Life Enhancing Alcohol-management Program (LEAP)

A Cluster-randomized Controlled Trial Testing the Effectiveness of the Life Enhancing Alcohol-management Program (LEAP) for Housing First Residents

People experiencing chronic homelessness comprise a small yet high-morbidity, high-cost subset of the larger homeless population and are disproportionately impacted by alcohol-related harm. Unfortunately, traditional abstinence-based treatment does not adequately reach or engage this population, and both firsthand (problems stemming from one's own alcohol use) and secondhand (problems stemming from others' alcohol use) alcohol-related harm persists even after housing attainment. There have therefore been calls for more flexible and client-centered approaches tailored to this population's needs. Housing First, which entails the provision of immediate, permanent, low-barrier, nonabstinence-based housing, is a response to this call. Research has shown that Housing First is associated with decreased alcohol use, alcohol-related harm, and publicly funded service utilization. Nonetheless, Housing First residents continue to experience both first- and secondhand alcohol-related harm. Thus, further community-based interventions are necessary. To this end, a pilot project was conducted in which researchers as well as Housing First residents, staff and management codeveloped, implemented, and initially evaluated the Life Enhancing Alcohol-management Program (LEAP). The LEAP entails low-barrier, community-level, house-wide resident programming-including leadership opportunities, activities, and pathways to recovery. At the 6-month follow up, LEAP participants reported significantly more engagement in meaningful activities than control participants (p < .001). Moreover, high levels of LEAP program engagement (>2 activities per month) predicted significant reductions in alcohol use and alcohol-related harm (ps < .01). To build on these promising findings, we propose a larger, cluster-randomized controlled trial of LEAP (N=160) as an innovative, community-based, and client-driven adjunct to Housing First. Analyses will test LEAP effectiveness in increasing engagement in meaningful activities, decreasing alcohol use, ameliorating both first- and secondhand alcohol-related harm, and improving quality of life. Engagement in meaningful activities will also be tested as a mediator of the LEAP effect on alcohol and quality-of-life outcomes. Finally, we will assess whether LEAP is associated with reduced costs stemming from participants' use of emergency health-care and criminal justice services.

