Anxiety,Bipolar,Depressive,and Schizophrenia Diagnoses Among Patients Receiving Addiction Treatment in the United States,2006-2022: A Descriptive Study

Abstract

Background:

This study examined specific mental health diagnoses groupings among patients admitted to specialty addiction treatment in the United States from 2006 to 2022.

Methods:

Joinpoint regression was used to examine annual data from the publicly available Treatment Episode Data Set – Admissions. Sample selection criteria included having a primary substance listed and any of the following conditions: (a) anxiety disorders, (b) bipolar disorders, (c) depressive disorders, and (d) schizophrenia and other psychotic disorders.

Results:

Prior to sample selection, across the full dataset of 31 152 649, cases with anxiety disorders accounted for 0.1%, bipolar disorders accounted for 0.3%, depressive disorders accounted for 0.4%, and schizophrenia and other psychotic disorders accounted for 0.1%. The total sample size was N = 224 051 with: (a) 13% for anxiety disorders (n = 29 111), (b) 30.3% for bipolar disorders (n = 67 926), (c) 43.4% for depressive disorders (n = 97 293), and (d) 13.3% for schizophrenia and other psychotic disorders (n = 29 721). Count data identified significant decreases including the following: 12.29% from 2008 to 2022 among the bipolar disorder group, 16.69% from 2019 to 2022 among the depressive disorder group, and 6.96% from 2006 to 2022 among the schizophrenia and other psychotic disorders group.

Conclusion:

This study has important implications for future research and clinical care related to co-occurring mental health and substance use disorders. Its findings demonstrate the need for further studies to examine how addiction treatment facilities are screening for and recording mental health disorder diagnoses. Future research is needed to clarify the prevalence of specific mental health diagnoses in the addiction treatment setting, including diagnoses such as anorexia nervosa and generalized anxiety disorder. This research would help determine specific treatment needs by using epidemiological data to provide a snapshot of the prevalence of these co-occurring conditions.

Keywords

addiction treatment anxiety disorders bipolar disorders depressive disorders dual diagnosis schizophrenia substance-related disorders

Highlights

This study examined trends in mental health diagnoses in addiction treatment.

Real-world addiction treatment episodes from the Treatment Episode Data Set – Admissions dataset were used.

Repeated cross-sectional data from 2006 to 2022 were included.

Trends of anxiety, bipolar, depressive, and schizophrenic disorders were examined.

Introduction

Twenty million adults in the United States had co-occurring mental health and substance use disorders in 2023.¹ According to the National Survey on Drug Use and Health, fewer than 1 in 5 (19%) received both mental health and substance use disorder treatment.¹ Another 2 in 5 received treatment for their mental health disorder only, and 4% received substance use disorder treatment solely.¹ When they receive treatment, the majority of individuals with co-occurring mental health and substance use disorders primarily receive mental health disorder treatment, and a minority receive addiction treatment.¹

The Substance Abuse and Mental Health Services Administration’s Treatment Episode Data Set – Admission (TEDS-A) provides an annual snapshot of admissions to US addiction treatment facilities that receive public funds. While previous studies have used TEDS data to examine mental health comorbidity among those receiving addiction treatment, these primarily used a binary (Yes/No) variable in the dataset that indicates whether any co-occurring mental health diagnosis is present.^2,3 For example, one study found that among substance use disorder treatment discharge data from 2009 to 2011, 28% of individuals had a co-occurring mental health condition, as indicated by this binary variable.

There is another variable in TEDS that also captures information about the mental health diagnoses of those receiving treatment. This underutilized variable, “DSMCRIT,” captures specific mental health groupings, such as anxiety disorders. One study used DSMCRIT to examine its association with treatment completion but reported high missingness within the variable.⁴ Even with the variable missingness,^4,5 however, these specific diagnostic groupings are helpful, as they include different clinical features describing the cluster of conditions within these categories. For example, based on diagnostic criteria, anxiety disorders have a common fear-based feature, whereas bipolar disorders may include hypomanic/manic and depressive features.⁶ Researchers have yet to use these data to describe cases and examine trends in substance use disorder admissions in relation to specific mental health disorder categories found in TEDS-A, such as anxiety disorders, bipolar disorders, depressive disorders, and schizophrenia and other psychotic disorders. Examples of these conditions include anxiety disorders: separation anxiety disorder; bipolar disorders: bipolar I disorder; depressive disorders: disruptive mood dysregulation disorder; and schizophrenia and other psychotic disorders: delusional disorder.⁶

