Attitudes toward opioid use disorder pharmacotherapy among recovery community center attendees

Abstract

Background:

Despite their proven efficacy, medications for opioid use disorder (MOUD) are underutilized. Negative beliefs and attitudes toward MOUD are quite common, yet rapidly expanding recovery community centers (RCCs) may offer a promising venue for fostering MOUD support as they operate under the maxim, “many pathways [to recovery], all should be celebrated” and are utilized mainly by those with opioid use disorder. The current study provides a first look at MOUD attitudes and their correlates in RCC attendees.

Methods:

The study conducted a cross-sectional survey (N=320) of recovering adults attending 31 RCCs across New England, assessing demographic, treatment, and recovery-relevant factors, as well attitudes (positive vs. negative) toward the use of agonist and antagonist MOUD. The study used frequencies and confidence intervals to obtain prevalence estimates for positive and negative attitudes toward agonist and antagonist MOUD, and to examine differences between them. Spearman correlations identified correlates of MOUD attitudes (at p < 0.10), and significant correlates were assessed for unique contributions via multivariable logistic regression.

Results:

Positive attitudes were common and more prevalent than negative attitudes for both agonist (positive: 71.4 [66.1, 76.3]%; negative: 28.6 [23.7, 33.9]%) and antagonist (positive: 76.5 [71.4, 81.1]%; negative: 23.5 [18.9, 28.6]%) MOUD, which did not differ. The study identified several correlates of MOUD attitudes at the p < 0.10 level, but only four variables emerged as unique predictors controlling for other correlates. Lifetime history of agonist MOUD treatment was uniquely associated with positive agonist attitudes (p=0.008), whereas greater social support for recovery was associated with positive antagonist attitudes (p=0.007). Lower educational attainment was uniquely associated with negative antagonist attitudes (p=0.005), and a greater degree of spirituality was related to negative attitudes toward both agonists (p=0.005) and antagonists (p=0.01).

Conclusions:

Findings reveal very high rates of positive MOUD attitudes among RCC participants, highlighting the potential for this growing tier of recovery support to foster acceptance and peer support for medication-facilitated recovery pathways. Correlates of attitudes further reveal opportunities for facilitating MOUD acceptance within and beyond the RCC network.

1. Introduction

In the United States, an estimated 10 million individuals misused opioids and about 2 million suffered from an opioid use disorder (OUD) within the past year (Substance Abuse and Mental Health Services Administration, 2019). Rising rates of opioid misuse and related disorders over the past two decades were paralleled by a dramatic increase in opioid-involved overdose deaths (i.e. ~47,000 deaths per year; Wilson et al., 2020). Efforts to address the opioid epidemic and the individual and societal problems that accompany it have in part focused on expanding the availability and use of evidence-based medications for OUD (MOUD).

Currently, there are three medications that the Food and Drug Administration has approved for the treatment of opioid use disorder, including two μ-opioid receptor agonists (i.e., methadone [full agonist] & buprenorphine [partial agonist]) and one μ-opioid receptor antagonist (i.e., naltrexone). These first-line, lifesaving medications are proven to help prevent opioid relapse and opioid-related mortality, increase treatment retention, and reduce drug-related risky behaviors (American Society of Addiction Medicine, 2020; Schuckit, 2016; National Institute on Drug Abuse, 2018; Kosten & George, 2002; Kampman & Jarvis, 2015; Alderks, 2017). Data from the National Survey of Substance Abuse Treatment Services (N-SSATS) suggest substantial increases in the availability and utilization of all three MOUDs in public and private opioid treatment programs over the last several years (Aledrks, 2017). Despite these increasing trends, evidence also suggests a persistent unmet need for MOUD treatment, with only about 20% of individuals with OUD receiving specialty addiction treatment (Substance Abuse and Mental Health Services Administration, 2019; Saloner & Karthikeyan, 2015).

Research has identified several barriers to MOUD receipt, including institutional, provider, policy, and financial barriers. At the individual-level, many potential barriers have been acknowledged (e.g., not recognizing need for treatment, lack of patient demand), but they are seldom systematically investigated (Madras et al., 2020). With efforts underway to enhance access and use of MOUD treatment, researchers must gain a better understanding of MOUD from the perspective of individuals with lived experiences of addiction and recovery. This cohort constitutes an important subset of stakeholders in the dissemination of MOUD, as they are likely to have an impact on MOUD provision and utilization. Therefore, one of the individual-level barriers that requires greater attention is MOUD attitudes and receptivity in the recovery community.

MOUD attitudes are presumed to be predominantly negative among recovering individuals, in general, with clinical commentaries and qualitative studies of providers’ perceptions highlighting the potential role of MOUD stigma within mutual-help organizations and psychosocial treatment programs (e.g., Wakeman & Rich, 2018; Richard et al., 2020). These skeptical MOUD attitudes are hypothesized to stem in part from agonist MOUD attitudes, which, in contrast with antagonists, are often colloquially misinterpreted as “replacing one addiction for another” (Woods & Joseph, 2018). Yet these reports are largely based on clinical opinion, rather than rigorous scientific evidence. Recent investigations, however, have begun to empirically characterize MOUD attitudes among recovering individuals, to determine whether these assumptions are true, and to identify patient characteristics impacting MOUD attitudinal preferences. A recent national survey, for example, revealed that only about 20% of individuals who resolved a significant substance use problem reported positive attitudes toward MOUD (Bergman et al., 2019). Negative MOUD attitudes among the general population of recovering individuals have the potential to adversely impact attitudes and service use among newly recovering peers (Tsai et al., 2019). Therefore, characterization of MOUD attitudes and the factors influencing them is essential for advancing our understanding of OUD treatment barriers and for further expanding MOUD treatment receipt.

Recovery community centers (RCCs) may offer a promising venue for fostering MOUD support. RCCs are part of an expanding infrastructure of recovery support services that implement professional and peer-led activities, which cater to the SUD community and those with OUD in particular (Kelly et al., 2020), and aim to reduce psychosocial barriers to long-term recovery (White et al., 2012). RCCs are founded on the principle of continuing care, with an emphasis on accrual of recovery capital (i.e., personal, social, environmental, and cultural resources that can be used to support recovery; Cloud & Granfield, 2008). These novel and expanding recovery support hubs are especially inclusive (Valentine, 2011) as they do not follow any particular recovery model (e.g., 12-step) and they operate under the maxim that there are “many pathways [to recovery], all should be celebrated”, including those that involve medication treatment. They also are frequented most often by those with OUD (Kelly et al., 2020). The inclusive nature of RCCs might, therefore, offer an environment that is supportive and accepting of individuals with OUD using medications. However, little is known about this important community-based recovery resource and MOUD attitudes have yet to be investigated in the context of RCCs.

The current study aimed to address this knowledge gap by exploring agonist and antagonist MOUD attitudes and their correlates among RCC attendees, attending almost three dozen centers in the northeastern United States. We had two primary research questions pertaining to the current work: (1) What is the prevalence of positive and negative attitudes toward agonist and antagonist MOUD? (2a) What are the demographic, clinical, and recovery-related correlates of MOUD attitudes, and (2b) which correlates are unique predictors of attitudes, controlling for other relevant correlates?

