Abstract
Background
Mental health treatment use remains low despite increasing major depressive episodes (MDE), suicidality, and substance use disorders (SUD).
Aims
To examine the associations of perceived treatment helpfulness with clinical characteristics (MDE, suicidality, and SUD) and treatment modality (medication, counseling/other treatment [COT], and both treatment [MCOT]).
Methods
We used the 2022 and 2023 U.S. National Survey on Drug Use and Health data on adult depression treatment users (N = 10,221) and multinomial and generalized ordinal logistic regression models.
Results
Among treatment users, 15% used medication only, 25% used COT only, and 60% used MCOT. Overall, 60% of medication users and 47% of COT users perceived their respective treatments as “a lot” or “extremely” helpful. MDE and suicidality were associated with a higher likelihood of using COT only and/or MCOT. Both MDE and suicidality were associated with lower odds of perceiving treatment as helpful among medication users (OR = 0.47, 95% CI = 0.39–0.57 for MDE; OR = 0.78, 95% CI = 0.66–0.91 for suicidality) as well as COT users (OR = 0.67, 95% CI = 0.56–0.80 for MDE; OR = 0.82, 95% CI = 0.70–0.97 for suicidality). Co-occurring alcohol and drug use disorders were also associated with lower odds of perceived helpfulness among COT users. The use of both medication and COT was associated with higher odds of perceiving either modality as helpful. These patterns did not differ meaningfully for individuals with MDE.
Conclusions
The lower perceived treatment helpfulness among individuals with complex clinical presentations underscores the need for more intensive, personalized, and sustained interventions. Comparison between treatment recipients’ perceived helpfulness and objective clinical effectiveness is also needed.
Supplementary Information
The online version contains supplementary material available at 10.1007/s44192-026-00370-2.
Keywords: Depression, Treatment, Medication, Counseling, Treatment helpfulness
Introduction
Epidemiologic data indicate a rising burden of depression and suicidality in the United States. Among individuals aged 12 and older, the prevalence of depression, as measured by a score ≥ 10 on the Patient Health Questionnaire (PHQ-9) [1], increased from 8.2% during 2013–2014 to 13.1% between August 2021 and August 2023 [2]. Other data sources also showed an increasing trend in major depressive episodes (MDE), especially among adolescents and young adults [3]. In a 2023 National Center for Health Statistics survey, 5.3% of U.S. adults reported past-year suicidal thoughts, with elevated rates among those aged 18–44 and with lower income [4]. Each year, approximately 1.1–1.4% of adults make suicide plans, and 0.5–0.6% attempt suicide, with increasing rates observed among Black Americans [5–7].
Mental illness often co-occurs with substance use disorders (SUD). In 2023, 7.9% of adults had co-occurring mental illness and SUD, with the highest prevalence among those aged 26–49 [8]. An estimated 35% of adults with a mental disorder also had a SUD [9]. Suicidal ideation is especially elevated among individuals with co-occurring conditions. For example, between 2009 and 2020, the prevalence of suicidal thoughts among adults with opioid use disorder rose from 22.8 to 29.8%, with substantially higher rates among those with co-occurring MDE and other SUDs [10].
Despite the high prevalence of depression, suicidality, and co-occurring SUD, treatment rates remain low. Brody and Hughes [2] found that only 43.0% of females and 33.2% of males with depression between August 2021 and August 2023 received counseling or therapy with a mental health professional in the preceding 12 months, even though 87.9% of individuals reported impairment in work, home, or social functioning due to their depressive symptoms. Goodwin et al. [3] also found no corresponding increase in mental health service use despite rising rates of MDE. Mental health treatment uptake is also limited among those experiencing suicidality. According to the 2021 National Survey on Drug Use and Health (NSDUH), just 61% of adults with serious suicidal thoughts accessed mental health services in the past year, and only 37% of those who did not receive care perceived a need for it [11]. A previous NSDUH-based study also found that three-fourths of those who did not receive mental health treatment despite serious suicidal thoughts did not feel the need for treatment [12]. Na et al. [10] observed no increase in mental health service use despite rising suicidal ideation among individuals with opioid use disorder. An earlier study of adult depression care based on the 2008–2014 NSDUH showed that comorbid SUD modestly but significantly decreases the likelihood of receiving past-year depression care for MDE, as 55.4% of adults with MDE and co-occurring SUD received depression care, compared to 60.1% of those without SUD [13].
The low rate of mental health service use among individuals with depression, suicidality, and co-occurring SUD is concerning, given the evidence base of available treatments. A meta-analysis found all 21 reviewed antidepressants to be more effective than placebo in treating major depression [14]. A review of systematic reviews and meta-analyses similarly reported small to medium effect sizes for antidepressants and at least medium-sized effects for various psychotherapies, with combined pharmacotherapy and psychotherapy recommended for more severe or chronic cases [15]. Beyond symptom reduction, patients’ perceived helpfulness of treatment is an essential indicator of care quality and acceptability. Research suggests that perceived helpfulness may be a key predictor of future service use, particularly among populations with historically low engagement in mental health care [16].
Research on the perceived helpfulness of depression care remains limited. Findings from the World Mental Health Surveys indicated that among individuals with lifetime major depression, just over two-thirds reported ever receiving treatment they considered helpful, with older age and higher educational attainment associated with greater odds of perceiving treatment as helpful [17]. In a recent study of French adults with past-year suicidal ideation, Husky et al. [18] found that a larger proportion perceived medication to be helpful compared to psychotherapy, and perceived helpfulness was higher among those with major depression and among those who received both medication and psychotherapy. However, individuals who had recently attempted suicide were less likely to perceive medication as helpful.
Among U.S. adults, analyses of the 2008–2012 and 2008–2014 NSDUH data showed that those with MDE were more likely to perceive treatment as helpful when it was received from specialty mental health providers rather than general practitioners [12, 19]. Kuramoto-Crawford et al. [12] also noted that individuals with suicidal ideation had lower odds of perceiving treatment as helpful. Using 2015–2016 NSDUH data, Alang and McAlpine [16] found that combined treatment involving both medication and counseling was associated with higher perceived helpfulness among adults with MDE. However, poor self-rated health and severe mental illness were associated with lower perceived helpfulness of medication, while substance use problems also correlated with decreased odds of rating counseling or a combination of counseling and medication as helpful.
In this study, we analyzed a nationally representative sample of U.S. adults who received any depression treatment in the past year to examine the relationships between key clinical features, MDE, suicidality, and SUD in particular, and two outcomes: (1) the type of treatment received (medication only, counseling or other treatment only, or both medication and counseling/other treatment), and (2) how helpful the treatment was perceived. Based on previous research, we hypothesized that MDE, suicidality, and any SUD would be associated with a higher likelihood of receiving (H1a) counseling/other treatment only or combined medication and counseling/other treatment, compared to medication only, and (H1b) combined medication and counseling/other treatment, compared to counseling/other treatment only; and (H2) lower odds of perceiving the treatment as helpful, after accounting for sociodemographic factors and self-rated health. The results aim to expand the evidence base on patient-reported experiences with depression treatment and inform strategies to improve care delivery and engagement.
Methods
Data source
Data for this study came from the 2022 and 2023 NSDUH, funded by the U.S. SAMHSA. NSDUH is an annual epidemiologic survey of the civilian, non-institutionalized population, age 12 years and older, to measure the prevalence of behavioral health indicators and treatment, physical/functional health, and healthcare utilization. Because of the COVID-19 pandemic, data collection in 2022 and 2023 was done using both in-person and web-based modes. The overall percentages of interviews that were completed via the web were 40.7% in 2022 and 36.0% in 2023 [20]. The combined number of adult (age 18+) respondents in the 2022 and 2023 NSDUH was 92,233 (47,100 in 2022 and 45,133 in 2023). The analytic sample in this study was 10,221 adults who reported receiving any depression care in the 12 months preceding the survey. All participants in NSDUH provided informed consent and/or assent for data collection and the use of de-identified data in research studies [20]. The authors’ institutional review board exempted this study from review based on secondary data analysis (#00007908).