The current examined the 4 most prevalent mental health disorder diagnostic groupings found in the TEDS-A from 2006 to 2022: (a) anxiety disorders, (b) bipolar disorders, (c) depressive disorders, and (d) schizophrenia and other psychotic disorders.⁷ This study was conducted because previous studies examining mental health disorders in the TEDS-A have largely focused on a binary Yes/No variable.^2,3,5 Therefore, this study contributes to the literature by focusing on specific mental health diagnostic categories. Although the DSMCRIT variable has limitations, including missing data and being a mutually exclusive variable that is also shared with substance use disorder diagnoses, it provides insights into noted (a) anxiety disorders, (b) bipolar disorders, (c) depressive disorders, and (d) schizophrenia and other psychotic disorder admissions co-occurring with substance use disorders in addiction treatment.

Methods

Dataset

This study examined the TEDS-A, 2006 to 2022, a repeated cross-sectional dataset of state-reported addiction treatment episodes.⁷ As different treatment facilities provide the data, there is variability in how they are collected. However, each state reports its data to the Substance Abuse and Mental Health Services Administration using uniform variables and values. The full dataset contains 31 152 649 treatment episodes. A variable that identifies the primary substance for a specific treatment episode was used to select cases with a primary substance listed, thereby excluding those with missing data or no substance indicated. Cases with a primary substance selected as a proxy to identify admissions in which co-occurring substance use disorders and one of the 4 mental health disorder groupings were present. There were 29 587 659 cases with a primary substance listed. Using DSMCRIT, a variable that indicates mental health diagnosis grouping, we selected treatment episodes with diagnoses of anxiety disorder, bipolar disorder, depressive disorder, or schizophrenia/other psychotic disorders, which resulted in the final analytic sample of N = 224 051 cases. Two more mental health disorder categories were included in the DSMCRIT variable, “attention deficit/disruptive behavior disorders” and “other mental health condition.” These 2 categories were excluded from the analysis because the “attention deficit/disruptive behavior disorders” variable had low counts during the analytic period (n = 6464), resulting in low counts in some analyses in this study, and the “other mental health condition” category did not provide specificity in the mental health conditions. The number of cases in the 4 diagnostic groups examined in this study were as follows: (a) anxiety disorders (n = 29 111; 13%), (b) bipolar disorders (n = 67 926; 30.3%), (c) depressive disorders (n = 97 293; 43.4%), and (d) schizophrenia and other psychotic disorders (n = 29 721; 13.3%). The dataset lists a maximum of 1 of these conditions per case.

Measures

The variables included in this study described aspects of the treatment episode admissions, such as their sociodemographic characteristics. Some values were collapsed because of small cell sizes.

Age

This variable described the age at admission with the following values: 12 to 17, 18 to 29, 30 to 39, 40 to 49, and 50 years and older.

Mental Health Diagnosis

This variable identified which of the 4 study diagnostic groups a case belonged to: anxiety disorders, bipolar disorders, depressive disorders, or schizophrenia and other psychotic disorders.

Primary Substance

Primary substances included alcohol, benzodiazepines, cannabis, cocaine/crack, methamphetamine, other amphetamines, opioids, and other primary substance. Some substance categories were combined due to low cell counts. One example is nonprescription methadone, which was combined into the category of opioids due to low cell counts when annually examining the n = 614 total cases with this substance listed.

Polysubstance Use

In addition to the primary substance variable, the dataset also has a secondary substance variable. Cases with a secondary substance listed were classified as “Yes” for polysubstance use, and those without were classified as “No.”

Race and Ethnicity

Two variables were combined to create this variable, one that identified ethnicity and another that identified race. Any case identified as Hispanic was listed as Hispanic of Any Race. Other categories in this variable included Black or African American, White, and Another Race.

