Between Care and Control: Examining Surveillance Practices in Harm Reduction

Abstract

As harm reduction programs and services proliferate, people who use drugs (PWUD) are increasingly subjected to surveillance through the collection of their personal information, systematic observation, and other means. The data generated from these practices are frequently repurposed across various institutional sites for clinical, evaluative, epidemiological, and administrative uses. Rationales provided for increased surveillance include the more effective provision of care, service optimization, risk stratification, and efficiency in resource allocation. With this in mind, our reflective essay draws on empirical analysis of work within harm reduction services and movements to reflect critically on the impacts and implications of surveillance expansion. While we argue that many surveillance practices are not inherently problematic or harmful, the unchecked expansion of surveillance under a banner of health and harm reduction may contribute to decreased uptake of services, rationing and conditionalities tied to service access, the potential deepening of health disparities amongst some PWUD, and an overlay of health and criminal-legal systems. In this context, surveillance relies on the enlistment of a range of therapeutic actors and reflects the permeable boundary between care and control. We thus call for a broader critical dialogue within harm reduction on the problems and potential impacts posed by surveillance in service settings, the end to data sharing of health information with law enforcement and other criminal legal actors, and deference to the stated need among PWUD for meaningful anonymity when accessing harm reduction and health services.

The growing recognition of the value of harm reduction has begun to shift the relationship between people who use drugs (PWUD) and health and social services. In the context of the North American overdose and drug poisoning crisis, some jurisdictions have seen the expansion of harm reduction programs, recognizing that PWUD have faced systemic neglect, violence from both medical and criminal legal systems, and widespread deprivation. PWUD are routinely subject to a range of oppressive surveillance practices within mainstream healthcare settings (e.g., hospitals) (Guta et al., 2021) and in public (e.g., law and policing) (van der Meulen et al., 2021; van der Meulen & Chu, 2022). While new health initiatives serving PWUD are welcomed, this context has also generated a range of surveillance practices with potential impacts that demand critical appraisal. Legal identification requirements to access supervised consumption sites (Passafiume, 2021) and systematic “wellness checks” in shelter settings (Marquez et al., 2020) are but two examples of how routinized surveillance is implicated in the daily realities of PWUD.

In this essay, we draw upon our collective expertise as researchers, harm reduction workers, and participants of PWUD social movements. This combined experience comprises political advocacy, community-based research (including on the impacts of drug criminalization, barriers to health services, etc.), and harm reduction service provision and access (such as needle and syringe programs, supervised consumption, overdose prevention in residential settings, safer supply programs, and street-based outreach). We overlay relevant evidence derived from lived and practice-based experience with recent sociopolitical and programmatic developments. By blending personal observation from sustained and long-term engagement in harm reduction with broader empirical analysis of a range of programs and policies, we identify emerging surveillant trends and provide a critical appraisal of their implications. This approach involves an orientation toward the future and the possibilities contained therein, while recognizing not all are foregone conclusions (Rhodes & Lancaster, 2021). Our analysis draws in particular on the institutional and policy environment where we are located (Ontario, Canada) to interrogate the presumed benign and inherent good of surveillance practices that fall under a banner of harm reduction and health care. Canada provides a critical site to explore this tension as a leader in harm reduction research and programing. To this end we take cues from critical scholars who have called for greater interrogation of interventions targeting PWUD, which claim to be caring, benevolent, or therapeutic (McKim, 2008; Moore, 2011). We argue that surveillance expansion within harm reduction functions through appeals to care, optimization, and resource allocation. We further argue that surveillance expansion generates specific harms and potential impacts for people who use drugs, exacerbating existing barriers to health. In so doing, we address the absence of work within harm reduction and drug policy that contends with, and critically interrogates, surveillant practices.

While surveillance is frequently conceptualized in relation to state power, population management, and centralized rule resulting in “profoundly unequal” and “debilitating impacts” (van der Meulen & Heynen, 2019, p. 4), it is also a generative concept that allows us to examine the various practices on the continuum between care and control in healthcare, and to render visible those practices operating in a grey area between the two (Armstrong, 1995). As such, this essay draws on a critical social sciences tradition that emphasizes the denaturalization of accepted social forms (Sayer, 2009). While we raise critical questions about surveillance, we do not regard all surveillance practices as inherently punitive or harmful. For example, some epidemiological and bio-behavioral surveillance data has enabled a more comprehensive understanding of drug-using practices that have been instrumental to leveraging resources, initiating programs, and developing responsive health policy, a recognition reflected in other areas of health surveillance (French & Smith, 2013). Data collection can be primarily benign, motivated by establishing reach of services, managing inventory of resources, or other operational concerns. In other instances, however, public health officials have provided bio-behavioral overdose “hot spot” surveillance data to police services which use the information to target and arrest PWUD (Canning et al., 2021). In this way, surveillance in harm reduction can actuate both “care” and “control” (Moore, 2011).

This essay aims to make the tension between care and control more visible and explicit. Surveillance is our analytic framework precisely because of its capacity to underscore how banal or “caring” practices can slip into or be used for coercive and/or repressive purposes, regardless of intent. We raise questions and make connections not to categorically denounce these practices but to offer tools in the interest of drawing attention to the implications of collecting personal data, the systematic observation by peers, front line workers, and health care providers, the increased circulation of sensitive health information, and more.

