Abstract

Context: Although New York State is a generous provider of substance-use treatment, people who ask for help have difficulty accessing services. If the laws are on the books, the agency is there to act, and the options are available, why is treatment so hard to get?

Methods: The authors conducted 87 open-ended interviews and observed local task force meetings in Sullivan County, New York. They open coded data, identifying key topics and themes.

Findings: Even though New York is a best-case scenario for treatment, people who seek help cannot always access it. The state sees the problem as a lack of beds or information about beds, but people on the ground face real barriers that make it difficult to get treatment, including the medical model of detoxification, admissions criteria, staff shortages, and other life complications.

Conclusions: Contrary to the policy literature, this article shows that policies may fail not because they are poorly designed or implemented but because the policy itself does not address the actual underlying problem. Furthermore, in the case of opioids, it shows how misplaced solutions can hide evidence of the underlying problem, exacerbating the very issue that policy makers want to fix.

Between 1999 and 2017, nearly 400,000 people died from opioid overdoses—roughly two-thirds of all drug-overdose deaths in the country—prompting the federal government to declare opioid use a national public health emergency (HHS 2017). In 2017, more than 3,000 New Yorkers died from an opioid overdose: a rate of 16.1 deaths per 100,000 people, which is higher than the national average of 14.6 (NIDA n.d.). Yet, New York State is a best-case scenario in terms of addressing the opioid epidemic. The state has removed major barriers to seeking treatment: It requires licensed facilities to provide services regardless of ability to pay and forbids insurance companies from requiring preauthorization or limiting treatment to 28 days without the right to appeal. New York even has an agency dedicated specifically to substance-use services, the Office of Addiction Services and Supports (OASAS).1 And, compared to other states, it has more public treatment options. For all of its advantages, however, we heard over and over again from people who work with or have a substance-use disorder about the difficulty accessing services. If the laws are on the books, the agency is there to act, and options are available, why is treatment so hard to get?

To answer this question, we talked to more than 80 people across New York State: policy makers in the state capital (Albany) as well as people on the frontlines of the opioid epidemic in rural Sullivan County. In discussions with local officials and families, we heard repeatedly that beds—or open slots for treatment at the doctor's office, treatment facility, or hospital—were a problem. Yet the state's online treatment locator tool showed plenty of openings, and people who worked at treatment facilities had space. We find that even in a well-resourced state like New York, services available in a database can be difficult or impossible to access in person, what we call the illusion of services. Contrary to the policy literature, we show that policies may fail not because they are poorly designed or implemented, but because the policy itself does not address the underlying problem, and, furthermore, the policy can hide evidence of the problem, exacerbating the very issue that policy makers want to fix.

Why Policies Fail: The Illusion of Services

Why do policies in place fail?2 Existing research suggests policies fail because they may be weak, poorly designed, or poorly implemented. However, even strong, well-designed, well-implemented policies may fail if policy makers misunderstand the underlying problem.

Policies may be designed to fail. Or, at the very least, they are designed to be inefficient or ineffective to achieve other goals. Robert Saldin (2017) illustrates how—although completely unsustainable—long-term care coverage passed as part of the Affordable Care Act of 2010 not because it was sound policy (it was not), but because its inclusion lowered projections of the overall cost of health care. Similarly, Eva Bertram (2015) demonstrates how expansions to the welfare state in the 1960s and 1970s came with conservative vehicles to thwart more generous policies and to permit future retrenchment. In sum, policies may be designed for purposes other than what their goals state.

Alternatively, policies may fail because policy makers are incentivized to please constituents rather than best address a problem. Policies with upfront preventative costs are less attractive to voters than policies that pay large amounts after the fact (Gailmard and Patty 2009), even though “investing in preparedness produces a large social benefit” (Healy and Malhotra 2009: 388). Political officials are encouraged to design poor policies because citizens and residents penalize them for taking long-term, more cost-effective action upfront.

Policies may also fail because they are poorly implemented. While top-down studies see implementation as the faithful execution of a policy's goals (Bardach 1977; Hogwood and Gunn 1997; Pressman and Wildavsky 1984; Van Meter and Van Horn 1975) and bottom-up scholars see street-level bureaucrats as policy makers, implementation scholars bridging this divide systematically lay out what may lead to better or worse implementation outcomes: features of the policy, the organization and people administering it, and politics. Policy features like traceability of problems being addressed, the extent to which a statute coherently structures the implementation process, and nonstatutory variables, such as media, public support, resources, commitment, and leadership skills (Sabatier and Mazmanian 1980), are important. Administrative features matter, too, such as institutional capacity of organizations responsible for making programs work and qualifications of those people in charge (Goggin 1990) as well as conflict and ambiguity surrounding certain policy implementation decisions (Matland 1995). Finally, electoral, group, and administrative politics (Manna and Moffitt 2019) also play a role.

Policies may fail when the wrong solution is coupled to an existing problem. John Kingdon (1984) famously explained the three streams of the policy process: problem, politics, and policy. Policy entrepreneurs wait for a window of opportunity to open, so they may couple their desired solution to whatever problem is in the national spotlight. In the words of David Rochefort and Roger Cobb (1993: 58): “the solution begets the problem.” But Kingdon, whose focus is on agenda setting, does not look at what outcomes arise when solutions do not fit the problem they are coupled with.

Researchers know that policy makers have imperfect information and foresight into what will happen in the future. Policy makers have bounded rationality, and they rely on habits and routine decision making (March and Simon 1958; Simon 1947). Policy makers, furthermore, can be myopic, without the ability “to clearly see the horizon of the future policy environment in which impacts of the policy will develop” (Nair and Howlett 2017: 104). Given the complexity of many problems—even with the best intentions—policy makers may not affix an adequate solution to a problem. Anecdotal evidence shows misplaced policy solutions, especially in public health: cholera epidemics wrongly attributed to miasma (bad air) rather than contaminants in water (Freedman 2008) or HIV/AIDS thought to be transmitted through routine household contact rather than blood (Shilts 1987). Yet, policies may fail—even in the short term—because policy makers do not fully understand the problem and craft a solution that does not match it. As Hogwood and Gunn (1997: 219) explain, a “policy may be based upon an inadequate understanding of a problem to be solved, its causes and cure; or of an opportunity, its nature, and what is needed to exploit it” (see also Bardach 1977; Pressman and Wildavsky 1984).