Region

This variable describes the regions of the United States, along with US territories. The following categories for this variable are Midwest, Northeast, South, West, and US territories.

Sex

The categories for this variable were Female and Male.

Treatment Type

This variable described the type of treatment service/setting: hospital detoxification (withdrawal management), free-standing residential detoxification, hospital non-detoxification residential, short-term residential, long-term residential, intensive outpatient, non-intensive outpatient, and outpatient detoxification. Some treatment types were combined due to small cell sizes in accordance with appropriate data protection techniques.

Year

This variable described the year of admission, which ranged from 2006 to 2022.

Analysis

Study analyses used Joinpoint Regression Software,⁸ SPSS,⁹ and R¹⁰ with the ggplot2¹¹ specific package. Annual percentages of the 4 mental health diagnosis groups were examined among the full dataset of 31 152 649 treatment episodes. The annual percentage of missingness on the DSMCRIT variable across the full dataset was also examined. Univariable analyses were conducted among the analytic sample of N = 224 051 cases. Joinpoint regression was used to examine annual count data for the following diagnoses: (a) anxiety disorders, (b) bipolar disorders, (c) depressive disorders, and (d) schizophrenia and other psychotic disorders. Annual counts rather than annual percentages were examined because the annual percentages showed little variability. The input file contained 3 variables, [a] year, [b] diagnosis, and [c] count. In the input file each year from 2006 to 2022 contained one of the 4 mental health disorder groups (eg, anxiety disorders) and a count corresponding to the number of cases with that condition for a specific year. The joinpoint regression models were specified with count dependent variables, constant variance (homoscedasticity): uncorrelated, “yes” as the default log transformation, and using the empirical quantile method for the final selected model to identify more precise confidence intervals. Diagnosis was used as a by variable in the Joinpoint Regression Software. Weighted Bayes Information Criterion was selected as the data-driven Bayesian Information Criterion methods. Empirical Quantile was selected for the final selected model, which utilizes confidence intervals to identify significance rather than P-values, such as a confidence interval not included zero is significant. No advanced analysis tools such as pairwise comparison or jump model/comparability ratio model were selected for these analyses. Joinpoint regression analysis has utility in addiction research as it segments different portions (joinpoints) of temporal data, then determines whether these joinpoints are significant if the 95% confidence interval (CI) does not include zero.

The University of North Carolina at Chapel Hill Institutional Review Board conducted ethical review and identified this study as not meeting the criteria for human subjects research.

Results

Total Reported Prevalence

Across the full dataset of 31 152 649 treatment episodes, cases with anxiety disorders accounted for 0.1%, bipolar disorders accounted for 0.3%, depressive disorders accounted for 0.4%, and schizophrenia and other psychotic disorders accounted for 0.1% of the total treatment episodes. The annual percentages of cases with these diagnoses across the entire dataset may be found in Table 1. Missingness on the DSMCRIT variable was highest in 2007 at 60.8% and lowest in 2020 at 21.2%. Data on annual missingness may be seen in Supplemental Table 1.

Table 1.

Annual Percentages of Cases Flagged With Anxiety, Bipolar, Depressive, or Schizophrenia and Other Psychotic Disorder Diagnoses Among the Entire Sample of Addiction Treatment Admissions From 2006 to 2022; N = 31 152 649.

Year	Mental health diagnosis
Year	Anxiety disorders (%)	Bipolar disorders (%)	Depressive disorders (%)	Schizophrenia and other psychotic disorders (%)
2006	0.1	0.4	0.6	0.2
2007	0.1	0.4	0.6	0.2
2008	0.1	0.4	0.5	0.2
2009	0.1	0.3	0.4	0.1
2010	0.1	0.3	0.4	0.1
2011	0.1	0.3	0.4	0.1
2012	0.1	0.3	0.4	0.1
2013	0.1	0.3	0.4	0.1
2014	0.2	0.2	0.4	0.1
2015	0.1	0.1	0.2	0.1
2016	0.1	0.1	0.3	0.1
2017	0.1	0.1	0.3	0.1
2018	0.1	0.1	0.2	0.1
2019	0.1	0.1	0.3	0.1
2020	0.2	0.2	0.3	0.1
2021	0.1	0.1	0.2	0.1
2022	0.1	0.4	0.6	0.2

The entire analytic sample included 31 152 649 cases, of which 224 051 cases had any of the 4 mental health diagnoses presented in this table.