Although they represent only 15% of the larger homeless population, chronically homeless people utilize substantially more services. This finding is understandable given that chronic homelessness is characterized by long or frequent episodes of homelessness paired with medical, psychiatric, and substance use disorders. Although epidemiologic data for chronically homeless people are scarce, studies conducted with the larger homeless population indicate that 80% of homeless people report current alcohol use, and 38% have severe alcohol use disorders (AUDs). This disproportionately problematic use results in alcohol-attributable mortality that is 6 to 10 times higher than in the general US population. Alcohol-related harm impacts the affected individual; however, it also has secondhand effects on the larger community as well, including caretaking burden, noise complaints, verbal altercations, and physical and sexual assault. Unfortunately, the most widely available approach-individual-level, abstinence-based treatment-does not effectively engage or treat this population. In our prior research, chronically homeless people with AUDs indicated they were not interested in abstinence-based approaches, having experienced a mean of 16 abstinence-based treatment episodes in their lifetimes. Further, our research has shown that improvements in alcohol outcomes in this population are associated with intrinsic motivation for change but not with abstinence-based treatment attendance. Instead, chronically homeless people with AUDs have indicated that they prefer community-based, harm-reduction approaches that support their own self-defined pathways to recovery. They are particularly interested in creative and socially engaging activities that bear personal meaning. As applied to alcohol use, harm reduction refers to a broad range of compassionate and pragmatic approaches applied at the individual, community, population or policy levels that aim to reduce alcohol-related harm and improve quality of life (QoL) for affected people and their communities. Housing First, also referred to as harm-reduction housing, is one such approach. Housing First entails the provision of immediate, permanent, low-barrier, supportive housing without preconditions such as alcohol abstinence or treatment attendance. Our own and others' research has shown Housing First to be associated with reductions in alcohol-related harm as well as publicly funded service utilization and cost (e.g., emergency medical services, emergency department, jail) for people experiencing chronic homelessness. Despite these positive outcomes, many Housing First residents still experience alcohol-related harm due to their own and their neighbors' alcohol use. There is thus a need for further interventions to address alcohol use in this setting. In response to this need, our research team used a community-based participatory research approach to work together with Housing First residents, management and staff to develop and initially test the effectiveness of a community-level intervention, the Life Enhancing Alcohol-management Program (LEAP) to improve alcohol and quality-of-life (QoL) outcomes for residents living in Housing First settings (K01AA021147; PI: Clifasefi). We first conducted needs assessments with residents, staff, and management and then formed a community advisory board to oversee the development, implementation, and evaluation of the LEAP. Together, we developed LEAP values, processes, and components. LEAP components for residents included leadership opportunities, LEAP activities, and pathways to recovery. Once developed, the LEAP was tested in a nonrandomized controlled pilot (N=116) with residents at 3 Housing First sites: 2 sites served as services-as-usual control sites and 1 received LEAP. Findings were promising: LEAP participants reported significantly more engagement in meaningful activities than control participants. This finding is important because engagement in meaningful activities is associated with improved medical, psychiatric, and substance-use outcomes. In within-subjects analyses, LEAP participants showed significant pre-post reductions in alcohol use and alcohol-related problems. These changes showed a dose-response effect based on participants' attendance at LEAP activities: high levels of LEAP programming engagement (>2 activities per month) predicted significant reductions in alcohol quantity and alcohol-related harm (ps < .01). To establish a more definitive evidence base for LEAP, we propose to test LEAP effectiveness using a 2-arm, 12-month, cluster-randomized controlled trial at 10 Housing First sites (N=160). Sites will be randomized to the services-as-usual control or LEAP conditions. Quantitative analyses will test LEAP effectiveness in improving participants' alcohol and QoL outcomes from baseline through the 3, 6-, and 12-month follow-up assessments. The specific aims are to test: LEAP effectiveness in reducing alcohol use and alcohol-related harm and improving QoL. Compared to controls, LEAP participants will report less alcohol use; less first- and secondhand alcohol-related harm; and improved health-related and general QoL over the follow-up. Group differences in participants' engagement in meaningful activities as well as its role as a mediator of changes on alcohol and QoL outcomes. Over time, it is expected that LEAP participants will report more engagement in meaningful activities than control participants. It is expected that greater engagement in meaningful activities will explain the hypothesized positive LEAP effect on outcomes. LEAP effects on costs associated with healthcare and criminal justice service utilization (i.e., emergency medical services, emergency department services, jail). Compared to control participants, LEAP participants will show greater decreases in service utilization costs over time.

10 Housing First sites will be optimally matched and block randomized on a rolling basis to either LEAP or services-as-usual control conditions.

The LEAP is a comprehensive, multidimensional, community-level intervention offered in Housing First settings that was developed using a CBPR framework. The LEAP principles, process, and components were conceptualized and operationalized by the CAB in the pilot study, and are outlined in a recently published articles in the American Psychologist and Journal of Community Psychology. Briefly, LEAP principles include recognizing and respecting community as a key unit of identity versus solely as individuals in a traditional research hierarchy; embracing cultural humility; practicing pragmatism and compassion; aiming to build equitable research partnerships with communities; acknowledging and promoting community strengths; and supporting relevant, sustainable, positive and community-led change.

The Alcohol and Substance-use Frequency Assessment questions were adapted from the Addiction Severity Index (McLellan, Kushner, Metzger, & Peters, 1992) and will be used to assess frequency of use of alcohol and other substance use

The Alcohol Quantity and Use Assessment (AQUA) was created by the research team for previous studies with a similar population and will be used to record the quantity of alcohol consumed on participants' heaviest, typical, and lightest drinking days in the past month. Scores are expressed in number of standard drinks with higher numbers indicating heavier drinking.

The Short Inventory of Problems (SIP-2R) is a psychometrically reliable and valid, 15-item, Likert-scale questionnaire that measures experience with social, occupational, and psychological harms related to one's own alcohol use. Scores range from 0 to 45 with higher scores indicating more severe alcohol-related harm.

Secondhand effects of alcohol will be measured using a newly developed, psychometrically sound questionnaire, augmented by items from prior validated surveys. The 8, dichotomous items include feeling unsafe; being harassed, bothered, called names or insulted; being pushed, hit or assaulted; having one's living space disturbed; having one's property stolen or damaged; experiencing unwanted sexual advances; having to take care of another resident who drank too much; and having sleep or other activities interrupted. The summary score will be used to reflect overall experience of secondhand alcohol-related harm as an outcome variable. Scores range from 0 to 30 with higher scores indicating more severe secondhand alcohol-related harm.