2. Materials and methods

2.1. Sample and procedure

The study recruited three-hundred and thirty-six individuals between February 2016 and October 2017, from 31 RCCs across the New England region of the United States (see Kelly et al., 2020 for detailed recruitment methods). Eligible participants were adult (aged 18+ years) RCC attendees, currently seeking or in recovery from an alcohol or other drug problem. Participants were recruited via fliers and in-person announcements at their RCCs and study staff invited interested individuals to complete an online survey via REDCap (Harris et al., 2009). At the start of survey administration, a pre-screening questionnaire assessed study eligibility, and we obtained electronic informed consent. The study staff offered participants a $10.00 gift card for survey completion. Of the 336 individuals who completed the survey and were deemed valid responders, 320 responded to questions regarding MOUD attitudes (agonists: n=318; antagonists: n=315), and we include them in the current report. The majority of the sample responded to both agonist and antagonist survey items (n=313); five individuals completed agonist questions only, and two completed antagonist questions only. The Partners HealthCare Institutional Review Board reviewed and approved all study procedures.

2.2. Measures

2.2.1. Medication attitudes

Participants completed survey items regarding their attitudes toward various pharmacotherapies. Given limited research to date on medication attitudes among recovering individuals, the study took all items directly from, and operationalized according to, a large-scale national survey of adults in recovery from alcohol and other drug problems (i.e. The National Recovery Study (NRS); Kelly et al., 2017; Bergman et al., 2020), thereby providing context with respect to the broader population of recovering individuals in the United States. Questions regarding medication attitudes in the NRS were based on Rychtarik and colleagues’ (2000) survey of twelve-step attendees’ attitudes toward anti-relapse/craving medications for alcohol use problems. Using a Likert scale (1=strongly disagree, 2=disagree, 3=somewhat disagree, 4=somewhat agree, 5=agree, 6=strongly agree), participants rated their level of agreement with two statements regarding MOUD: “It is a good idea for someone with an OPIOID problem to take…” (1 – agonist attitudes) “… a substitute opioid medication like Suboxone or methadone to help them stop using” and (2 – antagonist attitudes) “… an opioid blocking medication like naltrexone/Vivitrol to help them stop using”. To enhance meaningful data interpretation, we subsequently dichotomized responses, with positive attitude defined as any level of agreement (strongly agree, agree, or somewhat agree) and negative attitude defined as any disagreement (strongly disagree, disagree, somewhat disagree).

To provide comparison and context for MOUD-related attitudes, the study also evaluated attitudes for medications used to treat alcohol problems (MAUD) and emotional problems (Mmood). Using the same Likert scale, participants rated their agreement with the following statements: “It is a good idea for someone with an ALCOHOL problem to take a medication to help them stop drinking” and “It is a good idea for someone with an EMOTIONAL problem to take a medication to help”. Consistent with MOUD attitudes, the study dichotomized MAUD and Mmood Likert scale responses.

2.2.2. Demographics

Study staff asked RCC attendees to provide demographic information, including age, gender, race, ethnicity, educational attainment (BA or higher; high school or less; some college or other degree), employment status (unemployed; part time [including irregular work]; full time [35+ hrs/week]) over the past-90 days, and past-year household annual income (< $10,000; $10,000–$49,000; ≥ $50,000). The study also asked participants to indicate the extent to which they considered themselves to be a religious person and a spiritual person (via separate Likert scales; from 1=not at all religious/spiritual to 4=very religious/spiritual; Idler et al., 2003).

2.2.3. Criminal justice involvement

The study assessed current legal involvement with a single-item question from the TCU (Institute of Behavioral Research, 2007), indicating current legal status as “none”, “on probation only”, “on parole only”, “on probation and parole”, “awaiting charge, trial or sentence”, “outstanding warrant”, “case pending”, or “other” (ultimately dichotomized, with “none” reflecting “no current legal involvement” and all other response categories reflecting “current legal involvement”). Participants also indicated (yes/no) whether they had ever been arrested (Institute of Behavioral Research, 2008), participated in drug court, and recommended or required by legal authorities to attend substance use disorder treatment (Institute of Behavioral Research, 2001). The study also assessed number of lifetime arrests (coded “0” for those who had no lifetime history of arrest; Institute of Behavioral Research, 2008).

2.2.4. Substance use histories & recovery-related measures

Study staff presented participants with a list of 15 substances (i.e., alcohol, cannabis, cocaine, heroin, unprescribed methadone, unprescribed buprenorphine, unprescribed other opioids, hallucinogens, synthetic marijuana, amphetamine, methamphetamine, benzodiazepines, barbiturates, inhalants, and steroids), and asked participants to indicate whether they had ever regularly used (i.e., at least once per week) and/or were currently using each substance (Dennis et al., 2003). Participants also identified their primary substance (i.e., drug of choice). The study re-grouped substance categories into five drug classes: (1) alcohol, (2) cannabis, (3) cocaine, (4) opioids (i.e. heroin, unprescribed methadone/buprenorphine/other opioids), and (5) other (i.e. all other drug categories). Participants could endorse multiple regularly used substances, but only one primary substance. The study subsequently used responses to questions regarding regular substance use and current use to assess the total number of substances used regularly across the lifetime and current abstinence (yes/no), respectively.

Study staff showed participants three definitions and asked them to indicate which one best fit their definition of recovery; we recoded the responses as (1) total abstinence (i.e., “abstinence from all drugs/alcohol”) and (2) moderate/nonproblematic use (i.e., original response was “abstinence from only those drugs/alcohol with which I had a problem” or “nonproblematic/moderate use of drugs/alcohol, including those with which I had a problem”). The study assessed recovery identity via a single-item measure with dichotomous (yes/no) response options (i.e. “Would you describe yourself as being in recovery?”). Individuals who identified as being in recovery also indicated their length of time in recovery from addiction.

2.2.5. History of treatment & recovery service utilization

We grouped lifetime use of treatment/recovery services into 5 dichotomous categories (yes/no): formal treatment (i.e., drug/alcohol detoxification, outpatient addiction treatment, inpatient/residential treatment); recovery support services (RSS; i.e., sober living, recovery high schools, college recovery programs/communities, faith-based recovery services, state/local recovery community organizations); mutual-help organizations (MHOs; i.e., Alcoholics Anonymous (AA), Narcotics Anonymous (NA), other 12-step-based services [e.g., Marijuana Anonymous (MA), Cocaine Anonymous (CA), Crystal Methamphetamine Anonymous (CMA)], SMART Recovery, LifeRing Secular Recovery, Moderation Management, Celebrate Recovery, Women for Sobriety, Secular Organizations for Sobriety, “other”); anti-relapse/craving pharmacotherapy for alcohol problems (MAUD; i.e., acamprosate, nalmefene, topiramate, disulfiram, baclofen, oral naltrexone); and anti-relapse/craving pharmacotherapy for opioid problems, (MOUD; agonists: methadone, levomethadyl acetate, buprenorphine, buprenorphine-naloxone; antagonists: oral naltrexone, long-acting injectable naltrexone). The study used these categories to assess lifetime use of formal treatment services and RSS, lifetime receipt of MAUD and agonist/antagonist MOUD, as well as lifetime prevalence of regular (≥ 1 time per week) MHO attendance. Additional questions addressed the total number of times individuals participated in formal treatment, MHO attendance within the past-90 days (yes/no), number of MHO meetings attended in the past-90 days, and lifetime treatment in an emergency department for alcohol/drug problems and other mental health problems (yes/no).