Sample Characteristics

Depressive disorders were the most prevalent diagnoses (n = 97 293; 43.4%), followed by bipolar disorders (n = 67 926; 30.3%). Across the entire sample, the majority of cases were male (n = 116 802; 52.1%) and white (n = 142 950; 63.8%). Approximately 74% (n = 164 598) of the cases were admitted to non-intensive outpatient treatment, and 13% (n = 29 822) of the cases were admitted to intensive outpatient treatment. US territories account for 0.3% (n = 655) of the cases in this sample. Regarding regions of the United States in which the treatment admissions occurred, approximately 31% (n = 69 281) were in the Midwest, 7% (n = 15 693) were in the Northeast, 52% (n = 117 134) were in the South, and 10% (n = 21 288) were in the West. Table 2 provides the characteristics of each of the 4 diagnostic groups.

Table 2.

Characteristics of the Sample With Anxiety, Bipolar, Depressive, or Schizophrenia and Other Psychotic Disorder Diagnoses During Addiction Treatment Admissions From 2006 to 2022; N = 224 051.

Variable	Anxiety disorders		Bipolar disorders		Depressive disorders		Schizophrenia and other psychotic disorders
	N = 29 111		N = 67 926		N = 97 293		N = 29 721
	n	%	n	%	n	%	n	%
Treatment type
Free-standing residential detoxification	1018	3.5	1679	2.5	2264	2.3	653	2.2
Short-term residential	1815	6.2	4323	6.4	6603	6.8	2089	7.0
Long-term residential	854	2.9	2425	3.6	2766	2.8	1587	5.3
Intensive outpatient	3875	13.3	9182	13.5	13 049	13.4	3716	12.5
Non-intensive outpatient	21 350	73.3	49 886	73.4	71 899	73.9	21 463	72.2
Other treatment type	199	0.7	431	0.6	712	0.7	213	0.7
Gender
Female	14 083	48.4	36 287	53.4	48 497	49.8	8286	27.9
Male	15 014	51.6	31 606	46.5	48 754	50.1	21 428	72.1
Missing	14	0.0	33	0.0	42	0.0	7	0.0
Age
12-17 y	1542	5.3	3038	4.5	5924	6.1	177	0.6
18-29 y	10 188	35.0	21 905	32.2	26 766	27.5	7906	26.6
30-39 y	8681	29.8	19 842	29.2	24 873	25.6	8221	27.7
40-49 y	5473	18.8	15 463	22.8	22 828	23.5	7830	26.3
50 y and older	3227	11.1	7678	11.3	16 902	17.4	5587	18.8
Race and ethnicity
Black or African American	2785	9.6	10 922	16.1	18 003	18.5	11 812	39.7
White	21 868	75.1	46 150	67.9	62 889	64.6	12 043	40.5
Hispanic of any race	1765	6.1	3457	5.1	5147	5.3	1666	5.6
Another race	1349	4.6	2425	3.6	4239	4.4	1575	5.3
Missing	1344	4.6	4972	7.3	7015	7.2	2625	8.8
Primary substance
Alcohol	10 662	36.6	26 121	38.5	42 148	43.3	12 831	43.2
Benzodiazepines	678	2.3	819	1.2	1272	1.3	155	0.5
Cannabis	6041	20.8	15 464	22.8	19 587	20.1	7095	23.9
Cocaine/Crack	1685	5.8	9092	13.4	10 031	10.3	4737	15.9
Heroin	3337	11.5	5518	8.1	7144	7.3	1249	4.2
Methamphetamine	2061	7.1	3312	4.9	4714	4.8	1991	6.7
Other amphetamines	258	0.9	463	0.7	603	0.6	213	0.7
Other opiates and synthetics	3375	11.6	5466	8.0	9365	9.6	783	2.6
Other primary substance	1014	3.5	1671	2.5	2429	2.5	667	2.2
Polysubstance use
No	13 401	46.0	25 018	36.8	41 325	42.5	11 056	37.2
Yes	15 710	54.0	42 908	63.2	55 968	57.5	18 665	62.8

Trend Analysis

As seen in Figure 1, no significant trend was identified for the anxiety disorder group.

