in progress

Final Research Plan

Screening for Depression, Anxiety, and Suicide Risk in Children and Adolescents

August 13, 2020

Recommendations made by the USPSTF are independent of the U.S. government. They should not be construed as an official position of the Agency for Healthcare Research and Quality or the U.S. Department of Health and Human Services.

This figure is an analytic framework depicting the key questions within the context of the populations, interventions, comparisons, outcomes, time frames, and settings (PICOTS) relative to the effectiveness and harms of screening and treatment for depression, anxiety, and suicide risk in children and adolescents. The figure illustrates screening of adolescents and children in populations representative of primary care, followed by possible harms of screening; diagnosis of major depressive disorder, anxiety disorder, or high risk for suicide; treatment; possible harms of treatment; and health outcomes.

Abbreviation: KQ=key question.

  1. Do depression, anxiety, or suicide risk screening programs in primary care or comparable settings result in improved health outcomes in children and adolescents?
  2. Do instruments to screen for depression, anxiety, or suicide risk accurately identify children and adolescents with depression, anxiety, and increased risk of suicide in primary care or comparable settings?
  3. What are the harms associated with screening for depression, anxiety, or suicide risk in primary care or comparable settings in children and adolescents?
  4. Does treatment (psychotherapy, pharmacotherapy, or collaborative care) of depression, anxiety, or suicide risk result in improved health outcomes in children and adolescents?
  5. What are the harms of treatment (psychotherapy, pharmacotherapy, or collaborative care) in children and adolescents who are treated for depression, anxiety, or suicide risk?

Contextual questions will not be systematically reviewed and are not shown in the Analytic Framework.

  1. What is the diagnostic yield from screening for depression, anxiety, or suicide risk in typical primary care practice settings?
  2. What are the minimal clinically important differences (the smallest value of benefit to patients) for symptoms and functioning on the most common instruments used to measure response to treatment of depression, anxiety, or suicide risk?
  3. What are the U.S. Food and Drug Administration boxed warnings for pharmacotherapy for the treatment of depression, anxiety, or suicide risk in children and adolescents?
  4. What psychotherapies other than cognitive behavioral therapy are used to treat anxiety in children and adolescents?
  5. What is the effectiveness of evidence-based treatment in children and adolescents with persistent depressive disorder and depressive disorders not otherwise specified?
  6. What proportion of children and adolescents who screen positive for depression, anxiety, or increased suicide risk engage with care (i.e., return for clinical evaluation and treatment)?

The Research Approach identifies the study characteristics and criteria that the Evidence-based Practice Center will use to search for publications and to determine whether identified studies should be included or excluded from the Evidence Review. Criteria are overarching as well as specific to each of the key questions.

Criteria Include Exclude
Condition definition Major depressive disorder, as defined by DSM criteria (present in at least 50% of the enrolled study population)

Anxiety disorders include generalized anxiety disorder, social anxiety disorder, panic disorder, agoraphobia, separation anxiety disorder, and selective mutism

Definitions for increased risk of suicide may vary by study but may include suicidal ideation (suicidal thoughts or plan for suicide), history of suicide attempts (nonfatal, self-directed, and potentially injurious behavior that is intended to result in death), and deliberate self-harm

Included studies may address these conditions individually or in combination

Other mental health disorders (e.g., obsessive compulsive disorder, posttraumatic stress disorder, psychotic disorders, bipolar disorder, cyclothymia, adjustment disorder with depressed mood), persistent depressive disorder/dysthymia, disruptive mood dysregulation disorder, premenstrual dysphoric disorder, substance/medication-induced depressive disorder, depressive disorder due to another medical condition, and depression not otherwise specified; substance/medication-induced anxiety disorder, anxiety disorder due to another medical condition, and anxiety not otherwise specified
Population KQs 1–3: Children and adolescents (mean age ≤18 years). Studies may include:
  • Unselected primary care population
  • Primary care patients without known depression, anxiety disorders, or increased risk of suicide (including deliberate self-harm)
  • Comparable community-based population

KQs 4, 5: Children and adolescents (age ≤18 years) with major depressive disorder, anxiety disorders, or increased risk of suicide

A priori priority populations of interest include by age (children vs. adolescents), race/ethnicity, gender identity, and sexuality