In the case of New York State, we find that a well-crafted and well-implemented policy solution, the New York State OASAS bed-locator tool, does exactly what it was designed to do: keep a tally of open treatment beds in the state. But the bed-locator tool does not actually fix the problem that people on the ground face: access to these open beds. Furthermore, the tool not only gives the impression that the state has addressed the problem, it also provides data to support state officials' claims, effectively masking the real problem and thwarting efforts to address it.

A Method of Listening

Like many states, New York has an opioid problem. But, unlike many other states, New York has taken comprehensive action to address it. New York requires all state-licensed facilities to provide treatment to anyone who wants it regardless of ability to pay. The state forbids insurance preauthorization for substance-use treatment services, and it does not allow 28-day insurance limits on treatment services without a process of appeal. Comparatively, the state has a wealth of resources, including an agency specifically devoted to substance-use services, as well as one of the largest drug-treatment systems in the United States. In 2016, New York admitted more people into state substance-use treatment programs than any other state (SAMHSA 2018). To understand why it is still difficult to access treatment, we examined rural Sullivan County, which has been heavily hit by the opioid epidemic, in greater depth.

Sullivan County has 75,500 residents spread across an area the size of Rhode Island. Although rural is often a mistaken euphemism for white, Sullivan has sizable African American (9.9%) and Hispanic (16.6%) populations (table 1). Roughly three-quarters of its population live in a rural part of the county, where access to health care, healthy foods, and employment is difficult. Like many rural communities across the country, Sullivan County's main industries (agriculture and tourism) receded, and the population has continued to decline.

Like many rural communities across the country, Sullivan County has a serious opioid problem. It has some of the state's highest emergency department admissions (NYSDOH 2017), highest overdose death rates (CDC Wonder n.d.), and highest opioid prescribing rates.3 Emergency department visits and hospitalizations are higher than average for upstate New York (figure 1), and overdose deaths continue to climb (figure 2).

Why are there so many opioid-related emergency department visits? Why are deaths still on the rise? And what prevents policy makers from doing more about it?

To better understand the problem, we used a method of listening (Cramer 2016), conducting open-ended, in-depth interviews with law enforcement, lawyers, judges, providers, doctors, nurses, social workers, local government officials, activists, families, and people in recovery. Our Sullivan interviewees were identified through a stratified four-snowball sample. We started with key informants' contacts in (a) the Sullivan County Prescription Drug Task Force, (b) health care community, (c) activist community, and (d) local leadership. From these four starting points, we asked interviewees who else we should speak with. We also interviewed a select sample of state policy officials relevant to the opioid epidemic and elected officials representing Sullivan. Interviews lasted forty-five minutes to two hours, and most (though not all) were taped and transcribed. Between December 2017 and January 2019, for this portion of the project, we talked to 87 people in upstate New York: 46 were based in Sullivan, 20 in neighboring Orange County (where many Sullivan residents go for substance-use services), 9 in organizations that served both counties, and 12 state officials in Albany. Additionally, we attended two Sullivan County Drug Task Force meetings and two meetings of the Sullivan County Perinatal Drug Task Force, as well as public forums. We open-coded (Emerson, Fretz, and Shaw 2011) the interview data we collected, identifying key topics and themes that came up frequently in our discussions.

Our analysis revealed: (1) the appearance of services that are available on paper but unattainable in practice and (2) the disconnect between the right issue—opioid epidemic—but the wrong solutions—more beds and more information—rather than structural barriers that prevent access to treatment. In this article we discuss how one policy solution—the bed-locator tool—not only fails to help people as intended but can actually mask the underlying problem. Agency officials and providers suggest more information will help people understand the many services that are available to them. However, these additional solutions are based on an illusion of services rather than the actual problem at hand: the specific barriers that keep people from being able to access treatment when they ask for it (see, e.g., Weaver 2015).

The Beds Problem

From the first day we stepped into Sullivan County, we heard about the problem of beds in our conversations with everyone from grassroots organizers (“there are no beds” [field notes 171211]) to the Commissioner of Health and Family Services (“getting a bed is a wait” [field notes 180118]). “Beds” does not always mean physical beds. It can be, in the language of OASAS, slots in an outpatient clinic. Beds is shorthand for treatment (un)availability in the community. One local official explained, “We hear all the time they're coming and saying ‘I need help,’ and we are like calling everyone we could find and there's no beds anywhere to send people. . . . So, I think that to me, that there's just a lack of long-term treatment beds” (interview 42). The beds discussion also suggests the opioid problem in local communities is much bigger than the available solutions. A local health officer told us how the hospital does not have a detoxification unit, there are thirteen beds in Sullivan's main city, Monticello, a youth facility in Fallsburg, and a third facility for women. “That is it. Two others went out of business. You combine all these factors: dynamite” (interview 66).

State officials heard about a lack of beds, too. The 2016 State Heroin and Opioid Task Force, chaired by Lt. Governor Kathy Hochul and OASAS Commissioner Arlene González-Sánchez, “repeatedly heard” about a shortage of treatment beds (Heroin Opioid Task Force 2016: 13). It recommended “the State take steps to increase the number of treatment beds and expand the type of treatment beds” (Heroin Opioid Task Force 2016: 13). The governor subsequently signed into law comprehensive reforms based on the task force recommendations a mere two weeks later, including an additional 270 treatment beds and 2,335 treatment slots, adding to more than 12,500 treatment beds across New York (NYSGO 2016b). New opioid treatment programs had opened in Albany, Buffalo, the Bronx, Peekskill, Plattsburgh, Syracuse, Rome, and Watertown, while residential treatment beds were added to facilities in Staten Island, Albany, Niagara, Suffolk, and Westchester (NYSGO 2016a).

