose, 2007; Sunder Rajan, 2020b). Moral economies demonstrate that affective and sociocultural values are integral to and constitutive of knowledge production, technology adoption, and expert practice in biomedicine (Daston, 1995; Löwy, 2015; Roscoe, 2015). The pragmatic strand asks how values are made, seeking to undo the divide between economic and sociocultural understandings of value by examining the construction, coordination, and contestation of values in a given context (Dussauge et al., 2015; Fochler, 2016). Pragmatic analyses of the life sciences and biomedicine have shown that valuation is intertwined with and influenced by epistemologies and professional hierarchies. New ways of knowing redefine what is considered worthy of being known. Higher-status groups such as physicians and public health experts make valuation decisions that have trickle-down effects on the practices of lower-status groups, such as patients and technicians (Löwy, 2015; Pinel, 2021; Roscoe, 2015). In the context of biomedicine and technology, patients’ valuation practices remain understudied, with scholarship focusing on scientists, technicians, physicians, and public health experts instead. This article adopts a pragmatic approach to juxtapose patient valuations of EEGs with those of practitioners.

In a related line of work, scholars of disability and chronic illness from feminist STS, anthropology, and sociology have shown that patients value medical technology for multiple reasons. In contrast to scholarship on objectification and medicalization that has portrayed women as either passive recipients of oppressive technologies or ‘inherently suspicious of and resistant to technological interventions’ (Lock & Kaufert, 1998, pp. 1–2), feminist scholars theorize that, given the ‘dearth of social supports for mothering’ (Fox & Worts, 1999, p. 326), low- and middle-class women appreciate the care and social support they receive through fetal monitors and other advanced technologies. Hence, women desire and actively seek medical technology and intervention during pregnancy and childbirth (Cussins, 1996; Davis-Floyd, 1994; Fox & Worts, 1999; Lock & Kaufert, 1998; Thompson, 2005). Further, technological objectification can serve as a vehicle for patient agency (Cussins, 1996; Thompson, 2005). STS scholars have also shown that patients appreciate technologized interactions with medical practitioners because they help patients reach a desired goal, cope with uncertainty, normalize their condition, and address disruptions in hospital wards (Davis-Floyd, 1992; Dumit, 2004; Wheeler et al., 2020). This article builds on and extends these arguments, showing that extended interventions can help patients feel cared for in a system that does not appear to value their time.

This article makes the following contributions: Showing that expert witnessing augments patient agency via objectification, I expand feminist STS arguments beyond childbirth and demonstrate their applicability to EEGs and the larger category of extended interventions. I extend valuation studies scholarship to the study of patient valuations and infrastructural technologies. Patient valuations of infrastructural technologies destabilize taken-for-granted professional valuations and enable critical intervention into debates on automation and efficiency. Further, I develop the notion of temporal economies to account for the many stakeholders of medical technology and bridge pragmatic and moral economic approaches to valuation. Finally, I make practical recommendations to help practitioners and patients acknowledge the therapeutic value of extended interventions.

Background

Electroencephalograms (EEGs) are the primary diagnostic test for epilepsy, a chronic illness and disability characterized by recurrent and unpredictable seizures. Physicians use EEGs to monitor the patient’s brain activity for seizures through electrodes attached to standard locations on the scalp. EEGs sample, amplify, record, and display the electrical activity at these locations through the medium of continuous waveforms (henceforth, brainwaves). The location, morphology, and electrical properties of a waveform provide EEG readers with cues about its clinical significance (Ebersole et al., 2014). EEGs are ordered for durations ranging from 45 minutes to several days, depending on whether the goal is to study brain activity, rule out abnormalities, or to capture specific events. In the latter case, the patient may be deprived of sleep and/or medication to increase the likelihood of recording a seizure. This article focuses on long-term EEGs (henceforth, extended EEGs), where patients are monitored for 24 hours or more.

EEGs are a particularly powerful ‘mode of ordering’ in epilepsy care (Law, 1994). Until the late nineteenth century, epilepsy remained a contested object between neurology and psychiatry (Hacking, 1998). By making brainwaves visible, the development of EEGs helped place epilepsy ‘firmly within the discipline of neurology’ (Jacoby & Baker, 2010, p. 616). Today, EEGs are integral to the habitus of neurologists and fundamental to all stages of epilepsy care. Physicians use EEG monitoring to inform decisions about diagnosis, first-line treatment, invasive procedures, and continuing care. In addition to detecting seizures, EEGs can help localize seizure activity, i.e., identify specific regions in the brain that seizures originate from. Localization carries implications for treatment and prognosis. To date, EEGs are the only technology capable of representing the brain’s electrical activity in a reliable manner, at a reasonable cost, and at a high temporal resolution. EEGs are also a significant source of revenue for neurology facilities in the United States, generating 50–75% more billable work value per unit of time than office visits (Donofrio et al., 2015). Thus, EEGs occupy a central role in the experience, practice, and organization of epilepsy medicine.

Methods and context

This article draws on fifteen months of ethnographic fieldwork in the Midwestern region of the United States, working with twenty-five people with epilepsy and thirty-six epilepsy care practitioners. The bulk of this fieldwork took place in the comprehensive epilepsy program of a large research and teaching hospital that I shall call hospital X. Hospital X is an academic medical center that regularly ranks among the nation’s best hospitals. Its comprehensive epilepsy program is accredited by the National Association of Epilepsy Centers as a level four referral facility offering the highest standard of specialized medical and surgical evaluation and treatment for epilepsy, seeing over 3000 patients each year.

