Structured teaching

A023 remains a broad practice family. Describe the actual organising support before deciding what evidence is relevant. Keep three questions separate: whether a focused work system changes performance in a particular task; whether a specified classroom or family package from the TEACCH Autism Program at the University of North Carolina improves outcomes against its actual comparison; and whether a person experiences the support as useful and freely usable. One question cannot answer the others.

The findings below remain attached to the components, programmes and populations studied. Proposed mechanisms are reasons to investigate a change, not demonstrated effects. No effect estimate applies to generic structured teaching as a whole.

On 12 September 2026, targeted searches followed named primary studies, the two inherited 2025 review leads, official TEACCH identity and implementation pages, and a focused-practice review. Searches looked for nulls, active comparisons, dose, fidelity (delivery of the approach as intended), original methods, family burden and recipient accounts. References led to original papers and the author institution’s study brief. Paywalls were respected. One publicly author-uploaded manuscript was read in full; publisher abstracts were used only at their actual access level.

This was not an exhaustive database review. No pooled estimates were recalculated, no study authors or recipients were contacted, and no unavailable methods were supplied from assumption. The Zhu review and the author-institution brief were read as extracted full text; their page images were not available for visual checking. Full text was not obtained for the Hume/Odom, Boyd or Family Implemented TEACCH for Toddlers (FITT) journal papers. A separate Hume/Plavnick/Odom 2012 abstract was a supplementary lead, not a second fully examined component study. An unrelated “Going Mobile” parent-intervention lead was excluded after identity checking.

Programme and institution names: LEAP means Learning Experiences: An Alternative Program for Preschoolers and Parents. FITT means Family Implemented TEACCH for Toddlers. NMS denotes non-model-specific education. UNC is the University of North Carolina; FPG is its Frank Porter Graham Child Development Institute. These labels identify different programmes or sources, not equivalent interventions.

S01 — UNC TEACCH Autism Program — Certification¹

• Identity/version: Current Structured TEACCHing terminology, checked 12 September 2026
• Source role: Developer identity and implementation guidance
• Access: Full official web page; no complete manual inspected
• Design: Not an outcome study
• Population/setting: Autism programme; no study sample
• Intended and actual delivery: Four areas: physical organisation, individual schedules, work/activity systems and material/visual structure. Certification describes practitioner and consultant roles. Actual delivery not studied.
• Comparison/co-supports: Not applicable
• Verified outcomes/reporter/time: No independently established efficacy outcome from this source.
• Fidelity/adaptation: Named approach has its own competencies; a tidy room or checklist does not establish fidelity.
• Burden/unwanted effects: Preparation/training are relevant planning considerations; no quantified burden or adverse-event study here.
• Limits: Developer account supports identity, not a comparative efficacy judgement or an assumption that every autistic person understands visual material.
• Checked locator: Page sections: Structured TEACCHing; certification levels. Accessed complete page.

S02 — UNC TEACCH Tips — Visual Schedule²

• Identity/version: Visual-schedule tip, current page
• Source role: Developer practical guidance
• Access: Full official web page
• Design: Not an outcome study
• Population/setting: No study sample
• Intended and actual delivery: Object, photograph, picture and written formats are described; selection depends on understanding. Shows completion and what follows; delivered dose not evaluated.
• Comparison/co-supports: Not applicable
• Verified outcomes/reporter/time: No controlled outcome accepted.
• Fidelity/adaptation: Individualised format is a design principle, not evidence of implementation fidelity in a local setting.
• Burden/unwanted effects: Set-up and updating require consideration; no measured harm rate.
• Limits: The page’s benefit language is not adopted as proof of reduced anxiety.
• Checked locator: Complete tip, particularly Choosing/creating a visual schedule and using it.