Not everybody believes that there is a bed problem. Staff at treatment facilities told us they have open beds, but people with substance-use disorders just do not know about them. At a local task force meeting we attended, the facilitator noted how difficult it is for pregnant moms to navigate the different kinds of treatment and the different rules for getting in: “There's a maze of bureaucracy. Layers upon layers of service. If the experts can't navigate it, what do you expect from these moms?” (field notes 180711). A local official also expressed frustration with how complicated it is to get help: “There's all these great programs, but what good are they if no one knows about them? . . . You can create the best program in the world, but, if no one knows about it, what good is it? There's a disconnect between the programs being offered and what people know about them” (interview 57).

OASAS, too, saw an information problem. “At OASAS we had been hearing that there were no beds available,” explained an addictions specialist in an online tutorial. “So we thought there was a real need to have a quick, easy tool out there for people to be able to find beds” (OASAS 2017). The OASAS Treatment Availability Dashboard allows people to search for treatment beds across New York State based on distance, patient characteristics, and type of program. Anyone with access to the Internet can run a search, link to a description of the services offered, and download the results in a PDF file (OASAS n.d.-b). At any given time, New Yorkers can go online and see what facilities treat what kinds of addictions and whether there is space for them. “We've gotten fabulous feedback on this,” explained one agency official: “I've had people email me and call me directly, telling me how wonderful it is and how it's helped them. We have many providers that are using it to find placement for clients that they can't actually serve. We've had other states contact us because they're interested in how we developed this and they'd like to create something similar in their state. So, it's been very well received” (OASAS 2017).

To further deal with what it saw as a lack of information, OASAS created an ombudsman program in 2018 to “educate individuals, families, and health care providers on their legal rights to coverage, help them to access treatment and services and . . . investigate and resolve complaints regarding denial of health insurance coverage” (OASAS 2018). In 2019, OASAS launched a “Know the Facts” campaign “to dispel myths, provide facts, and raise awareness about addiction services in New York State, and will help direct people to addiction services and help” (OASAS 2019).

For New York State's treatment providers and OASAS, both of which work day in and day out on substance use, the refrain about lack of beds is not simply an infrastructure problem, it is an information problem. Even though the bed locator can help people find treatment more easily, and even though the number of treatment beds is growing, people still struggle to get help. We asked one provider why:

Patricia Strach: So, one of the things that I keep trying to figure out as I talk to folks at the state and they say “there's a thousand open beds, plus open slots, so there isn't a problem.”

Nonprofit Executive: Yeah. There's not a problem. There's open slots, yeah, there's open beds. I get emails every day from providers, “we have beds, we have beds, we have beds.” Sullivan County has beds, the crisis center up there, I get emails every day from them, Catholic Charities has beds, all the providers have openings in their out-patient slots. And then we have another provider who wants to put a 200-bed rehab in the community. And I was like . . . it's not like we don't, it's like that's not the problem. That's not the problem.

Patricia Strach: So, what is the problem? . . .

Nonprofit Executive: I don't know, I honestly don't know. (interview 95)

How can some community members and local officials believe there are no beds available while agency officials believe there is a 1000-bed surplus? The answer is the illusion of services. Yes, there are open beds. But, no, people on the ground cannot easily access them. One mother we spoke to explained how her daughter was turned away from the same treatment facility three times—the first because her daughter was on antidepressants, the second because her daughter needed to detoxify from fentanyl, and the third time because the hospital detoxified her with methadone, which made her ineligible for a bed because the treatment facility considered methadone a drug (even though the treatment facility referred her to that hospital in a neighboring county to detoxify). Being turned away from a residential facility because of drugs may seem counterintuitive. Nonetheless, in many of these facilities patients have to be detoxed before admissions, and, in some facilities, methadone is considered a drug rather than a medication. This young woman's plight illustrates how hard it is to access services even when the bed-locator tool shows an open slot at a treatment facility.

In the next section, we lay out barriers that keep people from accessing available services—the medical model for detoxification, admissions criteria, staff shortages, and other life challenges—showing how easy it is for people who need help, like the daughter above, to be turned away. The bed-locator tool in and of itself is not a bad thing. It allows people who need services to find them. But lack of information is not the main obstacle keeping people out of treatment, lack of access is. The dashboard does not address this more difficult and intractable problem.

The Illusion of Service in the Opioid Epidemic

Why are services so hard to access? Although state officials may be right—there are plenty of treatment beds across New York—solutions like the bed-locator tool fail to address the underlying problem: concrete barriers that bar access to treatment, even when beds are available.

The Medical Model Limits Access to Hospital Services

Under federal law, hospitals are required to provide appropriate medical care for emergency medical conditions, where the absence of medical care could be expected to place the health of the individual in serious jeopardy or serious impairment to bodily functions (CMS n.d.). People who have substance-use disorders routinely go to the hospital emergency department for help. However, withdrawal is typically not treated as an emergency medical condition, and emergency care is expensive. In 2008, New York State enacted new guidelines for detoxification, shifting to a community-based model of care for withdrawal (OASAS n.d.-a). As a result, many hospitals—including Sullivan County's only hospital—do not provide detoxification services.

Most hospitals treat addiction under a medical model of care, where individuals have to be experiencing physical withdrawal before they are admitted. Essentially, people who come to the emergency department for help are sent home unless they are experiencing, as one hospital social worker described, very painful symptoms: “shakes, dilated [pupils], sweats, whole body aches, severe body aches, restlessness, it's really hard for them to stay still ‘cause their aches are so bad. Body twitching, like their legs will twitch or their arms will twitch, is a sign. . . . Typically, they are nauseous, they're vomiting, they have diarrhea” (interview 65). People who have more complicated cases have a greater likelihood of being admitted to a hospital, such as patients with comorbid medical conditions, for example, diabetes or hypertension (SAMHSA 2005).

Although federal flexibility allows physicians at hospitals to treat emergency withdrawal with medications like buprenorphine, and although a state waiver allows hospitals to convert every medical bed into a detoxification bed (NYSDOH 2018), hospitals still turn patients away because they do not meet the medical criteria for withdrawal. Being admitted is so unlikely that the head of one nonprofit warns people about the difficulty. “We kind of ask [clients], because if they don't meet the admission criteria, which has nothing to do with treating addiction—it has to do with health—if they don't meet that, they are not going through an emergency room, and they're not getting into a hospital bed” (interview 95).

