Interaction effects in the association between methadone maintenance therapy and experiences of racial discrimination in U.S. healthcare settings

George Pro; Nick Zaller

doi:10.1371/journal.pone.0228755

. 2020 Feb 6;15(2):e0228755. doi: 10.1371/journal.pone.0228755

Interaction effects in the association between methadone maintenance therapy and experiences of racial discrimination in U.S. healthcare settings

George Pro ^1,^*, Nick Zaller ²

Editor: Stéphanie Baggio³

PMCID: PMC7004348 PMID: 32027723

Abstract

Background

Disparities in methadone maintenance therapy (MMT) outcomes have received limited attention, but there are important negative outcomes associated with MMT that warrant investigation. Racial discrimination is common in healthcare settings and affects opioid use disorder (OUD) treatment and comorbidities. However, race/ethnicity alone may not fully explain experiences of discrimination. MMT remains highly stigmatized and may compound the effect of race/ethnicity on discrimination in healthcare settings. We sought to quantify differential associations between MMT and experiences of racial discrimination between racial/ethnic groups in a U.S. national sample.

Methods

We used the National Epidemiologic Survey on Alcohol and Related Conditions-III (2012–2013) to identify a subset of individuals with a lifetime OUD who had ever used MMT (survey n = 766; weighted population n = 5,276,507). We used multivariable logistic regression to model past-year experience of racial discrimination in a healthcare setting. We included an interaction term between race/ethnicity and MMT status to identify the odds of discrimination (MMT vs. no MMT [referent]) within racial/ethnic groups. We used survey procedures with weights to account for the parent study’s complex survey design.

Findings

Twenty-two percent of our sample experienced racial discrimination in a healthcare setting in the past year. Discrimination was more common among those who had ever used MMT (x² = 10.00, p = 0.001) and racial/ethnic minorities (x² = 23.15, p<0.001). The interaction effect was much stronger than the main effects of race/ethnicity and MMT status. MMT status (versus no MMT) was positively associated with discrimination among Blacks (aOR = 3.93, 95% CI = 3.87–3.98, p<0.001), Whites (aOR = 2.25, 95% CI = 2.23–2.27, p<0.001), and Latino/Latinas (aOR = 1.59, 95% CI = 1.55–1.62, p<0.001). Among American Indian/Alaska Natives (AI/AN), those who had used MMT had over thirty times the odds of racial discrimination, compared to their non-MMT counterparts (aOR = 32.78, 95% CI = 31.16–34.48, p<0.001).

Conclusion

Race/ethnicity alone did not sufficiently account for racial discrimination in healthcare settings among those with a lifetime OUD. MMT status was strongly associated with racial discrimination among AI/AN. Our strong interaction effect is indicative of an additional barrier to health services utilization among AI/AN, which has important implications for OUD treatment outcomes and comorbidities. Health promotion programs aimed at increased adoption of MMT are promising, but should be considered in the context of racial/ethnic disparities, drug use and MMT stigma, and implicit biases in clinical settings.

Background

In the U.S., the prevalence of heroin and opioid misuse has risen substantially over the past decade [1], disproportionately impacting Whites [2, 3], women [4], and less wealthy and underinsured populations [5]. Opioid misuse is commonly linked to community availability of opioids, perceptions of harms, and structural factors such as economic hardship [6, 7], but the causal linkages may be bidirectional for many of these factors.

Broadly, the racial/ethnic and gender trends in opioid misuse closely mirror those observed in the contemporaneous rise in heroin- and opioid-related overdose deaths [8, 9]. However, rates of overdose deaths are actually higher among non-Whites in some U.S. states. For example, the rates of overdose deaths among Blacks are more than ten points higher than Whites in Iowa (White, 7 per 100,000; Black, 21 per 100,000), Missouri (White, 16 per 100,000; Black, 32 per 100,000), and Washington D.C. (White, 9 per 100,000; Black, 60 per 100,000) [10]. Relatedly, the rate of change in overdose deaths also varies by state and by race/ethnicity. For example, in Minnesota, the rate of change has been the greatest among American Indian/Alaska Natives (AI/AN) (123% increase), followed by Blacks (105%) and Whites (72%) [11]. Opioid use is also positively associated with arrest and incarceration [12–14], resulting in the overrepresentation of opioid use disorders (OUDs) in incarcerated populations [15].

Methadone maintenance therapy (MMT) is one type of a broad range of medication-assisted therapies (MAT), and many research reports vary in terms of the specific type of medication assistance being studied. While the focus of the current study is MMT, a general understanding of other types of MAT is useful in developing a broad understanding of disparities in OUD treatment. MMT for the treatment of OUD is effective in reducing withdrawal symptoms and maintaining abstinence [16]. Medication-assisted therapies in general, and MMT specifically, are positively associated with better health and social outcomes across groups, including a decreased frequency of injecting opioids, less involvement in criminal activity, a decrease in infectious disease exposure, improved work productivity and employment, and other functional outcomes [17–19]. The use of MMT has increased steadily since the early 2000s [20], although reports of racial/ethnic disparities in MMT and MAT utilization are mixed. For example, Krawczyk and colleagues [21] found that White heroin users were the least likely to receive medication-assisted treatment (either methadone or buprenorphine), even though Whites have demonstrated the highest prevalence of OUD in recent years. Conversely, other findings have illustrated a lower likelihood of both MMT-specific and MAT uptake among Blacks and Hispanics [22, 23].

In the U.S., experiences of racial discrimination in healthcare settings are more common among racial/ethnic minorities than among Whites. [24, 25]. MMT status—itself a source of stigma and bias—may exacerbate racial discrimination and impact treatment retention [26]. In a subsample of people who use illicit drugs, McKnight and colleagues [27] found that perceptions of institutional racial discrimination in healthcare settings negatively affected health services utilization, and that Latino/Latina drug users were the least likely to perceive institutional racism. However, the authors did not disaggregate their findings by opioid-specific substances or participation in MMT. In addition to racial/ethnic discrimination, stigmas toward drug use and opioid-based medication treatment are also common experiences among MMT participants [28]. Drug use stigma is a barrier to substance use treatment utilization among MMT participants, and the cumulative stigmas of age, race, and poverty compound this relationship [29]. Furthermore, healthcare professionals providing non-OUD services to MMT clients are largely untrained in addiction medicine, which plays a role in the perpetuation of stigma towards OUDs in healthcare and clinical settings [30].

Importantly, studies on racial/ethnic differences in the effects of MMT on discrimination are lacking in the public health literature. However, several studies in MMT settings have illustrated the importance of focusing on racial discrimination. For example, Black and Latino/Latina patients are more likely than Whites to receive methadone doses lower than the recommended amount [31, 32]. In considering differences in MMT-related factors between racial/ethnic groups, AI/AN have been largely excluded from MMT research. However, and of particular importance to this study, AI/AN MMT clients have cited institutional discrimination and experiences of prejudice in non-Indian Health Service settings as barriers to accessing and complying with MMT services [33].

Volkow and colleagues [34] recently and thoughtfully called for greater involvement of healthcare professionals in improving access to medication-assisted therapies and addressing comorbidities. However, this insight into the future of opioid treatment should also include consideration of possible negative MMT outcomes. Racial discrimination is common in general healthcare settings, but there may be underlying mechanisms driving discriminatory behavior in addition to race/ethnicity. MMT status may compound healthcare-based racial discrimination. In the current study, we sought to identify group differences in the association between MMT status and experiences of racial discrimination in healthcare settings between AI/ANs, Blacks, Latino/Latinas, and Whites with lifetime heroin or opioid use disorders. We hypothesized that 1) MMT status would be positively associated with racial discrimination, 2) racial/ethnic minority statuses would each be positively associated with racial discrimination and group odds ratio estimates would vary in magnitude, compared to Whites, and 3) the interaction effects would be stronger than the main effects alone.

Methods

Data source

We used the National Survey on Alcohol and Related Conditions-III (NESARC-III) (2012–2013). A general description of NESARC-III and data collection methods have been described elsewhere [35]. In short, NESARC-III provides a nationally representative sample with variables describing substance use disorders and treatment services utilization, as well as many social and environmental indicators related to public health. The dataset includes estimates for population weights and strata, reflective of the parent study’s complex sampling methods. The household survey response rate was 72%. Analyses of complex survey data requires the use of weights to compensate for variable probabilities of selection and differential nonresponses rates by group. NESARC-III provides pre-calculated sampling weights designed to calibrate the sampling distribution for major subgroups defined by region, sex, age, and race/ethnicity [35]. In the full, unrestricted dataset (survey n = 36,309; weighted population n = 235,411,957), less than 0.1% of those without an OUD had a record of MMT participation. Thus, in order to focus on MMT participants as a subset of those with OUD, we restricted the dataset to include only individuals with records indicating a lifetime OUD (inclusive of both heroin use disorder and opioid use disorder) and complete data for all study variables (survey n = 766; weighted population n = 5,276,507).

Variables

Our outcome of interest was a binary indicator of whether an individual experienced racial discrimination in a healthcare setting in the past year (yes or no [referent]). Response options to discrimination survey questions included very often, fairly often, sometimes, almost never, and never. Our binary variable included those who endorsed no experience (never) or any experience (very often, fairly often, sometimes, or almost never) with racial discrimination in a healthcare setting. Survey respondents who reported any experience with either of the following two questions were coded as having ever experienced racial discrimination in a healthcare setting: 1) During the last 12 months, about how often did you experience discrimination in your ability to obtain healthcare or health insurance coverage because of your race or ethnicity? 2) During the last 12 months, about how often did you experience discrimination in how you were treated when you got care because of your race or ethnicity? The type of healthcare setting was not specified in these two NESARC-III questions, leaving the interpretation of what constitutes a healthcare setting up to the survey respondent.

The main effects of interest were MMT treatment utilization and race/ethnicity. MMT status included whether an individual ever attended MMT treatment services (ever or never [referent]). We did not consider past-year MMT utilization alone due to small sample size and subsequent regression parameter estimation problems caused by small covariate cell sizes. Also, NESARC-III does not capture utilization of other agonist therapies such as buprenorphine. Our racial/ethnic groups included White (referent), Black, Latino/Latina, and American Indian/Alaska Native. We did not include the Asian/Native Hawaiian/Other Pacific Islander group in our analyses because of its small sample size, and for whom several cross-tabulation cells were equal to zero.

We also considered several covariates in our study, based on a priori knowledge of factors that likely confound the relationships between MMT, race/ethnicity, and discrimination in healthcare settings. We included sex (male [referent] or female), age group (18–29 [referent], 30–39, 40–49, 50+ years old), geography (urban [referent] or rural), sexual orientation minority status (heterosexual/straight [referent], or gay/lesbian, bisexual, or not sure), opioid use disorder type (heroin use disorder only [referent], illicit/prescription opioid use disorder only, or both), past-year dual mental health disorder (none [referent] or any), ever homeless in the past year (never [referent] or ever), enrollment in Medicare or Medicaid (yes or no [referent]), and any history of incarceration (yes or no [referent]). We also considered past-year non-opioid substance use disorder diagnoses (none [referent], alcohol use disorder only, drug use disorder only, or both alcohol and drug use disorders), using pre-defined diagnostic variables provided by NESARC which are based on criteria for substance use disorders outlined by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition.

Analysis

We used SAS software (v9.4) [36] and applied survey weights to all analyses. We used multivariable logistic regression to model racial discrimination in healthcare settings. Specifically, we used an interaction term between MMT and race/ethnicity to calculate the associations between MMT and racial discrimination within racial/ethnic groups. Within-group adjusted odds ratios, 95% confidence intervals, and p-values were generated using LSMEANS and SLICEDIFF options using PROC GLIMMIX. For our fitted model, following a strategy of purposeful selection of covariates for multivariable regression outlined by Hosmer and Lemeshow [37], we carried forward variables with chi-square p-values less than 0.20 from bivariate tests.

Results

The majority of individuals with a lifetime OUD were White (79%), younger (31% 18–29 years, 22% 30–39 years), resided in rural areas (75%), and were not covered by Medicare or Medicaid (66%) (Table 1). A clear minority ever utilized MMT (9%), and nearly half had a co-occurring substance use disorder (44%) and had ever been incarcerated (46%).

