sociology essay

American Sociological Review 2014, Vol. 79(5) 866 –887 © American Sociological Association 2014 DOI: 10.1177/0003122414549552 http://asr.sagepub.com

Open 24 hours a day, 7 days a week, and unable to refuse care, public hospitals and their ERs have increasingly borne the burden of caring for the urban poor (Albrecht, Slo- bodkin, and Rydman 1996; Andrulis and Duchon 2007). Private hospitals are often unable or unwilling to pay for uncompensated care and thus transfer their poor patients to public ERs (Gifford, Manheim, and Cowper 2002). Moreover, health clinics are unevenly distributed and often lack the specialty ser- vices offered by the public ER (Lewin and Altman 2000). Consequently, public hospitals now provide 65 percent of uncompensated

care within the largest metropolitan areas (Bazzoli et al. 2005). Some claim that more poor people pass through the public ER than through any other public institution (Gordan 1999; Malone 1998).1

Seemingly unrelated to changes in ER care are three ongoing trends in crime control and

549552 ASRXXX10.1177/0003122414549552American Sociological ReviewLara-Millán 2014

aUniversity of California-Berkeley

Corresponding Author: Armando Lara-Millán, University of California- Berkeley, Department of Sociology, 1410 Barrows Hall, Berkeley, CA 94720 E-mail: [email protected]

Public Emergency Room Overcrowding in the Era of Mass Imprisonment

Armando Lara-Millána

Abstract Although sociologists have long studied public emergency rooms (ERs) and their service delivery to the urban poor, little is known about how crime control and criminal justice contact affect ER admission decisions—a critical oversight given the documented increase in incarceration rates, intensification of policing, and proliferation of crime control language in public institutions. Using ethnographic methods and a count of ER admission decisions (N = 1,114), this article describes the continual rushing and delaying of medical resources to patients based on their perceived criminality or actual relationship to the criminal justice system. These dynamics develop through four processes: (1) widespread administration of pain medication, which associates waiting patients with criminal narcotics; (2) deployment of criminal stigma along race and gender lines to select some qualified patients over others; (3) patrolling of waiting rooms by police; and (4) rushing of beds to vast numbers of arrestees, inmates, and witnesses. Key to these findings is that if the urban poor happen to be wards of the criminal justice system, they receive rushed health care resources, but if they enter health care organizations on their own accord they are policed, delayed, and deterred from accessing care.

Keywords criminal justice, health care, mass imprisonment, medical sociology, organizations, urban poverty

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

Lara-Millán 867

criminal justice. First, the United States has experienced a dramatic, highly concentrated rise in incarceration rates. The number of peo- ple incarcerated in the United States grew sev- enfold over the past four decades, and this growth was concentrated among the minimally educated, non-white male population living in urban areas (Clear 2007; Garland 2001; West- ern and Pettit 2010). Second, an enlarged net- work of policing now affects the social lives of the urban poor outside of prison (Goffman 2009; Rios 2011). Third, scholars have docu- mented the proliferation of crime control lan- guage or “technologies, discourses, and metaphors of crime and criminal justice” (Simon 2007:4) in many non–criminal justice settings and policy domains (Muhammad 2010; Simon 2007; Soss, Fording, and Schram 2011).

Because the urban poor are disproportion- ately likely to enter the public ER and to be affected by these crime control trends, the following question emerges: How do incar- ceration, policing, and crime control lan- guage interact with the admission process of the public ER? Building on medical sociolo- gy’s study of health care organizations, this article draws from 13 months of fieldwork to identify the dual effects of three crime control trends on the contemporary urban public ER. These dual effects are defined as the continu- ous rushing and delaying of some medically qualified patients over others based on their perceived or actual relationship to criminality or the criminal justice system. Specifically, health care workers, operating in the context of high incarceration rates, intensive policing, and pervasive criminal stigma, rush health care resources to inmates and arrestees while delaying health care resources to members of the general public. Members of the general public who are perceived to represent crimi- nal elements, especially in relationship to narcotics, are especially at risk for delay.

These dual effects—the delaying of gen- eral patients and the rushing of medical resources to inmates/arrestees—have impli- cations for population health and the study of spillover effects more generally. Regarding population health, this study has three

implications. First and foremost, this study is the first to document how crime control trends diminish health care access for the urban poor who have no contact with criminal justice and only perceived criminality. Second, this study expands and offers support for recent conten- tions that prison is becoming a primary social service provider to the poor (Comfort 2007). Because of diminished social services outside of prison, incarceration may be the primary means by which the urban poor gain access to services like health care. Finally, this study elucidates how the provision of health care services to individuals in custody is organiza- tionally linked to diminished access for the non-incarcerated. To understand this organi- zational link, one must move beyond a con- sideration of incarceration trends and detail how they interact with policing and crime control language among health care workers in a major health care access point like the public ER. This organizational link is cor- roborated by two seemingly contradictory trends in population health: incarceration offers some health benefits to incarcerated African American men, but it is associated with worse health outcomes for ex-inmates, their families, and aggregate measures of community health (Patterson 2010; Wilde- man 2011; see also reviews in Schnittker, Masssoglia, and Uggen 2011; Wildeman and Muller 2012). In other words, a comprehen- sive analysis of health care for the urban poor in the United States is incomplete without reference to these dual dynamics.

By bringing a focus on organizations and health care workers into dialogue with con- temporary research on crime control and crim- inal justice, this study also identifies new research avenues in the study of incarcera- tion’s spillover effects on wider society. Most often, the effects of incarceration on economic inequality, education, neighborhood attain- ment, health, and family dynamics are meas- ured through the presence of ex-inmates in families and communities (e.g., Hagan and Foster 2012; Wakefield and Wildeman 2014; Western and Pettit 2010). In contrast, this study shows that dynamics and characteristics

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

868 American Sociological Review 79(5)

of organizations, such as hospitals, can pro- duce an independent effect on families and communities. These findings are relevant to scholars and policymakers seeking to under- stand and account for health inequality, as well as those seeking to reduce the costs asso- ciated with high incarceration rates, policing, and expanded crime control language.

ThEORETICAL BACKgROund Sociology of Emergency Rooms

Medical sociology’s study of how resources are distributed to patients is closely related to the labeling and categorization of patients, their illnesses, and their moral worth (Dobran- sky 2011). Like their counterparts in other human service organizations, medical profes- sionals use discretion in carrying out the daily work of medicine (Lipsky 1980; Prottas 1979; Schwartz 1972). Medical tenets, which pur- port objectivity and the absence of bias, sit alongside wider cultural beliefs about patients that inform practices often unacknowledged by medical professionals (Chiarello 2013; George and Dundes 1978; Heimer and Staffen 1995; Timmermans 1998; Waitzkin 1989).

Roth (1972) and Jeffery (1979) extended this discussion to the public ER and prompted a number of sociological studies (e.g., Hinze et al. 2009; Kurz 1987; Lincoln 2006; Mizrahi 1985; Vassey 2001). These scholars document three findings relevant for this article. First, the designation of “undeserv- ing” patients is linked to wider prejudices about poverty, dependency on welfare, and race (Lincoln 2006; Roth 1972; Schwartz 1972). Medical professionals use cues given during clinical encounters to make presump- tions about their patients’ social status (Roth 1972). These studies, conducted in the 1970s, found that in contrast to stigma about poverty and welfare, criminal stigma was reserved for a small subset of the patient population. Medical staff assigned criminal stigma only to arrestees and to patients who presented to the ER for drug overdose (Jeffery 1979; Roth 1972).

Second, ER staff value patients they con- sider “urgent” (with more critical medical conditions) and strongly disfavor “non-urgent” patients (who, presumably, can be treated in outpatient settings) (Jeffery 1979; Mannon 1976; Mizrahi 1985; Roth 1972; Schwartz 1972; Vassey 2001). Professionals believe the categories of urgent and non-urgent are well specified and rationally demarcated. Indeed, medical journals are replete with attempts to understand the motivations of non-urgent patients who supposedly choose to use the ER as a source of primary care (see Durand et al. 2011; Krug 1999). In these studies, casting non-urgent patients as inappropriate candi- dates for emergency care is crucial to how medical professionals make sense of their patients and overcrowding.

Third, these studies underspecify the mechanisms by which stigma about patients actually delays ER admission. Roth (1972) focuses on use of aggressive verbal com- mands against stigmatized patients. He leaves unclear, however, whether such patients receive the care they seek in spite of their often harsh treatment. Jeffery (1979) describes how staff delay the treatment of “undeserv- ing” patients, but he provides evidence only for alcoholics sent to back rooms (to be dealt with when staff members have the time) and non-urgent patients sent to separate waiting rooms. Jeffery notes that when crowding increases, some urgent patients are pushed to these separate waiting rooms, but he says nothing about how this blurring unfolds. Vas- sey (2001) studies how staff’s labels affect patients throughout an ER visit. Patients are subject to delays in x-rays, delays in transfer to inpatient wards, and other less visible instances of feet dragging. However, the exact mechanisms by which stigma affects the admissions process remain underspecified.

Sociology of Imprisonment, Policing, and Crime Control Language

Criminologists and scholars studying the social control of the urban poor document three trends relevant to this inquiry: a dramatic rise

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

Lara-Millán 869

in incarceration rates, intensification of polic- ing, and a proliferation of crime control lan- guage. First, since the 1970s, the adult incarceration rate jumped from 100 people per 100,000 to its current rate in excess of 700 people per 100,000 (Western and Pettit 2010). This is a larger rate than any other country in the world. Among other factors, deindustrial- ization, the breakdown of informal social con- trols, stricter sentencing, the criminalization of narcotics, and the enlargement of prison capac- ity and law enforcement budgets contributed to this growth (Bertram et al. 1996; Western 2006).

