Abstract
Hypermarginalized populations, such as homeless drug users with acute health problems, are subject to multiple intersecting adversities that result in social exclusion and chronic suffering. Despite this population's high need for health and social services, institutions provide services that are fragmented and often punitive, contributing to further marginality. In this article, we present a hybrid methodological approach that combines clinical social work and ethnography in a study of intensive case management for HIV-positive indigent adults in Oakland, California. We investigate two primary research questions. First, we consider the challenges this population faces in navigating institutions to meet their basic needs, and we demonstrate how organizational irrationality has severe consequences for this population. Second, we grapple with the question of how to ethically engage hypermarginalized participants in research by presenting a clinically informed intervention that is responsive to individual vulnerabilities and also enhances our understanding of institutional failure.
“Hypermarginality” refers to a historically contingent social positioning in which inequities coalesce to shape everyday experience. On a concrete level, hypermarginality manifests in individuals as a complex matrix of social exclusion and chronic suffering, including homelessness and housing instability, drug use, serious mental illness, poor health, inadequate access to basic social and medical services, and repeated incarceration. Over time, these phenomena become intensely acute and entangled and must be confronted simultaneously, both analytically and practically, as they are contemporary syndemics (Singer and Clair 2003). However, for the urban poor in the United States, there exists a particular paradox that has grave consequences: hypermarginalized populations have extremely high and interconnected needs for health and social services, yet the institutions that ostensibly provide those services are not only grossly fragmented but often extremely punitive, and contribute to further marginality by hampering, discouraging, or oppressing individuals seeking help (Lopez 2011, 2014).
In this article, we grapple with two interrelated research questions, one empirical and one methodological. Empirically, we investigate hypermarginalized individuals’ processes of institutional navigation, analyzing the challenges that they face as they try to access the basic resources necessary for survival and the ways in which various social institutions introduce or contribute to these challenges. Through a qualitative study of destitute adults diagnosed with human immunodeficiency virus (HIV), we find that when people with complex difficulties encounter medical or social service institutions that cannot address their interwoven needs, they risk becoming categorized as “resistant to treatment” (Buckley and Bigelow 1992) and being further excluded from assistance. A lack of coordination across service institutions significantly exacerbates this risk, as “failure” to perform as required for one service can become grounds for denial of another; for example, serious mental health issues can result in an eviction, which in turn becomes a barrier to obtaining subsidized housing. In addition, hypermarginalized people's contact with the criminal justice system is frequent and heavy, and the various restrictions and mandates placed on them through arrest warrants, criminal records, probation, and parole can pose insurmountable obstacles to meeting requirements for services; an individual with a felony drug conviction, for instance, is ineligible for food stamps. Thus, while hypermarginalized populations interface repeatedly with correctional, medical, and social service institutions, these encounters may merely patch them up from one crisis to the next, or even drive them further into a downward spiral, rather than providing coordinated services to elevate their quality of life.
Methodologically, we pursue the question of how to gather rich, detailed data on the experiences of a highly vulnerable, mobile, and oppressed group. Although information about these individuals can be captured to some extent when they are incarcerated, hospitalized, sheltered, or taken to the coroner's office, there are significant obstacles to enrolling them in research that seeks to follow them across various social institutions, and especially into the interstices between institutions. They often cannot be reached through permanent addresses, directory-listed phone numbers, informational flyers, or other standard procedures for participant recruitment. When they are located through targeted outreach efforts or their contact with a criminal justice, medical, or social service institution, their material circumstances and psychological complexities pose exceptional difficulties for staying in touch. And if a confluence of resourcefulness, rapport, and flat-out luck results in study retention, profound questions arise regarding the ethics of gathering data on the deepest layers of human suffering through non-interventionist, observational research.
Yet without research on people at the margin of the margins, we consistently overlook a small but deeply affected population whose inclusion might alter the results of standard research on urban poverty, and thus we remain ill equipped to determine how to address the literal matters of life and death that afflict society's poorest members. The physician Emily Wang, the sociologist and demographer Christopher Wildeman, and their colleagues have demonstrated that systematically excluding incarcerated individuals from participation in medical research results in lower enrollment and retention rates for African American men in cross-sectional and prospective cohort studies, thereby jeopardizing the validity of conclusions for this population (Wang and Wildeman 2011; Wang et al. 2014). They recommend the careful development of protocols for ethically including prisoners in medical studies, given the long-term benefits of understanding this population's specific health issues. Following this logic, adapting methodological strategies to reach people at the nexus of mental illness, substance use, criminal justice involvement, homelessness, and chronic health conditions will help expand poverty research to cover those at the very bottom of the socioeconomic ladder and thereby illuminate channels for the development of meaningful support services and policy reform to reach the most severely deprived.
In our study of hypermarginalized HIV-positive people's experiences interfacing with an array of social institutions, we developed a methodological model of close collaboration between a clinical social worker and two ethnographers. The social worker provided intensive case management to research participants with the primary objectives of engaging them in regular medical care with an HIV specialist who was part of the research team, increasing adherence to HIV medications, and facilitating the continuity of HIV treatment as participants cycled in and out of correctional facilities. A corollary goal of the improvement in HIV-related health was to help stabilize participants by connecting them to permanent housing, government entitlements (such as Supplemental Security Income, or SSI), and other social services. The social worker therefore worked closely with participants to access and navigate what the medical anthropologist Kim Hopper has termed the “institutional circuits” that populate the lives of the poor: courts, jails, prisons, hospitals, shelters, welfare offices, and the like (Hopper et al. 1997). This process permitted the ethnographers to follow the trajectories of participants step by step as they entered, occupied, and exited a gamut of social institutions; indeed, the collaboration with the social worker provided access to information and situations typically beyond the reach of nonclinician researchers, such as legal documents, medical reports, and hospital examining rooms. We thus were able to undertake our inquiry from an embedded, organizational perspective that shed light on the specific sites of fragmentation within critical services and the implications of this fragmentation for the hypermarginalized.
Intervention research is well established in social work, as is the use of qualitative methods to inform social work practices, and much research on poverty is generated by ethnographers within the discipline of social welfare (Fairbanks 2009; Floersch, Longhofer, and Suskewicz 2014; Gilgun and Sands 2012). We therefore make no pretense that the mere combination of social work and ethnography was particularly innovative. However, our approach introduced an unconventional third party into the traditional researcher-participant dyad—an embedded social worker who through the provision of practical assistance and therapeutic support to participants served as a bridge between them and the ethnographers, while simultaneously becoming a “key informant” through her own participation in navigating the correctional, medical, and social institutions tasked with managing the poor.
We organize this article around the elaboration of three primary strengths afforded by this hybrid social work–ethnography approach. First, we discuss how the social worker's clinical skills and therapeutic background enabled her to form sustainable relationships with participants that could weather periods when they suffered severe mental health deterioration in response to the overwhelming stressors of extreme poverty. Although ethnographic research on people with mental health issues exists (Biehl and Eskerod 2005; Scheper-Hughes 1981), social scientists generally are underequipped to handle psychosis and other serious mental illness or mental health crises among participants, especially when mental health is not the primary focus of a study and the researchers are not clinically trained. In the face of significant mental distress, this lack of preparedness can easily translate into psychological and physical dangers for participants and researchers alike. Having a clinician who was the primary point of study contact permitted us to remain in touch with mentally ill participants through therapeutically supportive means. Ethically, this enabled us to assist and advocate for highly vulnerable participants during times of particularly dire need. Methodologically, our approach permitted us to gather data on participants’ encounters with mental health services and, perhaps more importantly, with correctional and social service institutions that were not equipped to meet their mental health needs. These data greatly contributed to our substantive findings about the severe impact of organizational irrationality and disjuncture on people for whom mental health is one among many critical issues.