Measures
Past-year MDE, suicidality, and SUD: MDE was determined in the NSDUH based on DSM-5 criteria [21], which require the presence of five or more depressive symptoms during the same two-week period (coded yes vs. no). Suicidality was assessed with three questions asking whether, at any time in the past 12 months, respondents had: (1) seriously thought about trying to kill themselves, (2) made a plan to do so, or (3) attempted suicide. The plan and attempt questions were asked regardless of whether the respondent reported suicidal thoughts. For this study, we created a binary suicidality variable coded as yes (= 1) if respondents endorsed any of the three items (i.e., suicidal thoughts or behaviors) and no (= 0) if they endorsed none.
The SUD questions in NSDUH are also based on DSM-5 diagnostic criteria. In this study, we used the NSDUH’s four-category SUD variable: No SUD, alcohol use disorder only, drug use disorder only, and both alcohol and drug use disorders. Drugs included marijuana, cocaine (including crack), heroin, hallucinogens, inhalants, methamphetamine, and any use of prescription stimulants, tranquilizers or sedatives (e.g., benzodiazepines), and pain relievers. Thus, drug use disorder may have included more than one SUD diagnosis (i.e., polydrug use disorders), and those categorized as having both alcohol and drug use disorders may have included individuals who met criteria for alcohol use disorder plus one or more additional SUDs. We included nicotine dependence (yes or no) based on the Fagerström Test [22].
Past-year depression treatment: Regardless of MDE classification, adult NSDUH respondents who reported depressive symptoms were asked if they (1) took prescription medication for their mood; and (2) saw/talked to the following people for their depressive feelings: a general practitioner/family doctor; other medical doctor like a cardiologist, gynecologist, urologist; psychologist; psychiatrist or psychotherapist; social worker; counselor; other mental health professional like a mental health nurse; nurse, occupational therapist, or other health professional; religious advisor; another healer; herbalist, chiropractor, acupuncturist, or massage therapist; or another type of helping professional (i.e., counseling/other treatment). In this study, we categorized past-year depression treatment into three mutually exclusive groups: (1) medication only, (2) counseling/other treatment (COT) only, and (3) both medication and COT (MCOT).
Perceived helpfulness of depression treatment: Individuals who received depression treatment were asked to rate the helpfulness of their treatment using two questions: “During the past 12 months, how much has prescription medication for mood helped you?” and “During the past 12 months, how much has treatment/counseling helped you?” Response options were: not at all, a little, some, a lot, and extremely. For the multivariable analyses, the not at all and a little categories were combined to create a four-level ordinal outcome, enhancing model parsimony and fit. We also examined binary outcomes (a lot/extremely vs. all others) for a supplementary analysis.
Sociodemographic and health characteristics: These included age group (18–25, 26–34, 35–49, 50–64, 65+); gender (male vs. female); race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, other); education (college degree vs. no degree); employment status (full-time, part-time, unemployed, other); family income (< poverty line, < 200% of poverty line, ≥ 200% of poverty line); health insurance (yes; no); and self-rated health (poor = 1 through excellent = 5).
Analysis
We used Stata/MP 19.5’s svy function (College Station, TX) with the subpop command in all analyses to account for NSDUH’s multi-stage, stratified sampling design to ensure that variance estimates incorporate the full sampling design. For this study’s 2-year pooled data set, we created adjusted person-level weights by dividing the final person-level analysis weights by the number of pooled years (two), following NSDUH guidelines for combining weights across multiple survey years. All estimates are weighted except for sample sizes. First, we used Pearson’s χ2 tests to compare the prevalence and 95% confidence intervals (CI) of MDE, suicidality, SUD, and depression treatment use in 2022 and 2023. We also examined treatment use among adults with MDE, suicidality, and SUD. Second, focusing on those who received depression treatment, we used Pearson’s χ2 tests and ANOVA to compare sociodemographic, self-rated health, and clinical characteristics among three groups: medication only, COT only, and MCOT. Third, we used Pearson’s χ2 tests to compare perceived helpfulness between medication and COT, comparing medication only with MCOT and COT only with MCOT.
Fourth, we fitted a multinomial logistic regression model to test H1 (associations between clinical characteristics and treatment modality [COT only and MCOT vs. medication only, and MCOT vs. COT only]). Fifth, for H2 (associations between clinical characteristics and perceived helpfulness of medication and COT), we fitted two generalized ordinal logistic regression models, using Stata’s gologit2 command, one for medication helpfulness and another for COT helpfulness, with the four-level ordinal response categories of helpfulness as the dependent variable. We fitted the models for those with and without MDE, and then for those with MDE, to further examine whether the associations may differ among those with MDE. Each model included a variable indicating whether the individual received both treatments. We used the gologit2 procedure because it tests the proportional odds assumption for each independent variable and allows for partial proportional odds modeling. Specifically, it estimates different odds ratios across thresholds for variables that violate the proportional odds assumption, while constraining odds ratios to be equal across thresholds for variables that meet the assumption [23]. Additionally, to enhance the interpretability of effect sizes, we estimated marginal predicted probabilities for key comparisons (i.e., the adjusted probabilities of rating treatment as “a lot” or “extremely” helpful). For supplementary sensitivity analysis, we also fitted logistic regression models with dichotomous (a lot of/extreme vs. other) helpfulness as the dependent variable. Multicollinearity among covariates was not a concern based on a cut-off of 2.50 using the variance inflation factor test [24]. We present multinomial logistic regression results as relative risk ratios (RRR) with 95% CI, generalized ordinal logistic regression results as the log-odds of the cumulative probabilities of the ordinal categories, shown as odds ratios (OR) with 95% CI, and binary logistic regression results as OR with 95% CI. Significance was set at p < 0 .05.
Results
Prevalence of MDE, suicidality, SUD, and depression care
Table 1 shows that the prevalence of MDE (8.7% in 2022 and 8.6% in 2023), suicidality (5.4% in 2022 and 5.1% in 2023), and SUD (8.2% in 2022 and 8.1% in 2023) did not significantly differ between the two years. Meanwhile, the prevalence of treatment (8.6% in 2022 and 9.3% in 2023), especially COT, was significantly higher in 2023 than in 2022. Treatment use among individuals with MDE increased from 61.2% in 2022 to 67.3% in 2023. However, treatment rates among those with suicidality (41.9% in 2022; 44.5% in 2023) and those with SUD (18.3% in 2022; 18.2% in 2023) remained low and showed no significant change over the two years.
Table 1.