Some characterize this dramatic growth as “mass imprisonment” insofar as the incarcer- ation rate is now high enough and concen- trated enough to affect an entire demographic group (Garland 2001): a large proportion of the incarcerated are poor, are people of color (Western 2006), and come from concentrated urban areas (Clear 2007). As Clear (2007:9– 10) writes, “The ubiquity of prison touches almost everybody in these neighborhoods. Every family has a member with limited labor market options; every family knows someone struggling to stay out of jail, many mothers are raising children whose fathers have a prison record” (see also Comfort 2008). Research has focused on returning inmates’ diminished social status and its effect on a host of individual- and community-level out- comes, including, but not limited to, access to public aid (Mauer and Chesney-Lind 2002), voting (Manza and Uggen 2006), employ- ment (Pager 2008), parenting and children (Wakefield and Wildeman 2014), education (Hagan and Foster 2012), health (Schnittker et al. 2011), and inequality (Western 2006).

Second, scholars have documented the intensification of policing in these neighbor- hoods (Goffman 2009; Rios 2011). Goffman (2009) describes a vast network of post- incarceration policing, surveillance, and intimidation that shapes the practices and lived experience of the urban poor. Young men and their partners live in a state of gener- alized fear and mistrust—even reporting reluctance to visit hospitals for fear of

re-arrest. Rios (2011) documents that, even before their first arrest, young black and Latino boys are subject to police scrutiny— regardless of whether they have committed crimes. Moreover, increased police presence and surveillance have been documented in urban public schools (Hirschfield 2008; Kup- chik 2012) and welfare institutions (Gustaf- son 2011). Regarding the latter, Gustafson (2011) reports that recipients’ names are rou- tinely shared with law enforcement, and pub- lic aid offices have been used as sites for sting operations.

Finally, scholars have documented the pro- liferation of crime control language and crim- inal stigma as transforming a variety of laws, policies, public and private institutions, and the everyday practices of policymakers and citizens (Garland 2002; Muhammad 2010; Simon 2007; Soss et al. 2011; Wacquant 2009). Simon (2007:4) articulates three aspects of this proliferation: (1) citizens are now legitimized when they act to prevent behaviors viewed as criminal; (2) crime con- trol is used as legitimizing motivation for many non-criminal policy interventions; and (3) “the technologies, discourses, and meta- phors of crime and criminal justice” reshape how policymakers and citizens view a pleth- ora of social problems. Indeed, Garland (2002:201) writes that welfare and criminal justice institutions “share the same assump- tions, harbor the same anxieties, deploy the same stereotypes, and utilize the same recipe for the identification of risk and the allocation of blame.” For their part, Hays (2004) and Soss and colleagues (2011) emphasize the use of punitive sanctions in the welfare system, including the use of drug testing and criminal records to exclude clients. The expansion of incarceration, policing, and crime control lan- guage in the lives of the urban poor—the primary users of public ERs—all point to the necessity of this inquiry.

METhOds I studied a large public ER attached to a gen- eral hospital with more than 200 beds. While

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

870 American Sociological Review 79(5)

detailed identifying information must remain vague so as not to jeopardize the anonymity of the institution, this ER is consistent with the flagship public ERs of Chicago, Los Angeles, and New York City, which have average wait times fluctuating between 6 and 12 hours (Katz 2012; Quintanilla 2006; Rabin 2011). It is located in a metropolitan statisti- cal area with more than 1.5 million people. The hospital’s catchment area contains zip codes with at least 20 percent poverty rates. Nearly all of its patients are publicly insured or uninsured. More than half of its patients are people of color. Over the past 20 years, the hospital was downsized by at least a quar- ter of its staffed beds and, as a result, now faces overcrowding unprecedented in its his- tory. Downsizing is typical among large pub- lic ERs in major urban areas: nationwide, staffed beds in public hospitals declined 27 percent between 1980 and 2002 (Andrulis and Duchon 2007).

I conducted approximately 700 hours of participant-observation over 13 months. I observed triage staff at four gatekeeping points: (1) during assessments of incoming patients and subsequent assignment of urgency rankings; (2) as some patients were selected over others for beds; (3) arrivals of police arrestees and jail inmates; and (4) as police and nurses conducted reassessments (re-checks of patient vital signs) in waiting lobbies.

Fieldwork outings lasted eight to twelve hours—a period roughly coinciding with the time it took for a patient to receive a bed. Some weeks I conducted such outings three times per week, other weeks I conducted only one outing. In total, I logged 532 hours divided equally between the morning, day, night, and graveyard shifts—whatever shifts I requested were approved. The remaining approximately 200 hours were conducted during the graveyard shift.

During observations, I wrote shorthand notes to identify general events. I used these notes to orient detailed audio recordings, which I conducted in private every four hours. These audio recordings were transcribed after the project was completed. The excerpts

reported in this article are taken from these audio recordings.2

To gauge the prevalence of my observa- tions, I also constructed a unique dataset of admission decisions (N = 1,114). The count offers one measure of how many admissions are affected by the logics described in this article. The count required access—afforded by the ethnographic method—to how nurses think about patients at the moment of an admissions decision.

I constructed the dataset by accompanying head nurses as they examined medical records and patients, debated with other nurses over patients, and made final admissions deci- sions. This deep entrenchment allowed me to build up knowledge about the pool of poten- tial patients head nurses select from when beds became available. Each time a bed became available, I recorded nurses’ explana- tions for why one patient was chosen over another. I instructed nurses to decline speak- ing about decisions when they were too busy or when they did not feel comfortable talking about them.

I categorized decisions into six possible types: (1) Medical decision: patient chosen because of medical criteria (e.g., vital signs, medical history, or visual cues of pain). (2) Time-of-arrival decision: patient chosen because they arrived earlier than another patient. (3) Police/jail arrival decision: patient chosen because they were a police or jail arrival. (4) Criminal stigma decision: patient chosen because of another patient’s criminal stigma (e.g., drug addiction or general crime). (5) Fire/ambulance decision: patient chosen because of fire department or ambulance involvement. (6) Unknown: nurses were unclear or unwilling to explain.

It is important to note that I categorized decisions that used a mixture of medical and any other decision type as medical decisions. This provided the greatest allowance for medi- cal decision-making, including instances when I strongly suspected non-medical criteria played a role but nurses would not confirm it.3 Two key features of this dataset must be kept in mind. First, this moment of decision-making

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

Lara-Millán 871

occurred after much negotiation, including upgrading and downgrading of patients’ rank- ings. Second, the count represents only those decisions made during my observations; as such, it is not a sample of all decisions. Instead, in the tradition of ethnographic “counting” (Becker 1998), I utilized the count to guard against observations straying from the most important happenings of the field site.

sETTIng And ORgAnIzATIOnAL PROCEssEs The Rational System of Triage and the Harrowing Wait

In the ERs of the 1970s studied by Schwartz (1972), Roth (1972), Prottas (1979), and Jef- fery (1979), non-medical clerks were in charge of admission decisions. The rules for vetting incoming patients were unspecific, learned on the job, and applied inconsistently. In the contemporary ER, registered nurses and nurse practitioners now conduct triage when admitting patients. This professional- ization contributes to the purported objectiv- ity and rationality of the admission process.4

When patients first enter the ER, they pass through a security checkpoint (with x-ray machines and metal detectors) and are directed to an initial screening room. This screening room contains 3 to 10 interview booths (depending on staffing). Patients typically wait an hour to be called to a booth. The room is tense, as patients are told very little about the admissions process: every few minutes, security guards wrangle with weary patients who demand to be informed of their fate.

Eventually, nurses call each patient to a booth, conduct oral interviews, and adminis- ter vital signs checks. In these interviews, nurses push for specific symptoms and the reasons patients chose to seek care on that particular day. Patients, however, often annoy nurses with the wider social circumstances of their pain (e.g., lack of housing, unemploy- ment, and stress) and more general health problems. Finally, vital signs checks (blood

pressure, heart rate, and temperature meas- ures) are given to patients. The vital signs tests are a key source of triage’s purported objectivity:

“It’s how we know whether or not a patient is faking, if they are really in pain,” said one hospital administrator, “it also allows us to differentiate between patients with the same presenting medical issues. If one has a heart rate over 200 and another one with 195 we can make a distinction. The machine does not lie.”

These oral interviews and the vital signs checks constitute the rational rules that guide nurses in assigning urgency rankings.

Assigning patients urgency rankings (A, B, C, or D) is the critical goal of these inter- views. The order patients are to be admitted is based on their ranking; the sickest should always receive beds first. Most patients are given a ranking of B (urgent) or C (non- urgent) and sent to two different waiting rooms. The C-list patients, who constitute the vast majority, are sent to the front room lobby where they are placed on the “general wait list.” These patients are to be admitted based on their order of arrival.

The front room waiting lobby—where C-list patients wait—is an unpredictable set- ting. At times, patients sleep and patiently await their admission. At other times, it is a chaotic space with police officers threatening the removal of patients and confrontations between frustrated patients and nurses. Patients normally line the walls and sit on the floor; however, the room becomes noticeably over- crowded when new patients are only able to stand in walkways and obstruct nurse traffic. The bathroom always has a long line. Children are a constant source of noise. For most patients, the 8- to 18-hour wait is harrowing.

Critical to the fate of C-list patients is the process of “reassessment.” Reassessment refers to re-checks of patients’ vital signs dur- ing the wait. The goal is to monitor changing conditions and upgrade or downgrade patients’ rankings accordingly. Reassessment is particularly important because nurses have

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

872 American Sociological Review 79(5)

the discretion to expunge patients from the wait list if they do not respond when called for reassessment. This process results in the expulsion of many C-list patients.

