Provider views of harm reduction versus abstinence policies within homeless services for dually diagnosed adults

Abstract

Harm reduction is considered by many to be a legitimate alternative to abstinence-based services for dually diagnosed individuals, yet there is limited understanding of how varying approaches affect front-line practice within supportive housing services for homeless adults. This paper examines how front-line providers working with individuals who have experienced homelessness, serious mental illness, and addiction view policies of harm reduction versus abstinence within two distinctly different approaches to homeless services: the traditional or ‘treatment first’ (TF) approach that requires abstinence and the more recent Housing First (HF) approach that incorporates harm reduction. As part of a federally-funded qualitative study, 129 in-depth interviews conducted with 41 providers were thematically analyzed to understand how providers view a harm reduction versus abstinence approach. Themes included: (a) harm reduction as a welcomed alternative; (b) working with ambiguity; and (c) accommodating abstinence. Drawing on recovery principles, we consider the broader implications of the findings for behavioral health care with this population.

Introduction

The majority of programs serving homeless adults with serious mental illness and co-occurring substance abuse require detoxification and ongoing abstinence in order to receive services including transitional and permanent housing.¹ Yet abstinence-based, congregate housing, with its ‘all or nothing’ approach, has not been effective at engaging or retaining chronically homeless adults accustomed to a transient life on the streets.^2,3

Although harm reduction began as a grass-roots effort to limit the spread of HIV and hepatitis B infections in the 1980s,⁴ it is now considered a legitimate alternative to an abstinence-based approach to addressing drug use in a broad array of health and mental health services.⁵ Within homeless services harm reduction became a key element of homeless services with the rise of ‘housing first.’^6,7 Housing first is an approach that stands in contrast to more traditional programs that follow a residential continuum or graduated approach. Within traditional programs, temporary shelter is initially offered followed by transitional housing before permanent housing can be attained. Graduating through these stages of housing requires that individuals meet treatment goals that include prolonged abstinence.⁸ Housing First, as the name implies, starts by placing people directly into permanent housing and uses a harm reduction approach, which effectively de-links housing and treatment. Although assistance with controlling substance use is offered, abstinence is not mandated as a requirement for maintaining one’s housing. Whether permanent housing comes in the form of congregate settings such as ‘wet housing’ for groups of individuals who use alcohol or independent apartment living that is scattered throughout the community, harm reduction has been shown to improve a variety of outcomes at reduced costs within a housing first context.^6,7,9 Yet its adoption is far from widespread.^{1, 10}

The embedding of harm reduction within the Housing First [HF] model is not a foregone conclusion---traditional programs could have ostensibly adopted harm reduction (and some have in recent years). However, there are obvious challenges to implementing harm reduction in congregate housing, e.g., the contagion effects and disruptions of co-habiting users. And there are understandable differences between tolerating abuse of alcohol versus illegal drugs in congregate settings. Local governments and neighborhoods are far less likely to approve of the latter than the former (hence the presence of ‘wet housing’ for those addicted to alcohol is recognized more so than ‘drug-tolerant housing’). In contrast, HF that relies on scatter-site apartments renders harm reduction a matter between the consumer and his/her service provider. If the substance abuse becomes problematic—endangering the health or safety of the consumer or other building residents—the HF program will strongly urge treatment and rehab with the understanding that consumers can return to their apartment once discharged from treatment and/or their condition improves. As originally implemented, if the landlord proceeds with eviction (on the same grounds as any tenant’s behavior might warrant), HF programs provide equivalent housing, otherwise stated as ‘re-housing’ the consumer.¹¹

Harm reduction, like many other aspects of service provision, is subject to discretionary decision-making in real-world practices. Except in the most rigid zero-tolerance programs such as therapeutic communities, staff may choose to overlook signs of intoxication or give private warnings to the consumer. However, a stricter regime dependent on random urine testing and close surveillance is clearer in implementation and enforcement.