Prospective patients are often told to leave and come back when the symptoms are more severe. The hospital social worker above told us these conversations are difficult: “I'll say like ‘come back, like, can you hang out somewhere?’ [They respond:] ‘I'm homeless, no car, where do you want me to go, it's winter, what am I supposed to do?’ And those are valid points! Like, what are they supposed to do?” Many simply leave before they are admitted.

Admissions Criteria Limit Access to Treatment Beds

Unlike emergency departments, which are open 24 hours a day, community treatment facilities may only be open during business hours. As one provider explained, “If we have clients who work or have childcare issues, 9–5 Monday through Friday might not work” (interview 75). Yet, according to a state official, most “people don't usually say ‘I want to go to treatment’ Monday through Friday, 9–5” (field notes 190928). People who come for help after hours are not seen because the office is closed and there is no qualified staff member to admit them.

Even when treatment facilities are open, they do not accept every patient who comes through their doors. Facilities may turn people away if they fail to meet age or sex requirements for beds. Some beds are available only for adults whereas other beds are only available to a particular sex (male-only beds are not available to women, for example).

Treatment facilities may also turn people away if they cannot provide the appropriate level of care. For instance, not all facilities are licensed to provide medically supervised detoxification. Because withdrawal can involve complications and is often accompanied by potentially fatal side effects, treatment centers typically refer out for detoxification. But people seeking help do not always know the difference. As one local official explained, “Somebody who is experiencing addiction, or a family member who is making those calls, they can get turned away. They don't know the levels of care. It might not be the right level of care and then all of a sudden they're getting turned away because they need a higher level of care or a lower level of care, in which case they may not get into treatment” (interview 22).

People who need help may be turned away because they go to the wrong facility: they are not sick enough to be seen at an emergency department even if they are too sick or the wrong age/gender for a community treatment facility.

Staffing Shortages Make Accessing Care Difficult

Even when there are facilities with beds to treat people, severe staffing shortages mean those beds are, in practice, inaccessible to people who need them. Increased coverage for mental health means greater demand for psychologists and psychiatrists without an adequate supply (Olfson 2016). Furthermore, there are not enough physicians to provide medication-assisted treatment (MAT), which gives patients prescription drugs to block the effects of opiate withdrawal. MAT—through methadone, buprenorphine, or naltrexone—is the most effective treatment for opioid-use disorder when combined with intensive mental health and behavioral counseling. Yet, more than half of US counties lack physicians who can prescribe buprenorphine, leaving 30 million people without access (Rosenblatt et al. 2015).

The staff problem is widespread: a lack of social workers, credentialed alcoholism and substance abuse counselors (CASACs), and nurse practitioners. Half of agencies specializing in substance use say they have difficulty filling open positions, primarily because of a lack of qualified applicants (Hoge et al. 2013, citing Ryan, Murphy, and Krom 2012). Turnover is high (19% nationally, but 40% in some reports) because of low pay, few benefits, and heavy caseloads, as well as the stigma of working with addictions (Hoge et al. 2013). In Sullivan, social workers can carry caseloads of 70 or 80 clients (interview 60).

Staff shortages mean that beds can remain empty. A nonprofit worker in neighboring Orange County, whose job it is to connect clients to services, explained how one local treatment facility “is a great place. Clients are really happy with the treatment.” But, as much as she would like to place people there, “no one is answering the phone. You have to leave a message, and nobody gets back to you” (field notes 180711). If there is no receptionist to answer the phone or nurse to do intake, people who need help cannot get it at the facility regardless of what the bed-locator tool indicates.

The problem is not lack of beds, but staff shortages that make those beds unavailable to people who ask for help. As one person who worked for a nonprofit provider explained, “This is the frustration of the treatment programs . . . they keep expanding access to treatment, but you can't find a nurse practitioner to write buprenorphine” (interview 51). What few people there are who specialize in addiction services are hard to hire, and county governments and nonprofits cannot compete with hospitals and for-profit providers who can afford to pay higher salaries. Unable to offer a competitive salary, Sullivan County had difficulty filling five vacancies including four social work positions and a CASAC position. The county cannot pay well, because, as one government official observed, you “can't get something out of a dry well” (interview 60).

Although federal rules and regulations could alleviate the problem of staff shortages—by incentivizing people to pursue training as addiction specialists, for example—existing policies do just the opposite: they make it harder for people to find a provider who can prescribe MAT. Under existing law, physicians, dentists, veterinarians, physician assistants, nurse practitioners, and nurse midwives in New York can prescribe opioids, but MAT requires specialized clinics, trainings, and authorization. Methadone is a Schedule II drug, available only through highly regulated clinics, which patients have to visit daily when they first start methadone maintenance. Buprenorphine, a Schedule III drug, can be prescribed in physicians' offices, but it requires practitioners to obtain a DEA waiver, which includes an 8-hour training for doctors and an additional 16 hours of online training for physicians' assistants and nurse practitioners (CRS 2019). Ironically, it is far easier to prescribe opioids than the medication-assisted treatment to help people stop using them. As one state official explained, “If you want pills, limp into the ER and when they ask you how much pain you have say seven and you'll get it. But if you want methadone it's regulated as if it were weapons-grade plutonium. Clinics have to keep it in a huge safe. A vault. It prevents access” (field notes 190928).

In addition to limiting which medical personnel can provide MAT, federal regulations also limit how many patients they can treat. During the first year of buprenorphine certification, physicians can have up to 30 patients under treatment at one time. After a year, they can apply to have the number increased to 100, then 275 (SAMHSA n.d.-b). In 2016, less than 4% of physicians were waivered to prescribe buprenorphine in the US (Wakeman and Rich 2018). Of the 55,000 physicians who can prescribe buprenorphine in the US, 72% are 30-Patient Certified, 20% are 100-Patient Certified, and 8% are 275-Patient Certified (SAMHSA n.d.-c). Even if every physician prescribed at the limit there would still be more patients than treatment slots. But most physicians do not prescribe to the limit (Jones, Campopiano, and McCance-Katz 2015). More than 30 million people (10% of the population) do not have access to a single prescriber of medications for addiction treatment—the overwhelming majority (21 million) are in rural areas (Rosenblatt et al. 2015).