Patients given B rankings are considered “urgent” and are sent to a separate waiting room called the “observation waiting room.” To gain access to this room, patients pass through a separate security checkpoint. In the observation waiting room, head nurses delib- erate over patients who are near the point of admission. This separate waiting room allows nurses to maintain continual observation of B-list patients as they wait. Because these patients are considered more urgent than their C-list counterparts, they are, technically, always to be admitted before C-list patients— regardless of arrival time.

It is in this observation room that the head nurse makes final admissions decisions. When beds become available, the nurse gets a phone call from a bed coordinator with information on how many beds are free. During such moments, many head nurses claim to remem- ber all patients and the seriousness of their conditions. The information is built up from reviewing records, their own observations, and conversations with other staff. Once a decision is made, patients are called and, if they have not left in the interim, led to a bed.

There is also a “back room” entrance to the ER. This entrance is available for patients brought by ambulance, fire departments, police departments, and sheriffs from local jails. This entrance has a separate nurse whose primary responsibility is to negotiate access to beds for these public safety arrivals. It has a small waiting lobby where there is almost always a congregation of public safety offic- ers. Nurses befriend these officers after repeated interactions, and it is a collegial set- ting. It is also a gauntlet of sorts for incoming arrestees and inmates, who are often taunted and questioned about criminal activity.

Two Conundrums of Overcrowding

Triage staff must find solutions to two organi- zational conundrums created by extreme

overcrowding. The first is the rule that B-list patients (more urgent patients) should always receive beds before C-list patients. A problem emerges when, because so many B-list wor- thy patients arrive to the ER, the B-list is never empty. Therefore, technically, in these cases, no one on the C-list should ever receive a bed. If staff do not find informal means of thinning the C-list, the general lobby will swell even further and the hospital risks los- ing accreditation.5

The second conundrum is that there are more B-list patients whom triage guidelines categorize as urgent and in need of a bed than there are available beds. That is, nurses must choose between patients who, according to medical criteria, equally qualify for beds. This constitutes insufficient specificity in the con- ferral of “urgent” rankings (Prottas 1979). For example, a head nurse is told that two beds are now open, but the nurse previously deter- mined that three patients were all equally in need of beds. Over the course of their approxi- mately eight-hour wait, all three patients had at least one concerning EKG (a diagnostic tool that measures the electrical rhythms of the heart), some shortness of breath, intermittent chest tightness, and high pulse rates. A head nurse said in such a situation,

These patients . . . are most likely not going to have a heart attack during their time at this hospital. But, you never know, and it only takes one patient to die here in the waiting room to spark a lawsuit. From my perspective, with the information I have on them and what they have provided me with, they all have an equal chance, albeit a small chance, of that happening.

The nurse is faced with a conundrum: Which two patients get a bed and which patient will continue to wait?

What is at stake when a bed is given to one patient over another? Those who are made to wait are more likely to leave the ER before being seen by a physician (National Public Health and Hospital Institute 2011). Although available data show no medical differences

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

Lara-Millán 873

between those who leave and those who con- tinue to wait, patients who leave are far more likely to develop serious medical problems in the short term (Baker, Stevens, and Brook 1991). The consequences for long-term health are unknown. While the medical community is at a loss for why some patients “choose” to leave the ER (Arendt et al. 2003), in what fol- lows, I show that patients often leave because of dynamics related to crime control and criminal justice.

RushIng And dELAyIng hEALTh CARE In ThE ERA Of MAss IMPRIsOnMEnT

In this section, I detail the dual dynamics by which the ER in this study is able to manage extreme overcrowding. I view these dynam- ics not as a planned strategy, but as the out- come of managing patients in the context of pervasive crime control language, the pres- ence of police, and high incarceration rates. Health care workers utilize the lines of action made available to them to find what they believe to be reasonable solutions to the prob- lems of overcrowding.

The dual effects of crime control trends on the public ER are defined as the continuous rushing and delaying of some medically qual- ified patients over others based on their per- ceived or actual relationship to criminality or the criminal justice system. These dual effects unfold through four processes. First, in the front room entrance, general patients are delayed through the widespread administra- tion of pain medication and subsequent con- ferral of criminal stigma. Specifically, pain medication temporarily alters patients’ vital signs and provides staff with valid medical reasons for delaying patients. Nurses justify this practice through the belief that most wait- ing patients only enter the ER to obtain the very opiates that staff rely upon to manage overcrowding. Second, in the front room entrance, nurses deploy criminal stigma along race and gender lines to compensate for ubiq- uitous “missing medical information” about

incoming patients. In the absence of medical information about patients, triage staff make use of criminal stigma to delay some simi- larly situated patients over others. Third, hospital-based police patrol the front room lobby, pressuring some patients to stop seek- ing care, which expunges patients from the wait list. Police justify this thinning of the wait list by combining crime control language with an interest in solving hospital over- crowding. Fourth, in the back room entrance to the ER, police and jailers bring a signifi- cant number of arrestees, inmates, and wit- nesses who are rushed to hospital beds ahead of general patients—even when they are less medically qualified for those beds. As a result, front room patients lose beds to the endless supply of drug arrests and inmates being rushed medical resources in the back room entrance. I will expand on all four processes in turn and illustrate them from common exam- ples taken during the course of fieldwork.

Before doing so, I present the count of admission decisions described in the methods section. Of the 1,114 decisions to fill a bed, I categorized 349 (31 percent) as “medical,” 152 (14 percent) as “time-of-arrival,” 290 (26 percent) as “criminal stigma,” and 157 (14 percent) as “police/jail arrivals.” Of the remaining decisions, I categorized 22 (2 per- cent) as fire department/ambulance and 144 (13 percent) as unknown.6 Close to half of admissions (45 percent) were based on pur- ported objective triage criteria (e.g., medical and time-of-arrival decisions); criminal stigma and police/jail decisions represent 40 percent of decisions.

These data can also be grouped to distin- guish medical decisions from decisions that utilize discretion. Doing so is analytically useful, because when nurses are unable to use medical criteria to make an admission deci- sion, they must use other means to inform their decisions (e.g., time-of-arrival, criminal stigma, and police/jail arrivals). To that end, removing medical decisions from the pool leaves a population of 765 admissions. Of these discretionary decisions, nurses used time-of-arrival 20 percent of the time,

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

874 American Sociological Review 79(5)

criminal stigma 38 percent of the time, and police/jail arrivals 20 percent of the time.7 Grouping the last two categories together, which constitute the crime control dynamics described in this article, accounts for 58 per- cent of all discretionary decisions. Thus, when no medical criteria present themselves, nurses are much more likely to invoke crimi- nal stigma or to choose a police arrestee than to simply pick the next patient in line. To summarize, while objective triage protocols (medical and time-of-arrival) represent the majority of admission decisions, decisions influenced by the dynamics studied here (criminal stigma and police/jail arrivals) rep- resent nearly as substantial a proportion of all decisions and are the most important when nurses must use discretion.

Delaying General Patients: Pain Medication and the Expansion of Criminal Stigma

Previous studies of pain medication document how people from low social-status groups are less likely to be prescribed pain medication at the point of hospital discharge (Hinze et al. 2009; Todd et al. 2000). In stark contrast stands the world of triage, where pain medica- tion is ubiquitous and plays a central role in managing the front room waiting list patients. During the course of the wait, many patients are administered heavy opiate analgesics.

The widespread administration of pain medication allows staff to manage over- crowding in two ways. First, pain medication temporarily alters waiting patients’ vital signs (the primary means by which triage staff identify the sickest patients) and thus pro- vides nurses with valid medical reasons for selectively delaying the admission of some patients over others. After receiving pain medication (which temporarily makes patients feel better), some patients leave the ER—an outcome that shrinks the wait list. Second, administration of pain medication creates the potential for triage staff to invoke the stigma of narcotics addiction and criminality about any patient. Triage staff rationalize delays by

explaining that some who seek care do so only to obtain narcotics. Thus, the act of seek- ing health care in the front room entrance is understood as an act of criminality.

The following excerpt describes an event that took place in the observation room and illustrates the processes whereby a patient, on the verge of obtaining a bed because he has significant indicators of chest pain, was moved from urgent to non-urgent. Triage guidelines at this hospital dictate that rapidly changing blood pressure, high heart rates, and suspicious EKGs should be upgraded.

“Yeah, it is true. He fits the criteria for a bed if that’s what you’re asking,” Arthur, the head nurse, said to me after I asked him about a patient who had just been complain- ing about his wait time. “But what that guy doesn’t understand is that we gave him pain medication. And he’s been here for what . . . [looking at the computer screen] six hours. He got reassessed and his vitals have changed, he can wait awhile longer at this point. . . . The pain meds are doing what they are supposed to do.”

At that point, the patient came back up to Arthur and, while touching Arthur’s fore- arm, said, “If I die here, it’s your fault.” Arthur, aghast, looked down at his arm and said, “Sir, you need to take your arm off of me right now, I will have the sheriffs come down here and remove you.” The patient walked away.

Several hours later and after two rounds of bed openings had passed, Arthur called the patient for a re-check of his vital signs. He was nowhere to be found. It was clear he was one of the many patients that left the ER prematurely. “HA!” said Arthur.

I shot Arthur a perplexed look and said, “What? He wasn’t as serious as he claimed?”

Arthur replied, “Well [pause] . . . he defi- nitely needed a bed at some point. I had him on-deck for about four rounds [nurses’ word

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

Lara-Millán 875

for instances when beds became available]. He had a concerning EKG, the other one was okay after, and his sweating and heart rate were a bit high before the pain meds, so he meets the criteria . . . but . . .” Sensing Arthur was reluctant to elaborate I said, “It must be tough, there are a few people on this list just like him, right?”