For persons with serious mental illness, the use of alcohol and drugs has been characterized as ‘self-medication’ of psychotic symptoms¹² as well as a diversion from the harsh realities of life on the streets.¹³ Once initiated, substance dependence presents the usual challenges of habituation with the added complications of psychiatric symptoms and, for the subset who become chronically homeless, histories of trauma and adversity.¹⁴

The appeal of leaving the streets and obtaining services, leveraged by an offer of temporary housing (and the promise of eventual permanent housing), would appear undeniable. But when that offer is predicated upon immediate abstinence and adherence to a set of rules—curfews, daily supervision, mandatory attendance at day treatment, no visitors, urine testing, etc.—engagement can falter. Consumers view this ‘all or nothing’ proposition as leaving them with few options beyond complying or returning to homelessness.¹⁵

The idea of mental health recovery that has been embraced by grass-roots advocacy groups and policy makers at the highest levels of government,¹⁶ is made complicated by co-morbid substance use.¹⁷ Mental health recovery proponents have largely been silent on how this obstacle might or should be overcome (beyond promoting consumer choice and empowerment in decision-making).^{18, 19} On the other hand, the Substance Abuse and Mental Health Services Administration recently updated its definition of mental health recovery by taking the position that “abstinence is the safest approach for those with substance use disorders.”²⁰

Few would disagree that substance use and relapse can inhibit the larger process of mental health recovery^21,22 especially for those experiencing homelessness. What remains poorly understood are how cycles of rehab and relapse are dealt with ‘on the ground’ by providers working in homeless services for adults with serious mental illness.²³ After all, consumer driven services are fundamental to mental health recovery¹⁶ but this does not mean that providers must condone an individual’s choice to abuse drugs. Rather than supporting or ‘enabling’ substance abuse, however, the goal of harm reduction is to limit or prevent negative consequences associated with it. As such, it offers a non-judgmental, pragmatic approach to working with individuals who use and abuse substances.²⁴

The rise of HF has thrown into sharp relief the pivotal role of harm reduction in a new arena--homeless services. Yet this role has received little attention from researchers.²⁵ Abstinence based residential programs are based upon a behaviorist therapeutic community model featuring a structured peer-led environment steeped in a system of rewards and punishments.^26,27 In contrast, harm reduction is characterized as a pragmatic approach, i.e., a set of practices such as needle-exchange programs for intravenous opiate users, methadone for heroin users or nicotine patches for cigarette smokers.²⁸ It is also based on a stages of change model that recognizes people who are actively using may still be working towards recovery.²⁹

This report examines how front-line providers understand policies of harm reduction versus abstinence within two distinctly different approaches to homeless services: the relatively recent Housing First (HF) approach or the more traditional service approach sometimes referred to as ‘treatment first’ (TF) by way of comparison.³⁰ Both approaches are designed to serve an array of needs beyond housing to promote stability and a move toward independent living.¹⁶ Our research questions include: (1) How do providers understand a harm reduction approach? (2) What do providers view as the benefits and challenges to abstinence based or harm reduction approach? (3) In what ways do providers apply these different approaches to promoting mental health recovery (i.e. consumer driven services)?

Methods

Sampling

The 41 front-line providers included in this study were part of the New York Services Study, a longitudinal qualitative study of new consumer enrollees of programs serving homeless adults with co-occurring psychiatric and substance use disorders. Twenty providers were from a program that used the Housing First model and 21 providers came from three other programs that were part of the residential continuum “Treatment First” model. Providers were recruited through their client’s participation in the study during a 1 year recruitment period in which all new enrollees of services were invited to participate (individuals without an Axis I diagnoses of serious mental illness such as schizophrenia or bi-polar disorder and a history of substance abuse were excluded). 83 client-participants gave informed consent to be interviewed and to have their provider at the program be interviewed – all of whom consented. Most providers (n=30) had multiple clients enrolled in the study. Both client and provider participants were paid $30 per interview and all study protocols were approved by the authors’ university institutional review board.