Table 1. Weighted descriptive characteristics of individuals with lifetime opioid use disorders (NESARC-III, 2012–2013) (survey n = 766; weighted population n = 5,276,507).

		Past-year experience with racial discrimination in healthcare settings
Variables	Total	Ever experienced 22.48%	Never experienced 77.52%	x²	p
	%	col %	col %
Methadone treatment status				10.00	0.002
Never utilized	90.93	83.86	92.95
Ever utilized	9.07	16.13	7.04
Race/ethnicity				23.15	<0.001
White	78.86	64.45	83.05
Black	9.21	17.01	6.94
Latina/Latino	9.29	14.27	7.84
American Indian/Alaska Native	2.64	4.27	2.17
Sex				5.04	0.023
Male	54.05	62.91	51.48
Female	45.94	37.09	48.51
Age group				2.46	0.482
18–29	30.97	25.52	32.55
30–39	21.64	23.01	21.25
40–49	19.05	19.62	18.89
50+	28.33	31.85	27.30
Urbanicity				0.43	0.510
Urban	22.57	24.91	21.89
Rural	77.42	75.09	78.10
Sexual orientation minority				0.00	0.983
No	91.72	91.67	91.73
Yes	8.28	8.32	8.27
Opioid use disorder type				0.26	0.880
Opioid use disorder only	79.39	80.98	78.93
Heroin use disorder only	9.81	8.84	10.10
Both opioid and heroin use disorders	10.78	10.17	10.96
Past-year non-opioid substance use disorder diagnoses				2.45	0.458
None	55.64	51.87	56.74
Alcohol use disorder only	23.28	26.40	22.37
Drug use disorder only	8.14	10.72	7.39
Both AUD and DUD	12.94	11.00	13.50
Past-year dual disorder diagnosis				23.34	<0.001
None	40.45	22.76	45.59
Any	59.54	77.24	54.41
Ever homeless in the past year				3.46	0.063
No	90.07	85.22	91.49
Yes	9.92	14.78	8.51
Medicare or Medicaid coverage in the past year				5.54	0.019
No	65.54	56.45	68.18
Yes	34.45	43.55	31.82
Ever been incarcerated				7.64	0.006
No	53.71	42.52	56.97
Yes	46.29	57.48	43.03

Open in a new tab

Notes: Percentages are weighted, column sums may not equal 100.00; Rao-Scott x²

Twenty-two percent of those with a lifetime OUD experienced racial discrimination in a healthcare setting in the past year. Experiences of racial discrimination were more likely among those who had ever used MMT (p = 0.002), racial/ethnic minority groups (p<0.001), men (p = 0.023), those with a past-year dual disorder diagnosis (p<0.001), those who were homeless in the past year (0.063), those enrolled in Medicare or Medicaid (p = 0.019), and those with a history of incarceration (p = 0.006).

Table 2 shows results from our fitted model. Only sex, past-year dual disorder diagnosis, past-year homeless status, enrollment in Medicare/Medicaid, and incarceration history were carried forward from bivariate analyses. Individuals with a lifetime OUD who had ever utilized MMT had more than four times higher odds of experiencing racial discrimination in healthcare settings, compared to their non-MMT counterparts (adjusted odds ratio [aOR] = 4.28, 95% confidence interval [95% CI] = 4.21–4.35, p<0.001). All racial/ethnic minority groups had higher odds of discrimination than Whites, and the association was particularly pronounced among AI/AN (aOR = 4.99, 95% CI = 4.86–5.12, p<0.001) and Blacks (aOR = 3.93, 95% CI = 3.87–3.98, p<0.001).

Table 2. Multivariable logistic regression modeling past-year experience with racial discrimination in healthcare settings among individuals with lifetime opioid use disorders (NESARC-III, 2012–2013) (survey n = 766; weighted population n = 5,276,507).

Variables	aOR	95% CI	p
Main effects
MMT status
Never utilized MMT	1.00
Ever utilized MMT	4.28	(4.21, 4.35)	<0.001
Race/ethnicity
White	1.00
Black	3.93	(3.87, 3.98)	<0.001
Latino/Latina	2.16	(2.13, 2.19)	<0.001
American Indian/Alaska Native	4.99	(4.86, 5.12)	<0.001
Interaction effects
*MMT status Race/ethnicity** Within-group comparisons of MMT status (Ever vs. Never [ref])
White	2.25	(2.23, 2.27)	<0.001
Black	2.87	(2.79, 2.94)	<0.001
Latino/Latina	1.59	(1.55, 1.62)	<0.001
American Indian/Alaska Native	32.78	(31.16, 34.48)	<0.001
Covariates
Sex
Male	1.00
Female	0.51	(0.50, 0.52)	<0.001
Past-year dual disorder diagnosis
None	1.00
Any	3.38	(3.36, 3.40)	<0.001
Ever homeless in the past year
No	1.00
Yes	1.21	(1.20, 1.22)	<0.001
Medicare or Medicaid coverage in the past year
No	1.00
Yes	1.35	(1.34, 1.36)	<0.001
Ever been incarcerated
No	1.00
Yes	1.29	(1.28, 1.30)	<0.001

Open in a new tab

The magnitude of associations between MMT and racial discrimination in healthcare settings varied widely between racial/ethnic groups. Among Blacks, Whites, and Latino/Latinas, those who ever utilized MMT (versus never) had higher odds of experiencing racial discrimination (Black—aOR = 2.87, 95% CI = 2.79–2.94, p<0.001), (White—aOR = 2.25, 95% CI = 2.23–2.27, p<0.001), (Latino/Latina—aOR = 1.59, 95% CI = 1.55–1.62, p<0.001). The odds of racial discrimination were especially pronounced among AI/AN, such that those who had ever used MMT demonstrated more than 30 times higher odds of discrimination, compared to their non-MMT counterparts (aOR = 32.78, 95% CI = 31.16–34.48, p<0.001).

With respect to other covariates of interest, women demonstrated lower odds of experiencing racial discrimination in healthcare settings than men (aOR = 0.51, 95% CI = 0.50–0.52, p<0.001). Those with past-year dual disorder diagnoses (aOR = 3.38, 95% CI = 3.36–3.40, p<0.001), had been homeless in the past year (aOR = 1.21, 95% CI = 1.20–1.22, p<0.001), were enrolled in Medicare/Medicaid (versus not enrolled) (aOR = 1.35, 95% CI = 1.34–1.36, p<0.001) and had a history of incarceration (versus no history) (aOR = 1.29, 95% CI = 1.28–1.30, p<0.001) had higher odds of experiencing racial discrimination in a healthcare setting.

Discussion

We found evidence of a strong interaction effect of race/ethnicity on the association between MMT and racial discrimination in healthcare settings. Previous research has demonstrated that MMT is effective in treating OUDs, with decades of research documenting positive clinical and social outcomes among diverse patient populations. However, MMT status may also have negative consequences. We sought to better understand racial/ethnic differences in the association between MMT and experiences of racial discrimination in general healthcare settings. Specifically, we found that MMT status was independently and positively associated with experiences of racial discrimination in our adjusted model. Among AI/AN, the association between MMT status and experiences of racial discrimination was positive and very strong.

Reports on disparities in MMT outcomes are rare. Rather, disparities research tends to focus on antecedents to MMT utilization, including barriers to treatment initiation and implicit bias in treatment and healthcare settings. Our findings contribute to the MMT literature by framing discrimination as an unintended consequence associated with MMT and by focusing on differential racial/ethnic effects of MMT-related discrimination. Our study complements a recent meta-analysis by Fitzgerald and Hurst [38], who demonstrated occasions of implicit bias among healthcare providers, ranging from predominately race-based biases to biases against intravenous drug users, patients with mental health problems, and people with multiple marginalizing characteristics. Our study also aligns with work by Hansen and Roberts [39], who identified that MMT is often portrayed using politicized imagery of poor, Black, and Hispanic treatment-seekers. In stark contrast to the stigma surrounding MMT, Lagisetty and colleagues [40] recently demonstrated that buprenorphine is associated with White treatment clients who pay with private insurance or self-pay for services. We expand on the current literature by measuring the compounding effects of MMT and race/ethnicity on racial discrimination. In this sense, race/ethnicity alone does not sufficiently explain race-based biases in healthcare settings.

We found that women were less likely than men to perceive racial discrimination in healthcare settings. Rowan-Szal and colleagues [41] found that women in MMT were more likely than men to build positive rapport with clinic staff, and women generally demonstrated higher levels of motivation to successfully complete treatment. Relatedly, women are more likely than men to remain in MMT services for up to 90 days [42], while several other studies have demonstrated better MMT outcomes among women [43, 44]. Despite the broad gender differences in MMT outcomes, more research is needed to identify gender differences within racial/ethnic groups. In addition, experiences of racial discrimination may also vary by gender when stratified by racial/ethnic group. Future research is needed to identify disparate trends in discrimination and MMT by both race/ethnic and gender groups.

MMT clients have complex healthcare needs and typically present with multiple comorbidities [45, 46]. For example, results from our descriptive analyses indicated that nearly 60% of OUD treatment participants also had a dual mental health diagnosis in the past year. This finding was expected and aligns with previous reports that dual mental health and psychiatric disorders are exceptionally high among individuals with OUD [47, 48]. Importantly, additional research has pointed to the intersectional relationships between mental health problems, MMT utilization, and the experience of social inequities, in particular among MMT patients who also struggle with mental health diagnoses [49]. Future research may begin to unpack the complex relationships between race/ethnicity, mental health, stigma, discrimination, and MMT services utilization.

The interaction between MMT and race/ethnicity highlights an additional barrier to receiving general health services. Experiences of discrimination negatively affect compliance and further utilization of general healthcare [24, 50], which may worsen MMT comorbidities by delaying or foregoing participation in other specialty services. Specifically, AI/AN have demonstrated poorer health on multiple metrics relative to their non-AI/AN counterparts [51], which may be partially due to the convergence in AI/AN communities of the opioid epidemic, multiple barriers to behavioral and general health services, and the under-treatment of comorbid conditions. The treatment and resolution of comorbidities is effective when addressed in parallel to the treatment of OUD or in integrated behavioral health clinical settings [52, 53] but racial discrimination by health providers is associated with low health services utilization [24]. Thus, it is reasonable to infer that resolving co-occurring health problems in integrated healthcare settings is a critical yet particularly challenging obstacle to OUD treatment success in general, and success among AI/AN in particular.

Our findings also align with recent calls to incorporate AI/AN subsamples into research addressing opioid- and treatment-related outcomes. While the Indian Health Service (IHS) has made recent and notable strides in improving OUD treatment outcomes [54, 55], research into determinants of treatment outcomes among AI/AN has not kept up. While the implementers of the NESARC-III survey worked closely with tribal leaders to reach a nationally representative sample of AI/AN participants [35], it was not possible to discern whether MMT was prescribed through the IHS in this dataset. However, given the IHS’ emphasis on culturally-centered health services [56], as well as AI/ANs’ preferences for IHS-specific health services and the incorporation of traditional indigenous elements into health services delivery [57, 58], it is reasonable to suggest that our strong findings of racial/discrimination in healthcare settings among AI/AN may be driven, in part, by AI/AN biases inherent in non-IHS services [51, 59]. Importantly, the majority of AI/AN do not receive health services through the IHS [56], and only 1% of substance abuse treatment providers identify as AI/AN [60]. Given this relatively low coverage of IHS services, the risk for experiences of discriminatory behavior may be higher among AI/AN than other groups. Broadly, the findings from this study may inform future research into treatment services within AI/AN communities. For example, scientific investigations may be warranted that address components of the ongoing National Institutes of Health initiative entitled Interventions for Health Promotion and Disease Prevention for Native American Populations. These programs are specifically aimed at increasing health services utilization and improving health through 1) community-driven intervention and research processes, 2) multi-sectoral research and practice partnerships, and 3) support for emerging tribal research infrastructures [61].

Limitations

NESARC-III only includes information on MMT—there are no indicators for other medication-assisted therapies (buprenorphine, naltrexone) that are also of interest in opioid treatment outcomes research. The association between MMT and racial discrimination likely differs depending on the specific type of medication assistance. In addition to methadone, future research may compare differential effects of buprenorphine and naltrexone on OUD treatment outcomes.