Second, researchers working with the hypermarginalized must continually confront moments of crisis as participants struggle for survival. As ethnographers such as Philippe Bourgois, Jeffrey Schonberg, and Teresa Gowan have shown, help can be extended during these moments with sensitivity, generosity, and urgency, particularly when time has been invested in developing strong rapport and trust in relationships (Bourgois 1995; Bourgois and Schonberg 2009; Gowan 2010). Incorporating a social worker into the research team provides a formal structure for clinically informed crisis intervention, with the critical advantage of gaining access to medical records, legal files, and other sources of important information that typically are not available to ethnographers but that can be leveraged to assist participants. We elucidate how this advantage played out in our study and reflect on the opportunities offered by these moments to analyze crisis situations from a cross-institutional perspective by giving us an understanding of how information gaps can disrupt continuity of care for suffering people. In addition, our experiences responding to participants’ crises raised broader questions regarding the hegemonic “non-intervention” mandate present in much social science research. We grapple with the broader ramifications of a particular form of clinically trained intervention as not only ethically mandated but as central to enhancing the means and lenses through which we understand complex social problems.
Finally, we expound on how the ethnographer–social worker–participant triad reorients the researcher's gaze from perpetual observation of catastrophe and misery to active documentation of attempts to mobilize social service resources, obtain physical and mental health care, and undertake other supportive mediation. Intervening to achieve well-being and assessing better ways to do so are commonplace in clinical social work, which takes human suffering as its starting point and applies itself to relieving this distress (Brandell 2010; Wachholz and Mullaly 1997). By contrast, other social science disciplines concentrate on describing, correlating, explaining, or theorizing human suffering as the primary analytic object: the conditions that produce it, the behaviors in which it manifests, and its impact on relationships and institutions such as family life or the economy. We would argue that this tight analytic focus on a “suffering subject” (Lopez 2014; Robbins 2013) and the mandate to not intervene can limit research findings and mask a deeper level of complexity.
The vast majority of the participants in our study were disconnected from the services they desperately needed because, at first glance, they had no mailing address or consistent phone number, they missed appointments, and they struggled to obtain and keep track of documents. However, once the social worker actively intervened to resolve these barriers, massive institutional irrationalities and dysfunctions continued to block participants’ access to health care, housing, SSI, and other resources. Helping people through the first steps of connecting to services permitted us to comprehend much more clearly the underlying structural forces at work in their institutional encounters. Thus, applying a social science theoretical lens to the clinical intervention process enriched our analysis of the (mis)management of suffering by social institutions by giving us a unique vantage point from which to, in the words of professor of urban studies Robert Fairbanks (2012, 547), “elucidate translations of policy mandates in local contexts, [and chart] local variations and complex pathways as well as edges, weak spots, contestations, contradictions, and sites of breakdown/failure.” As such, the knowledge we gained in our study enabled us to locate the mechanisms through which structural violence operates for hypermarginalized individuals at the specific junctures where organizational irrationality intersects with already entrenched and complex adversity. Our thinking thus advanced from striving to understand how and why people suffer in the abandoned spaces of the city to investigating what happens when they are brought into contact with the front line of the institutions that handle the impoverished.
METHODS
Our data come from the qualitative component of a larger study of HIV testing and treatment among people who use drugs and are involved with the criminal justice system. For the main study (led by authors Alex Kral and Jennifer Lorvick), we conducted a quantitative survey with and provided HIV testing to 2,424 people in Oakland, California, who were age eighteen or over and had used crack cocaine or injected any drug in the previous six months. Participants were recruited through targeted sampling methods, with a highly experienced outreach worker conducting face-to-face recruitment in public settings (Kral et al. 2010; Watters and Biernacki 1989; Watters, Cheng, and Kral 1997). The study had two phases. First, all participants received rapid HIV testing and counseling and completed a computer-assisted personal interview administered by a trained interviewer. Next, those who tested HIV-positive (the majority of whom were already aware of their HIV status) were offered enrollment in an intervention study. HIV-positive participants who were not receiving HIV medical care were offered participation in an intensive case management intervention. Those who were already receiving medical care for HIV were enrolled as a comparison group. Both the intervention and the comparison group participants returned for a survey and blood draw (for future HIV viral load testing) every three months until the study's conclusion in December 2014. In our analyses, we determined that the comparison group was too different from the intervention group to draw any meaningful conclusions; the fact that comparison group participants were already enrolled in HIV care at baseline tended to indicate that they had also managed to connect to basic services, apply for and receive SSI, and take other steps toward stability. Therefore, in this article we discuss only the intervention group.
A total of nineteen participants were enrolled in the intervention between November 2011 and August 2013; all of them received intensive case management through December 2014. Participants included four cis-gender women, two transgender women, and thirteen men. At the time of study enrollment, participants’ ages ranged from twenty-six to sixty-five, with five people in their forties and seven people age fifty or older. Sixteen identified themselves as African American, one as Latino, one as Native American and African American, and one as white. No one had stable housing: a dozen people slept in parks, on the sidewalk, in homeless encampments, or in cars, and the rest were temporarily staying in shelters, transitional living facilities, and family members’ residences. Three participants were illiterate. Extreme poverty and social dislocation were not new to this group; during a bio-psycho-social intake questionnaire, the social worker (author Christina Powers) documented extensive histories of trauma, addiction, incarceration, mental illness, and victimization for the majority of participants. The overall portrait of our participants that emerged was one of current life conditions being an extension of decades of intense hardship.
The amount of time the social worker spent with each participant varied widely depending on the individual's need in general (for example, people with a higher level of cognitive functioning often required less hands-on help than those with a lower level of ability) and on the given time (for example, less time was spent with participants during their incarceration, but more time was spent with them immediately after their release from custody). During periods of high need, it was not uncommon for her to spend several hours a day three to four days a week with one participant. In contrast to many social workers who are tethered to a specific office, our case management model supported the social worker's full liberty to literally “meet people where they were at”—which included homeless encampments, parks, and street corners—and to accompany them to medical, social service, and other appointments. She also had authorization to visit them when they were incarcerated in the local county jail in order to develop treatment plans for when they would be released. She kept case notes on each interaction with a participant and met weekly with a clinical supervisor who cosigned the case notes and provided supervisory guidance and the opportunity to confer about specific participants or issues. In addition, she held weekly case conferences with an HIV medical provider who was a part of the research team in order to discuss participants’ health issues, such as medication adherence, psychiatric developments, and emergency room visits.
Ethnographic data collection (conducted by authors Megan Comfort and Andrea Lopez) began several months after the case management intervention began. We had an initial sense of the serious challenges facing our participants, nearly all of whom were “triple diagnosed” with HIV, serious mental illness, and substance use issues. We also understood from conversations with the social worker how important it was that participants learn to trust her as a clinician, and we hesitated to introduce extra parties into their interactions for fear of confusing participants regarding her role as their primary support person and advocate. We therefore began our qualitative inquiry by gathering data from the social worker herself in the form of weekly, audio-recorded debriefs during which she reviewed her caseload person by person and her calendar day by day with the ethnographers, a procedure we followed for the duration of the case management intervention. In addition to these meetings, several times a week the social worker emailed the ethnographers updates on participants and key events as well as photographs of locales, signs, forms, and other items of interest.