Data Collection Procedures

Study protocols included multiple in-depth interviews with providers: baseline interviews within a month of their client’s enrollment in the study and follow-up interviews either six-months later or when their client left the program, whichever came first (follow-up interviews were not conducted if baseline provider interviews closely coincided with a client’s departure). In total, the 41 providers participated in 129 interviews (79 baseline and 50 follow-up). Even with a small provider sample, significant group differences found HF providers to be predominantly white compared to higher percentages of African-Americans and Latinos providers in TF (p=.018). Although not significant, there were also higher percentages of graduate level providers within the HF program [see table 1].

Table 1.

Demographic Characteristics of Providers

	Housing First, n=20	Treatment First, n=21	Total n=41
Gender
Male	9 (45%)	7 (33%)	16 (39%)
Female	11 (55%)	14 (67%)	25 (61%)
Race/Ethnicity
White	12 (60%)	6 (29%)	18 (44%)
African American	5 (25%)	9 (43%)	14 (34%)
Latina/o	1 (5%)	6 (29%)	7 (17%)
Other	2 (10%)	0	2 (5%)
Highest educational degree
Graduate	13 (65%)	7 (33%)	20 (49%)
Bachelor	4 (20%)	9 (43%)	13 (32%)
Associate	2 (10%)	5 (24%)	7 (17%)
High School	1 (5%)	0	1 (2%)
Length of employment
< 1 year	6 (30%)	9 (43%)	15 (37%)
1–3 years	6 (30%)	7 (33%)	13 (32%)
> 3 years	8 (40%)	5 (24%)	13 (32%)

Semi-structured interviews were conducted by four trained interviewers familiar with the mental health service system usually in a private office at the provider’s agency. The interviews lasted approximately 30–45 minutes, with interviewers asking providers both general questions about their work experience, as well as client-specific questions. Interviewers were trained to ask additional probing questions based on providers’ answers. Some of the questions included: What is working here like for you? What’s your approach to working with clients who have serious mental illness along with substance use disorders? What are some of the challenges you face on the job? Questions and probes about harm reduction or abstinence were asked only when providers initiated discussion of those topics. However, given our sampling criteria of substance abuse history, dealing with the prospect of substance abuse (whether currently using or in the past) was a common topic in the interviews. All interviews were transcribed verbatim and entered into ATLAS/ti software.

Data Analysis

Thematic analysis³¹ was employed that consisted of the following process: (1) generating codes to be attached to similar quotes or topics within the transcripts for data reduction; (2) reviewing these codes and associated quotes to identify themes that fit with the data across all of the transcripts; and (3) identifying both positive and negative examples or qualifications to the themes to ensure that they comprehensively and accurately represent the data

Close to half of the transcripts (n=59) were independently coded by two members of the research team and then compared in order to reach a consensus about the appropriateness of assigning a particular code to a given passage and on the nature of the themes emerging from the data. Examples of codes include: accessing housing, substance use, discretionary power, and system obstacles/problems. Ongoing memo-writing was used in the development of themes that allowed for the exploration of ideas and the documentation of analytic decision.³² The first author continued to code the remaining 70 transcripts to further refine thematic analysis by identifying supporting and negative cases in the raw data, making sure that multiple transcripts from the same provider were not exclusively used in the testing of themes to avoid relying disproportionately on a single provider’s experience. ATLAS.ti software was used in this process to help separate and sort coded material based on program type.

Several strategies for rigor were employed including peer-debriefing within the data collection and analytic processes, independent co-coding of transcripts, refinement of themes through negative case examples, and the use of memo-writing to aid in the development of ideas as well providing a decisional audit trail.³³

Results

Although HF providers discussed a harm reduction and abstinent based approach that indicated experiences with both, TF providers rarely acknowledged an alternative to an abstinent based approach. This resulted in the following three themes that largely reflect HF provider perspectives.