NESARC-III data is cross-sectional in nature, limiting interpretations of study results to associations. We cannot infer causal relationships with this data. Also, the NESARC-III sample includes only survey participants based in the US. While OUD and MMT treatment disparities may be widespread globally, we cannot generalize our findings to other global settings. Finally, NESARC-III data ranges from 2012–2013. There have been multiple public health and policy initiatives aimed at preventing OUD and reducing the harm associated with it since the collection of NESARC-III data. Thus, it is possible that a replicated study using more current data may show differences in the size and strength of associations.

Conclusion

MMT is an effective OUD treatment strategy, but there remains a need to investigate negative MMT outcomes. We found that MMT status is associated with experiences of racial discrimination in general healthcare settings. Race/ethnicity moderated this relationship, such that AI/AN who had ever used MMT had drastically higher odds of past-year discrimination, compared to the effect of MMT within all other racial/ethnic groups. Our findings highlight racial discrimination as a possible barrier to healthcare services utilization among groups with multiple marginalized statuses—namely AI/AN with a history of OUD and MMT use. Furthermore, it is important to note the challenges of interpreting one expression of bias (racial discrimination) in the midst of other latent factors that may also affect how racial discrimination is both expressed and perceived in healthcare settings.

Promoting culturally tailored healthcare may reduce implicit biases towards underrepresented patient populations. Integrating mental and behavioral health services into general medical practices may strengthen providers’ understanding of addiction and treatment, as well as their ability to treat OUD-related comorbidities. Health promotion interventions should continue the focus on increasing MMT uptake, while also working to dissolve many of the underlying systemic and discriminatory barriers to treatment and other health services utilization.

Data Availability

The National Epidemiologic Survey on Alcohol and Related Conditions-III (NESARC-III) data cannot be shared because its use is restricted by the National Institute of Alcohol Abuse and Alcoholism (NIAAA). Access to the dataset is granted by NIAAA to researchers who successfully apply. Once obtained, NIAAA does not grant dataset users permission to share NESARC-III data. The authors of the current study did not have any special access privileges to the dataset that other applicants would not have. Interested researchers will be able to replicate the results of this study by following the protocol outlined in the current paper’s Methods section. Researchers who are interested in obtaining the NESARC-III dataset through NIAAA can find more information here: https://www.niaaa.nih.gov/research/nesarc-iii/nesarc-iii-data-access.

Funding Statement

The authors received no specific funding for this work.

References

1.Jones C, Logan J, Gladden RM, Bohm M. Vital Signs: Demographic and substance use trends among heroin users—United States, 2002–2013. Morbidity and Mortality Weekly Report. 2015;64(26):719–25. [PMC free article] [PubMed] [Google Scholar]
2.Friedman J, Kim D, Schneberk T, Bourgoins P, Shin M, Celious A, et al. Assessment of racial/ethnic and income disparities in the prescription of opioids and other controlled medications in California. JAMA Internal Medicine. 2019;E-publication ahead of print; Published online on February 11, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Martins S, Santaella-Tenorio J, Marshall B, Maldonado A, Cerda M. Racial/ethnic differences in trends in heroin use and heroin-related risk behaviors among nonmedical prescription opioid users. Drug and Alcohol Dependence. 2015;151(1):278–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Serdarevic M, Striley C, Cottler L. Gender differences in prescription opioid use. Current Opinion in Psychology. 2018;30(4):238–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Compton W, Jones C, Baldwin G. Relationship between nonmedical prescription-opioid use and heroin use. The New England Journal of Medicine. 2016;374:154–63. 10.1056/NEJMra1508490 [DOI] [PubMed] [Google Scholar]
6.Rigg K, Monnat S. Urban vs. rural differences in prescription opioid misuse among adults in the United States: Informing region specific drug policies and interventions. International Journal of Drug Policy. 2015;26(5):484–91. 10.1016/j.drugpo.2014.10.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Keyes K, Cerda M, Brady J, Havens J, Galea S. Understanding the rural-urban differences in nonmedical prescription opioid use and abuse in the United States. American Journal of Public Health. 2014;104(2):e52–e9. 10.2105/AJPH.2013.301709 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Rudd R, Paulozzi L, Bauer M, Burleson R, Carlson R, Dao D, et al. Increases in heroin overdose deaths—28 states, 2010–2012. Morbidity and Mortality Weekly Report. 2014;63(39):849–54. [PMC free article] [PubMed] [Google Scholar]
9.Evans E, Kelleghan A, Li L, Min J, Huang D, Urada D, et al. Gender differences in mortality among treatment opioid depdent patients. Drug and Alcohol Dependence. 2015;155(1):228–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Henry J. Kaiser Family Foundation. Opioid Overdose Deaths by Race/Ethnicity. San Francisco, CA: 2017. [Google Scholar]
11.Minnesota Department of Health. Race rate disparity in drug overdose deaths. St. Paul, MN: Minnesota Department of Health, 2019. [Google Scholar]
12.Boutwell A, Nijhawan A, Zaller N, Rich J. Arrested on heroin: A national opportunity. Journal of Opioid Management. 2007;3:328–32. 10.5055/jom.2007.0021 [DOI] [PubMed] [Google Scholar]
13.Darke S, Mills K, Ross J, Williamson A, Havard A, Teesson M. The aging heroin user: Career length, clinical profile and outcomes across 36 months. Drug and Alcohol Review. 2009;28(3):243–9. 10.1111/j.1465-3362.2008.00024.x [DOI] [PubMed] [Google Scholar]
14.Becker W, Sullivan L, Tetrault J, Desai R, Fiellin D. Non-medical use, abuse and dependence on prescription opioids among U.S. adults: Psychiatric, medical and substance use correlates. Drug and Alcohol Dependence. 2008;94(1–3):38–47. 10.1016/j.drugalcdep.2007.09.018 [DOI] [PubMed] [Google Scholar]
15.Wakeman S. Why it’s inappropriate not to treat incarcerated patients with opioid agonist therapy. AMA Journal of Ethics. 2017;19(9):922–30. 10.1001/journalofethics.2017.19.9.stas1-1709 [DOI] [PubMed] [Google Scholar]
16.Connery HS. Medication-assisted treatment of opioid use disorder: Review of the evidence and future directions. Harvard Review of Psychiatry. 2015;23(2):63–75. 10.1097/HRP.0000000000000075 [DOI] [PubMed] [Google Scholar]
17.Corsi K, Lehman W, Booth R. The effect of methadone maintenance on positive outcomes for opiate injection drug users. Journal of Substance Abuse Treatment. 2009;37(2):120–6. 10.1016/j.jsat.2008.11.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Corsi K, Kwiatkowski D, Booth R. Predictors of postive outcomes for out-of-treatment opiate injectors recruited into methadone maintenance through street outreach. Journal of Drug Issues. 2002;32(3):999–1016. [Google Scholar]
19.Maglione M, Raaen L, Chen C, Azhar G, Shahidinia N, Shen M, et al. Effects of medication assisted treatment (MAT) for opioid use disroder on functional outcomes: A systematic review. Journal of Substance Abuse Treatment. 2018;89:28–51. 10.1016/j.jsat.2018.03.001 [DOI] [PubMed] [Google Scholar]
20.Alderks C. Trends in the use of methadone, buprenorphine, and extended-release naltrexone at substance abuse treatment facilities: 2003–2015 (update). Bethesda, MD: Substance Abuse and Mental Health Services Administration, 2017. [PubMed] [Google Scholar]
21.Krawczyk N, Feder KA, Fingerhood MI, Saloner B. Racial and ethnic differences in opioid agonist treatment for opioid use disorder in a U.S. national sample. Drug and Alcohol Dependence. 2017;178:512–8. 10.1016/j.drugalcdep.2017.06.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Hansen H, Siegel C, Wanderling J, DiRocco D. Buprenorphine and methadone treatment for opioid dependence by income, ethnicity and race of neighborhoods in New York City. Drug and Alcohol Dependence. 2016;164(1):14–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Lundgren L, Amodeo M, Ferguson F, Davis K. Racial and ethnic differences in drug treatment entry of injection drug users in Massachusetts. Journal of Substance Abuse Treatment. 2001;21(3):145–53. 10.1016/s0740-5472(01)00197-0 [DOI] [PubMed] [Google Scholar]
24.Lee C, Ayers S, Kronenfeld J. The association between perceived provider discrimination, health care utilization, and health status in racial and ethnic minorities. Ethnicity & Disease. 2009;19(3):330–7. [PMC free article] [PubMed] [Google Scholar]
25.Williams D, Wyatt R. Racial bias in health care and health: Challenges and opportunities. JAMA. 2015;314(6):555–6. 10.1001/jama.2015.9260 [DOI] [PubMed] [Google Scholar]
26.Earnshaw V, Smith L, Copenhaver M. Drug addiction stigma in the context of methadone maintenance therapy: An investigation into understudied sources of stigma. International Journal of Mental Health and Addiction. 2013;11(1):110–22. 10.1007/s11469-012-9402-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.McKnight C, Shumway M, Masson C, Pouget E, Jordan A, Des Jarlais D, et al. Perceived discrimination among racial and ethnic minority drug users and the association with health care utilization. Journal of Ethnicity in Substance Abuse. 2017;16(4):404–19. 10.1080/15332640.2017.1292418 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Anstice S, Strike C, Brands B. Supervised methadone consumption: Client issues and stigma. Substance Use and Misuse. 2009;44(6):794–808. 10.1080/10826080802483936 [DOI] [PubMed] [Google Scholar]
29.Conner K, Rosen D. "You’re nothing but a junkie": Multiple experiences of stigma in an aging methadone maintenance population. Journal of Social Work Practice in the Addictions. 2008;8(2):244–64. [Google Scholar]
30.Olsen Y, Sharfstein J. Confronting the stigma of opioid use disorder—and its treatment. JAMA. 2014;311(14):1393–4. 10.1001/jama.2014.2147 [DOI] [PubMed] [Google Scholar]
31.D’Aunno T, Pollack H. Changes in methadone in treatment practices: Results from a national panel study, 1988–2000. Journal of the American Medical Association. 2002;288(7):850–6. 10.1001/jama.288.7.850 [DOI] [PubMed] [Google Scholar]
32.Pollack H, D’Aunno T. Dosage patterns in methadone treatment: Results from a national survey, 1988–2005. Health Services Research. 2008;43(6):2143–63. 10.1111/j.1475-6773.2008.00870.x [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Venner K, Donovan D, Campbell A, Wendt D, Rieckmann T, Radin S, et al. Future directions for medication assisted treatment for opioid use disorder with American Indian/Alaska Natives. Addictive Behaviors. 2018;86:111–7. 10.1016/j.addbeh.2018.05.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Volkow N, Jones E, Einstein E, Wargo E. Prevention and treatment of opioid misuse and addiction: A review. JAMA Psychiatry. 2019;76(2):208–16. 10.1001/jamapsychiatry.2018.3126 [DOI] [PubMed] [Google Scholar]
35.Grant B, Chu A, Sigman R, Amsbary M, Kali J, Sugawara Y, et al. Source and Accuracy Statement: National Epidemiologic Survey on Alcohol and Related Conditions-III (NESARC-III). Bethesda, MA: National Institute on Alcohol Abuse and Alcoholism, 2014. [Google Scholar]
36.SAS Institute Inc. Cary, NC: SAS Institute Inc., 2013, Cary, NC.
37.Hosmer D, Lemeshow S. Applied Logistic Regression. 3 ed Hoboken, NJ: John Wiley & Sons, Inc.; 2013. [Google Scholar]
38.FitzGerald C, Hurst S. Implicit bias in healthcare professionals: A systematic review. BMC Medical Ethics. 2017;18(19):1–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Hansen H, Roberts S. Two Tiers of Biomedicalization: Methadone, Buprenorphine, and the Racial Politics of Addiction Treatment. Netherland J, Editor: Emerald Group Publishing Limited; 2012. [Google Scholar]
40.Lagisetty P, Ross R, Bohnert A, Clay M, Maust D. Buprenorphine treatment divide by race/ethnicity and payment. JAMA psychiatry. 2019; E-publication ahead of print, May 08 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Rowan-Szal G, Chatham L, Joe G, Simpson D. Services provided during methadone treatment: A gender comparison. Journal of Substance Abuse Treatment. 2000;19(1):7–14. 10.1016/s0740-5472(99)00091-4 [DOI] [PubMed] [Google Scholar]
42.Kelly S, O’Grady K, Mitchell S, Brown B, Schwartz R. Predictors of methadone treatment retention from a multi-site study: A survival analysis. Drug and Alcohol Dependence. 2011;117(2–3):170–5. 10.1016/j.drugalcdep.2011.01.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Hser Y, Anglin D, Liu Y. A survival analysis of gender and ethnic differences in responsiveness to methadone maintenance therapy. International Journal of the Addictions. 1990;25:1295–315. 10.3109/10826089009068465 [DOI] [PubMed] [Google Scholar]
44.Jimenez-Trevino L, Saiz P, Barcia-Portilla M, Diaz-Mesa E, Sanchez-Lasheras F, Buron P, et al. A 25-year follow-up of patients admitted to methadone treatment for the first time: Mortality and gender differences. Addictive Behaviors. 2011;36(12):1184–90. 10.1016/j.addbeh.2011.07.019 [DOI] [PubMed] [Google Scholar]
45.Ngo H, Tait R, Hulse G. Hospital psychiatric comorbidity and its role in heroin dependence treatment outcomes using naltrexone implant or methadone maintenance. Journal of Psychopharmacology. 2011;25(6). [DOI] [PubMed] [Google Scholar]
46.Conway K, Compton W, Stinson F, Grant B. Lifetime comorbidity of DSM-IV mood and anxiety disorers and specific drug use disorders: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Clinical Psychiatry. 2006;67(2):247–57. 10.4088/jcp.v67n0211 [DOI] [PubMed] [Google Scholar]
47.Roncero C, Barral C, R-C L, Perez-Pazos J, Martinez-Luna N, Casas M, et al. Psychiatric comorbidities in opioid-dependent patients undergoing a replacement therapy programme in Spain: The PROTEUS Study. Psychiatry Research. 2016;243:174–81. 10.1016/j.psychres.2016.06.024 [DOI] [PubMed] [Google Scholar]
48.Martinez-Luna N, Rodriguez-Cintas L, Esojo A, Palma-Alvarez R, Robles-Martinez M, Grau-Lopez L, et al. Harm reduction program use, psychopathology and medical severity in patients with methadone maintenance treatment. Adicciones. 2018;30(3):197–207. 10.20882/adicciones.897 [DOI] [PubMed] [Google Scholar]
49.Syme V, Browne A, Varcoe C, Josewski V. Harm reduction, methadone maintenance treatment and the root causes of health and social inequities: An intersectional lens in the Canadian context. Harm Reduction Journal. 2011;8(17):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Burgess D, Ding Y, Hargreaves M, van Ryn M, Phelan S. The association between perceived discrimination an dunderutilization of needed medical and mental health care in a multi-ethnic community sample. Journal of Health Care for the Poor and Underserved. 2008;19(3):894–911. 10.1353/hpu.0.0063 [DOI] [PubMed] [Google Scholar]
51.Bailey Z, Krieger N, Agenor M, Graves J, Linos N, Bassett M. Structural racism and health inequities in the USA: Evidene and interventions. The Lancet. 2017;389(10077):1453–63. [DOI] [PubMed] [Google Scholar]
52.Abraham A, Andrews C, Grogan C, D’Aunno T, Humphreys K, Pollack H, et al. The Affordable Care Act transformation of substance use disorder treatment. American Journal of Public Health. 2017;107(1):31–2. 10.2105/AJPH.2016.303558 [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Chou R, Korthuis P, Weimer M, Bougatsos C, Blazina I, Zakher B, et al. Medication-assisted treatment models of care for opioid use disorder in primary care settings. Technical Briefs, Agency for Healthcare Research and Quality; 2016;28. [PubMed] [Google Scholar]
54.Duvivier H, Gustafson S, Grueutman M, Jangchup T, Harden A, Reinhard A, et al. Indian Health Service pharmacists engaged in opioid safety initiatives and expanding access to naloxone. Journal of the American Pharmacists Association. 2017;57(2):S135–S40. [DOI] [PubMed] [Google Scholar]
55.Tipps R, Buzzard G, McDougall J. The opioid epidemic in Indian Country. The Journal of Law, Medicine & Ethics. 2018;46(2):422–36. [DOI] [PubMed] [Google Scholar]
56.Novins D, Aarons G, Conti S, Dahlke D, Daw R, Fickenscher A, et al. Use of evidence base in substance abuse treatment programs for American Indians and Alaska natives: Pursuing quality in the crucible of practice and policy. Implementation Science. 2011;6(63):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Redden S, Tracy S, Shafer M. A metaphor analysis of recovering substance abusers’ sensemaking of medication-assisted treatment. Qualitative Health Research. 2013;23(7):951–62. 10.1177/1049732313487802 [DOI] [PubMed] [Google Scholar]
58.Walls M, Johnson K, Whitbeck L, Hoyt D. Mental health and substance abuse services preferences among American Indian people of the Northern Midwest. Community Mental Health Journal. 2006;42(6):521–35. 10.1007/s10597-006-9054-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Walls M, Gonzalez J, Gladney T, Onello E. Unconscious biases: Racial microaggressions in American Indian health care. The Journal of the American Board of Family Medicine. 2015;28(2):231–9. 10.3122/jabfm.2015.02.140194 [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Rieckmann T, Farentinos C, Tillotson C, Kocarnik J, McCarty D. The substance abuse counseling workforce: Education, preparation and certification. Journal of Substance Abuse. 2011;32(4):180–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Jernigan V, Peercy M, Branam D, Saunkeah B, Wharton D, Winkleby M, et al. Beyond health equity: Achieving wellness within American Indian and Alaska Native communities. American Journal of Public Health. 2015;105(S3):S376–S9. [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0228755.r001