Figure 1.

Joinpoint regression analyses examining reported annual trends of (1) anxiety disorder: represented by circles in the figure, (2) bipolar disorder: represented by triangles in the figure, (3) depressive disorder: represented by diamonds in the figure, and (4) schizophrenia and other psychotic disorders: represented by x’s in the figure, diagnoses among addiction treatment episodes from 2006 to 2022. The line starting at the top represents depressive disorders. The second from the top starting line identifies bipolar disorders. The second from the bottom starting line represents schizophrenia and other psychotic diagnoses, which ranged from 2431 in 2006 to 738 in 2022. The line starting at the bottom represents anxiety disorders. An asterisk (*) indicates joinpoints identified as significant with a 95% confidence interval that does not include zero.

Among the bipolar disorder group, one joinpoint was identified, which included a significant decrease of 12.29% from 2008 to 2022 (CI = −17.39, −10.65). Among the depressive disorder group, one joinpoint was identified, which included a significant decrease of 16.69% from 2019 to 2022 (95% CI = −34.07, −6.31). Among the schizophrenia and other psychotic disorder group, no joinpoint was identified, which included a significant decrease of 6.96 for the full analytic period of 2006 to 2022 (95% CI = −9.09, −4.72). A significant decrease with no joinpoint identified indicates a consistent decline.

Supplemental Figures

After conducting the primary analyses, 4 supplemental a posteriori figures examining the sample’s most utilized treatment types, most reported primary substances, sex (female and male), and polysubstance use statuses (Yes and No) were created. These figures are labeled as Supplemental Figures 1 to 4.

Discussion

This study described the presence of anxiety, bipolar, depressive, and schizophrenia/other psychotic disorder diagnoses among admissions to addiction treatment facilities in the United States. Both the annual prevalence and annual counts of these conditions were examined. Significant decreases in the number of cases with bipolar disorders (from 2008 to 2022), depressive disorders (from 2019 to 2022), and schizophrenia or other psychotic disorders (from 2006 to 2022) were identified. However, there was not much variability in the annual prevalence of these cases when examining the annual percentages of these cases.

Although trends were identified in the annual count data, the lack of variability in the annual percentage data suggests these may be more a function of overall admission trends. Nationally representative data suggest that individuals with co-occurring mental health and substance use disorders primarily receive mental health treatment instead of addiction treatment.¹ Specifically, 19% of these persons received treatment for both, 2 in 5 received only mental health disorder treatment, and 4% received only substance use disorder treatment.¹ The presence of these co-occurring conditions can impact treatment engagement. For example, among substance use disorder treatment, having a mental health condition is associated with leaving treatment prematurely.^2,3 While individuals with co-occurring conditions may be more likely to enter the mental health treatment setting, several factors may influence substance use disorder treatment engagement and whether persons entering these treatment settings are even screened for mental health conditions.

Alongside persons with co-occurring conditions primarily entering mental health treatment, several factors may influence the low variability of these diagnoses in this sample. Limited resources and training can be a barrier to both screening and treating co-occurring mental health and substance use disorders.¹² For example, if there is a lack of providers with expertise in screening for and treating co-occurring conditions, facilities may be less likely to offer these services. As TEDS-A is a large surveillance dataset of addiction treatment admissions, reporting practices, staff screening, client self-report, sociodemographic and geographic factors that can influence who is screened and treated for specific conditions, and the level of funding at the state or local level all may impact what data from the state level are ultimately reported to the Substance Abuse and Mental Health Services Administration as the proprietors of the dataset. Future research is needed to identify the prevalence of specific mental health diagnoses in the addiction treatment setting nationally, including diagnoses such as anorexia nervosa and generalized anxiety disorder, to provide a more extensive snapshot of treatment needs. Along with the co-occurring mental health and substance use conditions that may be present in an addiction treatment setting, screening for multiple substances in the form of polysubstance use is also a prominent consideration.