Prevalence of adult major depressive episode, suicidality, substance use disorder, and depression treatment use in 2022 and 2023 (%)
| 2022 | 2023 | p | ||||
|---|---|---|---|---|---|---|
| Sample N | % (95% CI) | Sample N | % (95% CI) | |||
| Major depressive episode (MDE) | 47,100 | 8.7 (8.3–9.2) | 45,133 | 8.6 (8.2–9.6) | 0.622 | |
| Suicidality | 5.4 (5.0–5.8) | 5.1 (4.8–5.3) | 0.123 | |||
| Serious suicidal thoughts | 5.3 (4.9–5.6) | 4.9 (4.7–5.2) | 0.149 | |||
| Suicide planning | 1.5 (1.4–1.6) | 1.4 (1.3–1.6) | 0.587 | |||
| Suicide attempt | 0.7 (0.5–0.8) | 0.5 (0.5–0.6) | 0.139 | |||
| Alcohol or drug use disorder (SUD) | 47,100 | 8.2 (7.8–8.7) | 45,133 | 8.1 (7.7–8.5) | 0.726 | |
| Depression treatment use | 47,100 | 8.6 (8.1–9.1) | 45,133 | 9.3 (8.9–9.7) | 0.031 | |
| Prescription medication | 6.4 (5.9–6.8) | 7.0 (6.6–7.3) | 0.050 | |||
| Counseling/other treatmenta | 7.3 (6.9–7.8) | 7.9 (7.5–8.3) | 0.044 | |||
| Depression treatmentb use among adults who had | ||||||
| MDE | 5413 | 61.2 (58.4–63.9) | 5185 | 67.3 (65.2–69.2) | < 0.001 | |
| Suicidality | 3566 | 41.9 (38.6–45.3) | 3308 | 44.5 (41.9–47.2) | 0.185 | |
| SUD | 11,004 | 18.3 (16.6–20.2) | 10,832 | 18.2 (16.7–19.8) | 0.889 | |
aSaw/talked to a general practitioner/family doctor; other medical doctor like a cardiologist, gynecologist, urologist; psychologist; psychiatrist or psychotherapist; social worker; counselor; other mental health professional like a mental. health nurse; nurse, occupational therapist, or other health professional; religious advisor; another healer; herbalist, chiropractor, acupuncturist, or massage therapist; or another helping professional
bThe sample size for any depression treatment users in 2022–2023 was 10,221 (n = 5154 in 2022 and n = 5067 in 2023)
Probability values were calculated based on Pearson’s χ2 tests and denote the difference between 2022 and 2023
Characteristics of individuals by treatment modality
Table 2 shows that among individuals who received depression treatment in 2022 and 2023, 15.0% used medication only, 25.4% received COT only, and 59.6% used MCOT. Medication-only users included higher proportions of those 50 + years, non-Hispanic whites, those with income ≥ 200% of the poverty line, and those without SUD. COT-only users included a higher proportion of those < 50 years, Blacks and Hispanics, employed individuals, and those with “good” self-rated health. MCOT users consisted of a higher share of women than men and those with nicotine dependence.
Table 2.
Characteristics of adults who received past-year depression treatment by treatment modality
| Medication only 1290 (15.0%) |
COT only 2653 (25.4%) |
MCOT 6278 (59.6%) |
p | ||
|---|---|---|---|---|---|
| Survey year | 0.904 | ||||
| 2022 | 47.4 | 48.5 | 47.8 | ||
| 2023 | 52.6 | 51.5 | 52.2 | ||
| Age group (years, %) | < 0.001 | ||||
| 18–25 | 11.9 | 25.8 | 20.0 | ||
| 25–34 | 14.8 | 24.9 | 22.4 | ||
| 35–49 | 26.5 | 26.8 | 27.8 | ||
| 50–64 | 29.0 | 14.7 | 20.7 | ||
| 65+ | 18.0 | 7.8 | 9.1 | ||
| Gender (%) | 0.012 | ||||
| Female | 63.9 | 63.1 | 68.7 | ||
| Male | 36.1 | 36.9 | 31.3 | ||
| Race/ethnicity (%) | < 0.001 | ||||
| Non-hispanic white | 81.8 | 61.8 | 74.5 | ||
| Non-hispanic black | 3.3 | 13.4 | 6.3 | ||
| Hispanic | 9.0 | 16.0 | 13.5 | ||
| Other | 5.9 | 8.8 | 5.7 | ||
| College degree (%) | 39.8 | 38.5 | 36.1 | 0.303 | |
| Employment status (%) | < 0.001 | ||||
| Working full-time | 46.8 | 52.5 | 41.8 | ||
| Working part-time | 11.3 | 17.3 | 16.8 | ||
| Unemployed | 3.7 | 4.8 | 5.4 | ||
| Other | 38.2 | 25.4 | 36.0 | ||
| Income (%) | < 0.001 | ||||
| < Poverty line | 11.8 | 12.5 | 18.7 | ||
| < 200% poverty line | 19.5 | 21.6 | 21.2 | ||
| ≥ 200% poverty line | 68.7 | 65.7 | 60.1 | ||
| Health insurance coverage (%) | 93.4 | 91.5 | 94.6 | 0.037 | |
| Self-rated health (1, poor − 5, excellent), M (SE) | 3.13 (0.05) | 3.33 (0.03) | 3.05 (0.03) | < 0.001 | |
| Past-year major depressive episode (%) | 33.7 | 65.3 | 68.2 | < 0.001 | |
| Past year suicidality (%) | 12.5 | 20.7 | 30.4 | < 0.001 | |
| Past-year substance use disorder (%) | < 0.001 | ||||
| No substance use disorder | 66.4 | 64.9 | 61.6 | ||
| Alcohol use disorder only | 14.4 | 13.4 | 9.7 | ||
| Drug use disorder only | 13.9 | 13.8 | 19.6 | ||
| Both alcohol and drug use disorders | 5.3 | 8.0 | 9.1 | ||
| Nicotine dependence (%) | 8.2 | 7.5 | 11.9 | 0.005 | |
COT , Counseling/other treatment; MCOT , Medication & counseling/other treatment
Probability values were calculated based on Pearson’s χ2 tests for categorical variables and ANOVA for self-rated health
Among medication-only, COT-only, and MCOT users, 33.7%, 65.3%, and 68.2%, respectively, had MDE, with no significant difference between COT-only and MCOT users (F[1,50] = 1.79, p = 0.187). Additionally, 12.5%, 20.7%, and 30.4%, respectively, experienced suicidality, with significant differences among all groups (F[1,50] = 14.25, p < 0 .001 for medication-only vs. COT-only users; F[1,50] = 37.60, p < 0 .001 for COT-only vs. MCOT users). SUD rates did not differ significantly between medication-only and COT-only users (F[2.8,140.4] = 0.93, p = .0.425), but rates were significantly higher among MCOT users than among COT-only users (F[2.9,146.2] = 7.38, p < 0.001). Additional analysis showed that 42.2% of treatment users who had MDE, compared to 28.0% of those who did not have MDE, also had an SUD (11.7% alcohol use disorder only, 20.5% drug use disorder only, and 10.0% both alcohol and drug use disorders) (F(2.8, 141.9) = 29.83, p < 0.001).
Perceived treatment helpfulness
The top rows of Table 3 show that of all medication users, 15.3% perceived no/a little helpfulness of medication, 25.0% some helpfulness, 36.0% a lot of helpfulness, and 23.7% extreme helpfulness. A larger proportion of medication-only users (64.8%) compared to MCOT users (58.5%) perceived a lot of/extreme helpfulness of medication. Of all COT users, 26.8% perceived no/a little helpfulness of COT, 26.3% some helpfulness, 28.1% a lot of helpfulness, and 18.7% extreme helpfulness of COT. A larger proportion of MCOT users (50.0%), compared to COT-only users (39.5%), perceived a lot of/extreme helpfulness of COT.
Table 3.