Decision Letter 0

Rodrigo Marín-Navarrete

Transfer Alert

This paper was transferred from another journal. As a result, its full editorial history (including decision letters, peer reviews and author responses) may not be present.

25 Jul 2019

PONE-D-19-17408

Interaction effects in the association between methadone maintenance therapy and experiences of racial discrimination in healthcare settings

PLOS ONE

Dear Dr. Pro,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The manuscript which you submitted to PLOS ONE, has been reviewed. The reviewers comments are included at the bottom of this letter. The reviewers would like to see major revisions made to your manuscript before the final decision. Therefore, I invite you to respond to the reviewers comments as soon as posible.

We would appreciate receiving your revised manuscript by Sep 08 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Rodrigo Marín-Navarrete, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We noticed that you have chosen the subsection category “[FOR JOURNAL STAFF USE ONLY]” for your manuscript. Unfortunately, this is not a valid category. At this time, please choose one or more subsections that best represent the topic(s) of your study.

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thank you for the invitation to review this interesting manuscript.

The issue of discrimination, including racial discrimination, and its relationship with access to quality health care is an important one, and the NESARC dataset is a comprehensive and well-respected source. I offer the following comments in the hope of assisting the authors in fine tuning the manuscript.

The major source of non-clarity in the manuscript is the assumption that participation in MMT is the source of the discrimination being reported by people. In this paper, lifetime MMT participation is being effectively used as a proxy for having, or having had, an opioid use disorder (OUD). The discussion of drug-based stigma in the manuscript does not explain this and so is somewhat convoluted with discussion moving back and forth between the related but not equivalent issues of drug use, MMT and OUD. There is a literature that discusses the stigma attached to substance use disorders, and the idea that intra-group discrimination exists such that people who inject drugs (PWID) are discriminated against by other users of illicit drugs. Those who have ever participated in MMT are likely to be a subset of PWID and/or a subset of people with OUD.

*In order to differentiate between MMT participation per se and OUD, the author’s model would need to have adjusted for OUD, rather than excluding it from the covariates chosen.* I believe this is a significant issue with the model presented.

The links being sought between a lifetime experience of MMT and racial discrimination make less sense than potential links between identification of a person as someone with an OUD, or a PWID, and racial-based discrimination. It is also possible that the discrimination perceived as racial may in fact be an expression of substance-based discrimination – this concept is also relevant but not explored. There is also a literature that discusses the bi-directional nature of the relationship between substance use and discrimination. This is particularly important for this manuscript, as the relationships between discrimination and MMT participation/uptake are, rightly, of concern.

1. Background section

The literature discussed is entirely North American. This should be acknowledged in the text; differences in patterns of use and misuse vary world-wide.

Language used should be precise to avoid confusion, particularly in discussions of complex, inter-related topics. A number of examples are given below.

The term ‘drug abuse’ is not generally accepted in other regions. ‘Misuse’ is considered more acceptable.

The word ‘prevalence’ should be used in place of ‘incidence’ for heroin and opioid misuse, as the paper referenced does not measure incidence.

Similarly, poly substance use is a risk for overdose, rather than opioid misuse; the factors listed are linked to prevalence of misuse, rather than attributed to those factors, s the causal linkages may be bidirectional for many of them.

Paragraph 2 starts talking about rates of OD deaths, but gives an example of the rate of change in the number of OD deaths. These concepts are related but not interchangeable.

Similarly in paragraph 3, the authors switch between discussions of MMT (specifically methadone) and other MATs (medication assisted treatments, which include other OSTs).

In paragraph 4, “experiences of racial discrimination ….more common… racial/ethnic minorities”… but a comparison group is not cited.

McNight’s paper apparently found that discrimination differed between groups, but which groups?

In paragraph 5 (line 87) is it possible the text should read “differences in the effects on MMT of discrimination? The following sentence/s refers to the link between race and receipt of appropriate MMT.

In the final paragraph of the Background, it is somewhat tautological that racial/ethnic minority status would be associated with racial discrimination.

Methods:

Variables in NESARC describe substance use disorders etc, rather than ‘addressing’ them. Stratum is a singular noun – perhaps should be strata.

In line 136, perhaps use gender rather than ‘sex’, to avoid confusion with ‘sexual minority status’, which is presumed to refer to sexual orientation.

In the Analysis section, why was a cut-off of 0.2 used for the p-values for inclusion in the model, rather than the usual 0.05?

Results:

Does Table 2 represent a single multivariable model, adjusted for all listed variables and covariates? If so, why does line 200 in the Discussion refer to ‘models’?

Discussion:

In general, a note that the relationships explored in this analysis are associations, and causality cannot be assumed. Therefor (line 201) one variable should not be discussed as having an effect on another.

Line 208: the Fitzgerald article describes occasions of bias, rather than trends (changes).

Line 210: …”people with marginalized characteristics” should read "marginalizing characteristics"

Paragraph 3 of the Discussion draws some long bows without references, e.g. the experience of discrimination negatively affecting compliance (there are references for this) and the “worsening of MMT comorbidities” (unspecified) which may refer to comorbid conditions going untreated if people avoid healthcare in general after experiencing discrimination. Line 223 may read more clearly as “..the convergence in AI/AN communities of the opioid epidemic, multiple barriers… “ etc as the ‘opioid epidemic’ is not specific to those communities.

The final statements in this paragraph seem to suggest that treatment of conditions comorbid with OUDs must precede treatment for the OUDs: this is an outdated concept. There is considerable literature suggesting that it is more effective to address multiple co-occurring conditions in parallel or in integrated fashion, rather than sequentially.

The final comments relate to challenges faced in service delivery for AI/AN communities. There is something of a leap in stating that the findings from this study justify initiatives targeting Native American populations, however essential those initiatives are.

The Limitations section should include a statement noting that relationships cannot be construed as causal, and that data are from the USA only and may not generalise to other settings.

The Conclusion refers to issues that are worthy of attention but extrapolates somewhat from the findings, as it appears to infer that it is racial discrimination that is the major barrier to health service uptake, particularly OST, rather than this as perhaps an expression of the multiple other stigma associated with illicit drug use.

Finally - some attention should be paid to accuracy and consistency of formatting of the references.

Reviewer #2: This is article represents an important contribution to the literature on discrimination of patients with OUD. This paper demonstrates independent effect of Methadone Maintenance Therapy on racial / discrimination in general health care settings. Authors do a great job in explaining the objectives and the methods to support their hypotheses.

Some minor comments:

Background:

The authors could include some of the following references as part of the citations of the effects on the positive effects of medication-assisted treatment.