My fieldwork at hospital X consisted of 355 hours of participant observation, 34 formal interviews (64 minutes each on average), and countless go-along conversations with people with epilepsy, physicians specializing in epilepsy (henceforth, neurologists ² ), EEG technicians (henceforth, techs), and associated care practitioners. Hospital X’s neurology department operates a 24/7 EEG service that conducts, interprets, and reports on EEGs ordered by the hospital’s outpatient and inpatient neurology services, by other departments, and via external referral. The EEG service is based in one section of one floor of the hospital called the EEG lab. The EEG lab consists of eight rooms: four EEG prep rooms, two office-like rooms that each serve as shared workrooms for EEG techs and administrative support staff, a conference room, and a storage room for EEG supplies. Techs use EEG prep rooms to connect EEGs to patients undergoing extended EEG monitoring. Once prepped, techs move patients to private inpatient rooms located on other floors of hospital X. Offices for attending neurologists (physicians responsible for the care of neurology patients and for the training and supervision of trainee physicians and non-physician practitioners) and a shared workroom for neurologists-in-training are in a staff-only area a passage away from the EEG lab. I used the EEG lab as the focal point for my hospital-based fieldwork and paid particular attention to the work practices of ten epilepsy fellows—post-residency neurologists undergoing a year of intensive training in epilepsy.

I conducted semi-structured life narrative style interviews with 21 adults with epilepsy (72 minutes each on average), all of whom had undergone one or more EEGs. I also conducted four interviews with patients who had recently undergone EEG monitoring at hospital X (69 minutes each on average) and observed neurologists evaluate their EEGs. I used a combination of snowball, purposeful, and theoretical sampling through different phases of fieldwork, trying to prioritize diversity in age, gender, race/ethnicity, and income when possible (Bryant & Charmaz, 2012; Denzin & Lincoln, 2011). Field data were analysed through an iterative and inductive process consisting of regular analytic memoing and situational analysis, an interpretivist methodology that accounts for human and non-human stakeholders and the analyst’s partial and situated vision (Clarke, 2004; Haraway, 1988). I used data and methodological triangulation, member checking, long-term involvement, and mentor feedback to improve the depth, rigor, and validity of my findings (Creswell, 2007; Emerson et al., 2011; Flick, 2012). All names are anonymized for confidentiality. Quotes are lightly edited for readability and refer to me as the ethnographer. All research was approved by the research ethics review boards attached to hospital X and my institution.

EEG and idle time

This section illustrates how medical practitioners value EEGs through scenes from the EEG lab.

I was to meet tech Amelia in hospital X’s EEG lab at 10 am to ask her patient Olivia if she would interview with me. At 7:49 am, Amelia emailed, saying that Olivia had already arrived for her EEG because it was the only time that she could get a ride to the hospital (Olivia is not allowed to drive, due to seizures, and public transit is sparse in this part of the Midwest).

By the time I arrived, Amelia had already connected Olivia to an EEG machine. She led me to the prep room where Olivia waited to be assigned to an inpatient room. I tentatively stepped towards a raised hospital bed oriented to face away from the door. Lying in bed was Olivia, a woman with neatly braided brown hair that made the white wrap used to secure EEG electrodes stand out. Also present in the room were her mother and mother’s boyfriend.

When I asked Olivia if she would like to interview with me, she said, ‘sure, yeah, that’s fine’, with a weak smile, waving aside explanations about the voluntary nature of her participation and my lack of involvement in her care. ³ I extended a consent form towards Olivia, but she turned to her mother, who filled out the form. As Olivia signed, I noticed that her left arm was covered by a blanket and held a little stiffly, as though she were afraid of losing control over the arm (as can happen during a seizure). Olivia was undergoing EEG monitoring to have hospital X’s epilepsy specialists assess the frequency and severity of her seizures, at her neurologist’s suggestion. This evaluation was to be conducted over 24 hours, extending to 48 hours if she did not have seizures on the first day. Olivia and I agreed to get in touch after her EEG to schedule our interview.

I walked to the tech room a few steps away, excited and relieved to have Olivia participate in my research. Amelia was at her computer. I said, ‘I’m done. That was fast!’

Amelia:

Oh, you’re done interviewing?

Ethnographer:

No, the interview will take about an hour, so I’ll do it another day.

Amelia:

I mean, you can do it any time.

Ethnographer:

During the EEG?

Amelia:

Yeah, we don’t care.

Ethnographer:

But I don’t know who the admitting doctor is, and uh, I don’t want to step on anyone’s toes.

Amelia:

Trust me, they’re not going to care. ⁴

According to Amelia and other clinical interlocutors, the time that Olivia spends wired to an EEG is idle time—time waiting to be filled up, most ideally in service of biomedical research. Practitioners are compensated based on EEG duration (irrespective of seizures) and thus all EEG time generates baseline economic value for practitioners. However, idle time fails to advance the treatment, education, and research missions specific to academic medicine.

For practitioners, EEG monitoring time is valuable when it provides clinically significant information. Clinical significance is achieved primarily when patients have new types of seizures during monitoring. For new patients, all seizures are significant because they help practitioners solidify or rule out an epilepsy diagnosis, with implications for treatment. For patients with known epilepsy, new or previously undocumented seizures—such as generalized seizures for a patient diagnosed with focal epilepsy—are valuable because they alter the course of treatment. Patients with rare types of seizures or rare types of epilepsy additionally generate information of relevance to research, due to the possibility of publishing case reports in medical journals. The teaching mission of academic medicine benefits from seizures as well, because seizures provide medical trainees with opportunities to learn and practice diagnostic and treatment skills. That is, practitioners value capturing seizures on EEG, due to their usefulness in treatment and prognosis, medical education, and publishing, all of which generate value for individual practitioners and for hospital X. They prize new and rare seizures most of all, because they contribute to the hospital’s reputation as a premier academic medical center with high-quality care, instruction, and research.

Expert valuation of seizures over idle time is visible throughout epilepsy care. For new patients with possible seizures, physicians in hospital X’s emergency department (ER) order short 30-minute EEGs that may convert to extended monitoring based on the neurology team’s evaluation. Neurologists recommend extended EEGs for ER patients only if seizures are recorded during the short EEG. Similarly, most pre-surgical case conferences—multidisciplinary meetings of neurologists, neurosurgeons, neuropsychologists, and social workers to discuss the merits of invasive treatment for patients whose seizures do not respond to medication—are spent reviewing and discussing seizures recorded on EEG, whereas EEG time outside seizures is skipped entirely. This focus on seizure EEGs extends to training opportunities like didactic sessions and grand rounds, as well as research seminars and conferences.

