Graduated exposure

What supports using planned exposure within appropriately assessed treatment of fear/anxiety in children and young people, and what cannot be inferred for ordinary avoidance, school access or speaking differences?

The answer depends on the treatment, population, comparison group and outcome. Specific-phobia pathways, broader anxiety cognitive behavioural therapy (CBT) and a selective-mutism package are different records. This profile does not turn all avoidance into a therapeutic target, test a custom school ladder, or establish that exposure alone explains a package effect.

Eight material works are retained: three original randomized trial reports, two original reviews, two National Institute for Health and Care Excellence (NICE) guidelines and NHS public guidance. ASPECT’s quantitative and qualitative chapters are one work and share the same trial. Trial participants are not added to review totals; a source count is not a count of independent replications. No new pooled estimate, formal certainty rating or universal professional endorsement is supplied.

This is a targeted review of the stated sources, not an exhaustive systematic review or a new certainty assessment. Source-specific access limits remain in the records below.

The targeted search began with the partial article and the complete historical A015 evidence-profile section. Original phobia-treatment, autism-anxiety and selective-mutism trials were pursued, together with the original youth exposure-optimisation review, the original Cochrane summary and official clinical guidance. The search register (internal editorial record) preserves query groups and failures. This was not a registered or exhaustive search, independent screening exercise or reassessment of every trial in the reviews.

ASPECT, Wood and Oerbeck original methods/results and material limitations were read, together with the full Plaisted review. Oerbeck was available as original article text on an author-hosted page; unrelated citing-paper text was excluded. Cochrane was read at detailed original abstract/public-summary level. NICE recommendation text was available in indexed official-source results, but direct opening returned access errors; no fresh appraisal of full guideline evidence tables is claimed. The NHS page displays a 2023 review date and a February 2026 review-due date; access in September 2026 does not make it a newly reviewed publication.

The historical reference to a 2026 synthesis of 290 trials and a 75-study component association is not carried forward as verified evidence. The historical stakeholder scoping-review lead was identified but not comprehensively extracted here. A protocol or a review lead is not a newly verified result. The earlier manual-edition question also remains bounded: Oerbeck cites inspiration from Johnson and Wintgens (2007), but the manual was not itself accessed and that trial is not evidence for another edition or any locally assembled speaking programme.

The article retains its fear/anxiety scope and collaborative premise. It narrows the implication that a well-ordered sequence itself ensures learning or benefit. A hierarchy can organise a clinical plan; the shape of that hierarchy is not independently validated simply because a multicomponent treatment did well.

The non-inferiority design of ASPECT is especially important. It asks whether a specified shorter pathway performs within a predefined margin of usual CBT, on a particular outcome and time point. It does not ask whether exposure is better than no exposure, whether a tiny-step approach is superior to larger steps, or whether all participants improve. The observed task, diagnostic/interference measures and family accounts should stay distinct.

The autism and selective-mutism trials make generalisation limits concrete. They do not describe the same population or the same package. A person’s diagnosis alone does not identify the treatment component, while an apparent refusal or lack of speech does not establish an anxiety mechanism. The article therefore keeps assessment, current communication access and chosen participation goals in view.

Inhibitory learning is treated as a theoretical account with open developmental and clinical questions. Completion, momentary calm or a fall in a distress rating does not directly measure that proposed mechanism. The article does not claim that a fear memory has been erased or that a particular dose is neurologically necessary.

The following are study/source records. “Not reported” or “not extracted” means unknown in this review, not zero burden, no support or no effect. Recommendations are not recorded as if recipients actually received them.

1. A015-S01¹

• Identity and version: NICE CG159 (2013), Social anxiety disorder: recognition, assessment and treatment; current indexed recommendations
• Source role: Official clinical guidance, not an efficacy trial
• Design: Evidence-informed guideline; no trial n
• Population and setting: Children/young people with assessed social anxiety; developmental maturity matters
• Intended and actual delivery: Individual CBT 8–12 ×45 minutes; group 8–12 ×90 minutes, including exposure, psychoeducation and skills work. Recommendation, not observed receipt.
• Comparison and other support: Not a comparative study; parents may be involved
• Outcome, reporter and time point: No new effect estimate
• Fidelity and adaptation: Individual maturity and parent involvement specified
• Burden and unwanted effects: Session/time burden specified; no harms incidence estimate
• Limits: This indication-specific CBT guidance is not a standalone hierarchy or recommendation for every avoided task.
• Access: Indexed official recommendation text retrieved; direct page opening returned 403; no new full-guideline appraisal.
• Exact locators: Recommendations 1.5.3–1.5.6; assessment scope
• Overlap: One work/source record; not added to review participant or trial totals.