Ma J, Bao YP, Wang RJ, et al. Effects of medication-assisted treatment on mortality among opioids users: a systematic review and meta-analysis. Molecular psychiatry. 2018.

Maglione MA, Raaen L, Chen C, et al. Effects of medication assisted treatment (MAT) for opioid use disorder on functional outcomes: A systematic review. Journal of substance abuse treatment. 2018;89:28-51.

Methods:

It would be important to explain / operationalize racial discrimination in healthcare settings- do these include discrimination in all types of health care settings including substance abuse specialty treatment centers, that is centers where methadone might be dispense or does it exclude it? A clear explanation of what healthcare settings includes can help in interpreting the findings and understanding their full implication.

Discussion:

A stronger justification towards the findings in AI/AN is warranted.The statement “our strong findings of racial/discrimination in healthcare settings among AI/AN may be driven, in part, by AI/AN biases inherent in non-IHS services” requires a citation and a broader explanation to contextualize the findings.

Limitations:

Authors should acknowledge that the dataset ( NESARC- III) is from 2012-2013, and while racial discrimination in healthcare settings by individuals on MMT might be pervasive and not easy to change, mulitple national initiatives towards the opioid epidemic and education towards evidence based treatments within mainstream of care might shift the results observed, should we look at current data.

Reviewer #3: The paper could be accepted. However, a major review must be done and after that should be reviewed again.

General comments

There is some typos/ deleted references f.e MMT vs. no MMT [ref[), see all the mns.

The expression “mental health disorder diagnosis” should be change by Dual Disorders, in the text, tables, etc for a review of the use of this term see Szerman et al. Rethinking Dual disorders/pathology. Addictive Disorders & Their Treatment. 2013; 12 (1): 1–10 doi: 10.1097/ADT.0b013e31826e7b6a

It is necessary to use the term Latino/Latina? Or Latino includes all?

Abstract

It is clear according the result and discussion, but should be modified if there is new comments or results.

Introduction

Read: McKnight et al. Perceived discrimination among racial and ethnic minority drug users and the association with health care utilization. J Ethn Subst Abuse. 2017;16(4):404-419. doi: 10.1080/15332640.2017.1292418.

Methods & Results:

Page 6 “Our outcome of interest was a binary indicator of whether an individual experienced 120 racial discrimination in a healthcare setting in the past year (yes or no). Response options to 121 discrimination survey questions included very often, fairly often, sometimes, almost never, and 122 never. Our binary variable included those who endorsed no experience (never) or any experience…”

I suppose that that no racial discriminations means that the answer is “never” and yes means “very often, fairly often, sometimes, almost never”, in any of the questions, please clarify this sentence.

Could you show if there are differences between patients on MMT coming from heroin-dependence or opiate pain-killer dependence?

Discussion

I think that a short sentence summarizing all your relevant results it will do the reading more clear. After that you can discuss your results.

The presence of Dual disorders is high as expected is close to 60% last year, according other research among opiate-dependent patients see Roncero et al. Psychiatric comorbidities in opioid-dependent patients undergoing a replacement therapy programme in Spain: The PROTEUS study. Psychiatry Res. 2016 Sep 30;243:174-81. doi:10.1016/j.psychres.2016.06.024. and Martínez-Luna et al. Harm reduction program use, psychopathology and medical severity in patients with methadone maintenance treatment. Adicciones. 2018 Jan 15;30(3):197-207. doi: 10.20882/adicciones.897.

Do you any date or any comments if there is an association between dual disorders / raze-ethnicity and discrimination??

Could you speculate if experiences of racial discrimination were more likely among any raze of dual disorders opiate-dependent? Should be done an special intervention or screening on discrimination in this patients?

It seems that there is no discrimination according sex, it is important to point out this result.

Could you provide any idea for the clinicians on the fact of how avoid or deal with this discrimination

Bibliography

I suggest read and consider to included in the mns this papers, that I have mentioned before.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2020 Feb 6;15(2):e0228755. doi: 10.1371/journal.pone.0228755.r002

Author response to Decision Letter 0

15 Aug 2019

Please also see our 'Response to Reviewers' document, uploaded as part of this resubmission.

Rodrigo Marín-Navarrete, PhD, Academic Editor

PLOS ONE

National Institute of Psychiatry

Mexico City, Mexico

Dear Dr. Marín-Navarrete

Thank you for providing us with an opportunity to revise our manuscript, “Interaction effects in the association between methadone maintenance therapy and experiences of racial discrimination in healthcare settings”. The feedback provided by our three reviewers was encouraging and insightful.

Broadly, our major revisions had to do with clarifying how our hypotheses informed our model construction and covariate selection, more clearly narrating the implications of our results in the context of our strong odds ratio estimates among American Indians/Alaska Natives in the discussion section, and summarizing suggestions for public health practice without extrapolating beyond our findings in the conclusion.

Below we summarize our responses to the critiques. We have tracked changes in the revised manuscript and also provided a clean, untracked version (line numbers given throughout our response are in reference to the tracked changes version). Again, we thank you and the reviewers for improving the clarity and precision of our research.

Sincerely,

George Pro, PhD (Corresponding Author)

Postdoctoral Fellow

Center for Health Equity Research

george.pro@nau.edu http://nau.edu/CHER

(928).523.4267

General feedback

Reviewer 1

Thank you for your thoughtful and insightful review. As our paper deals with multiple related factors around OUD, MMT, and racial discrimination, we want to be careful in articulating how our hypotheses informed our model creation and selection of covariates.

Using NESARC-III data, we restricted the dataset to include only individuals who had a lifetime OUD (heroin use disorder or other opioid use disorder, n=766). Among our sample of individuals with any OUD, MMT participation is not necessarily a proxy indicator of OUD status, but rather a subset of those with an OUD. In this sense, OUD could not be excluded as a covariate, but instead defined our analytic sample. Our rationale for restricting our sample to those with OUD was that MMT participation would be unlikely among those without an OUD and somewhat of an anomaly in the dataset. Indeed, in a post-hoc analysis of the full, unrestricted dataset conducted for this review, less than 0.1% of those without an OUD received MMT. Thus, we restricted our dataset to include only those with an OUD, rather than adjusting for OUD using the full NESARC-III dataset. We have added a description of this rationale and the process we underwent to arrive at our final analytic sample in the methods section. (Lines 128-131)

We agree that intravenous opioid use is an important factor in the development of stigma and bias in healthcare settings and we appreciate the reviewer’s comment. We are discussing a future manuscript that addresses both racial and gender disparities in treatment outcomes among those who inject any illicit drug.

While we believe that MMT can certainly be considered a proxy for intravenous opioid use or OUD more generally, we note that there is substantial stigma associated with MMT itself (i.e., it is often viewed as substituting one drug for another, and is often associated with a poorer class of drug users). Therefore, we believe that MMT is itself an important factor contributing to racial bias and thus warrants investigation separate from intravenous drug use and/or OUD.

There is also a literature that discusses the bi-directional nature of the relationship between substance use and discrimination. This is particularly important for this manuscript, as the relationships between discrimination and MMT participation/uptake are, rightly, of concern.

We appreciate the feedback about the bidirectional nature of substance misuse and multiple factors related to discrimination and treatment services uptake. We have addressed this concern in our revisions, specifically in response to Comment #4.

As suggested by a previous comment, a future study addressing the relationship between discrimination and MMT may reverse the independent and dependent variables used in the current study. For example, in such a follow-up study, we will specifically examine the interaction between discrimination (our current outcome) and race/ethnicity and its association with MMT (our current predictor) as an outcome. However, while important, the strategy of looking at multiple combinations of independent and dependent variables as suggested by this review was beyond the scope of the current study design. In our conclusion, we address many of the complexities inherent in interpreting our findings, particularly the challenge of interpreting one expression of bias (racial discrimination) in the midst of other latent factors that may also affect how racial discrimination is both expressed and perceived in healthcare settings.

Reviewer 2

This is article represents an important contribution to the literature on discrimination of patients with OUD. This paper demonstrates independent effect of Methadone Maintenance Therapy on racial / discrimination in general health care settings. Authors do a great job in explaining the objectives and the methods to support their hypotheses.

Thank you for this acknowledgement.

Reviewer 3

The paper could be accepted. However, a major review must be done and after that should be reviewed again.

We appreciate your thorough review. We address each of your critiques below, and in the process have substantially strengthened our manuscript.

There is some typos/ deleted references f.e MMT vs. no MMT [ref[), see all the mns.

Our use of “[ref]” in the description of categorical variables refers to the referent group chosen for analyses of categorical variables. We have spelled out the word ‘referent’ in place of ‘ref’ so as not to be confused with a placeholder for a missing literature reference. (Lines 161-168)

Thank you for helping us use the correct terminology of dual disorders. We have replaced ‘mental health disorder diagnosis’ with ‘dual disorders’ throughout the manuscript text and tables.

It is necessary to use the term Latino/Latina? Or Latino includes all?

There are several ways in which social scientists label the Latino/Latina group in the U.S. We opted for Latino/Latina as it includes both male (Latino) and female (Latina) versions of this label, and generally refers to people descended from populations in Latin America. Other common labels include Hispanic (people who speak Spanish and come from Spanish-speaking populations) and Latinx (a non-gendered version of Latino/Latina).

Introduction

Reviewer 1

The literature discussed is entirely North American. This should be acknowledged in the text; differences in patterns of use and misuse vary world-wide.

We have reiterated throughout the text that both our literature review and data are U.S.-based. We also added an indication of our U.S. focus in the manuscript title.

The term ‘drug abuse’ is not generally accepted in other regions. ‘Misuse’ is considered more acceptable.

Thank you for pointing out these important differences. We have replaced ‘abuse’ with ‘misuse’, except in circumstances where we are referencing the work of others who used the term ‘abuse’ in their own work. (Line 49)

The word ‘prevalence’ should be used in place of ‘incidence’ for heroin and opioid misuse, as the paper referenced does not measure incidence.

We have replaced ‘incidence’ with ‘prevalence’ in our references to both the Jones et al. (2015) and Krawczyk et al. (2017) papers. (Lines 49 and 81)

Thank you for clarifying the inaccuracies in this sentence. We have changed ‘attributed’ to ‘linked’, and restructured this sentence to more accurately reflect the relationships between factors linked to misuse. (Lines 51-55)

Paragraph 2 starts talking about rates of OD deaths, but gives an example of the rate of change in the number of OD deaths. These concepts are related but not interchangeable.

In support of our statement that overdose death rates are higher among non-Whites in some states, we have added a Kaiser Family Foundation citation to illustrate state-level racial differences in the rate of overdose deaths. We also clarified the use of rate versus rate of change. (Lines 59-63)

Similarly in paragraph 3, the authors switch between discussions of MMT (specifically methadone) and other MATs (medication assisted treatments, which include other OSTs).

Thank you for pointing out the difference between MMT and MAT. We have made clear that MMT is the focus of this study, but is also only one type of MAT. The NESARC-III dataset only includes data on methadone-specific maintenance therapy. We further justified our use of citations that include different types of MAT. In the context of this MMT study, better understanding disparities in both MMT and other MAT types is useful in the overall narrative of OUD treatment disparities and discrimination in healthcare settings. (Lines 68-72)

In paragraph 4, “experiences of racial discrimination ….more common… racial/ethnic minorities”… but a comparison group is not cited.

We have added that the comparison being made is between racial/ethnic minority groups and Whites. (Line 85)

McNight’s paper apparently found that discrimination differed between groups, but which groups?

We have included additional findings from the McKnight paper, specifically that Latino/Latina drug users were the group least likely to perceive institutional racism in healthcare settings. (Lines 89-90)

In paragraph 5 is it possible the text should read “differences in the effects on MMT of discrimination? The following sentence/s refers to the link between race and receipt of appropriate MMT.

Thank you for pointing out the confusing structure of this sentence and paragraph. American Indian/Alaska Natives are excluded from OUD disparities research in general, and MMT research in particular, often due to small sample sizes or unavailable data for this group. We have clarified what we believe to be the important takeaway message in this paragraph – although there are few MMT studies that compare AI/AN to other groups, we do see some reports that some AI/AN MMT clients have reported institutional discrimination in the extant literature. (Lines 102-104)

In the final paragraph of the Background, it is somewhat tautological that racial/ethnic minority status would be associated with racial discrimination.