Helpfulness of prescription medication and counseling/other treatment (%)
| Those with or without MDE | ||||||
|---|---|---|---|---|---|---|
| Prescription medication | COT | |||||
| All 7564 (100%) |
Medication only 1288 (20.1%) |
MCOT 6276 (79.9%) |
All 8920 (100%) |
COT only 2649 (29.9%) |
MCOT 6271 (70.1%) |
|
| Not at all | 5.3 | 5.4 | 5.3 | 12.7 | 19.0 | 10.0 |
| A little | 10.0 | 10.7 | 9.9 | 14.1 | 17.4 | 12.7 |
| Some | 25.0 | 19.1 | 26.5 | 26.3 | 24.1 | 27.3 |
| A lot | 36.0 | 35.6 | 36.2 | 28.1 | 22.6 | 30.5 |
| Extremely | 23.7 | 29.2 | 22.3 | 18.8 | 16.9 | 19.5 |
| p | = 0.022 | < 0.001 | ||||
| Those with MDE | ||||||
|---|---|---|---|---|---|---|
| Prescription medication | COT | |||||
| All 4,942 (100%) |
Medication only 493 (11.0%) |
MCOT 4449 (89.0. %) |
All 6212 (100%) |
COT only 1766 (29.0%) |
MCOT 4446 (71.0%) |
|
| Not at all | 6.3 | 8.3 | 6.1 | 13.4 | 21.2 | 10.2 |
| A little | 12.6 | 17.1 | 12.0 | 16.3 | 21.9 | 14.1 |
| Some | 30.4 | 29.4 | 30.5 | 27.4 | 23.5 | 29.0 |
| A lot | 34.2 | 36.2 | 34.0 | 27.6 | 20.1 | 30.6 |
| Extremely | 16.5 | 9.0 | 17.4 | 15.3 | 13.3 | 16.1 |
| p | = 0.066 | < 0.001 | ||||
COT , Counseling/other treatment; MCOT , Medication & counseling/other treatment
Probability values were calculated based on Pearson’s χ2 tests
The bottom rows of Table 3 show perceived helpfulness among those with MDE. Among medication users with MDE, 45.5% of medication-only users and 51.4% of MCOT users perceived a lot/extreme helpfulness. Among COT users with MDE, 33.4% of COT-only users and 46.7% of MCOT users perceived a lot/extreme helpfulness.
Associations between clinical characteristics and treatment modality: multinomial logistic regression results
Table 4 shows that MDE was associated with a higher likelihood of COT only (RRR = 3.73, 95% CI = 2.86–4.87) and MCOT (RRR = 3.54, 95% CI = 2.87–4.37), compared to medication only. However, MDE was not significantly associated with the likelihood of MCOT versus COT only. Suicidality was associated with a higher likelihood of MCOT versus medication only (RRR = 2.04, 95% CI = 1.49–2.80) or MCOT versus COT only (RRR = 1.81, 95% CI = 1.52–2.15). Alcohol use disorder was associated with a lower likelihood of MCOT versus medication only (RRR = 0.57, 95% CI = 0.38–0.87), while drug use disorder was associated with a higher likelihood of MCOT versus COT only (RRR = 1.38, 95% CI = 1.09–1.76). The findings partly support H1a and H1b.
Table 4.
Associations of types of depression treatment with major depressive episode (MDE), suicidality, and substance use disorder (SUD): multinomial logistic regression results
| COT only versus medication only RRR (95% CI) |
MCOT versus medication only RRR (95% CI) |
MCOT versus COT only RRR (95% CI) |
|||
|---|---|---|---|---|---|
| Past-year MDE versus No MDE | 3.73 (2.86–4.87)*** | 3.54 (2.87–4.37)*** | 0.95 (0.77–1.18) | ||
| Past-year suicidality versus no suicidality | 1.13 (0.78–1.64) | 2.04 (1.49–2.80)*** | 1.81 (1.52–2.15)*** | ||
| Past-year SUD type: versus No SUD | |||||
| Alcohol use disorder | 0.94 (0.56–1.60) | 0.57 (0.38–0.87)*** | 0.79 (0.61–1.02) | ||
| Drug use disorder | 0.72 (0.48–1.10) | 0.99 (0.69–1.42) | 1.38 (1.09–1.76)** | ||
| Both alcohol and drug use disorders | 0.72 (0.47–1.10) | 1.12 (0.78–1.61) | 1.19 (0.84–1.67) | ||
| Survey year 2023 versus 2022 | 0.94 (0.69–1.27) | 0.98 (0.76–1.27) | 1.05 (0.90–1.21) | ||
| Age group: versus 18–25 years | |||||
| 26–34 | 0.88 (0.63–1.24) | 1.25 (0.93–1.68) | 1.41 (1.09–1.83)* | ||
| 35–49 | 0.61 (0.41–0.89)* | 0.92 (0.66–1.28) | 1.52 (1.21–1.90)** | ||
| 50–64 | 0.34 (0.22–0.53)*** | 0.61 (0.43–0.86)** | 1.80 (1.35–2.39)*** | ||
| 65+ | 0.30 (0.18–0.52)*** | 0.43 (0.27–0.69)** | 1.42 (0.88–2.29) | ||
| Male versus female | 1.03 (0.81–1.31) | 0.77 (0.63–0.95)* | 0.75 (0.62–0.90)** | ||
| Non-hispanic white | |||||
| Non-hispanic black | 5.75 (2.97–11.12)*** | 1.97 (1.03–3.77)* | 0.34 (0.26–0.45)*** | ||
| Hispanic | 2.04 (1.29–3.24)** | 1.50 (0.98–2.28) | 0.73 (0.57–0.95)* | ||
| Other | 1.55 (0.92–2.64) | 0.79 (0.49–1.28) | 0.51 (0.38–0.68)*** | ||
| College degree: versus no degree | 1.13 (0.88–1.45) | 1.19 (0.92–1.55) | 1.06 (0.88–1.27) | ||
| Employment status: versus Working full-time | |||||
| Working part-time | 1.21 (0.87–1.69) | 1.51 (1.08–2.11)* | 1.24 (1.00-1.55) | ||
| Unemployed | 0.74 (0.41–1.34) | 1.05 (0.61–1.81) | 1.42 (0.95–2.12) | ||
| Other | 0.81 (0.58–1.11) | 1.19 (0.88–1.61) | 1.48(1.20–1.82) | ||
| Income: versus < Poverty line | |||||
| < 200% poverty line | 1.17 (0.76–1.80) | 0.78 (0.54–1.14) | 0.67 (0.52–0.86)** | ||
| ≥ 200% poverty line | 1.12 (0.74–1.68) | 0.78 (0.53–1.14) | 0.70 (0.52–0.94)* | ||
| Health insurance coverage | 0.87 (0.53–1.42) | 1.57 (0.96–2.56) | 1.80 (1.23–2.64)** | ||
| Self-rated health | 1.29 (1.13–1.47)*** | 1.07 (0.94–1.21) | 0.83 (0.76–0.91)*** | ||
| Nicotine dependence | 1.07 (0.69–1.67) | 1.32 (0.81–2.16) | 1.23 (0.89–1.71) | ||
| Model statistics | Sample N = 10,213; population N = 22.9 million; design df = 50; F(46,5) = 6.85; p = .020 | ||||
COT = Counseling/other treatment; MCOT = Medication & counseling/other treatment
*p < 0.05; **p < .0.001; ***p < 0 .001
Of the covariates, ages 50 + were associated with a lower likelihood of COT only and MCOT, compared to medication only. Ages 35–49 were also associated with a lower likelihood of COT only, and male gender was associated with a lower likelihood of MCOT. Being Black was associated with a significantly higher likelihood of COT only (RRR = 5.75, 95% CI = 2.97–11.12) and MCOT (RRR = 1.97, 95% CI = 1.03–3.77), compared to medication only. Part-time employment was associated with a higher likelihood of MCOT. Higher self-rated health was associated with a higher likelihood of COT only. Ages 26–64 years, compared to ages 18–25 years, and having health insurance were associated with a higher likelihood of MCOT versus COT only. On the other hand, male gender, racial/ethnic minority status, income above the poverty line, and higher self-rated health were associated with a lower likelihood of MCOT versus COT only.