S03 — Hume & Odom (2007) — Effects of an individual work system on independent functioning³

• Identity/version: Individual work system; JADD 37:1166–1180; DOI 10.1007/s10803-006-0260-5
• Source role: Direct focused-component study
• Access: Primary publisher abstract only; subscription full text not accessed
• Design: Single-participant withdrawal design
• Population/setting: Three autistic participants; classroom and employment settings. Ages and access profiles not verified.
• Intended and actual delivery: Individual work system; exact teaching, dose and received support unavailable from abstract.
• Comparison/co-supports: Withdrawal phases; concurrent supports insufficiently described in accessible text.
• Verified outcomes/reporter/time: Abstract reports increased observed on-task behaviour/materials completed and reduced teacher prompts, maintained at one month. No numerical effect estimate adopted.
• Fidelity/adaptation: Detailed implementation and fidelity not verified.
• Burden/unwanted effects: Not reported in accessible abstract.
• Limits: Small, specific performance study; neither whole-programme validation nor evidence of subjective autonomy, general executive-function improvement or broad developmental effects.
• Checked locator: Abstract; journal publication metadata.

S04 — Welterlin et al. (2012) — The Home TEACCHing Program for toddlers with autism⁴

• Identity/version: Manualised, individualised Home TEACCHing; DOI 10.1007/s10803-011-1419-2
• Source role: Direct programme pilot
• Access: Complete author-upload manuscript text; publisher DOI metadata; no binary version-equivalence claim
• Design: Randomised waitlist pilot plus nested multiple-baseline probes
• Population/setting: 20 families; autistic children aged 2–3, North Carolina. Nested observation: six families, arranged in three pairs.
• Intended and actual delivery: Twelve weekly 90-minute clinician/parent sessions, modelling, prompting, autism education and homework. Individualised tasks; receipt outside observed sessions not fully quantified.
• Comparison/co-supports: Waitlist; other-service information collected from only half the sample.
• Verified outcomes/reporter/time: Three-month group developmental/adaptive and parent-stress comparisons were nonsignificant. Child work observations improved clearly across two of three pairs; one treatment child reversed.
• Fidelity/adaptation: Clinician observed integrity 87.5–100%; intervention families had only one baseline probe.
• Burden/unwanted effects: Family sessions/homework; child-rated acceptability and systematic adverse outcomes not established.
• Limits: Abstract says six pairs; Methods/Results specify three. Short probes, small sample and concurrent-service uncertainty restrict inference.
• Checked locator: Manuscript Methods pp8–12; Results pp12–15; Limitations pp17–18.

S05 — Boyd et al. (2014; online 2013) — Comparative efficacy of LEAP, TEACCH and non-model-specific special education⁵

• Identity/version: Established classroom models; DOI 10.1007/s10803-013-1877-9
• Source role: Direct active-comparator study
• Access: Primary abstract plus full author-institution research brief S06; full journal article paywalled
• Design: Quasi-experimental comparison, not random assignment
• Population/setting: 198 preschoolers; abstract describes 25 TEACCH, 22 LEAP and 28 non-model-specific (NMS) classrooms.
• Intended and actual delivery: Existing high-fidelity classrooms across a school year; full individual attendance and extra services not verified.
• Comparison/co-supports: Active, high-quality non-model-specific special education and LEAP; no untreated comparator.
• Verified outcomes/reporter/time: Abstract reports improvement across conditions; S06 provides the explicit between-model null finding.
• Fidelity/adaptation: Models selected for fidelity; detailed thresholds and modelling cannot be independently checked from the full article.
• Burden/unwanted effects: No child experience or adverse-event outcome verified in abstract.
• Limits: Within-group gains cannot establish superiority. Abstract classroom count and S06 teacher count are different reporting units; do not silently collapse them.
• Checked locator: Abstract and metadata; S06 is separately registered.

S06 — FPG Snapshot 70 (July 2013) — LEAP and TEACCH comprehensive treatment models⁶

• Identity/version: Summary of Boyd et al. study
• Source role: Official author-institution summary of S05; same cohort, not replication
• Access: Full two-page PDF extracted text; screenshot attempt failed
• Design: Institutional report of quasi-experiment
• Population/setting: 85 TEACCH/54 LEAP/59 NMS children aged 3–5: 198 at start, 185 at end. Teacher counts 25/22/27.
• Intended and actual delivery: At least six months between assessments; experienced trained teachers, model boosters and four fidelity/quality observations. Actual child-level dose not supplied.
• Comparison/co-supports: High-quality active programmes; classrooms maintained intended fidelity.
• Verified outcomes/reporter/time: Reports improvement in autism-severity, communication and fine-motor measures with no significant differences between models.
• Fidelity/adaptation: Confirms programme fidelity; full statistical methods and confidence intervals unavailable here.
• Burden/unwanted effects: No direct child account or systematic burden analysis supplied.
• Limits: No equivalence/noninferiority proof; selection limits causal inference. This brief supplements, rather than substitutes for, full-paper methods access.
• Checked locator: PDF pp1–2; Teachers, Students and Method; Child Gains.