2.2.6. Psychological well-being

To assess lifetime history of psychiatric diagnoses, the study staff presented participants with a list of 16 non-SUD psychiatric disorders and asked them to indicate with which they had ever been diagnosed (Dennis et al., 2002). The study measured social support for recovery with a modified version of the 9-item social support subscale from the Texas Christian University Client Evaluation of Self and Treatment (CEST-SS; current sample α = 0.92; Institute of Behavioral Research, 2007; Kelly et al., 2020), with higher scores indicating greater social support for recovery. We evaluated psychological distress with the Kessler-6; participants rated 6 symptoms on their frequency of occurrence (0 = “none of the time” to 4 = “all of the time”) over the past 30 days (sample α = 0.90; Kessler et al., 2003). The study measured recovery capital with the Brief Assessment of Recovery Capital (BARC-10; Vilsaint et al., 2017), a 10-item self-report inventory for which participants reported their extent of agreement (1 = “strongly disagree” to 6 = “strongly agree”) with various statements concerning their recovery, environmental support, and well-being (e.g., “I regard my life as challenging and fulfilling without the need for using drugs or alcohol”; sample α = 0.95). The study assessed quality of life with the EUROHIS-QOL (Schmidt et al., 2006), which required participants to rate 8-items on a one to five Likert-scale (e.g., 1 = very dissatisfied; 5 = very satisfied; sample α = 0.89). We evaluated self-esteem with a single-item measure (i.e. “I have high self-esteem”, rated as 1 = “not very true” to 5 = “very true”; Robins et al., 2001).

2.2.7. RCC experience

The study asked participants about the length of time (in years) since they started attending their respective RCC, the average amount of time (in hours) spent at the RCC during a typical visit, and the number of days that they attended their RCC in the past 90 days (examined as percent days). Survey items also asked whether individuals had used various RCC-provided services, including (1) mental and physical health services (i.e., mental health support, health insurance education, health/exercise/nutrition programs); (2) social services (i.e., recreational/social activities, opportunity to volunteer / “give back” to the center, employment assistance, education assistance, housing assistance, legal assistance, financial services, childcare services, family support services, recovery advocacy outreach and opportunities, basic needs assistance, expressive arts services, other social services/basic needs assistance); (3) addiction/recovery (nonpharmacotherapy) services (i.e., recovery coaching, peer-facilitated recovery support groups, mutual-help & “all recovery” groups, smoking cessation support, recovery advocacy outreach and opportunities, other addiction resources), and (4) medication for addiction treatment support services (e.g., Pathway Guide, MARS group).

2.3. Statistical analyses

Study staff calculated descriptive statistics (means, standard deviations, percentages, sample sizes) to characterize the RCC attendee sample. We used frequencies to obtain the prevalence of positive and negative attitudes toward agonist and antagonist MOUD, MAUD, and Mmood. Study staff calculated confidence intervals (95% CI) to inform potential differences between frequency estimates, with nonoverlapping confidence intervals indicating a significant difference (Schenker & Gentleman, 2001). We conducted separate Spearman correlational analyses to explore potential correlates of agonist and antagonist MOUD attitudes; we considered several demographic, criminal justice, psychological well-being, RCC experience, clinical- and recovery-related measures. Given the exploratory nature of the current investigation, the study initially identified significant bivariate Spearman correlations at a significance level of p < 0.10. We subsequently analyzed significant correlates in separate multivariable logistic regression models for agonist MOUD and antagonist MOUD (reference = negative attitude) to test for unique correlates (i.e. predictors) of positive and negative attitudes, controlling for all other factors in the model. In multivariable regression models, the study defined significance as p < 0.05. Odds ratios greater than one reflect greater likelihood of a positive attitude, whereas odds ratios less than one reflect greater likelihood of a negative attitude. Between 0% and 5% of data were missing for any given variable of interest. To ensure missing data did not significantly impact outcomes, the study re-analyzed multivariable logistic regression analyses with “missing” as a modeled category for categorical/dichotomous variables, and assessed missing continuous variables via multiple imputation (25 imputed datasets each; i.e. SAS Proc MI & Proc MIANALYZE); outcomes were unaffected and mirrored results reported herein. We performed all statistical analyses using SAS Version 9.4.

3. Results

3.1. Demographics

Table 1 presents descriptive statistics, including demographics, criminal justice history, psychological well-being, and recovery-related measures. The majority of participants were White (80%), with an annual household income of less than $50,000 (92%), and an average age of ~41 years. Opioids (39%) and alcohol (31%) were the most commonly reported primary substances; most participants (75%) had used more than one substance regularly throughout their lifetime. On average, participants had been attending their RCC for ~2.5 years, and visited the center on 46 of the past 90 days.

Table 1.

Descriptive Statistics (N = 320)

Variable		M / %	SD / n
Age		40.70	12.29
Gender (% female)		51%	159
Race (% White)		80%	254
Ethnicity (% Hispanic)		11%	35
Education
	BA or higher	14%	44
	Some college or other degree	36%	113
	High school or less	50%	160
Employment (past 90 days)
	Unemployed	48%	146
	Part-time (including irregular work)	30%	93
	Full-time (35+ hours/week)	22%	66
Income (total household past year)
	Less than $10,000	48%	147
	$10,000 to $49,999	44%	135
	$50,000 or more	8%	26
Religiosity (1=min; 4=max)		2.14	1.06
Spirituality (1=min; 4=max)		2.87	1.12
Current legal involvement (% yes)		24%	76
Lifetime recommendation by legal authorities to attend addiction Tx (% yes)		39%	125
Lifetime requirement by legal authorities to attend addiction Tx (% yes)		37%	117
Lifetime drug court participation (% yes)		9%	29
Lifetime history of arrest (% yes)		68%	216
Lifetime total number of arrests		5.13	11.66
Lifetime number of substances used regularly (≥ 1 time per week)
	None	5%	18
	1 substance	20%	63
	2 substances	21%	67
	3+ substances	54%	172
Lifetime regular alcohol use (% yes)a		66%	210
Lifetime regular cannabis use (% yes)a		57%	183
Lifetime regular cocaine use (% yes)a		50%	159
Lifetime regular opioid use (% yes)a		52%	166
Lifetime regular use of ‘other’ substances (% yes)a		37%	119
Primary substance
	Opioid	39%	110
	Alcohol	31%	87
	Cocaine	15%	43
	Cannabis	9%	25
	Other	6%	18
Recovery identity (% yes)		97%	306
Definition of recovery
	Total abstinence	85%	268
	Moderate / non-problematic use	15%	46
Length of time in recovery (in years)		4.17	7.18
	Less than a year	33%	100
	1-5 years	46%	139
	5+ years	21%	63
Current Abstinence (% no)		17%	54
Lifetime use of any formal Tx service (% yes)		57%	182
	Lifetime use of outpatient SUD Tx (% yes)a	38%	121
	Lifetime use of detoxification services (% yes)a	27%	87
	Lifetime use of residential SUD Tx (% yes)a	44%	142
Total number of times received any formal Tx		5.06	10.99
Lifetime receipt of MAUD (% yes)		11%	36
Lifetime receipt of agonist MOUD (% yes)		28%	90
Lifetime receipt of antagonist MOUD (% yes)		4%	14
Lifetime SUD treatment in Emergency Department (% yes)		43%	136
Lifetime mental health treatment in Emergency Department (% yes)		40%	127
Lifetime use of recovery support services (% yes)		51%	164
Lifetime history of any regular MHO attendance (1+ per week; % yes)		81%	258
Any MHO attendance past 90 days (% yes)		83%	267
Number of MHO meetings attended in the past 90 days
	Alcoholics Anonymous (AA)	32.56	49.14
	Narcotics Anonymous (NA)	18.77	35.40
	SMART Recovery	1.17	6.78
	Celebrate Recovery	0.90	6.15
	All other MHOs	5.82	22.65
Any formal (non-SUD) psychiatric diagnosis (lifetime; % yes)		49%	156
	Mood disorder (lifetime; % yes)a	37%	119
	Anxiety Disorder (lifetime; % yes)a	34%	109
	Other psychiatric disorder (lifetime; % yes)a	13%	43
	Multiple psychiatric diagnoses (lifetime; % yes)a	34%	108
Social support for recovery (CEST-SS; 9 items, 1-6 scale)		4.83	1.00
Recovery Capital (BARC-10; 10 items, 1-6 scale)		5.04	0.88
Quality of Life (EUROHIS-QOL; 8 items, 1-5 scale)		3.79	0.74
Self-esteem (1 item, 1-10 scale)		6.44	2.35
Psychological distress (Kessler-6, 6 items, 0-4 scale)		2.01	0.79
Length of RCC attendance (in years)		2.55	3.39
Percent days attended RCC in past 90 days		45.62	32.19
Length of typical RCC visit (in hours)		3.13	2.68
Use of RCC Mental & physical health services (% yes)		25%	79
Use of RCC Social services (e.g., housing, childcare, employment; % yes)		69%	221
Use of RCC Addiction/recovery (non-MAT) services (% yes)		94%	302
Use of RCC MAT support services (% yes)		15%	48