Associations between clinical characteristics and perceived treatment helpfulness: generalized ordinal logistic regression results
The first two columns in Table 5 show that among all treatment users, MDE and suicidality were associated with lower odds of higher perceived helpfulness among medication users (OR = 0.47, 95% CI = 0.39–0.57 for MDE; OR = 0.78, 95% CI = 0.66–0.91 for suicidality) as well as COT users (OR = 0.67, 95% CI = 0.56–0.80 for MDE; OR = 0.82, 95% CI = 0.70–0.97 for suicidality). Having both alcohol and drug use disorders was also associated with lower odds of higher perceived helpfulness (OR = 0.72, 95% CI = 0.56–0.91) among COT users. The results generally support H2.
Table 5.
Associations of helpfulness of depression treatment with major depressive episode (MDE), suicidality, and substance use disorder (SUD): generalized ordinal logistic regression (Gologit2) results
| Those with or without MDE | Those with MDE | ||||
|---|---|---|---|---|---|
| Medication helpfulness OR (95% CI) | COT helpfulness OR (95% CI) | Medication helpfulness OR (95% CI) | COT helpfulness OR (95% CI) | ||
| Past-year MDE versus No MDE | 0.47 (0.39–0.57)*** | 0.67 (0.56–0.80)*** | |||
| Past-year suicidality versus No suicidality | 0.78 (0.66–0.91)** | 0.82 (0.70–0.97)* | 0.81 (0.67–0.98)* | 0.82 (0.69–0.97)* | |
| Past-year SUD type: versus No SUD | |||||
| Alcohol use disorder | 0.86 (0.67–1.09) | 1.05 (0.86–1.28) | 0.83 (0.62–1.11) | 1.06 (0.84–1.33) | |
| Drug use disorder | 0.97 (0.79–1.19) | 0.89 (0.74–1.08) | 0.87 (0.68–1.09) | 0.93 (0.75–1.15) | |
| Both alcohol and drug use disorders | 0.80 (0.57–1.12) | 0.72 (0.56–0.91)** | 0.80 (0.61–1.05) | 0.66 (0.51–0.87)** | |
| Both medication and counseling/other treatment vs. medication only | 1.55 (1.11–2.18)* | 1.42 (1.03–1.95)* | |||
| Both medication and counseling/other treatment vs. counseling/other treatment only | 2.25 (1.94–2.62)*** | 2.74 (2.23–3.36)*** | |||
| Survey year 2023 versus 2022 | 0.89 (0.77–1.02) | 1.00 (0.89–1.12) | 0.89 (0.73–1.09) | 0.96 (0.81–1.14) | |
| Age group versus 18–25 years | |||||
| 26–34 | 1.11 (0.90–1.37) | 1.09 (0.93–1.28) | 1.17 (0.88–1.56) | 0.94 (0.71–1.26) | |
| 35–49 | 1.26 (1.01–1.67)* | 0.93 (0.77–1.13) | 1.16 (0.91–1.48) | 0.88 (0.71–1.09) | |
| 50–64 | 2.00 (1.34–2.98)** | 0.88 (0.72–1.08) | 2.03 (1.29–3.17)** | 0.80 (0.61–1.05) | |
| 65+ | 0.89 (0.60–1.34) | 0.80 (0.58–1.10) | 0.99 (0.60–1.63) | 0.85 (0.56–1.28) | |
| Male versus female | 0.74 (0.63–0.87)** | 0.82 (0.69–0.98)* | 0.75 (0.60–0.93)** | 0.83 (0.69–0.99)* | |
| Non-hispanic white | |||||
| Non-hispanic black | 0.62 (0.42–0.91)* | 1.11 (0.85–1.46) | 0.72 (0.51–1.01) | 1.15 (0.84–1.57) | |
| Hispanic | 0.59 (0.40–0.87)** | 1.22 (0.98–1.54) | 0.63 (0.40–0.99)* | 1.23 (0.92–1.66) | |
| Other | 0.67 (0.50–0.90)** | 1.01 (0.80–1.29) | 0.70 (0.49–1.01) | 0.92 (0.65–1.30) | |
| College degree versus no degree | 1.06 (0.89–1.27) | 1.25 (1.10–1.43)** | 1.09 (0.90–1.32) | 1.35 (1.07–1.69)* | |
| Employment status: working full-time | |||||
| Working part-time | 0.84 (0.66–1.05) | 0.96 (0.76–1.23) | 0.82 (0.62–1.09) | 1.04 (0.74–1.46) | |
| Unemployed | 0.97 (0.63–1.49) | 0.98 (0.70–1.36) | 1.01 (0.67–1.53) | 1.09 (0.74–1.61) | |
| Other | 0.98 (0.76–1.26) | 1.14 (0.95–1.37) | 0.86 (0.66–1.12) | 1.29 (1.03–1.61)* | |
| Income: versus < poverty line | |||||
| < 200% poverty line | 0.91 (0.69–1.20) | 0.91 (0.72–1.15) | 0.99 (0.71–1.37) | 1.05 (0.77–1.42) | |
| ≥ 200% poverty line | 1.25 (1.00-1.54)* | 1.01 (0.82–1.25) | 1.23 (0.95–1.58) | 1.08 (0.85–1.37) | |
| Health insurance coverage | 1.15 (0.79–1.65) | 1.42 (1.12–1.80)** | 1.05 (0.66–1.69) | 1.42 (1.08–1.88)* | |
| Self-rated health | 1.24 (1.12–1.36)*** | 1.23 (1.14–1.33)*** | 1.24 (1.09–1.40)** | 1.28 (1.17–1.41)*** | |
| Nicotine dependence | 0.72 (0.51–1.02) | 0.79 (0.60–1.04) | 0.72 (0.47–1.10) | 0.85 (0.59–1.23) | |
| Model statistics |
Sample N = 7557; population N = 17.1 million; design df = 50; F(32,19) = 14.30; p < .001 |
Sample N = 8912; population N = 19.5 million; design df = 50; F(26,25) = 9.20; p < .001 |
Sample N = 4935; population N = 10.5 million; design df = 50; F(27,24) = 4.65; p < .001 |
Sample N = 6205; population N = 13.1 million; design df = 50; F(31,20) = 5.76; p < .001 |
|
COT , Counseling/other treatment
*p < 0 .05; **p < 0 .01; ***p < .0.001
Gologit2 procedures used partial proportional odds modeling for the following categories of variables that violated proportional odds assumptions: Combined treatment receipt, the 50-64 age group, Hispanic ethnicity, and other work status in the medication helpfulness model, and combined treatment receipt in the counseling/other treatment helpfulness model for those with or without MDE; and the 50-64 age group and Hispanic ethnicity in the medication helpfulness model, and Combined treatment receipt, the 26-34 age group, college degree, and working part-time in the counseling/other treatment helpfulness model for those with MDE
Of the covariates, MCOT and higher self-rated health were associated with greater odds of higher perceived helpfulness among medication users (OR = 1.55, 95% CI = 1.11–1.28 for MCOT; OR = 1.24, 95% CI = 1.12–1.36 for self-rated health) as well as COT users (OR = 2.25, 95% CI = 1.94–2.62 for MCOT; OR = 1.23, 95% CI = 1.14–1.33 for self-rated health). Male gender was associated with lower odds of perceiving higher helpfulness for medication as well as COT. Additionally, ages 35–49, compared to 18–25 years, and income ≥ 200% of the poverty line, compared to income below the poverty line, were associated with greater odds of higher perceived medication helpfulness, whereas racial/ethnic minority status was associated with lower odds. For COT helpfulness, a college degree and having health insurance were associated with greater odds of higher perceived helpfulness.
The second two columns of Table 5 show that the results for those with MDE did not differ significantly from the results for those with or without MDE.
Predicted probabilities indicated modest differences by suicidality. Among individuals with MDE, those without suicidality had slightly higher adjusted probabilities of rating medication as “a lot” helpful (35.0% vs. 32.7%) or “extremely” helpful (17.4% vs. 14.7%). For COT, those without suicidality likewise showed higher probabilities of endorsing “a lot” (28.4% vs. 26.2%) or “extremely” helpfulness (16.2% vs. 13.7%) compared with those with suicidality.