As the social worker's relationships with the participants solidified, she began introducing the ethnographers to those who were mentally and physically well enough to handle the presence of an additional person. All participants had signed consent to be in the research study when they were enrolled in the intervention (and they were regularly reminded of the research component when they completed their quarterly interviews and blood draws), but at this first meeting the ethnographer would conduct an additional informed consent procedure. She would explain what participant observation would entail, clearly establish that she was not a case manager or therapist, and emphasize that the participant could ask her to leave at any time or refuse to participate in any recorded interviews without jeopardizing the receipt of case management services. The ethnographers then made arrangements for one of them to periodically accompany the participant and the social worker on case management–related outings, such as medical appointments.
The degree of direct contact between the ethnographers and the participants varied considerably, guided by the social worker's assessment of a participant's overall mental health stability and of how well the participant could differentiate between the ethnographer's role and the social worker's role; that is, the social worker ensured that the participant did not see the ethnographer as an additional case manager, a view that could have jeopardized feelings of trust and emotional safety if expectations were not met. For those who chronically struggled with serious mental disorders, our protocol was to gather data exclusively through the social worker's report and not risk further disrupting their mental fragility by direct observation or burden these participants with ethnographic interviewing. There were eight participants whom the ethnographers did not meet: four of them were either nearly always incarcerated or in very sporadic contact with the social worker during the study period. In the remaining eleven cases, the ethnography encompassed outreach visits, running errands such as obtaining a money order or picking up a prescription, and attending medical appointments, including mammograms and fittings for dentures and for a prosthetic eye, as well as routine HIV care. Over time, one or more in-depth recorded interviews were conducted with seven participants whom the social worker assessed to be mentally stable (for instance, not veering toward psychosis or paranoia); sometimes these interviews focused on a specific topic of analytical interest, and other times they were used as opportunities to gather more general narratives about a participant's past or present. Importantly, calibrating our ethnographic engagement was an iterative process; as called for by each participant's dynamic psychological state, we alternated between periods of initiating direct contact and periods of stepping back.
WEATHERING SEVERE MENTAL HEALTH DETERIORATION
Multiple participants in our study experienced episodes of psychological deterioration to the point of profound disorientation and psychosis. Often these episodes were characterized by auditory hallucinations (hearing voices, receiving messages from the television or radio) and extreme paranoia. Fluctuations in mental health contribute to the attrition of hypermarginalized individuals in research; the loss of contact with “reality” can easily result in a participant losing contact with researchers, depleting study samples of people who experience severe mental health difficulties. In addition, interacting with someone in such a compromised state without adequate clinical training to address mental health vulnerability raises serious ethical questions and poses risks to both the participant and the researcher.
For instance, Charlie1 was forty-six years old and had an AIDS diagnosis when he enrolled in the study in June 2012.2 Upon his release from jail a month earlier, a psychiatrist had diagnosed him with social phobia, paranoid-type schizophrenia, generalized anxiety, depressive disorder, cannabis and cocaine dependence, and alcohol abuse. Charlie was living in a homeless shelter when he joined the study, and two months later he moved into a transitional living facility for former prisoners. He reported having spent the majority of his adult life behind bars and never having had stable housing outside of a correctional facility, in large part because his felony record caused his applications to be rejected. With the social worker's help, he applied and was approved for subsidized housing through Housing Opportunities for Persons with AIDS (HOPWA); he moved into a modest studio apartment in January 2013. Lacking a bed, he slept on a donated comforter that he folded into the size of a cot, remarking that he was used to sleeping on small surfaces from his time in jail. The first time he entered his apartment, he stopped by the building manager's office and told him he was going upstairs, then became flustered when the manager informed him that he could come and go as he pleased. “I thought I had to check in,” he mumbled sheepishly to the social worker.
Despite the progress Charlie made in obtaining housing, the social worker observed a progressive decompensation of his mental health in late 2012 and early 2013. Charlie tried to self-medicate his intense anxiety by increasing his drug use, which in turn gave him reason to avoid checking in with his parole officer. He was reincarcerated briefly in early February for the technical parole violation of failing to report his new address and released on March 1 without a supply of his HIV or psychiatric medications.3 He returned to his studio apartment but was jailed again on May 14 on new charges, which were subsequently dropped. This time, he did not receive HIV or psychiatric medications during the two weeks he was incarcerated or upon release. In late June, Charlie experienced a psychotic break, believing that the police were spying on him via a microchip implanted in his body, a drone that followed him, and an airplane that hovered over his residence. He stopped entering his apartment through the building doorways and instead climbed through a window using a ladder, thinking that the hallway outside of his door was under surveillance. Subsequently, he started sleeping in the bushes in order to avoid the apartment altogether because he worried the police would find him there. He was reincarcerated yet again, and the social worker received a call from an attorney who during a legal visit had found Charlie hiding behind a chair with a correctional officer laughing at him. When the social worker visited him in jail shortly thereafter, Charlie had not received any HIV or psychiatric care during the two weeks he had been incarcerated, despite his pleas for medications to help make the drone go away. He was covered in sores from an untreated rash that he scratched until he bled, and he sobbed in distress as they spoke.
Being able to chart this series of events was critical in building our understanding of what constitutes “stable housing” for those whose existence is characterized by severe deprivation not only of economic resources but also of mental and physical wellness. In the absence of psychiatric treatment, and with the omnipresence of criminal justice surveillance, Charlie's mind transformed a safe place to call his own into a treacherous site where he felt exposed and hunted. His struggles clearly indicated his need for a therapeutically and medically supportive living environment as well as a respite from the unrelenting cycle of parole check-ins, police stops, and jail stays. Charlie's ordeal was also instructive for problematizing the concept of correctional settings as “public health opportunities” (Glaser and Greifinger 1993), given his clear need for medical intervention and the lack of response to his anguish when he was behind bars. Yet following a terrified man for the purpose of documenting his trajectory while he was caught in a web of hallucinations would have posed significant and potentially insurmountable ethical issues, and doing so could have compounded the already enormous suffering he was experiencing. Interacting with an ethnographer whom Charlie knew to be someone who was recording his conversations and taking notes on his activities could have aggravated his paranoia, and it would have been logical for him to view the researcher as another part of a malevolent surveillance system. Charlie might even have become violent if he felt the need to protect himself from what he perceived as a tightening circle of intrusion, or if he experienced auditory hallucinations commanding him to strike out. However, avoiding contact with Charlie during a time when he clearly needed intensive help would have left him even more exposed to harm, including eviction, victimization, or self-inflicted injury. In addition, his psychosis was jeopardizing his recently acquired and highly sought-after housing, without which he would be subject to more health and mental health risks.