Other Life Challenges Make Accessing Treatment Difficult

Other life and logistical challenges make it hard to access treatment, too. Family obligations can be an impediment to care. Mothers are particularly hard to get into inpatient treatment because they do not want to leave their children. As one local official explained,

It's easier to take a male out of a home than a female, especially as far as caregivers. When we try to treat women[, it] is very difficult for us to get them to comply with any level, or any higher level of care anywhere, ’cause they don't want to leave their kids and their responsibilities and everything they have to do. Not saying that men don't feel the same way. But, for some reason, we can leverage them a little bit easier (interview 60).

Sullivan County is fortunate to have a women-only facility, which takes pregnant women and women with children younger than 3 years old. Many mothers, however, have children older than that.

Lack of transportation, too, makes it difficult for people who suffer from addiction to get the services they need. Transportation came up in 32 of our interviews. Medicaid will pay for taxis to medical appointments but not to the pharmacy, grocery store, or work. One person explained, “Access to just normal health care is really limited and . . . that's even aggravated substantially by the distances that people need to travel. So, you may live in like Cochecton or wherever and it's 40 minutes to Monticello, and there's probably no primary care physician closer, you know, so that creates an issue for people” (interview 32).

Mental health issues are also a barrier to access. Even though substance use frequently co-occurs with mental health disorders—in 2017, 18.7 million American adults had a substance-use disorder, and 45.6% of them (8.5 million people) also had a co-occurring mental illness (SAMHSA n.d.-a)—mental health providers will often reject someone with a substance-use disorder, and substance-use providers will often reject someone with a mental health issue because of the complications medications like benzodiazepines and opioids present (interview 49). One doctor described it this way: “It's like a house with two fires. Fire is the addiction. The second is the psychiatric co-morbidity. You have to put out both fires for this to work” (field notes 190927). Yet, it rarely works. Only half of people who have a substance-use disorder or mental health issue receive treatment for either, but a small fraction receive treatment for both (HHS 2018).

Although New York State eliminated some of the most daunting limitations on access—such as ability to pay and some insurance restrictions—even in this well-resourced state, we have found an illusion of services: beds are available in the OASAS system, but people on the ground cannot access them. The illusion can be difficult to combat because computer systems show capacity to treat patients even when capacity is not there (lack of staff, limits on how many people health care professionals can and will treat). Furthermore, patients may be turned away because they do not meet the demographic characteristics, they have co-occurring conditions, or they simply choose to forgo treatment because they are stymied by family or transportation issues.

When Policies Are Created Based on a Disconnect

New York State lawmakers have not shied away from addressing the opioid epidemic. In 2012, New York created a prescription monitoring program (I-STOP), requiring real-time data reporting (“Duty to Consult,” Chapter 447 of NYS Laws 2012; Heroin and Opioid Task Force 2016). In 2014, the state passed comprehensive legislation to address opioids, including initiatives for new state police officers, increased criminal penalties for selling narcotics, insurance regulations to make care easier to access, and a public education campaign (NYSGO 2014). It passed comprehensive legislation, based on the Heroin and Opioid Task Force recommendations, again in 2016.

Still, addressing beds has been a large part of the state's strategy. The state's two-pronged approach of expanding the total number of treatment beds and creating a bed locator may have enhanced treatment supply and provided information to the community, but it does not necessarily address the fundamental reasons people do not get treatment. As we have shown, people face very specific barriers: a medical model of detoxification, admissions criteria, staff shortages, and life challenges. In other words, the gulf between what officials believe to be the problem (lack of beds or information about beds) and what actually causes it (structural barriers) undermines the effectiveness of their solutions. So why have state policy makers and officials focused on beds and information instead?

There are many potential reasons why legislators and executive agency officials might address the problem as we have explained it here. First and foremost, treatment is expensive, and people who are dependent on drugs are not a well-regarded target population (Schneider and Ingram 1993). Lack of action may reflect lack of interest. Interest, however, does not seem to be a problem in New York. Although the state could certainly do more to address opioids in local communities, it provided treatment services to approximately 234,000 individuals in 2015 (OASAS 2016) and, in 2016, allocated (from all government sources, including Medicaid) over $1.4 billion for OASAS to address the crisis (Heroin Opioid Task Force 2016: 2).

Alternatively, organizational logic could have driven behavior in one of two ways. Agency officials at OASAS may have simplified the problem so that it could be addressed with an administratively easy and inexpensive solution (Scott 1998), or policy makers could have used standard modes of addressing problems, based as much on habits and routines as specific analysis of a particular problem (March and Simon 1958; Simon 1947). But here, too, it seems unlikely because New York has taken a broad range of actions: legislative solutions with comprehensive reforms in 2014 and 2016 and executive agencies expanding Medicaid under the 2010 Affordable Care Act to ensure that every New Yorker can afford drug treatment.

Finally, legislators may have misunderstood what constituencies across the state want and, in return, what the appropriate state reaction might be (Brookman and Skovron 2018; Hertel-Fernandez, Mildenberger, and Stokes 2018), either because they chose to listen to interest groups over citizens (Gilens and Page 2014), or because some communities were better able to convey their concerns as more important over others (Konisky and Reenock 2013). In our discussions with local officials, providers, and community members in Sullivan, we heard repeated, unprompted references to a disconnect between what communities say is the problem and what policy makers hear. One local official explained, “There's a disconnect between what the community provides and what the state believes.” The official continued,

So now you have the state talking to the county. But yet, the information that's down here, the people that are in the trenches, doesn't get up there. It just doesn't. And then they make decisions based on a disconnect. And then people scream loud enough and in 10–20 years we come back around and are having the same argument all over again. . . . I really think they need to turn off their brains, turn on their ears. (interview 60)

Although political scientists have shown that elected officials misperceive public opinion (Brookman and Skovron 2018; Hertel-Fernandez, Mildenberger, and Stokes 2018), the disconnect this official describes may reflect a broader lack of understanding about the concrete challenges people seeking treatment face and a broader inability to address the real problem. Even though state policy makers have not shied away from the epidemic, people on the frontlines feel as though the “money's being wasted” (interview 14).