Presence of EtG, a direct metabolite of alcohol (cut-off > 500 ng/ml), which is indicative of heavy alcohol use within the past 3 days, will be used as a primary outcome.

We will collect administrative data from local county and state databases reflecting participants' utilization of emergency medical services, hospital, and jail and associated cost estimates. These data will be summed across types to create overall cost outcomes, which will serve as the primary outcome for specific aim 3.

The EuroQoL-5D-5L self-report measure yields 5 dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Additionally, the EQ-Visual Analog Scale (EQ-VAS) is a single item representing health-related quality of life via participants' general assessments of their current health, where 0 = the worst health imaginable and 100 = the best health imaginable.

The Quality of Life Scale (QOLS) elicits subjective assessments of general QoL across various domains (i.e., material and physical well-being; relationships; social, community, and civic activities; personal development and fulfilment; and recreation). Scores range from 16 to 112 where a higher score indicates higher quality of life. The QOLS is reliable and valid across various patient groups and cultures and is conceptually distinct from measures of health-related QoL, such as the EQ-5D-5L.

Attendance at other treatment and mutual-help groups will be assessed at each time point using single, dichotomous (y/n) items on the Personal Information Questionnaire (PIQ). These time-varying variables will be used as a secondary outcome and as a covariate of the treatment effects, as necessary.

The Meaningful Activity Participation Assessment (MAPA) is a 28-item, psychometrically validated tool designed to measure level of engagement in general life activities that bring meaning to people's lives. Respondents are presented with a list of various activities they may encounter in their day-to-day lives (e.g., socializing, writing, physical exercise, reading, prayer/meditation, community organization, computer use). Each activity is then rated on 2, 4-point Likert scales assessing the frequency with which they engage in that activity and the level of meaningfulness ascribed to each. The 2 scores for each item are combined multiplicatively, and a summary score, which reflects the overall level of engagement in meaningful activities, is formed. The score range from 0 to 672 with a higher score indicating greater participation in meaningful activities. The summary score will be used both as an outcome and as a potential mediator of the LEAP effect on alcohol and QoL outcomes.

Site-specific activity participation will be measured using activity sign-up sheets across all 10 participating Housing First sites. Site-specific activity sign-up sheets will document participants' engagement in all site-specific activities. Level of engagement will be summed to form simple counts of activities in which each participant was involved over their 12-month assessment course. Activities recorded will include but will not be limited to LEAP activities. Higher scores indicate higher engagement in site-specific activities. The summary score will be used both as an outcome and as a potential mediator of the LEAP effect on alcohol and QoL outcomes.

The ARM-R is a self-report measure that focuses on a relational understanding of well-being, framing resilience as an individual or community's ability to navigate to and negotiate for resources within their ecologies, while also acknowledging that it is up to the social-ecology to provide necessary resources in a way that individuals and communities value. The ARM-R was adapted from the Child and Youth Resilience Measure (CYRM), which was developed as part of the International Resilience Project (IRP) at the Resilience Research Centre (RRC), involving 14 communities in 11 countries. The ARM-R consists of 17 items and can be scored on 3 or 5 point Likert scales and 3 qualitative prompts.

Inclusion Criteria: At-risk drinking as established by the AUDIT-C at screening (cut-off scores: men > = 3, women > = 2 Having a history of chronic homelessness according to the widely accepted federal definition (i.e., having a psychiatric, medical, or substance use disorder paired with being homeless for a year or more or having 4 or more episodes of homelessness in the past 3 years) Being a current DESC client living in 1 of 10 participating Housing First sites Exclusion Criteria: Refusal or inability to consent to participation in research Constituting a risk to the safety and security of other clients or staff

The proposed research will involve people recruited from prominent Housing First programs serving chronically homeless people who are often multiply affected by psychiatric, medical and substance-use disorders. The proposed sample will be recruited from a highly vulnerable and marginalized population in a tight-knit urban community. Furthermore, the partnering agency is well-known for its housing approach, and regularly conducts tours to help inform other housing providers, policy makers, and other public stakeholders about this model to promote replicability. Thus, even with the removal of all identifiers, we believe that it could become difficult to fully protect the identities of participants and their data (including e-cigarette use and smoking which are not currently allowed in housing). For these reasons, we do not have immediate plans to share the data collected in the context of the proposed study.