As a clinician providing intensive case management, the social worker had a clear role that was recognized both by Charlie and by various institutional actors during this tumultuous time; as such, she was able to advocate on behalf of her client while also closely observing his experiences and communicating these observations to the research team. In contrast to ethnographers, who typically seek to limit the impact they have on situations in order to enter as fully as possible into participants’ social worlds, a social worker strives to build a “therapeutic relationship” with participants to effect changes in their perspectives and behaviors (McWilliams 2004). Using techniques such as cognitive behavioral therapy, motivational interviewing, and trauma-informed care, a social worker creates a “holding environment” for the relationship in which the individual can feel physically and psychologically safe (Altman 2009; Thompson and Cotlove 2005). For Charlie, this sense of a safe connection to the social worker persisted throughout his mental health decompensation, as evidenced when he asked permission to call her from jail, periodically dropped by her office to request help, and explained to her in detail the various ways in which he believed he was being followed by the police, his strategies for avoiding them, and what the voices he heard were saying to him. From an ethical standpoint, not only did the social worker's presence not incur more risk for Charlie, but she also provided support and worked with him to create a safety plan to mitigate further risks to his well-being.
Importantly, other institutional actors also reached out to the social worker during Charlie's travails: she received calls from or met with the manager of his apartment building, his community-based HIV doctor, the coordinator for HIV care in the county jail, and the attorney who found Charlie hiding behind a chair during the jail visit. Due to Health Insurance Portability and Accountability Act (HIPAA) protections, these interactions were only possible because earlier in their work together the social worker had obtained signed release forms from Charlie giving her permission to communicate with each person in her role as a clinician. Again, these contacts were multifunctional; they served the research purposes of keeping track of Charlie's whereabouts and also provided ways of verifying specific details of his experiences. (For example, the building manager discussed having found the ladder Charlie used to climb into his apartment, and the jail HIV coordinator confirmed that he had not received his medications.) In addition, having access to these institutional actors during a chaotic and troubled time provided the social worker with information and leverage that she could use to help Charlie (for example, asking his community-based HIV doctor to call the jail HIV coordinator regarding his medications, or negotiating with his building manager to avoid eviction). The social worker therefore served as a link between various silo-ed institutions with which Charlie interfaced—a link that simply did not exist in the normal operations of the institutions, despite the overlapping social problems that Charlie was facing. The social worker worked to establish relationships in different institutional arenas in order to facilitate the critical ongoing communication that service providers acknowledged was necessary, but was rarely seen, to stabilize someone with Charlie's complex profile.
INTERVENING IN CRISIS SITUATIONS
Living in a state of perpetual crisis often characterizes deep poverty, and many crises are related to health. From November 2011 through December 2014, five of the nineteen people enrolled in the case management intervention were hospitalized for illness, two were hospitalized for injuries suffered when they were assaulted (one was assaulted and hospitalized multiple times), four were placed on involuntary psychiatric custodial holds, two overdosed on heroin (and survived), and one was shot (and also survived; in total, seven participants had been shot in their lifetime). In addition, fourteen were incarcerated at least once; ten of them returned to jail over ten times.
Studying people who are cycling in and out of correctional, medical, and psychiatric facilities while undergoing life-threatening crises has prompted us to revisit the principles of ethical research and our responsibilities to participants. Institutional review boards (IRBs) typically consider potential “harm” in research as a risk stemming from the research itself—for example, a participant's discomfort with an interview question, a breach of confidentiality, or a side effect from a drug. More than once, however, we found ourselves challenged to intervene on behalf of a participant to protect him or her from harm from a source external to the research. For ethnographers, the imperative to take action under these circumstances can be clouded by questions as to whether it is scientifically appropriate to involve oneself in directing events that are under empirical study. Too much interference may disrupt the patterns and processes being investigated, and ethnographers who do intervene find themselves having to justify their acts (Rios 2011). Further, the likelihood that even ethnographers who are steeped in “applied” approaches to research lack the clinical skills to interact with a population with acute medical and mental health issues makes the question of “intervention” much more ethically complicated with hypermarginalized populations.
Social workers begin from an entirely different point of departure: their presence in a client's life is organized around intervention, and they are equipped with clear guidelines about the ethics of doing so as well as models of practices that have been implemented and analyzed in the discipline (National Association of Social Workers 2008). In our study, the frequency with which crises arose that required clinical intervention affirmed for us that in research with hypermarginalized participants, a protocol for the full “protection of research subjects” ideally would include a highly trained and experienced clinician with competency in crisis management.
A particularly salient example is Hook, who was twenty-six years old and had known he was HIV-positive for three years when he enrolled in the study in May 2012. A sharp dresser with a compact build, Hook had been released from San Quentin State Prison ten days prior to his first meeting with the social worker. He also had an extensive history of mental health issues, with diagnoses of schizoaffective disorder and bipolar disorder and regular auditory hallucinations of voices that he called “the committee.” He described himself as “raised by the State of California”: he was placed in foster care at age five, entered the juvenile justice system at fourteen, and had not spent more than three consecutive months out of correctional facilities during his adulthood. Ironically, the state was no longer officially responsible for Hook. He had been “realigned” from state parole to county probation as a “non-violent, non-serious, non-sexual offender” under relatively new legislation that aimed to reduce the California prison population.4 In an interview, Hook expressed bewilderment at being included in this group: “I'm a true violent offender! I've never been arrested for drugs or anything, just violence.” Being transferred out of the state system had made Hook ineligible for any of the mental health services or housing resources provided to parolees.
Four days after enrolling in our study, Hook was shot in the arm while standing outside of the homeless shelter where he had been staying. His assailant was a man confined to a wheelchair owing to his own gunshot injury years prior. The attack was captured by video surveillance, but despite camera footage and a name, no arrest was ever made. Hook was required to move out of the shelter after he was released from the hospital because staff feared that the perpetrator, who had been seen hanging out in the neighborhood and had specifically targeted Hook, might return to look for him. Two weeks later, during a meeting with the social worker, Hook requested that she take him to the county psychiatric hospital because he feared he was having a mental breakdown. After several hours there, Hook reported that he felt better, and because he did not meet the technical criteria for a “5150” involuntary hold, he was released from the hospital.5 The social worker took him directly to a transitional housing program for former prisoners and negotiated to have him admitted. If she had not been there, it is unclear where he would have slept that night, since the hospital did not have mechanisms in place to ensure that a person has somewhere to stay when released.
Over the next four months, Hook was in and out of touch with the social worker as he resumed using crack cocaine, was expelled from the transitional housing program, lost thirty pounds from his muscular frame, and stopped taking his HIV and psychiatric medications. During a visit with Hook at his mother's residence in mid-August, the social worker noted signs of an impending psychotic break, but he was unwilling to accompany her to see his doctor. In late September, Hook called the social worker from the county hospital where he was being treated for an as-yet unidentified illness, and she visited him there. The hospital staff reported that he had been threatening them and they had called security multiple times; that night Hook left the hospital against medical advice. The next day a physician called the social worker to tell her that Hook had been diagnosed with methicillin-resistant staphylococcus aureus (MRSA) in his bloodstream and that it would probably be swiftly fatal if left untreated. The social worker received this phone call only because Hook had signed the release of information form, which allowed communication between her and his doctors. Even knowing that the infection would likely be fatal, without the social worker, the hospital had no formal institutional procedures to locate Hook to inform him of his grave situation.
