We used data from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). Briefly, this nationally representative study of the US population was conducted in 2001–2002 and 2004–2005 among non-institutionalized adults (≥18 years of age) residing in households and group quarters. African-Americans, Hispanics, and individuals aged 18–24 years were oversampled. All participants were interviewed at home by trained lay interviewers who received extensive training and supervision. After complete description of the study to the participants, written informed consent was obtained. All procedures, including informed consent, received full ethical review and approval from the U.S. Census Bureau and U.S. Office of Management and Budget. Of the 43,093 respondents in 2001–2002, 34,653 were re-interviewed in 2004–2005. The response rate in 2001–2002 was 81% and 86.7% in 2004–2005 25 . For the present analysis we combined data from 2001–2002 and 2004–2005 and studied individuals with lifetime illegal drug use disorders and their access to treatment. Data were weighted at wave 2 to account for differential loss to follow-up and to be representative of the target population. This analysis includes 34,653 respondents who completed interviews at wave 1 and wave 2.

Participants’ lifetime psychiatric disorders were measured using the NIAAA Alcohol Use Disorder and Associated Disabilities Interview Schedule (AUDADIS), a structured interview designed to measure psychiatric disorders and associated conditions in large scale surveys 17 in 2001–2002 and 2004–2005. To meet criteria for a lifetime psychiatric diagnosis participants had to have a positive diagnosis at either NESARC wave (or both). Lifetime illegal drug use disorders were defined as abuse or dependence meeting DSM-IV criteria 26 for any of the following illegal drugs: opioids, sedatives, tranquilizers, amphetamine, cocaine, inhalants/solvents, hallucinogens, cannabis, and heroin.

Lifetime mood disorders were defined as major depression, dysthymia, manic disorder or hypomanic disorder meeting DSM-IV criteria; lifetime anxiety disorders were defined as panic disorder with or without agoraphobia, agoraphobia with no history of panic disorder, social phobia, specific phobia, or generalized anxiety meeting DSM-IV criteria 26 .

We divided the study population into four groups: 1) illegal drug use disorder without comorbid mood or anxiety disorders (n = 244), 2) illegal drug use disorder with comorbid mood disorder only (n = 264), 3) illegal drug use disorder with comorbid anxiety disorder only (n = 110), 4) illegal drug use disorder with comorbid mood and anxiety disorders (n = 496).

We examined two study outcomes: treatment for illegal drug use disorders among all those with an illegal drug use disorder and perceived unmet need for treatment among those having not received treatment. Service use was ascertained for each drug of abuse by the following question: “Have you ever gone anywhere or seen anyone for a reason that was related in any way to your use of medicines or drugs - a physician, counselor, Narcotics Anonymous, or any other community agency or professional?” Participants who answered positively were additionally asked to describe the type of treatment they received (outpatient, inpatient, detoxification, rehabilitation, social services). Moreover, participants were asked about perceived unmet need for treatment: “Was there ever a time when you thought you should see a doctor, counselor, or other health professional or seek any other help for your drug use, but you didn’t go?” Participants who met criteria for perceived unmet need for treatment were further asked about reasons for not receiving treatment. Variables ascertaining treatment for illegal drug use disorders and perceived unmet need for treatment were dichotomized. For the purpose of our descriptive analyses participants were split into four groups: 1) those who had an illegal drug use disorder and received treatment (n = 350); those with an illegal drug use disorder who did not receive treatment (n = 764); 2) those who did not receive treatment for an illegal drug use disorder and perceived an unmet need for treatment (n = 127); 3) those who did not receive treatment for an illegal drug use disorder and did not perceive an unmet need for treatment (n = 637).

Additionally, we studied access to treatment for illegal drug use disorders in relation to the type and number of drug of abuse.

Socio-demographic covariates included sex (female vs. male), age (35–54, ≥55 years vs. 18–34 years), educational level (<high school degree vs. ≥ high school), marital status (divorced/separated, widowed, never married vs. married), ethnicity (African American, Other non-Caucasian vs. Caucasian) and family income dichotomized using the lowest quartile of the distribution as the cut-off ($10,000-19000 vs. ≥$20,000/year).

Analyses examining the relationship between the presence of mood and anxiety disorders and a) treatment for an illegal drug use disorder (n = 1114) and b) perceived unmet need for treatment among participants who never received treatment for an illegal drug use disorder (n = 764) were conducted separately.

Because of a significant overlap between treatment for an illegal drug use disorder and treatment for mood (44.5%) and anxiety (28.5%) disorders, in secondary analyses we further controlled for lifetime treatment for mood and anxiety disorders. Furthermore, 74.9% of participants with lifetime illegal drug use disorder also had an alcohol-related disorder; in further analyses we controlled for comorbid alcohol-related disorder.

All analyses were carried out in a logistic regression framework using STATA/SE 12; odds ratios and 95% confidence intervals were weighted to account for sampling design. We used Taylor series estimation methods for variance estimation (STATA ¹svy¹ commands) to obtain proper standard error estimates for the cross-tabulations and logistic regressions 27 .