Associations between clinical characteristics and perceived treatment helpfulness: binary logistic regression results
Binary logistic regression models examining “a lot of/extreme” helpfulness (vs. all other responses), presented in Supplemental Table 1, show several consistent patterns across treatments. MDE and co-occurring alcohol and drug use disorders were associated with lower odds of perceiving both medication and COT as highly helpful. Suicidality and alcohol use disorder were also associated with lower odds of rating medication as highly helpful, although neither was associated with COT helpfulness. Use of medication and COT was associated with greater odds of perceiving COT as highly helpful. Male gender and Black race were associated with lower odds of reporting high medication helpfulness, whereas Hispanic ethnicity and having health insurance were associated with greater odds of perceiving COT as highly helpful. Finally, higher self-rated health was associated with greater odds of reporting high helpfulness for both medication and COT.
Discussion
This study shows that in 2022 and 2023, just under 9% and just over 5% of the U.S. adult population experienced MDE and suicidality, respectively, and a little over 8% had any SUD. The proportion of those who received depression treatment (medication only, COT only, or MCOT) increased from 8.6% in 2022 to 9.3% in 2023. Nearly 60% of those who received treatment reported MCOT, while 15% used medication only, and 25% used COT, which included heterogeneous providers (specialty, general medical, and complementary). Nearly 60% of medication users and nearly 47% of COT users perceived their respective treatments as a lot or extremely helpful.
Multivariable analyses showed that individuals with MDE, compared to those without MDE, were significantly more likely to use COT or MCOT rather than medication alone. Those with suicidality were also more likely to use MCOT, suggesting that the receipt of combined treatment may have been in response to greater treatment needs. On the other hand, individuals with alcohol use disorder were less likely to use MCOT. Relative to young adults, middle-aged and older adults were less likely to use COT, while Black and Hispanic individuals were more likely to do so. The analysis further indicated that both MDE and suicidality were associated with lower odds of perceiving treatment, whether medication or COT, as helpful. Co-occurring alcohol and drug use disorders were also associated with lower odds of perceiving COT as helpful. The use of both medication and COT was associated with higher odds of perceiving either modality as helpful. These patterns did not differ meaningfully among individuals with MDE.
Understandably, individuals with MDE and/or suicidality may be less likely to see their treatment as helpful, since those with more severe symptoms might respond less to treatment, especially short-term interventions, and feel the treatment was not adequate. Because self-rated health is positively associated with perceived treatment helpfulness, people with higher symptom burden and poorer self-rated health might perceive their treatment as less effective. Supporting this, a study of mental health patients in Quebec found a negative link between unmet needs and treatment satisfaction [25]. However, previous research on individuals with suicidal thoughts or behaviors shows mixed results, depending on the sample and treatment setting. For example, Hom et al. [26] found no significant difference in treatment satisfaction between those with and without suicidal ideation in a partial hospitalization program, with both groups rating treatment as “quite a bit” helpful on average. In contrast, a study of transgender and gender-diverse adults hospitalized for suicidality found that over half rated their hospital-based care as unhelpful to some degree.
The association between co-occurring alcohol and drug use disorders and lower perceived helpfulness of COT also likely reflects the complex and often unmet care needs experienced by this group. Prior research shows that antidepressant effectiveness among individuals with comorbid major depression and SUD is inconsistent and often conflicting [27–30]. Individuals with co-morbid mental disorders and SUD typically exhibit more severe symptom profiles and greater health-related social needs (e.g., unemployment, housing instability), yet they often receive less adequate depression treatment, experience poorer treatment continuity, and report higher unmet treatment needs [31–33]. A systematic review further indicates that people with dual mental and substance use disorders frequently experience cognitive impairments, particularly in attention and executive functioning [34], necessitating comprehensive and integrated psychiatric and addiction care. Collectively, these disparities in access, continuity, and quality of mental health and addiction services likely contribute to the lower perceived helpfulness of treatment reported by individuals with comorbid MDE and SUD.
Our finding that individuals who received MCOT, meaning both medication and COT, perceived either treatment as more helpful aligns with prior research among adults with MDE [16, 18]. A meta-analysis and a network meta-analysis have shown that combined treatment with medication and psychotherapy produces better outcomes than either modality alone [35, 36]. The higher perceived helpfulness among MCOT users may reflect synergistic benefits from combined treatments, suggesting that perceived helpfulness could serve as a marker of improved clinical effectiveness.
Our findings also indicate that men perceive both treatment options as less helpful. Prior research consistently shows that men are less likely to seek help for mental health issues, often because of reluctance to express emotions and health concerns, embarrassment, anxiety, fear, and difficulty communicating with healthcare providers [37, 38]. These help-seeking barriers may influence men’s views on treatment effectiveness, as such views are shaped by underlying beliefs about mental health care [39]. Additionally, women generally experience greater improvement than men [40]. Age-related differences in perceived medication helpfulness may also reflect known variations in pharmacological responses across age groups. A meta-analysis of patients with major depression found that those under age 21 and over 55 had slower and less effective responses to antidepressants compared to those 21–35 years, and gender differences in treatment response did not significantly change across age groups [40].
The significant association between racial/ethnic minority status and lower perceived helpfulness of medication aligns with previous research showing greater skepticism and ambivalence toward psychiatric medications in racial/ethnic minority groups [41–44]. These attitudes may stem not only from cultural beliefs and historical mistrust but also from clinical and contextual factors such as lower treatment effectiveness in some racial/ethnic groups and disparities in access to care, including insurance coverage [45, 46]. The positive link between income and perceived helpfulness of medication indicates that affordability influences perceptions of treatment helpfulness. The finding that college education and health insurance coverage are positively associated with perceived helpfulness of COT also underscores the impact of socioeconomic and structural factors on treatment experiences and perceived value.
This study has several limitations. First, because the study relies on cross-sectional epidemiological data, the observed relationships are correlational and cannot establish causation. Second, NSDUH does not provide information on treatment history or duration, which constrains our ability to conduct more nuanced analyses of perceived helpfulness. Third, the survey data lack detailed clinical information, such as episode severity and duration, medication dose or number of therapy sessions, antidepressant class, and adherence, all of which meaningfully influence both clinical outcomes and perceived treatment helpfulness. Fourth, the NSDUH’s COT combines multiple heterogeneous treatment modalities delivered across specialty and non-specialty settings. Because the helpfulness questions did not distinguish among these modalities or settings, we were unable to examine potential differences in perceived benefit across specific forms of care. Fifth, the NSDUH does not differentiate whether MDE is primary or secondary to substance use. This distinction is clinically important, as individuals with primary depression co-occurring with SUD (“dual disorders”) often experience more severe symptoms, greater functional and cognitive impairment, and more complex treatment needs than those whose depressive symptoms are substance-induced [34]. The inability to differentiate these subgroups may influence perceived treatment helpfulness and potentially obscure important variation in treatment experiences. Sixth, data on patient treatment preferences were unavailable, despite evidence that such preferences shape satisfaction, adherence, and outcomes [47]. Future research should incorporate more granular information on treatment type, intensity, and patient preferences to better elucidate treatment-specific determinants of perceived helpfulness.