The practitioner mode of valuation thus regards EEG monitoring as valuable primarily when it has clinical utility. We will see later that practitioner valuations fail to account for several important effects of EEG monitoring for patients.

Practitioner work during idle time

The devaluation of idle time by EEG practitioners occurs in a hospital context where the devaluation of patients’ time is a routine occurrence. But EEG practitioners’ devaluation of idle time does not extend to other parts of the patient experience; indeed, EEG practitioners actively work to reduce wait times for their patients.

Four hours after meeting Olivia, I sat with epilepsy fellows Betty and Jen as they wrote reports for the morning’s EEGs. Fellows spend most of their day evaluating EEGs in the EEG reading room. In practice, this means poring second-by-second over days and even weeks of brainwave recordings. Twice a day, fellows discuss EEG interpretations with the attending (the neurologist in charge) and write formal long form reports for the patient’s electronic health record.

During a brief lull, I asked Betty if she could tell me about Olivia to help me prepare for our interview. Opening Olivia’s EEG file, Betty said, ‘Oh you mean in terms of the EEG? I’ve looked at a little and so far, normal. I’ve already read part of it so this will be really quick.’ Betty scrolled through Olivia’s EEG at a fast pace: 10-12 right arrow clicks per second; each click taking us 4-5 seconds forward in the EEG. Almost without pausing, Betty added a label: ‘vertex?’ to highlight a brief period of V-shaped waveforms and shortly after, ‘vertex’, again. Betty played the accompanying video for a brief second and saw Olivia lying in bed, head to one side and eyes closed, looking sleepy or asleep. Her state of drowsiness confirmed that the vertex patterns were not cause for concern. Then, Betty added the label ‘ss’ for sleep spindles. I recognized the spindles a little later in the EEG: sinusoidal waves set close to each other, resembling a coiled spring. At some point, the screen abruptly turned black: There were lots of high-amplitude clusters of vertical lines in the waveforms. Betty scrolled slowly for a second, eyes on the waves, but then rushed forth again past many such screens, without playing the video. This flurry of black screens is typical of artifacts caused by patient movement such as walking. Having reached the end of the available EEG, Betty added a label ‘r-B’, to indicate to future readers that she had read up until that point in time. Turning to me, Betty said, ‘okay, so, [Olivia] has some sleep things but otherwise nice, normal stuff.’

EEG practitioners enact therapeutic value through time. Specifically, the amount of time practitioners spend on a screen of EEG data corresponds to their valuation of that portion of data, spending greater amounts of time evaluating data with potential clinical utility. Betty’s motions of speeding up and slowing down through Olivia’s EEG illustrate this expert construction of therapeutic value: Betty sped through the ‘nice, normal’ data carrying little utility for medical treatment, education, and research. Betty slowed down first on noticing vertex waves and then on seeing high-amplitude clusters of waveforms, because both can be clinically concerning patterns. After establishing that the patterns were not clinically relevant, Betty sped up through the rest of Olivia’s EEG. Tracing the rhythms of EEG reading thus reveals how practitioners construct therapeutic value from brainwaves.

Betty read over 60 minutes of Olivia’s EEG in about four minutes, or 1/15th of the EEG duration. As I thanked her, Amelia knocked on the door and said that ‘the ten o’clock patient’ (Olivia) had not yet been moved to an inpatient room. It had taken two hours longer than expected for the previous patient to vacate the room, and now that the room was ready, the charge nurse was refusing to admit Olivia. ‘Why?’, inquired Betty. Amelia explained that Olivia’s mother and mother’s boyfriend wanted to leave as soon as she was moved to an inpatient room, and the charge nurse was adamant that a pediatric patient could not be admitted without a caregiver present. Amelia had tried explaining that Olivia was an exception: she was 20 years old ⁵ and a ‘reliable history-taker’ (her primary neurologist said that she could be trusted to accurately report her seizures and to alert medical practitioners when a seizure began). However, the charge nurse refused to budge, with the implication that Olivia’s EEG would have to be canceled and rescheduled—after the techs having started her EEG four hours earlier.

Despite a very busy day, Betty consulted three attending neurologists (the EEG attending, the neurology wards attending, and Olivia’s neurologist), and called the charge nurse to request that she admit Olivia, adding that the wards attending would ‘personally make a phone call to approve’. The nurse relented and Olivia was moved to an inpatient room after a five-hour wait. Betty told me that when an attending offers to call the floor, ‘such things are usually a given’. Attending physicians are at the top of the clinical labor hierarchy and usually delegate to trainees the work of coordinating with nurses and techs. Attendings also make decisions influencing the work of other practitioners, for instance writing orders that nurses, techs, and pharmacists fulfill. The attending’s offer to personally call is thus a signal to the nurse that this request is important enough to set aside usual protocols. Recognizing that a neurologist’s time, labor, and words mean more and go farther than those of techs and patients, EEG practitioners regularly campaign to reduce patient wait times by negotiating to obtain rooms and appointments.

This section provided a glimpse into the practitioner experience of EEGs. This includes the work of sifting through, evaluating, and reporting on brainwave data, as well as the work of interprofessional negotiation across medical divisions of labor.

Being seen and valued on EEGs

My patient interlocutors valued EEG monitoring in ways that far outweighed its clinical utility. This value arose from a constellation of factors, including social, material, epistemic, and ontological aspects of EEG monitoring that remained with patients long after the procedure.

EEG monitoring is a period of high anxiety for patients, because they undergo this procedure for the express purpose of having seizures (on the route towards diagnosis and treatment). Seizures are unpredictable, recurrent, and difficult to experience. Patients must also take time off school, lose income, and arrange for care and transportation to undergo monitoring. In such a context, expert witnessing is therapeutic for patients. By expert witnessing, I refer to the ways that medical practitioners watch over, attend to, respond to, and acknowledge the patient during EEG monitoring. Annie said,

they stayed in the room with me and made sure I came out of the seizure fine. I was like, ‘Wow, okay.’ It was helpful for me to have them there with me, and then when they told me this is what was going on, I can accept it. They didn’t just abandon me and tell me I was making it all up.