As shown in Table 2, the most frequent types of treatment were a consultation with a physician, psychiatrist, psychologist or social worker (46.9%) and attendance of a drug rehabilitation program (10.2%).

Overall, 31.4% of study participants ever received any form of treatment for an illegal drug use disorder. As shown in Table 3, the likelihood of receiving such treatment was not associated with comorbid mood and anxiety disorders. The probability of illegal drug use disorder treatment was highest in participants who were older than 35 years and lowest in participants who were Native American, Asian, or Hispanic.

^aOdds ratios are weighted to account for sampling design and adjusted for all covariates described in the table.

In additional analyses we observed that the majority of participants who received treatment for illegal drug use disorders used one (33.0%) or two drugs (14.0%) and the number of drugs used was positively associated with treatment (p < 0.01). Moreover, treatment was more frequent among participants with a lifetime history of an alcohol-related disorder (multivariate OR: 1.93, 95% CI 1.34- 2.78), as well as among those who received treatment for mood disorders (multivariate OR: 1.72, 95% CI 1.13-2.61) or anxiety disorders (multivariate OR: 1.89, 95% CI 1.16-3.08).

Eleven percent of study participants who did not receive treatment reported unmet need for treatment for an illegal drug use disorder. As shown in Table 4, individuals who simultaneously had an illegal drug use disorder and a mood or anxiety disorder had higher levels of perceived unmet need for treatment than participants with no such comorbidities (bivariate ORs respectively: illegal drug use and mood disorder: 2.17, 95% CI: 1.22-3.85; illegal drug use and anxiety disorder: 2.37, 95% CI: 1.25-4.48; illegal drug use and mood and anxiety disorder: 2.74, 95% CI 1.66- 4.55). Other factors significantly associated with perceived unmet need for treatment included being African-American and between 35 and 54 years of age.

^bOdds ratios are weighted to account for sampling design and adjusted for all covariates described in the table.

After controlling for all socio-demographic covariates, comorbid mood and anxiety disorders were associated with an elevated likelihood of perceived unmet need for treatment (multivariate ORs respectively: illegal drug use and mood disorder: 2.21, 95% CI, 1.23- 4.10; illegal drug use and anxiety disorder: 2.38, 95% CI, 1.27-4.45; illegal drug use and mood and anxiety disorder: 2.90, 95% CI, 1.71-4.94). After adjusting for treatment for mood and anxiety disorders, individuals who had an illegal drug use disorder as well as an anxiety disorder or both mood and anxiety disorders had higher levels of perceived unmet need for treatment than participants with no such comorbidities (multivariate ORs respectively: illegal drug use and anxiety disorder: 2.34, 95% CI: 1.23- 4.44; illegal drug use and mood and anxiety disorders: 1.94, 95% CI: 1.09- 3.47).

After additionally controlling for lifetime alcohol-related disorder, only comorbid anxiety disorder remained significantly associated with an elevated likelihood of perceived unmet need for treatment (multivariate OR 2.02, 95% CI, 1.03-3.94).

Main reasons for not using treatment services were: the drug problem not being serious enough (37.3%), no willingness to seek treatment (20.0%), spontaneous recovery (11.3%) and social stigma (7.7%) (not shown). These did not differ between the three comorbidity groups (p = 0.15). The majority of participants who reported perceived unmet need for treatment used one (39.8%) or two (16.5%) drugs, primarily cannabis (40.5%), cocaine (27.4%), or heroin (7.4%). The number of drugs used was positively associated with perceived unmet need for treatment (p <0.01).

Our study, based on a nationally representative sample of the US adult population, shows that among individuals who have an illegal drug use disorder, comorbid mood and anxiety disorders are not associated with treatment for illegal drug use disorders. However, common psychiatric comorbidities predict an over twofold increase in perceived unmet need for treatment, which is not accounted for by socio-demographic characteristics, treatment for mood and anxiety disorders and comorbid alcohol use disorder. Overall, these findings suggest that individuals who concomitantly have an illegal drug use disorder and mood or anxiety disorders require special attention from mental health and addiction specialists; integrated mental health and substance use programs could prove effective in addressing their treatment needs.

Our result of no association between comorbid mood and anxiety disorders and treatment for illegal drug use disorders contrasts with findings of studies which simultaneously examined illegal drug use and alcohol-related disorders 21 . However, in line with prior research, we found that treatment for illegal drug use disorders was associated with the number of drugs used 21 , suggesting that the severity of substance use problems is a relevant predictor of treatment seeking.

Mood and anxiety disorders were associated with high levels of unmet need for illegal drug use disorder treatment. Overall, the magnitude of the association was similar whether the comorbid disorder was a mood or an anxiety disorder, suggesting that it is the presence of comorbidity rather than its type that is especially relevant. After controlling for treatment for mood and anxiety disorders as well as comorbid alcohol use disorder, only anxiety disorders were associated with an approximately two-fold increase in the likelihood of perceived unmet need for treatment. One possible explanation of this finding is that individuals who simultaneously have illegal drug use problems and mood disorders seek treatment for their mood disorder, because of lesser stigma, easier access, and simpler reimbursement schemes 21 . This does not appear to be the case of participants with comorbid anxiety disorders, who are less likely to seek mental health treatment.