2. A015-S02²

• Identity and version: Wright et al. (2022), One-session treatment compared with multisession CBT in children aged 7–16 years with specific phobias: ASPECT; HTA 26(42)
• Source role: Direct comparative treatment evidence plus nested recipient/parent/clinician accounts
• Design: Pragmatic assessor-blinded non-inferiority randomised controlled trial (RCT); ISRCTN19883421
• Population and setting: 268 randomized, 134/134; ages 7–16; 26 sites including 12 NHS trusts. Unsafe exposure or unsuitable clinical needs excluded case by case.
• Intended and actual delivery: one-session treatment (OST) intended 1-hour assessment plus up-to 3-hour treatment with graduated exposure and other CBT components. Among treated: mean 2.2 sessions/3.8 hours; CBT 5.9 sessions/6.1 hours.31% OST/27% CBT received none.
• Comparison and other support: Usual multisession CBT, also containing exposure; no no-exposure arm
• Outcome, reporter and time point: 6-month observed Behavioural Avoidance Task, an observed test of approaching the feared situation (BAT): 149 intention-to-treat (ITT)/113 per-protocol analyses. Standardized CBT − OST difference −0.123 (95% confidence interval (CI) −0.449, 0.202); per-protocol (PP) −0.204 (−0.579, 0.171); upper limits below prespecified 0.4 margin. No secondary between-arm differences.
• Fidelity and adaptation: Recordings sparse; 4 OST exposure sessions reviewed; 15 rated sessions overall: 8 high/7 moderate fidelity.
• Burden and unwanted effects: Four serious hospitalizations judged unrelated; no treatment-attributed events reported.27/34 approached children and young people (CYP) interviewed: many valued control, some felt rushed. Parents described fatigue/travel/work burdens; selected recipients, not all referrals.
• Limits: Missing BAT data/COVID/waiting times limit inference; assessed phobia packages, not step size, universal equivalence or isolated exposure.
• Access: Full original report: relevant Methods, Results, fidelity, harms and qualitative sections read.
• Exact locators: Ch 2 eligibility/interventions; Ch 4 intervention summaries, primary outcomes, fidelity, safety and qualitative Studies 1/2
• Overlap: ASPECT full report, journal reports and nested interviews share the trial; one source record, one trial cohort; interview participants not additional trial participants.

3. A015-S03³

• Identity and version: James, Reardon, Soler, James & Creswell (2020), Cognitive behavioural therapy for anxiety disorders in children and adolescents; CD013162.pub2
• Source role: Original systematic review; package-level contextual evidence
• Design: 87 randomised controlled trials (RCTs); search through October 2019
• Population and setting: 5,964 participants under 19 with anxiety diagnoses; no included samples with intellectual impairments
• Intended and actual delivery: Variable CBT formats with direct child/parent contact; exact exposure dose not isolated in this extraction
• Comparison and other support: Waitlist/no treatment, usual care, attention and alternative treatments distinguished
• Outcome, reporter and time point: Primary-diagnosis remission vs waitlist: 49.4% vs 17.8%, odds ratio (OR) 5.45 (95% CI 3.90–7.60), 39 studies/n 2697. Vs usual care: OR 3.19 (0.90–11.29), 8/n 487.
• Fidelity and adaptation: Heterogeneous formats; no consistent format/subgroup differences established
• Burden and unwanted effects: Attrition used as acceptability proxy; adverse-effect monitoring unclear despite none reported
• Limits: Not isolated exposure; imprecise active comparisons are not equivalence. Source authors’ quality judgements are not a new Atlas grade.
• Access: Original detailed abstract and public summary; all individual trial extractions not rerun.
• Exact locators: Main results; Selection criteria; Authors’ conclusions
• Overlap: Review overlaps youth CBT literature; not an extra trial cohort and no combined participant sum.