  • Adults (age ≥19 years)
  • Studies in which more than 50% of the population are age 19 years or older
  • Studies limited to populations that are not broadly generalizable to primary care populations (e.g., populations with mental health conditions other than anxiety, depression, and increased suicide risk); persons with treatment-resistant depression or anxiety; persons in residential, institutional, or inpatient settings; persons with developmental disorders (e.g., autism spectrum disorder or ADHD); persons in the midst of a suicidal crisis that are identified through their use of health care services related to a suicide attempt (e.g., in the emergency department); studies that require patients to have a specific clinical condition for enrollment (e.g., cancer, chronic illness, epilepsy)
Interventions KQs 1–3: Screening interventions with or without additional provider or patient-facing elements such as referral support, treatment guidelines, symptoms monitoring, and standardized treatment. Screening tools must be brief standardized instruments designed to identify persons with major depressive disorder, anxiety disorders, or an increased risk of suicide; self-report with or without parental report), clinician administered, or electronically delivered (<5 minutes if clinician administered, <15 minutes if self-administered) instruments are eligible

KQs 4, 5 (depression and suicide):

  • Counseling (e.g., psychotherapy, psychoeducation, suicide means restriction)
  • Care delivery models targeting improved mental health outcomes (e.g., collaborative care, care management)
  • First-line pharmacotherapy agents approved for pediatric use (e.g., duloxetine, fluoxetine, escitalopram, sertraline, fluvoxamine, clomipramine)
  • Combination therapies

KQs 4, 5 (anxiety):

  • Cognitive behavioral therapy (including exposure therapy)*
  • Eligible psychotherapy studies regardless of mode of intervention
  • First-line pharmacotherapy agents approved for pediatric use (e.g., clonidine, duloxetine, fluoxetine, escitalopram, sertraline, fluvoxamine)
  • Combination therapies
KQ 1–3: Studies reporting on a screening instrument that does not have established validity and scoring mechanism or thresholds for use within clinical practice

KQs 4, 5 (all disorders): Other treatment modalities (e.g., exercise, light therapy, transcranial magnetic nerve stimulation, electroshock treatment, diet and herbal supplements such as St. John’s wort and other complementary and alternative medicine, social marketing, policy, system-level interventions, or adjunctive agents to enhance the effects of antidepressants)

Interventions involving components that could not be replicated in most health care settings, including environmental components (media messages, public signage), interventions on groups in closed (preexisting) social networks (e.g., in daycares, schools), or those requiring the parent to have the target condition

Pharmacotherapeutic agents that are not FDA approved for pediatric use (e.g., paroxetine, vortioxetine)

KQs 4, 5 (anxiety): Psychotherapy other than cognitive behavioral therapy

Comparators KQs 1, 3 (screening): Usual care/no screening

KQ 2 (depression and anxiety): Clinical diagnosis based on structured clinical interview by qualified professional using standard diagnostic criteria in place at the time of the study (e.g., DSM-IV or DSM-5)

KQ 2 (suicide): Assessment of increased suicide risk based on clinical interview by qualified professional

KQs 4, 5 (psychotherapy and care delivery):

  • No intervention
  • Waitlist control (i.e., delayed treatment)
  • Attention control (i.e., receives interpersonal interaction but no other elements of the active intervention)
  • Usual care (e.g., referral to treatment, nonstandardized treatment, or unclear treatment services)

KQs 4, 5 (suicide risk): Treatment as usual (the provision of standard treatment services not governed by a study protocol, but at a duration and level of intensity consistent with active treatment interventions) are also eligible

KQs 4, 5 (pharmacotherapy): Placebo (including placebo along with psychotherapy, when compared with the active agent plus the same psychotherapy intervention [e.g., CBT plus placebo vs. CBT plus medication would be eligible])

KQs 1, 3: No comparator

KQ 2: Another screening instrument, nonstandardized clinical diagnosis (i.e., diagnosis not made based on existing DSM criteria at the time of the study)

KQs 4, 5: No comparator, active intervention (i.e., comparative effectiveness); for example, medication X vs. medication Y would not be eligible, nor would CBT plus medication X vs. CBT plus medication Y

KQ 4, 5 (anxiety and depression): Treatment as usual comparator groups where the comparator group receives standard treatment services that involve a reasonably standardized active intervention provided outside of a study protocol are not eligible
Outcomes KQs 1, 4:
  • Depression or anxiety symptoms, remission or diagnosis, or response
  • Suicide deaths, suicide attempts and deliberate self-harm, or suicidal ideation
  • All-cause mortality
  • Quality of life measured using validated scales or instruments
  • Functioning (using validated scales or instruments, days of missed school)

KQ 2:

  • Sensitivity, specificity, or data to calculate one or both
  • Negative predictive value, positive predictive value, area under the curve/area under the receiver operating characteristic/receiver operating characteristic, diagnostic odds/likelihood ratios, Youden's index

KQ 3:

  • False alarm
  • False reassurance

KQs 3, 5:

  • Treatment avoidance
  • Deterioration in patient-provider relationship
  • Labeling or stigma
  • Inappropriate/unnecessary treatment

KQ 5 (pharmacotherapy only):