Tx = treatment; SUD = substance use disorder; MAUD = alcohol use disorder medication; MOUD = opioid use disorder medication; MHO = mutual help organization; RCC = recovery community center; MAT = medication assisted treatment.

^aVariables are dichotomous (yes/no) and not mutually exclusive (i.e. participants were able to endorse > 1 response category).

3.2. Prevalence of positive and negative medication attitudes

Positive attitudes were more prevalent than negative attitudes for both agonist (positive: 71.4%, 95% confidence interval [CI: 66.1, 76.3%]; negative: 28.6%, 95%CI [23.7, 33.9%]) and antagonist (positive: 76.5%, 95%CI [71.4, 81.1%]; negative: 23.5%, 95%CI [18.9, 28.6%]) MOUDs. Consistent with MOUD, positive attitudes were also more prevalent than negative attitudes for MAUD (positive: 69.3%, 95%CI [63.9, 74.3%]; negative: 30.7%, 95%CI [25.7, 36.1%]) and Mmood (positive: 79.9%, 95%CI [75.1, 84.2%]; negative: 20.1%, 95%CI [15.8, 24.9%]; see Figure 1).

Figure 1.

Prevalence of Positive and Negative Medication Attitudes

Proportion with positive (strongly agree, agree, somewhat agree) and negative (somewhat disagree, disagree, or strongly disagree) attitudes toward agonist and antagonist medications for opioid use disorder (MOUD), as well as attitudes toward medications for alcohol use disorder (MAUD) and psychiatric medications for emotional problems (i.e. mental health conditions; Mmood); Participants had a greater likelihood of reporting a positive attitude than a negative attitude for all medication categories. Error bars represent 95% CI upper and lower bounds.

The prevalence of positive and negative attitudes was similar between agonist MOUD and antagonist MOUD. When compared to MAUD and Mmood, rates of positive and negative attitudes were comparable for agonist and antagonist MOUDs. Positive attitudes were more prevalent for Mmood than MAUD and negative attitudes were more prominent for MAUD than Mmood.

3.3. Correlates & predictors of MOUD attitudes

3.3.1. Agonist MOUD attitudes

Significant correlates (p < 0.10) of agonist MOUD attitudes in correlation analyses included: lifetime history of regular alcohol use (r = −0.15, p = 0.009), history of regular opioid use (r = 0.11, p = 0.049), abstinence-based definition of recovery (r = 0.13, p = 0.018), current abstinence (r = −0.10, p = 0.072), lifetime history of agonist MOUD treatment (r = 0.22, p < 0.0001), lifetime history of regular MHO attendance (r = −0.11, p = 0.042), number of AA meetings attended in the past 90 days (r = −0.17, p = 0.002), number of SMART Recovery meetings attended in the past 90 days (r = −0.11, p = 0.04), percent days of RCC attendance in the past 90 days (r = −0.10, p = 0.069), length of typical RCC visit (r = −0.19, p = 0.0007), utilization of RCC social services (r = −0.12, p = 0.031), and medication for addiction treatment support services (r = 0.13, p = 0.024), lifetime history of a mood disorder (r = 0.10, p = 0.083) or “other” disorder (r = 0.10, p = 0.066), self-esteem (r = −0.11, p = 0.048), and spirituality (r = −0.24, p< 0.0001).

When the study assessed significant correlates in multivariable regression analyses, only lifetime history of agonist MOUD treatment (p = 0.008) and spirituality (p = 0.005) remained significant (see Table 2). More specifically, prior receipt of agonist MOUD was uniquely associated with more positive attitudes toward agonist MOUD, whereas greater spirituality was associated with more negative agonist attitudes. Variables included in the full multivariable model collectively accounted for 22.9% of the variance in agonist attitudes (Nagelkerke’s R²=0.229).

Table 2.

Multivariable logistic regression model for agonist MOUD attitudes (N=300)

		(r-sq = 0.23; model p = 0.003)
		Positive Vs. Negative Attitude
Variable		OR	95% CI	p
Substance Use & Recovery
	Regular alcohol use (Ref = no)	0.53	[0.25, 1.09]	0.08
	Regular opioid use (Ref = no)	0.92	[0.48, 1.74]	0.79
	Definition of recovery (Ref = total abstinence)	2.28	[0.79, 6.59]	0.13
	Current Abstinence (Ref = no)	1.09	[0.45, 2.59]	0.85
Service Use Histories
	Lifetime receipt of agonist MOUD (Ref = no)	3.12	[1.34, 7.26]	0.008
	Lifetime history of any regular MHO attendance (1+ per week; Ref = no)	1.46	[0.54, 3.97]	0.46
	Number of AA meetings attended in past 90 days	1.00	[0.99, 1.01]	0.90
	Number of SMART Recovery meetings attended in past 90 days	1.01	[0.96, 1.06]	0.72
RCC Experience
	Percent days attended RCC in past 90 days	1.00	[0.99, 1.01]	0.77
	Length of typical RCC visit (in hours)	0.91	[0.81, 1.02]	0.10
	Used Social services (e.g., housing, childcare, employment; Ref = no)	0.70	[0.35, 1.42]	0.33
	Used MAT support services (Ref = no)	1.41	[0.46, 4.31]	0.55
Psychological Well-Being
	Mood disorder (Ref = no)	1.66	[0.89, 3.08]	0.11
	Other psychiatric disorder (Ref = no)	2.50	[0.86, 7.26]	0.09
	Self-esteem (1=min; 10=max)	1.02	[0.90, 1.17]	0.72
Demographics
	Spirituality (1=min; 4=max)	0.62	[0.44, 0.87]	0.005

Multivariable logistic regression model for agonist MOUD. Modeling the probability of indicating positive vs. negative agonist MOUD attitudes. Bold indicates statistical significance at p < 0.05. r-sq = Max-rescaled R-Square; MOUD = opioid use disorder medication; MHO = mutual help organization; AA = Alcoholics Anonymous; RCC = recovery community center; MAT = medication assisted treatment.