Our approach here is informed by notable scholars of surveillance studies. We employ Lyon’s (2018) definition of surveillance as “the operations and experiences of gathering and analyzing personal data for influence, entitlement or management” (p. 6), coupled with French and Smith’s (2013) understanding of “health surveillance” as “any monitoring, whether systematic or not, of health-related information” (p. 383). Both definitions are decidedly broad, underscoring that such practices are not always inherently harmful or repressive forms of social control, but rather, that surveillance can also encompass banal practices such as routine data collection (keeping in mind that information which may be deemed banal or neutral can in fact be sensitive and personal to some, or shift over time as circumstances change).

Through this, we mobilize the concept of the surveillant assemblage, characterized by networked flows and “driven by the desire to bring systems together” (Haggerty & Ericson, 2000, p. 610), to conceive of the range of loosely (inter)connected surveillance practices described below. This assemblage operates notably through the creation of data doubles which “serve as markers for access to resources, services and power in ways which are often unknown to its referent” (Haggerty & Ericson, 2000, p. 613), as well as through social sorting (Lyon, 2003) in which data are collected for classification, risk assessment, and intervention. Assemblage thinking has been advanced for understanding the complexity of drug use beyond the addictions paradigm (Duff, 2014). We emphasize the decentralized nature of surveillance in harm reduction, its expression as both lateral and vertical, and its capacity to be both relationally and technologically mediated.

These surveillance studies definitions and concepts direct our attention to the fact that, while harm reduction sites may be discreet, surveillance practices can simultaneously work as a whole: they share many of the same rationales and outcomes, yet often have diverse purposes (e.g., clinical, epidemiological, evaluative, administrative, regulatory, criminal legal, etc.). The ubiquity of these routine practices means that some harm reduction practitioners and scholars may be unaware of the extent to which (or even resist the notion that) surveillance is increasingly implicated in the functioning and governing of harm reduction programs, for better or for worse. Thus, this essay begins by conceptualizing harm reduction surveillance in relation to current tensions we have observed through our respective work, research, and experience in health and social services. Next, we provide an inventory of sites of surveillance in harm reduction programs that we see today, the inter-reliance of both technological and relational forms, and how data are repurposed in networked flows. This is followed by what we identify as the corresponding impacts and implications of surveillance in harm reduction settings, with particular attention to questions of access, disentitlement, and health disparity in the context of enduring criminalization. We draw on empirical examples to foster critical dialogue amongst those invested in harm reduction approaches, but we intentionally refrain from offering “one size fits all” policy prescriptions.

Conceptualizing Harm Reduction Surveillance

The concept of harm reduction lacks definitional consensus (Denis-Lalonde, 2019; Miller, 2001; Tammi, 2004), and reflects a largely contested and iterative terrain which includes a constellation of both institutional and community-driven practices (Smith, 2012). As a term, harm reduction is used by public health institutions as a descriptor to characterize programs and services that advance population health goals of reducing disease burden among specific priority populations of PWUD. It is simultaneously used by social movements to signal a set of values in individual and societal responses to drug use and drug trades, emphasizing humanism, self-determination, and the rejection of both abstinence as a precondition to service access and criminal law as a tool for the governance of drug use (Nowell & Masuda, 2020; Smith, 2016). Harm reduction services often represent a convergence of these two approaches—that is, institutional and social movement—with related visions animating philosophical and programmatic tensions, sometimes even within the same organization or program. But how do these visions of harm reduction orient themselves to surveillance?

An institutional vision of harm reduction relies largely on biobehavioral and epidemiological surveillance, emerging in part from surveillance medicine (Armstrong, 1995). Harm reduction, regarded as a pragmatic approach, found certain receptive audiences within public health and epidemiology interested in engaging marginalized groups and the reduction of disease burden among populations impacted by HIV and hepatitis C, though this institutional recognition of harm reduction has been highly uneven (Smith, 2012). Significantly, the entanglement of the institutional vision of harm reduction with surveillance is not limited to its entanglement with epidemiological approaches. Surveillance in health care settings is embedded in the routine functions of institutions, including clinical charting and other record-keeping practices, managing organizational liability, and evaluation and regulatory oversight. While varied forms of data collection and related surveillance practices are frequently a function of contractual obligations with funders or the state, the institutional vision of harm reduction instead often frames these practices as service coordination or continuity of care (e.g., the centralization of health records has been made possible through the growth of electronic medical records, see Perera, 2011). As harm reduction programs play an ever-greater role linking PWUD to clinical and other services, and are frequently collocated within the same settings, concerns have deepened over the collection of client information, the transfer or movement of data across databases, and institutional relationships with criminal legal systems and police.

Given the tendency for communities of PWUD and their advocates to reject the role of criminal law in regulating drug trades and management of drug-related harms, there tends to be a more critical orientation to surveillance within the social movement vision of harm reduction. A philosophical commitment to self-determination (e.g., drug use as choice), bodily autonomy, and the leadership of PWUD (e.g., as captured in the organizing principle Nothing About Us Without Us) further affirms a tendency to reject institutional modes of surveillance that might undermine this ethical position. As the use of epidemiological methods to monitor overdose grows, and harm reduction services become increasingly integrated into healthcare, the relationship between social movements and surveillance has grown more complex and ambiguous. Harm reduction workers, clients, and people with lived experience hired into “peer” employment roles are often at the forefront of efforts to push back against surveillance practices. But even within this more critical orientation, surveillance is frequently operationalized and tacitly endorsed, for instance through calls for greater data collection on patterns of drug use and locations of overdose, demonstrating that community-based and community-led harm reduction initiatives also carry with them the possibility for, and at times may directly replicate, surveillance-related practices. Surveillant dimensions are generally deemed to be necessary trade-offs to secure resources to respond to health crises or address health inequities.