With almost seamless transition, he said, “Yeah. And the thing is . . . this guy has been here a bunch. You saw that from his record. He’s got problems and can’t take care of himself: eat right, exercise, and go to the damn clinic. Blaming me if you die here? It’s your fault for not putting in the time to get it looked at. I mean, look, he took off . . . probably was here just to get the drugs.”

Although many patients leave the ER after receiving pain medication, many others con- tinue to wait, and this necessitates another way of ignoring qualifications for admission. In the example above, in addition to adminis- tering pain medication, the nurse cast this patient as a part of the undeserving poor. The patient is worthy of delay because Arthur depicts him as not having done anything to fix his health problems, which Arthur sur- mises from the patient’s purported unneces- sary use of the ER and blaming of Arthur for his predicament. This stigmatization is but- tressed by threats of arrest, ambiguous use of the word “record,” and presumed drug addic- tion. Immediately after the exchange, I asked Arthur to elaborate on his indictment of the patient as a drug addict:

Arthur replied, “The thing you have to understand: we have a bunch of drug users and drug-pushers in here. Half the time these people are complaining and are in here for these simple issues, they are just here for the free supply.”

The application of criminal stigma to patients post hoc obscures the true function of pain medication in the ER. To be clear, qualified patients are delayed and deterred from

accessing health care through the use of pain medication, and this delaying and deterrence is justified through the use of crime control language.

In Roth’s (1972) and Jeffery’s (1979) eth- nographies of public ERs in the 1970s, crimi- nal stigma was reserved for a small subset of patients: those brought to the ER for drug overdose or brought by police. In the contem- porary ER, as documented here, pain medica- tion creates the potential to indict all patients as drug addicts, regardless of their relation- ship to police or actual criminal drug use, and therefore undeserving of care. The rhetoric and narratives around drug peddling and addiction—which are a ubiquitous feature of contemporary crime control language (Simon 2007)—inform triage staff’s shared under- standing of their patient population and justi- fications for continuous delays.

Delays across Race and Gender: Deploying Criminal Stigma and Missing Medical Information

The previous section described how criminal stigma is applied to general patients on the front room wait list (via the administration of pain medication); I now describe how tri- age staff put such an understanding to work in selecting some patients over others for beds. Specifically, I demonstrate that triage staff utilize criminal stigma to solve the critical problem of “missing medical infor- mation.” As we will see, deployment of this criminal stigma occurs along race and gen- der lines.

Because staff have limited access to in- depth medical tests and time, they are faced with much “missing medical information” about the patients they are trying to rank. As a result, they often use discretion to admit patients on the “hunch” that missing medical information would confirm an urgent ranking. This use of discretion is, in part, informed by criminal stigma about patients.8 As an illus- tration, the following excerpt details how one woman moved from a less urgent ranking to immediately admissible:

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

876 American Sociological Review 79(5)

A young woman approached the nursing booth and asked the head nurse, named Melinda, how much longer she would have to wait. Melinda gave her standard response, “I’m sorry ma’am, I know you have been waiting a while, but you’re going to have to continue to wait; we try to give beds to the sickest patients.”

As the patient walked away, Frank, another nurse, whispered to Melinda, “Yeah, her vitals are pretty good. . . . I just reassessed her maybe an hour or two ago.” “Hmmm,” Melinda responded. “She can’t be faking that, her eyes are so blood red . . . she looks miserable, but when you look at her vitals they are perfect.” I see Melinda slowly going over the computer and pondering the patient’s record. “Let’s look at her labs,” she says, “I don’t know, can’t really upgrade her if she isn’t indicating anything. It’s weird, I don’t think I’ve seen her here before.”

At this point, Melinda, the head nurse, was unwilling to admit this patient despite the disconnect between visual cues of pain (the red eyes) and normal vital/lab results. Melinda referenced criminal stigma when she explained the delay with the phrase, “It’s weird, I don’t think I’ve seen her before.” In the culture of the triage staff, this phrase invokes the baseline assumption that most waiting patients fake illnesses to return to the ER and obtain narcotics. Although most patients have some kind of missing medical information, patients who can dissociate themselves from criminal stigma are admitted on “hunches” that missing information con- stitutes an urgent health problem. Several hours later this same patient asked Melinda if she could go outside:

“What for? You really shouldn’t go outside if your situation is serious . . .” responded Melinda. “Solemente quieria hablar con mi esposo [I only wanted to talk to my hus- band],” the woman said. Melinda, with a surprised look on her face said, “Your hus- band is waiting for you out there?” After the

patient said yes, Melinda told another nurse to go find the husband and bring him to the waiting room. . . . As the man walked in, with painter trousers on, he immediately struck me as one of the many immigrant construction workers in the area. Melinda leaned toward me so as to hide her com- ment, “I didn’t know she was married and that he was waiting out there for her this whole time . . . that’s so cute . . . [as the woman leans on her husband’s shoulder] . . . he really cares about her. . . . I’m pretty sure something must be up cause he wouldn’t be waiting if he’s got work tomorrow . . . [after a short pause] . . . something must really be wrong.”

A short while later, two beds opened up and the patient was admitted above six other people. I said to Melinda, “How did you know she wasn’t faking the pain? Her vitals were good right?”

“Yeah, it’s true, but I mean you saw her right, she looked miserable and when we can’t really tell if something is going on, that doesn’t mean something ISN’T going on. And the thing is you don’t really fake when the family waits for you like that. I asked them and he was in construction and he does have to work tomorrow. Because I told them that they would be here through the night and they stayed anyways. HE stayed with her . . . so you know . . .”

In this example, the head nurse moved from being willing to ignore the missing informa- tion to admitting this patient on a hunch. This move was made possible once the patient was able to demonstrate her romantic relationship with someone who was perceived to earn a living from non-criminal endeavors.

This excerpt points to how criminal stigma (or, in this case, the absence thereof ) is deployed according to the intersection of race and gender. Publicly, the ER is a color-blind environment. While medical professionals are willing to verbalize stigma around criminality and welfare-dependency, they are absolutely

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

Lara-Millán 877

unwilling to verbalize stigma about race and gender. Yet, over the course of my fieldwork, it became clear that ethnicity and gender mediate how some patients come to be asso- ciated or disassociated with criminal stigma in two main ways.

First, ethnicity is critical in determining how women’s relationship status associates them with criminal stigma. Specifically, when Latinas indicate that they are in relationships with Latino men (who triage staff largely con- sider to be involved in non-criminal employ- ment), triage staff are more willing to pursue missing medical information for such patients. In stark contrast stands the assessment of Afri- can American women. When African Ameri- can women associate themselves with African American men, triage staff are likely to assume such partnerships exist only to support illegal economic activity (e.g., prostitution or drug use) and are thus less likely to admit such patients on hunches. These assumptions draw on broader stereotypes of African American women (Collins 2000) and specific stereo- types of black men as criminals (Anderson 2012). This also suggests the strategic deploy- ment of Latinos as a “model minority” (Wortham, Mortimer, and Allard 2009).

Second, the presence of children and asso- ciated gender roles mediates men’s and wom- en’s relationship to criminal stigma. When women present to the emergency room with children, triage staff often assume the chil- dren were reared, in part, to increase access to public aid, and, by extension, the ER’s opi- oids. The staff’s willingness to fill in missing information for such patients is much lower. Yet, when adult men present to the ER with children, triage staff are more likely to assume they are “good fathers” and unassociated with criminality. In other words, for women, chil- dren are a sign of welfare dependence (and, by extension, ER abuse), whereas for men, children provide protection against criminal stigma. These gendered deployments of crim- inal stigma in the ER reflect how women’s relationships to men and children are recon- structed by state actors in frontline institu- tions (Haney 1996).

Police Patrols and Expelling General Patients in the Front Room

During the past decade, the number of official police units stationed at public hospitals has been on the rise (to be clear, I refer only to official police units, not private security guards). After remaining relatively stable throughout the 1990s, police forces devoted to patrolling public medical facilities jumped from 42 in 2000 to 64 in 2008, and the amount of sworn officers in those forces increased from 978 to 1,460.9 This number is not insignificant considering that the number of general acute care public hospitals in the 100 largest U.S. cities recently fell to just 70 (Andrulis and Duchon 2007). A common mis- conception among doctors in this ER is that the full-time police unit stationed at the hos- pital simply protects medical staff from angry patients and kicks out individuals who attempt to spend the night in the lobby. Far more important is police officers’ role in systemati- cally thinning the front room wait list of gen- eral patients.

When the waiting lobby becomes particu- larly overcrowded—that is, there are no more seats for new patients—the police unit begins, one by one, checking the IDs of each person waiting in the lobby and occasionally reading those names into walkie talkies. These names are read so that officers at remote computers can conduct background checks. On many occasions, I observed waiting patients who, after seeing this activity occurring on one end of the lobby, preemptively left the ER. Field interviews with such ex-patients indicate they left the ER because they did not want the police to discover some arrestable status (i.e., an outstanding warrant, probation violation, or status as an undocumented immigrant). This pressure to stop seeking health care is consistent with Goffman’s (2009) finding that young men “on the run” avoid public institu- tions for fear of re-arrest; it also aligns with work on undocumented immigrants’ fear of local deportations (Berk and Schur 2001). However, this policing also affects people who have little connection with criminal or

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

878 American Sociological Review 79(5)

undocumented statuses; that is, members of the general public are also pressured to stop seeking health care in the front room entrance. To understand this important point, it is vital to understand the connection between reas- sessment and policing.