In the social worker's clinical evaluation, Hook's mental health had decompensated to a point that he was unable to understand the gravity of his medical situation and therefore had not made an informed decision to refuse treatment when he left the hospital. After consultation with the full research team, she and an outreach worker who had known Hook for years through the homeless shelter went to look for him. They found him at a house known to be frequented by drug users, and the social worker explained the MRSA infection to him and his risk of sudden death. Drawing on established therapeutic models of intervention, she framed returning to the hospital for treatment with her support as a proactive choice that Hook could make to save his life. He decided that he would return, and upon arrival the social worker helped ensure that he was readmitted immediately rather than forced to wait in the extremely busy and chaotic emergency room. Soon thereafter, he nonetheless began exhibiting psychotic symptoms and acting menacing toward the medical personnel, at which point he was deemed a threat to others and forcibly restrained under the 5150 code by three deputies (two of whom he recognized from the county jail).6 The social worker was allowed to be by Hook's bedside once he had been administered antipsychotic medications, and she remained there into the evening, advocating for his needs and supporting him emotionally. When she went to visit him the next day, he was calm and no longer restrained, although he had been moved to a respiratory isolation unit and the social worker had to don a hospital gown, gloves, and facial mask before entering. An infectious disease specialist spoke with her and said that the MRSA had spread to Hook's lungs and would require between four and eight weeks of treatment with intravenous antibiotics.
As with Charlie, Hook's trajectory from prison through the homeless shelter to the streets with several loops through the emergency room provides rich data for analyzing correctional, social, and medical systems from the vantage point of the extremely poor. From being ejected from the shelter for being a gunshot victim to being allowed to walk away from life-saving medical treatment while submerged in psychosis, Hook's repeated encounters with “care” institutions deepened his deprivation and placed him at high risk of death. In retrospect, it is hard for us to fathom what our experience with Hook would have been like without the social worker guiding us. From the early days of his study involvement, Hook repeatedly cycled through medical and psychiatric settings that would have been off limits to someone who was not a clinician and did not have formal clearance under HIPAA to be privy to confidential information. When the physician notified the social worker of Hook's MRSA diagnosis and the danger it posed, he was only able to do so because of this clearance. The social worker's clinical understanding of Hook's mental health issues alerted her to the need to intervene quickly, and she possessed the therapeutic skills necessary to do so, along with on-the-ground knowledge of where he might be located. Chillingly, the most likely scenario for Hook's study participation had the social worker not been a part of the team would have been a rapid loss of contact as his psychosis set in, followed by news of his death from an untreated MRSA infection.
TURNING THE LENS ON POVERTY-PROCESSING INSTITUTIONS
Upon entering the study, twelve participants had been in correctional facilities within the last year, but their contact with other institutions was minimal. Only one person possessed an official California state identity (ID) card, three had makeshift correctional ID cards that identified them as parolees, and fifteen had no form of ID. Apart from one person who was able to work (tellingly, the same person who had an official ID), all participants were eligible for some form of government support, yet three were receiving nothing at all, and nine were receiving substantially less than they were due (for example, only food stamps when they were eligible for monthly income through SSI).
Impoverished people, especially those with the compounded issues of mental illness and drug use, are often held responsible for their inability to connect with social services, and indeed, it can be difficult to determine whether individual issues, such as illiteracy or a tendency to miss appointments, prevent them from accessing resources. In our study, the early steps of case management typically centered on securing basic survival needs, such as making doctor appointments, obtaining a state ID card, seeking a shelter bed, applying for SSI, and connecting to resources such as food pantries, free meal programs, and subsidized bus passes. The social worker's repeated undertaking of these activities—often multiple times with the same individual, as even brief incarcerations or hospitalizations resulted in lost ID cards and derailed application processes—provided the opportunity to observe how even a skilled social welfare professional working carefully to follow bureaucratic procedures, submit required paperwork, and attend mandatory appointments continually encountered major setbacks and institutional barriers.
Returning to Charlie's story provides an apt example. Upon release from his jail stay for the parole violation of not reporting the address of his new subsidized apartment, Charlie and the social worker attended a check-in meeting with his parole officer. At this meeting, they were informed that Charlie would not be allowed to use his apartment as his address of record for parole because it was in the county adjacent to the one in which he was required by parole to live. The social worker explained that Charlie would be homeless if he could not stay in the HOPWA housing and asked if special accommodations could be made, such as applying for a transfer of Charlie's county of parole. The parole officer refused and reiterated that Charlie would be considered to have violated his parole if he occupied his apartment.
The social worker's account of this conversation was invaluable for instructing us about the significant housing challenges faced by parolees (not to mention challenges faced by drug-using, mentally ill, HIV-positive parolees). The parole officer's unwillingness to help problem-solve the conundrum that a man could be condemned to homelessness when housing had been made available to him because of a medical condition, even when asked for such help by a case manager working with the parolee, brought into sharp relief the Kafka-esque ways in which highly bureaucratic systems create additional hardships for the people under their control (Lipsky 1980). The total disconnect between the parole administration and the low-income housing system, even when the social worker attempted to mediate between them, rendered what should have been a milestone in improving Charlie's quality of life into a further barrier to his well-being. Not incidentally, this conflict fueled Charlie's perceptions that the police were hunting him and greatly complicated the social worker's efforts to help him manage his mental health: whereas the microchip, drone, and airplane were hallucinations, it was utterly (sur)real that he was prohibited to reside in an apartment leased to him and for which he was paying rent, especially considering that he had been chronically homeless and adequate, affordable housing in the San Francisco Bay Area is extremely hard to come by.
Ideally, partnerships between parole officers and social workers could advance common goals of helping marginalized people obtain housing, access health care, find employment, and reconnect to family life. This possibility could occasionally be glimpsed when an “old-school” parole officer facilitated the social worker's request to refer a participant to drug treatment, as would have been more likely to happen in the past, when community supervision professionals commonly had a background in social work (Lynch 2000; Simon 1993). More often than not in our study, however, we documented not only a disintegration of the potential alliance of social and parole services but outright contention, typically when parole officers disallowed housing options or threatened reincarceration for drug use relapse and the social worker could no longer be confident that information she possessed would not be used against the parolee. These communication breakdowns generated additional work for the social worker, who had to help participants develop strategies for managing the problems stemming from parole restrictions, which also often affected others in the participant's network.7
Cadillac's case is illustrative. Cadillac was fifty-one years old when he enrolled in the study in September 2012. He had been diagnosed with HIV while in jail in 1994, the same year he was viciously assaulted with a baseball bat and suffered a traumatic brain injury that left him with significant cognitive impairment. Like Charlie, Cadillac had a viable residence that was outside of his county of parole—the house of his oldest sister, Sherry. Sherry cared deeply about Cadillac and struck up a regular, mutually supportive communication with the social worker by text and phone as soon as Cadillac signed consent for them to be in touch. In the autumn of 2013, when her brother was released after one of his many jail stays, she offered to let him live in a comfortable “in-law” unit connected to her house while he attended an outpatient drug treatment program during the day. The parole officer denied authorization for this arrangement because Sherry lived in a county that was adjacent to Cadillac's county of parole, not in it. Cadillac then turned to his sister Linda, whose residence was in his assigned county and therefore would be acceptable to the parole officer. Linda was less financially secure than Sherry and had struggled recently with her own drug addiction; nevertheless, she hesitantly agreed to let her brother sleep in the living room, all the while expressing misgivings about the destabilizing impact he might have on her sobriety. Soon thereafter, the parole officer came to the residence for a routine check, but he forgot to bring Linda's apartment number. When Linda returned home from work, she was mortified to learn that the parole officer had knocked on nearly every one of her neighbors’ doors looking for her brother. Furious, she called the parole officer and said that Cadillac could no longer live with her. The parole officer retorted that being homeless was a violation of Cadillac's parole, and if he no longer had a residential address, a warrant would be issued for his arrest. In the wake of this conversation, the social worker and Sherry were caught in a flurry of attempts to find an available shelter bed in Cadillac's county of parole while trying also to persuade Linda not to put her brother out until he had a place to go, especially given that her embarrassment and anger were a result of the parole officer's actions, not Cadillac's. At her limit and worried about her own health, Linda refused. The shelters were full. The social worker strategized with Cadillac how to safely store and remember to take his HIV medications when living outside. And Sherry painstakingly assembled a backpack for her brother with clean socks, disinfectant wipes, lip balm, a blanket, and other survival items for life on the streets.