How do service users and PWUD orient themselves to surveillance? The desire for discretion, confidentiality, and anonymity among those seeking harm reduction services is well established (Klein, 2007; Loue et al., 1995; World Health Organization, 2007). Reasons for this are varied, including the potential intrusion of child apprehension authorities (Boyd, 2016) or fear of police profiling and harassment for having been identified as a PWUD (Wood et al., 2003), as well as to shield against stigma and discrimination (Wolfson et al., 2021) or to protect against criminalization (i.e., sex work, drug selling). Evidence from the province of Alberta reveals that only 36% of prospective service users would access supervised consumption services if asked for ID (Hyshka et al., 2016). Where surveillant practices are tolerated among PWUD (such as collecting nominative information), some studies have observed a correlation between the level of economic and racial marginalization and willingness to accept surveillance measures (see Xavier et al., 2021). While further research is needed to fully understand the dynamics at play, this raises troubling questions concerning the potential role that social deprivation and systemic racism play in acquiescence to intrusive practices in health and the racial and social capital often required to contest or refuse such intrusions.

Health- and harm-reduction-related surveillance of PWUD is embedded within the broader surveillance of poor, criminalized, and racialized communities. Invasive monitoring of those accessing public, disability, and social assistance services (Eubanks, 2018; Maki, 2021; Monahan, 2008), as well as racial constructions of risk that pervade both criminal legal systems (Hannah-Moffat, 2019) and medicine (Dryden & Nnorom, 2021), structure the forms of surveillance experienced by PWUD across health care settings. For instance, receiving social assistance places individuals into a highly gendered and racialized network of bureaucratic surveillance, creating opportunities for greater scrutiny and intervention by enforcement agencies, as with child protection/apprehension (Boyd, 2019). This is consistent with claims that policing and surveillance functions have been amplified among welfare and social services in the context of neoliberal economic restructuring and austerity (Wacquant, 2001) and continuous with historic forms of colonial surveillance of substance use among Indigenous peoples (Genosko & Thompson, 2006). One area in which the broader intersection with poverty surveillance is evident is in social and supportive housing settings, where surveillance practices include CCTV cameras, security personnel, unannounced room inspections, guest logs, and collaborative relationships with police (e.g., turning over of security camera footage upon request) (Boyd et al., 2016; Fast & Cunningham, 2018; Parsell, 2016).

While intersections and convergences with other surveillance forms are multiple, this essay focuses on the surveillance to which PWUD are subjected when seeking and accessing harm reduction services. These services have been implemented mainly due to years of activism by PWUD and allies, fighting for governments to address the devastating toll and loss of human life wrought by drug prohibition and a contaminated illicit supply. As members of PWUD movements and drug policy researchers, we too have advocated in support of such programs. In no way do we seek to undermine these efforts, particularly as these gains are often precarious. Instead, this essay takes up calls for critical engagements—both theoretical and material—with harm reduction, particularly during crisis and “uncertain political times” (Watson et al., 2020).

Locating Surveillance in Harm Reduction Programs

You wake up. It’s time to head to the pharmacy for your observed daily dose of slow-release oral morphine. They pull up your file, watch you swallow the pill, chart that you’ve taken your dose, and fill your prescription. You appreciate that the person on staff just lets you swallow the pill and didn’t insist on opening the capsule and sprinkling it on pudding or asking to inspect your mouth as some other pharmacy staff do. They tell you they need a confirmation from your social assistance worker that you still have medication coverage so they can update your file. While you’re at the pharmacy, you pick up a few naloxone kits for your neighbor who doesn’t feel comfortable asking the pharmacist—she’s dealing with child protective services and is unsure who might have access to this information when it’s documented. They ask for your health insurance card again—they explain it’s so they can bill the Ministry of Health. Later, you drop by the supervised consumption site to see some friends. You walk in and are asked to provide your code—a combination of your birth year, letters from your first and last name, and part of your postal code. A friend is being interviewed by a peer researcher. Among the questions: “Do you share needles with anyone?” and “When was the last time you overdosed?” When your friend is finished and gets their $10 voucher, you both leave and head back to your place. You live in a subsidized housing unit run by an agency. Your friend needs to sign in and there are CCTV cameras everywhere. Though after spending eight years on the waiting list to get an apartment you feel like you can hardly complain.

This vignette, based on an amalgam of our direct experiences and observations, illustrates the range of surveillant practices that PWUD may face in a typical day while accessing harm reduction and social services. In this section, then, we provide a partial inventory of surveillance in these and other harm reduction settings, many of which have emerged or gained traction in the context of the overdose crisis.

Needle and syringe programs (NSPs): Lack of anonymity is widely cited as a key barrier among PWUD when accessing NSPs, and best practice guidelines caution against requirements for identifying information (Klein, 2007; Loue et al., 1995; Strike et al., 2013, 2015). In Ontario, NSP providers assign a unique identifier combining year of birth, gender, and a four-letter alphabetic code consisting of the first two letters of the client’s first name and final two letters of their last name (ONHRDB, 2021), effectively creating a data double that can be mobilized for other purposes (e.g., service planning). While clients are not generally denied services for refusing to provide this information, these data are required by the provincial health authority. The first three characters of the client’s postal code are also requested in some cases. While providing pseudonymous information is permitted, this option is often not expressly communicated, and practices vary widely between NSP sites. For most NSPs in Ontario, data are then input into a privately developed third-party database called Neo 360, a UK-based company (ONHRDB, 2021). Harm reduction service providers retain technical ownership of the data. A privacy impact assessment that was conducted in 2017 did not find technical violations but did identify several risks, including data security and challenges related to communicating data collection practices to clients (ONHRDB, 2017).