Expert witnessing is important throughout EEG monitoring, while patients await seizures, during seizures, and while recovering from seizures. Outside the immediacy of seizures, patients worried constantly that their brainwaves showed seizures, particularly because some types of seizures resemble common bodily sensations such as muscle spasms from exhaustion. Amira noted the double-bind of being sufficiently EEG-literate to be concerned but not literate enough to be certain about one’s worries:

I see my brainwaves on the [display] and I think … I’m wondering, am I having a seizure or am I not. Is there anything going on? I don’t know exactly how to read it. … If I see something that looks kind of weird, I get anxious. I will be like ‘Oh my god, is that a spike?’ I will be too anxious to ask, and I will keep reading and getting anxious for the next day to come to see if [doctors] say it is seizure activity. … I don’t know how to read it. I’m not a neurologist. Maybe I just saw the wrong thing—maybe it’s normal because I was walking and my brain was just getting active when I was moving.

Amira found it comforting that techs and neurologists were watching over her during this challenging time: ‘they help me. They take care. They take the time to watch me in the cameras. … They don’t want to rush through things, which is really good. It’s good that they take time.’ Analyses of cardiac monitoring and other technologies of time have shown that although these technologies can enable patients to feel cared for, expert monitoring can also make patients feel surveilled and experimented upon (Greene, 2022; Oudshoorn, 2015; Pollock, 2008). None of my EEG patient interlocutors expressed such concerns, calling for reflection on the approaches taken by various biomedical specialties to making monitoring productive, and in the case of EEGs, agentive. As the rest of this section will show, EEG practitioners use several ways to create relational opportunities for connection to counter the uncertainty and isolation of epilepsy via monitoring. In contrast to other medical specialties that value speedy action given the conditions they treat, epilepsy practitioners adopt a ‘wait and watch’ ethos more suited to the recurrent and nature of epilepsy. This ethos is articulated in extended monitoring, as Amira noted, ensuring that EEG patients do not feel rushed.

Not being rushed is a rare experience in medical settings. Jill equated the pace of typical doctor’s appointments with express checkout lanes: ‘[W]henever I go to the [doctor] it feels like I’m going through the quick checkout at a grocery store. I want to talk more sometimes, and it just feels like I am not getting that time.’ That is, patients are used to medical practitioners not having or making the time to attend to and acknowledge their experiences. This makes the extended witnessing that patients experience during EEG monitoring that much more special.

Patients are also used to being ignored or dismissed in generalist and emergency care settings, particularly when their seizure symptoms differ from the mainstream imaginary of grand mal or generalized tonic clonic seizures (that involve whole-body spasms and a loss of consciousness). Zainab complained, ‘it’s happened several times where I’m in the ER and I’m like, “I’m having a seizure,” and they are like, “that’s not a seizure,” because it’s a partial seizure. They only know tonic clonic [seizures].’ In contrast, EEG practitioners are trained to recognize diverse seizure symptoms, as professionals who specialize in epilepsy. Leah echoed Zainab, adding, ‘I have a memory of describing an episode. [The generalist] just looked at me like I was crazy. … I really only feel sane when I’m seeing an epilepsy specialist.’ Patients were comforted when EEG practitioners took the time to acknowledge and validate their experiences, as Isabel shares: ‘It was comforting to know that you for sure had a seizure, … that all the stuff that I’m feeling is for a reason. I’ve had no answers for so long, that just having answers now, I’m comfortable with that. It’s not a mystery anymore.’

Part of this comfort, and an important function of EEG monitoring, comes from reinstating a nature/culture boundary and thereby helping patients become comfortable with epilepsy. Wendy, for example, experiences a strong sense of déjà vu during her seizures. That is, she feels as though she can predict what is about to happen around her, having already experienced it in the past. EEG monitoring helped Wendy delink her déjà vu seizures from clairvoyance and mysticism, and redefine epilepsy as a natural tendency of her brain:

After you’re diagnosed on EEG, you realize that [the seizure] isn’t clairvoyance. This is your brainwaves doing a funky dance up there. My term for it is a brain arrythmia. If you say seizure, to a lot of people, there’s still a fear of that word. I try and get away from that and say that brainwaves are doing a funky dance. It’s a brain arrythmia. I look at it like people have heart murmurs, people have brain arrythmias.

By locating pathology within the brain and making brainwaves visible, EEG monitoring places the patient’s seizures firmly within the realm of the natural, and dissociates them from the sociohistorical connotations of deviance, divinity, possession, and moral failure that accompany the terms seizure and epilepsy in mainstream culture.

Extended EEGs helped legitimize an epilepsy diagnosis for several interlocutors. As devices that visualize brain activity, EEGs can establish the organic etiology of a patient’s seizures, in effect arbitrating the boundary between epilepsy (a neurological condition) and psychogenic non-epileptic seizures (PNES)—a psychiatric condition that mimics epilepsy. Epileptic seizures usually produce abnormal EEG patterns while PNES do not, although this distinction is not quite so neat in practice: People with epilepsy can have normal EEGs and people without epilepsy can have abnormal EEGs (Binder & Salinsky, 2007; Institute of Medicine, 2012), showing that the boundary between neurology and psychiatry remains blurred. For Annie, extended EEG monitoring led to an epilepsy diagnosis after a long struggle with a PNES misdiagnosis. Annie recounts:

they weren’t able to catch my seizures on any of my short EEGs. So, for the first two years after I had seizures, they told me these are non-epileptic seizures. That was frustrating, just feeling like I was totally by myself and had no one to speak up for me. … [The EEG duration] wasn’t long enough for my brain to show what was going on.

In such cases, extended EEGs helped patients prove that they are physically ill by displaying the neurological basis of seizures, thereby distancing patients from the stigma associated with psychiatric conditions. Patients were comforted by this ontological shift facilitated by EEG monitoring (‘it was just a huge relief to know, to have an explanation, and to not be crazy’, per Elie). At times, epilepsy patients develop PNES due to the trauma of epileptic seizures and other stressors; here too, interlocutors found EEGs helpful because ‘they were able to prove that there’s physical evidence that it’s happening, and they were able to differentiate between the two [types of seizures]’ (Leah).