In analyzing the penal regulation of poverty in neoliberal societies, the sociologist Loïc Wacquant points to how the social ills afflicting the destitute are cast as problems to be managed by the disciplinary, regulatory “Right Hand” institutions of Pierre Bourdieu's “bureaucratic field,” and he notes that police, courts, and corrections figure prominently among these state actors (Bourdieu 1992, 1993; Wacquant 2010, 2014). Charlie's and Cadillac's housing struggles are apt examples of Right Hand management: despite vigorous efforts by the social worker and their family members to connect them to supportive “Left Hand” institutions, the final authority rested with the parole officers. Yet, even when participants were handled by hospitals and social service programs, the chronic draining of funding from Left Hand institutions under neoliberal governance left them exposed to neglectful, irrational, and punishing treatment.
Take the case of Crystal, who was thirty-five years old in March 2012 when she joined the study. At that time, Crystal was on probation and parole, although she spent the vast majority of her time in the study in jail; over the ensuing months, we documented her being released and reincarcerated in as few as four days. A survivor of a childhood and adolescence saturated with physical, sexual, and emotional abuse, Crystal had been living on and working the streets since age eleven and had been using crack and heroin since fourteen. Her social network had remained relatively insular and continued to include a man several decades older than her with whom she had been sexually involved since arriving on the streets and who regularly assaulted her. After one attack, Crystal was found bloody, disoriented, and wearing only her underwear by a passing motorist, who stopped to help and phoned the social worker at Crystal's request. The social worker called an ambulance and met Crystal at the hospital, where they spent the afternoon and evening waiting for examinations and X-ray results. Around 10:00 PM, Crystal's wounds were determined not to need further treatment and she was discharged. The hospital had provided her with a standard-issue gown, but no other clothes. Because Crystal had not wanted to press charges against her attacker (which she fatalistically told the social worker would do nothing more than put her at risk for retaliation), she was not provided with a placement in a domestic violence shelter. She thus was sent out into the dead of night barefoot, in a hospital gown and underwear, with no money, no ID, and no place to go.
The social worker managed to track down a women's shelter that agreed to admit Crystal that night despite her lack of an ID card as required (the social worker provided this from a photocopy the next day, along with clothes and shoes for Crystal) and her decision not to press charges (which became grounds for Crystal being asked to leave the shelter several days later). Had Crystal been on her own, she would have had to figure out a way to traverse the five miles between the hospital and the general area where she usually stayed, and when she arrived there—bruised, exhausted, barely dressed, and barefoot—she would probably have been dependent on her assailant for help, since he was her primary social connection and she had no other resources. When spelled out in detail, such institutional abandonment plainly reads as egregious and stunningly counterproductive. In Crystal's case, however, employees at an overcrowded public hospital were contending with limited resources in a chaotic environment, and they did not violate formal protocol regarding the minimal requirements for intervention (Ansell 2011). This situation nonetheless contributed to the multiple compounded physical and psychological assaults Crystal endured.
Yeheskel Hasenfeld, a professor of social welfare, distinguishes between “people processing” and “people changing” institutions: the former have relatively short-term contact with clients and deploy their activities on the boundaries of the organization, while the latter sustain long-term contact with their wards (students, patients, prisoners) and locate their activities within the center of the organization (Hasenfeld 1972). Over the two and a half years of our study, we observed innumerable instances of what could be called institutional “poverty-processing,” whereby standard procedures, bureaucratic adherence to regulations, and other forms of “business as usual” worked against, jeopardized, or failed to protect someone in need of care to the point of effacing his or her personhood. There was the time when Crystal, who had been creating a plan with the social worker to enter a residential drug treatment program immediately upon her release from jail so that she would not resume her substance use, was suddenly set free without advance warning late on the Wednesday afternoon before Thanksgiving. Unsurprisingly, Crystal had relapsed when the social worker returned to work after the four-day weekend. Being released early from jail before a major holiday may on its surface seem compassionate, but it is much more complicated for people facing compounded adversities who do not have family to welcome them home. If there had been established coordination between service providers inside and outside of the jail, it would probably have been determined that early release on a long weekend and the risks posed in terms of drug use and personal safety would actually be antithetical to the established treatment plan.
Or consider the saga of PeeWee, who decided to turn himself in for an outstanding warrant as a means of getting sober. He binged on his final stash of drugs, poured his heart out to the social worker during a “last supper” at a fast-food restaurant, and then was paradoxically turned away at the jail because the fingerprint scanning machine was broken and they could not verify his identity. Doggedly returning the next day, he was again refused, this time because the clerk erroneously claimed that there was no warrant issued for him. Two weeks later, on the heels of a drug run that left him gaunt and with open sores on his lips from smoking crack, he asked the social worker to take him to the jail for a third time; there he was finally admitted on the original warrant. This scenario points to the perplexing way in which institutional engagement is unpredictable for this population and policies (or lack thereof) create further instability.
And then there was Moan, who after months of homelessness was notified that he had qualified for subsidized housing and was encouraged to come immediately to look at his new apartment. Hours after viewing the studio, as he gathered his scarce belongings in preparation for moving in the next day, the social worker received a call from an administrator who told her that Moan could not live in the building after all because he had failed the eviction check. Eviction records are commonplace among the extremely poor, whose prior evictions are often due to irregular and unreliable sources of income that must be juggled across multiple competing demands; this blemish then perpetuates their housing instability (Desmond 2012). Indeed, in some more progressive areas (such as certain subsidized housing buildings in San Francisco), eviction history is not weighed in housing eligibility because it is assumed to be one of the major bureaucratic barriers keeping people from securing housing. Despite the social worker's vehement protests, Moan was barred from the apartment building and continued to live on the street. Six months later, he was admitted to a respite bed in a homeless shelter owing to his diminishing health and multiple hospitalizations.
The list of instances of failed institutional coordination and de facto punitive institutional practices we compiled is a long one. There is the common practice of releasing people from county jail in the middle of the night, with no option other than to wait huddled at the bus stop until dawn when public transportation starts running. There is the federal Shelter Plus Care program's requirement that an individual be in a homeless shelter or live on the streets in order to be eligible for a subsidized housing unit; thus, while waiting to ascend the eighteen-month waiting list, applicants are technically not allowed to live with family members or to scrape together a temporary housing option. There is the program's further demand that, after being on its waiting list for a year or longer, homeless people must be reachable by phone and then able to produce scads of documents (including photo ID and proof of income) within one week in order to advance to the next step, even though they are not formally connected to a social worker through the housing program to help them prepare for this hurdle.