Supervised consumption services (SCSs): These programs have been the focus of critical appraisal, most notably for the regulatory constraints under which they operate (Fischer et al., 2004; Foreman-Mackey, 2019; Kolla et al., 2020; Scher, 2020). Surveillance is manifest at several levels, including through the supervision of drug use, injection practices, and related behavioral mandates such as prohibitions on drug sharing/splitting, restrictions on peer-assisted injection, ¹ interdictions against injection in some regions of the body, and the disallowance of consumption via inhalation (Ranger, 2021; Small et al., 2011). Surveillance within SCS functions as a tool to regulate and curtail practices deemed as surpassing acceptable risk thresholds or as presenting an organizational liability. Prohibitions such as these are generally understood within the policy literature and by SCS staff as necessary constraints produced by external legal and policy environments (Small et al., 2011). A 2012 supervised consumption assessment study found that prospective service users preferred an intake process that did not require either legal identification or an anonymous identification code (Bayoumi et al., 2012). Despite this, those accessing SCSs in Ontario are assigned the same unique identifier as when accessing an NSP. Similarly, and as noted above, while the possibility of providing pseudonymous information exists, this option is often poorly understood by PWUD and practices across SCSs vary considerably.

Overdose-response interventions: Among the most common overdose response interventions that have emerged or been scaled up in recent years are community and pharmacy-based take-home naloxone distribution programs; witnessed drug use and “wellness checks” to monitor for overdoses in shelter settings; and formalized “spotting” interventions consisting of accompanied use via telephone or app (Bardwell et al., 2018). In Ontario, pharmacists request identification as proof of enrollment in provincial health insurance when distributing naloxone, in which logged information is attached to an individual’s pharmacy file. While they may “in limited circumstances” provide naloxone to those who do not wish to provide identification (Ministry of Health and Long Term Care of Ontario, 2021), this policy is often poorly understood by pharmacy staff and clients. Regarding wellness checks, these became commonplace in shelter-hotels and other congregate settings that opened in the context of the COVID-19 pandemic. While some checks are voluntary and initiated at residents’ request, others have raised concerns regarding unannounced and unwanted inspections. For example, we have witnessed shelter operators using information collected through residents’ acquisition of sterile drug use supplies to identify them for later checks, without resident request. Information gathered through monitoring practices initiated in service of safety are redeployed to deny service and can lead to resident evictions. “Wellness checks” conducted without consent can comprise a level of surveillance prone to abuse while normalizing privacy violations and can lead to isolated drug use to avoid detection. And lastly, formalized phone and app spotting interventions that support people who use drugs alone collect geo-locational information so that trained volunteers or workers can trigger a response (e.g., alerting a neighbor, calling EMS) if an overdose takes place or is suspected. Data collection, storage, and governance practices diverge significantly. These initiatives generally involve community-based organizations, health authorities, researchers, and private sector actors (Perri et al., 2021). A similar concept in development are wearable biometric sensors with remote sensing technology (Roth et al., 2021). Research has noted that criminalization may limit uptake of formalized services (Perri et al., 2021), underscoring the ongoing need for longstanding informal models among PWUD (e.g., friends spotting friends) based on negotiated safety.

“Peer” workers in harm reduction services: The growing recognition of the importance of involving people with lived experience of drug use has prompted significant shifts within harm reduction services, resulting from sustained calls from drug users’ rights movements (Canadian HIV/AIDS Legal Network, 2005). This shift has generated discussion and debate regarding professionalization, remuneration, and working conditions, especially in light of peers’ experiences of disproportionate workplace surveillance due to entrenched stigma, perceptions of untrustworthiness, and systemic discrimination (Kennedy et al., 2019; Michaud et al., 2016; Olding et al., 2021). Additionally, peers are often enlisted into the institutional relations and practices inherent in these settings, including data collection, record keeping, and compliance with policies and expectations regarding client behavior and conduct. “Peer navigator” roles encompass a wide range of practices and ethical orientations, ranging from accompaniment and advocacy that is voluntary and initiated at the request of a client, to highly directive mandates to incite, encourage, or persuade forms of behavioral compliance (e.g., treatment adherence) (Cunningham et al., 2018). Initiatives of this kind underscore the central role of relational dynamics in harm reduction surveillance—at once horizontal by virtue of a shared experience and identity as a PWUD, and vertical due to an institutional power differential—comprising forms of lateral surveillance (Andrejevic, 2002). Thus, peer involvement risks instrumentalizing the shared experience and social location of PWUD to achieve narrowly defined institutional goals. At their worst, these practices exploit trust and solidarity among PWUD.

Prescription opioid programs: Opioid agonist treatment (OAT)—and methadone in particular—has faced sustained critique for its punitive and surveillant functions, including urine drug screens, daily observed dosing, restrictions on take-home doses, and prescription monitoring programs, placing enormous constraints on autonomy and mobility. These practices are generally regarded as stemming from regulatory constraints and stigma from clinicians (Bourgois, 2000; Fischer, 2000; Fraser & valentine, 2008; Treloar & valentine, 2013). One example of embedded surveillance in OAT is the recruitment of pharmacists to monitor for “aberrant behaviors” regarded as evidence of diversion of opioid agonist medication (Bach & Hartung, 2019), a shift from passive surveillance (e.g., information gathering) to active surveillance (e.g., intervening or governing behavior directly) (Holmgren et al., 2020). Others have raised similar concerns regarding the enrollment of physicians by law enforcement in disciplinary interventions aimed at opioid users in service of prescription monitoring (Hussain & Bowker, 2021). We raise OAT here not to reiterate calls for systemic and regulatory changes that are well documented elsewhere (see Crawford, 2013; Frank, 2021; McEachern et al., 2019; McNeil et al., 2020), but to note the extent to which surveillance is embedded in the design of OAT and to foreground how these practices are exported into other health and harm reduction services.