4. A015-S04⁴

• Identity and version: Plaisted, Waite, Gordon & Creswell (2021), Optimising Exposure for Children and Adolescents with Anxiety, obsessive-compulsive disorder (OCD) and post-traumatic stress disorder (PTSD); CC&FPR 24: 348–369
• Source role: Original systematic review; optimisation limits and theory context
• Design: 29 articles; PsycINFO/Medline search November 2019
• Population and setting: Youth anxiety/fear/OCD/PTSD; clinical and nonclinical studies; varied age ranges
• Intended and actual delivery: Exposure procedures within varied protocols; before/during/after-exposure factors
• Comparison and other support: Mixed comparisons and within-treatment associations, not uniformly randomized component tests
• Outcome, reporter and time point: No significant optimisation finding replicated at same timepoint with same methodology; follow-up 4 days–1 year
• Fidelity and adaptation: Weaknesses included fidelity, withdrawals and power; only 2 generalisation assessments
• Burden and unwanted effects: Side effects often insufficiently reported
• Limits: Inhibitory learning is a proposed account with developmental uncertainties; associations do not prove component causation or justify removing genuine access/safety supports.
• Access: Full original article Methods, Tables 1–3, Discussion/limitations read; underlying studies not individually rerun.
• Exact locators: Introduction; Methods; Quality Ratings; Table 3; Discussion; Strengths and Limitations
• Overlap: Review study reports may overlap wider CBT literature; no pooled cross-source count.

5. A015-S05⁵

• Identity and version: Wood et al. (2020; online 22 November 2019), Cognitive Behavioral Treatments for Anxiety in Children With Autism Spectrum Disorder; JAMA Psychiatry 77: 474–483
• Source role: Direct comparative whole-package evidence; programme developers among authors
• Design: Three-arm randomized assessor-blinded trial; NCT02028247
• Population and setting: 167 randomized, 145 completed; verbal autistic children 7–13, intelligence quotient (IQ) ≥70; 3 US universities
• Intended and actual delivery: 16 weekly sessions: Coping Cat 60 minutes, Behavioural Interventions for Anxiety in Children with Autism (BIACA) 90 minutes with more parent involvement/modular supports; exposure in both. Actual mean dose not extracted.
• Comparison and other support: Adapted CBT vs standard CBT vs usual services/referrals for 4 months; coexisting medication/services present
• Outcome, reporter and time point: Post-treatment assessor Pediatric Anxiety Rating Scale, a clinician-rated measure of anxiety severity (PARS): BIACA 2.13 vs Coping Cat 2.43 (p=.04), treatment as usual (TAU) 2.93 (p<.001); secondary outcomes parent-rated
• Fidelity and adaptation: Sampled-session required-topic adherence 97%/96%; weekly expert supervision
• Burden and unwanted effects: Two serious events judged unrelated; unequal time/parent burden; direct recipient experience not the reported primary outcome
• Limits: Unequal dose/components; mostly male and selected verbal/IQ range. Not an exposure-only trial or evidence for all autistic/communication-access profiles.
• Access: Full original publisher Methods, Results, Tables and Limitations read.
• Exact locators: Participants; Interventions; Therapist Fidelity; Table 2; Results adverse events; Limitations
• Overlap: One work/source record; not added to review participant or trial totals.

6. A015-S06⁶

• Identity and version: NICE CG170 (2013), Autism spectrum disorder in under 19s: support and management; current indexed recommendations
• Source role: Official clinical/access guidance, not efficacy trial
• Design: Guideline; no trial n
• Population and setting: Autistic under 19s with coexisting anxiety; verbal/cognitive engagement requirement explicitly stated for CBT recommendation
• Intended and actual delivery: Consider adapted group/individual CBT; visual/concrete material, parent help, breaks/interests; guidance rather than measured receipt
• Comparison and other support: Not a comparative study
• Outcome, reporter and time point: No new effect estimate
• Fidelity and adaptation: Adapting delivery recommended; not proof every adapted protocol is effective
• Burden and unwanted effects: Attention to communication, pain, environment and safeguarding; no harms incidence
• Limits: Eligibility wording describes this CBT guidance’s scope; it does not deny support to people outside that scope.
• Access: Official indexed recommendations retrieved; direct page 403. Full evidence guideline not reconstructed.
• Exact locators: Recommendations 1.4.1 and 1.7.2–1.7.3
• Overlap: One work/source record; not added to review participant or trial totals.