Having an embedded social worker allowed us to distinguish the features of poverty-processing institutions from the litany of personal difficulties that are levied against the extremely poor as explanations for their own misery. Before we undertook this research, we and many others were well aware that illiteracy, drug use, mental health crises, and sickness make it hard for impoverished people to follow through on the myriad applications and processes they are obliged to complete in order to receive the basic elements for leading safe, healthy lives, and that those living outside the reach of health and social welfare institutions are tremendously vulnerable (Bourgois 2009; Braveman and Egerter 2008; Phelan et al. 2004; Sidel 1998; Wildeman and Muller 2012). In principle, introducing the case management component should have mitigated or resolved many related obstacles: the social worker's phone was never cut off, she was cognitively equipped to fill out countless forms, she stored original paperwork in her filing cabinet and electronic records in her computer, and her bulging calendar reminded her of every appointment and deadline. These skills and resources did help, to be sure: eight participants were approved for SSI, and two received presumptive SSI benefits for a period of time; nine obtained housing; nearly all saw a doctor, and many began taking HIV medications; several of them managed to spend dramatically longer periods of time outside of correctional facilities before being reincarcerated; and (not to be taken for granted) everyone stayed alive. And yet, for each accomplishment there were multiple setbacks as the social worker and participants navigated the maze of underfunded, rigidly bureaucratized institutions that served more as gatekeepers withholding the protections of social integration than as conduits to those protections. These setbacks would have been challenging for anyone, but in the context of the chronic suffering, mental health crises, and acute health emergencies of our participants, their emotional and physical tolls were even more dire.
CONCLUSION
Through our hybrid approach to studying the experiences of HIV-positive, indigent adults navigating various institutional contexts with the assistance of a clinical social worker, we were able to deeply probe questions of how and why very poor and very sick people continue to be severely deprived of the resources they need. This research yielded substantive findings with concrete policy implications and generated methodological reflections on the ethics of conducting research with hypermarginalized populations. Ethnographically documenting the provision of intensive case management by an experienced clinician permitted us to enrich our analysis with a perspective from the “inside” and to gain information about the precise sticking points, glitches, and breakdowns of institutions rather than having to settle for superficial explanations such as “the patient did not show up for the appointment” or “the client's phone has been disconnected.” Directly intervening in participants’ lives highlighted that the social worker–participant dyads encountered major obstacles when a single institution could not address the participant's needs beyond a relatively narrow purview and that cross-institutional collaboration was required. Highly salient in these scenarios was organizational irrationality, not only in the inability of single institutions to take a broader perspective on the complex problems that afflicted their patients or clients, but also in the massive disconnect between institutions—to the point that they undermined each other by operating at cross-purposes.
The social worker's efforts were paramount in building our understanding of the irrationality at play. The fact that a skilled professional working full-time to obtain services for her clients by bridging agencies, fostering connections, and trying to reconcile the requirements of multiple entities met with considerable opposition and formidable stumbling blocks forcefully revealed that correctional, medical, and social service institutions have not been designed to be comprehensible and accessible, least of all to hypermarginalized individuals. This tops our list of policy implications stemming from this research: just as university systems are expected to be intelligible to their students and retirement programs provide clear counsel and support to their investors, institutions delivering services to extremely poor people should be obliged to make their procedures understandable to and usable by those who show up on their doorstep. Having social workers be a part of this effort would be ideal, particularly for people who are new to the institution or in crisis, but the widespread need for them could be alleviated if procedures were more transparent, coherent, and explained in clear language. Flexibility with regard to the specific issues of poverty (for example, lacking access to a phone or being unable to make photocopies or fill out complicated forms) is also critical.
Our findings also support the necessity of increasing institutional capacity in public hospitals, social services, and community supervision settings (for example, probation and parole) to more fully serve a hypermarginalized clientele with complex and overlapping needs. Narrow service mandates necessarily lead to “poverty-processing,” as there is no way to meaningfully improve the well-being of a severely deprived patient or client by focusing on one problem in isolation. Under these circumstances, overwhelmed service providers may become fatalistic and reduce their assistance to the bare minimum, knowing that their efforts will soon be undone by forces outside of their control. Care and treatment of an individual become reduced to managing the immediate issue at hand as swiftly and economically as possible, without violating rules but also without consideration of other conditions that contribute to his or her suffering. The impoverished are continually processed through the “institutional circuits” (Hopper et al. 1997) that may provide temporary relief but lack the breadth and resources to effect actual change. The resulting frustration often destabilizes the client's health and mental health even further. We can imagine the difference it might have made, for example, had a psychiatrist been included in Hook's treatment team as soon as he was admitted to the hospital with what turned out to be a MRSA infection. With appropriate support and advocacy, he might never have left the hospital against medical advice, and his extended stay for intravenous antibiotics could have provided an opportunity for therapeutic services such as substance use treatment, mental health counseling, and stabilization on antipsychotic medications.
When a “one-stop shop” for comprehensive services is not feasible, it is critical to establish coordination and dialogue across institutions at the local and state levels. In the course of the three years of our study, the social worker was able to establish strong professional ties with several people in key positions at the county jail, Oakland's main homeless shelter, a legal clinic for indigent clients, and a nurse at an HIV clinic. These partnerships provided a glimpse of how efficiently integrated services could operate when there were willing parties on both sides and clients had signed HIPAA releases. For instance, the HIV services coordinator at the county jail, rather than requiring the participant to place a collect call from a public phone monitored by correctional officers, would call the social worker while the participant was in her office to coordinate care, and the three would talk on speakerphone. The homeless shelter coordinator would contact the social worker when a respite bed opened to ask whether a participant was in need. The social worker would alert attorneys at the legal clinic when a participant was incarcerated so that they could schedule a meeting at the jail and applications for housing or government benefits did not get derailed by missed appointments. And the nurse would send a brief update to the social worker after conducting a home visit to let her know if a participant seemed destabilized. Importantly, all of this cooperation occurred informally: the individual actors reached out to each other, obtained the necessary releases, and communicated among themselves without direction from the organizational level to do so. Although highly beneficial to the participants—not to mention a time-saving collegial resource for the professionals—there was neither institutional support for this kind of service integration nor any official mechanisms for creating or sustaining it. When an individual left a job, the collaborative tie was broken, and it was not always possible to reestablish it. Institutional integration therefore must be created with attention to developing formal cross-agency policies, providing training to employees on collaborative protocols, and dedicating adequate financial resources to these processes.
Our hybrid approach not only generated data from an embedded perspective that enriched the policy relevance of our findings but also provided very direct assistance to some people living in deep poverty. Such a methodology is familiar to social work and public health researchers but is less accepted as “scientific” by social scientists, who consider it ethically problematic. There is a school of thought in the social sciences that intervention invalidates findings because the researcher becomes implicated in the processes under study, altering them in ways that might not have “naturally” occurred. We engage this argument on several levels. First, as discussed at length in this article, intervention can move research into new realms of complexity by permitting us to probe beneath the obvious: merely documenting that an illiterate person with mental health issues and a criminal justice record has a hard time filling out the required paperwork for his SSI and housing applications is not particularly innovative or helpful. But realizing that SSI and housing remain elusive to him even with the support of a clinical social worker sheds light on institutional operations, structural violence, and the types of reforms necessary to address hypermarginality. In sum, with an appropriate research design, intervention does not impede our learning but rather helps us learn more.