Thank you for pinpointing the ambiguous nature of this part of our hypothesis. We added language to clarify our hypothesized association between race/ethnicity and discrimination – i.e., compared to Whites, we hypothesized that racial/ethnic minority groups would have higher odds of racial discrimination in healthcare settings, and the odds ratio estimates would vary in magnitude by racial group. (Lines 115-118)

Reviewer 2

The authors could include some of the following references as part of the citations of the effects on the positive effects of medication-assisted treatment.

Ma J, Bao YP, Wang RJ, et al. Effects of medication-assisted treatment on mortality among opioids users: a systematic review and meta-analysis. Molecular psychiatry. 2018. Maglione MA, Raaen L, Chen C, et al. Effects of medication assisted treatment (MAT) for opioid use disorder on functional outcomes: A systematic review. Journal of substance abuse treatment. 2018;89:28-51.

Thank you for bringing these references to our attention. We have added a reference to the positive effect of MAT on functional outcomes (Maglione et al., 2018). (Line 77)

The meta-analysis by Ma et al. (2018) indeed demonstrates a positive effect of MAT on reduced mortality. However, only four out of 30 studies included in their meta-analysis were based in the United States. We previously responded to Reviewer 1’s concern (Comment #1) that our data source is entirely based in North America, and that disparities in OUD and MAT utilization vary worldwide. Our response to Reviewer 1 was that we have made it clear throughout our manuscript that our literature review, data, and discussion are all in the context of North America. In this sense, a reference to Ma et al. may be out of place within our focus on MAT in the US.

Reviewer 3

Thank you for reiterating the importance of the McKnight et al. (2017) article. In response to a comment by Reviewer 1 above (Comment #8), we have added to our introduction the specific findings of racial/ethnic differences in perceived institutional racial discrimination. Specifically, Latino/Latina drug users were the least likely to perceive institutional racism in health service settings. (Lines 89-90)

Methods

Reviewer 1

Variables in NESARC describe substance use disorders etc, rather than ‘addressing’ them.

We have changed ‘addressing’ to ‘describing’. (Line 125)

Stratum is a singular noun – perhaps should be strata.

Indeed, in this sentence we intended to use a plural noun. We have changed ‘stratum’ to ‘strata’. (Line 127)

In line 136, perhaps use gender rather than ‘sex’, to avoid confusion with ‘sexual minority status’, which is presumed to refer to sexual orientation.

While we do our best to recognize the differences in the application of ‘sex’ versus ‘gender’, the NESARC-III questionnaire asks respondents to report their sex, including only two response options of male or female. Therefore, we have opted to continue using ‘sex’ for this study, instead of ‘gender’, in order to more accurately reflect the survey questionnaire.

We also clarified the construction of our sexual minority status variable, so as to avoid any confusion with other covariates. Survey respondents were asked about their sexual orientation, with response options including heterosexual or straight [referent], or gay/lesbian, bisexual, or not sure. We categorized respondents who identified as non-heterosexual as sexual orientation minorities.

In the Analysis section, why was a cut-off of 0.2 used for the p-values for inclusion in the model, rather than the usual 0.05?

While there are many useful strategies for variable selection in multivariable regression models, we followed the strategy of ‘purposeful selection of covariates’ outlined in the textbook Applied Logistic Regression by David Hosmer and Stanley Lemeshow (2013). In using this approach, we recognize that a cutoff of the traditional α=0.05 level may fail to identify variables known to be important, and that may demonstrate a significant degree of confounding in the presence of other covariates. At the same time, a threshold of α=0.20 is low enough as to avoid overfitting a model. We have included a description of this approach in our methods section, as well as a reference for the source material. (Lines 175-177)

Reviewer 2

The two NESARC-III survey questions that ask about racial discrimination in healthcare settings do not specify the specific type of healthcare setting. We have added a statement in this paragraph that the type of healthcare setting was not specified in these survey questions, leaving the interpretation of what constitutes a healthcare setting up to the survey respondent. Additionally, we restructured our description of the contents of these survey questions for clarity. (Lines 140-149)

Reviewer 3

Thank you for helping us clarify how we constructed our discrimination variable. We have added that our ‘any’ discrimination variable level includes very often, fairly often, sometimes, or almost never. Also, please see our response to Reviewer 2 (Comment #17), which also addresses strategies to more clearly articulate the construction of this variable.

Could you show if there are differences between patients on MMT coming from heroin-dependence or opiate pain-killer dependence?

We appreciate your inquiry into the differences between heroin and opioid painkiller dependence, as our current sample is restricted to anyone who had a lifetime heroin or other opioid use disorder.

In recognition of the clinical and behavioral differences between the two drug types, we opted to include an additional covariate in our analysis, indicating whether an individual had a heroin use disorder only, an opioid use disorder only, or both. We added the prevalence and chi-square estimates to Table 1 (Line 191). We found that the specific opioid use disorder type was not significantly associated with past-year experiences of racial discrimination in healthcare settings (x2 = 0.26, p = 0.8801). This variable did not meet our selection criteria for inclusion in our fitted model, leaving our fitted model unchanged.

Results

Reviewer 1

Does Table 2 represent a single multivariable model, adjusted for all listed variables and covariates? If so, why does line 200 in the Discussion refer to ‘models’?

Thank you for catching this typo. Table 2 represents a single multivariable model with an interaction term. We have changed ‘models’ to ‘model’. (Line 230)

Discussion

Reviewer 1

Thank you for helping us correct this mistake. We have changed ‘effect of’ to ‘association between’. (Line 228)

Line 208: the Fitzgerald article describes occasions of bias, rather than trends (changes).

We have changed ‘trends in implicit biases’ to ‘occasions of implicit bias’. (Line 239)

Line 210: …”people with marginalized characteristics” should read "marginalizing characteristics"

Thank you. We have changed ‘marginalized’ to ‘marginalizing’. (Line 242)

We have added two references in support of our claim that discrimination negatively effects compliance, including Lee et al. (2009) and Burgess et al. (2008). Including these references more effectively grounds our findings in the extant literature. We also added to this point by suggesting that treatment addressing MMT comorbidities may be delayed or foregone as a result of prior experiences with discrimination. (Lines 254-256)

Line 223 may read more clearly as “..the convergence in AI/AN communities of the opioid epidemic, multiple barriers… “ etc as the ‘opioid epidemic’ is not specific to those communities.

Thank you for helping to clarify this sentence. We have changed ‘the convergence of the opioid epidemic in AI/AN communities…’ to ‘the convergence in AI/AN communities of the opioid epidemic…’. (Line 258)

We have removed the outdated citations and replaced them with references that support the modern way practitioners think about OUD treatment and integrated behavioral health settings. The Abraham et al. (2017) article discusses the importance of integrated care for the treatment of OUD, as well as how the Affordable Care Act has been key in moving treatment services forward towards more integrated service delivery models. We also included a citation by Chou et al. (2016), which outlines strategies for improved care of people with OUDs in integrated primary care settings. (Lines 260-266)

We agree that promoting disease prevention initiatives among AI/AN is important, but concluding that our study justifies the existence of those initiatives is indeed beyond the scope of our findings. Rather than remove this section entirely, we thought that an alternative framing around AI/AN populations would serve as an appropriate end to the discussion section. Instead of stating that our findings justify such AI/AN-focused programs, we added that our findings may inform future research into specific programs addressing MAT, health services, and improved health within AI/AN communities. (Lines 279-285)

Reviewer 2

A stronger justification towards the findings in AI/AN is warranted. The statement “our strong findings of racial/discrimination in healthcare settings among AI/AN may be driven, in part, by AI/AN biases inherent in non-IHS services” requires a citation and a broader explanation to contextualize the findings.

We have added two articles as references to support our statement that our AI/AN findings may be driven by biases in non-IHS settings (Bailey et al., 2017; Walls et al., 2015). Both articles demonstrate racial and institutional baises experienced by AI/AN. Bailey et al. looked at health equities in the US in general but highlighted inequities particularly relevant to AI/AN. Walls et al. looked at racial microaggressions experienced by AI/AN, and demonstrated correlations between microaggressive/discriminatory behaviors towards AI/AN and poor health outcomes. We have also added some additional explanation to contextualize these references, including an examination of IHS services coverage among AI/AN and the potential for a high risk of experiencing discriminatory behavior in non-IHS services. (Lines 277-281)

Reviewer 3

I think that a short sentence summarizing all your relevant results it will do the reading more clear. After that you can discuss your results.

Thank you for helping us begin our discussion section with a succinct description of our main finding. We have reorganized this first paragraph to begin with summarizing our main finding, then delving into specific interaction findings with a focus on our strong AI/AN results. (Lines 223-233)

Do you any date or any comments if there is an association between dual disorders / raze-ethnicity and discrimination??

Could you speculate if experiences of racial discrimination were more likely among any raze of dual disorders opiate-dependent? Should be done and special intervention or screening on discrimination in this patients?

Thank you for pointing out the high prevalence of past-year dual mental health diagnoses among OUD treatment participants. In our fitted model and among our covariates, the odds of experiences of racial discrimination were highest among those with dual mental health diagnoses (versus none) compared to the estimates of all other covariates. Clearly, dual diagnoses play an integral role in MMT services utilization and healthcare-based discrimination. We have added a line in the discussion that reiterates the importance of our descriptive finding. (Lines 251-252)

Yes, it is possible to look at differences in the association between discrimination and dual disorder status, dependent on race/ethnicity. One approach to solve this problem would be adding a second interaction term to our fitted model, which would be race/ethnicity by dual diagnosis status (similar in structure to our original race/ethnicity by MMT interaction). However, while this is an important research question addressing a gap in the MMT disparities literature, we believe that the inclusion of a second interaction term would distract from the stated scope of this paper – i.e., differences in associations between MMT and discrimination within racial/ethnic groups. Instead, we plan on pursuing a future manuscript that will look at disparities in MMT utilization based on dual disorder status.

Limitations

Reviewer 1

The Limitations section should include a statement noting that relationships cannot be construed as causal, and that data are from the USA only and may not generalise to other settings.

Thank you for pointing out some of the inherent limitations of this North American-based, cross-sectional dataset. We have added a paragraph articulating the limits of causal inference, as well as our study’s limited generalizability. (Lines 297-300)

Reviewer 2

Authors should acknowledge that the dataset ( NESARC- III) is from 2012-2013, and while racial discrimination in healthcare settings by individuals on MMT might be pervasive and not easy to change, multiple national initiatives towards the opioid epidemic and education towards evidence based treatments within mainstream of care might shift the results observed, should we look at current data.

We have included a section that points out the possibility of different findings if a similar or replicated study were based on more current data. (Lines 300-304)

Conclusion

Reviewer 1

Thank you for your insight into making our conclusion stronger. We have restructured this section to better reflect the findings and implications of our study. In particular, we noted the challenges of interpreting one expression of stigma (racial discrimination) in the midst of other unobserved or latent factors that may also affect how racial discrimination is both expressed and perceived in healthcare settings. (Lines 311-323)

Finally - some attention should be paid to accuracy and consistency of formatting of the references.

We have reviewed and updated our reference list for accuracy and consistency, and to the best of our knowledge all of our references follow the PLOS ONE guidelines, specifically using the Vancouver citation style.

Also, in response to Reviewer 3’s general feedback, we have changed the way we refer to reference groups for categorical variables in the methods section, so as to avoid being mistaken as a placeholder for a missing citation.

PLoS One. doi: 10.1371/journal.pone.0228755.r003

Decision Letter 1

Stéphanie Baggio

9 Dec 2019

PONE-D-19-17408R1

Interaction effects in the association between methadone maintenance therapy and experiences of racial discrimination in U.S. healthcare settings

PLOS ONE

Dear Dr. Pro,

We would appreciate receiving your revised manuscript by Jan 23 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

We look forward to receiving your revised manuscript.

Kind regards,

Stéphanie Baggio

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

Unfortunately, two of the previous reviewers were unable to assess the manuscript again. I assessed it myself and I also added some additional (minor) comments.