Extended EEGs also provide a forum for patients and caregivers to learn about managing epilepsy from expert practitioners. Seizures can be traumatic to experience and witness, and so patients and caregivers are often overcome with panic and fear during seizures. Franco and his wife learnt about seizure first aid through the EEG team’s calm and watchful approach:

All of a sudden, someone’s standing in the monitoring room with you and you’re saying, ‘What’s the matter?’ ‘Well, you’re about to have a seizure.’ Then, as you snap out of [the seizure], they’re holding down your arms and making sure that nothing happens. Then you snap out of it and you’re back to normal and they say, ‘Are you okay? Count how many fingers I’m holding up. Everything good?’ They come check on you in five or ten minutes and say, ‘Okay, everything good?’ ‘Yeah.’ ‘All right.’

Expert witnessing during EEG monitoring thus provides the time, space, and opportunity for patients and caregivers to learn about seizure response and recovery. Learning about seizure response and recovery is particularly vital for patients who have newly begun experiencing seizures and for the 30% of epilepsy patients whose seizures do not respond to medical treatment (World Health Organization, 2023). Expert interactions during EEGs are thus focused on seizures, in contrast to other interventions like dialysis where experts address machine failures and other concerns secondary to chronic kidney disease (Hamdy, 2008; Russ et al., 2007).

Additionally, during EEG monitoring, patients gain access to material and social comforts that are ordinarily inaccessible, particularly to women, disabled people, and low-income people, and this helps patients feel a sense of agency and control in a system that does not appear to value their time. For instance, although it is common for EEG patients to wait two hours before being admitted to an inpatient room (Olivia had an exceptionally long delay), my patient interlocutors made no mention of the delay. Instead, they complimented hospital X’s care team and facilities.

For Amanda, a working mother, EEG monitoring serves as time away from the familial responsibilities and chores that usually occupy her time and energy:

for me, it’s almost like vacation. You bring your computer, you can catch up on Netflix and everything else. For me, it’s almost relaxing. Isn’t that terrible to say? But it’s true. Granted, they keep you up, they take you off your meds, and sometimes they don’t get great data. They never really get good data from me when I’m on [EEG]. … It’s a very different environment. I’m not making dinner every night, making sure the dog is fed, making sure the dog’s taken out and walked, making sure the laundry’s done. It’s a very unrealistic setting for an adult woman, frankly. … usually, I’ve got a teenager, a household, and a puppy to take care of.

EEG monitoring provides Amanda and other women with free time, a rare comfort given the gendered nature of domestic and familial labor. Women value this time of leisure even when deprived of sleep and anti-seizure medication (to induce seizures) and even when monitoring does not yield clinically useful data.

Further, EEG monitoring is set up to give patients control over food, visitation, and activities. This is rare in institutional contexts where patients rarely have agency in these matters, as feminist and disability rights activists and scholars have shown (Davis-Floyd, 1987; Hingsburger, 2013). EEG patients can order freely from the inpatient meal service menu, bring their own food and drinks, and have visitors at any time during the procedure. Cathy explains:

You had your own hospital room, and you had a little sofa chair to sit on. You can watch TV. In my case, I took my laptop computer, I could play games. … They would let me have visitors at any time of the day 24/7. They let me take my own snacks. I was able to take Pepsi and I had a little fridge in my room to have it in. I’m a chocoholic and I was able to have candy to snack on.

EEG monitoring also provided patients an opportunity to bond with loved ones and indulge in rare pleasures. For Alex:

My older sister would stay up with me all night long and we would drink Dr Pepper. We are not allowed to have pop in my family. It is very, ‘do not drink sugary anything.’ … So, I was like, ‘oh, this is awesome.’ We would make sand art. … I felt so cool because I got to hang out with my sister, and I’m the pesky little eight-year-old that nobody wants to be around.

For Isabel, EEGs were occasions to hang out with her father: ‘They had an Xbox in there, so my dad and I played video games. We would binge watch some shows. … Buttered popcorn usually throws me into [a seizure], so, they delivered to my room, buttered popcorn.’ Thus, EEG monitoring in hospital X is set up to make patients feel a sense of agency.

The generous spatial and technological configurations of EEG patients’ rooms add to this sense of agency, particularly for low-income patients. Per Amira: ‘The speakers were really cool. … The bed was cool to play with and adjust. The view was really nice outside. I liked the room, it’s really nice. Outside the windows was really nice too—big and wide. When it’s nighttime, the lights shine.’ Living in a thoughtfully designed space and interacting with new technologies helped Amira find pleasure in the weird ⁶ experience of being wired to a machine and confined to a room for extended periods, despite the perceived expansion of time (‘I’m only there for a day, but it makes me feel like I’m there for a week’).

Patients undergoing extended interventions do not always receive private rooms and control over food, visitation, and activities. For example, dialysis is conducted in shared spaces that do not permit food, drink, or visitors. A confluence of technological needs and disease characteristics makes these comforts possible in EEG monitoring: Private rooms are thought to improve data accuracy by reducing electrical interference and noise, and epileptic seizures are thought to be unaffected by food, visitors, and other external stimuli (Schachter, n.d.; Tatum, 2014).

EEG practitioners also make patients feel comforted and valued via small but notable gestures. Rylie, who was initially apprehensive when ‘they put a bunch of sticky electrodes on me for the EEG’, felt better when the tech ‘gave me a teddy bear and put some stickers on the teddy bear’. Alex ‘got a little pillow that said, “you’re a star patient”’ when they underwent EEG monitoring as a child. Similarly, Annie, who lives alone, appreciated that staff facilitated relational opportunities for connection: ‘people came by with a service dog and asked, “Do you want to pet the dog?” And so that would be a half hour of my time. Or just different people that came around with different things. I really appreciated that. It was like, “Wow, human contact.”’