As previously described, reassessment is the process by which nurses are mandated to re-check waiting patients’ vital signs at regu- lar intervals. Hospital policy dictates that if patients do not respond when their names are called for reassessment two or more times, they can be removed from the waiting list. If patients return after missing their name call, they can be forced to re-register and start the waiting process over again. Nursing staff con- sider this job the least desirable because it rarely results in upgrading of patients’ condi- tions and usually leads to confrontations with frustrated patients.

When nurses are in charge of reassess- ment, they are reluctant to expunge patients from the waiting list. Depending on whether a patient has been given a good moral evalu- ation, nurses will strike a patient from the list after two, three, or four chances. Moreover, nurses will go out of their way to look for patients in the bathrooms, hallways, or even outside the hospital before expunging a patient.

Police buttress the link between reassess- ment and expulsion in the following two ways. First, they use policing tactics to pres- sure individuals to actually leave the ER when nurses remove them from the wait list. Absent such tactics, patients can sometimes plead their way back onto the waiting list when dealing directly with nurses. Second, when overcrowding reaches critical levels and the waiting list swells, police take over the work of reassessment directly. During such instances, a police unit calls out the names of every C-list patient waiting in the front room lobby. When police are in charge of calling out names, reassessments occur more frequently. The calls are made from one location and take about five seconds to read each name three times. Patients inevitably miss their call. Police see buttressing the

effectiveness of reassessment as one of the primary ways they “keep the waiting line in check,” as one officer described it to me:

Sometimes the nurses give them a little lee- way. We just make sure it happens. . . . These people need to be in this room if they are really sick . . . if they are off, out of the ER, god knows what they are up to.

If the nurses do this a lot of times the people will come crawling back and start hassling them. . . . Saying they were here the whole time. But if we do it, the people are gonna be less likely to challenge what we say. . . . Lots of times if the person is really challenging us, we will just get on the radio and call the guys down [other officers]. We can just sur- round the suspect, stare him down, pressure, and they will remove themselves. We know how to deal with people that are disrespect- ful, that’s what we do for a living.

Another officer described his role as follows:

The hospital has rules about the waiting list. The list would just be endless if we didn’t do this. These people come in here week after week, everybody knows they don’t really need to be here. And the conditions, the overcrowding in here, can be dangerous, people lose their cool. And these types of people, this is just a fact I’m not trying to be a dick, crime is always going to be an issue. You have drug peddlers in here. So it helps with safety as well. If we didn’t enforce these rules or like make them actually work, these people would really be able to abuse the waiting list.

Two key points stand out from these excerpts. First, police justify their role in expunging general patients by utilizing the same under- standing of patients as the nurses: the crimi- nalized patient. When the police officer says, “if they are off, out of the ER, god knows what they are up to,” he is demonstrating the shared idea that some patients are not really sick and are perpetrating crime. Patients, who

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

Lara-Millán 879

in this context have nothing to do with the criminal justice system, are redefined as sus- pects. When these patient-suspects attempt to pressure police for re-enlistment onto the waiting list, the police respond with arrest tactics to ensure expulsion. When threatened with arrest, jailing, and a criminal record, even the most resilient patients leave the pub- lic ER. Second, the officers conjoin their tra- ditional roles of ensuring public safety with the mandate to solve the organizational conundrums of overcrowding. When the officer prefaces his crime control statement with, “the hospital has rules about the waiting list. The list would just be endless if we didn’t do this,” he indicates his understanding that policing works, in part, to solve overcrowd- ing problems in the front room waiting lobby.

Rushing Health Care to Inmates and Arrestees

The frequency with which police and jailers bring arrestees, inmates, suspects, and wit- nesses to the back room entrance of the ER significantly affects the front room waiting line. Arrestees and inmates who arrive in the back room entrance are provided with beds before general patients in the front room— often despite having non-urgent medical con- ditions. This is a counter-intuitive finding, as previous research suggests the “mark” of the criminal justice system precludes the provi- sion of services (Mauer and Chesney-Lind 2002). Instead, nurses justify favoring police and jail arrivals because they believe the escorting public safety officials cannot be made to wait the 12 or more hours that would be standard for such non-urgent patients. In contrast, when fire departments and ambu- lance officials bring patients through the back room entrance, they simply drop them off into the front room and these patients enter the normal triage process. Police and jailers, however, cannot relinquish custody of patients they bring to the ER.

During the 1990s, the frequency of police and jail arrivals to this ER began to exceed the amount of space reserved for inmates. The 15

to 20 beds in this ER’s jail ward—the beds in the ER specifically devoted to in-custody patients—are perpetually filled. Discussion with officers in the back room entrance of this ER revealed that police bring individuals to the hospital for two general reasons. First, public safety officials are legally obligated to provide treatment for “serious illness or injury” suf- fered during custody (Sundararajan et al. 2012). Second, as documented here, police bring arrestees, suspects, and witnesses to the ER to interrogate them.10 The following excerpt captures the admission of a police arrival. It is an important example because it illustrates both police understandings of their reasons for bringing the arrestee into the ER and the nurse’s reasons for favoring the patient:

Two police officers were talking with a young man strapped and handcuffed to a gurney, kind of joking with him in a way. “You make good money. I know you have a better load than I do,” said the police officer to the man. “We know you’re hustling, we seen you on that corner before, don’t lie.”

At this, I assumed the man was one of many brought here for possession of narcotics. He struck me as aloof in a way that instantly made me think he was high on drugs, and I could tell the officers were trying to get him to admit to selling . . . which carries a much stiffer sentence.

“We gotta get you off the streets man . . . you keep hurtin’ people, your family . . . let’s just take you out of there . . .” said the other officer to the man.

After a brief silence (I was very scared to say anything), I managed to ask: “Why is he here, guys?”

The policeman looked up at me in surprise, looked over to the back room charge nurse, who nodded to him, and then said, “He’s here ’cause he’s high, I mean he’s drunk too, so I doubt anything is wrong with him per se . . . we just HAVE to take him here.”

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

880 American Sociological Review 79(5)

“So it’s like, what, protocol to bring him down here?” I asked.

“Well it’s not so much that, I mean that’s true, but that is kind of up to our judgment. A lot of these guys are high when we take them in, you know like we can pretty much find stuff people are doing illegal just by talking to them . . . and we’re trying to crackdown on that area.”

Shortly after this conversation, a nurse escorted the police officers to a bed where they would sit with the man for the next few hours. As soon as they left, I walked over to the back room charge nurse to ask him about the admission, “So do the cops ever have to like wait around a bunch? It’s kind of surpris- ing a bed was open so quickly . . . right?” The nurse replied,

Look, there is a reason they have a charge nurse back here in the ambulance entrance and one in the front . . . these cops are com- ing in all the time with drug guys, sometimes the alcoholics, but mainly the drugs. I have a relationship with those guys, a working rela- tionship. I know what they go through. It’s hard out there. It’s a courtesy to them. . . . I mean legally they can’t just drop that guy off here, ’cause normally if that guy walked through the doors, right, he would be waiting a LONG time. There’s no way we can have those cops waiting in the ER for 12 hours when they know nothing’s wrong with the guy . . . it would basically kill the police . . . it’s just a thing they have to do.

The institutional preference for policing often comes into conflict with the goals of the front room triage staff. On many occasions, head nurses in the front room complained to me about patients whose ailments were so com- mon that had they walked through the front door they would simply be told to wait. As one front room nurse told me,

We give the patients the cops bring in prior- ity, I mean it’s good because it helps [the

cops] out, they are doing a job for the city, but I mean, in a way it kind of backs us up here. The people they bring in aren’t really sick, they wouldn’t qualify in any situation for the type of fast admission they get.

It is important to point out that triage staff are ambivalent in how they view the rushing of inmates and arrestees. While nurses express reluctance about the preference for police and their wards, they also accept public safety arriv- als because they maintain professional relation- ships with public safety staff. Ultimately, general patients in the front room lose beds to the endless supply of arrestees and inmates. The loss of beds occurs in a context in which, as described earlier, longer waits put pressure on the 21 percent of patients who leave the ER before being seen by a physician.

dIsCussIOn And COnCLusIOns This study details a dual effect on admission decisions: the rushing of medical resources for public safety wards and the delaying of resources for the general public and those with perceived criminality. These results pro- vide the first empirical evidence of how orga- nizational decision-making in health care settings both contributes to the health care access of the incarcerated and negatively affects the health care access of the non- incarcerated urban poor. In so doing, I bring together medical sociology’s study of health care workers and sociological work on incar- ceration and social control of the urban poor.

This dual effect suggests that the urban poor’s access to health care is mediated in countervailing ways by incarceration, polic- ing, and crime control language. Incarcera- tion affects health care access when a nontrivial number of inmates and arrestees are prioritized for beds in the back room, regardless of the severity of their ailments, resulting in a loss of beds for more medically serious general patients in the front room. Tri- age staff simultaneously decry the loss of beds—which adds to their difficulty in

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

Lara-Millán 881

admitting patients from the front room—and accept public safety arrivals out of profes- sional courtesy.

Regarding policing, police patrol the front room waiting lobby, which decreases access to health care resources for the non-incarcerated urban poor. While police periodically run background checks on waiting patients, which pressures patients with arrestable statuses to leave the ER early, I also detailed how police participate in vital-signs reassessment, which ends up expelling patients who have no offi- cial contact with the criminal justice system. Police justify these practices with a combined concern for crime control and helping the hospital solve its severe overcrowding. Recall that, according to triage protocol, C-list patients (who pack the front room waiting lobby) are technically never supposed to receive beds ahead of the always full B-list. Without police patrolling the front room, there would be little way for triage staff to resolve the C-list.