The current study identified polysubstance use at rates ranging from 54% to 63% among the 4 diagnostic groups analyzed. This demonstrates the importance of considering polysubstance use alongside co-occurring mental health conditions.^13,14 The prevalence of not only co-occurring mental health diagnoses but also co-occurring substance use disorders highlights the complex clinical reality of treating multiple conditions. Many providers do not offer co-occurring services, and those that do may be limited by factors such as institutional policy or lack of staff.¹² In addition, it is widely recognized that structural barriers can prevent individuals from accessing co-occurring treatment for mental health and substance use disorders.^15-17 For this reason, clients with co-occurring conditions may also require case management or other targeted social interventions—like childcare or legal support—that can facilitate access to treatment.¹⁵

Among the 4 mental health diagnostic groups analyzed in this study, alcohol and cannabis were the 2 most reported primary substances. This is not surprising as—excluding tobacco use disorder—alcohol and cannabis use disorders are the most prevalent substance use disorders in the United States, with 10% of individuals meeting the criteria for alcohol use disorder and 6.8% for cannabis use disorder in 2023.¹ However, in the larger dataset used for this study prior to the sample selection criteria, heroin was the second most reported primary substance, surpassing cannabis. While the prevalence of cannabis surpassed heroin in this sample, the combined category of opioids, which included heroin and other opiates and synthetics, accounted for the third largest primary substance. The other opiates and synthetics category was larger than heroin as the category for the primary substance, likely due to the significant presence of fentanyl on the market during this period.^18-20 These findings align with other studies that have identified high co-occurrence of mental health disorders with opioid use disorder.²¹

This study faced some limitations related to the quality and quantity of data available in the TEDS-A dataset. One limitation of using the TEDS-A is that it only contains data about treatment episodes in facilities that receive public funds, which may not be generalizable to individuals or treatment episodes more broadly. There is also a potential for bias in chart data used for clinical purposes, including self-report biases. Nevertheless, the TEDS-A offers the most robust publicly available addiction facility–level data on annual treatment episode admissions in the United States. Other limitations of the study are related to the DSMCRIT variable used to identify the 4 diagnostic groups. These include that it has high levels of missing data (44%),⁴ is shared across substance-related and addictive disorders as possible values on the variable (eg, alcohol intoxication), and only permits one diagnosis per case. Furthermore, bias is introduced because facilities documenting these specific diagnoses may differ from those that do not. However, DSMCRIT is the only variable in the TEDS-A that provides specific mental health diagnostic groupings. Whereas other studies of this dataset utilized the binary Yes/No variable that identifies the presence of co-occurring mental health disorders in general,^2,3 using the DSMCRIT variable allowed the current study to examine specific mental health disorder groupings such as anxiety and depressive disorders.

Conclusion

This study has important implications for future research and clinical care related to co-occurring mental health and substance use disorders. Its findings demonstrate the need for further studies to examine how addiction treatment facilities are screening for mental health disorder diagnoses. Future research is needed to identify the prevalence of specific mental health diagnoses in the addiction treatment setting, including diagnoses such as anorexia nervosa and generalized anxiety disorder. This research would help determine specific treatment needs by using epidemiological data to provide a snapshot of the prevalence of these co-occurring conditions. The current study highlights that, while TEDS-A is a prominent dataset of addiction treatment episodes, it could be improved by the inclusion of variables that indicate which groups of mental health diagnoses were present for a case.

Supplemental Material

sj-docx-5-saj-10.1177_29767342261445244 – Supplemental material for Anxiety, Bipolar, Depressive, and Schizophrenia Diagnoses Among Patients Receiving Addiction Treatment in the United States, 2006-2022: A Descriptive Study

Supplemental material, sj-docx-5-saj-10.1177_29767342261445244 for Anxiety, Bipolar, Depressive, and Schizophrenia Diagnoses Among Patients Receiving Addiction Treatment in the United States, 2006-2022: A Descriptive Study by Orrin D. Ware, David A. Fiellin and Mark P. McGovern in Substance Use & Addiction Journal

