This cross-sectional study assesses the likelihood of depression treatment in adults after the 2016 US Preventive Services Task Force recommendation for depression screening.
Key Points
Question
Is the 2016 US Preventive Services Task Force grade B recommendation for routine adult depression screening associated with the likelihood of depression treatment among adults?
Findings
This cross-sectional study with a policy evaluation in 372 962 adults found that the recommendation was associated with an increased likelihood of psychotherapy and decreased likelihood of antidepressant use among adults reporting some level of psychological distress, no changes among adults reporting no distress, and decreased probability of seeing a prescriber for psychotherapy.
Meaning
These findings suggest a positive association between routine adult depression screening and psychotherapy among adults in need of treatment and no association with unnecessary treatment among adults with low levels of distress.
Abstract
Importance
Fewer than one-third of adults with depression receive any treatment for depression. It is important to evaluate the association between the 2016 US Preventive Services Task Force (USPSTF) recommendation and depression treatment nationally.
Objective
To examine changes in the likelihood of psychotherapy and antidepressant use after the USPSTF-recommended routine adult depression screening.
Design, Setting, and Participants
This policy evaluation used a single-group, interrupted time series design with logistic regression. Medical Expenditure Panel Survey Household Component data from 2008 to 2023 were analyzed from November 3, 2025, to March 31, 2026. The sample included adults aged 18 years and older.
Exposure
2016 USPSTF grade B recommendation for routine adult depression screening.
Main Outcomes and Measures
Likelihood of psychotherapy or antidepressant use immediately (level change) and over time (post-USPSTF slope change [trend]) using an interrupted time series design and logistic regression models.
Results
The analysis included 372 962 adults. Using the Medical Expenditure Panel Survey person-level survey weights, this corresponded to a nationally representative sample of 244 316 623 adults (median age, 47 years; IQR, 32-61 years; 51.6% female). The recommendation was associated with a significant level increase (odds ratio [OR], 1.18; 95% CI, 1.02-1.37) and post-USPSTF trend (OR, 1.08; 95% CI, 1.04-1.11) increase in the likelihood of psychotherapy only, level (OR, 0.88; 95% CI, 0.82-0.94) and post-USPSTF trend (OR, 0.98; 95% CI, 0.96-0.99) decreases in the likelihood of antidepressants only, and no significant changes in combined treatment among adults overall. By psychological distress level, it was associated with a significant level increase in the likelihood of psychotherapy only (OR, 1.63; 95% CI, 1.10-2.41) and post-USPSTF trend decrease in the likelihood of antidepressants only (OR, 0.94; 95% CI, 0.89-0.98) among adults reporting serious distress, significant level (OR, 1.38; 95% CI, 1.15-1.66) and post-USPSTF trend (OR, 1.05; 95% CI, 1.01-1.10) increases in the probability of psychotherapy only, and level (OR, 0.88; 95% CI, 0.81-0.96) and post-USPTF trend (OR, 0.96; 95% CI, 0.94-0.98) decreases in the likelihood of antidepressants only among adults reporting mild or moderate distress, and no significant changes among adults reporting no distress. There was a significant level decrease (OR, 0.79; 95% CI, 0.70-0.89) in the probability of seeing a prescriber for psychotherapy after the recommendation.
Conclusions and Relevance
This study found that the USPSTF recommendation for routine adult depression screening was associated with increased likelihood of psychotherapy and decreased likelihood of antidepressant use among adults reporting some level of psychological distress. Findings suggest no association with unnecessary treatment among adults with low levels of distress.
Introduction
Depression is common among US adults, and its prevalence increased during the COVID-19 pandemic.1,2 Depression is one of the leading causes of disability and is associated with both adverse health and economic outcomes.3,4,5 Yet, fewer than one-third of adults with depression receive any depression treatment.6 Barriers to access of depression treatment include stigma, cost, mental health clinician availability, and differential provision of both diagnoses and treatment.7,8,9 These barriers affect multiple points in the care continuum from screening, diagnosis, and referral to treatment-seeking and receipt of care.
Multiple federal policies have been enacted to address barriers and improve access to mental health treatment by reducing cost-sharing and increasing coverage to parity with medical care.10,11,12 In 2016, the US Preventive Services Task Force (USPSTF) issued a grade B recommendation for routine adult depression screening to advance universal access to screening, diagnosis, and referral to treatment.13 This was an expansion of its 2009 recommendation for nonpregnant adults, which advised against routine adult depression screening in the absence of staff-assisted supports.14 Under the Patient Protection and Affordable Care Act (ACA), private health insurance companies must provide coverage with no patient cost-sharing for USPSTF grade A or B recommendations.12 In addition, the ACA incentivized state Medicaid programs to cover all USPSTF-recommended grade A or B preventive services with a 1–percentage point increase in their federal match rate.15 The Centers for Medicare & Medicaid also authorized no-cost coverage for annual depression screening among Medicare beneficiaries.16
The effects of these policies, however, remain unclear. The 2009 USPSTF recommendation was associated with increased depression screening, but estimated rates among adults remained below 6.5% prior to the 2016 expansion.17,18,19 It is not clear whether the 2016 USPSTF recommendation for routine depression screening led to substantial or sustained rate increases over time.18 Routine screening may help reduce disparities in depression screening, and screening is associated with increased depression identification and treatment.18,19,20,21 Evidence is mixed, however, about whether routine screening leads to more accurate diagnoses, and some clinicians voiced concerns that implementing the 2016 USPSTF recommendation would result in false-positive results, unnecessary treatment, and inappropriate antidepressant prescribing practices.22,23,24 Evaluations of parity policies also found mixed results for whether mental health service use increased after implementation.25,26,27,28,29,30 A previous analysis,31 however, found that cost-sharing parity combined with routine depression screening was associated with increases in outpatient mental health service use among Medicare beneficiaries with depression.