S07 — Turner-Brown et al. (2019; online 2016) — Preliminary efficacy of Family Implemented TEACCH for Toddlers⁷

• Identity/version: FITT, a six-month parent-mediated package; DOI 10.1007/s10803-016-2812-7
• Source role: Direct programme randomised trial
• Access: Primary publisher abstract only; subscription full text not accessed
• Design: Randomised comparison
• Population/setting: 50 autistic children under three and their parents in abstract; final analytic denominator not verified.
• Intended and actual delivery: Planned 20 in-home 90-minute sessions plus four parent groups. Actual attendance/home practice unavailable.
• Comparison/co-supports: Six months of community services as usual; details of other services not verified.
• Verified outcomes/reporter/time: Abstract: parent stress/wellbeing and child social-communication treatment effects; global child measures null. No effect size, reporter specification or confidence interval imported.
• Fidelity/adaptation: Fidelity and adaptations not verified from accessible abstract.
• Burden/unwanted effects: Substantial scheduled family contact; refusal, unwanted effects and unmeasured family workload remain unverified.
• Limits: Different package and duration from S04. Li table lists 49 children; do not replace the primary abstract’s 50 with a guessed attrition explanation.
• Checked locator: Publisher abstract and publication chronology.

S08 — Zhu et al. (2025) — Efficacy of structured teaching program for rehabilitation of children with ASD⁸

• Identity/version: Pak J Med Sci 41(9):2658–2666; DOI 10.12669/pjms.41.9.12634
• Source role: Primary systematic review/meta-analysis
• Access: Complete publisher PDF text; screenshot failed
• Design: Ten controlled studies: five randomised controlled trial (RCT), five nonrandomised; search to 31 January 2025
• Population/setting: Reported mean ages 2.5–9.1; small samples and heterogeneous settings.
• Intended and actual delivery: Varied TEACCH-labelled packages/doses; no single protocol.
• Comparison/co-supports: Standard care/other controlled conditions; components and co-supports differ.
• Verified outcomes/reporter/time: Communication standardised mean difference (SMD) −0.33 [−1.99,1.33]; social 0.12 [−0.61,0.85]; daily living −0.09 [−1.62,1.44]; cognition 0.07 [−0.43,0.57]: nonsignificant. Fine/gross motor estimates positive.
• Fidelity/adaptation: Risk/heterogeneity and inconsistent delivery restrict transfer.
• Burden/unwanted effects: No usable pooled recipient-experience/harms conclusion.
• Limits: Do not infer ineffectiveness from imprecision or adopt speculative mechanisms/add-on recommendations. Table dose discrepancies require original checks; no pooled score assigned to generic A023.
• Checked locator: Methods; Table I; Results and forest plots; Discussion/Limitations.

S09 — Li et al. (2025) — Effectiveness of TEACCH-based interventions in improving adaptive skills⁹

• Identity/version: Transl Pediatr 14:3263–3280; DOI 10.21037/tp-2025-466
• Source role: Primary systematic review/meta-analysis
• Access: Full publisher HTML including main tables; supplement and all original studies not fully reanalysed
• Design: Search to 10 May 2024; 20 studies/920 narratively, only 16 pooled (8 RCT/8 nonrandomised); four one-arm excluded from pooling.
• Population/setting: Varied child samples/settings; outcome denominators vary.
• Intended and actual delivery: Different TEACCH-based packages; inconsistent descriptions.
• Comparison/co-supports: Active education, waitlist/usual services and added components; no uniform comparison.
• Verified outcomes/reporter/time: Communication g = 0.21 [0.03,0.39]; daily living 0.39 [0.04,0.75]; social 0.75 [0.32,1.18]. Motor trim-and-fill −0.02 [−0.21,0.17]; broad benefits not uniform.
• Fidelity/adaptation: Selection bias, sparse blinding and heterogeneous protocols.
• Burden/unwanted effects: Parent stress pooled; child experience/harms not established.
• Limits: Unresolved internal reporting: cognition describes five studies but gives a study count (K) of 6; subgroup counts differ from the main K; motor intervals differ by section; statistical heterogeneity (I²) of 88.2% is described as low. No local reanalysis or optimal-dose inference.
• Checked locator: Methods; Table1; Results; Table2; publication-bias and limitations sections.

