Help seeking behaviors of Latinos with substance use disorders who perceive a need for treatment: substance abuse versus mental health treatment services

Abstract

Objective:

Perceiving a need for substance abuse treatment is a strong predictor of substance abuse treatment utilization among those with substance use disorders (SUD). Studies have also found that when persons with SUD perceive a need for treatment, they are more likely to use mental health treatment rather than substance abuse treatment. Substance abuse treatment utilization is low among Latinos, relative to other racial/ethnic groups. It is unknown if Latinos with SUD who perceive a need for treatment are more likely to use mental health or substance abuse treatment.

Methods:

Data were pooled from the National Survey on Drug Use and Health survey years 2014, 2015, 2016, and 2017. Analyses were limited to adult Latino participants who met DSM-IV criteria for a past-year substance use disorder (n=1,889). Multinomial logistic regression analyses examined the role of perceived treatment need on the past-year use of (1) no treatment, (2) substance abuse treatment only, and (3) mental health treatment only. Important covariates included socio-demographics, problem severity, and poor mental health status.

Results:

Only 5% of Latinos with SUD reported perceiving a need for treatment. Treatment utilization was also low: 83% reported not using any treatment in the past-year. In multinomial logistic regressions, compared to not using any treatment, Latinos with SUD who perceived a need for substance abuse treatment were more likely to report using mental health treatment only. Perceiving a need for treatment and using substance abuse treatment services only was not statistically different from not using any treatment.

Conclusions:

Findings underscore the need for better integration of substance abuse and mental health treatment services to address the health needs of Latinos.

1. Introduction

Latinos with substance use disorders (SUD) use substance abuse treatment services at low rates. National estimates indicate that only about 3-12% of Latinos with SUD seek help (Guerrero et al., 2013; Pinedo et al., 2018; Vaeth et al., 2017). Latinos are also less likely to use treatment services than other racial/ethnic groups (Guerrero et al., 2013; Pinedo et al., 2018). Importantly, the recent passage of Affordable Care Act (ACA) and the Mental Health Parity & Addiction Equity Act have significantly increased coverage, accessibility, and affordability to substance abuse treatment services. Recent research has found that health insurance rates among Latinos, including those with SUD, have increased since the passage of the ACA (Creedon and Cook, 2016; Manuel, 2017; Pinedo, 2019). Substance abuse treatment services are now more accessible to Latinos, and the general population, than ever. However, recent research also indicates that increased access to treatment services have not resulted in increased use of these services, particularly among Latinos (Ali et al., 2015; Creedon and Cook, 2016; Pinedo, 2019). One national study using 2015-2017 data from the National Survey on Drug use and Health (NSDUH) found that among Latinos with SUD, only 10% reported using a treatment service in the past-year (Pinedo, 2019). Further, among those with health insurance, Latinos were still significantly less likely than their White counterparts to use substance abuse treatment services. This suggests that other barriers (e.g., stigma, cultural barriers) may be at play and expanding coverage alone may not be adequate enough to increase use of treatment services among Latinos and eliminate existing health disparities (Guerrero et al., 2013; Witbrodt et al., 2014).

An important predictor of treatment utilization is having a perceived need for treatment. Those who recognize having a problem with alcohol and/or drug use and perceive a need for treatment are more likely to seek help than those who do not (Ali et al., 2015; Choi et al., 2014; Grella et al., 2009). For example, one national study among persons with SUD found that having health insurance was associated with increased use of substance abuse treatment services, but only if participants perceived a need for treatment (Ali et al., 2015). Participants who did not perceive a need for treatment were less likely to use treatment services regardless if they had health insurance or not. Along these lines, another study examining Latino-White disparities in the use of substance abuse treatment found that disparities were explained by perceived treatment need (Pinedo and Villatoro, under review). Among those who perceived a need for treatment, differences in the use of treatment services between Latinos and Whites with SUD were non-significant. Thus, perceived treatment need is a significant predictor of treatment utilization among Latinos as other racial/ethnic groups.