Supplemental Material

sj-tiff-1-saj-10.1177_29767342261445244 – Supplemental material for Anxiety, Bipolar, Depressive, and Schizophrenia Diagnoses Among Patients Receiving Addiction Treatment in the United States, 2006-2022: A Descriptive Study

Supplemental material, sj-tiff-1-saj-10.1177_29767342261445244 for Anxiety, Bipolar, Depressive, and Schizophrenia Diagnoses Among Patients Receiving Addiction Treatment in the United States, 2006-2022: A Descriptive Study by Orrin D. Ware, David A. Fiellin and Mark P. McGovern in Substance Use & Addiction Journal

Supplemental Material

sj-tiff-2-saj-10.1177_29767342261445244 – Supplemental material for Anxiety, Bipolar, Depressive, and Schizophrenia Diagnoses Among Patients Receiving Addiction Treatment in the United States, 2006-2022: A Descriptive Study

Supplemental material, sj-tiff-2-saj-10.1177_29767342261445244 for Anxiety, Bipolar, Depressive, and Schizophrenia Diagnoses Among Patients Receiving Addiction Treatment in the United States, 2006-2022: A Descriptive Study by Orrin D. Ware, David A. Fiellin and Mark P. McGovern in Substance Use & Addiction Journal

Supplemental Material

sj-tiff-3-saj-10.1177_29767342261445244 – Supplemental material for Anxiety, Bipolar, Depressive, and Schizophrenia Diagnoses Among Patients Receiving Addiction Treatment in the United States, 2006-2022: A Descriptive Study

Supplemental material, sj-tiff-3-saj-10.1177_29767342261445244 for Anxiety, Bipolar, Depressive, and Schizophrenia Diagnoses Among Patients Receiving Addiction Treatment in the United States, 2006-2022: A Descriptive Study by Orrin D. Ware, David A. Fiellin and Mark P. McGovern in Substance Use & Addiction Journal

Supplemental Material

sj-tiff-4-saj-10.1177_29767342261445244 – Supplemental material for Anxiety, Bipolar, Depressive, and Schizophrenia Diagnoses Among Patients Receiving Addiction Treatment in the United States, 2006-2022: A Descriptive Study

Supplemental material, sj-tiff-4-saj-10.1177_29767342261445244 for Anxiety, Bipolar, Depressive, and Schizophrenia Diagnoses Among Patients Receiving Addiction Treatment in the United States, 2006-2022: A Descriptive Study by Orrin D. Ware, David A. Fiellin and Mark P. McGovern in Substance Use & Addiction Journal

Footnotes

Acknowledgements

The authors thank Lydia Rose Rappoport-Hankins for her editing expertise.

Author Contributions

Orrin D. Ware: Conceptualization, Methodology, Software, Validation, Formal Analysis, Investigation, Data Curation, Writing – Original Draft, Writing – Review & Editing, Visualization; David A. Fiellin: Methodology, Investigation, Writing – Original Draft, Writing – Review & Editing; Mark P. McGovern: Methodology, Investigation, Writing – Original Draft, Writing – Review & Editing, Supervision, Funding Acquisition.

ORCID iD

Orrin D. Ware

Ethical Considerations

The University of North Carolina at Chapel Hill Institutional Review Board conducted ethical review and identified this study as not meeting the criteria for human subjects research.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding was received from the National Institute on Drug Abuse (NIDA) P50 Center for Dissemination & Implementation at Stanford (C-DIAS) (P50DA05544072).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental Material for this article is available online at the SAJ website .

References

Substance Abuse and Mental Health Services Administration. Key substance use and mental health indicators in the United States: results from the 2023 National Survey on Drug Use and Health (HHS Publication No. PEP24-07-021, NSDUH Series H-59). Center for Behavioral Health Statistics and Quality, Substance Abuse and Mental Health Services Administration. Accessed January 10, 2026. https://www.samhsa.gov/data/report/2023-nsduh-annual-national-report

Friesen

Kurdyak

The impact of psychiatric comorbidity on treatment discontinuation among individuals receiving medications for opioid use disorder. Drug Alcohol Depend. 2020;216:108244. doi:10.1016/j.drugalcdep.2020.108244