  • Serious adverse effects
  • Withdrawals due to adverse effects
  • Suicidality
All other outcomes
Outcome assessment timing No minimum followup Not applicable
Setting KQs 1–3: Recruitment from:
  • Primary care settings (e.g., pediatrics, family medicine, or school-based health clinics)
  • Virtual or community setting, if population comparable to general primary care (i.e., focus on “healthy” children or adolescents, or broad spectrum of medical and mental health conditions in rates comparable to primary care setting)
  • General emergency departments
  • Studies recruiting participants from schools, primary care, or virtual or community settings but conducting the study elsewhere (i.e., mental health clinics) are also eligible

KQs 4, 5: Treatment in:

  • Primary care or specialty clinics, including school-based health clinics
  • Virtual or community-based settings
  • General EDs are eligible for recruitment of patients to an intervention; however, interventions delivered solely/entirely within an ED setting are not eligible
KQs 1–3: Recruitment from:
  • Established patients at mental health clinics
  • Inpatient/residential facilities
  • Correctional facilities
  • Psychiatric emergency departments

Studies conducting school-wide or communitywide screening are not eligible

KQs 4, 5: Treatment in:

  • Correctional facilities
  • Schools (involving schoolwide interventions)
  • Inpatient/residential facilities
  • Psychiatric emergency departments
Study design KQs 1, 3: RCTs, CCTs

KQ 2: Studies of diagnostic test accuracy**

KQ 3: RCTs, CCTs, observational studies

KQ 4: RCTs

KQ 5:

  • RCTs
  • Systematic reviews of comparative cohort and case-control observational studies

Harms of pharmacotherapy only: large (>1,000 participants) comparative cohort and case-control observational studies published after identified systematic reviews that include observational studies

All other study designs

KQ 2: Psychometric development and internal (e.g., split sample) validation studies of new instruments; case-control studies (i.e., studies that limit the study sample to only those with and without known mental health symptoms)

KQs 1–4: Systematic reviews of RCTs (reviews will only be used to identify relevant studies)

Study geography Primary studies that primarily take place in countries categorized as “Very High” on the 2019 Human Development Index (as defined by the United Nations Development Programme) Reviews in which >50% of included studies take place in countries not categorized as “Very High” on the Human Development Index
Publication language English Any language other than English
Quality rating Fair- or good-quality studies Poor-quality studies

* We will summarize the effect of other non-CBT interventions for anxiety as a contextual question, using a best-evidence approach.
** We will catalog all studies reporting on instruments that otherwise meet all eligibility criteria, but our synthesis will focus on the instruments that are reported in more than one study.

Abbreviations: ADHD=attention deficit hyperactivity disorder; CBT=cognitive behavioral therapy; CCT=controlled, clinical trial; DSM=Diagnostic and Statistical Manual of Mental Disorders; ED=emergency department; FDA=U.S. Food and Drug Administration; KQ=key question; RCT=randomized, controlled trial.

The draft Research Plan was posted on the USPSTF website for public comment from April 30, 2020 to May 27, 2020. Regarding suggested edits to the KQs, one commenter noted that screening is intended to identify those at increased risk for any of the eligible conditions (anxiety, depression, suicide risk, or a combination), not just those at increased risk of suicide. In response, the USPSTF edited the key questions to remove the qualifier “increased risk of suicide” and instead refers to “suicide risk.” One comment suggested a focus on implementation barriers rather than the effectiveness of screening. Although we agree that these factors are important, the review is intended to support a screening recommendation. Commenters suggested edits to the contextual questions (CQs) to improve clarity. In response, we revised CQ 2 to clarify that the term “minimal clinically important differences” refers to the smallest value of benefit to patients. We revised CQ 3 and CQ 4 to specify that the population of interest is children and adolescents. We qualified CQ 5 as being limited to evidence-based treatments and clarified that engagement with care in CQ 6 refers to returning for clinical evaluation and treatment.

Regarding suggested edits to the inclusion and exclusion criteria, one commenter suggested focusing on screen-detected populations in reviewing treatment literature. Although we agree that for treatment questions, screen-detected populations are ideal, the evidence is likely to be extremely sparse. As a result, we are not restricting treatment studies to screen-detected populations alone. One commenter suggested excluding clomipramine because it is not a first-line therapy; in response, we excluded clomipramine. Some reviewers asked about the exclusion of active comparators for treatment questions. The USPSTF considers comparative effectiveness to be outside of its scope. Commenters suggested several outcomes; in response, we have clarified that we will include validated outcomes for prespecified outcomes for KQ 1 and KQ 4 and have added false alarm and false reassurance to outcomes for KQ 3. Some comments focused on priority populations; the review will report on priority populations defined by age, sex, race/ethnicity, gender identity, and sexuality. Some comments highlighted the importance of context, applicability, type of intervention, and type of disorder. We will consider these factors in the analysis.