The expansion of pharmacological options for opioid users has grown in recent years, propelled largely by the overdose crisis, yet this growth has been uneven, faced difficulties in scale up, been subject to considerable contestation, and is inadequate to address community needs (Tyndall, 2020). In some jurisdictions, this has included increased availability of pharmaceutical opioids (such as hydromorphone, slow-release oral morphine), either in oral or injectable formulations to PWUD deemed at high risk for overdose, often under a banner of injectable opioid agonist therapy (iOAT) or safer supply (Tyndall, 2020). Safer supply in Canada is currently only prescribed by clinicians (Bonn et al., 2020) and generally requires recipients to be diagnosed with opioid use disorder. Certain programs require that all safer opioid supply doses be consumed in SCSs or specialized clinics under supervision, producing substantial barriers for those whose schedules or consumption-related needs do not align with operational requirements. In some SCS settings, injection of prescribed opioids is accommodated and managed on site, while others permit take-home oral doses recognizing that many will choose to crush and inject (British Columbia Centre on Substance Use, 2020).

In several respects, safer supply programs depart from many of the surveillant and punitive functions of opioid agonist treatment (e.g., no requirement for abstinence, greater flexibility regarding take-home doses among some programs, etc.) and provide some tentative first steps in health-based approaches that take a critical orientation to unwanted surveillance. However, the regulatory apparatus in which prescription-based safer supply is embedded partially constrains these intentions. Due to the complex array of federal, provincial, professional, and organizational regulations on opioid prescribing, clinical charting is extensive and observed dosing of certain medications is required. Additionally, such initiatives are often subject to prescription drug monitoring programs, despite growing evidence of the community-level protective effects of diversion of prescription medication (Bardwell et al., 2021; Doernberg et al., 2019).

Surveillance and data collection in safer supply and iOAT programs serve multiple purposes, fulfilling clinical, bureaucratic, regulatory, epidemiological, and evaluative needs. Specific safer supply programs use biometric information, such as handprints, for identification purposes (Tyndall, 2020). Within the current regulatory and programmatic context, iOAT is only available as an observed model (except for some COVID-related exemptions) (Oviedo-Joekes, 2021). Risk stratification to determine eligibility in the context of constrained resources and stringent requirements may serve as a form of routinized social sorting. Entitlement to a regulated, pharmaceutical-grade supply is, in the process, rendered contingent on a corresponding obligation to accept medical surveillance. Safer supply and iOAT prescribers and PWUD alike are advocating for community-led approaches to de-medicalize access to a safer supply of opioids (e.g., Vancouver drug buyers’ club) (Lekhtman, 2021), which would go a long way in addressing many among the surveillant practices embedded within—and endemic to—more medicalized approaches.

Key Features of Harm Reduction Surveillance

By examining this assemblage of surveillant practices in the above harm reduction settings, we can begin to trace how technologically mediated forms of surveillance (e.g., data collection and institutional data sharing) are facilitated by relational ones (e.g., the enlistment of supportive and therapeutic actors), often with “the desire to gain total knowledge of clients” (Fraser & valentine, 2008, p. 80). McKim’s (2008, 2017) examination of surveillance in residential treatment programs provides a valuable case in point, showing how personal relationships, intimate knowledge of clients, and therapeutic alliance are instrumental in cultivating adherence to institutional or program goals. In these and other examples we see service providers being recast and repositioned as surveillance workers responsible for the extraction of data and information from PWUD, reflecting the porous border separating care and control. Herein we can see a broader social process at play in which therapeutic workers take on surveillant functions, and at times engage in penalizing practices toward PWUD. (See Hussain & Bowker, 2021, for the enlistment of physicians by law enforcement in prescription drug monitoring and governance.) Frequently, personal health data collected by these surveillance workers are repurposed across institutional sites or mobilized for other uses, creating a data double whose utility serves to blur the distinction between information acquired for clinical, evaluative, and epidemiological reasons. And while the linking of discrete datasets and the repurposing of clinical data are longstanding, these practices can circumvent established principles of informed consent by obscuring how data are recirculated (McClelland et al., 2020; O’Byrne & Holmes, 2009). Indeed, many have noted the inadequacies of existing consent and individual privacy-based approaches in an era of big data (Corrigan et al., 2009).

Data collection practices are not always inherently intrusive and should not be collapsed with data leaks or violations of privacy policies. Yet, we contend that harm reduction services are characterized by a substantial grey area in which routinized data collection and data circulation practices can slip easily from benign to problematic, stemming from, among other factors, the lack of clarity and transparency around the purposes and uses of personal information which is collected (including among workers), and the heightened stakes in the context of ongoing criminalization of drug use and drug trade involvement. Routine data collection and data sharing by harm reduction services pose two distinct problems: an increased likelihood of breaches and violations; and a disregard for the stated preferences and needs expressed by PWUD—as reflected in best practices guidelines—for anonymity when accessing services.