Conclusion

Policies fail for many reasons, including poor design and/or poor implementation. In this article, we show that even a well-designed and well-implemented policy can fail when it does not address the underlying problem. We documented how beds, which are supposedly open per the computer screen, are not accessible to people who show up at the door, what we call the illusion of services. Although the state has put resources into creating beds and a bed-locator tool to provide information, these solutions do not address some of the main reasons why individuals who want help cannot get it: concrete barriers that limit access. In the case of the opioid epidemic in New York, the bed-locator tool provides data about a surplus of available help, but that help is unavailable to people who try to access it.

Although the bed-locator tool can be a valuable part of a broader strategy, it can also mask the true problem. It creates an illusion of services. Agency officials can point to open beds, and they believe that help is available when it is not. For example, in a conversation with an agency official we gave examples of what we have heard on the frontlines, including the following:

Patricia Strach: We talked to a mom, she's been in this 15 years, and we said, “Well, you know, what supports are there?” And she walked out of the room and then came back after she composed herself and said, “There's nothing.”

OASAS Official: There is.

The illusion of services can be more frustrating than having no services at all. In the case of the latter, it is clear what is not available and what people do not have access to. But it is exasperating to people on the frontlines of the opioid epidemic to see services that are supposedly available be just out of reach. The experience can fuel the idea of a disconnect between the governmental response and what people in communities need. As one mother put it, “Stop putting information out there where it looks like you're doing something and you're really not. . . . Stop the bullshit” (interview 14).

The illusion of services can prevent policy makers from putting the right solutions in place. To end the illusion, policy makers could target solutions to address specific problems that prevent more people from getting the help they need. First, policy makers could address the medical model of care by ensuring each county has a detoxification facility with easy and open access. Facilities might be the local hospital emergency department, which would be required to take patients needing detoxification services, or they could be a dedicated 24-7 building. Policy makers could create and enforce standard protocols for treating substance-use disorder in medical facilities, much like standard protocols for chest pain. Second, policy makers could address admissions criteria that lead community facilities to turn patients away, creating a system of care so everyone who asks for help is able to receive it. Third, policy makers could address staffing needs by incentivizing people to go into addiction-related specializations, especially in underserved areas (including rural communities). State policy makers could work with federal policy makers to standardize regulations on opioids and MAT, so it would not be easier to prescribe opioids than the medication to help people with opioid-use disorder. Fourth, state policy makers could reduce the barriers that people with co-occurring conditions face (hospitals should be able to treat these). Solutions like these, however, will be hard to pursue as long as the illusion of services is in place.

In this article, we examined substance-use treatment in Sullivan County. Yet, the concept of illusion of services is applicable to a broad array of failures in public-service provision, especially fragmented policy areas like health. For example, a guarantee for reproductive health services means very little if there are few or no clinics in a state or doctors to perform procedures (Dresser 2008; Feleder et al. 2019). Furthermore, the disconnect people feel between what government provides and what they need is not limited to Sullivan. In our broader research, we heard about a disconnect from people on the frontlines of the opioid epidemic in New York City, Syracuse, and Albany, too.

Not being able to access services is a problem. But with opioids, the consequences are especially dire. Every time someone is turned away from a supposedly available bed, the community loses an opportunity to save a life. The illusion of services means that people who try—and fail—to access services are invisible to state officials making public policies. The state-run systems to track treatment services show open slots, whether or not people on the ground can actually access them. Policies may be well designed and executed but they do not address—and in some cases exacerbate—a broader problem that policy makers wish to fix.

Acknowledgments

We thank the people on the frontlines in Sullivan and Orange Counties and the state officials in Albany for giving their time and for sharing their stories with us. We also thank the attendees at the Politics of the Opioid Epidemic conference, the Politics and History Group at the University at Albany, the JHPPL reviewers for their helpful comments, and the Rockefeller Institute of Government for supporting this project. Any errors are our own.

Notes

1.

OASAS changed its name in 2019 from Office of Alcoholism and Substance Abuse Services to Office of Addiction Services and Support.

2.

Our discussion here builds on the excellent work of Joseph Popcun (2018).

3.

Prescriptions were 66.5 per 100 people in 2017, down from 106.8 in 2012. See CDC drug overdose data at www.cdc.gov/drugoverdose/maps/rxrate-maps.html.