Crime control language shapes access to health care by allowing triage staff to accom- plish and justify the delay of qualified patients seeking care in the front room. Criminal stigma is assigned to waiting patients through the widespread administration of pain medi- cation; nurses develop a baseline understand- ing that patients seek care to access these opiates, and their denial of such patients is thus viewed as a kind of crime control. This baseline understanding is useful in reconcil- ing the most pertinent organizational conun- drum faced by triage staff: there are many more B-list patients who qualify for admis- sion than can be accommodated by the ER. When faced with such a conundrum and unable to find valid medical reasons to favor one qualified patient over another, medical staff invoke criminal stigma—often mediated by race and gender—to make final admission decisions. Without the invocation of crime and crime control, nursing staff would be unable to resolve the problems associated with an always full B-list.

While the main effect of the three crime control trends is decreasing access to health

care resources for the non-incarcerated urban poor, incarceration mediates this relationship by increasing access for the urban poor who happen to be in custody. A count (N = 1,114) of admission decisions corroborated the ubiq- uity of these observations: medical criteria influenced the majority of admission deci- sions, but criminal stigmatization of general patients and prioritization of arrestees and inmates influenced nearly as many decisions and were employed more often when nurses needed to use discretion.

Consequences for Population Health

The dual effects of crime control trends on the public ER are consequential for understand- ing the population health of the urban poor. First, the idea that the urban poor have more access to health care if they are incarcerated expands on Comfort’s (2007:271) contention that prison is becoming a “primary distributor of social services to the poor.” In the context of a diminished welfare state, America’s vast prison system may be the urban poor’s only source of what Comfort (2007:286) calls “carceralized aid”—access to health care, food, shelter, childcare, and other meager services offered in custody settings. In rela- tion to health care, this contention is sup- ported by incarcerations’ unexpected improvement of the mortality of incarcerated black men; that is, compared to their peers outside of prison, incarcerated black men are less likely to die. Patterson (2010) contends that while the protective effects of imprison- ment (protection from homicide or car acci- dents) explain some of incarceration’s effect on mortality, mandated in-custody health care offered to inmates also produces incarcera- tion’s positive effect on health (see also the discussion in Wildeman 2011). This develop- ment is possible both because state criminal justice spending is negatively associated with state health/welfare spending (Beckett and Western 2001; Ellwood and Guetzkow 2009) and because inmates have little access to health care outside of prison and often resume treatments once they enter into custody

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

882 American Sociological Review 79(5)

settings (Wilper et al. 2009). The rushing effects studied here suggest that the dynamics of public health care organizations also help explain inmates’ increased protection against death.

Second, these findings corroborate and expand on the idea that, outside of prison, ex- inmates are less likely to access public insti- tutions. Brayne (2014) uses nationally representative data to show that formerly incarcerated individuals are much less likely to utilize what she calls “surveilling institu- tions”—institutions, like hospitals, that tend to record information about their clients. Goffman (2009) elucidates ex-inmates’ per- spectives on this avoidance, documenting that they fear being re-arrested in hospitals. This study expands on these findings by showing that ex-inmates (and undocumented persons) fear hospitals for good reason: the public ER is a site of police patrols, and ERs systemati- cally rely on such patrols to solve their over- crowding problems. These findings may contribute, in part, to trends in population health: ex-inmates have far higher rates of mortality, communicable diseases, and disa- bility and, in general, have poor long-term health (Massoglia 2008; Schnittker and John 2007; Wildeman 2011).

Third, by taking a broader view of crime control trends—not simply incarceration or policing, but also expanded crime control language among health care professionals— this research is the first to document an organizational explanation for crime control’s additive effect on access to health care for the urban poor who have no criminal justice con- tact. When the urban poor seek care in the public ER, regardless of their criminal status, they will find fewer beds available to them and simultaneously face delays, policing, and deterrence from accessing health care. More- over, their medical diagnosis and chances of gaining admission will, in part, be shaped by crime control language that is pervasive among health care workers.

Note that the urban poor who have no direct criminal justice contact may have indirect con- tact through familial relationships with inmates

or ex-inmates. Indeed, romantic partners and children of ever-incarcerated persons are far more likely to experience communicable dis- eases, cardiovascular disease, poor mental health, and higher mortality (for reviews of this large body of research, see Schnittker et al. 2011; Wildeman and Muller 2012). Scholars have theorized that it is these individuals’ con- tact with the ever-incarcerated that results in poor health (i.e., through exposure to stress, diseases vectors, and the reduced earning power of ex-inmates). In contrast, the findings here suggest that this group’s negative health outcomes may also be due to crime control dynamics of health care organizations.

In short, the dual effect in the public ER reconciles and offers an organizational expla- nation for countervailing demographic trends in population health: incarceration’s positive effects on health outcomes for the incarcerated urban poor and negative effects on health outcomes for the urban poor outside of prison. By viewing crime control trends as more than just incarceration—studying how incarcera- tion interacts with policing and crime control language in an organizational setting—it is possible to identify these countervailing demographic trends as concordant. The organ- izational dynamics of health care settings and the decision-making of health care workers are potential mechanisms through which crim- inal justice trends contribute to the population health of the urban poor.

This dual effect has important implications for policymakers. As a result of an aging prison population and inmate rights litigation (McDonald 1999; Simon 2013), between 2001 and 2008 median state spending on inmate health care rose 52 percent (Chokshi 2013). In addition to adequately addressing inmates’ health care needs, policymakers must consider this spending’s effect on health care for individuals outside of custody. For instance, police patrol expenditures or crime crackdowns might also include assessments of their potential effect on local medical resources. Potential crime control benefits must be weighed against potential health con- sequences. Should health care for individuals

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

Lara-Millán 883

in custody and those not in custody be pro- vided in separate settings or, perhaps, more rationally linked? And, given high rates of recidivism among ex-inmates, what are the added economic costs of inmates who—after finding it difficult to obtain health care out- side of prison—return to prison with the same or worse health? Finally, although it is beyond the scope of this article, policymakers might consider the ethical and human rights impli- cations of criminal violations—and all the lifelong legal and economic consequences they entail—becoming de facto gateways to health care for the urban poor. In short, given expanded crime control efforts in U.S. cities, health care for inmates/arrestees and health care for the urban poor must not be consid- ered separate policy domains.

Contribution to Spillover Effects: Organizations, Not Just Returning Inmates

What does a focus on organizational dynam- ics imply about the study of spillover effects of imprisonment? Much of the scholarship on imprisonment’s broader impact on health (Schnittker et al. 2011), economic inequality (Harris, Evans, and Beckett 2010; Western and Pettit 2010), neighborhood attainment (Massoglia, Firebaugh, and Warner 2013), education (Hagan and Foster 2012), democ- racy (Manza and Uggen 2006), children and family dynamics (Wakefield and Wildeman 2014), and other features of U.S. life largely conceive of the inmate as the mechanism of influence for incarceration’s effects. Essen- tially, it is the inmate who receives negative effects from having ever been incarcerated and then exposes family members to these effects. Because the sheer number of inmates and ex-inmates is so large and so concen- trated in specific demographic groups and urban areas, they are likely to have an effect observable at aggregate measures.

By bringing a focus on organizations and medical sociology’s study of health care workers into dialogue with the sociology of imprisonment and crime control, this article

suggests that the spillover effects of incar- ceration may be conveyed not only by former inmates. Instead, researchers might also emphasize the role that organizational deci- sion-making plays in conveying incarcera- tions’ spillover effects into communities. Hypothetically, even while returning inmates attempt to avoid the usual negative influences of having been incarcerated—for example, by finding stable employment, reintegrating into daily family life, or removing other ex- inmates from their social networks—inmates and their families could still be negatively affected by crime control’s spillover effects on organizations.

Attention to organizational context high- lights the need for additional research. Long- term ethnographic research may help disentangle how organizational dynamics interact with other sources of disadvantage and how they accumulate over the life course. Statistical analyses could isolate the effect of types or characteristics of organizations on specific outcomes, such as morbidity, house- hold earnings, or educational attainment. For instance, in the case of educational outcomes, beyond variables that measure ever-incarcer- ated parents or delinquent peers, scholars could consider the presence of police in schools. Such organizational context varia- bles may affect people who have little to no formal contact with criminal justice.

Research might also compare the specifics of the organizational dynamics found here— rushing and delaying resources—with similar processes in other settings. Auyero (2012), for example, focuses on the continual and endless waiting the Argentinean poor must endure to obtain welfare checks, housing sub- sidies, identification cards, and court rulings. For Auyero, key features of waiting as a mode of statecraft are (1) spontaneous postpone- ments that come without warning and are given little or no explanation and the (2) inex- plicable granting of resources to some waiters over others. While Auyero (2012) contends that individuals selected for postponement and those selected to receive resources are seemingly chosen at random with little logic,

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

884 American Sociological Review 79(5)

the case of the urban public ER in the United States suggests there is discernible organiza- tional logic for why some groups of poor people are chosen for postponements over others. Through similar comparisons, schol- ars might build a more general theory of the logic of rushing and delaying in public insti- tutions that serve the poor.

The results of this study suggest that crime control trends in medical settings are an important component of how health care resources are distributed. The findings show that crime control mediates the urban poor’s access to health care in countervailing ways, making access easier for those in custody and more difficult for those not in custody. By focusing on a health care organization that primarily serves the urban poor—the public ER—I have laid the foundation for a deeper understanding of the degree to which incar- ceration, policing, and crime control language fuel health and health inequality.