A welcomed alternative

TF providers discussed how an abstinence approach resulted in many consumers either being mandated to intensive treatment such as detox or rehab or dropping out of services altogether. At times these providers tried to accommodate consumers by finding a middle ground. As one TF provider explained, “So we really didn’t know that she- you know, we suspected it [drug use] but I don’t think we really pursued it because we didn’t want to lessen her chance of getting housing. A minority of TF providers questioned the approach, “I think that these kinds of places are inherently revolving doors…I don’t know if there needs to be a total revamping of the rules or if there needs to just not be any or change some of the rules that they have…” These providers did not, however, invoke harm reduction as an alternative approach.

When weighing alternatives, HF providers endorsed harm reduction over abstinence. As one provider described, “here we use the harm reduction model instead of abstinence, so that’s been a powerful eye opener to how you can really help people to facilitate change.” It is important to note that within the context of programs that provide housing, harm reduction meant that housing was not contingent on sobriety. This resonated with many of the HF providers, with one provider explaining,

“it changed my perspective on everything because I worked at a drug TC program before, where abstinence is definitely a must before they even try to get you housing and stuff like that. And I worked at other like I said mental health agencies where you have a drug problem, that’s the criteria, you have to come in and go through a detox or whatever the case is, and then we’ll start. And that deters a lot of people.”

Several HF providers discussed feeling liberated, with one person identifying, “I’m a harm reductionist so it’s a great place to work. The freedom that you have to work with your client, I haven’t been able to find anywhere else, and I don’t know any of my friends who get to work so freely.”

When endorsing a harm reduction approach, most HF providers noted the advantage that consumers would be more honest about their substance use while staying engaged in services. As one HF provider explained, “there are all these variables that might impact the person’s willingness to be open [about substance use]. So recognizing all those, and that most of what we are told is probably not 100% honest, and just being comfortable with that and not taking it personally.” Another HF provider explained that harm reduction allowed them to treat their consumers “with respect, recognizing and pointing out that a stigma exists, that I understand that they might not be so willing to be so honest with me because that’s the reality of the world we live in.” In fact, the goal for many HF providers using harm reduction was to foster an open dialogue about substance use,

Yeah, you can come and tell me you’ve used. And I’m actually kind of happy when people come in and say… we had a harm reduction group and this woman had relapsed. She goes, “Aren’t you mad at me?” I said, “No, I’m really happy because it’s a huge success that you came in and told us. And if you keep using, then you keep using.” And she’s like, “But I don’t want to.” And I thought that’s the thing people have to decide for themselves.

In part because of a more honest relationship with consumers, HF providers using harm reduction often described their job as ‘less stressful’, since as one person explained, “it’s not my expectation, and the client needs to see that. When you’re badgering someone and constantly-trying to reinforce something [abstinence] a lot of times you’re pushing a client.”

Working with ambiguity

Although harm reduction was preferable to the HF providers who discussed both approaches, many noted ongoing challenges. Expressing some insecurity, one HF provider confessed, “Sometimes I think, “oh, I’m not doing enough.” And maybe I’m gonna be called out on this, like, “oh, you’re the substance abuse specialist. What’s your problem?” Because I’m not getting people into treatment, but to me, that’s not what this is about.” TF providers also expressed doubt, with one stating, “I think that the whole system is inherently flawed because if it wasn’t the people wouldn’t need to go to treatment so many times.” Although there was agreement that residential drug treatment could help consumers with abstinence, providers described widespread concern that this wouldn’t last upon discharge. As one TF person noted, “To put them in a treatment facility is not a natural, organic kind of…you can learn how to not use when you’re on lockdown and your next door neighbor is gonna rat on you.”

Using harm reduction in community settings meant working with consumers to make difficult choices. In discussing a particular consumer who had previously been evicted when his apartment had been taken over by drug-using peers, one HF provider describes,

“That’s his harm reduction strategy is, everything else is replaceable but once I let them in here [to my apartment] and that jeopardizes my housing…and especially in his case where he does have a suicidal history…he knows how dangerous a proposition that is. So that for him is where he can draw the line.”