References

References
Bardach, Eugene.
1977
.
The Implementation Game: What Happens after a Bill Becomes a Law
.
Cambridge, MA
:
MIT Press
.
Bertram, Eva.
2015
.
The Workfare State: Public Assistance Politics from the New Deal to the New Democrats
.
Philadelphia
:
University of Pennsylvania Press
.
Brookman, David E., and Skovron, Christopher.
2018
. “
Bias in Perceptions of Public Opinion among Political Elites
.”
American Political Science Review
112
, no.
3
:
542
63
.
CDC (Centers for Disease Control and Prevention)
. n.d. “
Multiple Cause of Death Data
.” CDC Wonder. wonder.cdc.gov/mcd.html (accessed
November
13
,
2019
).
CMS (Centers for Medicare and Medicaid Services)
.
2012
. “
Emergency Medical Treatment and Labor Act (EMTALA)
.”
March
26
. www.cms.gov/Regulations-and-Guidance/Legislation/EMTALA/index.html.
Cramer, Katherine J.
2016
.
The Politics of Resentment: Rural Consciousness in Wisconsin and the Rise of Scott Walker
.
Chicago
:
University of Chicago Press
.
CRS (Congressional Research Service)
.
2019
. “
Opioid Treatment Programs and Related Federal Regulations.
June
12
. fas.org/sgp/crs/misc/IF10219.pdf.
Dresser, Rebecca.
2008
. “
From Double Standard to Double Bind: Informed Choice in Abortion Law
.”
George Washington Law Review
76
, no.
6
:
1599
622
. www.gwlr.org/wp-content/uploads/2012/08/76-6-Dresser.pdf.
Emerson, Robert M., Fretz, Rachel I., and Shaw, Linda L.
2011
.
Writing Ethnographic Fieldnotes
. 2nd ed.
Chicago
:
University of Chicago Press
.
Feleder, Florencia, Gowing, Katie, Mendez, Kadie, Taylor, Vanessa, and Weiss, Megan.
2019
. “
Beyond Roe: The State of Sexual and Reproductive Healthcare in New York State
.”
Rockefeller Institute of Government
,
January
2
. rockinst.org/issue-area/beyond-roe-the-state-of-sexual-and-reproductive-healthcare-in-new-york-state.
Freedman, David A.
2008
. “
On Types of Scientific Enquiry: The Role of Qualitative Reasoning
.” In
The Oxford Handbook of Political Methodology
, edited by Box-Steffensmeier, Janet M., Brady, Henry E., and Collier, David,
300
18
.
New York
:
Oxford University Press
.
Gailmard, Sean W., and Patty, John W.
2018
. “
Preventing Prevention
.”
American Journal of Political Science
63
, no.
2
:
342
52
. doi.org/10.1111/ajps.12411.
Gilens, Martin, and Page, Benjamin I.
2014
. “
Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens
.”
Perspectives on Politics
12
, no.
3
:
564
81
.
Goggin, Malcolm L.
1990
.
Implementation Theory and Practice: Toward a Third Generation.
Glenville, IL
:
Scott Foresman and Co
.
Healy, Andrew, and Malhotra, Neil.
2009
. “
Myopic Voters and Natural Disaster Policy
.”
American Political Science Review
103
, no.
3
:
387
406
.
Heroin and Opioid Task Force, State of New York
.
2016
. “
Combatting the Heroin and Opioid Crisis: Heroin and Opioid Task Force Report
.”
June
9
. www.governor.ny.gov/sites/governor.ny.gov/files/atoms/files/HeroinTaskForceReport_3.pdf.
Hertel-Fernandez, Alexander, Mildenberger, Matto, and Stokes, Leah C.
2018
. “
Legislative Staff and Representation in Congress
.”
American Political Science Review
113
, no.
1
:
1
18
.
HHS (US Department of Health and Human Services)
.
2017
. “
HHS Acting Secretary Declares Public Health Emergency to Address National Opioid Crisis
.”
October
26
. www.hhs.gov/about/news/2017/10/26/hhs-acting-secretary-declares-public-health-emergency-address-national-opioid-crisis.html.
HHS (US Department of Health and Human Services), Office of the Surgeon General
.
2018
.
Facing Addiction in America: The Surgeon General's Spotlight on Opioids
.
Washington, DC
:
US Department of Health and Human Services
.
Hoge, Michael A., Stuart, Gail W., Morris, John, Flaherty, Michael T., Paris, ManuelJr., and Goplerud, Eric.
2013
. “
Mental Health and Addiction Workforce Development: Federal Leadership Is Needed to Address the Growing Crisis
.”
Health Affairs
32
, no.
11
:
2005
12
.
Hogwood, Brian, and Gunn, Lewis.
1997
. “
Why ‘Perfect Implementation’ is Unattainable
.” In
The Policy Process: A Reader
, edited by Hill, Michael,
217
25
.
New York
:
Routledge
.
Jones, Christopher M., Campopiano, Melinda, Baldwin, Grant, and McCance-Katz, Elinore.
2015
. “
National and State Treatment Need and Capacity for Opioid Agonist Medication-Assisted Treatment
.”
American Journal of Public Health
105
, no.
8
:
e55
e63
.
Kingdon, John.
1984
.
Agendas, Alternatives, and Public Policies
.
Boston
:
Little, Brown, and Company
.
Konisky, David, and Reenock, Christopher.
2013
. “
Examining Sources of Compliance Bias in Policy Implementation
.”
Journal of Politics
75
, no.
2
:
506
19
.
March, James G., and Simon, Herbert.
1958
.
Organizations
.
New York
:
Free Press
.
Manna, Paul, and Moffitt, Susan L.
2019
. “
Traceable Tasks and Complex Policies: When Politics Matter for Policy Implementation
.”
Policy Studies Journal
,
June
17
. doi.org/10.1111/psj.12348.
Matland, Richard E.
1995
. “
Synthesizing the Implementation Literature: The Ambiguity-Conflict Model of Policy Implementation
.”
Journal of Public Administration Research and Theory
5
, no.
2
:
145
74
.
Nair, Sreeja, and Howlett, Michael.
2017
. “
Policy Myopia as a Source of Policy Failure: Adaptation and Policy Learning under Deep Uncertainty
.”
Policy and Politics
45
, no.
1
:
103
18
.
NIDA (National Institute on Drug Abuse)
.
2019
. “
Opioid-Involved Overdose Deaths
.”
New York Opioid Summary
,
March
. www.drugabuse.gov/opioid-summaries-by-state/new-york-opioid-summary.
NYSDOH (New York State Department of Health)
.
2017
.
Opioid Annual Report
.
October
. www.health.ny.gov/statistics/opioid/data/pdf/nys_opioid_annual_report_2017.pdf.
NYSDOH (New York State Department of Health)
.
2018
. DAL 18-05: “
Time Limited Waiver to Provide Detoxification Services in Excess of Bed/Patient Days Thresholds
.”
March
2
. www.health.ny.gov/professionals/hospital_administrator/letters/2018/2018-03-02_dhdtc_dal_18-05_waiver_detox_services.htm.
NYSGO (New York State Governor's Office)
.
2014
. “