1. Please report p<.001 (three decimals) in the abstract, results, and tables.

2. Even if the study’s methods have been published elsewhere, the response rate should be mentioned.

3. Which kind of survey weights did you use?

4. More information is needed for the measures: a reference for the racial discrimination scale and which kind of diagnosis (DSM-5?) for addictive disorders. This is especially important because the definition changed between DSM-IV and DSM-5.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #3: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #3: The paper could be accepted but yhe authors dosen´t respond all my concerns. However, a moderate review must be done and after that should be reviewed again.

This paper provides some evidence on the role of race and ethnicity on discrimination among opiate-dependent patients in MMT. Twenty-two percent of the sample experienced racial discrimination in a healthcare setting. This is a relevant an a not well-studied topic and the findings of the paper should be taking account for the future. However, there is a lack of some relevant factors that must be detail and explain. The authors confirm my guess that the dual-diagnosis patients have more discrimination. Authors doesn´t replay all my concerns and comments so I think that the paper should be reviewed again.

General comments

Abstract

It is clear according the result and discussion, but should be modified if there is new comments or results.

Introduction

That´s ok

Methods & Results:

Could you show if there are differences between patients on MMT coming from heroin-dependence or opiate pain-killer dependence?

“We appreciate your inquiry into the differences between heroin and opioid painkiller

dependence, as our current sample is restricted to anyone who had a lifetime heroin or

other opioid use disorder.”

The pain killers are opioids so I guess that you want to said.

“We appreciate your inquiry into the differences between heroin and opioid painkiller

dependence, as our current sample is restricted to anyone who had a lifetime heroin or

other (illegal) opioid use disorder.” Please if it Ii right change explain it in the paper.

Discussion

The first short sentence summarizing all your relevant results it is very clear.

I agree with the new sentence, but, should be point out that confirm previous findings, r please add the references and the idea that it is expected

For example, results from our descriptive analyses indicated that, as expected (references..) , nearly 60% of OUD treatment participants also had a dual mental health diagnosis in the past year.

Your results “Experiences of racial discrimination were more likely among …those with a past-year mental health) diagnosis (p<0.0001)”. I believe that is an important factor an any comments should be added on this topic. Do you any date or any comments if there is an association between dual disorders / raze-ethnicity and discrimination?? Could you speculate if experiences of racial discrimination were more likely among any raze of dual disorders opiate-dependent? Should be done an special intervention or screening on discrimination in this patients?

Your Response “… it is possible to look at differences in the association between discrimination and dual disorder status, dependent on race/ethnicity. One approach to solve this problem

would be adding a second interaction term to our fitted model, which would be race/ethnicity by dual diagnosis status (similar in structure to our original race/ethnicity by MMT interaction). However, while this is an important research question addressing a gap in the MMT disparities literature, we believe that the inclusion of a second interaction term would distract from the stated scope of this paper – i.e., differences in associations between MMT and discrimination within racial/ethnic groups. Instead, we plan on pursuing a future manuscript that will look at disparities in MMT utilization based on dual disorder status. “

I understand that due to the space it is not possible to show here the dates that you have. I encourage you to do the second paper¡¡, because the dual-disorders patients for sure are suffering the double-discrimination and double-stigma and it is important to show it to the scientific and clinicians community. But at least in the paper it important to describe the relevance of this topic and in the last paragraph to add something as in the future the interaction between the racial and dual-disorders diseases should be studied… or something like that.

You should analyze not only the dual-disorders, if it is possible ( according the “n” of the sample, psychotic versus affective disorders vs personality disorders… obviously depending of the prevalence in your sample

It seems that there is no discrimination according sex, it is important to point out this result.

Conclusions:

Take in account my previous suggestions

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

PLoS One. 2020 Feb 6;15(2):e0228755. doi: 10.1371/journal.pone.0228755.r004

Author response to Decision Letter 1

3 Jan 2020

METHODS

1. Even if the study’s methods have been published elsewhere, the response rate should be mentioned.

We have included a sentence describing the household survey response rate of 72%.

2. Which kind of survey weights did you use?

We have clarified that we used survey weights pre-calculated by NESARC which are designed to calibrate the sampling distribution for major subgroups defined by region, sex, age, and race/ethnicity.

3. More information is needed for the measures: a reference for the racial discrimination scale and which kind of diagnosis (DSM-5?) for addictive disorders. This is especially important because the definition changed between DSM-IV and DSM-5.

We have clarified that the reference group for our binary discrimination indicator is ‘no’. We have also included additional text describing the pre-defined NESARC diagnostic variables based on DMS-5 criteria.

4. Could you show if there are differences between patients on MMT coming from heroin-dependence or opiate pain-killer dependence?

Our sample is restricted to only those who meet the criteria for heroin use disorder or illicit opioid use disorder, based on DSM-5 criteria. Our analysis includes a covariate differentiating between opioid use disorder, illicit opioid use disorder, or both. In our bivariate analyses (Table 1), we found no association between disorder type and discrimination in a healthcare setting (x2=2.45, p=0.458). Thus, this variable was not carried forward to our fitted model. While the association between disorder type and discrimination may vary by MMT status, as you suggest, we had no theoretical or algorithmic justification to explore this relationship as disorder type demonstrated a very weak relationship with our outcome.

RESULTS

5. Please report p<.001 (three decimals) in the abstract, results, and tables.

We have changed the decimal place of our p-values to only include three decimal places.

DISCUSSION

6. The presence of Dual disorders is high as expected is close to 60% last year, according other research among opiate-dependent patients see Roncero et al. Psychiatric comorbidities in opioid-dependent patients undergoing a replacement therapy programme in Spain: The PROTEUS study. Psychiatry Res. 2016 Sep 30;243:174-81. doi:10.1016/j.psychres.2016.06.024. and Martínez-Luna et al. Harm reduction program use, psychopathology and medical severity in patients with methadone maintenance treatment. Adicciones. 2018 Jan 15;30(3):197-207. doi: 10.20882/adicciones.897. You should point out that confirm previous findings, r please add the references and the idea that it is expected.

Thank you for helping us provide more context around our strong finding of dual mental health disorders and experiences of discrimination. We have cited the articles by Roncero, et al. and Martinez-Luna, et al.

7. Your results “Experiences of racial discrimination were more likely among those with a past-year mental health diagnosis (p<0.0001)” are an important factor and any comments should be added on this topic. Do you any date or any comments if there is an association between dual disorders / raze-ethnicity and discrimination? I understand that due to the space it is not possible to show here the dates that you have. I encourage you to do the second paper!!

Thank you for encouraging us to pursue this topic further. A second paper addressing dual mental health diagnoses is forthcoming. We have added comments and citations to highlight the importance of mental health comorbidities in among our target population.

8. It is important to describe the relevance of dual mental health disorders, and in the last paragraph to add something as in the future the interaction between the racial and dual-disorders diseases should be studied… or something like that.

We agree that this is an important topic for future research. We have added text outlining a possible future research trajectory which could be designed to answer questions about the relationships between race/ethnicity, mental health, stigma, discrimination, and MMT services utilization.

9. It seems that there is no discrimination according sex, it is important to point out this result.

In our fitted model, we found that women have lower odds of experiencing racial discrimination, compared to men (aOR=0.51, p<0.001). We have added a paragraph highlighting this finding, and provided additional citations to contextualize these results within the larger body of extant literature addressing differential MMT outcomes by gender.

PLoS One. doi: 10.1371/journal.pone.0228755.r005

Decision Letter 2

Stéphanie Baggio

23 Jan 2020

Interaction effects in the association between methadone maintenance therapy and experiences of racial discrimination in U.S. healthcare settings

PONE-D-19-17408R2

Dear Dr. Pro,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Stéphanie Baggio

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #3: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

PLoS One. doi: 10.1371/journal.pone.0228755.r006

Acceptance letter

Stéphanie Baggio

30 Jan 2020

PONE-D-19-17408R2

Interaction effects in the association between methadone maintenance therapy and experiences of racial discrimination in U.S. healthcare settings

Dear Dr. Pro:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Stéphanie Baggio

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

[pone.0228755.ref001] 1.Jones C, Logan J, Gladden RM, Bohm M. Vital Signs: Demographic and substance use trends among heroin users—United States, 2002–2013. Morbidity and Mortality Weekly Report. 2015;64(26):719–25. [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref002] 2.Friedman J, Kim D, Schneberk T, Bourgoins P, Shin M, Celious A, et al. Assessment of racial/ethnic and income disparities in the prescription of opioids and other controlled medications in California. JAMA Internal Medicine. 2019;E-publication ahead of print; Published online on February 11, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref003] 3.Martins S, Santaella-Tenorio J, Marshall B, Maldonado A, Cerda M. Racial/ethnic differences in trends in heroin use and heroin-related risk behaviors among nonmedical prescription opioid users. Drug and Alcohol Dependence. 2015;151(1):278–83. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref004] 4.Serdarevic M, Striley C, Cottler L. Gender differences in prescription opioid use. Current Opinion in Psychology. 2018;30(4):238–46. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref005] 5.Compton W, Jones C, Baldwin G. Relationship between nonmedical prescription-opioid use and heroin use. The New England Journal of Medicine. 2016;374:154–63. 10.1056/NEJMra1508490 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref006] 6.Rigg K, Monnat S. Urban vs. rural differences in prescription opioid misuse among adults in the United States: Informing region specific drug policies and interventions. International Journal of Drug Policy. 2015;26(5):484–91. 10.1016/j.drugpo.2014.10.001 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref007] 7.Keyes K, Cerda M, Brady J, Havens J, Galea S. Understanding the rural-urban differences in nonmedical prescription opioid use and abuse in the United States. American Journal of Public Health. 2014;104(2):e52–e9. 10.2105/AJPH.2013.301709 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref008] 8.Rudd R, Paulozzi L, Bauer M, Burleson R, Carlson R, Dao D, et al. Increases in heroin overdose deaths—28 states, 2010–2012. Morbidity and Mortality Weekly Report. 2014;63(39):849–54. [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref009] 9.Evans E, Kelleghan A, Li L, Min J, Huang D, Urada D, et al. Gender differences in mortality among treatment opioid depdent patients. Drug and Alcohol Dependence. 2015;155(1):228–35. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref010] 10.Henry J. Kaiser Family Foundation. Opioid Overdose Deaths by Race/Ethnicity. San Francisco, CA: 2017. [Google Scholar]

[pone.0228755.ref011] 11.Minnesota Department of Health. Race rate disparity in drug overdose deaths. St. Paul, MN: Minnesota Department of Health, 2019. [Google Scholar]

[pone.0228755.ref012] 12.Boutwell A, Nijhawan A, Zaller N, Rich J. Arrested on heroin: A national opportunity. Journal of Opioid Management. 2007;3:328–32. 10.5055/jom.2007.0021 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref013] 13.Darke S, Mills K, Ross J, Williamson A, Havard A, Teesson M. The aging heroin user: Career length, clinical profile and outcomes across 36 months. Drug and Alcohol Review. 2009;28(3):243–9. 10.1111/j.1465-3362.2008.00024.x [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref014] 14.Becker W, Sullivan L, Tetrault J, Desai R, Fiellin D. Non-medical use, abuse and dependence on prescription opioids among U.S. adults: Psychiatric, medical and substance use correlates. Drug and Alcohol Dependence. 2008;94(1–3):38–47. 10.1016/j.drugalcdep.2007.09.018 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref015] 15.Wakeman S. Why it’s inappropriate not to treat incarcerated patients with opioid agonist therapy. AMA Journal of Ethics. 2017;19(9):922–30. 10.1001/journalofethics.2017.19.9.stas1-1709 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref016] 16.Connery HS. Medication-assisted treatment of opioid use disorder: Review of the evidence and future directions. Harvard Review of Psychiatry. 2015;23(2):63–75. 10.1097/HRP.0000000000000075 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref017] 17.Corsi K, Lehman W, Booth R. The effect of methadone maintenance on positive outcomes for opiate injection drug users. Journal of Substance Abuse Treatment. 2009;37(2):120–6. 10.1016/j.jsat.2008.11.004 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref018] 18.Corsi K, Kwiatkowski D, Booth R. Predictors of postive outcomes for out-of-treatment opiate injectors recruited into methadone maintenance through street outreach. Journal of Drug Issues. 2002;32(3):999–1016. [Google Scholar]

