Aversion to Injection Limits Acceptability of Extended-Release Naltrexone Among Homeless, Alcohol-Dependent Patients

Abstract

Background

Ending homelessness is a major priority of the Department of Veteran Affairs (VA), and alcohol use can be a barrier to stable housing. Clinical trials suggest that depot extended-release naltrexone (XR-NTX) is efficacious in reducing alcohol consumption among alcohol-dependent subjects.

Methods

An open-label, randomized pilot study sought to examine the feasibility and effectiveness of XR-NTX versus oral naltrexone to improve alcohol consumption and housing stability among homeless, alcohol-dependent veterans at the Providence Veteran Affairs Medical Center.

Results

Of 215 potential candidates approached over a 16-month recruitment period, only 15 agreed to consider study entry and 7 were randomized. The primary reasons given for refusal were not wanting an injection; fear of needles; and not wanting to change drinking habits. Only 1 participant in the XR-NTX group returned after the first injection. Three participants in the oral naltrexone group attended all 7 visits and had good outcomes.

Conclusions

Although XR-NTX has demonstrated efficacy in reducing heavy drinking, limited acceptance of the injection might reduce its effectiveness among homeless, alcohol-dependent patients.

INTRODUCTION

Veterans are at disproportionate risk for homelessness (1), and up to 70% of homeless veterans meet criteria for alcohol dependence (2). Ending homelessness is a priority for the Department of Veteran Affairs (VA), but alcohol use can be a barrier to housing. Clinical trials suggest that depot extended-release naltrexone (XR-NTX) is efficacious in reducing alcohol consumption among alcohol-dependent subjects (3), but XR-NTX is expensive and has restricted formulary availability in VA hospitals. We fielded a small, randomized study to evaluate the feasibility and pilot effectiveness of XR-NTX versus oral naltrexone over 4 months among alcohol-dependent, homeless veterans. We hypothesized that XR-NTX would be superior to oral naltrexone in reducing alcohol consumption.

METHODS

This open-label, 24-week pilot study at the Providence Veteran Affairs Medical Center (VAMC) attempted to randomize 20 homeless, alcohol-dependent veterans to XR-NTX versus oral naltrexone to compare effects on alcohol consumption, housing stability, and other outcomes over 6 months. The goal was to test the feasibility and to estimate effect sizes for the design of a future comparative effectiveness trial. The institutional review board (IRB) at the Providence Veteran Affairs Medical Center approved the protocol.

The study population was homeless, alcohol-dependent veterans who had 12 hours to 12 months of sobriety. Inclusion criteria were homelessness (4); past-year alcohol abuse or dependence by DSM-IV (Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition) criteria; last reported drink 12 hours to 12 months prior; age 18 to 64 years; eligible for VA services; and no contraindications to naltrexone. Potential candidates were identified by the Homeless Coordinator or homeless clinic staff, and then approached by an experienced research assistant who explained study procedures, screened for eligibility, and obtained consent. Recruitment techniques include those that we and others have used successfully with homeless populations (5, 6), including facilitating trust with the research staff through introductions by the homeless services staff; careful explanation of the study’s intent and procedures; appealing to altruism; targeted outreach at community sites; and offering payment for interviews. Participants could receive up to $215 in gift cards (the IRB would not permit cash). Bus passes were provided. In response to inadequate recruitment, we expanded community recruitment sites, abbreviated interviews, and expanded inclusion criteria to reach those who moved from transitional to permanent housing 3 months previously.

Consenting participants were randomized to XR-NTX, 380 mg via intramuscular injection every 4 weeks or oral naltrexone 50 mg daily over a 16-week treatment period (i.e., at baseline, then weeks 4, 8, 12, and 16). Block randomization balanced housing status (transitionally housed versus seeking transitional housing) and last alcohol use (more or less than 7 days prior).

Participants saw a study nurse or physician every 4 weeks during the 16-week treatment period; medical management adherence counseling was provided at each visit (7). All subjects were referred for existing VA substance abuse treatment; attendance was not required. Other referrals were made per standard practice.

A research assistant performed assessment interviews at baseline, weeks 4, 8, 12, and 16 during the treatment period, and at 2 follow-up visits (weeks 20 and 24). Subjects were paid $25 for phlebotomy at baseline and weeks 4, 16, and 24; $15 for each research assessment during the treatment period; and $20 for each follow-up assessment at weeks 20 and 24. During each assessment, participants reported alcohol consumption and housing status in the prior 4 weeks using a Timeline Followback Calendar (8).