Elevated rates of perceived unmet need for treatment for illegal drug use related disorders may be due to the scarcity of specialized services, absence of adequate programs for simultaneous treatment of co-occurring illegal drug use and psychiatric disorders, as well as stigma. Moreover, perceptions of the severity of the drug use problem and of treatment effectiveness may also play a role. Currently available treatment services for people who have illegal drug use problems other than crack/heroin may not be sufficient, leading to low access and high perceived unmet need for treatment. Furthermore, when illegal drug use disorders are not addressed by mental health services not specialized in addiction, individuals may have the impression that their drug use problems are not serious enough to require treatment or that adequate treatment is not available.

We need to acknowledge several study limitations. First, illegal drug use and psychiatric disorders, as well as treatment for illegal drug use disorder and perceived unmet need for treatment were studied over participants’ lifetime and we are not able to assess trajectories of substance use and treatment. Research shows that the association between addictive behaviors and mood and anxiety disorders is bidirectional, but common psychiatric disorders most frequently precede onset of substance dependence 28 . Thus, in our study mood and anxiety disorders probably preceded treatment for illegal drug use disorders. Second, data on illegal drug use were collected by federal employees and could be underreported. Nonetheless, the NESARC study interviews were confidential and the reliability of the AUDADIS-IV is good to excellent, therefore the influence of information bias on our measures should be limited 29 30 31 . Prior to fieldwork, Census employees received extensive training in the AUDADIS, which was created to be administered by non-clinicians, similar to several standardized epidemiological interviews 30 . Moreover, the reliability of the AUDADIS have previously been verified and described 31 . Third, access to treatment for illegal drug use disorders and unmet need for treatment were ascertained using a single measure specific to the NESARC study, previously used by other researchers 32 . Fourth, the study population did not include individuals who are institutionalized, who may have elevated rates of both illegal drug use disorders and unmet treatment needs. Therefore, the association between mood and anxiety disorders and unmet need for illegal drug use treatment may be stronger than we report. Fifth, we were not able to examine whether access to treatment for illegal drug use disorders varied between rural and urban areas, by state, or type of treatment unit.

Our study also has strengths. First, we studied a nationally representative sample of the US adult population which limits the possibility of selection bias. Second, psychiatric disorders were assessed using a structured diagnostic questionnaire, which evaluates the presence of clinically relevant disorders as measured by DSM-IV criteria 26 . Third, we investigated factors associated with service use, but also unmet perceived need for treatment for illegal drug related disorders.

National surveys indicate that only half of US mental health centers offer treatment programs for individuals with ‘dual diagnosis’ and this proportion did not significantly change during the past decade 6 . Lack of progress in this area can be explained by a) insufficient clinical knowledge, b) the absence of standardized screening procedures to identify individuals who have an illegal drug use disorder 33 34 , and c) incomplete treatment guidelines in case of dual diagnosis and insufficient organizational support 35 . Patients themselves repeatedly report that illegal drug use disorders are often ignored and not treated by mental health care providers.

According to a proposed four-quadrant continuity-of-care model, a) persons with low severity psychiatric and substance use disorders should be treated in the primary health care system (quadrant I); b) persons with high severity psychiatric disorders and low severity substance use disorders should be treated in the mental health system (quadrant I); c) persons with low severity psychiatric disorders and high severity substance use disorders should be treated in the addiction treatment system (quadrant III); d) persons with high severity psychiatric and substance use disorders need to use multiple treatment systems and make more frequent use of emergency and inpatient services (quadrant IV) 36 . This relies on a well-functioning referral and collaboration system between the point-of-contact (primary care, emergency departments, etc.) and specialized care. Number of studies conducted in the last 15 years provided evidence for the effectiveness of integrated treatment for comorbid psychiatric and substance use disorders 37 38 39 . Nevertheless the integration of mental health and substance use treatment services has produced uneven results due to a number of barriers such as the absence of regulations, the lack of appropriate financial resources 40 , the lack of sufficiently experienced staff 41 , high rates of stuff turn over 42 , and the lack of tools and implementation strategies 43 . In a context of low availability of integrated treatment, progress has been made with the development and continued validation of external rating tools (DDCAT -the Dual Diagnosis Capability in Addiction Treatment and DDCMHT-the Dual Diagnosis Capability in Mental Health Treatment) to determine the capability of substance abuse and mental health service providers to treat co-occurring disorders 44 45 . Previous evaluations using the DDCAT showed that the majority of substance abuse programs were rated less than ‘capable’ and only few programs achieved or exceeded the ‘capable’ level 46 47 . Overall, growing evidence stresses the importance of routine screening for illegal drug use disorders in patients with other psychiatric disorders and the integration of addiction services into psychiatric and medical treatment settings to promote symptom reduction and stable remission.