Finally, patients are made to feel that they are contributing something of value by undergoing EEG monitoring. In addition to asking patients to participate in research studies, this sense of value is reinforced by teaching patients that confirmatory EEGs provide valuable information. Amira, for example, stopped having seizures after a change in medication six months earlier. Her neurologist ordered an EEG to verify that she really was not seizing. As expected, Amira did not seize during the EEG. Despite the procedure’s weirdness and inconvenience, to Amira the EEG was ‘worth it to get the information’. If Amira knew she was not having seizures, what new information did the EEG provide? ‘It showed [my neurologist], okay, it looks like the medication is working.’ Thus, neurologists teach patients to value the EEG’s ability to verify patient testimony even when doing so does not change the course of treatment.

To summarize, extended EEGs have long-lasting therapeutic effects for patients irrespective of clinical utility. Short and outpatient EEGs did not evoke similar narratives from my interlocutors. Access to expert witnessing and other comforts facilitated by extended monitoring are thus key to patient valuations of EEG.

Discussion and conclusion

Patients value extended EEGs beyond their clinical utility. Specifically, patients derive multi-dimensional therapeutic effects separate from the EEG’s intended diagnostic purpose. Given the uncertainty, trauma, and stigma of seizures, patients value having medical experts witness and care for seizures in a comfortable environment that offers opportunities for connection. This therapeutic value remains with patients far beyond the duration of the procedure and helps them reorient to life with epilepsy. In contrast, for medical practitioners, the primary value of EEGs lies in recording seizures because normal EEGs rarely generate information of significance for medical practice, research, and teaching. This section explores the theoretical and practical implications of these findings: First, I examine the transferability of the insight on the value of extended EEGs to other technologies. Then, I discuss the article’s contributions to scholarship on valuation. I conclude with practical recommendations to acknowledge the therapeutic value of extended interventions.

Therapeutic value beyond epilepsy and EEGs

Patients derive therapeutic value from the extended technologized care of EEG monitoring. This insight applies to the broader category of extended interventions with some stipulations.

As I found in the case of EEGs, sociological analyses show that expert witnessing during extended diagnostic and curative interventions comforts patients irrespective of clinical utility (Davis-Floyd, 1992; Wheeler et al., 2020). These valuations are attributed to three important functions of expert witnessing: helping patients cope with uncertainty around their condition (such as contractions in childbirth), addressing annoying or disruptive technologies (such as loud alarms from infusion pumps attached to patients and their ward-mates), and helping patients reach a desired goal (such as an uncomplicated birth). Fetal monitors—devices used to monitor uterine activity during childbirth—present a close parallel to EEGs, sharing several characteristics including their use of waveforms to represent physiological activity, their ability make hidden worlds (wombs and brains) accessible to practitioners, their pervasive presence in the relevant specialty (obstetrics and neurology), and the extended nature of monitoring. As with EEGs, patient valuations of fetal monitors are consistently positive: More than 70 of the 100 women interviewed by anthropologist Robbie Davis-Floyd valued fetal monitors, despite clinical literature showing that fetal monitors do not improve birth outcomes (Cartwright, 2013; Davis-Floyd, 1992). Patient valuations of intravenous infusions and other curative interventions are similarly positive (Wheeler et al., 2020). This article confirms these findings and extends their applicability to diagnostic and curative extended interventions.

In palliative extended interventions, on the other hand, patient valuations are ambivalent and contingent. Patients do not ascribe therapeutic value to palliative interventions when patient and practitioner goals are misaligned and when expert witnessing does not address the underlying condition. In such cases, patients question the intervention’s therapeutic potential, and the intervention is valued only by patients for whom its positive effects outweigh its negative effects. Analyses of cochlear implants present an example of misaligned treatment goals between patients and practitioners: Whereas Deaf people and caregivers define treatment success based on the patient’s ability for social connection, language (including sign language), and emotional health, practitioners focus narrowly on the ability to produce speech; as a result, cochlear implants remain a contested topic in the Deaf community (Blume, 1997, 2000; Mauldin, 2019). In other extended palliative interventions, such as cardiac and nerve stimulation implants, in-vitro fertilization (IVF), respiratory support, and dialysis, expert witnessing functions differently: Patients meet practitioners during brief clinic visits and expert interactions address something other than the underlying condition. In dialysis, for example, experts intervene primarily to address concerns secondary to chronic kidney disease, namely machine failures and cannulation pain (Hamdy, 2008; Russ et al., 2007). When expert witnessing is short and not centrally about the condition, there is considerable variation in patient valuations, with patients valuing interventions primarily when their positive impacts outweigh side-effects (Cussins, 1996; Dalibert, 2016; Kaufert & Locker, 1990; Kaufman, 2015; Perrotta & Hamper, 2021).

Dialysis machines—palliative interventions used to manage chronic kidney disease—present an interesting counterpoint to EEGs, because patient valuations of dialysis decline in proportion to the length of time they spend connected to the machine (Hamdy, 2008; Russ et al., 2005, 2007). Beyond a point, patients see dialysis as a frustrating process that reduces their quality of life and drains time and resources, rather than a life-preserving therapeutic. Three key differences between EEGs and dialysis machines explain the difference in patient valuation: the intervention’s ability to advance the patient towards a defined goal; the nature of expert witnessing; and intervention-enabled modes of being. Dialysis is a maintenance therapy without a defined endpoint, whereas EEGs are conducted with the explicit goal of recording seizures. Epilepsy patients see longer EEGs as more beneficial because they advance patients towards this goal. In contrast, dialysis takes patients away from the goal of improved health: Patients ‘know that the more time they spend on dialysis, the sicker they are getting, and the less they will benefit from a kidney transplant’ (Hamdy, 2008, p. 558). Further, expert interactions with dialysis patients address secondary concerns like machine failure and pain at the cannulation site (Hamdy, 2008; Russ et al., 2007). That is, expert witnessing during dialysis is not directly related to kidney disease—it is as though machines and cannulas receive greater expert attention than patients and their chronic condition. Thus, dialysis provides patients with fewer opportunities than do EEGs to feel seen or witnessed as a kidney disease patient, engage in sensemaking, or learn about managing their condition. Next, EEGs are conducted in private rooms where patients can eat, drink, invite visitors, and move with some degree of freedom. Dialysis is conducted in shared and semi-private spaces where patients are not allowed food, drink, or visitors, and patients are required to sit or lie down throughout the procedure. Thus, patients have limited use for the comforts and time away from responsibilities afforded by dialysis. Taken together, dialysis patients receive neither disease-related expert witnessing nor the space and time necessary to benefit from extended interventions. As a result, patient valuations revolve around dialysis-imposed restrictions on their ability to participate in personal, social, and professional activities. Dialysis, then, serves as a boundary case that sets the limit for the therapeutic value of extended interventions: we should expect an intervention’s therapeutic value to decline once its life-sustaining potential becomes less salient than its imposed limitations on patients’ lives.