RESULTS

Over a 16-month recruitment period, 215 potential candidates were approached in the community and the homeless clinic at the Providence VAMC, of whom only 15 agreed to be screened after an explanation of study procedures. The primary reasons for refusal were not wanting to be injected with a medication; fear of needles; and not wanting to change drinking habits. Of the 15 screened, 5 were excluded (1 did not meet criteria for homelessness, 1 did not meet criteria for alcohol abuse or dependence, 1 had untreated delusional thinking, and 2 had opioid-requiring pain); 3 eligible participants did not appear for their baseline interview and could not be found; only 7 were randomized.

Of the 7 randomized, 5 received oral naltrexone and 3 were randomized to XR-NTX. Mean age was 50. All were male. Three lived in transitional housing, 2 in substance abuse treatment facilities, and 2 were unsheltered. Five participants were Caucasian, 1 African American, and 1 both African American and Caucasian.

Of those in the XR-NTX group, only 1 participant returned after the first injection and he remained sober through week 16 (Table 1). Only 3 participants attended all 7 visits; all 3 were in the oral naltrexone group—2 of the 3 were abstinent and the third maintained controlled drinking (Table 1).

TABLE 1.

Self-Reported Alcohol Consumption During the 4 Weeks Prior to Each Research Assessment, in Standard Drinks

		Treatment period				Follow-up
Assignment	Baseline	Week 4	Week 8	Week 12	Week 16	Week 20	Week 24
XR-NTX	0	0	0	0	0	Lost to follow-up	—
XR-NTX	0	Lost to follow-up	—	—	—	—	—
XR-NTX	85	Lost to follow-up	—	—	—	—	—
Oral NTX	0	0	0	0	0	0	0
Oral NTX	12	3	4	12	0	2	4
Oral NTX	0	0	10	0	0	0	0
Oral NTX	495	891	374	Lost to follow-up	—	—	—

DISCUSSION

Despite our extensive experience with research in this population and attempts to facilitate enrollment (5, 6), only 15 of 215 alcohol-dependent, homeless veterans would enter a study that involved an intramuscular injection. Of those randomized to the XR-NTX, only 1 of 3 participants returned for a second injection. The patients’ stated aversion to injection appears to have contributed to the low overall rate of recruitment and poor retention in the XR-NTX group. Oral naltrexone was more accepted, but a small sample prevents firm conclusions.

The limited acceptability of XR-NTX here contrasts with a study in a detoxification unit in Portland, Maine, where 60 alcohol-dependent, homeless patients were administered XR-NTX on discharge and referred for repeat injections monthly at a Healthcare for the Homeless clinic (9). In that study it was reportedly “rare for a patient to decline” participation, but refusal rates were not provided. Greater acceptability possibly resulted from greater motivation among those recruited from a detoxification unit, but the typical patient still received only 2 to 3 months of injections.

In the current study, almost everyone expressed an aversion to injection. Fear of injections is common, and has been found among 22% to 30% of adults in clinical settings (10–12). Nonetheless, this aversion here might be a manifestation of ambivalence or distrust. Given ambivalence toward recovery, the injection might have provided a pretext for wavering motivation among alcohol-dependent persons. Alternatively, distrust could underlie the aversion. Anecdotally, several participants expressed concern that “the government” might inject them with something. Distrust might also have been with the researchers, although this research team has been very successful recruiting from this population and in these settings for protocols that did not involve an injection (5, 6). Most of the screening refusals occurred after the Homeless Coordinator, a social worker with over 10 years of experience serving the homeless population, explained that the protocol included an injection.

Regardless of the true reasons underlying the aversion to injection, the need for an injection might be sufficient to negate the theoretical adherence advantage of monthly XR-NTX over the daily dosing of oral naltrexone. In trying to field this study, we have come to question the assumption underlying our study and the manufacturer’s marketing that a form of naltrexone injected monthly will be more effective than an oral form dosed daily because the monthly injection will increase adherence. The limited acceptability—for whatever reason—of an injection limits the effectiveness of a formulation that was efficacious for alcohol dependence in controlled trials with a selected population that entered that study precisely because they had no problem receiving an injection. It appears that many alcohol-dependent patients will not accept an injection at all, and many who accept an initial injection will not adhere past 2 to 3 months. To realize fully the potential adherence benefit of the monthly injectable form of naltrexone, future effectiveness studies and efforts to implement it must include novel strategies to recruit and engage vulnerable, alcohol-dependent populations and to overcome the aversion to injection.