The sites of surveillance in harm reduction programs detailed above also share key rationales, including:

by invoking institutional or administrative demands (e.g., regulatory compliance, bureaucratic requirements, organizational policy);

by invoking the best interest of the service user (e.g., safety, therapeutic benefits);

to gain a more comprehensive view or understanding of the client (e.g., therapeutic alliance);

by invoking the exceptional status of PWUD (e.g., as disorganized, chaotic, unreliable);

by invoking community safety (e.g., the purported safety of non-drug using community members);

to allocate resources (e.g., determinations of eligibility or deservedness);

to generate more “reliable” data, positioned relative to the purported inadequacies of existing evidence and the presumed inherent unreliability of self-report among PWUD (Lancaster et al., 2019); and

by appealing to the need for greater service optimization (e.g., continuity of care, cross-sectoral collaboration).

Taken as a whole, these rationales point to processes whereby potential privacy violations are deemed justifiable, surveillance practices are deemed reasonable, and institutional demands are privileged over client preference for low barrier access. Indeed, the mobilization of these rationales effaces concerns among PWUD, propelling barriers to access services (Greene et al., 2022). An interrogation of the rationales forwarded to justify surveillance underscore their politically or ideologically motivated official logics, as a recent review of SCS in Alberta has illustrated (Greene et al., 2022). We argue that this is far from an inevitable or necessary component of delivering health and harm reduction services to PWUD, as several innovative low threshold models providing fully anonymous access demonstrate (Wallace et al., 2019). A critical examination of the role of surveillance practices within harm reduction settings requires one to question these rationales and the presumptions upon which they rely (see Table 1).

Table 1.

Characteristics of Surveillance Across Health Care and Harm Reduction Settings.

Site of surveillance	Surveillance practice	Key rationales	Therapeutic actors enlisted	Purposes of surveillance/data collection
Needle and syringe programs	Data collection, lateral surveillance	Administrative demands; Generation of “reliable” data; Service optimization	Harm Reduction and/or “peer” workers	Administrative; evaluative
Supervised consumption services	Data collection, observed use, lateral surveillance	Administrative demands; Best interest; Generation of “reliable” data; Service optimization	Nurses; “peer” workers	Administrative; evaluative; clinical
Safer Opioid Supply programs / iOAT	Data collection, observed use, biometrics, prescription monitoring / “diversion prevention”	Administrative demands; Best interest; Exceptional status of PWUD; community safety; Resource allocation	Nurses; “peer” workers	Administrative / regulatory; clinical; evaluative
Pharmacy-based naloxone distribution	Data collection	Administrative demands; Generation of “reliable” data	Pharmacists	Administrative / regulatory
Overdose-response interventions in shelter settings (e.g., wellness checks)	Data collection, lateral surveillance	Administrative demands; Best interest; Gain comprehensive view of client	Front-line workers	Clinical, administrative/bureaucratic
Overdose-response interventions via phone lines and apps	Data collection, lateral surveillance	Best interest	Volunteers	Clinical
“Peer” workers	Data collection, lateral surveillance	Gain a comprehensive view of client; Service optimization	“Peer” workers	Clinical

Potential Implications for PWUD

Surveillance practices in harm reduction settings can have varied and far-reaching problematic effects on PWUD. While these are myriad, in this section we explore what we regard as the five most significant: the re-identification of PWUD through data breaches; the use of health data by police in both enforcement and criminal proceedings; the erosion of anonymity thresholds resulting in the non-uptake of harm reduction services; the creation of restrictive risk assessment eligibility and conditionality resulting in disentitlement; and the leveraging of personal information for behavioral compliance. Some of these implications are in the process of unfolding, whereas others are more future-oriented (Rhodes & Lancaster, 2021). We recognize that some scenarios described below focus on potential impacts that as of yet have not taken hold in the context of harm reduction; where this is the case, we root our discussion in empirically grounded examples that have played out in other areas of health policy (e.g., HIV).

To begin, especially concerning are the increasingly common data breach incidents of sensitive health information (Seh et al., 2020). Two recent and high-profile examples of ransomware or cyberattacks in Canada involve the breach of the Newfoundland provincial health system that was described as the “worst in Canadian history” (CBC, 2021) and the breach of information regarding about 15 million Ontario and British Columbia clients of the private testing company LifeLabs, including medical diagnoses, personal information, and blood test results (Gollom, 2019). Given these and other data breaches, PWUD have legitimate concerns regarding the privacy of their health information as collected by and held at harm reduction services like NSPs or SCSs. These concerns are often allayed by assurances that personal data are either anonymized or de-identified. However, re-identification is increasingly possible as data collection and computing capacity increases (Tanner, 2017). As the surveillant assemblage grows more tightly networked, risks increase. Exposure of health information, or even service access, stands to have particularly troublesome effects for PWUD, ranging from the denial of insurance coverage to employment discrimination to issues with child custody (see Dovey, 2017, for the denial of health insurance to people using PrEP as HIV prevention method). The ongoing criminalization of drug use and entrenched social stigma faced by PWUD implicate much broader-reaching consequences of surveillance than with most other information leaks or re-identification risks.