Krawczyk

Feder

Saloner

Crum

Kealhofer

Mojtabai

The association of psychiatric comorbidity with treatment completion among clients admitted to substance use treatment programs in a U.S. national sample. Drug Alcohol Depend. 2017;175:157-163. doi:10.1016/j.drugalcdep.2017.02.006

Baird

Cheng

Xia

Use of machine learning to examine disparities in completion of substance use disorder treatment. PLoS One. 2022;17(9):e0275054. doi:10.1371/journal.pone.0275054

Acharya

Kar

Ally

Soar

Predicting co-occurring mental health and substance use disorders in women: an automated machine learning approach. Appl Sci. 2024;14(4):1630.

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed., text rev. Accessed January 10, 2026. https://dsm.psychiatryonline.org/doi/book/10.1176/appi.books.9780890425787

Substance Abuse and Mental Health Services Administration. Data from: treatment episode data set admissions (TEDS-A) concatenated 2006-2022: public use file. 2024.

Joinpoint Regression Program, Version 4.9.1.0 . Statistical Methodology and Applications Branch, Surveillance Research Program; 2022.

IBM SPSS Statistics: Version 31.0 . 2025.

10.

R: A Language and Environment for Statistical Computing . R Foundation for Statistical Computing; 2025. Accessed January 10, 2026. https://www.R-project.org/

11.

Wickham

ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag; 2016.

12.

Adeniran

Quinn

Wallace

, et al. A scoping review of barriers and facilitators to the integration of substance use treatment services into US mainstream health care. Drug Alcohol Depend Rep. 2023;7:100152. doi:10.1016/j.dadr.2023.100152

13.

Ellis

Rabinowitz

Ware

Wells

Dunn

Huhn

AS.

Patterns of polysubstance use and clinical comorbidity among persons seeking substance use treatment: an observational study. J Subst Use Addict Treat. 2023;146:208932. doi:10.1016/j.josat.2022.208932

14.

Ware

Manuel

Huhn

AS.

Adults with opioid and methamphetamine co-use have lower odds of completing short-term residential treatment than other opioid co-use groups: a retrospective health services study. Front Psychiatry. 2021;12:784229. doi:10.3389/fpsyt.2021.784229

15.

Farhoudian

Razaghi

Hooshyari

, et al. Barriers and facilitators to substance use disorder treatment: an overview of systematic reviews. Subst Abuse. 2022;16:11782218221118462. doi:10.1177/11782218221118462

16.

McGovern

Xie

Segal

Siembab

Drake

RE.

Addiction treatment services and co-occurring disorders: prevalence estimates, treatment practices, and barriers. J Subst Abuse Treat. 2006;31(3):267-75. doi:10.1016/j.jsat.2006.05.003

17.

Spivak

Strain

Spivak

, et al. Integrated dual diagnosis treatment among United States mental health treatment facilities: 2010 to 2018. Drug Alcohol Depend. 2020;213:108074. doi:10.1016/j.drugalcdep.2020.108074

18.

Han

Yan

Zheng

Khan

Yuan

The rising crisis of illicit fentanyl use, overdose, and potential therapeutic strategies. Transl Psychiatry. 2019;9(1):282. doi:10.1038/s41398-019-0625-0

19.

Armenian

Barr-Walker

Lynch

KL.

Fentanyl, fentanyl analogs and novel synthetic opioids: a comprehensive review. Neuropharmacology. 2018;134(Pt A):121-132. doi:10.1016/j.neuropharm.2017.10.016

20.

Friedman

Shover

CL.

Charting the fourth wave: geographic, temporal, race/ethnicity and demographic trends in polysubstance fentanyl overdose deaths in the United States, 2010-2021. Addiction. 2023;118(12):2477-2485. doi:10.1111/add.16318

21.

Santo

Jr Campbell

Gisev

, et al. Prevalence of mental disorders among people with opioid use disorder: a systematic review and meta-analysis. Drug Alcohol Depend. 2022;238:109551. doi:10.1016/j.drugalcdep.2022.109551

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

1.58 MB

0.00 MB

0.03 MB

1.24 MB

1.29 MB

1.21 MB