Acknowledgments I thank Mary Pattillo for invaluable support and guid- ance. Also at Northwestern, I am grateful to Gary Alan Fine, Ann Shola Orloff, Celeste Watkins-Hayes, Monica Prasad, Arthur Stinchcombe, Elizabeth Onasch, and all the members of the Ethnography Workshop. Lara Medina, David Harding, Kerry Dobransky, Lynne Haney, Robert Vargas, Loïc Wacquant, Annette Lareau, Colin Jerolmack, Christopher Wildeman, Marcus Hunter, Megan Comfort, Stephen Sweet, Margaret Andersen, Nicole Van Cleve, Elijah Anderson, Camila Gripp, and the anonymous reviews gave invaluable suggestions.

notes 1. From 1996 through 2006, annual ER visits in the

United States increased from 90.3 to 119.3 million (32 percent), while the number of ERs dropped by 12 percent, resulting in a 78 percent increase in vis- its per ER between 1995 and 2003 (Pitts et al. 2006; Wilper et al. 2008). The situation is much worse in public emergency rooms located in the nation’s largest cities (General Accounting Office 2003).

2. Access to extended field notes will be provided on the author’s personal website at a future date.

3. A future study could presumably weight each factor. 4. No uniform method of triage currently exists. This

section draws from fieldwork. 5. Who are nurses accountable to? Typical of street-

level bureaucracies, there is a great deal of organi- zational buffer between managers and staff (Prottas

1979). Staff’s primary fear is misdiagnosing the rare “true emergency” patients and embroiling the hospital in a lawsuit. Additionally, the ER is peri- odically visited by state health agencies that have the power to cite the ER and disaccredit the hospi- tal. This ER was cited for dangerous overcrowding twice during my fieldwork; it is unclear how over- crowded the ER would need to be to actually garner disaccreditation.

6. Fire department/ambulance decisions are low because triage staff utilize medical criteria for patients brought to the ER by these agencies; these patients were therefore grouped in the medical decision-making category. Many of these arrivals are dropped off and enter into the normal front room triage process.

7. Fire department/ambulance were 3 percent of dis- cretionary decisions; 19 percent of discretionary decisions were unknown.

8. An association with criminality refers to associ- ating oneself, knowingly or not, with any type of engagement with the criminal justice system or the world of drug use, distribution, and violent crime. Criminality is a complex status, but in triage there is little time for making distinctions between types of crime.

9. I gathered these figures from the 1996, 2000, 2004, and 2008 versions of the Census on Law Enforce- ment dataset (Reaves 2011). The American Hos- pital Association annual survey and the National Hospital Ambulatory Medical Care Survey do not collect such data, so it is unknown how many of these police forces are in public hospitals located in high poverty, urban areas.

10. This information was garnered through field interviews with police officers in the back room entrance. These are the general reasons police offi- cers gave for bringing arrestees to public ERs. This does not shed light on the vetting process by which police or jails select some wards over others for ER care. This study is an examination of ER admis- sions and the reception of police referrals. Police decision-making is studied elsewhere (e.g., see Tep- lin and Pruett 1992).

References Albrecht, Gary L., David Slobodkin, and Robert J.

Rydman. 1996. “The Role of Emergency Depart- ments in American Health Care.” Pp. 289–318 in Research in the Sociology of Health Care, edited by J. J. Kronenfeld. Greenwich, CT: JAI Press.

Anderson, Elijah. 2012. “The Iconic Ghetto.” ANNALS of the American Academy of Political and Social Sci- ence 642(1):8–24.

Andrulis, Dennis P. and Lisa M. Duchon. 2007. “The Changing Landscape of Hospital Capacity in Large Cities and Suburbs: Implications for the Safety Net in Metropolitan America.” Journal of Urban Health 84(3):400–414.

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

Lara-Millán 885

Arendt, Katherine W., Annie T. Sadosty, Amy L. Weaver, Christopher R. Brent, and Eric T. Boie. 2003. “The Left-Without-Being-Seen Patients: What Would Keep Them from Leaving?” Annals of Emergency Medicine 42(3):317–23.

Auyero, Javier. 2012. Patients of the State: The Politics of Waiting in Argentina. Durham, NC: Duke Univer- sity Press.

Baker, David W., Carl D. Stevens, and Robert H. Brook. 1991. “Patients Who Leave a Public Hospital Emer- gency Department without Being Seen By a Physi- cian: Causes and Consequences.” Journal of the American Medical Association 266(8):1085–90.

Bazzoli, Gloria J., Ray Kang, Romana Hasnain-Wynia, and Richard C. Lindrooth. 2005. “An Update on Safety-Net Hospitals: Coping with the Late 1990s and Early 2000s.” Health Affairs 24(4):1047–56.

Becker, Howard S. 1998. Tricks of the Trade: How to Think About Your Research While You’re Doing It. Chicago: University of Chicago Press.

Beckett, Katherine and Bruce Western. 2001. “Govern- ing Social Marginality: Welfare, Incarceration, and the Transformation of State Policy.” Punishment and Society 3(1):43–59.

Berk, Marc L. and Claudia L. Schur. 2001. “The Effect of Fear on Access to Care among Undocumented Latino Immigrants.” Journal of Immigrant Health 3(3):151–56.

Bertram, Eva, Morris Blachman, Kenneth Sharpe, and Peter Andreas. 1996. Drug War Politics: The Price of Denial. Berkeley: University of California Press.

Brayne, Sarah. 2014. “Surveillance and System Avoid- ance: Criminal Justice Contact and Institutional Attachment.” American Sociological Review 79(3):367–91.

Chiarello, Elizabeth. 2013. “How Organizational Context Affects Bioethical Decision-Making: Pharmacists’ Management of Gatekeeping Processes in Retail and Hospital Settings.” Social Science & Medicine 98:319–29.

Chokshi, Niraj. 2013. “State Spending on Prison Health Care is Exploding. Here’s Why.” Washington Post, October 30.

Clear, Todd R. 2007. Imprisoning Communities: How Mass Incarceration Makes Disadvantaged Communi- ties Worse. New York: Oxford University Press.

Collins, Patricia H. 2000. Black Feminist Thought: Knowledge, Consciousness, and the Politics of Empowerment. New York: Routledge.

Comfort, Megan. 2007. “Punishment beyond the Legal Offender.” Annual Review of Law Social Science 3:271–96.

Comfort, Megan. 2008. Doing Time Together: Love and Family in the Shadow of the Prison. Chicago: Univer- sity of Chicago Press.

Dobransky, Kerry. 2011. “Labeling, Looping, and Social Control: Contextualizing Diagnosis in Mental Health Care.” Pp. 111–31 in Sociology of Diagnosis (Advances in Medical Sociology, Vol. 12), edited by

P. J. McGann and D. J. Hutson. Bingley, UK: Emerald Group Publishing Limited.

Durand, Anne-Claire, Stéphanie Gentile, Bénédicte Devictor, Sylvie Palazzolo, Pascal Vignally, Patrick Gerbeaux, and Roland Sambuc. 2011. “ED Patients: How Nonurgent Are They? Systematic Review of the Emergency Medicine Literature.” American Journal of Emergency Medicine 29(3):333–45.

Ellwood, John W. and Joshua Guetzkow. 2009. “Foot- ing the Bill: Causes and Budgetary Consequences of State Spending on Corrections.” Pp. 207–238 in Do Prisons Make Us Safer? edited by S. Raphael and M. A. Stoll. New York: Russell Sage Press.

Garland, David, ed. 2001. Mass Imprisonment: Social Causes and Consequences. London, UK: Sage.

Garland, David. 2002. Culture of Control: Crime and Social Order in the Contemporary Society. Chicago: University of Chicago Press.

General Accounting Office. 2003. Hospital Emergency Departments. Report to the Ranking Minority Member, Committee on Finance, U.S. Senate, Washington, DC.

George, Victoria and Alan Dundes. 1978. “The Gomer: A Figure of American Hospital Folk Speech.” Journal of American Folklore 91(359):568–81.

Gifford, Blair, Larry Manheim, and Diane Cowper. 2002. “Unforeseen Policy Effects on the Safety Net: Med- icaid, Private Hospital Closures and the Use of Local VAMCs.” Pp. 45–55 in Social Inequalities, Health and Health Care Delivery (Research in the Sociology of Health Care, Vol. 20), edited by J. J. Kronenfeld. Greenwich, CT: Emerald Group Publishing Limited.

Goffman, Alice. 2009. “On the Run: Wanted Men in a Philadelphia Ghetto.” American Sociological Review 74(3):339–57.

Gordon, James A. 1999. “The Hospital Emergency Department as a Social Welfare Institution.” Annals of Emergency Medicine 33(3):321–25.

Gustafson, Kaaryn S. 2011. Cheating Welfare: Public Assistance and the Criminalization of Poverty. New York: New York University Press.

Hagan, John and Holly Foster. 2012. “Children of the American Prison Generation: Student and School Spillover Effects of Incarcerating Mothers.” Law & Society Review 46(1):37–69.

Haney, Lynne. 1996. “Homeboys, Babies, Men in Suits: The State and the Reproduction of Male Dominance.” American Sociological Review 61(5):759–78.

Harris, Alexes, Heather Evans, and Katherine Beckett. 2010. “Drawing Blood from Stones: Legal Debt and Social Inequality in the Contemporary United States.” American Journal of Sociology 115(6):1753–99.

Hays, Sharon. 2004. Flat Broke with Children: Women in the Age of Welfare Reform. New York: Oxford Uni- versity Press.

Heimer, Carol and Lisa Staffen. 1995. “Interdepen- dence and Reintegrative Social Control: Labeling and Reforming ‘Inappropriate’ Parents in Neonatal Intensive Care Units.” American Sociological Review 60(5):635–54.