HF providers acknowledged, however, that developing harm reduction strategies can be difficult since consumers may still be reticent to openly discuss their substance use,

“He might feel that we will shut down his apartment if he’s not doing well, which I think we try our hardest to tell people that that’s not really the case, that apartments only get shut down due to behavioral things that threaten the apartment, not necessarily just because of drug use.”

Harm reduction, therefore, was sometimes seen as a specific strategy but also seen as a more general and unspecified approach,

“the way we operate it’s harm reduction approach so it’s very vaguely wherever that client is is where we want to meet them where they’re at and not try to impose what we think is best of them because if they don’t want abstinence it doesn’t matter how much we bring it out, so we just help them kid of manage what they’re doing”

The ambiguity that providers experienced within a harm reduction approach was mostly missing from the TF providers. As one explains, things were more clearcut,

“You say you want an apartment, you want to open up a bank account, you want entitlements…so you have to draw the picture so that they can see it. In order for you to do all these things, then drugs are not an option. The first step is getting them into detox. Then giving reasons why we need to get them into detox. From there, you’ve got inpatient and outpatient rehabilitation because those are the issues of x, y, and z. If you give them that picture, some of them will take it and some of them will not come back. That’s the risk we take because we have to let them know that it’s not gonna be easy.”

While policies were more straightforward within the TF programs, roughly one-third of TF providers expressed ambiguity in terms of whether they would enforce certain policies. As one TF provider explained, “I didn’t try to force him into rehab. I took his word that he’s not using. He said, ‘I’ll give you some urine right now.’ I took some urine, but just for him to know that I could.” Often within TF programs it was providers’ discretionary power that determined whether a consumer would be given a second or third chance to remain in the program after relapse.

Accommodating abstinence

Although the two approaches are often considered opposites, harm reduction was not seen as precluding abstinence. As one HF provider explains, “for many of our participants, a harm reduction can be a beneficial treatment strategy to use. And I guess, as we’ll soon know, many people that think of harm reduction as precluding the possibility, for many people, of abstinence…obviously that’s part of harm reduction.” In fact, HF providers attempted to make accommodations for consumers choosing abstinence,

while harm reduction encompasses obviously people who are using and who are currently sober, it doesn’t mean that it’s any easier for someone who is sober either through choice or through external mandate to listen to someone talk about active drug use. So I think that that’s one of the things that we’ve talked about and one of the ways that we’ve negotiated that is doing one on one counseling with people who are working towards sobriety or currently sober.

There were rare occasions (and consumers) for whom harm reduction didn’t work. Speaking of one of these exceptions, a HF provider opines, “he cannot-he doesn’t have the power, the harm reduction is not working. For dependent clients, harm reduction is not working. I can say that. Even though there’s so many times I’ve said that to other people and they’ve said “Oh you are judging people” or “that’s not right”, but this is the reality.” This belief that abstinence can be accommodated within a harm reduction framework speaks to flexibility in its implementation.

While harm reduction was y viewed as a sustainable strategy over time, it was also an effective engagement tool in the short run, “insofar as you have to hide your drug use, you might be inclined to hide this aspect of your life and that aspect of your life, and once people feel as though they have to hide certain things, it turns into a slippery slope.” This also allowed HF providers to address co-occurring disorders with several noting that while “theoretically you address both at the same time”, it also “depends on which is more overt, the mental health [illness] or the addiction.” In this context, harm reduction was seen as allowing HF providers to be continually supportive, “You know, one client may drink a quart of liquor a day and you know if he went down to a fifth, then that’s progress, not because you know, but it would also be good for me to try to help them see that there’s progress and encourage them.” Another HF provider explained, “one of the things about harm reduction that I love is that any incremental change is a huge deal.” Such gradual change does not preclude consumers from achieving and maintaining abstinence.