Further, novel research suggests that when individuals with SUD are ready to seek help, they may avoid substance abuse treatment services all together. One national study among persons with SUD and no mental health disorders (n=18,600) found that those who perceived a need for treatment were significantly more likely to use mental health treatment services rather than a substance abuse treatment program (Creedon and Cook, 2016). Individuals may perceive their SUD as a mental health problem (Khantzian, 1997), or may want to avoid using substance abuse treatment for fear of being stigmatized (Corrigan et al., 2005; Pescosolido et al., 1999; Pinedo et al., 2018). Nonetheless, the majority of studies investigating treatment utilization among persons with SUD typically focus on the use of informal (e.g., mutual help groups) and specialty substance abuse treatment (e.g., in/out-patient services, rehabilitation), and overlook mental health treatment (Guerrero et al., 2013; Witbrodt et al., 2014; Zemore et al., 2014). This scientific gap limits our understanding of the help-seeking behaviors of persons with SUD.

Moreover, it is unknown if Latinos’ help seeking behaviors mirror those of the general population given important cultural and socio-demographic differences. It is worth noting that Latinos also use mental health treatment services at low rates and are less likely to use them than other racial/ethnic groups (Alegria et al., 2002; Wang et al., 2005). Nonetheless, it unknown which services, mental health or substance abuse treatment, Latinos with SUD are more likely to use when they are ready to seek help for an alcohol or drug problem. Such knowledge is warranted to inform tailored strategies aimed at increasing the use of treatment among Latinos in need and who are ready to seek help. To address this critical gap in the knowledge base we undertook the present study using a nationally representative sample of Latinos with SUD. The objective of this study was to investigate if Latinos with SUD who perceive a need for treatment are more likely to use substance abuse treatment or mental health treatment services. Given that Latinos report low utilization of both substance abuse and mental health treatment, in general and relative to other racial/ethnic groups, we have no a prior hypothesis and consider this an exploratory study. We draw on pooled data (2014-2017) from the National Survey on Health and Drugs (NSDUH), which provides a current and well-powered dataset to investigate this research question.

2. Method

2.1. Data source and study population

The NSDUH is a national survey conducted annually by the Substance Abuse and Mental Health Services Administration (SAMHSA) in English and Spanish. Participants are randomly recruited using stratified, multistage, probability sampling methods. A national sampling frame of the 50 states, including the District of Columbia, is used to recruit participants. Those who are 12 years of age or older, non-institutionalized, and reside in the United States are considered eligible for the survey. The survey collects data on socio-demographic characteristics, alcohol and drug use, mental health status, and treatment utilization, among other health-related topics. Additional details about the survey methodology and survey items are publicly available (Center for Behavioral Health Statistics and Quality, 2014, 2015, 2016, 2017). De-identified and publicly available NSDUH data from survey years 2014 to 2017 were pooled for the present analysis. Given the objective of the current study, the analytic sample was limited to adult (i.e., 18 years of age or older) Latinos who met diagnostic criteria for SUD. Additionally, because individuals with SUD with criminal histories are commonly mandated to substance abuse treatment (Lê Cook & Alegría, 2011), we excluded participants who reported currently being on parole or probation to facilitate the interpretation of the findings. Our final analytic sample was 1,889 adult Latino participants with SUD and without criminal histories.

2.2. Measures

The dependent variable for this analysis was the type of treatment used in the past year: no treatment, substance abuse treatment only, or mental health treatment only. All categories were mutually exclusive. We excluded participants who indicated using both mental health and substance abuse treatment services due to a small sample size. Substance abuse treatment services include seeking help from: in/out-patient services from a rehabilitation facility, hospital, or rehabilitation center; private doctor’s office; prison/jail; or mutual self-help groups (e.g., Alcoholics Anonymous, Narcotic Anonymous). Mental health treatment refers to in/out patient services in a clinical or non-clinical setting and the use of any prescription medication to treat a mental or emotional condition.

The primary independent variable of interest was having a perceived need for treatment. This variable was constructed using the following question: “During the past 12 months, did you need treatment or counseling for your alcohol or drug use?” Participants who answered affirmatively to either question were characterized as perceiving a need for treatment. The NSDUH includes variables for participants that met DSM-IV diagnostic criteria for an alcohol use disorder (AUD) or drug use disorder (DUD). Participants who met diagnostic criteria for AUD or DUD were coded as having SUD. More information on how these variables are measured and constructed can be found via their publicly available public use codebook (Center for Behavioral Health Statistics and Quality, 2014, 2015, 2016, 2017).