3.3.2. Antagonist MOUD attitudes

Significant correlates of antagonist MOUD attitudes included: history of regular alcohol use (r = −0.09, p = 0.093), regular cocaine use (r = 0.10, p = 0.072), or regular opioid use (r = 0.16, p = 0.005); history of a primary opioid problem (r = 0.13, p = 0.018), lifetime utilization of residential SUD treatment (r = 0.10, p = 0.066), lifetime receipt of agonist MOUD treatment (r = 0.20, p = 0.0004), length of typical RCC visit (r = −0.19, p = 0.0007), utilization of RCC social services (r = −0.11, p = 0.044), and medication for addiction treatment support services (r = 0.13, p = 0.020); lifetime history of a mood disorder (r = 0.10, p = 0.075), anxiety disorder (r = 0.13, p = 0.019), or multiple co-occurring psychiatric disorders (r = 0.11, p = 0.045); degree of social support for recovery (r = 0.20, p = 0.075); education (r = 0.12, p = 0.033); and spirituality (r = −0.14, p = 0.014).

Multivariable regression models revealed three unique correlates of antagonist MOUD attitudes, including social support for recovery (p = 0.007), education (p = 0.005), and spirituality (p = 0.014; see Table 3). Greater social support was associated with more positive antagonist attitudes. Compared to those with a college degree or higher, individuals with “some” collegiate attainment or a noncollege degree, and individuals with high school or lower education were more likely to have negative antagonist attitudes. Similar to agonist attitudes, greater spirituality was uniquely associated with more negative antagonist attitudes. Variables included in the full multivariable model collectively accounted for 28.9% of the variance in antagonist attitudes (Nagelkerke’s R²=0.289).

Table 3.

Multivariable logistic regression model for antagonist MOUD attitudes (N=304)

			(r-sq = 0.29; model p = 0.0002)
			Positive Vs. Negative Attitudes
Variable			OR	95% CI	p
Substance Use & Recovery
	Regular alcohol use (Ref = no)		0.58	[0.27, 1.23]	0.16
	Regular cocaine use (Ref = no)		1.56	[0.81, 3.02]	0.18
	Regular opioid use (Ref = no)		1.13	[0.46, 2.74]	0.79
	Primary substance: opioid (Ref = no)		1.60	[0.60, 4.22]	0.35
Service Use Histories
	Lifetime use of residential SUD Tx (Ref = no)		1.34	[0.68, 2.64]	0.39
	Lifetime receipt of agonist MOUD (Ref = no)		1.64	[0.64, 4.18]	0.30
RCC Experience
	Length of typical RCC visit (in hours)		0.92	[0.82, 1.03]	0.13
	Used social services (e.g., housing, childcare, employment; Ref = no)		0.54	[0.25, 1.20]	0.13
	Used MAT support services (Ref = no)		2.87	[0.84, 9.83]	0.09
Psychological Well-Being
	Mood disorder (Ref = no)		1.61	[0.63, 4.12]	0.32
	Anxiety disorder (Ref = no)		2.38	[0.75, 7.57]	0.14
	Multiple psychiatric diagnoses (Ref = no)		0.80	[0.21, 3.10]	0.74
	Social support for recovery (CEST-SS; 1=min; 6=max)		1.49	[1.12, 1.99]	0.007
Demographics
	education (Ref = BA or higher)				0.005
		High school or less	0.13	[0.04, 0.46]	0.002
		Some college or other degree	0.23	[0.07, 0.80]	0.02
	Spirituality (1=min; 4=max)		0.66	[0.47, 0.92]	0.01

Multivariable logistic regression model for antagonist MOUD. Modeling the probability of indicating positive vs. negative antagonist MOUD attitudes. Bold indicates statistical significance at p < 0.05. r-sq = Max-rescaled R-Square; SUD = substance use disorder; Tx = treatment; MOUD = opioid use disorder medication; RCC = recovery community center; MAT = medication assisted treatment.

3.4. Exploratory subsidiary analyses: Spirituality

Given commonly used 12-step MHOs have spirituality as a central part of their philosophy and practices, to determine whether MHO participation accounted for outcomes regarding spirituality, we conducted separate multivariable logistic regression models, with spirituality and various 12-step MHO measures as predictor variables of agonist and antagonist MOUD attitudes. In these analyses, spirituality consistently remained a significant correlate (all ps ≤ 0.01), and none of the MHO variables were significantly associated with agonist or antagonist MOUD attitudes (all ps > 0.15). To determine whether spirituality outcomes were specific to MOUD attitudes, or more general medication attitudes, the study conducted identical analyses with Mmood attitudes as the outcomes measure. Outcomes mirrored those of MOUD; spirituality emerged as a significant correlate of Mmood (p = 0.0003) and remained significant with the introduction of various MHO measures (all ps ≤ 0.0004). MHO only emerged as a significant factor in analyses examining regular AA attendance (p = 0.02), with regular lifetime AA attendance being associated with more positive Mmood attitudes. See Table 4 for spirituality and MHO exploratory analyses.

Table 4.

Exploratory multivariable logistic regression models: Assessing the influence of MHO participation on spirituality outcomes for agonist MOUD, antagonist MOUD, and Mmood attitudes.