In addition to potential data breaches, the sharing of information about PWUD between public health, law enforcement, and the criminal legal system represents another troubling example of surveillance health data being used in problematic ways. Take, for example, the Connecticut Statewide Opioid Response Directive (or “SWORD”) where near-real-time localized overdose surveillance data gathered by Emergency Medical Services and shared with public health was subsequently made accessible to local police who used the information to target and arrest people who use and sell drugs (Canning et al., 2021; see also Michaud et al., 2022). This example illustrates how information collected for “health” reasons becomes repurposed in service of prohibition and interdiction efforts. In Ontario, a collaboration between Toronto Public Health and Toronto Police Service saw plainclothes officers accompanying harm reduction workers on outreach activities (Kivanç, 2015). While the Toronto initiative later folded due to outcry among PWUD and allies, evidence suggests that partnerships between harm reduction services and police institutions are becoming increasingly common overall (see Beletsky et al., 2011; Morrissey et al., 2019; Silverman et al., 2012; White et al., 2021). This points to a problematic overlay of health and criminal legal systems and institutionalized forms of coordination that meld care and control (Goetz & Mitchell, 2006). The emergence of novel health data sharing protocols between public health and law enforcement in the context of COVID-19 containment efforts provides further illustration of the movement of health data between criminal legal and health systems with troublesome effects (Luscombe & McClelland, 2021; Molldrem et al., 2021).

Of particular concern is that health records in Canada are subject to subpoena and can be used in court proceedings and police investigations (Sanders, 2015). The potential use of such records is worrisome considering the ongoing prohibition of the personal possession of a controlled substance, and the laying of manslaughter charges for those who provide drugs that result in an overdose death (Heighton, 2019; Hrymak, 2018). Currently, there is little to safeguard against these and other potential criminal legal consequences. While some might question the frequency of such forms of institutional coordination or data sharing practices, we suggest that the question of frequency is of secondary concern given the collateral consequences on transparency and trust between service providers and PWUD. Greene and colleagues (2022) have noted that the mere perception of the convergence between policing and harm reduction services is sufficient to operate as a powerful disincentive. Efforts to meaningfully address the problems posed by such forms of institutional collaboration and data sharing across legal and health spheres require an unconditional termination of such practices.

The recognition among harm reduction service providers of the specific legal vulnerabilities of PWUD given their criminalized status has historically informed a high degree of confidentiality and anonymity, identified as a core component of low-barrier or “low-threshold” service provision (Toronto Central Local Health Integration Network, 2018). However, the increase of surveillant practices in harm reduction settings appears to be steadily eroding anonymity thresholds, as illustrated by growing demands for identification requirements and data collection when accessing NSPs and SCSs (Passafiume, 2021). The normalization of data capture undermines the principle of low barrier access and erodes trust between service providers and the constituencies they serve, resulting in disincentivizing access (McLean, 2013). As one example, an evaluation of a harm reduction and overdose prevention program in the city of Toronto found that lack of discretion and privacy were cited by PWUD—especially young people, those living in zero-tolerance residential settings, individuals involved in drug selling, and parents of young children—as a significant obstacle to accessing services (Kolla & Penn, 2020), a finding echoed in other settings including SCS (Bardwell et al., 2020). This exemplifies the unevenness with which different populations experience surveillance, and further suggests that surveillance avoidance represents a key motivator for “non-take-up” of services (Warin, 2014). Other more commonly recognized factors, such as distance from harm reduction sites and local police practices, compound these experiences.

The desire for anonymity and the corresponding failure of many health and harm reduction programs to ensure that anonymity risks deepening health disparities. This is especially the case among individuals who already access services at lower rates due to systemic racism, fear of child protective service involvement, or criminalization wholesale (Bardwell et al., 2019; Goodman et al., 2017; Wolfson et al., 2021), a reality particularly true for Black communities (Silverman et al., 2012) and women (Kolla et al., 2020; Shirley-Beavan, 2020). Ongoing legacies of medical racism (Dryden & Nnorom, 2021; Nelson, 2011) coalesce with routine surveillance practices, structuring individuals’ health system avoidance. This sustains social sorting on a broader population level and risks reinforcing notions of “deserving” and “undeserving” service users (Fischer et al., 2004, p. 364) while stratifying PWUD into those who accept being subject to surveillance and those who decline.

In many settings, surveillance has become a precondition to program access: only those who can be identified and documented as meeting an established risk threshold are eligible for certain services, often irrespective of assessments of therapeutic benefit. Restrictive eligibility criteria are prevalent among services that lack sufficient resources to meet demand, a situation endemic in parts of the sector. To manage demand and efficiently allocate limited resources, programs often employ risk assessments to identify those deemed at “greatest risk” and then tie such conditionality to eligibility (Bharat et al., 2021; Sandino, 2020). In so doing, harm reduction programs amass significant and sensitive personal information about PWUD. Safer supply and iOAT programs illustrate this dynamic particularly clearly. Given their significantly limited availability and capacity due in part to the generally high level of hesitancy among medical prescribers (Woo, 2021), safer supply programs have instituted restrictive eligibility criteria to account for restricted program capacity and overwhelming demand.

The limiting of supervised injectable OAT programs (siOAT) in British Columbia on “extreme cases” is another key example (Guta et al., 2017). These regimes of conditionality divide PWUD who are susceptible to overdose mortality into groups determined by their relative level of risk. This triage process represents an instantiation of social sorting and the classifying drive of contemporary surveillance (Lyon, 2003; Moore & Hannah-Moffat, 2005) while also constituting a form of bureaucratic disentitlement (Lipsky, 1984), whereby the availability of benefits or services is “conditioned” to manage demands on health system resources. While some may note the existence of triage practices in other areas of health care, such as hospital emergency departments, harm reduction services—and health promotion and prevention measures more broadly—have historically been governed by principles of universal access. The systemic historical under-resourcing of harm reduction has created a slippage to an embrace of triage logics motivated by the management of health system resources. The expansion of such logics—normalized in the context of austerity—demands critical interrogation.