Despite these limitations, the findings have important implications for enhancing depression treatment engagement and outcomes. The associations of lower perceived helpfulness of treatment with MDE, suicidality, and SUD highlight the need for more intensive, personalized, and ongoing interventions for those with complex clinical presentations. Given the greater perceived helpfulness of MCOT, expanding access to integrated care models that provide both pharmacotherapy and counseling or other treatments may improve patient satisfaction and clinical results. Tailoring treatment approaches to address gender, racial/ethnic, and socioeconomic disparities is also crucial. The lower perceived helpfulness among men and Black, Hispanic, and other racial/ethnic groups for medication suggests a need for culturally sensitive engagement strategies and trust-building efforts. Additionally, the role of affordability, as reflected by income, education, and insurance status, underscores the importance of reducing structural barriers to high-quality, acceptable care. These findings support initiatives to enhance the personalization, accessibility, and cultural responsiveness of mental health services to increase both perceived and actual effectiveness of depression treatment. Finally, further research on perceived helpfulness and its links to treatment engagement and outcomes is necessary to improve access to mental health services for those with high symptom burdens. Research is also needed to compare treatment recipients’ perceived helpfulness with objective assessments of clinical effectiveness. While patient perspectives are essential for understanding engagement, satisfaction, and perceived benefit, they may not always align with clinical indicators of treatment adequacy, symptom improvement, or therapeutic progress. Accordingly, perceived helpfulness should be interpreted alongside measures of actual treatment effectiveness to provide a more complete understanding of depression care quality.
Supplementary Information
Below is the link to the electronic supplementary material.
Author contributions
study conceptualization: NGC & CNM; data management: NGC; data analysis: NGC; interpretation: NGC & CNM; manuscript draft: NGC; final editing: NGC & CNM.
Funding
The authors did not receive any funding for this study.
Data availability
The data used in this study (The National Survey on Drug Use and Health) are publicly available: [https://www.samhsa.gov/data/data-we-collect/nsduh-national-survey-drug-use-and-health/datafiles](https:/www.samhsa.gov/data/data-we-collect/nsduh-national-survey-drug-use-and-health/datafiles).
Declarations
Ethics approval and consent to participate
The University of Texas at Austin’s Institutional Review Board approved this study (approval # STUDY00007908) of the secondary analysis of public-domain data. Not applicable.
Consent for publication
Both authors agree to publication.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
20
- 1.Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–13. 10.1046/j.1525-1497.2001.016009606.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Brody DJ, Hughes JP. Depression prevalence in adolescents and adults: United States, August 2021–August 2023. NCHS Data Brief. 2025;(527):1–8. Available from: https://www.cdc.gov/nchs/data/databriefs/db527.pdf. [DOI] [PMC free article] [PubMed]
- 3.Goodwin RD, Dierker LC, Wu M, Galea S, Hoven CW, Weinberger AH. Trends in U.S. depression prevalence from 2015 to 2020: the widening treatment gap. Am J Prev Med. 2022;63(5):726–33. 10.1016/j.amepre.2022.05.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Singichetti B, Wang J, Lee R, Ballesteros MF, Mack KA. Notes from the field: suicidal thoughts and knowing someone who died by suicide among adults-United States, 2023. MMWR Morb Mortal Wkly Rep. 2025;74(12):213–5. 10.15585/mmwr.mm7412a4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ivey-Stephenson AZ, Crosby AE, Hoenig JM, Gyawali S, Park-Lee E, Hedden SL. Suicidal thoughts and behaviors among adults aged ≥ 18 years -United States, 2015–2019. MMWR Surveill Summ. 2022;71(1):1–19. 10.15585/mmwr.ss7101a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Samples H, Cruz N, Corr A, Akkas F. National trends and disparities in suicidal ideation, attempts, and health care utilization among U.S. adults. Psychiatr Serv. 2025;76(2):110–9. 10.1176/appi.ps.20230466. [DOI] [PubMed] [Google Scholar]
- 7.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. Rockville (MD): center for behavioral health statistics and quality, substance abuse and mental health services administration. 2024. Report No.: PEP24-07-021, NSDUH Series H-59. Available from: https://www.samhsa.gov/data/report/2023-nsduh-annual-national-report.
- 8.Substance Abuse and Mental Health Services Administration. Co-occurring mental health and substance use services. Rockville (MD): center for mental health services, substance abuse and mental health services administration. 2024. Report No.: PEP24-01-008. Available from: https://library.samhsa.gov/sites/default/files/issue-brief-co-occurring-pep24-01-008.pdf.
- 9.National Institute on Drug Abuse. Co-occurring disorders and health conditions. Bethesda (MD): National Institute on Drug Abuse; 2024. Available from: https://nida.nih.gov/research-topics/co-occurring-disorders-health-conditions#mental.
- 10.Na PJ, Bommersbach TJ, Petrakis IL, Rhee TG. National trends of suicidal ideation and mental health services use among US adults with opioid use disorder, 2009–2020. EClinicalMedicine. 2022;54:101696. 10.1016/j.eclinm.2022.101696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Choi NG, Marti CN, Choi BY. Mental health treatment use, perceived treatment need, and reasons for non-use among U.S. adults with serious suicidal thoughts during the COVID-19 pandemic. Community Ment Health J. 2024;60(5):972–84. 10.1007/s10597-024-01249-8. [DOI] [PubMed] [Google Scholar]
- 12.Kuramoto-Crawford SJ, Han B, Jacobus-Kantor L, Mojtabai R. Differences in patients’ perceived helpfulness of depression treatment provided by general medical providers and specialty mental health providers. Gen Hosp Psychiatry. 2015;37(4):340–6. 10.1016/j.genhosppsych.2015.04.006. [DOI] [PubMed] [Google Scholar]
- 13.Han B, Olfson M, Mojtabai R. Depression care among depressed adults with and without comorbid substance use disorders in the United States. Depress Anxiety. 2017;34(3):291–300. 10.1002/da.22592. [DOI] [PubMed] [Google Scholar]
- 14.Cipriani A, Furukawa TA, Salanti G, Chaimani A, Atkinson LZ, Ogawa Y, et al. Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: a systematic review and network meta-analysis. Lancet. 2018;391(10128):1357–66. 10.1016/S0140-6736(17)32802-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Simon GE, Moise N, Mohr DC. Management of depression in adults: a review. JAMA. 2024;332(2):141–52. 10.1001/jama.2024.5756. [DOI] [PubMed] [Google Scholar]
- 16.Alang S, McAlpine D. Treatment modalities and perceived effectiveness of treatment among adults with depression. Health Serv Insights. 2020;13:1178632920918288. 10.1177/1178632920918288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Harris MG, Kazdin AE, Chiu WT, Sampson NA, Aguilar-Gaxiola S, Al-Hamzawi A, et al. Findings from world mental health surveys of the perceived helpfulness of treatment for patients with major depressive disorder. JAMA Psychiatry. 2020;77(8):830–41. 10.1001/jamapsychiatry.2020.1107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Husky MM, Léon C, du Roscoät E, Vasiliadis HM. Perceived helpfulness of medication and psychotherapy among adults with suicidal ideation. Int Rev Psychiatry. 2024;36(4–5):503–7. 10.1080/09540261.2024.2318252. [DOI] [PubMed] [Google Scholar]
- 19.Han B, Olfson M, Mojtabai R. Depression care among adults with co-occurring major depressive episodes and substance use disorders in the United States. J Psychiatr Res. 2017;91:47–56. 10.1016/j.jpsychires.2017.02.026. [DOI] [PubMed] [Google Scholar]
- 20.Substance Abuse and Mental Health Services Administration. Center for behavioral health statistics and quality. 2023 National survey on drug use and health (NSDUH): methodological summary and definitions. Rockville (MD): substance abuse and mental health services administration; 2024. Available from: https://www.samhsa.gov/data/report/2023-methodological-summary-and-definitions.
- 21.American Psychiatric Association. Diagnostic and statistical manual of mental disorders (5th ed.) 2013. 10.1176/appi.books.9780890425596.