7. A015-S07⁷

• Identity and version: Oerbeck et al. (2014; online 26 October 2013), A randomized controlled trial of a home and school-based intervention for selective mutism—defocused communication and behavioural techniques; CAMH 19: 192–198
• Source role: Direct small package trial, delivered with developer supervision
• Design: Randomized trial, 12 treatment/12 waiting; all 24 analysed at 3 months
• Population and setting: Norwegian children 3–9; six bilingual; excluded pervasive developmental disorder, IQ<50, other active selective mutism (SM) treatment/medication and non-Norwegian-speaking parents
• Intended and actual delivery: 21 sessions: 3 weekly home sessions then twice-weekly school sessions; defocused communication, graded behavioural work and rewards
• Comparison and other support: Both groups initial parent/staff psychoeducation; waitlist then no further contact for 3 months
• Outcome, reporter and time point: Teacher School Speech Questionnaire, completed by teachers (SSQ) time×group p=.029; mother school/total improved; home/public subscale group differences nonsignificant
• Fidelity and adaptation: Manual checklists and session supervision; no independent fidelity rating
• Burden and unwanted effects: No systematic harms/recipient experience established in report; home/school staffing and speech-contingent rewards require scrutiny
• Limits: Unblinded outcomes, small sample; package not isolated exposure. Cited 2007 Johnson/Wintgens manual inspiration is not a trial of any current edition.
• Access: Complete original article text via author-hosted ResearchGate; website’s page-range metadata differs, original printed pp 192–198 used. Independent editorial review additionally checked exclusions and primary/secondary findings in official publisher full text on 12 September 2026.
• Exact locators: pp 193–195 Methods/intervention; p 196 Outcome; pp 197–198 Limitations
• Overlap: One work/source record; not added to review participant or trial totals.

8. A015-S08⁸

• Identity and version: NHS, Selective mutism; page last reviewed 17 February 2023, next review due 17 February 2026
• Source role: Official public identity/access guidance
• Design: Not a comparative study
• Population and setting: Situation-specific inability to speak; communication/development/language context matters
• Intended and actual delivery: Reduce speaking pressure; accessible assessment and coordinated support; graded techniques described
• Comparison and other support: Not applicable
• Outcome, reporter and time point: No new trial effect extracted
• Fidelity and adaptation: Individual communication routes and environmental changes described
• Burden and unwanted effects: Avoid pressure/bribery/public praise; retain nonverbal communication
• Limits: Past displayed review-due date; consulted 12 September 2026, not certified as updated 2026. Not current manual validation.
• Access: Full public page read.
• Exact locators: Diagnosing selective mutism; Treating; Advice for parents; review footer
• Overlap: One work/source record; not added to review participant or trial totals.

ASPECT provides direct youth interview material, distinct from parent views and outcome ratings. Its consenting, treated interview sample cannot represent those who declined treatment, never started or did not agree to interview. The source contains both accounts of useful control and collaboration and concerns about pace. Parent accounts add practical and recovery burdens. These are findings about experience; they are not randomized tests of which interpersonal ingredient caused improvement.²

The other trial records do not provide an equivalent direct account from every recipient. Teacher or parent ratings of speech and anxiety should not stand in for whether a child felt able to refuse, understood a task or valued the outcome. Missing adverse-effect reporting, or completion of treatment, cannot settle those questions.

For proposed use, the article’s editorial safeguards require accessible agreement and refusal, retained communication and essential support, and attention to actual danger or unmet need. These are conditions for responsible review. They are not presented as independently proven treatment components or as a retrospective claim that all studied packages followed the same principles. In particular, the historical selective-mutism package’s speech-contingent rewards are described rather than silently removed from its identity or recommended wholesale.

Direct topic 01 remains warranted by the defined fear/anxiety purpose. Conditional 03 applies only to an assessed speaking-anxiety application. Conditional 06 applies to the person’s valued or necessary safe participation goal. Neither conditional tag turns silence, sensory barriers or an unwanted outing into an indication.