Blood transfusions present a different kind of boundary condition where patients do not experience therapeutic value from technology. Patients do not notice anything special about the care they receive during transfusion, although they do appreciate the attention and reassurance that transfusion nurses provide (Adams & Tolich, 2011; Fitzgerald et al., 1999). This attenuated valuation is likely due to two reasons: expert witnessing is already intensive prior to transfusion and transfusion is a relatively short intervention. Transfusion patients are usually severely ill: For example, transfusion is indicated in cancer treatment, major surgery, and emergency care. As a result, patients are under intensive expert monitoring even prior to transfusion. That is, expert witnessing during transfusion seems like a mere continuation of the intensive care patients were already receiving. Hence, patients may not notice a difference in care during transfusion. Further, transfusion is a relatively short intervention, lasting up to four hours, and this may be too short a period for patients—particularly patients who are already severely ill—to experience noticeable therapeutic value comparable to that of extended EEGs. The case of transfusion suggests that patients do not notice the therapeutic value of extended interventions without a marked and extended difference in care and context.

To summarize, the therapeutic value of extended interventions is predicated on the following conditions: Expert interactions should help patients learn about their condition (e.g., reduce the uncertainty of epilepsy), the setting in which the intervention is conducted must offer patients the possibility of privacy, pleasure, and a change of context (e.g., patients can eat as and when they want), and spending more time on the machine must increase the likelihood of attaining a defined goal (e.g., fewer seizures). Even with these provisions, patients are unlikely to benefit from the expert witnessing and change of context facilitated by extended interventions when: The patient is already under intensive care (e.g., transfusion), the patient is well-adjusted to their condition (e.g., a patient at comfort with the uncertainty of epilepsy), the patient’s survival depends on familiarity and routine (e.g., routines are vital for some neurodivergent people), and practitioners fail to sufficiently attend to patient experiences and concerns (as can happen in overburdened settings).

My conceptualization of therapeutic value is not captured by patient effects alone; it requires interventions to work. However, what it means for an intervention to work may not align with normative expectations around utility, accuracy, and outcomes. Extended diagnostics must have the potential to generate data with clinical utility, although this potential does not need to be actualized in every instance. For example, my physician interlocutors valued extended EEGs for providing reliable data that helped make decisions in the face of uncertainty. Instead of abandoning EEGs because of their limited accuracy, physicians learned to account for their limitations. These expert valuations are mirrored in other extended diagnostics such as cardiac and fetal monitors (Cartwright, 2013; Stonington, 2020). For patient interlocutors, the potential of EEG monitoring to help treat their own seizures, naturalize their epilepsy, and improve care for future patients, was important to their valuations of EEGs.

My proposal on therapeutic value applies equally to digital and non-digital interventions. The push towards digitization in medicine and technology heightens the urgency of expansive notions of therapeutic value because digitization sets the stage for automation and valuation is inextricable from automation: In healthcare, automation requires deciding when the labor of physicians and other human experts generates value over waste and inefficiency. Expansive notions of therapeutic value are imperative to including marginalized patients—especially low-income, disabled, and women patients—in debates on automation and the future of healthcare.

Patients, infrastructural technologies, and temporal economies

Uncovering a mismatch between patient and practitioner valuations of EEG monitoring, I make three contributions to scholarship on valuation: First, I show that centering patient valuations of technology destabilizes well-established expert valuations. Second, I show that analyses of infrastructural technologies enable critical intervention into current debates on automation and efficiency. Third, I contribute a methodological proposal for studying valuation practices.

By putting patient valuations in dialogue with expert valuations of technology, I demonstrate one way of developing critical interventions that can be translated to medical and technological practice to remedy the absence of patients in decision-making around technology. Centering patient valuations adds a crucial missing element to valuation studies of technology in medicine and the life sciences, and calls into question the validity of well-established expert valuations like idle time.

Further, I demonstrate how current conversations on automation and efficiency in healthcare are rooted in values embedded in routine and infrastructural technologies and practices. Focusing on EEG monitoring (instead of novel seizure detection algorithms, for instance) helps me show that the utilitarian notions of value underlying automated systems re-enact pervasive beliefs around value and waste in neurology. Importantly, these beliefs do not always align with practice: EEG monitoring practices help patients derive therapeutic value during idle time. Thus, in addition to uncovering the stakes and socio-technical background of automation, my focus on infrastructural technologies presents an opportunity for critical intervention into debates on automation and efficiency, e.g., making explicit the therapeutic value of idle time in extended technologized care.

Next, I present a methodological proposal to develop what I call temporal economies of medical technology. I propose that time is a particularly useful lens to access the complex arrangement of values in the practice of western academic medicine. Attending to how various stakeholders value time reveals intertwined social, political, moral, and economic valuations that are fundamental to medical practice. In this article, a temporal analysis of epilepsy patient and practitioner valuations unveils a temporal economy of EEG monitoring. By temporal economy, I refer to an emergent system characterizing how various stakeholders contribute time to derive value in ways that stand in well-defined relationship to one another. I use the word ‘economy’ in its older sense, following Daston: ‘[economy] refers not to money, markets, labor, production, and distribution of material resources, but rather to an organized system that displays certain [explicable] regularities’ (Daston, 1995, p. 4). My conceptualization of temporal economies is related to, but different from, Daston’s notion of moral economies of science. A moral economy of science refers to a stable system of emotional forces that characterizes particular collectives of scientists and their ways of knowing. Moral economies explain ‘how scientists at a given time and place dignify some objects of study at the expense of a great many others, trust some kinds of evidence and reject other sorts, and cultivate certain mental habits, methods of investigation, and even characters of a distinctive stamp’ (Daston, 1995, p. 23). For example, scientists who cherish quantification cultivate and demonstrate impartiality and impersonality to facilitate trust and collaboration with scientists from other backgrounds (Daston, 1995). In contrast to the single community focus of moral economies, temporal economies aspire towards multi-stakeholder analyses of valuation.