- 22.Heatherton TF, Kozlowski LT, Frecker RC, Fagerström KO. The Fagerström test for nicotine dependence: a revision of the Fagerström tolerance questionnaire. Br J Addict. 1991;86(9):1119–27. 10.1111/j.1360-0443.1991.tb01879.x. [DOI] [PubMed] [Google Scholar]
- 23.Williams R. Understanding and interpreting generalized ordered logit models. J Math Sociol. 2016;40(1):7–20. 10.1080/0022250X.2015.1112384. [Google Scholar]
- 24.Allison P. How can you safely ignore multicollinearity?. Statistical horizons. 2012;5(1):1–2. [Google Scholar]
- 25.Fortin M, Bamvita JM, Fleury MJ. Patient satisfaction with mental health services based on andersen’s behavioral model. Can J Psychiatry. 2018;63(2):103–14. 10.1177/0706743717737030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hom MA, Bowers EM, Björgvinsson T, Beard C. Comparing treatment outcomes and patient satisfaction among individuals presenting to psychiatric care with and without recent suicidal ideation. J Clin Psychol. 2021;77(11):2455–72. 10.1002/jclp.23197. [DOI] [PubMed] [Google Scholar]
- 27.Agabio R, Trogu E, Pani PP. Antidepressants for the treatment of people with co-occurring depression and alcohol dependence. Cochrane Database Syst Rev. 2018;2018(4):CD008581. 10.1002/14651858.CD008581.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.De Filippis S, Martinotti G, Nicoletti F, Mastrostefano A, Trovini G, Pugliese A, Di Nicola M. Major depression in comorbidity with substance use disorders: patients’ features and clinical-neurobiological rationale of antidepressant treatments. Curr Neuropharmacol. 2025;23(3):256–75. 10.2174/1570159X22666240827165327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hides L, Quinn C, Stoyanov S, Kavanagh D, Baker A. Psychological interventions for co-occurring depression and substance use disorders. Cochrane Database Syst Rev. 2019;2019(11):CD009501. 10.1002/14651858.CD009501.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Torrens M, Tirado-Muñoz J, Fonseca F, Farré M, González-Pinto A, Arrojo M, et al. Clinical practice guideline on pharmacological and psychological management of adult patients with depression and a comorbid substance use disorder. Adicciones. 2022;34(2):128–41. 10.20882/adicciones.1559. [DOI] [PubMed] [Google Scholar]
- 31.Coughlin LN, Pfeiffer P, Ganoczy D, Lin LA. Quality of outpatient depression treatment in patients with comorbid substance use disorder. Am J Psychiatry. 2021;178(5):414–23. 10.1176/appi.ajp.2020.20040454. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Chen LY, Crum RM, Martins SS, Kaufmann CN, Strain EC, Mojtabai R. Service use and barriers to mental health care among adults with major depression and comorbid substance dependence. Psychiatr Serv. 2013;64(9):863–70. 10.1176/appi.ps.201200289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Adan A, Torrens M. Special issue: diagnosis and management of addiction and other mental disorders (dual disorders). J Clin Med. 2021;10(6):1307. 10.3390/jcm10061307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Miguel N, Marquez-Arrico JE, Jodar M, Navarro JF, Adan A. Neuropsychological functioning of patients with major depression or bipolar disorder comorbid to substance use disorders: a systematic review. Eur Neuropsychopharmacol. 2023;75:41–58. 10.1016/j.euroneuro.2023.06.006. [DOI] [PubMed] [Google Scholar]
- 35.Cuijpers P, Noma H, Karyotaki E, Vinkers CH, Cipriani A, Furukawa TA. A network meta-analysis of the effects of psychotherapies, pharmacotherapies and their combination in the treatment of adult depression. World Psychiatry. 2020;19(1):92–107. 10.1002/wps.20701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Karyotaki E, Smit Y, Holdt Henningsen K, Huibers MJ, Robays J, de Beurs D, Cuijpers P. Combining pharmacotherapy and psychotherapy or monotherapy for major depression? A meta-analysis on the long-term effects. J Affect Disord. 2016;194:144–52. 10.1016/j.jad.2016.01.036. [DOI] [PubMed] [Google Scholar]
- 37.Liddon L, Kingerlee R, Barry JA. Gender differences in preferences for psychological treatment, coping strategies, and triggers to help-seeking. Br J Clin Psychol. 2018;57(1):42–58. 10.1111/bjc.12147. [DOI] [PubMed] [Google Scholar]
- 38.Yousaf O, Grunfeld EA, Hunter MS. A systematic review of the factors associated with delays in medical and psychological help-seeking among men. Health Psychol Rev. 2015;9(2):264–76. 10.1080/17437199.2013.840954. [DOI] [PubMed] [Google Scholar]
- 39.Jorm AF, Christensen H, Medway J, Korten AE, Jacomb PA, Rodgers B. Public belief systems about the helpfulness of interventions for depression: associations with history of depression and professional help-seeking. Soc Psychiatry Psychiatr Epidemiol. 2000;35(5):211–9. 10.1007/s001270050230. [DOI] [PubMed] [Google Scholar]
- 40.Strawn JR, Mills JA, Suresh V, Mayes T, Gentry MT, Trivedi M, Croarkin PE. The impact of age on antidepressant response: a mega-analysis of individuals with major depressive disorder. J Psychiatr Res. 2023;159:266–73. 10.1016/j.jpsychires.2023.01.043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Cooper LA, Gonzales JJ, Gallo JJ, Rost KM, Meredith LS, Rubenstein LV, et al. The acceptability of treatment for depression among African-American, hispanic, and white primary care patients. Med Care. 2003;41(4):479–89. 10.1097/01.MLR.0000053228.58042.E4. [DOI] [PubMed] [Google Scholar]
- 42.Green BL, Watson MR, Kaltman SI, Serrano A, Talisman N, Kirkpatrick L, Campoli M. Knowledge and preferences regarding antidepressant medication among depressed Latino patients in primary care. J Nerv Ment Dis. 2017;205(12):952–9. 10.1097/NMD.0000000000000754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Schnittker J. Misgivings of medicine? African Americans’ skepticism of psychiatric medication. J Health Soc Behav. 2003;44(4):506–24. [PubMed] [Google Scholar]
- 44.Vargas SM, Cabassa LJ, Nicasio A, De La Cruz AA, Jackson E, Rosario M, et al. Toward a cultural adaptation of pharmacotherapy: Latino views of depression and antidepressant therapy. Transcult Psychiatry. 2015;52(2):244–73. 10.1177/1363461515574159. [DOI] [PubMed] [Google Scholar]
- 45.Marazziti D, Mucci F, Avella MT, Palagini L, Simoncini M, Dell’Osso L. The increasing challenge of the possible impact of ethnicity on psychopharmacology. CNS Spectr. 2021;26(3):222–31. 10.1017/S1092852920000073. [DOI] [PubMed] [Google Scholar]
- 46.Olagunju AT, Wang J, Edet B, Onwuameze OE, Macaluso M. Racial and ethnic considerations for the clinical practice of psychopharmacology and research methodology: a narrative review of the growing body of literature. J Psychiatr Pract. 2025;31(2):56–64. 10.1097/PRA.0000000000000845. [DOI] [PubMed] [Google Scholar]
- 47.Lindhiem O, Bennett CB, Trentacosta CJ, McLear C. Client preferences affect treatment satisfaction, completion, and clinical outcome: a meta-analysis. Clin Psychol Rev. 2014;34(6):506–17. 10.1016/j.cpr.2014.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data used in this study (The National Survey on Drug Use and Health) are publicly available: [https://www.samhsa.gov/data/data-we-collect/nsduh-national-survey-drug-use-and-health/datafiles](https:/www.samhsa.gov/data/data-we-collect/nsduh-national-survey-drug-use-and-health/datafiles).