Researchers have called for studies to examine impact of routine screening on depression treatment. To our knowledge, no study has yet evaluated the impact of the 2016 USPSTF recommendation on depression treatment nationally. Psychotherapy and antidepressant medications are recommended as single-modality and combination treatments for depression.32 The purpose of this study was to examine whether the likelihood of psychotherapy and antidepressant use among US adults changed after the 2016 USPSTF recommendation for routine adult depression screening. We hypothesized a priori that the 2016 recommendation would promote increased screening in primary care settings, which, in turn, would lead to increased antidepressant prescribing by primary care practitioners (PCPs).33 We also hypothesized a smaller but positive increase in psychotherapy among those with positive screening results who were referred to specialty practitioners since demand would increase but use would be limited by mental health practitioner shortages.34,35
Methods
Data Source
This cross-sectional study analyzed publicly available data from the Medical Expenditure Panel Survey (MEPS) Household Component files from 2008 to 2023. Emory University determined that this study did not require institutional review board review because it is not research with human participants, nor is it a clinical investigation as defined in the federal regulations. The study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. The MEPS sample includes noninstitutionalized US citizens who participated in the previous year’s National Health Interview Survey.36 Household survey participants are interviewed about their health, health care utilization, and expenditures. We constructed a pooled analytic dataset as MEPS supports valid pooling of multiple years despite an overlapping panel design.37 We applied MEPS person-level weights adjusted for the pooled period and the pooled variance structure to generate nationally representative estimates, address potential issues in sample independence and correlations, and calculate appropriate standard errors.37,38
Outcome Measures
Our outcomes of interest included psychotherapy and antidepressant use. We defined psychotherapy as outpatient or office-based visits coded as Psychotherapy/Mental Health Counseling using the variable VSTCTGRY in the MEPS Event files. We identified antidepressants using the variable RXDRGNAM in the MEPS Prescription Medication files and specific drug names (eAppendix A in Supplement 1). We assessed use of psychotherapy only, antidepressants only, and a combination of both as 3 mutually exclusive treatment categories.
Statistical Analysis
We used an interrupted time series (ITS) design and estimated logistic regression models to assess changes in the likelihood of psychotherapy and antidepressant use associated with the USPSTF recommendation immediately in 2016 (level change) and over time from 2016 to 2023 (post-USPSTF slope change [trend]) (eAppendix B in Supplement 1). Our unit of analysis was person-year. We examined changes in the probability of treatment by category among adults overall and stratified by sociodemographic characteristics commonly associated with differences in mental health service use, including age, sex, race and ethnicity, health insurance, educational attainment, family income, US region, and self-reported psychological distress. Psychological distress was identified using the variable K6SUM42 in the MEPS Full-year Consolidated files, which provides a summed Kessler index (K6) score based on survey participant responses rating their feelings during the past 30 days.36 The K6 scale is a validated measure used to assess nonspecific psychological distress.39,40 In accordance with previous studies, scores of 13 or greater were classified as serious psychological distress, scores from 1 to 12 as mild or moderate, and scores of 0 as none.41,42 We further examined mean psychotherapy and antidepressant use by distress category.
We conducted robustness checks with a 1-year lag to account for potential delayed implementation of the recommendation. Given impacts of the COVID-19 pandemic on mental health service use and delivery, we conducted sensitivity analysis modeling a second interruption in 2020 to check baseline results within the context of this unexpected external disruption.43,44
To provide additional insight into our treatment results, we examined psychotherapy practitioners by type using the variables SEEDOC, DRSPLTY, and MEDPTYPE in the MEPS Event files. We assessed the proportion of psychotherapy practitioners seen over time by type and prescribing authority and the likelihood of seeing a prescriber for psychotherapy after the USPSTF recommendation (eAppendix C in Supplement 1). In addition to COVID-19 sensitivity analysis, we reran the prescriber ITS model with a second interruption in 2014 to model implementation of multiple federal parity laws that may have increased demand for mental health services and affected provider (clinician or facility) availability. All statistical analyses were conducted in Stata, version 19 (StataCorp LLC), or RStudio, version 5.1 (R Foundation), from November 3, 2025, to March 31, 2026. We considered 2-sided P < .05 to be statistically significant.
Results
The analysis included 372 962 individuals from the 2008-2023 MEPS, which corresponded to a nationally representative sample of 244 316 623 adults aged 18 years and older. The median age of the sample was 47 (IQR, 32-61) years. The largest proportion of the sample was aged 18 to 34 years (29.7% vs 25.0% aged 35-49 years, 25.3% aged 50-64 years, and 20.0% aged ≥65 years), female (51.6% vs 48.4% male), reported mild or moderate psychological distress (49.6% vs 3.6% serious distress and 28.1% none), had any private insurance (68.2% vs 21.1% public only and 10.7% uninsured), had attained a high school diploma or General Educational Development (GED) (48.2% vs 12.9% 0-12 years and no high school diploma, 19.5% bachelor’s degree, and 18.6% graduate degree), had a family income of 400% or greater of the federal poverty level (FPL) (42.9% vs 15.3% with income <125% FPL, 12.6% with income 125% to <200% FPL, and 29.2% with income 200% to <400% FPL) and resided in the South (37.6% vs 21.1% Midwest, 17.8% Northeast, and 23.5% West). For self-reported race and ethnicity, 15.8% identified as Hispanic, 11.8% as non-Hispanic Black, 64.0% as non-Hispanic White, and 8.4% as non-Hispanic other or multiple races (Table). Information was not ascertained on psychological distress for 18.7% of the sample and educational attainment for 0.8% of the sample. Stratified by treatment category, there were statistically significant differences across sociodemographic characteristics.
Table. Sociodemographic Characteristics of US Adults Aged ≥18 Years (2008-2023) Overall and by Psychotherapy and Antidepressant Treatment Category.
| Characteristic | Participants, No. (%) | P value | |||
|---|---|---|---|---|---|
| Adults ≥18 y | Treatment category | ||||
| Psychotherapy only | Antidepressants only | Psychotherapy and antidepressants | |||
| Unweighted No. | 372 962 | 5994 | 32 445 | 5503 | NA |
| Weighted No. | 244 316 623 | 4 798 143 | 23 852 365 | 4 069 191 | NA |
| Age, median (IQR), y | 47 (32-61) | 38 (28-52) | 55 (42-67) | 43 (31-56) | <.001 |
| Age group | |||||
| 18-34 y | 72 568 375 (29.7) | 1 947 652 (40.6) | 3 673 226 (15.4) | 1 303 953 (32.0) | <.001 |
| 35-49 y | 61 044 443 (25.0) | 1 450 179 (30.2) | 5 224 871 (21.9) | 1 253 482 (30.8) | |
| 50-64 y | 61 826 661 (25.3) | 973 719 (20.3) | 7 965 056 (33.4) | 1 094 134 (26.9) | |
| ≥65 y | 48 877 144 (20.0) | 426 592 (8.9) | 6 989 212 (29.3) | 417 623 (10.3) | |
| Sex | |||||
| Female | 126 111 094 (51.6) | 2 762 183 (57.6) | 16 396 078 (68.7) | 2 642 479 (64.9) | <.001 |
| Male | 118 205 529 (48.4) | 2 035 960 (42.4) | 7 456 287 (31.3) | 1 426 712 (35.1) | |
| Psychological distress | |||||
| Serious | 8 782 699 (3.6) | 494 893 (10.3) | 2 520 549 (10.6) | 801 213 (19.7) | <.001 |
| Mild or moderate | 121 291 398 (49.6) | 2 795 505 (58.3) | 14 861 967 (62.3) | 2 363 465 (58.1) | |
| None | 68 571 740 (28.1) | 427 382 (8.9) | 3 059 459 (12.8) | 200 130 (4.9) | |
| Inapplicable/cannot be computed | 45 670 786 (18.7) | 1 080 363 (22.5) | 3 410 390 (14.3) | 704 383 (17.3) | |
| Race and ethnicity | |||||
| Hispanic | 38 637 686 (15.8) | 559 560 (11.7) | 1 762 053 (7.4) | 373 363 (9.2) | <.001 |
| Non-Hispanic Black | 28 795 491 (11.8) | 496 193 (10.3) | 1 364 189 (5.7) | 299 694 (7.4) | |
| Non-Hispanic other/multiple races | 20 505 666 (8.4) | 319 013 (6.6) | 958 527 (4.0) | 226 615 (5.6) | |
| Non-Hispanic White | 156 377 781 (64.0) | 3 423 376 (71.3) | 19 767 596 (82.9) | 3 169 519 (77.9) | |
| Health insurance | |||||
| Any private | 166 668 029 (68.2) | 3 433 317 (71.6) | 15 688 147 (65.8) | 2 648 952 (65.1) | <.001 |
| Public only | 51 462 091 (21.1) | 1 159 717 (24.2) | 7 158 622 (30.0) | 1 226 853 (30.1) | |
| Uninsured | 26 186 503 (10.7) | 205 108 (4.3) | 1 005 596 (4.2) | 193 386 (4.8) | |
| Educational attainment | |||||
| 0-12 y (no diploma) | 31 597 043 (12.9) | 460 090 (9.6) | 2 532 050 (10.6) | 405 481 (10.0) | <.001 |
| High school diploma/GED | 117 682 334 (48.2) | 1 869 952 (39.0) | 12 258 989 (51.4) | 1 789 460 (44.0) | |
| Bachelor’s degree | 47 682 516 (19.5) | 1 252 800 (26.1) | 4 211 982 (17.7) | 923 298 (22.7) | |
| Graduate degree | 45 491 794 (18.6) | 1 187 847 (24.8) | 4 727 178 (19.8) | 935 990 (23.0) | |
| Not ascertained or inapplicable | 1 862 936 (0.8) | 27 454 (0.6) | 122 165 (0.5) | 14 961 (0.4) | |
| Family income | |||||
| <125% FPL | 37 338 654 (15.3) | 830 538 (17.3) | 4 231 805 (17.7) | 914 597 (22.5) | <.001 |
| 125% to <200% FPL | 30 803 324 (12.6) | 490 443 (10.2) | 3 195 816 (13.4) | 512 559 (12.6) | |
| 200% to <400% FPL | 71 318 626 (29.2) | 1 208 335 (25.2) | 6 922 198 (29.0) | 1 045 078 (25.7) | |
| ≥400% FPL | 104 856 019 (42.9) | 2 268 827 (47.3) | 9 502 545 (39.8) | 1 596 957 (39.2) | |
| US region | |||||
| Midwest | 51 486 174 (21.1) | 1 096 830 (22.9) | 6 208 175 (26.0) | 965 592 (23.7) | <.001 |
| Northeast | 43 581 417 (17.8) | 1 115 480 (23.2) | 3 889 109 (16.3) | 941 996 (23.1) | |
| South | 91 781 775 (37.6) | 1 345 944 (28.1) | 8 940 345 (37.5) | 1 294 579 (31.8) | |
| West | 57 467 258 (23.5) | 1 239 888 (25.8) | 4 814 736 (20.2) | 867 024 (21.3) | |
Abbreviations: FPL, federal poverty level; GED, General Educational Development; NA, not applicable.
Psychotherapy Only
Baseline results indicated significant level (odds ratio [OR], 1.18; 95% CI, 1.02-1.37) and post-USPSTF trend (OR, 1.08; 95% CI, 1.04-1.11) increases in the likelihood of psychotherapy only after the USPSTF recommendation (Figure 1). Robustness checks with a 1-year lag also found statistically significant level and post-USPSTF trend increases (eFigure 1 in Supplement 1). In the COVID-19 sensitivity analysis, only the post-USPSTF trend increase was statistically significant. Results from stratified analysis that were statistically significant and robust to COVID-19 sensitivity analysis indicated significant post-USPSTF trend increases in the probability of psychotherapy only among the following sociodemographic categories: age 35 to 49 years (OR, 1.08; 95% CI, 1.01-1.15) and 50-64 years (OR, 1.11; 95% CI, 1.03-1.20), female (OR, 1.07; 95% CI, 1.03-1.12), male (OR, 1.09; 95% CI, 1.03-1.14), non-Hispanic White (OR, 1.08; 95% CI, 1.04-1.12), any private insurance (OR, 1.11; 95% CI, 1.06-1.15), high school diploma or GED (OR, 1.07; 95% CI, 1.02-1.13), family income 400% of the FPL or greater (OR, 1.13; 95% CI, 1.07-1.19), and resided in the Midwest (OR, 1.14; 95% CI, 1.07-1.20) and in the South (OR, 1.09; 95% CI, 1.01-1.18) (eTables 1 and 2 in Supplement 1).
Figure 1. Dot Plot Showing the Probability of Psychotherapy and Antidepressant Use After the USPSTF Recommendation Among US Adults Aged ≥18 Years.
Level and post-trend odds ratio results with 95% CIs for the probability of psychotherapy only, antidepressants only, and combination treatment after the USPSTF recommendation from the baseline interrupted time series models. USPSTF indicates US Preventive Services Task Force.
Antidepressants Only
There were significant level (OR, 0.88; 95% CI, 0.82-0.94) and post-USPSTF trend (OR, 0.98; 95% CI, 0.96-0.99) decreases in the likelihood of antidepressants only after the USPSTF recommendation (Figure 1). Level-change decreases remained significant in the 1-year lag model and COVID-19 sensitivity analysis (eFigure 1 in Supplement 1). Stratified results that were statistically significant and robust to COVID-19 sensitivity analysis indicated level decreases in the probability of antidepressants only among the following sociodemographic categories: aged 50 to 64 years (OR, 0.76; 95% CI, 0.68-0.85), male (OR, 0.79; 95% CI, 0.70-0.89), non-Hispanic White (OR, 0.86; 95% CI, 0.79-0.93), any private insurance (OR, 0.85; 95% CI, 0.78-0.92), high school diploma or GED (OR, 0.80; 95% CI, 0.73-0.88), and resided in the South (OR, 0.90; 95% CI, 0.81-0.99) (eTables 3 and 4 in Supplement 1). Robust results also indicated significant post-USPSTF trend decreases in the probability of antidepressants only among the following categories: age 65 years and older (OR, 0.93; 95% CI, 0.90-0.96), female (OR, 0.97; 95% CI, 0.95-0.99), uninsured (OR, 0.89; 95% CI, 0.82-0.97), and family income less than 125% of the FPL (OR, 0.97; 95% CI, 0.93-1.0). In contrast to the overall pattern, baseline results indicated a significant level increase in the likelihood of antidepressants only among adults who identified as non-Hispanic other or multiple races (OR, 1.35; 95% CI, 1.02-1.79) and adults with 0 to 12 years of education (no diploma) (OR, 1.25; 95% CI, 1.05-1.48), but only the latter were robust to COVID-19 sensitivity analysis.
Combined Psychotherapy and Antidepressants
There were no statistically significant changes in the likelihood of combined psychotherapy and antidepressant use among adults overall in the baseline model (Figure 1). Results from sensitivity analysis indicated a significant level increase in the 1-year lag model (OR, 1.20; 95% CI, 1.02-1.41) and post-USPSTF trend increase in the COVID-19 model (OR, 1.10; 95% CI, 1.03-1.19) (eFigure 1 in Supplement 1). Stratified results that were statistically significant and robust to COVID-19 sensitivity analysis indicated post-USPSTF trend increases in the probability of combined treatment among adults with any private insurance (OR, 1.07; 95% CI, 1.02-1.12) and family income of 400% or more of the FPL (OR, 1.07; 95% CI, 1.00-1.14) (eTables 5 and 6 in Supplement 1).
Treatment by Psychological Distress Level
When examined by psychological distress level, results indicated significant changes associated with the USPSTF recommendation among adults who self-reported serious distress and mild or moderate distress (Figure 2). Among those reporting serious psychological distress, baseline results indicated a significant level increase in the likelihood of psychotherapy only (OR, 1.63; 95% CI, 1.10-2.41) and a significant post-USPSTF trend decrease in the likelihood of antidepressants only (OR, 0.94; 95% CI, 0.89-0.98) after the recommendation (eTable 7 in Supplement 1). These results, however, were not robust to COVID-19 sensitivity analysis (eTable 8 in Supplement 1). Baseline results among adults with mild or moderate psychological distress indicated significant level (OR, 1.38; 95% CI, 1.15-1.66) and post-USPSTF trend (OR, 1.05; 95% CI, 1.01-1.10) increases in the probability of psychotherapy only. For the likelihood of antidepressants only, results indicated significant level (OR, 0.88; 95% CI, 0.81-0.96) and post-USPSTF trend (OR, 0.96; 95% CI, 0.94-0.98) decreases among this group. Among adults with mild or moderate distress, the post-USPSTF trend results remained robust to COVID-19 sensitivity analysis for both treatment categories. The baseline model results indicated no significant changes associated with the USPSTF recommendation in any treatment category among adults reporting no psychological distress. Additionally, there were no significant changes in combination treatment across psychological distress categories.
Figure 2. Dot Plot Showing the Probability of Psychotherapy and Antidepressant Use After the USPSTF Recommendation Among US Adults Aged ≥18 Years by Psychological Distress Level.
Level and post-trend odds ratio results with 95% CIs for the probability of psychotherapy only, antidepressants only, and combination treatment after the USPSTF recommendation by category of psychological distress level. USPSTF indicates US Preventive Services Task Force.
Psychotherapy by Practitioner Prescribing Authority
Results show a significant shift in psychotherapy from visits with prescribers (70.6% in 2008) to nonprescribers (59.4% in 2023) (Figure 3). Baseline results estimating the probability of seeing a prescriber for psychotherapy indicate a significant level decrease associated with the USPSTF recommendation (OR, 0.79; 95% CI, 0.70-0.89) that was robust to sensitivity analysis modeling the 2014 parity laws and COVID-19 pandemic onset (Figure 4). Within the nonprescriber category, the proportion of psychotherapy visits with psychologists decreased over time and increased with social workers and mental health counselors (eTable 9 in Supplement 1). Among prescribers, the proportion of psychotherapy visits with psychiatrists and primary care physicians decreased by approximately half from 2008 to 2023.
Figure 3. Line Graph Showing Psychotherapy/Mental Health Counseling Providers by Prescribing Authority.
The proportion of US adults who saw a provider (clinician or facility) with prescribing authority vs a nonprescribing provider for psychotherapy/mental health counseling from 2008 to 2023, with vertical dashed lines at 2014 to indicate implementation of multiple federal mental health parity policies; at 2016 to indicate release of the USPSTF routine depression screening recommendation; and at 2020 to indicate the onset of the COVID-19 pandemic. Prescribers included primary care physician, psychiatrist, other physician (MD), nurse practitioner, and physician assistant. Nonprescribers included psychologists, social worker, mental health counselor/marital or family therapist (category added in 2018), and other provider (non-MD). Multiple federal parity policies: The Patient Protection and Affordable Care Act coverage of mental health as an essential health benefit, the expansion of parity under the Mental Health Parity and Addiction Equity Act to individual and small group markets, the expansion of Medicaid coverage for low-income adults, and full mental health parity for Medicare beneficiaries that all went into effect in 2014. USPSTF indicates US Preventive Services Task Force.
Figure 4. Dot Plot Showing the Probability of Psychotherapy/Mental Health Counseling Visit With a Prescriber Among US Adults Aged ≥18 Years With Any Therapy.
Level and post-trend odds ratio results with 95% CIs for the probability of seeing a prescriber for psychotherapy after the USPSTF recommendation from the baseline interrupted time series model and sensitivity analysis with added interruptions for 2014 parity laws and the 2020 onset of the COVID-19 pandemic. USPSTF indicates US Preventive Services Task Force.
Discussion
In this policy evaluation, we assessed whether the USPSTF recommendation for routine adult depression screening was associated with changes in the likelihood of psychotherapy or antidepressant use among US adults. Results indicated a positive association between the recommendation and likelihood of psychotherapy only. In contrast, we found a negative association between the recommendation and likelihood of antidepressants only. Additionally, changes in the overall probability of combination treatment among adults were nonsignificant. These findings contrast with our a priori hypothesis that the USPSTF recommendation would be associated primarily with an increase in antidepressant use, especially among adults with more serious conditions for whom antidepressants are typically regarded as a first-line treatment.
Our hypothesis was premised on the fact that the depression screening guideline was developed for, and was likely to most directly affect, care in primary care settings. PCPs prescribe the majority of psychotropic medications despite recent declines among PCPs and increases in prescribing among psychiatrists and psychiatric nurse practitioners.33,45 Thus, the finding that the proportion of psychotherapy visits with primary care physicians decreased by half over time could partially explain the decrease in probability of antidepressant use. The significant decrease in the likelihood of psychotherapy with any prescriber after the USPSTF recommendation could provide further explanation. The concomitant increase in the likelihood of psychotherapy suggests, however, that findings could also reflect patient preferences for psychotherapy vs pharmacotherapy.46 Additionally, psychotherapy is considered a good initial treatment approach for people with less serious distress, and the majority of the study cohort reported mild or moderate psychological distress. Thus, our study results could also signify greater coordination between PCPs and mental health practitioners for appropriate care.
Notably, our results indicate that changes in psychotherapy and antidepressant use occurred foremost among adults who indicated some level of psychological distress. Findings among adults who reported mild or moderate distress were both significant and robust to sensitivity analysis. This suggests that the USPSTF recommendation was associated with treatment among adults who were most likely to benefit from depression screening since they may not present as readily or obviously needing treatment as adults with serious distress. The recommendation also was associated with significant changes among adults who reported serious psychological distress, although sensitivity analysis results suggest that the COVID-19 pandemic had a moderating influence. Despite declines after COVID-19, however, this group had a much higher mean use of psychotherapy and antidepressants annually than adults with mild or moderate distress (eFigure 2 and eTable 10 in Supplement 1). Findings that the USPSTF recommendation was not associated with any significant treatment changes among adults who reported no psychological distress suggest that implementation of routine depression screening did not lead to false-positive results and unnecessary treatment.
Although most stratified results aligned with overall results, there were a few notable exceptions. In contrast to findings of decreased likelihood of antidepressants only among adults overall, results that were robust to sensitivity analysis suggested a significant level increase in likelihood among adults who had attained 0 to 12 years of education (no high school diploma). This positive association, however, was not sustained over time in the post-USPSTF trend. Results for combination treatment among adults overall were not significant. Yet, stratified findings that were robust to sensitivity analysis suggested that the USPSTF recommendation was associated with significant increases in the likelihood of combined psychotherapy and antidepressant use post-USPSTF among adults with any private insurance and those with a family income of 400% or greater of the FPL. Moreover, stratified results suggested that the USPSTF recommendation was not associated with significant changes in the likelihood of treatment among many sociodemographic categories, including adults with public-only insurance. These findings may be used to generate hypotheses for future research examining potential heterogeneity in policy effects.
Limitations
This study has limitations. Although our study used robust methods to assess changes in the likelihood of depression treatment after the USPSTF recommendation, our analysis is subject to several limitations. First, the single-group ITS design generates a synthetic counterfactual from preintervention level and trend estimates in the absence of a valid comparison group. Results, therefore, should be interpreted as associative vs causal. Second, our outcomes may have included some nondepression treatment as psychotherapy and antidepressants can be used for other conditions such as anxiety and smoking cessation. Since this measurement limitation would apply to both the pre- and postintervention periods, however, we would not expect it to bias ITS results. Third, we assumed consistent prescribing authority across provider types. Our results therefore do not reflect potential state variation in prescribing authority among some provider types. Last, MEPS added the mental health counselor category only in 2018, and the COVID-19 pandemic affected MEPS data collection, which could have reduced the statistical precision of some estimates.
Conclusions
This study found that the USPSTF recommendation for routine adult depression screening was associated with an overall increase in the likelihood of psychotherapy only and a decrease in the likelihood of antidepressant use only among US adults who reported some level of psychological distress. Stratified analysis suggests heterogeneity in the likelihood of treatment among different sociodemographic subgroups, with some experiencing no significant changes following the recommendation. Findings that the recommendation was not associated with significant treatment changes among adults who reported no psychological distress suggest that concerns about unnecessary treatment following routine depression screening were not realized.
eAppendix A. Drug names used to identify antidepressants in analysis
eAppendix B. Interrupted time series design
eAppendix C. Providers by prescribing authority category
eFigure 1. One-year lag and COVID sensitivity analysis model results for probability of psychotherapy and antidepressant use after the USPSTF recommendation among U.S. adults aged ≥18 years
eFigure 2. Mean use among U.S. adults ages ≥18 years by Level of Psychological Distress, 2008-2023
eTable 1. Probability of psychotherapy/ mental health counseling use only after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 2. COVID-19 sensitivity analysis for probability of psychotherapy/ mental health counseling use only after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 3. Probability of antidepressant use only after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 4. COVID-19 sensitivity analysis for probability of antidepressants use only after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 5. Probability of combined psychotherapy/ mental health counseling and antidepressant use after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 6. COVID-19 sensitivity analysis for probability of combined psychotherapy and antidepressant use after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 7. Probability of use by psychological distress level
eTable 8. COVID-19 sensitivity analysis for probability of use by psychological distress level
eTable 9. Psychotherapy/ Mental health counseling providers among U.S. adults aged ≥ 18 years with any therapy visits, 2008-2023
eTable 10. Mean use among U.S. adults ages ≥18 years by Level of Psychological Distress, 2008-2023
Data Sharing Statement
References
- 1.Daly M, Sutin AR, Robinson E. Depression reported by US adults in 2017-2018 and March and April 2020. J Affect Disord. 2021;278:131-135. doi: 10.1016/j.jad.2020.09.065 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ettman CK, Abdalla SM, Cohen GH, Sampson L, Vivier PM, Galea S. Prevalence of depression symptoms in US adults before and during the COVID-19 pandemic. JAMA Netw Open. 2020;3(9):e2019686. doi: 10.1001/jamanetworkopen.2020.19686 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Greenberg P, Chitnis A, Louie D, et al. The economic burden of adults with major depressive disorder in the United States (2019). Adv Ther. 2023;40(10):4460-4479. doi: 10.1007/s12325-023-02622-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Greenberg PE, Fournier AA, Sisitsky T, et al. The economic burden of adults with major depressive disorder in the United States (2010 and 2018). Pharmacoeconomics. 2021;39(6):653-665. doi: 10.1007/s40273-021-01019-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Katon WJ. Clinical and health services relationships between major depression, depressive symptoms, and general medical illness. Biol Psychiatry. 2003;54(3):216-226. doi: 10.1016/S0006-3223(03)00273-7 [DOI] [PubMed] [Google Scholar]
- 6.Olfson M, Blanco C, Marcus SC. Treatment of adult depression in the United States. JAMA Intern Med. 2016;176(10):1482-1491. doi: 10.1001/jamainternmed.2016.5057 [DOI] [PubMed] [Google Scholar]
- 7.Kamal R, Cox C, Rousseau D; Kaiser Family Foundation . Costs and outcomes of mental health and substance use disorders in the US. JAMA. 2017;318(5):415-415. doi: 10.1001/jama.2017.8558 [DOI] [PubMed] [Google Scholar]
- 8.Corrigan PW, Druss BG, Perlick DA. The impact of mental illness stigma on seeking and participating in mental health care. Psychol Sci Public Interest. 2014;15(2):37-70. doi: 10.1177/1529100614531398 [DOI] [PubMed] [Google Scholar]
- 9.Dobalian A, Rivers PA. Racial and ethnic disparities in the use of mental health services. J Behav Health Serv Res. 2008;35(2):128-141. doi: 10.1007/s11414-007-9097-8 [DOI] [PubMed] [Google Scholar]
- 10.Paul Wellstone and Pete Domenici Mental Health Parity and Addiction Equity Act of 2008. HR 6983. 110th Cong (2008). Accessed August 17, 2025. https://www.congress.gov/bill/110th-congress/house-bill/6983
- 11.Medicare Improvements for Patients and Providers Act of . 2008. HR 6331. 110th Cong (2008). Accessed August 18, 2025. https://www.congress.gov/bill/110th-congress/house-bill/6331
- 12.Patient Protection and Affordable Care Act. HR 3590. 111th Cong (2010). Accessed August 18, 2025. https://www.congress.gov/bill/111th-congress/house-bill/3590
- 13.Siu AL, Bibbins-Domingo K, Grossman DC, et al. ; US Preventive Services Task Force (USPSTF) . Screening for depression in adults: US Preventive Services Task Force recommendation statement. JAMA. 2016;315(4):380-387. doi: 10.1001/jama.2015.18392 [DOI] [PubMed] [Google Scholar]
- 14.US Preventive Services Task Force . Screening for depression in adults: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2009;151(11):784-792. doi: 10.7326/0003-4819-151-11-200912010-00006 [DOI] [PubMed] [Google Scholar]
- 15.Gates A, Ranji U, Snyder L. Coverage of preventive services for adults in Medicaid. 2014. Accessed May 19, 2026. https://www.kff.org/affordable-care-act/coverage-of-preventive-services-for-adults-in-medicaid/
- 16.Centers for Medicare & Medicaid Services . Screening for depression in adults. Accessed March 25, 2026. https://www.cms.gov/medicare-coverage-database/view/ncacal-decision-memo.aspx?proposed=N&NCAId=251
- 17.Bhattacharjee S, Goldstone L, Vadiei N, Lee JK, Burke WJ. Depression screening patterns, predictors, and trends among adults without a depression diagnosis in ambulatory settings in the United States. Psychiatr Serv. 2018;69(10):1098-1100. doi: 10.1176/appi.ps.201700439 [DOI] [PubMed] [Google Scholar]
- 18.Jackson JL, Kuriyama A, Bernstein J, Demchuk C. Depression in primary care, 2010-2018. Am J Med. 2022;135(12):1505-1508. doi: 10.1016/j.amjmed.2022.06.022 [DOI] [PubMed] [Google Scholar]
- 19.Samples H, Stuart EA, Saloner B, Barry CL, Mojtabai R. The role of screening in depression diagnosis and treatment in a representative sample of US primary care visits. J Gen Intern Med. 2020;35(1):12-20. doi: 10.1007/s11606-019-05192-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Garcia ME, Hinton L, Neuhaus J, Feldman M, Livaudais-Toman J, Karliner LS. Equitability of depression screening after implementation of general adult screening in primary care. JAMA Netw Open. 2022;5(8):e2227658. doi: 10.1001/jamanetworkopen.2022.27658 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pfoh ER, Janmey I, Anand A, Martinez KA, Katzan I, Rothberg MB. The impact of systematic depression screening in primary care on depression identification and treatment in a large health care system: a cohort study. J Gen Intern Med. 2020;35(11):3141-3147. doi: 10.1007/s11606-020-05856-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Mojtabai R. Universal Depression Screening to Improve Depression Outcomes in Primary Care: Sounds Good, But Where Is the Evidence? American Psychiatric Association; 2017. [DOI] [PubMed] [Google Scholar]
- 23.Skerrett P. Depression screening for adults and adolescents has benefits, but don’t ignore the downsides. STAT News. Accessed September 11, 2025. https://www.statnews.com/2016/02/08/depression-screening-benefits-downsides/
- 24.McLintock K, Foy R, House A, Alderson SL. A policy of universal screening for depression: caution needed. BMJ. 2016;353:i2174. doi: 10.1136/bmj.i2174 [DOI] [PubMed] [Google Scholar]
- 25.Drake C, Busch SH, Golberstein E. The effects of federal parity on mental health services use and spending: evidence from the Medical Expenditure Panel Survey. Psychiatr Serv. 2019;70(4):287-293. doi: 10.1176/appi.ps.201800313 [DOI] [PubMed] [Google Scholar]
- 26.Fung V, Price M, Nierenberg AA, Hsu J, Newhouse JP, Cook BL. Assessment of behavioral health services use among low-income Medicare beneficiaries after reductions in coinsurance fees. JAMA Netw Open. 2020;3(10):e2019854-e2019854. doi: 10.1001/jamanetworkopen.2020.19854 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cook BL, Flores M, Zuvekas SH, et al. The impact of Medicare’s mental health cost-sharing parity on use of mental health care services. Health Aff (Millwood). 2020;39(5):819-827. doi: 10.1377/hlthaff.2019.01008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Fung V, Price M, McDowell A, et al. Coverage parity and racial and ethnic disparities in mental health and substance use care among Medicare beneficiaries. Health Aff. 2023;42(1):83-93. doi: 10.1377/hlthaff.2022.00624 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Burns ME, Dague L, Saloner B, et al. Implementing parity for mental health and substance use treatment in Medicaid. Health Serv Res. 2020;55(4):604-614. doi: 10.1111/1475-6773.13309 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Lawrence MB. Parity is not enough! Mental health, managed care, and Medicaid. J Law Med Ethics. 2020;48(3):480-484. doi: 10.1177/1073110520958872 [DOI] [PubMed] [Google Scholar]
- 31.Tetlow SM, Phillips VL, Hockenberry JM. Medicare parity and outpatient mental health service use and costs among beneficiaries with depression. JAMA Netw Open. 2025;8(5):e258491-e258491. doi: 10.1001/jamanetworkopen.2025.8491 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Craighead WE, Dunlop BW. Combination psychotherapy and antidepressant medication treatment for depression: for whom, when, and how. Annu Rev Psychol. 2014;65:267-300. doi: 10.1146/annurev.psych.121208.131653 [DOI] [PubMed] [Google Scholar]
- 33.Hughes PM, Annis IE, McGrath RE, Thomas KC. Psychotropic medication prescribing across medical providers, 2016–2019. Psychiatr Serv. 2024;75(5):477-480. doi: 10.1176/appi.ps.20230156 [DOI] [PubMed] [Google Scholar]
- 34.Butryn T, Bryant L, Marchionni C, Sholevar F. The shortage of psychiatrists and other mental health providers: causes, current state, and potential solutions. Int J Acad Med. 2017;3(1):5-9. doi: 10.4103/IJAM.IJAM_49_17 [DOI] [Google Scholar]
- 35.Phillips L, Tucker S. A closer look at the mental health provider shortage. Couns Today. 2023;65(11):25-39. [Google Scholar]
- 36.Agency for Healthcare Research and Quality . Information on the health status of Americans, health insurance coverage, and access, use, and cost of health services. Accessed August 13, 2025. https://datatools.ahrq.gov/meps-hc/
- 37.Agency for Healthcare Research and Quality . MEPS HC-036: 1996-2023 Pooled Variance Estimation File (August 2025). Accessed October 23, 2025. https://meps.ahrq.gov/data_stats/download_data/pufs/h036/h36u23doc.shtml
- 38.Agency for Healthcare Research and Quality . Accounting for clustering in the analysis of MEPS data: frequently asked questions (FAQ). Accessed July 20, 2026. https://meps.ahrq.gov/survey_comp/hc_clustering_faq.pdf
- 39.Umucu E, Fortuna K, Jung H, et al. A national study to assess validity and psychometrics of the Short Kessler Psychological Distress Scale (K6). Rehabil Couns Bull. 2022;65(2):140-149. doi: 10.1177/00343552211043261 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Kessler RC, Barker PR, Colpe LJ, et al. Screening for serious mental illness in the general population. Arch Gen Psychiatry. 2003;60(2):184-189. doi: 10.1001/archpsyc.60.2.184 [DOI] [PubMed] [Google Scholar]
- 41.Olfson M, Wang S, Wall M, Marcus SC, Blanco C. Trends in serious psychological distress and outpatient mental health care of US adults. JAMA Psychiatry. 2019;76(2):152-161. doi: 10.1001/jamapsychiatry.2018.3550 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Olfson M, McClellan C, Zuvekas SH, Wall M, Blanco C. Psychotherapy trends in the United States. Am J Psychiatry. 2025;182(5):483-492. doi: 10.1176/appi.ajp.20240492 [DOI] [PubMed] [Google Scholar]
- 43.Pierce BS, Perrin PB, Tyler CM, McKee GB, Watson JD. The COVID-19 telepsychology revolution: a national study of pandemic-based changes in U.S. mental health care delivery. Am Psychol. 2021;76(1):14-25. doi: 10.1037/amp0000722 [DOI] [PubMed] [Google Scholar]
- 44.Zhu JM, Myers R, McConnell KJ, Levander X, Lin SC. Trends in outpatient mental health services use before and during the COVID-19 pandemic. Health Aff (Millwood). 2022;41(4):573-580. doi: 10.1377/hlthaff.2021.01297 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Gordon BS, Martin K. Primary care providers prescribe the majority of antidepressants and anxiolytics for people with employer sponsored insurance. 2026. March 31, 2026. Accessed May 21, 2026. https://healthcostinstitute.org/all-hcci-reports/primary-care-providers-prescribe-the-majority-of-antidepressants-and-anxiolytics-for-people-with-employer-sponsored-insurance/
- 46.McHugh RK, Whitton SW, Peckham AD, Welge JA, Otto MW. Meta-analysis patient preference for psychological vs pharmacologic. J Clin Psychiatry. 2013;74(6):595-602. doi: 10.4088/JCP.12r07757 [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
eAppendix A. Drug names used to identify antidepressants in analysis
eAppendix B. Interrupted time series design
eAppendix C. Providers by prescribing authority category
eFigure 1. One-year lag and COVID sensitivity analysis model results for probability of psychotherapy and antidepressant use after the USPSTF recommendation among U.S. adults aged ≥18 years
eFigure 2. Mean use among U.S. adults ages ≥18 years by Level of Psychological Distress, 2008-2023
eTable 1. Probability of psychotherapy/ mental health counseling use only after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 2. COVID-19 sensitivity analysis for probability of psychotherapy/ mental health counseling use only after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 3. Probability of antidepressant use only after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 4. COVID-19 sensitivity analysis for probability of antidepressants use only after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 5. Probability of combined psychotherapy/ mental health counseling and antidepressant use after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 6. COVID-19 sensitivity analysis for probability of combined psychotherapy and antidepressant use after the USPSTF recommendation among U.S. adults aged ≥18 years
eTable 7. Probability of use by psychological distress level
eTable 8. COVID-19 sensitivity analysis for probability of use by psychological distress level
eTable 9. Psychotherapy/ Mental health counseling providers among U.S. adults aged ≥ 18 years with any therapy visits, 2008-2023
eTable 10. Mean use among U.S. adults ages ≥18 years by Level of Psychological Distress, 2008-2023
Data Sharing Statement