Governor Cuomo Signs Legislation to Combat Heroin, Opioid, and Prescription Drug Abuse Epidemic
.”
June
24
. www.governor.ny.gov/news/governor-cuomo-signs-legislation-combat-heroin-opioid-and-prescription-drug-abuse-epidemic-0.
NYSGO (New York State Governor's Office)
.
2016a
. “
Governor Cuomo Announces Expanded Online Tool to Connect New Yorkers to Addiction Treatment Services
.”
December
6
. www.governor.ny.gov/news/governor-cuomo-announces-expanded-online-tool-connect-new-yorkers-addiction-treatment-services.
NYSGO (New York State Governor's Office)
.
2016b
. “
Governor Cuomo Signs Legislation to Combat the Heroin and Opioid Crisis
.”
June
22
. www.governor.ny.gov/news/governor-cuomo-signs-legislation-combat-heroin-and-opioid-crisis.
OASAS (New York State Office of Alcoholism and Substance Abuse Services)
. n.d.-a. “
Medically Managed Detoxification Reform
.” www.oasas.ny.gov/admin/hcf/mmdreform.cfm (accessed
October
24
,
2019
).
OASAS (New York State Office for Alcoholism and Substance Abuse Services)
. n.d.-b. “
Treatment Availability Dashboard
.” findaddictiontreatment.ny.gov/ (accessed
October
24
,
2019
).
OASAS (New York State Office of Alcoholism and Substance Abuse Services)
.
2016
.
Statewide Comprehensive Report 2015–2019
. www.oasas.ny.gov/pio/commissioner/documents/OASASStatewidePlan20152019.pdf (accessed
October
24
,
2019
).
OASAS (New York State Office of Alcoholism and Substance Abuse Services)
.
2017
. “
Learn about the Treatment Availability Dashboard
.”
May
11
. www.youtube.com/watch?v=mkdfnQFhEmA.
OASAS (New York State Office of Alcoholism and Substance Abuse Services)
.
2018
. “
OMH, OASAS, and DFS Announce New Program and Regulations to Help New Yorkers Access Insurance Coverage for Substance Abuse and Mental Health Services
.”
October
19
. apps.cio.ny.gov/apps/mediaContact/public/view.cfm?parm=7B67BF05–0A7F-7047–33DF96DD8A99F9CE&backButton.
OASAS (New York State Office of Alcoholism and Substance Abuse Services)
.
2019
. “
New York State Office of Alcoholism and Substance Abuse Services Announces New Campaign
.”
February
11
. www.oasas.ny.gov/pio/press/20190211OASASAnnouncesknowthefactscampaign.cfm.
Olfson, Mark.
2016
. “
Building the Mental Health Workforce Capacity Needed to Treat Adults with Serious Mental Illnesses
.”
Health Affairs
35
, no.
6
:
983
90
. doi.org/10.1377/hlthaff.2015.1619.
Popcun, Joseph.
2018
. “
Defining and Understanding Ordinary Failures in Public Policy
.” Unpublished paper.
Pressman, Jeffrey, and Wildavsky, Aaron.
1984
.
Implementation: How Great Expectations in Washington Are Dashed in Oakland
.
Berkeley
:
University of California Press
.
Rochefort, David A., and Cobb, Roger W.
1993
. “
Problem Definition, Agenda Access, and Policy Choice
.”
Policy Studies Journal
21
, no.
1
:
56
71
.
Rosenblatt, Roger A., Andrilla, C. Holly A., Catlin, Mary, and Larson, Eric H.
2015
. “
Geographic and Specialty Distribution of US Physicians Trained to Treat Opioid Use Disorder
.”
Annals of Family Medicine
13
, no.
1
:
23
26
.
Ryan, Olivia, Murphy, Deena, and Krom, Laurie.
2012
.
Vital Signs: Taking the Pulse of the Addiction Treatment Workforce, A National Report, Version 1
.
Kansas City, MO
:
Addiction Technology Transfer Center National Office in residence at the University of Missouri-Kansas City
.
Sabatier, Paul, and Mazmanian, Daniel.
1980
. “
The Implementation of Public Policy: A Framework of Analysis
.”
Policy Studies Journal
8
, no.
4
:
538
60
.
Saldin, Robert P.
2017
.
When Bad Policy Makes Good Politics
.
New York
:
Oxford University Press
.
SAMHSA (Substance Abuse and Mental Health Services Administration)
. n.d.-a. “
Key Substance Use and Mental Health Indicators in the United States: Results from the 2017 National Survey on Drug Use and Health
.”
September
. www.samhsa.gov/data/sites/default/files/cbhsq-reports/NSDUHFFR2017/NSDUHFFR2017.pdf (accessed
November
13
,
2019
).
SAMHSA (Substance Abuse and Mental Health Services Administration)
. n.d.-b. “
Apply for a Practitioner Waiver
.” www.samhsa.gov/medication-assisted-treatment/buprenorphine-waiver-management/increase-patient-limits (accessed
October
24
,
2019
).
SAMHSA (Substance Abuse and Mental Health Services Administration)
. n.d.-c. “
Practitioner and Program Data
.” www.samhsa.gov/programs-campaigns/medication-assisted-treatment/training-materials-resources/physician-program-data (accessed
October
24
,
2019
).
SAMHSA (Substance Abuse and Mental Health Services Administration)
.
2005
. “
Substance Abuse Treatment for Persons With Co-Occurring Disorders
.” Treatment Improvement Protocol (TIP) Series 42, Report No. (SMA) 05-3992. www.ncbi.nlm.nih.gov/books/NBK64197/ (accessed
October
24
,
2019
).
SAMHSA (Substance Abuse and Mental Health Services Administration)
.
2018
. “
Treatment Episode Data Set (TEDS) 2016: Admissions to and Discharges from Publicly Funded Substance Use Treatment
.”
August
. www.samhsa.gov/data/sites/default/files/2016_Treatment_Episode_Data_Set_Annual.pdf.
Schneider, Anne, and Ingram, Helen.
1993
. “
Social Construction of Target Populations: Implications for Politics and Policy
.”
American Political Science Review
87
, no.
2
:
334
37
.
Scott, James.
1998
.
Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed
.
New Haven, CT
:
Yale University Press
.
Shilts, Randy.
1987
.
And The Band Played on: Politics, People, and the AIDS Epidemic.
New York
:
St. Martin's Press
.
Simon, Herbert A.
1947
.
Administrative Behavior.
New York
:
MacMillan
.
Van Meter, Donald, and Van Horn, Carl.
1975
. “
The Policy Implementation Process: A Conceptual Framework
.”
Administration and Society
6
, no.
4
:
445
88
.
Wakeman, Sarah E., and Rich, Josiah D.
2018
. “
Barriers to Medications for Addiction Treatment: How Stigma Kills
.”
Substance Use and Misuse
53
, no.
2
:
330
33
. doi.org/10.1080/10826084.2017.1363238.
Weaver, R. Kent.
2015
. “
Getting People to Behave: Research Lessons for Policy Makers
.”
Public Administration Review
75
, no.
6
:
806
16
.