A final noteworthy implication of harm reduction surveillance is how personal information gathered through surveillance practices can be leveraged for behavioral compliance. Coercive OAT practices provide a window into how sensitive information and intimate knowledge that clinicians hold about PWUD can be mobilized to gain desired outcomes or achieve greater compliance (e.g., withholding take-home doses, threatening involuntary treatment discontinuation, using a client’s fear of withdrawal to incite adherence to abstinence, etc.) (Bourgois, 2000; Crawford, 2013; Neale, 2013). The more intimate knowledge that is held by health care practitioners, the greater the possibilities for abuse given the acute power differential between provider and client. Intimate knowledge and personal data of PWUD amassed in service of aligning behavior with institutional prerogatives is a dynamic vividly at play in recent reforms of SCSs in several Canadian provinces. Ontario and Alberta, for example, have each engaged in politically motivated reviews of SCS (Livingston, 2021), creating a new regulatory framework that is treatment-focused and “recovery oriented” (Alberta Health, 2021). In Alberta, this includes a systematic mandatory offer for addiction treatment and a “commitment that referral processes are tracked for each client” (Alberta Health, 2021, p. 7). Recently tabled federal and proposed provincial legislation ² stand to expand the possibilities for coerced drug treatment and to bolster the role of law enforcement in treatment referrals, resulting in greater police discretion in determining who will be subject to criminal sanctions and who will receive therapeutic intervention. The shifting legal environment regarding treatment raises a troubling set of questions for harm reduction services. How might harm reduction be positioned within these new legal arrangements? What might the contemporary legal environment portend regarding institutional coordination and convergence of health and criminal legal systems? Will surveillance data collected in harm reduction service settings be leveraged in service of criminal legal governance?

Conclusion

Our analysis has mapped the ways harm reduction services have taken on a range of surveillant functions in recent years, propelled by sociopolitical and programmatic developments, as well as by innovations in big data and the growing recognition and adoption of harm reduction by state and institutional actors. This has produced an assemblage of surveillant practices, characterized by the alignment of diverse interventions to extract information that is then repurposed in several decentralized but coordinated flows. We suggest that this acceleration of surveillance within harm reduction is operationalized through specific appeals to care, optimization, and allocation. Understanding the rationales upon which harm reduction surveillance relies helps inform a “reflexive stance regarding these possibly counter-intuitive dimensions” (Fischer et al., 2004, p. 364), grounded in a sustained engagement and accountability to impacted people. Such an approach is needed to fully comprehend the sociomaterial and ethical implications of increasingly interconnected drug using, harm reduction and surveillance (Duff, 2016; Guta et al., 2020). We further contend that the expansion of harm reduction surveillance generates specific discernible impacts for PWUD which stand to undermine privacy, accelerate criminalization, and exacerbate barriers to health care.

Some may regard the proliferation of surveillance practices as a necessary tradeoff to safeguard human life in a time of overdose crisis and to secure urgently needed resources. In contrast, others may regard surveillance as an inevitable byproduct of institutional recognition and integration, or indeed, the medicalization of drug use. The exploration of recent shifts in harm reduction above points to the central role of surveillance in both criminal legal and medical models. A critical project that envisions harm reduction beyond surveillance necessarily involves looking beyond societal responses that criminalize and medicalize drug use. While we are attentive to the health-related needs of PWUD and welcome the development of health and other services that address those needs, we echo calls for the development and expansion of non-medicalized responses as well (Emerson & Haden, 2021; Kaye, 2012).

Surveillance thrives on the exceptional circumstances of crisis. Yet, effective harm reduction and overdose prevention measures depend upon an attentiveness to surveillance concerns and the associated need for anonymity and discretion—well established in the literature and by PWUD. An appreciation of the myriad impacts and implications refutes the position that individual privacy rights fulfilled through technocratic processes should be the sole or primary cornerstone of our response. Data justice (Dencik et al., 2019) and data sovereignty (Hummel et al., 2021) provide potential models that hold promise for contending with some of the many problems of surveillance in harm reduction settings. This requires sustained ethical analyses regarding the purported benefits and collateral consequences of surveillance that centers the needs and experiences of PWUD (Molldrem & Smith, 2020). Indeed, our research here reveals at least three significant gaps in understandings of surveillance in harm reduction that would benefit from future studies: (1) acceptance or coercion of surveillance practices when accessing health and social services, including who is more inclined to tolerate such practices as a precondition of access and for what reasons, as well as the potentially related disparities in access and uptake; (2) how harm reduction workers, including those with lived experience of drug use and/or who are in peer positions, navigate the relational and ethical implications of surveillance practices in their jobs, and the strategies they employ to mitigate adverse effects; and (3) the remapping of the relationship between PWUD and social services, particularly the tension between entitlement and corresponding duties and obligations concerning relinquishing data, or treatment adherence.

To this end, we advocate for a proactive engagement that contends with the potential and realized problems posed by surveillance. We thus call for a broader critical dialogue within harm reduction, the cessation of data sharing between harm reduction and criminal legal actors, and for critical research with public health that “orients to epistemological and political differences” (Mykhalovskiy et al., 2019, p. 523). We further echo calls for meaningful anonymity for PWUD when accessing health and harm reduction services and for a deeper engagement among researchers and communities of PWUD for the specificity of their knowledge practices (Alexandridis et al., 2020). Moving beyond a strictly defensive posture about surveillance in harm reduction settings, we also invite broader future-oriented and speculative work (Rhodes & Lancaster, 2021), particularly that which foregrounds PWUD agency and sovereignty in imagining service access beyond surveillance.