Problem severity and poor mental health status (i.e., psychological distress) were included as covariates given that they have been strongly associated with perceived treatment need and treatment utilization (Alegria et al., 2002; Ali et al., 2015; Mojtabai et al., 2002; Schmidt et al., 2007). Psychological distress was also specifically included to control for individuals with a co-occurring mental health condition. To construct a variable for problem severity we used the following survey items that specifically focused on social consequences related to family, law, and employment. Participants were asked if they had experienced problems in the past year due to their drinking or drug use (asked separately for each substance), including: (1) problems at home or school (e.g., neglecting their children, missing work or school, doing a poor job at work or school, losing a job or dropping out of school), (2) trouble with the law, (3) problems with friends and family, and (4) continued substance use despite problems with family and friends. All responses were yes/no. Scores ranged from 0 to 21. Responses were added and averaged, resulting in a composite variable where higher scores indicate greater problem severity. The cronbach’s alpha for this variable was 0.82, indicating good internal consistency.

Psychological distress was assessed via the six-item K6 Distress Scale (Kessler et al., 2002). This is a validated scale that measures distressed symptoms, including feelings of hopelessness, restlessness, depressiveness, and worthlessness. The K6 Distress Scale has shown consistency across diverse samples, including Latinos, and is especially preferred when screening for mood or anxiety disorders (Pratt, Dey, & Cohen, 2007; Prochaska, Sung, Max, Shi, & Ong, 2012; Swartz & Lurigio, 2006). Participants self-reported on the frequency of each feeling in the past month with response options ranging from none of the time (0 points) to all the time (4 points). Scores were summed with a possible range between 0 and 24; higher scores indicate higher psychological distress. The cronbach’s alpha for this variable was 0.97, indicating excellent internal consistency.

Lastly, socio-demographic variables included in our analysis included biological sex (male vs. female), age (18-25 years, 26-34 years, 35-49 years, or 50 years and over), marital status (married vs. widowed, divorced, separated, or single), employment status (employed full/part-time vs. unemployed), highest educational attainment (less than high school, graduated high school, some college, or graduated college), annual household income (Less than $20,000, $20,000 - $49,999, $50,000 - $74,999, or $75,000 or more), urbanicity (large metro area, small metro area, or non-metro area), and insurance status (Medicare, Medicaid, military health care, or private health insurance vs. none).

2.3. Analysis

Statistical analyses were conducted using Stata v.15 software and weighted to adjust for the complex survey sampling design. The sample weight for the pooled dataset was created using guidelines provided by the Center for Behavioral Health Statistics and Quality (Center for Behavioral Health Statistics and Quality, 2014, 2015, 2016, 2017). Briefly, combining each individual sample weight year and dividing by the total number of years of combined survey data, in this case four, created the final pooled sample weight. More detailed information on the use of sample weights and sample design variables are publicly available (Center for Behavioral Health Statistics and Quality, 2014, 2015, 2016, 2017). To adjust for the design effects, STATA’s svy commands were used for all analyses. First, we generated weighted descriptive statistics for our sample, which were stratified by perceived treatment need. Variables were tested for bivariate association using chi-square tests and t-tests, for dichotomous and continuous variables, respectively. Following, we conducted a multinomial logistic regression model using past-year treatment type (i.e., no treatment, substance abuse treatment only, mental health treatment only) as the dependent variable. This model controlled for socio-demographic characteristics, problem severity, psychological distress, and perceived treatment need. This model allowed us to explore the association between types of treatment services used when perceiving a need for treatment.

3. Results

Sample characteristics stratified by perceiving (vs. not perceiving) a need for treatment are displayed in Table 1. Overall, 95% of Latinos with SUD reported not perceiving a need for substance abuse treatment; 5% reported perceiving a need for substance abuse treatment. On average, those who perceived a need for treatment were more likely to be between 35-49 years of age, report higher problem severity, and greater psychological distress. Within the context of treatment utilization, the majority of Latinos (83%) reported not having used any treatment service in the past year. Among those who perceived a need for treatment, 32% reported using mental health treatment services only and 13% reported using substance abuse treatment services only. Overall, those who perceived a need for treatment were more likely to have used either treatment service than those who did not.

Table 1.

Weighted percentages of selected sample characteristics of adult Latinos with past-year substance use disorders stratifying by perceived treatment need, National Survey on Drug use and Health, weighted n=1,998,656, unweighted n=1,889, 2014-2017.