	Agonist MOUD Attitudes			Antagonist MOUD Attitudes			Mmood Attitudes
	Model 1 (N=315)			Model 1 (N=312)			Model 1 (N=316)
	(r-sq = 0.08; model p < 0.0001)			(r-sq = 0.03; model p = 0.009)			(r-sq = 0.08; model p = 0.0003)
Variable	OR	95% CI	p	OR	95% CI	p	OR	95% CI	p
Spirituality	0.59	[0.46, 0.77]	<0.0001	0.71	[0.55, 0.92]	0.009	0.58	[0.43, 0.78]	0.0003
	Model 2 (N=315)			Model 2 (N=312)			Model 2 (N=316)
	(r-sq = 0.08; model p = 0.003)			(r-sq = 0.04; model p = 0.027)			(r-sq = 0.08; model p = 0.001)
Spirituality	0.60	[0.46, 0.79]	0.0002	0.68	[0.52, 0.91]	0.009	0.56	[0.41, 0.77]	0.0003
Any regular MHO attendance (lifetime; REF=no)	0.91	[0.42, 1.99]	0.82	1.29	[0.59, 2.84]	0.52	1.21	[0.52, 2.84]	0.66
	Model 3 (N=315)			Model 3 (N=312)			Model 3 (N=316)
	(r-sq = 0.08; model p = 0.0003)			(r-sq = 0.04; model p = 0.028)			(r-sq = 0.08; model p = 0.0007)
Spirituality	0.59	[0.45, 0.77]	<0.0001	0.69	[0.53, 0.91]	0.008	0.55	[0.41, 0.75]	0.0001
Past 90-day MHO attendance (REF=no)	1.03	[0.47, 2.22]	0.95	1.29	[0.59, 2.81]	0.53	1.56	[0.69, 3.54]	0.29
	Model 4 (N=314)			Model 4 (N=311)			Model 4 (N=315)
	(r-sq = 0.09; model p = 0.0002)			(r-sq = 0.04; model p = 0.034)			(r-sq = 0.10; model p = 0.0001)
Spirituality	0.62	[0.47, 0.80]	0.0003	0.71	[0.54, 0.93]	0.01	0.53	[0.39, 0.71]	<0.0001
Regular AA attendance (lifetime; REF=no)	0.73	[0.41, 1.29]	0.28	1.04	[0.58, 1.88]	0.89	2.02	[1.10, 3.70]	0.02
	Model 5 (N=315)			Model 5 (N=312)			Model 5 (N=316)
	(r-sq = 0.09; model p = 0.0001)			(r-sq = 0.04; model p = 0.01)			(r-sq = 0.08; model p = 0.001)
Spirituality	0.58	[0.45, 0.75]	<0.0001	0.69	[0.54, 0.90]	0.006	0.57	[0.43, 0.77]	0.0003
Regular NA attendance (lifetime; REF=no)	1.45	[0.87, 2.42]	0.16	1.50	[0.87, 2.58]	0.15	1.10	[0.63, 1.94]	0.73
	Model 6 (N=315)			Model 6 (N=312)			Model 6 (N=316)
	(r-sq = 0.08; model p = 0.0003)			(r-sq = 0.04; model p = 0.02)			(r-sq = 0.09; model p = 0.0004)
Spirituality	0.59	[0.46, 0.77]	<0.0001	0.69	[0.53, 0.90]	0.006	0.55	[0.41, 0.75]	0.0001
# of AA meetings attended (past 90 days)	1.00	[0.99, 1.01]	0.91	1.00	[0.99, 1.01]	0.40	1.01	[0.99, 1.01]	0.15
	Model 7 (N=315)			Model 7 (N=312)			Model 7 (N=316)
	(r-sq = 0.08; model p = 0.0003)			(r-sq = 0.04; model p = 0.03)			(r-sq = 0.08; model p = 0.0009)
Spirituality	0.59	[0.46, 0.77]	<0.0001	0.71	[0.55, 0.92]	0.01	0.58	[0.43, 0.78]	0.0004
# of NA meetings attended (past 90 days)	1.00	[0.99, 1.01]	0.96	1.00	[0.99, 1.01]	0.82	1.00	[0.99, 1.00]	0.34

Exploratory multivariable logistic regression models for agonist MOUD, antagonist MOUD, and Mmood attitudes (positive vs. negative). Modeling the influence of MHO participation on spirituality outcomes for agonist and antagonist MOUD attitudes, as well as Mmood attitudes (for reference). Bold indicates statistical significance at p < 0.05. r-sq = Max-rescaled R-Square; Spirituality: 1=min, 4=max; MHO = mutual help organization; MOUD = opioid use disorder medication; Mmood = medication for emotional problems; AA = Alcoholics Anonymous; NA = Narcotics Anonymous.

4. Discussion

Recovery community centers (RCCs) are one of the fastest growing entities among a new dimension of recovery support services in the United States. Adding to a newly emerging RCC literature, the current study is one of the first to characterize MOUD attitudes and their correlates among RCC attendees.

4.1. Prevalence of positive and negative attitudes toward MOUD

Investigation revealed predominantly positive MOUD attitudes, with more than 70% of RCC attendees exhibiting positive agonist and antagonist attitudes. Although systematic investigation of MOUD attitudes is lacking, research to date has revealed substantially less support for MOUD outside of the RCC milieu, particularly among the recovery community. For example, only about 20% of recovering individuals in the general U.S. population reported positive MOUD attitudes (Bergman et al., 2019). Therefore, our results highlight the potential for this growing tier of recovery support to foster acceptance and peer support for MOUD-facilitated pathways to recovery.

Furthermore, prevalence of positive and negative attitudes was comparable among agonist MOUD, antagonist MOUD, MAUD, and Mmood, suggesting a general positive attitudinal schema without specificity to a medication’s pharmacological properties or indications for use. Prior work has conversely revealed more prominent negative attitudes for antagonist MOUD than agonist MOUD, which the authors argued might be explained by greater overall exposure to agonist concepts in treatment and recovery contexts (Bergman et al., 2019). Though speculative, the inclusive nature of RCCs and the predominance of opioid use disorder among its attendees might amplify exposure to antagonist concepts and patients, thereby enhancing positive attitudinal perceptions of antagonist as well as agonist MOUD.

The OCORT Community Attitudes Survey found that about 70% of residents in a rural Appalachian county disproportionately affected by OUD agreed that medication treatment (Vivitrol or Suboxone) combined with counseling is effective for OUD (Beachler et al., 2020). Consistently, individuals who provide care for OUD patients (e.g., physicians, nurses, community health workers) appear to generally support the use of MOUD and recognize their benefits (Cioe et al., 2020). For example, one study demonstrated that about 70% of providers in New York State had positive attitudes toward methadone (McNeely et al., 2000). A national survey of primary care physicians observed a similar outcome, with the majority (62–77%) reporting agonist MOUD as an effective treatment for individuals with OUD (62-77%) and about half reporting antagonist MOUD as effective (Kennedy-Hendricks et al., 2020). Conversely, a study of American college students found that only 11–14% of individuals perceived MOUD (buprenorphine, methadone, naltrexone) as helpful or very helpful for treating OUD, whereas a significantly greater proportion (71%) perceived psychosocial counseling as helpful or very helpful (Andraka-Christou et al., 2020). A recent systematic review also revealed predominantly negative MOUD attitudes among staff, key stakeholders, and individuals involved in the criminal justice system (Grella et al., 2020). Therefore, MOUD attitudes appear to vary among nonrecovering populations, and contexts in which individuals have enhanced exposure to, or education around, OUD and/or MOUD (e.g., physicians, residents of geographic regions with high rates of OUD) might stimulate more positive MOUD attitudes.

Additional research should help us to better understand attitudinal differences within various treatment and recovery settings, as well as attitudes among the general population. Nonetheless, RCCs appear to house a group of stakeholders with astoundingly high levels of MOUD support, which may constitute an ideal environment for OUD patients to receive recovery support during medication treatment. Indeed, RCCs offer a venue with a high degree of reciprocal social exchange, whereby members give and receive help from one another, and this component appears to be a highly liked part of RCC infrastructure (Kelly et al., 2020). By offering a supportive and inclusive environment, wherein recovering peers hold positive medication attitudes, OUD patients considering or receiving MOUD treatment may feel more welcome and, in turn, more likely to engage in RCC participation. RCC attendance is associated with longer recovery durations and greater recovery capital, which work to lower psychological distress, and enhance quality of life as well as self-esteem (Kelly et al., 2020). Therefore, RCCs might be an appropriate venue for MOUD patients to feel supported by recovering peers and take full advantage of recovery support services, to ultimately benefit their treatment and recovery outcomes.

4.2. Correlates & Predictors of MOUD Attitudes

When we examined correlates of MOUD attitudes, several relationships emerged at a significance level of p < 0.10. However, when we subsequently assessed these variables for their unique contributions, controlling for all other correlates, only four factors were uniquely associated with positive and/or negative attitudes: (1) history of agonist MOUD treatment, (2) social support for recovery, (3) spirituality, and (4) educational attainment.