Discussion

Front-line providers working with individuals who have experienced homelessness, serious mental illness, and substance abuse encounter problems associated with an abstinence based approach – most notably that consumers disengage from services.²³ Our findings demonstrate that HF providers viewed harm reduction as a more effective way to work with consumers both in the short-run and the long-run. Their TF counterparts, working within an abstinence-only approach, acknowledged the challenges yet did not mention harm reduction as a possible (or necessarily desirable) approach.

It is noteworthy that, while harm reduction practices such as needle exchange presuppose drug use and derive their effectiveness from reducing its harmful consequences, the HF version of harm reduction entailed no such assumptions and was considered applicable to the pursuit of abstinence. That TF providers did not invoke similar flexibility is a reflection of the uniform and invariant policy under which they operated. Thus, the use of discretionary power to overlook mild indicators of substance use—present in the narratives of both HF and TF providers—is far more meaningful for the latter group.

Understanding harm reduction as existing on a continuum of practices tailored to individual needs and capabilities highlights the wide variation of situations that HF providers encounter within their day-to-day practice. It is not surprising that these providers encounter ambiguous situations in their work. Clinical supervision can help address this ambiguity, yet there are no clear guidelines of best practices within a harm reduction approach.^25,34 Similarly, little attention is given to the exercise of discretionary power by TF providers expected to adhere to abstinence-only policies.^8,35

Although it is acknowledged that “abstinence is the safest approach for those with substance use disorders,”²⁰ recovery is often described as a hard, fraught individual journey.³⁶ Harm reduction permits providers to engage with and support consumers who are in the midst of this process. Within the context of homeless services, researchers have argued that abstinence approaches elevate moral worthiness over clinical effectiveness in decisions over who gains access to housing, i.e., it is reserved for those who earn it through sobriety and compliance with treatment.³⁷ Endorsing harm reduction represents a shift in perspectives away from a conceptualization of addiction as a moral shortcoming, yet such thinking may continue to influence policy decisions and individual provider attitudes.¹⁰ Congregate ‘wet housing’ for individuals whose primary issue is abusing alcohol has brought to light these varying perspectives^7,38 yet ‘drug-tolerant’ housing continues to exist under a “don’t ask, don’t tell” policy, which makes pathways to recovery less clear for those living in supportive housing who abuse illicit substances.

Strengths and Limitations

Within this study the use of qualitative methods allowed for insight into provider views of harm reduction versus abstinence policies within homeless services for dually diagnosed adults. Nevertheless, the study relied exclusively on what providers say rather than observing what they do, and this may have led us to rely on incomplete accounts from some providers fearful of revealing too much and jeopardizing their position. Using observational methods could address this limitation in future research³³ by exploring in situ how providers react to incidents of substance use and deploy discretionary power if and when it is available.

This study is also limited by its geographical location insofar as harm reduction is still relatively rare in homeless services and the study sample of providers is not representative of the national population of providers in the field. At the same time, we note that several strategies for rigor were pursued including independent co-coding, peer debriefing and negative case analyses.

Implications for Behavioral Health

Harm reduction has become widely used beyond its origins in needle exchange and its incorporation into services for homeless adults with dual diagnoses has given providers a broader palette of options to deal with substance abusing clients. As described in this study, harm reduction can be more broadly interpreted to include the pursuit of abstinence; it is also compatible with consumer-driven, recovery-oriented services.

Providers and policymakers working within an abstinence-only approach may benefit from careful consideration of whether and how harm reduction can be adapted to working with consumers who are often challenging to engage. This would require expanded clinical supervision and further development of harm reduction expertise within supportive housing programs. And, as HF spreads (within the United States and internationally), it would behoove those in charge of implementation to ensure that abstinence is included as a goal as long as it is consonant with consumer choice.

The invariance of abstinence approaches opens the door to program dropout and to early (and sometimes unnecessary) referrals to expensive detox and rehab facilities. Recovery from substance abuse is rarely a linear all-or-nothing proposition and the addition of co-occurring serious mental illness and previous homelessness complicates this process considerably.²² As providers grapple with multiple problems, singular or narrow solutions hamper their ability to address these problems flexibly and in response to the client’s status and wishes.