[pone.0228755.ref019] 19.Maglione M, Raaen L, Chen C, Azhar G, Shahidinia N, Shen M, et al. Effects of medication assisted treatment (MAT) for opioid use disroder on functional outcomes: A systematic review. Journal of Substance Abuse Treatment. 2018;89:28–51. 10.1016/j.jsat.2018.03.001 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref020] 20.Alderks C. Trends in the use of methadone, buprenorphine, and extended-release naltrexone at substance abuse treatment facilities: 2003–2015 (update). Bethesda, MD: Substance Abuse and Mental Health Services Administration, 2017. [PubMed] [Google Scholar]

[pone.0228755.ref021] 21.Krawczyk N, Feder KA, Fingerhood MI, Saloner B. Racial and ethnic differences in opioid agonist treatment for opioid use disorder in a U.S. national sample. Drug and Alcohol Dependence. 2017;178:512–8. 10.1016/j.drugalcdep.2017.06.009 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref022] 22.Hansen H, Siegel C, Wanderling J, DiRocco D. Buprenorphine and methadone treatment for opioid dependence by income, ethnicity and race of neighborhoods in New York City. Drug and Alcohol Dependence. 2016;164(1):14–21. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref023] 23.Lundgren L, Amodeo M, Ferguson F, Davis K. Racial and ethnic differences in drug treatment entry of injection drug users in Massachusetts. Journal of Substance Abuse Treatment. 2001;21(3):145–53. 10.1016/s0740-5472(01)00197-0 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref024] 24.Lee C, Ayers S, Kronenfeld J. The association between perceived provider discrimination, health care utilization, and health status in racial and ethnic minorities. Ethnicity & Disease. 2009;19(3):330–7. [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref025] 25.Williams D, Wyatt R. Racial bias in health care and health: Challenges and opportunities. JAMA. 2015;314(6):555–6. 10.1001/jama.2015.9260 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref026] 26.Earnshaw V, Smith L, Copenhaver M. Drug addiction stigma in the context of methadone maintenance therapy: An investigation into understudied sources of stigma. International Journal of Mental Health and Addiction. 2013;11(1):110–22. 10.1007/s11469-012-9402-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref027] 27.McKnight C, Shumway M, Masson C, Pouget E, Jordan A, Des Jarlais D, et al. Perceived discrimination among racial and ethnic minority drug users and the association with health care utilization. Journal of Ethnicity in Substance Abuse. 2017;16(4):404–19. 10.1080/15332640.2017.1292418 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref028] 28.Anstice S, Strike C, Brands B. Supervised methadone consumption: Client issues and stigma. Substance Use and Misuse. 2009;44(6):794–808. 10.1080/10826080802483936 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref029] 29.Conner K, Rosen D. "You’re nothing but a junkie": Multiple experiences of stigma in an aging methadone maintenance population. Journal of Social Work Practice in the Addictions. 2008;8(2):244–64. [Google Scholar]

[pone.0228755.ref030] 30.Olsen Y, Sharfstein J. Confronting the stigma of opioid use disorder—and its treatment. JAMA. 2014;311(14):1393–4. 10.1001/jama.2014.2147 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref031] 31.D’Aunno T, Pollack H. Changes in methadone in treatment practices: Results from a national panel study, 1988–2000. Journal of the American Medical Association. 2002;288(7):850–6. 10.1001/jama.288.7.850 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref032] 32.Pollack H, D’Aunno T. Dosage patterns in methadone treatment: Results from a national survey, 1988–2005. Health Services Research. 2008;43(6):2143–63. 10.1111/j.1475-6773.2008.00870.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref033] 33.Venner K, Donovan D, Campbell A, Wendt D, Rieckmann T, Radin S, et al. Future directions for medication assisted treatment for opioid use disorder with American Indian/Alaska Natives. Addictive Behaviors. 2018;86:111–7. 10.1016/j.addbeh.2018.05.017 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref034] 34.Volkow N, Jones E, Einstein E, Wargo E. Prevention and treatment of opioid misuse and addiction: A review. JAMA Psychiatry. 2019;76(2):208–16. 10.1001/jamapsychiatry.2018.3126 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref035] 35.Grant B, Chu A, Sigman R, Amsbary M, Kali J, Sugawara Y, et al. Source and Accuracy Statement: National Epidemiologic Survey on Alcohol and Related Conditions-III (NESARC-III). Bethesda, MA: National Institute on Alcohol Abuse and Alcoholism, 2014. [Google Scholar]

[pone.0228755.ref036] 36.SAS Institute Inc. Cary, NC: SAS Institute Inc., 2013, Cary, NC.

[pone.0228755.ref037] 37.Hosmer D, Lemeshow S. Applied Logistic Regression. 3 ed Hoboken, NJ: John Wiley & Sons, Inc.; 2013. [Google Scholar]

[pone.0228755.ref038] 38.FitzGerald C, Hurst S. Implicit bias in healthcare professionals: A systematic review. BMC Medical Ethics. 2017;18(19):1–18. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref039] 39.Hansen H, Roberts S. Two Tiers of Biomedicalization: Methadone, Buprenorphine, and the Racial Politics of Addiction Treatment. Netherland J, Editor: Emerald Group Publishing Limited; 2012. [Google Scholar]

[pone.0228755.ref040] 40.Lagisetty P, Ross R, Bohnert A, Clay M, Maust D. Buprenorphine treatment divide by race/ethnicity and payment. JAMA psychiatry. 2019; E-publication ahead of print, May 08 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref041] 41.Rowan-Szal G, Chatham L, Joe G, Simpson D. Services provided during methadone treatment: A gender comparison. Journal of Substance Abuse Treatment. 2000;19(1):7–14. 10.1016/s0740-5472(99)00091-4 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref042] 42.Kelly S, O’Grady K, Mitchell S, Brown B, Schwartz R. Predictors of methadone treatment retention from a multi-site study: A survival analysis. Drug and Alcohol Dependence. 2011;117(2–3):170–5. 10.1016/j.drugalcdep.2011.01.008 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref043] 43.Hser Y, Anglin D, Liu Y. A survival analysis of gender and ethnic differences in responsiveness to methadone maintenance therapy. International Journal of the Addictions. 1990;25:1295–315. 10.3109/10826089009068465 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref044] 44.Jimenez-Trevino L, Saiz P, Barcia-Portilla M, Diaz-Mesa E, Sanchez-Lasheras F, Buron P, et al. A 25-year follow-up of patients admitted to methadone treatment for the first time: Mortality and gender differences. Addictive Behaviors. 2011;36(12):1184–90. 10.1016/j.addbeh.2011.07.019 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref045] 45.Ngo H, Tait R, Hulse G. Hospital psychiatric comorbidity and its role in heroin dependence treatment outcomes using naltrexone implant or methadone maintenance. Journal of Psychopharmacology. 2011;25(6). [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref046] 46.Conway K, Compton W, Stinson F, Grant B. Lifetime comorbidity of DSM-IV mood and anxiety disorers and specific drug use disorders: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Clinical Psychiatry. 2006;67(2):247–57. 10.4088/jcp.v67n0211 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref047] 47.Roncero C, Barral C, R-C L, Perez-Pazos J, Martinez-Luna N, Casas M, et al. Psychiatric comorbidities in opioid-dependent patients undergoing a replacement therapy programme in Spain: The PROTEUS Study. Psychiatry Research. 2016;243:174–81. 10.1016/j.psychres.2016.06.024 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref048] 48.Martinez-Luna N, Rodriguez-Cintas L, Esojo A, Palma-Alvarez R, Robles-Martinez M, Grau-Lopez L, et al. Harm reduction program use, psychopathology and medical severity in patients with methadone maintenance treatment. Adicciones. 2018;30(3):197–207. 10.20882/adicciones.897 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref049] 49.Syme V, Browne A, Varcoe C, Josewski V. Harm reduction, methadone maintenance treatment and the root causes of health and social inequities: An intersectional lens in the Canadian context. Harm Reduction Journal. 2011;8(17):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref050] 50.Burgess D, Ding Y, Hargreaves M, van Ryn M, Phelan S. The association between perceived discrimination an dunderutilization of needed medical and mental health care in a multi-ethnic community sample. Journal of Health Care for the Poor and Underserved. 2008;19(3):894–911. 10.1353/hpu.0.0063 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref051] 51.Bailey Z, Krieger N, Agenor M, Graves J, Linos N, Bassett M. Structural racism and health inequities in the USA: Evidene and interventions. The Lancet. 2017;389(10077):1453–63. [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref052] 52.Abraham A, Andrews C, Grogan C, D’Aunno T, Humphreys K, Pollack H, et al. The Affordable Care Act transformation of substance use disorder treatment. American Journal of Public Health. 2017;107(1):31–2. 10.2105/AJPH.2016.303558 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref053] 53.Chou R, Korthuis P, Weimer M, Bougatsos C, Blazina I, Zakher B, et al. Medication-assisted treatment models of care for opioid use disorder in primary care settings. Technical Briefs, Agency for Healthcare Research and Quality; 2016;28. [PubMed] [Google Scholar]

[pone.0228755.ref054] 54.Duvivier H, Gustafson S, Grueutman M, Jangchup T, Harden A, Reinhard A, et al. Indian Health Service pharmacists engaged in opioid safety initiatives and expanding access to naloxone. Journal of the American Pharmacists Association. 2017;57(2):S135–S40. [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref055] 55.Tipps R, Buzzard G, McDougall J. The opioid epidemic in Indian Country. The Journal of Law, Medicine & Ethics. 2018;46(2):422–36. [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref056] 56.Novins D, Aarons G, Conti S, Dahlke D, Daw R, Fickenscher A, et al. Use of evidence base in substance abuse treatment programs for American Indians and Alaska natives: Pursuing quality in the crucible of practice and policy. Implementation Science. 2011;6(63):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref057] 57.Redden S, Tracy S, Shafer M. A metaphor analysis of recovering substance abusers’ sensemaking of medication-assisted treatment. Qualitative Health Research. 2013;23(7):951–62. 10.1177/1049732313487802 [DOI] [PubMed] [Google Scholar]

[pone.0228755.ref058] 58.Walls M, Johnson K, Whitbeck L, Hoyt D. Mental health and substance abuse services preferences among American Indian people of the Northern Midwest. Community Mental Health Journal. 2006;42(6):521–35. 10.1007/s10597-006-9054-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref059] 59.Walls M, Gonzalez J, Gladney T, Onello E. Unconscious biases: Racial microaggressions in American Indian health care. The Journal of the American Board of Family Medicine. 2015;28(2):231–9. 10.3122/jabfm.2015.02.140194 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref060] 60.Rieckmann T, Farentinos C, Tillotson C, Kocarnik J, McCarty D. The substance abuse counseling workforce: Education, preparation and certification. Journal of Substance Abuse. 2011;32(4):180–90. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0228755.ref061] 61.Jernigan V, Peercy M, Branam D, Saunkeah B, Wharton D, Winkleby M, et al. Beyond health equity: Achieving wellness within American Indian and Alaska Native communities. American Journal of Public Health. 2015;105(S3):S376–S9. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Interaction effects in the association between methadone maintenance therapy and experiences of racial discrimination in U.S. healthcare settings

George Pro

Nick Zaller

Roles

Abstract

Background

Methods

Findings

Conclusion

Background

Methods

Data source

Variables

Analysis

Results

Table 1. Weighted descriptive characteristics of individuals with lifetime opioid use disorders (NESARC-III, 2012–2013) (survey n = 766; weighted population n = 5,276,507).

Table 2. Multivariable logistic regression modeling past-year experience with racial discrimination in healthcare settings among individuals with lifetime opioid use disorders (NESARC-III, 2012–2013) (survey n = 766; weighted population n = 5,276,507).

Discussion

Limitations

Conclusion

Data Availability

Funding Statement

References

Decision Letter 0

Rodrigo Marín-Navarrete

Roles

Transfer Alert

Author response to Decision Letter 0

Decision Letter 1

Stéphanie Baggio

Roles

Author response to Decision Letter 1

Decision Letter 2

Stéphanie Baggio

Roles

Acceptance letter

Stéphanie Baggio

Roles

Associated Data

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases