Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Apr 1.
Published in final edited form as: Am J Prev Med. 2021 Nov 17;62(4):511–518. doi: 10.1016/j.amepre.2021.09.008

Diagnoses and Treatment After Depression Screening in Primary Care Among Youth

Kira E Riehm 1, Emily Brignone 2, Elizabeth A Stuart 1, Joseph J Gallo 1,3, Ramin Mojtabai 1
PMCID: PMC8940608  NIHMSID: NIHMS1751721  PMID: 34801332

Abstract

Introduction:

Depression screening is universally recommended for adolescents presenting in primary care settings in the U.S. However, little is known about how depression screening affects the likelihood of being diagnosed with a mental disorder or accessing mental health care over time.

Methods:

This longitudinal cohort study used insurance claims data from adolescents who attended a well-visit between 2014 and 2017. Propensity score matching was used to compare adolescents who were screened for depression to similar unscreened adolescents. Diagnoses and treatment uptake were examined over 6-month follow-up and included depression diagnoses, mood-related diagnoses, antidepressant medications, any mental health medication, and psychotherapy. Heterogeneity of associations by sex was also examined. Analyses were conducted from December 2020 to June 2021.

Results:

The sample included 57,732 adolescents (mean age of 14.26 years, 48.9% female). Compared with adolescents who were not screened for depression, adolescents screened for depression were 30% more likely to be diagnosed with depression (risk ratio [RR]=1.30, 95% CI=1.11, 1.52) and 17% more likely to receive a mood-related diagnosis (RR=1.17, 95% CI=1.08, 1.27), but were not more likely to be treated with an antidepressant medication (RR=1.11, 95% CI=0.82, 1.51), any mental health medication (RR=1.15, 95% CI=0.87, 1.53), or psychotherapy (RR=1.13, 95% CI=0.98, 1.31). Associations were generally stronger among female adolescents.

Conclusions:

Adolescents who were screened for depression during a well-visit were more likely to receive a diagnosis of depression or a mood-related disorder in the 6 months following screening. Future research should explore methods for increasing treatment uptake following screening.

INTRODUCTION

Depression is a leading cause of morbidity and functional impairment among adolescents.1 Left untreated, adolescent depression can have adverse consequences for well-being in adulthood, including chronicity of depressive symptoms, onset of other mental health disorders, incidence of somatic conditions, and premature mortality.2,3 Together, the negative outcomes associated with depression and rising rates over time4,5 demand action to reduce the burden of this mental health problem among adolescents.

Prior studies have found that primary care providers may fail to identify up to two thirds of adolescents with depression.6 Screening has been proposed as a method of improving mental health via identification and treatment of adolescents not suspected of having depression and who have not already been diagnosed with depression.7,8 Universal depression screening is currently recommended in primary care settings for adolescents by the U.S. Preventive Services Task Force9 and American Academy of Pediatrics,10 but these recommendations are controversial, with some researchers pointing to a lack of evidence for improved outcomes following screening.7,11

At present, there are no RCTs that compare depression screening to usual care in primary care settings among adolescents.11 Instead, evidence for depression screening is often inferred from studies of collaborative care or education programs for providers.12–14 To better inform clinical guidelines, decision makers have called for research to examine the longitudinal outcomes of depression screening.9

In the absence of evidence from RCTs, the application of causal inference methods to observational data provides an important opportunity to estimate the effect of depression screening on outcomes among adolescents. Because physicians may be more likely to screen adolescents they suspect are depressed,15 to compare diagnoses and treatment uptake between screened and non-screened adolescents, propensity scores are used in this study to account for characteristics associated with being screened in the first place. The objectives of this study are: (1) to examine the prospective association of being screened for depression during routine well-visits with diagnoses and treatment uptake among adolescents and (2) to determine if associations vary by sex.

METHODS

Study Sample

Healthcare claims and pharmacy claims data were obtained from Highmark Health, a non-profit healthcare organization.16 The data include Blue Cross Blue Shield members with commercial (large group, small group, and participants who acquired individual or small group insurance through Affordable Care Act Marketplace) or Medicare Advantage insurance. The largest membership is in Pennsylvania, West Virginia, and Delaware, but members live across the U.S. owing to large national accounts.

This study used a retrospective cohort design as part of a larger project with the aim of examining diagnoses, health services use, and other processes following depression screening. Participants who met the following criteria were included: (1) had a well-visit (V20.2, V70.0, Z00.121, Z00.129, Z00.00, Z00.01, 99384, 99394, 99385, or 99395) in primary care or general pediatric settings after January 1, 2014 and before December 31, 2017 (hereafter referred to as the index well-visit), (2) were continuously enrolled for at least 6 months before and 2 years after the index well-visit, and (3) were aged 12–18 years (the age group to which the U.S. Preventive Services Task Force guidelines apply) at the time of the index well-visit. For participants with >1 well-visit during the study period, the first well-visit was selected to maximize available follow-up time. Adolescents who had a diagnosis of depression, an antidepressant prescription, or psychotherapy treatment in the 6 months prior to the index well-visit were excluded.17 Because the interest was in proximal clinical events occurring soon after depression screening, a follow-up period of 6 months after the index well-visit was selected, which mirrors prior studies.18

Of the 281,463 adolescents with a well-visit recorded between 2014 and 2017, a total of 57,732 were included in the propensity score–matched sample. A flow diagram is displayed in Appendix Figure 1.

Measures

Depression screening was defined according to a combination of ICD-9, ICD-10, Current Procedural Terminology, and Healthcare Common Procedure Coding System codes. Similar to a prior study,17 adolescents were coded as having been screened for depression during the index well-visit if ≥1 of the following codes was recorded: G0444, G8510, 96127, 99420, V79.0, or Z13.89. A description of each code, and the proportion of well-visits with each code recorded, is presented in Appendix Table 1.

Because general medical providers tend to use non-specific diagnostic categories to identify depressive disorders,19 2 diagnoses were used in this study: a narrower diagnosis of depression, and a broader definition of any mood-related diagnosis (including depression and other conditions). Depression diagnoses were identified by ICD-9 codes 296.20–296.25, 296.30–296.35, 300.4, 309.0, 309.1, 309.28, and 311 and ICD-10 codes F31.3–F31.6, F32.0–32.9, F33.0–33.3, F33.8, F33.9, F34.1, F34.8, F34.9, F38.0, F38.1, F38.8, F39, F41.2, and F99.20 Mood-related diagnoses were identified with ICD-9 codes 296, 300, 307‒309, 311, and 313 and ICD-10 codes F30–F48, F93, and F99.14

Treatment uptake was ascertained from Current Procedural Terminology codes and included medication treatment as well as psychotherapy. Antidepressant medications and any mental health medication (antidepressants, mood-stabilizers, anti-anxiety medications, anti-psychotic medications, and medications to treat attention–deficit hyperactivity disorder) were examined separately; a list of medications included is provided in Appendix Table 2. A list of the Current Procedural Terminology codes included to identify psychotherapy is in Appendix Table 3.

As previously noted, depression screening may be conducted selectively rather than randomly in primary care settings.15 Therefore, characteristics likely to be associated with both selection for depression screening and depression outcomes based on existing literature were selected as covariates.15,17 Age (years) and sex (male and female) were obtained from claims data. Race/ethnicity (Asian, Black, Hispanic/Latino, other, and White) was estimated using the wru R package, which applies Bayesian methods to generate predicted probabilities for each race/ethnicity category for a given person based on geolocation and other individual characteristics.21 Rurality was defined using Rural–Urban Commuting Area codes from 2010 (the most recent available), which are provided at the ZIP code level; these codes were aggregated into 3 categories (urban, large rural city/town, and small/isolated rural town). Prior emergency health services use was defined as the count of encounters with emergency or urgent care services in the 6 months prior to the index well-visit. Prior routine health services use was defined as the count of encounters with outpatient, primary care, and preventive services in the 6 months prior to the index well-visit. Physical health comorbidities were defined by the count of encounters in the 6 months prior to the index well-visit where an ICD code for a given category of the Charlson Comorbidity Index was recorded. Data for ZIP code–level median household income were obtained from the American Community Survey from 2006 to 2010.22 The medical specialties of the healthcare providers performing the well-visits were aggregated into 3 categories (child- or adolescent-specific, family-specific, and other primary care).

Statistical Analysis

Propensity score matching was used to account for confounding of the association between screening and diagnoses and treatment uptake by the aforementioned covariates. Propensity scores are a causal inference technique that can be applied to observational data to emulate some qualities of RCTs and estimate treatment effects when full-scale RCTs are not feasible.23 Nearest neighbor matching without replacement was used, selecting 3 unscreened adolescents for each screened adolescent. Matches were selected based on a propensity score estimated using a logistic regression model with depression screening as the dependent variable and the covariates as independent variables. To account for hypothesized moderation by sex, interaction terms between sex and each covariate were included in the propensity score model and exact matching was used for sex.24 Figure 2 displays standardized mean differences for all covariates before and after matching. Appendix Table 4 displays associations between each covariate and depression screening. After matching, standardized mean differences for all covariates were <0.1, indicating that the matching procedure improved balance between screened and unscreened adolescents on observed covariates. The matched sample (n=57,732) was composed of 14,433 screened and 43,299 unscreened adolescents.

Figure 2.

Figure 2.

Open in a new tab

Plot of absolute standardized mean differences for covariates, before (circles) and after (triangles) propensity score matching.

Descriptive statistics were calculated both the complete sample and the propensity score–matched sample. Changes in the proportion of adolescents screened over time were also examined.

To estimate the effect of depression screening on each outcome, a series of log-binomial regression models using the matched sample with depression screening as the independent variable was estimated. The same series of models including an interaction term between depression screening and sex was then estimated. Coefficients for the depression screening variable were exponentiated for interpretation as risk ratios (RRs) representing the average effect of screening among those who were screened (i.e., the average treatment effect among the treated). To estimate absolute measures of association, models for each outcome were re-estimated using logistic regression models with an identity link to convey risk differences. SEs in all models were estimated using cluster-robust variance estimates to account for matched clusters.

The study period coincided with the implementation of ICD-10 in the U.S. on October 1, 2015. To test the sensitivity of the results to the change in coding system, the models were re-estimated with an interaction term between depression screening and an indicator variable for whether an index well-visit occurred before or after October 1, 2015. The sensitivity of significant parameter estimates to unmeasured confounding was tested by calculating e-values.25

Statistical significance was assessed at p<0.05. All analyses were conducted using R (R studio, version 1.2.5042; R, version 4.0.0). This study was reviewed by the Johns Hopkins IRB and was determined to be human subjects research that meets criteria for exemption under Category 4 (i.e., secondary analysis of existing, de-identified data).

RESULTS

Descriptive characteristics are displayed in Table 1. The sample of 248,354 adolescents had a mean age of 14.26 years (SD=2.06) and 121,432 (48.9%) were female. In the 6-month period following the index well-visit, 4,730 (1.9%) adolescents received a depression diagnosis, 13,591 (5.5%) received a mood-related diagnosis, 1,330 (0.5%) were treated with antidepressants, 5,182 (2.1%) were treated with a medication for any mental health condition, and 6,394 (2.6%) were treated with psychotherapy. Descriptive statistics for the timing of diagnoses and treatment uptake over 6-month follow-up are displayed in Appendix Table 5.

Table 1.

Descriptive Characteristics for Complete and Matched Samples, N (%)

Variable Complete sample (n=248,354) n (%) Matched sample (n=57,732)
Not screened (n=43,299) n (%) Screened (n=14,433) n (%)
Covariates
 Agea 14.26 (2.06) 13.68 (1.98) 13.69 (1.96)
 Sex
  Female 121,432 (48.9) 21,375 (49.4) 7,125 (49.4)
  Male 126,922 (51.1) 21,924 (50.6) 7,308 (50.6)
 Race/Ethnicitya,b
  Asian 0.02 (0.11) 0.03 (0.12) 0.03 (0.12)
  Black 0.07 (0.17) 0.07 (0.18) 0.08 (0.18)
  Hispanic/Latino 0.07 (0.18) 0.05 (0.16) 0.05 (0.17)
  Other 0.02 (0.04) 0.01 (0.03) 0.01 (0.04)
  White 0.82 (0.28) 0.83 (0.26) 0.83 (0.27)
 Rurality
  Urban 199,569 (80.4) 37,914 (87.6) 12,670 (87.8)
  Large rural city/Town 30,039 (12.1) 3,883 (9.0) 1,237 (8.6)
  Small/Isolated rural town 18,746 (7.5) 1,502 (3.5) 526 (3.6)
 Year of index well-visit
  2014 117,743 (47.4) 6,968 (16.1) 2,314 (16.0)
  2015 54,054 (21.8) 10,362 (23.9) 3,373 (23.4)
  2016 40,024 (16.1) 11,467 (26.5) 3,775 (26.2)
  2017 36,533 (14.7) 14,502 (33.5) 4,971 (34.4)
 Emergency health services usea 0.18 (0.52) 0.21 (0.58) 0.21 (0.54)
 Routine health services usea 0.85 (1.84) 0.72 (1.50) 0.73 (1.61)
 Physical health comorbiditiea 0.12 (1.50) 0.09 (0.65) 0.10 (0.84)
 ZIP-level median incomea 55,886.52 (19,086.55) 58,219.29 (20,148.41) 58,107.51 (19,806.49)
 Provider specialty
  Child- or adolescent-specific 159,647 (64.3) 38,984 (90.0) 12,935 (89.6)
  Family-specific 67,957 (27.4) 2,640 (6.1) 909 (6.3)
  Other primary care 20,750 (8.4) 1,675 (3.9) 589 (4.1)
 Diagnoses and treatment uptake
  Depression diagnosisc 4,730 (1.9) 761 (1.8) 329 (2.3)
  Mood-related diagnosisc 13,591 (5.5) 2,426 (5.6) 948 (6.6)
  Antidepressant medicationc 1,330 (0.5) 224 (0.5) 83 (0.6)
  Any mental health medicationc 5,182 (2.1) 972 (2.2) 374 (2.6)
  Psychotherapyc 6,394 (2.6) 1,295 (3.0) 488 (3.4)
Open in a new tab
a

Reported as mean (SD).

b

Race/ethnicity is a predicted probability based on the wru package in R.

c

Cells report the count and proportion of adolescents with the outcome.

The percentage of adolescents screened for depression during the index well-visit increased across the study period (Figure 1). In 2014, approximately 2.0% of adolescents were screened for depression during the index well-visit; by contrast, 13.6% of adolescents who had their index well-visit in 2017 were screened.

Figure 1.

Figure 1.

Open in a new tab

Percentage of adolescents screened for depression at the index well-visit from 2014‒2017.

Results of the propensity score–matched analyses are displayed in Table 2. Compared with adolescents not screened for depression, those who were screened during the index well-visit were 30% more likely to be diagnosed with depression (RR=1.30, 95% CI=1.11, 1.52) and 17% more likely to receive a mood-related diagnosis (RR=1.17, 95% CI=1.08, 1.27) across 6-month follow-up. These relative differences corresponded to very small, but statistically significant, absolute differences (depression diagnosis: risk difference=0.52%, 95% CI=0.17%, 0.87%; mood-related diagnosis: risk difference=0.97%, 95% CI=0.46%, 1.47%) (Appendix Table 6). Being screened for depression was not associated with prescriptions for antidepressant medication or any mental health medication, nor with psychotherapy.

Table 2.

Propensity-Score Adjusted Association of Depression Screening With Diagnoses and Treatment Uptake, in the Matched Sample (n=57,732) and Stratified by Sex

Outcome Main association of depression screening With interaction term between depression screening and sex
RR (95% CI) Females
RR (95% CI)		 Males
RR (95% CI)		 p-value for difference
Depression diagnosis 1.30 (1.11, 1.52) 1.42 (1.17, 1.71) 1.12 (0.87, 1.44) 0.172
Mood-related diagnosis 1.17 (1.08, 1.27) 1.26 (1.12, 1.40) 1.06 (0.93, 1.22) 0.070
Antidepressant medication 1.11 (0.82, 1.51) 1.28 (0.89, 1.84) 0.85 (0.51, 1.44) 0.286
Any mental health medication 1.15 (0.87, 1.53) 1.23 (0.99, 1.53) 1.10 (0.92, 1.32) 0.633
Psychotherapy 1.13 (0.98, 1.31) 1.22 (1.05, 1.43) 1.02 (0.84, 1.23) 0.175
Open in a new tab

Notes: Boldface indicates statistical significance (p<0.05).

In stratified analyses (Table 2), female adolescents who were screened for depression were significantly more likely to be diagnosed with depression (RR=1.42, 95% CI=1.17, 1.71), to receive a mood-related diagnosis (RR=1.26, 95% CI=1.12, 1.40), and to be treated with psychotherapy (RR=1.22, 95% CI=1.05, 1.43), compared with female adolescents not screened for depression. These same associations were not significant among male adolescents. However, interactions terms between depression screening and sex were not statistically significant for any outcome.

Interaction terms between depression screening and an indicator variable for whether an index well-visit occurred before or after October 1, 2015 were not significant in models for any outcome. E-values for significant parameter estimates are presented in Appendix Table 7. The observed RRs of 1.30 and 1.17 for depression diagnoses and mood-related diagnoses could be explained away by an unmeasured confounder that was associated with both depression screening and diagnoses by RRs of 1.92 and 1.62, respectively, above and beyond the measured confounders, but weaker confounding could not do so. Given other aspects of the study design, namely the exclusion of those with recent depression diagnoses or depression treatment, these e-values indicate that unmeasured confounding is possible but unlikely.

DISCUSSION

In this study, adolescents who were screened for depression during a well-visit were more likely to subsequently be diagnosed with depression or a mood-related condition over 6-month follow-up. However, these increases were very modest, corresponding to absolute differences of only 0.5% for depression diagnoses and 1.0% for mood-related diagnoses. In addition, adolescents who were screened were not more likely to fill a prescription for an antidepressant or other mental health medication, nor receive psychotherapy. Associations were generally stronger among female compared with male adolescents. Finally, the proportion of adolescents screened for depression during the index well-visit increased over time.

This study builds on prior findings by using rigorous causal inference methods to examine longitudinal diagnoses and treatment uptake following depression screening in a large, population-based sample of adolescents. Screening rates reported in prior studies are heterogeneous; the rate of screening in this study (5.8%) was higher than that in a sample of privately insured adolescents (1.8%),17 but far lower than that in a pediatric primary care network in Pennsylvania (76.0%).26 Although numerous studies have assessed the efficacy of depression screening with observational data,14,27–29 only one of these directly compared screened to unscreened adolescents.18 Chisholm et al.18 found that screening was associated with increased medical and behavioral services use; the findings of the current study differ, which may be due to the inclusion of additional covariates, and the exclusion of adolescents who were already diagnosed or being treated for depression. Future research could build on the current study by examining the effectiveness of screening among subgroups of adolescents that may have a higher prevalence of depression, such as youth with chronic medical conditions, for whom access to care for depression might be especially important.

There are 2 possible responses to the findings in this study. First, increased resources could be devoted to improving treatment uptake and engagement following screening. Prior studies have noted the central role that parents play in the depression screening process; active approaches that provide psychoeducation to parents and involve them in routine treatment decisions may help ensure that treatment needs are met.30 Preliminary studies of “warm hand-offs,” where an adolescent is introduced to a behavioral health provider at the time of referral, have been found to increase the likelihood of attendance at follow-up appointments.31,32 Other strategies have focused on integration of psychosocial screening tools into electronic health records and the use of decision support systems.29,33–35 Digital and Internet-based interventions are increasingly found to be acceptable and efficacious among youth with depressive symptoms and could provide another avenue for engaging in treatment.36 Policy-based efforts are also critical for increasing the supply of mental health providers, as well as employing consultation models whereby psychiatrists can guide the management of mental health problems by primary care providers.37 Finally, the coronavirus disease 2019 (COVID-19) pandemic has resulted in rapid rollout of telehealth services, which could be adopted in the long term and may help overcome logistical barriers to care.38

The second possible response to these findings, and those of numerous other studies, is that the U.S. Preventive Services Task Force could reconsider their recommendation that depression screening be conducted universally in primary care settings. Mounting evidence from RCTs conducted in adults,39 as well as experiences with universal depression screening programs in other countries,40 suggest that depression screening in unselected populations does not lead to measurable improvements in mental health outcomes. In the absence of clear benefits, there are concerns that screening could waste scarce healthcare resources, which is reflected in decisions by the Canadian Task Force on Preventive Health Care and the United Kingdom National Screening Committee to recommend against routine depression screening in primary care settings.41

There was evidence of heterogeneity by sex: Associations between depression screening and diagnoses and treatment uptake tended to be stronger among female adolescents compared with male adolescents. A well-replicated finding is that male individuals tend to report greater mental health stigma, which serves as a substantial deterrent to help seeking.42 In the context of depression screening, male patients may be more reluctant than female patients to engage in follow-up care. Another explanation is that male individuals may express depression differently from female individuals, with greater symptoms of anger, irritability, and emotional numbness.43 These symptoms may not be included in traditional depression screening measures and may therefore not be considered when diagnosing depression in male adolescents.

Limitations

Some limitations of this study should be noted. The sample used in this study is based primarily in the Eastern U.S. and may not generalize to the country as a whole. The results also may not generalize to adolescents with insurance provided outside of Blue Cross Blue Shield. Health insurance claims are limited in the amount of information they can provide about a given medical encounter. For example, the results of depression screening were not accessible. Although validated exposure and outcome definitions were used where possible, the accuracy of claims data in reflecting certain procedures and health outcomes remains an area of active research.44 Adolescents in the sample may have accessed mental health treatment not covered by insurance, which would not have been captured by claims data. Finally, it is possible that the screening identified in this study may actually be selective and in response to clinical suspicion of depression,8 though the sensitivity analyses suggested that residual confounding appears to be minimal. Strengths of this study include the use of a large, longitudinal sample to examine rare outcomes and the application of rigorous causal inference methods to estimate effects.

CONCLUSIONS

This study found that adolescents who were screened for depression during a well-visit were more likely to receive a diagnosis of depression or a mood-related disorder in the 6 months following screening; however, they were not more likely to be treated with medication or psychotherapy. Future studies should examine long-term health outcomes of depression screening, as well as the implementation of strategies to increase treatment uptake among adolescents.

Supplementary Material

1

ACKNOWLEDGMENTS

Ms. Riehm was supported by a Ruth L. Kirschstein National Research Service Award from the National Institute of Mental Health (1F31MH124330-01) and by a Doctoral Foreign Study Award from the Canadian Institutes of Health Research.

Ms. Riehm conceptualized and designed the study, cleaned and analyzed the data, drafted the initial manuscript, and reviewed and revised the manuscript. Dr. Brignone assembled the data set and reviewed the manuscript. Dr. Stuart, Dr. Gallo, and Dr. Mojtabai assisted in designing the analytic approach, interpreting results, and reviewing the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

CRediT Statement

Kira Riehm: Conceptualization, Methodology, Formal Analysis, Writing – Original Draft Emily Brignone: Data Curation, Writing – Review & Editing Elizabeth A Stuart: Conceptualization, Writing – Review & Editing, Supervision Joseph J Gallo: Conceptualization, Writing – Review & Editing, Supervision Ramin Mojtabai: Conceptualization, Writing – Review & Editing, Supervision

All authors have no potential conflicts of interest to disclose.

REFERENCES

  • 1.Thapar A, Collishaw S, Pine DS, et al. Depression in adolescence. Lancet. 2012;379(9820):1056‒1067. 10.1016/s0140-6736(11)60871-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Clayborne ZM, Varin M, Colman I. Systematic review and meta-analysis: adolescent depression and long-term psychosocial outcomes. J Am Acad Child Adolesc Psychiatry. 2019;58(1):72‒79. 10.1016/j.jaac.2018.07.896. [DOI] [PubMed] [Google Scholar]
  • 3.Leone M, Kuja-Halkola R, Leval A, et al. Association of youth depression with subsequent somatic diseases and premature death. JAMA Psychiatry. 2021;78(3):302‒310. 10.1001/jamapsychiatry.2020.3786. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Mojtabai R, Olfson M, Han B. National trends in the prevalence and treatment of depression in adolescents and young adults. Pediatrics. 2016;138(6):e20161878. 10.1542/peds.2016-1878. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Keyes KM, Gary D, O’Malley PM, et al. Recent increases in depressive symptoms among US adolescents: trends from 1991 to 2018. Soc Psychiatry Psychiatr Epidemiol. 2019;54(8):987‒996. 10.1007/s00127-019-01697-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Burns BJ, Costello EJ, Angold A, et al. Children’s mental health service use across service sectors. Health Aff (Millwood). 1995;14(3):147‒159. 10.1377/hlthaff.14.3.147. [DOI] [PubMed] [Google Scholar]
  • 7.Mojtabai R Universal depression screening to improve depression outcomes in primary care: sounds good, but where is the evidence? Psychiatr Serv. 2017;68(7):724‒726. 10.1176/appi.ps.201600320. [DOI] [PubMed] [Google Scholar]
  • 8.Speechley M, Kunnilathu A, Aluckal E, et al. Screening in public health and clinical care: similarities and differences in definitions, types, and aims - a systematic review. J Clin Diagn Res. 2017;11(3):LE01‒LE04. 10.7860/jcdr/2017/24811.9419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Siu AL. Screening for depression in children and adolescents: US Preventive Services Task Force recommendation statement. Pediatrics. 2016;137(3):e20154467. 10.1542/peds.2015-4467. [DOI] [PubMed] [Google Scholar]
  • 10.Zuckerbrot RA, Cheung A, Jensen PS, et al. Guidelines for Adolescent Depression in Primary Care (GLAD-PC): Part I. Practice preparation, identification, assessment, and initial management. Pediatrics. 2018;141(3):e20174081. 10.1542/peds.2017-4081. [DOI] [PubMed] [Google Scholar]
  • 11.Roseman M, Saadat N, Riehm KE, et al. Depression screening and health outcomes in children and adolescents: a systematic review. Can J Psychiatry. 2017;62(12):813‒817. 10.1177/0706743717727243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Bakken S, Jia H, Chen ES, et al. The effect of a mobile health decision support system on diagnosis and management of obesity, tobacco use, and depression in adults and children. J Nurse Pract. 2014;10(10):774‒780. 10.1016/j.nurpra.2014.07.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Fallucco EM, James L, Smotherman C, et al. Impact of experiential training with standardized patients on screening and diagnosis of adolescent depression in primary care. J Adolesc Health. 2019;65(1):57‒62. 10.1016/j.jadohealth.2018.12.022. [DOI] [PubMed] [Google Scholar]
  • 14.Gadomski AM, Scribani MB, Krupa N, et al. Do the Guidelines for Adolescent Preventive Services (GAPS) facilitate mental health diagnosis? J Prim Care Community Health. 2014;5(2):85‒89. 10.1177/2150131914520711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Samples H, Stuart EA, Saloner B, et al. 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. 10.1007/s11606-019-05192-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Brignone E, George DR, Sinoway L, et al. Trends in the diagnosis of diseases of despair in the United States, 2009–2018: a retrospective cohort study. BMJ Open. 2020;10(10):e037679. 10.1136/bmjopen-2020-037679. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Sekhar DL, Ba DM, Liu G, et al. Major depressive disorder screening remains low even among privately insured adolescents. J Pediatr. 2019;204:203‒207. 10.1016/j.jpeds.2018.07.086. [DOI] [PubMed] [Google Scholar]
  • 18.Chisolm DJ, Klima J, Gardner W, et al. Adolescent behavioral risk screening and use of health services. Adm Policy Ment Health. 2009;36(6):374. 10.1007/s10488-009-0245-8. [DOI] [PubMed] [Google Scholar]
  • 19.Rushton JL, Felt BT, Roberts MW. Coding of pediatric behavioral and mental disorders. Pediatrics. 2002;110(1):e8. 10.1542/peds.110.1.e8. [DOI] [PubMed] [Google Scholar]
  • 20.Fiest KM, Jette N, Quan H, et al. Systematic review and assessment of validated case definitions for depression in administrative data. BMC Psychiatry. 2014;14:289. 10.1186/s12888-014-0289-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Imai K, Khanna K. Improving ecological inference by predicting individual ethnicity from voter registration records. Polit Anal. 2016;24(2):263‒272. 10.1093/pan/mpw001. [DOI] [Google Scholar]
  • 22.Michigan Population Studies Cemter. Zip Code Characteristics: Mean and Median Household Income. https://www.psc.isr.umich.edu/dis/census/Features/tract2zip/. Accessed January 21, 2021.
  • 23.Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46(3):399‒424. 10.1080/00273171.2011.568786. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Green KM, Stuart EA. Examining moderation analyses in propensity score methods: application to depression and substance use. J Consult Clinical Psychol. 2014;82(5):773‒783. 10.1037/a0036515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.VanderWeele TJ, Ding P. Sensitivity analysis in observational research: introducing the E-value. Ann Intern Med. 2017;167(4):268‒274. 10.7326/m16-2607. [DOI] [PubMed] [Google Scholar]
  • 26.Davis M, Rio V, Farley AM, et al. Identifying adolescent suicide risk via depression screening in pediatric primary care: an electronic health record review. Psychiatr Serv. 2020;72(2):163‒168. 10.1176/appi.ps.202000207. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Hacker K, Arsenault L, Franco I, et al. Referral and follow-up after mental health screening in commercially insured adolescents. J Adolesc Health. 2014;55(1):17‒23. 10.1016/j.jadohealth.2013.12.012. [DOI] [PubMed] [Google Scholar]
  • 28.Hacker KA, Penfold RB, Arsenault LN, et al. Behavioral health services following implementation of screening in Massachusetts Medicaid children. Pediatrics. 2014;134(4):737‒746. 10.1542/peds.2014-0453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Sudhanthar S, Thakur K, Sigal Y, et al. Improving validated depression screen among adolescent population in primary care practice using electronic health records (EHR). BMJ Qual Improv Rep. 2015;4(1):u209517.w203913. 10.1136/bmjquality.u209517.w3913. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Kelleher KJ, Hoagwood K. Beyond blame: parents as partners. J Pediatr. 2015;167(4):795‒796. 10.1016/j.jpeds.2015.07.039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Gurney BA, German M, Keller K, et al. Increasing behavioral health appointment attendance using warm handoffs in an integrated primary care setting. Behav Ther (N Y N Y). 2020;43(1):14‒19. [Google Scholar]
  • 32.Germán M, Hsu-Walklet T, Gurney BA, et al. “Nice to meet you”: a quality improvement project to increase warm handoffs. Clin Pract Pediatr Psychol. 2020;8(3):247‒256. 10.1037/cpp0000357. [DOI] [Google Scholar]
  • 33.Murphy JM, Stepanian S, Riobueno-Naylor A, et al. Implementation of an electronic approach to psychosocial screening in a network of pediatric practices. Acad Pediatr. 2021;21(4):702‒709. 10.1016/j.acap.2020.11.027. [DOI] [PubMed] [Google Scholar]
  • 34.Honigfeld L, Macary SJ, Grasso DJ. A clinical care algorithmic toolkit for promoting screening and next-level assessment of pediatric depression and anxiety in primary care. J Pediatr Health Care. 2017;31(3):e15‒e23. 10.1016/j.pedhc.2017.01.008. [DOI] [PubMed] [Google Scholar]
  • 35.Arauz-Boudreau A, Riobueno-Naylor A, Haile H, et al. How an electronic medical record system facilitates and demonstrates effective psychosocial screening in pediatric primary care. Clin Pediatr. 2019;59(2):154‒162. 10.1177/0009922819892038. [DOI] [PubMed] [Google Scholar]
  • 36.Radovic A, Reynolds K, McCauley HL, et al. Parents’ role in adolescent depression care: primary care provider perspectives. J Pediatr. 2015;167(4):911‒918. 10.1016/j.jpeds.2015.05.049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Findling RL, Stepanova E. The workforce shortage of child and adolescent psychiatrists: is it time for a different approach? J Am Acad Child Adolesc Psychiatry. 2018;57(5):300‒301. 10.1016/j.jaac.2018.02.008. [DOI] [PubMed] [Google Scholar]
  • 38.Badawy SM, Radovic A. Digital approaches to remote pediatric health care delivery during the COVID-19 pandemic: existing evidence and a call for further research. JMIR Pediatr Parent. 2020;3(1):e20049. 10.2196/20049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Thombs BD, Markham S, Rice DB, et al. Does depression screening in primary care improve mental health outcomes? BMJ. 2021;374:n1661. 10.1136/bmj.n1661. [DOI] [PubMed] [Google Scholar]
  • 40.Burton C, Simpson C, Anderson N. Diagnosis and treatment of depression following routine screening in patients with coronary heart disease or diabetes: a database cohort study. Psychol Med. 2013;43(3):529‒537. 10.1017/s0033291712001481. [DOI] [PubMed] [Google Scholar]
  • 41.Thombs BD, Saadat N, Riehm KE, et al. Consistency and sources of divergence in recommendations on screening with questionnaires for presently experienced health problems or symptoms: a comparison of recommendations from the Canadian Task Force on Preventive Health Care, UK National Screening Committee, and US Preventive Services Task Force. BMC Med. 2017;15:150. 10.1186/s12916-017-0903-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Clement S, Schauman O, Graham T, et al. What is the impact of mental health-related stigma on help-seeking? A systematic review of quantitative and qualitative studies. Psychol Med. 2015;45(1):11‒27. 10.1017/s0033291714000129. [DOI] [PubMed] [Google Scholar]
  • 43.Oliffe JL, Phillips MJ. Men, depression and masculinities: a review and recommendations. J Mens Health. 2008;5(3):194‒202. 10.1016/j.jomh.2008.03.016. [DOI] [Google Scholar]
  • 44.Townsend L, Walkup JT, Crystal S, et al. A systematic review of validated methods for identifying depression using administrative data. Pharmacoepidemiol Drug Saf. 2012;21(S1):163‒173. 10.1002/pds.2310. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1751721-supplement-1.pdf (208.5KB, pdf)

RESOURCES