Second, we must consider the ethics of non-interventionist research with hypermarginalized populations. Is it ethical for us, as individuals with careers, research funds, and informed ideas about how to improve people's lives, to be bystanders documenting other human beings’ suffering when we have the option to do otherwise? We take the stance that in the current era of poverty research, scholars should challenge themselves and each other to develop methodologically rigorous and ethically grounded studies that confer concrete benefits on participants, such as linkage to services, direct assistance, and other forms of help. To be sure, the mode of intervention in each study must be tailored to the researchers conducting it and the resources available and designed with extreme caution and sensitivity. Underestimating the level of skill and expertise required to meet the needs of hypermarginalized people risks exacerbating their already entrenched and overwhelming issues. Our experienced clinical social worker was equipped with over a decade of training in the unique needs of this population, trauma-informed care, and crisis intervention. Clearly, graduate students conducting solo fieldwork who have no clinical training should not attempt to provide “case management” simply because they “want to help.” Direct services should only be provided with appropriate clinical training and not simply with the intention of “doing good” for research participants.
Scholars at every level of seniority, however, in concert with IRBs, could think about the obligations we may have to participants beyond protecting them from potential harms of the research. Many ethnographers and others conducting field-based studies regularly face this question when participants need a ride, a meal, help finding a service agency's phone number, or some other low-grade favor. Although appropriate responses are usually not codified in the research protocol, it is commonplace for researchers in the field to provide these resources out of feelings of ethical compulsion and social norms of reciprocity, even if they often downplay or do not mention such responses in publications, owing to concerns about scientific acceptability. We encourage more dialogue and openness around these issues, in the vein of the sociologist Victor Rios's (2011) argument that practical support should be inherent to the study of disadvantaged groups. Acknowledging that it is ordinary to lend a hand when participants are in need would legitimate such interactions, animate discussions with colleagues and IRB members about best practices for engaging in them, and lay the groundwork for the broader inclusion of intervention in social science research.
Validating and adopting “compassionate practices” that can be used by individual researchers is only a first step. As Rios (2011) has noted, simple acts of help are unlikely to produce actual change in participants’ lives, even if they are meaningful interpersonally and ethically. The solid foundation of knowledge developed by previous poverty research and the desperate circumstances of contemporary hypermarginalized populations call us to move forward and embrace intervention research that can improve participants’ well-being and identify concrete policy recommendations for real structural change. Such research requires adequate funding to hire qualified practitioners and support them to do difficult work. In our study, we did not encounter any drawbacks to providing the intervention, perhaps in large part because our social worker had extensive training and experience and we were diligent about providing her with clinical supervision, a mileage allowance so she could be mobile, paid vacation and sick leave, and other optimal work conditions so that she could do her job well and avoid burnout. Conducting a cost-benefit analysis was beyond the scope of our study, but future research could compare the cost of resources dedicated to high-quality staff support and the expenses avoided or curtailed through reduced recidivism, decreased need for emergency room care, and other cost savings that accompany improvements in a population's social stability and mental and physical health.
We note that our efforts to stabilize our participants—to move them off the street and into housing, to provide them with a steady income by signing them up for SSI, to improve their health by connecting them with medical care—were an uphill battle and at best produced incremental improvements rather than radical transformation. We take this to be a substantive finding, but not a methodological justification. Had our intervention vastly ameliorated people's circumstances and thereby significantly changed empirical outcomes, our study design would have been equipped to understand the processes by which that happened. Again, this possibility highlights the utility of making intervention itself the focus of study in order to excavate and explicate the institutional conditions, structures, and relationships that promote or inhibit change, rather than fixating on suffering subjects with no intention of altering their plight. With the collaboration of skilled practitioners, appropriate ethical guidelines, and a rigorous study design, there is no reason for researchers to hold back from working to effect social change.
Acknowledgments
This research was supported by funding from the National Institutes of Health (R01MH094090, PI: Kral; R01DA033847, PI: Comfort; and R01MD007679, PI: Lorvick). We thank Jay Borchert, Elizabeth Kita, Reuben Miller, Victor Rios, Megan Tompkins-Stange, Damian T. Williams, and the participants in the Russell Sage Foundation “Severe Deprivation in America” conference for their invigorating and helpful feedback on drafts of the manuscript. We are particularly grateful to special issue editor Matthew Desmond for his detailed comments and insightful suggestions. And we are tremendously appreciative of and indebted to the people who participated in our research, and hope that we have done justice to the experiences they shared with us.
FOOTNOTES
↵1. All names are pseudonyms. We thank “Cadillac” for suggesting names for this purpose.
↵2. Acquired immunodeficiency syndrome (AIDS) is a late stage in HIV disease. Taking HIV medications can maintain the virus in a latent stage for decades, during which HIV-positive people do not experience symptoms of illness. If someone with an AIDS diagnosis does not receive medical treatment, life expectancy falls to one to three years.
↵3. Under the Eighth Amendment, jails and prisons are required to provide health care, including prescription medications, to incarcerated people. Policies for providing a supply of medication for people upon release from custody vary according to the jurisdiction of the correctional facility. The policy for the county jail in which our study participants were detained was to provide a thirty-day supply of medications for people leaving the jail. Nonetheless, we documented dozens of instances when people were discharged without any of their prescribed medications.
↵4. California Assembly Bill 109 is referred to by the California Department of Corrections and Rehabilitation as the “cornerstone of California's solution to reduce overcrowding, costs, and recidivism.” Implementation of “realignment,” as it is colloquially known, began in October 2011.
↵5. Under section 5150 of the California Welfare and Institutions Code, a clinician or a criminal justice authority can confine people against their will for seventy-two hours if they are considered to be a danger to themselves or others owing to a mental health disorder. The term “5150” is commonly used colloquially to refer to being placed on such a hold (for example, “She can't come today, she was fifty-one-fifty-ed”) and as a synonym for “crazy” (for example, “When he told me I couldn't have what I wanted, I went fifty-one-fifty”).
↵6. For a discussion of the overlap of criminal justice authorities in correctional facilities and public hospitals, see the sociologist Armando Lara-Míllan's (2014) study of emergency room waiting areas.
↵7. Though we have discussed the issue of confidentiality at various points in this article, we want to reemphasize here that any communication or collaboration conducted between the social worker and the parole officer was done by the request of the clients and with a signed release form. In accordance with clause 1.07(c) of the National Association of Social Workers (NASW) Code of Ethics, the social worker disclosed the “least amount of information necessary to achieve the desired purpose” per the client's request and with a signed release.
- Copyright © 2015 by Russell Sage Foundation. All rights reserved. Printed in the United States of America. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the publisher. Reproduction by the United States Government in whole or in part is permitted for any purpose. This research was supported by funding from the National Institutes of Health (R01MH094090, PI: Kral; R01DA033847, PI: Comfort; and R01MD007679, PI: Lorvick). We thank Jay Borchert, Elizabeth Kita, Reuben Miller, Victor Rios, Megan Tompkins-Stange, Damian T. Williams, and the participants in the Russell Sage Foundation “Severe Deprivation in America” conference for their invigorating and helpful feedback on drafts of the manuscript. We are particularly grateful to special issue editor Matthew Desmond for his detailed comments and insightful suggestions. And we are tremendously appreciative of and indebted to the people who participated in our research, and hope that we have done justice to the experiences they shared with us. Direct correspondence to: Megan Comfort at mcomfort{at}rti.org, Andrea M. Lopez at alopez{at}rti.org, Christina Powers at cpowers.contractor{at}rti.org, Alex H. Kral at akral{at}rti.org, and Jennifer Lorvick at jlorvick{at}rti.org; all authors at RTI International, 351 California St., Suite 500, San Francisco, CA 94014.
Open Access Policy: RSF: The Russell Sage Foundation Journal of the Social Sciences is an open access journal. This article is published under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.