4.2.1. Agonist MOUD experience

Lifetime history of agonist MOUD treatment was specifically associated with more positive agonist MOUD attitudes, which highlights the importance of lived experience in attitudinal perceptions of MOUD. This finding also speaks to the efficacy of agonist medications for OUD treatment and reveals potential opportunities for facilitating MOUD acceptance among recovering peers. Individuals who have experience with agonist MOUD might have greater opportunity to realize their evidence-based benefits and, in turn, may be more likely to develop a positive perception of agonist treatment. Studies conducted among the broader SUD population provide evidence in support of our finding. Rieckmann and colleagues (2007) investigated MOUD attitudes among residential and outpatient treatment clients, 81% of which had a primary opioid problem, and demonstrated that methadone patients had significantly more positive attitudes toward methadone than other clients. These findings were corroborated by another investigation, showing more positive attitudes in methadone patients relative to acute detoxification patients (Liu et al., 2013). In a study of medication attitudes among social workers (Bride et al., 2013), positive MOUD attitudes were related to higher ratings of its perceived effectiveness, which were associated, in-turn, with greater acceptance of MOUD use in substance use treatment. Furthermore, exposure to agonist MOUD in the context of the social worker’s treatment facility predicted higher ratings of perceived effectiveness. Therefore, a link appears to exist between MOUD exposure and related attitudes, such that personal experience of MOUD receipt and service providers’ exposure to and enhanced hands-on knowledge of MOUD dispensing/patients is associated with more positive attitudes, and these factors might play a dynamic role in related but independent perceptions (e.g., MOUD effectiveness) and support for MOUD patients. Nonetheless, one could speculate alternative mechanisms by which patient exposure is related to MOUD attitudes. For example, individuals with a history of MOUD receipt may have experienced stigma or judgement when undergoing treatment and, as a result, might recognize the importance of social support. Indeed, research suggests that interpersonal support from family members, treatment providers, and most relevant, other individuals who have had experience with MOUD, can facilitate MOUD treatment-seeking and outcomes, whereas low quality social support and broader social stigma are reported barriers to MOUD treatment and hinder recovery outcomes (Hewell et al., 2017). With limited research addressing MOUD attitudes, additional research is needed to better understand the factors that modulate this relationship. Research conducted outside of the treatment and recovery services sector has alternatively failed to reveal MOUD experience as a unique correlate of agonist attitudes (Bergman et al., 2019), suggesting that this relationship may be partially dependent on the population and setting of interest.

At least in the context of treatment and recovery services, sharing lived experiences of successful MOUD treatment with others may ultimately help to educate peers and alter their MOUD perceptions. Indeed, previous studies have demonstrated the important role of peers in recovery. For example, research has shown recovery coaches (i.e., peers sharing the lived experience of addiction and recovery) can motivate behavioral changes in other individuals with substance use disorder (Jack et al., 2018). Leveraging this model of peer support and education might, therefore, help to motivate attitudinal changes among individuals and within treatment/recovery settings with less positive MOUD attitudes. Given the noncausal nature of these data, individuals could have alternatively had positive attitudes prior to agonist MOUD receipt, which predisposed them to gravitate toward MOUD treatment in the first place. Regardless, outcomes would suggest that attitudes remain generally positive after agonist MOUD exposure. Therefore, positive MOUD experiences may help to sustain or foster positive attitudes in MOUD patients, who could ultimately help to spread positive perceptions by sharing their lived experiences with recovering peers. Further investigation is needed to prospectively characterize MOUD attitudinal changes across the lifetime, as a function of MOUD and other treatment/recovery experiences, and to evaluate whether former and current MOUD patients can aid attitudinal changes in recovering peers.

Interestingly, lifetime receipt of MOUD did not emerge as a unique correlate of antagonist attitudes. In a previous investigation, Bergman et al. (2019) examined a nationally representative sample of individuals who had resolved an alcohol or drug problem, and revealed that lifetime MOUD receipt was associated with antagonist, rather than agonist, MOUD attitudes. However, the study only observed this relationship when it categorized attitudes as positive, negative, or neutral, with MOUD experience predicting more positive as well as more negative antagonist attitudes, relative to neutral ones. Therefore, evaluation of antagonist attitudes in the context of an MOUD experience may require more fine-grained analysis. Antagonists (e.g., naltrexone) are shown to be as effective as agonists (e.g., buprenorphine) after induction (Lee et al., 2018), but antagonists are somewhat more difficult to initiate. Naltrexone induction requires abstinence and about 40% of patients who need detoxification do not successfully initiate naltrexone (Jarvis et al., 2018). Nonadherence to MOUD is also a common hurdle—about half of the patients who successfully initiate extended release injectable naltrexone discontinue treatment within six months (Jarvis et al., 2018). The difficulties of detoxification and induction might lead one to have a negative antagonist attitude, but patients who adhere to treatment and experience naltrexone’s benefits might have more positive attitudes. Consistent with national prevalence estimates (Kelly et al., 2017; Hoffman et al., 2020), the majority of RCC attendees did not endorse lifetime history of antagonist MOUD receipt. Therefore, heterogeneous treatment experiences and outcomes among a relatively small subgroup may have limited our ability to detect a relationship between antagonist receipt and attitudes. Additional research that targets individuals with antagonist treatment histories can help to determine whether and how MOUD experiences influence antagonist attitudes.

4.2.2. Social support for recovery

Interestingly, greater social support for recovery was associated with more positive antagonist attitudes. Why this measure correlated with antagonist but not agonist attitudes is unclear. Individuals with adequate support from family, friends, and significant others may have more faith in the efficacy of antagonist medications. As noted, antagonists are more difficult to initiate due in part to required abstinence prior to induction (Lee et al., 2018) and social support may help to facilitate treatment motivation and adherence, with the idea that successful antagonist treatment is possible with the abundant support of close interpersonal contacts translating to greater recovery capital (Cooper & Nielsen, 2017). Alternatively, these outcomes might reflect a belief that agonists represent a less optimal clinical course to pursue because they are still “opioids” and may therefore implicitly convey a notion of “continued use”. Qualitative research could be helpful in uncovering the exact reasons for this association between more positive social support for recovery and more support for antagonists rather than agonist MOUD.