	Perceived treatment need
Variable	Total	No (95%)	Yes (5%)	P-value
Male	69%	69%	60%	0.282
Age
18-25 years	34%	35%	13%	0.042
26-34 years	29%	29%	30%
35-49 years	24%	24%	34%
50 years and over	13%	13%	23%
Married	31%	31%	39%	0.405
Employed	57%	58%	50%	0.329
Educational attainment				0.334
Less than high school	22%	22%	19%
Graduated high school	29%	28%	40%
Some college	31%	31%	26%
Graduate college	18%	18%	15%
Total Family Income				0.070
Less than $20,000	21%	20%	32%
$20,000 - $49,999	39%	38%	48%
$50,000 - $74,999	14%	14%	8%
$75,000 or more	26%	27%	12%
Insured	75%	75%	75%	0.915
Urbanicity				0.212
Large metro	68%	68%	55%
Small metro	26%	25%	37%
Non-metro	6%	6%	8%
Mean problem severity score (SD)	1.31 (1.79)	1.23 (1.67)	3.33 (2.98)	≤0.001
Mean psychological distress score (SD)	9.55 (7.17)	9.38 (7.14)	13.58 (6.88)	≤0.001
Past year treatment use				≤0.001
No treatment	83%	85%	54%
Substance abuse treatment only	4%	4%	13%
Mental health treatment only	13%	11(18 %	32%

Table 2 shows results from the multinomial logistic regression model. Results show that when controlling for socio-demographic factors, problem severity, and psychological distress, those who perceived a need for treatment had more than twice the relative risk (Relative Risk Ratio (RRR): 2.47; 95% Confidence Interval (CI): 1.22-6.02) of using mental health treatment services only as compared to those who reported not using any treatment. The association between perceiving a need for treatment and using substance abuse treatment services only, relative to not using any treatment, was non-significant. Additionally, greater problem severity was positively associated with substance abuse treatment, whereas greater psychological distress was positively associated with mental health treatment utilization.

Table 2.

Multinomial logistic regression model examining the role of perceived need for treatment on the types of treatment services used in the past year among adult Latinos with past-year substance use disorders, National Survey on Drug use and Health, weighted n=1,938,731, unweighted n=1,843, 2014-2017.

	Substance abuse treatment only vs. none		Mental health treatment only vs. none
	Relative Risk Ratio	95% Confidence Interval	Relative Risk Ratio	95% Confidence Interval
	Variables
Perceived a need for treatment (vs. not)	2.68	0.48-14.75	2.47*	1.22-6.02
Male	1.38	0.53-3.56	0.84	0.48-1.47
Age
18-25 years (ref)
26-34 years	1.39	0.53-3.67	1.99	1.12-3.52
35-49 years	3.01*	1.26-7.14	1.73	0.84-3.59
50 years and over	3.29	0.97-11.11	3.48*	1.55-7.80
Married	0.70	0.23-2.21	1.23	0.71-2.11
Employed	0.87	0.42-1.79	0.50**	0.29-0.83
Educational attainment
Less than high school (ref)
Graduated high school	0.85	0.35-2.04	1.29	0.74-2.26
Some college	0.65	0.24-1.77	2.07	1.05-4.07
Graduate college	2.32	0.79-6.79	2.57**	1.27-5.20
Total Family Income
Less than $20,000 (ref)
$20,000 - $49,999	0.66	0.29-1.52	0.49*	0.27-0.88
$50,000 - $74,999	0.72	0.22-2.39	0.98	0.43-2.23
$75,000 or more	0.24	0.06-0.98	1.06	0.53-2.11
Insured	0.83	0.39-1.76	1.74*	1.07-2.83
Urbanicity
Large metro (ref)
Small metro	1.00	0.43-2.31	0.94	0.58-1.52
Non-metro	0.78	0.22-2.67	0.86	0.46-1.59
Mean problem severity score (SD)	1.24*	1.03-1.50	0.97	0.85-1.10
Mean psychological distress score (SD)	1.02	0.94-1.11	1.16***	1.13-1.19

4. Discussion

Using a nationally representative data from 2014-2017 among adult Latinos with SUD, this study examined the types of treatment services Latinos use when perceiving a need for substance abuse treatment. Overall, utilization of treatment services among Latinos was low: 83% reported not using any treatment in the past year. Data also indicated that when Latinos with SUD perceive a need for treatment, they favor mental health treatment over substance abuse treatment services. Among those who perceived a need for treatment, 32% reported using mental health treatment only, while 13% reported using substance abuse treatment only. This finding is also aligned with prior research that has found higher use of mental health treatment versus substance abuse treatment among persons with SUD (Ali et al., 2015). However, our study is unique in that no prior work has exclusively examined if Latinos are more likely to use substance abuse or mental health treatment services when they perceive a need for treatment. Thus, our finding is novel and has not been previously documented in the substance abuse treatment utilization literature.