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

886 American Sociological Review 79(5)

Hinze, Susan W., Noah J. Webster, Heidi T. Chirayath, and Joshua H. Tamayo-Sarver. 2009. “Hurt Running From Police? No Chance of (Pain) Relief: The Social Construction of Deserving Patients in Emergency Departments.” Pp. 235–61 in Research in the Sociol- ogy of Health Care, edited by J. Kronenfeld. Green- wich, CT: Emerald Group Publishing Limited.

Hirschfield, Paul J. 2008. “Preparing for Prison? The Criminalization of School Discipline in the USA.” Theoretical Criminology 12(1):79–101.

Jeffery, Roger. 1979. “Normal Rubbish: Deviant Patients in Casualty Departments.” Sociology of Health & Ill- ness 1(1):90–107.

Katz, Mitchell. 2012. “Status Report on Key Indicators of Progress, Hospital Operations, and Other Issues Related to the New LAC+USC Medical Center— Progress Report #81.” Los Angeles County, CA, Department of Health Services, October 2.

Krug, Steven E. 1999. “Access and Use of Emergency Services: Inappropriate Use versus Unmet Need.” Clinical Pediatric Emergency Medicine 1(1):35–44.

Kupchik, Aaron. 2012. Homeroom Security: School Dis- cipline in an Age of Fear. New York: New York Uni- versity Press.

Kurz, Demie. 1987. “Emergency Department Responses to Battered Women: Resistance to Medicalization.” Social Problems 34(1):69–81.

Lewin, Marion E. and Stuart Altman, eds. 2000. Ameri- ca’s Health Care Safety Net: Intact but Endangered. Washington, DC: National Academies Press.

Lincoln, Alisa. 2006. “Psychiatric Emergency Room Decision-Making, Social Control and the ‘Undeserv- ing Sick.’” Sociology of Health & Illness 28(1):54–75.

Lipsky, Michael. 1980. Street-Level Bureaucracies: Dilemmas of the Individual in Public Agencies. New York: Russell Sage.

Malone, Ruth E. 1998. “Whither the Almshouse? Over- utilization and the Role of the Emergency Depart- ment.” Journal of Health Politics, Policy and Law 23(5):795–832.

Mannon, James M. 1976. “Defining and Treating ‘Prob- lem Patients’ in a Hospital Emergency Room.” Medi- cal Care 14(12):1004–1013.

Manza, Jeffrey and Christopher Uggen. 2006. Locked Out: Felon Disenfranchisement and American Democracy. New York: Oxford University Press.

Massoglia, Michael. 2008. “Incarceration, Health, and Racial Disparities in Health.” Law and Society Review 42(2):275–306.

Massoglia, Michael, Glenn Firebaugh, and Cody Warner. 2013. “Racial Variation in the Effect of Incarceration on Neighborhood Attainment.” American Sociologi- cal Review 78(1):142–65.

Mauer, Marc and Meda Chesney-Lind, eds. 2002. Invis- ible Punishment: The Collateral Consequences of Mass Imprisonment. New York: The New Press.

McDonald, Douglas. 1999. “Medical Care in Prisons.” Crime and Justice: A Review of Research 26:427–78.

Mizrahi, Terry. 1985. “Getting Rid of Patients: Con- tradictions in the Socialisation of Internists to the

Doctor-Patient Relationship.” Sociology of Health & Illness 7(2):214–35.

Muhammad, Khalil G. 2010. Condemnation of Black- ness: Race, Crime, and the Making of Modern Urban America. Cambridge, MA: Harvard University Press.

National Public Health and Hospital Institute (NHHI). 2011. Strategies to Improve ED Overcrowding: Data from the National Public Health and Hospital Insti- tute ED Throughput Study. Washington, DC: National Public Health and Hospital Institute.

Pager, Devah. 2008. Marked: Race, Crime, and Finding Work in an Era of Mass Incarceration. Chicago: Uni- versity of Chicago Press.

Patterson, Evelyn J. 2010. “Incarcerating Death: An Analysis of Mortality in United States’ State Correc- tional Facilities.” Demography 47(3):587–607.

Pitts, Stephen, Richard Niska, Jainmin Xu, and Catha- rine Burt. 2006. Hospital Ambulatory Medical Care Survey. Hyattsville, MD: National Center for Health Statistics.

Prottas, Jeffrey M. 1979. People Processing: The Street- Level Bureaucrat in Public Bureaucracies. Lexing- ton, MA: Lexington Books.

Quintanilla, Ray. 2006. “Old Ills Plague New Hospital.” Chicago Tribune, January 15.

Rabin, Roni. 2011. “Fewer Emergency Rooms Avail- able.” New York Times, May 17.

Reaves, Brian A. 1998/2002/2007/2011. Census of State and Local Law Enforcement Agencies. Washington, DC: U.S. Department of Justice.

Rios, Victor M. 2011. Punished: Policing the Lives of Black and Latino Boys. New York: New York Uni- versity Press.

Roth, Julius A. 1972. “Some Contingencies of the Moral Evaluation and Control of Clientele: The Case of the Hospital Emergency Service.” American Journal of Sociology 77(5):349–64.

Schnittker, Jason and Andrea John. 2007. “Enduring Stigma: The Long-Term Effects of Incarceration on Health.” Journal of Health and Social Behavior 48(1):115–30.

Schnittker, Jason, Michael Masssoglia, and Christopher Uggen. 2011. “Incarceration and the Health of the African American Community.” Du Bois Review 8(1):133–41.

Schwartz, Barry. 1972. Queuing and Waiting: Studies in the Social Organization of Access and Delay. Chi- cago: University of Chicago Press.

Simon, Jonathan S. 2007. Governing Through Crime: How the War on Crime Transformed American Democracy and Created a Culture of Fear. New York: Oxford University Press.

Simon, Jonathan S. 2013. “The Return of the Medical Model: Disease and the Meaning of Imprisonment from John Howard to Brown v. Plata.” Harvard Civil Rights-Civil Liberties Law Review 48:217–56.

Soss, Joe, Richard C. Fording, and Sanford F. Schram. 2011. Disciplining the Poor: Neoliberal Paternalism and the Power of Race. Chicago: University of Chi- cago Press.

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from

Lara-Millán 887

Sundararajan, Radhika L., Andrew L. Miner, William Paolo Jr., and William Chiang. 2012. “Police Detain- ees in The Emergency Department: Who Do We Admit?” Emergency Medicine 3(1):WMC002849.

Teplin, Linda A. and Nancy S. Pruett. 1992. “Police as Streetcorner Psychiatrist: Managing the Mentally Ill.” International Journal of Law and Psychiatry 15(2):139–56.

Timmermans, Stefan. 1998. “Social Death as Self- Fulfilling Prophecy: David Sudnow’s Passing On Revisited.” Sociological Quarterly 39(3):453–72.

Todd, Knox H., Christi Deaton, Anne P. D’Adamo, and Leon Goe. 2000. “Ethnicity and Analgesic Practice.” Annals of Emergency Medicine 35(1):11–16.

Vassey, Carine. 2001. “Categorisation and Micro-Rationing: Access to Care in a French Emergency Department.” Sociology of Health & Illness 23(5):615–32.

Wacquant, Loïc. 2009. Punishing the Poor: The Neolib- eral Government of Social Insecurity. Durham, NC: Duke University Press.

Waitzkin, Howard. 1989. “A Critical Theory of Medical Discourse: Ideology, Social Control, and the Process- ing of Social Context in Medical Encounters.” Jour- nal of Health and Social Behavior 30(2):220–39.

Wakefield, Sarah and Christopher Wildeman. 2014. Chil- dren of the Prison Boom: Mass Incarceration and the Future of American Inequality. New York: Oxford University Press.

Western, Bruce. 2006. Punishment and Inequality in America. New York: Russell Sage.

Western, Bruce and Becky Pettit. 2010. “Incarceration & Social Inequality.” Daedalus 139(3):8–19.

Wildeman, Christopher. 2011. “Imprisonment and (Inequality in) Population Health.” Social Science Research 41(1):74–91.

Wildeman, Christopher and Christopher Muller. 2012. “Mass Imprisonment and Inequality in Health and Family Life.” Annual Review of Law and Social Sci- ence 8:11–30.

Wilper, Andrew P., Steffie Woolhandler, Wesley Boyd, Karen E. Lasser, Danny McCormick, David H. Bor, and David Himmelstein. 2009. “The Health and Health Care of U.S. Prisoners: Results of a Nation- wide Survey.” American Journal of Public Health 99(4):666–72.

Wilper, Andrew P., Steffie Woolhandler, Karen E. Lasser, Danny McCormick, Sarah L. Cutrona, David H. Bor, and David U. Himmelstein. 2008. “Waits to See an Emergency Department Physician: U.S. Trends and predictors, 1997–2004.” Health Affairs 27(2):84–95.

Wortham, Stanton, Katherine Mortimer, and Elaine Allard. 2009. “Mexicans as Model Minorities in the New Latino Diaspora.” Anthropology and Education Quarterly 40(4):388–404.

Armando Lara-Millán is an Assistant Professor of Sociology at the University of California-Berkeley and a Robert Wood Johnson Scholar in Health Policy Research at UC Berkeley/UCSF. He is interested in how order is reproduced in highly uncertain and chaotic situations. He is completing a manuscript on how public hospitals and county jails are able to, despite disastrous underfunding, provide services to large numbers of people and create an illusion of policy success. Armando’s research is the recipient of awards from the National Science Founda- tion, Max Planck Institute for the Study of Societies, Ford Foundation, SSSP, and the ASA.

at FRANKLIN & MARSHALL COLG LIB on September 10, 2016asr.sagepub.comDownloaded from