4.2.3. Spirituality

A greater degree of spirituality was associated with more negative agonist and antagonist MOUD attitudes, whereas religiosity was unrelated to MOUD perceptions (i.e., did not emerge as a correlate of agonist or antagonist MOUD attitudes at a significance level of p < 0.10). Spirituality is arguably conceptually broader than religiosity; it often involves dynamic processes with personal and experiential components that can exist independent of religion (Mueller et al., 2001). Scholars have proposed various definitions of spirituality and, although consensus has not been reached, overlapping themes of modern definitions suggest spirituality is related to the constructs of connectedness (e.g., to others, nature, higher power), pursuit of meaning and purpose, transcendentalism, interpersonal values, and individual well-being (Mueller et al., 2001). Nonetheless, spirituality is an arguably subjective experience (Tuncay, 2007). Although 12-step MHOs (e.g., NA, AA) are explicitly grounded on spiritual ideology, exploratory analyses revealed that 12-step participation did not account for the relationship between spirituality and MOUD attitudes. This finding might be explained by the fact that spirituality in the context of the 12-steps focuses on a “higher power” or “god” (Kelly et al., 2011). Within the framework of health and medicine, spirituality has also been defined in the context of “mind-body”; from this perspective, spirituality controls the mind, the mind controls the body, and all three culminate to govern health (Tuncay, 2007). Furthermore, one’s belief in “god or a higher power” might be replaced or supplemented with nature, arts, and/or social environment, with aspects of intrapersonal and interpersonal spirituality via connections to one’s self and others, respectively (Tuncay, 2007). Thus, spiritual identity is a complex construct, seemingly dependent on personal beliefs and social/environmental milieu. In the context of medication, spiritually guided mind-body principles would translate to a belief that medication is unnecessary because the mind has the ability to heal the body, and investigations among HIV patients suggest this perspective can be detrimental to medication initiation and adherence (Kremer et al., 2009). Importantly, research has linked spirituality to enhanced psychological and physical health, as well as superior recovery from a variety of medical and psychiatric disorders (Dedeli & Kaptan, 2013). Therefore, spirituality generally appears to be a protective factor that should be embraced in the context of addiction recovery, and incorporating mind-body guided education to correct misperceptions of MOUD might ultimately help spiritual individuals to view MOUD more positively without challenging their spiritual experiences, beliefs, and practices. However, we are aware of only one other study assessing/demonstrating this negative relationship between spirituality and MOUD attitudes (Bergman et al., 2019), and additional research would need to replicate this finding and identify the specific aspects of spirituality that contribute to MOUD perceptions.

4.2.4. Education

Educational attainment also emerged as a unique correlate of antagonist MOUD attitudes. Having a bachelor’s degree or higher was associated with more positive perceptions. Although our findings were specific to antagonists and the reasons for this are unclear, the observed relationship between MOUD attitudes and education is generally supported by prior research. Greater educational attainment among treatment providers is associated with more positive perceptions of naltrexone, buprenorphine (Fitzgerald & McCarty, 2009), and methadone (Aletraris et al., 2016) for OUD treatment. Although the majority of publications to-date have focused on providers’ attitudes, patient-focused research has also suggested a potential link between education and MOUD attitudes. For example, residential and outpatient treatment clients with college degrees reported more positive attitudes toward buprenorphine than those with lower educational attainment (Rieckmann et al., 2007). One might speculate that individuals with Bachelor’s degrees and higher educational attainment may be more likely to have greater recovery capital (e.g., employment, financial support) and enhanced resources for educating themselves about naltrexone (e.g., via private treatment providers) and, therefore, may have more positive perceptions about how naltrexone can be integrated into a comprehensive treatment plan to support a positive recovery trajectory. Furthermore, studies suggest that higher educational attainment is associated with longer naltrexone treatment durations and, in-turn, better treatment outcomes (Saxon et al., 2018). In line with the homophily principal of social circle structures, interpersonal contact occurs at a higher rate among individuals with similar, rather than dissimilar, characteristics (McPherson et al., 2001). Therefore, recovering individuals with higher educational attainment are more likely to have other highly educated recovering individuals in their social circles, and may therefore be exposed to successful naltrexone treatment experiences through their social contacts with similarly high levels of educational attainment, who would therefore be more likely to complete a full treatment episode and experience the benefits of antagonist MOUD. Moreover, naltrexone has received substantially less media coverage than buprenorphine and methadone (Kennedy-Hendricks et al., 2019), which would require individuals to seek other resources for educating themselves about naltrexone—more highly educated individuals may be more likely to have the skills to access and comprehend such information outside of a layman’s media context. However, other studies have failed to reveal a relationship between education and MOUD attitudes among recovering individuals (e.g., Bergman et al., 2019) and additional research is needed to better understand how, and in what context, education relates to MOUD perceptions.

4.3. Limitations

Outcomes should be interpreted with consideration of important limitations inherent to cross-sectional survey designs. Participants answered questions about lifetime receipt of medication treatment without consideration of detailed treatment regimens. Given the influence of MOUD treatment duration and dose on therapeutic outcomes, and the potential for therapeutic outcomes to influence MOUD attitudes, additional research should evaluate the relationship between various treatment aspects (e.g., dose, duration, use of other support services, clinic experiences) and MOUD attitudes. Although results highlight the potential for RCCs to foster MOUD acceptance, findings are not causal and may alternatively be a function of sample selection and the nature of our cross-sectional design. Accordingly, whether positive MOUD attitudes preceded MOUD receipt is unclear, making these individuals more likely to seek out and engage in pharmacotherapy, or if MOUD exposure subsequently facilitated positive perceptions. Longitudinal research should evaluate the directionality and mechanisms underlying this relationship. Furthermore, whether individuals with more positive MOUD attitudes are selecting into RCCs or if those with more negative attitudes are opting out of RCCs due to a conflict with their maxim of many pathways to recovery is unclear and merits further study. Findings regarding spirituality also require more comprehensive investigation to better understand which aspects of spirituality account for more negative MOUD attitudes and to evaluate 12-step participation in a more detailed manner, as the current study did not assess participants’ definitions of spirituality or detailed histories of engagement/experience with the 12-step community (e.g., meeting engagement, membership roles). Given relatively limited investigation of MOUD attitudes to date, the primary questions used to assess MOUD attitudes in this study have not yet been scientifically validated and results should therefore be considered preliminary. The study sample was recruited from 95% of RCCs in the New England area and additional research should determine if these findings apply to RCCs in other regions of the United States. Future studies of MOUD correlates that assess a broader range of RCCs and other predictor variables, including additional continuous measures not captured by the current study, will ultimately help to determine if these findings apply to RCCs in regions outside of New England, and elucidate additional factors related to MOUD attitudes.

5. Conclusion

Negative MOUD attitudes constitute barriers to treatment receipt (Khazaee-Pool et al., 2018). Given the importance of expanding MOUD treatment to address the opioid epidemic, research must gain a better understanding of MOUD attitudes among relevant stakeholders in MOUD dissemination and utilization, including individuals with lived experience of addiction and recovery. The current study provides a first look at MOUD attitudes and their correlates in RCC attendees. Our findings indicate predominantly positive agonist and antagonist MOUD attitudes among RCC attendees, highlighting the potential for this growing tier of recovery support to foster acceptance and peer support for medication-facilitated pathways to recovery. A greater degree of spirituality and less educational attainment were associated with more negative MOUD attitudes. Alternatively, lifetime receipt of agonist MOUD treatment and greater social support for recovery were markers of positive MOUD attitudes, revealing opportunities for facilitating MOUD acceptance within the recovery support framework. In line with models of peer support, leveraging lived experiences of successful MOUD treatment by sharing them with others might ultimately help to motivate attitudinal changes in recovering peers with less favorable perceptions of MOUD. Additional research will need to replicate the findings of this preliminary investigation and elucidate the factors underlying these relationships. Continued focus on this area of research could ultimately help to identify system- and individual-level methodologies to promote positive MOUD attitudes among relevant stakeholders and, in turn, enhance interest/engagement in MOUD treatment and boost peer support for MOUD patients.

Highlights.

• RCC attendees had primarily positive attitudes toward agonist/antagonist medication

• History of agonist treatment predicted positive agonist attitudes

• Greater social support for recovery predicted positive antagonist attitudes

• Greater spirituality predicted negative attitudes independent of 12-step involvement

• Lower educational attainment predicted negative antagonist attitudes