When taken together, about 45% of those who perceived a need for treatment received any treatment (i.e., mental health or substance abuse treatment) in the past year. This prevalence is more than four times the national average (~10%), when only accounting for substance abuse treatment utilization (Ali et al., 2015; Pinedo, 2019). This suggests that treatment utilization is more likely when Latinos recognize needing treatment for an alcohol or drug problem. Perceived treatment need is closely linked to problem recognition, though two separate constructs. Individuals must first recognize having an alcohol or drug problem before they consider or perceive a need for treatment. Regardless, individuals can also recognize having a substance use problem without perceiving that they need treatment. Research has found that problem recognition among those with SUD, including Latinos, is low (Choi et al., 2014; Pinedo et al., 2018). Efforts and initiatives to increase problem recognition, including awareness on the efficacy of treatment services, among Latinos may potentially influence individuals’ perceived need for treatment and encourage use. However, our findings suggest that increasing problem recognition and perceived treatment need may not be sufficient to increase use of substance abuse treatment services.

Why Latinos who perceived a need for alcohol or drug treatment were more likely to use mental health treatment versus substance abuse treatment is not immediately clear. A couple potential explanations may account for this. One, SUDs are commonly perceived as a mental health problem (Khantzian, 1997). Therefore, Latinos who want help may be more inclined to seek out a mental health provider. Second, SUDs are generally viewed as an individual choice, rather than an illness, and more stigmatizing than having a mental health disorder (Crapanzano, Hammarlund, Ahmad, Hunsinger, & Kullar, 2019; Yang, Wong, Grivel, & Hasin, 2017). Population-based studies have found that public opinion towards those with SUD are more negative and harsher than views towards persons with a mental illness. As compared to those with a mental health disorder, those with SUD are viewed as more blameworthy, violent, and dangerous (Corrigan et al., 2005; Pescosolido et al., 1999). Similarly, a qualitative study among persons with SUD found that Latino participants commonly avoided substance abuse treatment services for fear of being stigmatized, despite recognizing a need for treatment (Pinedo et al., 2018). Thus, Latinos with SUD may prefer mental health treatment services to reduce stigma-related fears. It is worth nothing that greater problem severity was associated with substance abuse treatment utilization, whereas greater psychological distress was associated with mental health treatment utilization. It may be that Latinos with co-occurring mental health problems may seek mental health treatment and those experiencing greater substance use-related problems may seek substance abuse treatment. However, the NSDUH does not ask participants regarding reasons for using types of treatment services. Future research should consider investigating reasons behind treatment type preferences to determine if other factors such as availability of services, cost, cultural factors, or language barriers, among other reasons, may explain why Latinos’ prefer mental health treatment over substance abuse treatment when seeking help for an alcohol or drug problem.

Findings from this study need to be interpreted within some limitations. First, it is possible that participants may have underreported their substance use and past year treatment use given the stigma surrounding these topics. Second, due to a lack of data on Latino subgroups (e.g., Mexican, Cuban, Puerto Rican), we were unable to test for any differences between outcomes and independent variables by subgroups. Future studies should explore these differences given that Latinos are not a homogenous group and important differences in socio-demographic characteristics, substance using behaviors, and treatment seeking behaviors have been documented (Black and Markides, 1993; Caetano et al., 2009; Randolph et al., 1998). Despite these limitations, our findings provide novel findings to the evidence base regarding treatment utilization among Latinos with SUD using a well-powered and representative sample.

Latinos are a significant segment of the United States population who are disproportionally impacted by substance use-related disparities (Guerrero et al., 2013; Vaeth et al., 2017; Witbrodt et al., 2014), which are compounded by low utilization rates of treatment services (Pinedo, 2019). Increasing use of efficacious substance abuse treatment services is a viable strategy to reduce existing health disparities. However, given that a substantial proportion of Latinos who sought help were more likely to use mental health treatment services calls for better integration of substance abuse and mental health treatment services, which often operate independently. Such integration can specifically tailor treatment plans to the needs of Latinos, leading to optimal care.

Highlights.

• Only 5% of Latinos with substance use disorders (SUD) reported perceiving a need for treatment.

• Latinos with SUD who perceived a need for treatment were more likely to use mental health treatment only versus substance abuse treatment or no treatment.

• Among Latinos with SUD, perceived treatment need was unrelated to substance abuse treatment utilization.