A temporal economy accounts for the experiences and circumstances of multiple stakeholders. When the patient Olivia missed two days of work for EEG monitoring that yielded no diagnostic insight, epilepsy fellow Betty spent endless hours staring at brainwaves on a screen with little reward or satisfaction in terms of patient care and research. Examined individually, these observations tell us about the values and constraints of individual stakeholders, without describing how these values and constraints interact in the ecosystem of medicine. Therefore, temporal economies require analysts to record and compare the relationship between time and value for multiple stakeholders. This includes noticing how patients like Olivia frequently trade significant time away from income-generating and educational opportunities and spend long periods connected to a weird machine to generate data that may help treat their seizures. When they do so, patients receive expert attention for a fraction of that time and derive therapeutic effects regardless of the EEG’s diagnostic utility. The knowledge enterprise of academic medicine depends on access to such patients for research and teaching.

Simultaneously, analysts must attend to how Betty and other epilepsy fellows must evaluate EEGs and write reports from 8 am–6 pm every day, with barely time for lunch. The epilepsy fellowship—a year-long stint of computer-facing and bureaucratic-seeming work—comes immediately after neurology residency, a four-year-long immersion in patient-facing and definitively clinical work. This transition from caring for patients to staring at brainwaves makes the fellowship a difficult and tedious year, but the apprenticeship enables fellows to qualify for the higher skill level and salary associated with the subspecialty of epileptology. Ultimately, EEGs provide clinical experience and economic value, and offer fellows and attendings the opportunity to advance their teaching and research records. Analysts must also attend to the often-invisible work and motivations of EEG techs like Amelia, who supply long hours of physical, interactional, and analytical labor: connecting, disconnecting, and monitoring patients and EEGs; interfacing with patients, caregivers, nurses, and physicians; and archiving EEG data for posterity, without formal recognition on occasions when their work informs medical decisions and publications. People from socioeconomically disadvantaged backgrounds value the stable employment, health insurance, and retirement benefits offered by medical centers, and the tech role is an accessible path to the medical profession (requiring a high school diploma and basic cardiopulmonary resuscitation training). Thus, each group of stakeholders depends on the time contributed by others to derive value in ways that stand in well-defined relationship to each other.

This juxtaposition and holding-together of patient and practitioner valuations reveals constraints faced by each group and provides analytical insight into tensions between groups. Temporal economies thus bridge moral economic approaches to valuation (focused on a single community of experts) with the pragmatic strand of valuation studies (that accounts for tensions between the valuations of multiple stakeholders). While recent analyses of medical technology have emphasized their financial and legal value, a temporal analysis uncovers expansive notions of value, such as the therapeutic value of being seen through a diagnostic technology.

In the contemporary life sciences and biomedicine, ‘the abstractions that represent value are more and more distantly coupled (ontologically and materially) from their materialist bases’ (Sunder Rajan, 2020a, p. 439). By allowing us to investigate how and when these ties are established, examined, and loosened—that is, by allowing us to reconnect abstractions to their materialist bases—the idea of temporal economies provides a tool to read against the grain of the utilitarian calculus and reformulate therapeutic value in terms that matter to patients and practitioners.

Practical implications

Medical practitioners (especially trainees, such as residents and fellows) should be taught about the therapeutic benefits of extended interventions and taught to recognize that providing agentic and relational opportunities that make patients feel seen and cared for is a fundamental part of the care and service they provide. Patients and caregivers should be taught that the opportunity to learn more about their chronic condition under the careful supervision of experts is part of the reason to undergo extended monitoring. Chronic conditions like epilepsy cannot be cured: for example, 30% of patients continue to have seizures despite treatment, and the rest can have seizures due to hormonal fluctuation, stress, sleep deprivation, missed medication, head injuries, and infections (Brodie, 2010; Krauss, 2014; World Health Organization, 2023). By providing a controlled environment for patients to seize under expert care, EEGs could be seen as doing the work of therapy (in addition to diagnosis). That is, EEG time could be considered just as valuable as treatment, instead of being dismissed as idle time. Indeed, patients and practitioners are rarely idle during idle time: While patients and caregivers engage in sensemaking, rest, and recreation, practitioners juggle EEG duties with feuds for hospital resources. Therefore, idle time could be reconceptualized as demanding active work from patients (as has been argued of the time of diagnosis by scholars of disability and chronic illness (Charmaz, 1991; Jain, 2007)) and medical practitioners alike.

Hospital administrators and technologists should take note of the resources and practices that generate therapeutic value for patients, as well as factors that limit therapeutic value. One way of making these therapeutic resources and practices explicit and thus legible for administrators and technologists is by changing reporting conventions for extended interventions. For example, EEG results are currently communicated via dry textual reports describing the presence and absence of characteristic waveforms and other physiological indicators. In addition, reports could recount how the patient’s experiences were witnessed and cared for. In chronic conditions, such reports could help normalize the patient’s condition and familiarize patients and caregivers with strategies to manage acute events such as seizures. More generally, administrators and technology developers should account for multiple and contested valuations of time in attempts to determine efficiency gaps and opportunities for automation. When defining and measuring value, waste, and efficiency, practitioners should look beyond readily quantifiable outcomes and account for benefits that may be harder to estimate, such as the therapeutic value of witnessing.

Would patients find extended monitoring therapeutic if automated systems were doing the work of care? This article suggests not, although this remains to be seen as automation finds its way into the EEG lab and beyond.