All three specific linked targets were read in full and their hashes are in paired-link decisions (internal editorial record). A verifiable E091 error needs separate correction: its paragraph and reference label describe a step-size experiment but link to ASPECT’s one-session-versus-multisession comparison. A015 explicitly avoids inheriting that claim. E030’s A017/A018 development-priority labels are historical target maintenance, not evidence that current working copies are absent. No Encyclopaedia source was edited.

These definitions describe the record structure; they are not additional findings or filled person-level data.

• Identity and version: Name the component or programme, its edition or protocol where relevant, and the purpose being evaluated.
• Source role: Separate direct comparative evidence, descriptive implementation findings, guidance, theoretical support and firsthand experience.
• Population and setting: Retain the studied age, communication/access characteristics and setting. State when an intended application lies outside them.
• Intended and actual delivery: Record the planned components and dose separately from what was received, by whom, and with what support or interruptions.
• Comparison and other support: Name the actual comparison and concurrent changes. If there is no comparison group or it is unreported, say so rather than supply an assumed usual-care condition.
• Outcome and reporter: Record the outcome, time point and reporter. Keep the young person’s account separate from proxy judgement, and access or participation separate from task performance.
• Fidelity and adaptation: Record whether the core components were delivered and describe adaptations; distinguish an adapted application from the protocol tested.
• Burden and unwanted effects: Record time, effort, practical access, cost or workload where known, alongside refusal, fatigue, distress, restriction and other unwanted consequences. Unreported burden is not evidence of no burden.

1. 1.

   NICE CG159 (2013), Social anxiety disorder: recognition, assessment and treatment; current indexed recommendations. Original source. Reviewed 12 September 2026. Indexed official recommendation text retrieved; direct page opening returned 403; no new full-guideline appraisal.

   ↩1
2. 2.

   Wright et al. (2022), One-session treatment compared with multisession CBT in children aged 7–16 years with specific phobias: ASPECT; HTA 26(42). Original source; DOI 10.3310/IBCT0609. Additional original access 1; Additional original access 2. Reviewed 12 September 2026. Full original report: relevant Methods, Results, fidelity, harms and qualitative sections read.

   ↩1 ↩2
3. 3.

   James, Reardon, Soler, James & Creswell (2020), Cognitive behavioural therapy for anxiety disorders in children and adolescents; CD013162.pub2. Original source; DOI 10.1002/14651858.CD013162.pub2. Reviewed 12 September 2026. Original detailed abstract and public summary; all individual trial extractions not rerun.

   ↩1
4. 4.

   Plaisted, Waite, Gordon & Creswell (2021), Optimising Exposure for Children and Adolescents with Anxiety, OCD and PTSD; CC&FPR 24: 348–369. Original source; DOI 10.1007/s10567-020-00335-z. Reviewed 12 September 2026. Full original article Methods, Tables 1–3, Discussion/limitations read; underlying studies not individually rerun.

   ↩1
5. 5.

   Wood et al. (2020; online 22 November 2019), Cognitive Behavioral Treatments for Anxiety in Children With Autism Spectrum Disorder; JAMA Psychiatry 77: 474–483. Original source; DOI 10.1001/jamapsychiatry.2019.4160. Additional original access 1. Reviewed 12 September 2026. Full original publisher Methods, Results, Tables and Limitations read.

   ↩1
6. 6.

   NICE CG170 (2013), Autism spectrum disorder in under 19s: support and management; current indexed recommendations. Original source. Reviewed 12 September 2026. Official indexed recommendations retrieved; direct page 403. Full evidence guideline not reconstructed.

   ↩1
7. 7.

   Oerbeck et al. (2014; online 26 October 2013), A randomized controlled trial of a home and school-based intervention for selective mutism—defocused communication and behavioural techniques; CAMH 19: 192–198. Original source; DOI 10.1111/camh.12045. Additional original access 1; Additional original access 2. Reviewed 12 September 2026. Complete original article text via author-hosted ResearchGate; website’s page-range metadata differs, original printed pp 192–198 used. Independent editorial review additionally checked exclusions and primary/secondary findings in official publisher full text on 12 September 2026. Official publisher cross-check.

   ↩1
8. 8.

   NHS, Selective mutism; page last reviewed 17 February 2023, next review due 17 February 2026. Original source. Reviewed 12 September 2026. Full public page read.

   ↩1