S10 — Hume et al. (2021) — Evidence-Based Practices: Third Generation Review¹⁰

• Identity/version: JADD 51:4013–4032; DOI 10.1007/s10803-020-04844-2
• Source role: Primary review of focused practices; contextual, not programme efficacy
• Access: Full publisher article text; separate table request failed; correction read
• Design: Positive-evidence classification review; 972 eligible articles, 1990–2017; 28 practice categories.
• Population/setting: Autistic participants birth–22; studies across settings.
• Intended and actual delivery: Variable focused practices; not a uniform package/dose.
• Comparison/co-supports: Group and single-case designs. Comprehensive programme models explicitly excluded.
• Verified outcomes/reporter/time: Classification includes visual supports; no generic structured-teaching pooled effect. The inherited exact “65” count is not carried forward without independently accessible table verification.
• Fidelity/adaptation: Quality criteria used; classifications are not local fidelity certification.
• Burden/unwanted effects: No estimate of how recipients value A023 from this review.
• Limits: Positive-evidence selection/classification cannot supply an overall benefit–harm rate. Broad components are not automatic validation of a combination. Correction specifies Ayres Sensory Integration; no sensory-treatment efficacy imported.
• Checked locator: Evidence-Based Intervention Approaches; Methods; Results; correction paragraph.

The controlled comparisons deserve more weight than improvement from baseline alone. An active, high-quality comparator tests a different question from waiting for a home programme. Neither a nonsignificant comparison nor a wide interval proves equivalence. Nor should a promising small study be silently converted into a broad claim about communication, emotional regulation or learning.

The two newer reviews overlap in older studies; they are not independent new trial cohorts. Their eligibility, outcome grouping, comparisons and TEACCH-based packages differ. Review discrepancies are retained for scrutiny rather than resolved by choosing the more positive estimate. The source register records primary-access limits and the unresolved reporting issues that prevent confident dose or subgroup prescriptions.

S04’s direct observations concern brief teaching sessions and are separate from its formal group tests. S05/S06 are two publications about one comparative school study. S07’s published abstract supports a mixed-outcome description but does not supply verified outcome-specific sample sizes, reporters, fidelity or harm assessment. No unpublished methods have been inferred to reconcile denominators across papers.

Parent stress is a parent outcome, not the child’s account of acceptability. Teacher prompting and completed materials are performance observations, not direct measures of the young person’s influence over decisions. The selected evidence does not justify saying that all recipients experience structure as calming, respectful or empowering, or that they find it restrictive. Those remain questions to ask directly, with accessible communication and an option to decline.

Locally record preparation, teaching time, upkeep, home practice, effort of using the arrangement and any unwanted restriction. Record who can revise the system, how requests for help/pause/change/refusal are understood, whether meaningful choice remains, and whether an unchanged activity is itself the difficulty. These are editorial safeguards and evaluation questions, not claims that a particular programme has a measured adverse-effect rate.

The review has not established the best amount of structure, which specific component accounts for package outcomes, long-term recipient-valued benefits, comparative costs, or transfer to every non-autistic learner and setting. Individualised use can be considered as an adjustment without pretending that those research gaps have been closed.

• 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, frequency and duration (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 condition (comparator) 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 someone else’s judgement on their behalf (proxy report), and access or participation separate from task performance.
• Delivery as intended (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.

   UNC TEACCH Autism Program — Certification. Source.

   ↩1
2. 2.

   UNC TEACCH Tips — Visual Schedule. Source.

   ↩1
3. 3.

   Hume & Odom (2007) — Effects of an individual work system on independent functioning. Source.

   ↩1
4. 4.

   Welterlin et al. (2012) — The Home TEACCHing Program for toddlers with autism. Source.

   ↩1
5. 5.

   Boyd et al. (2014; online 2013) — Comparative efficacy of LEAP, TEACCH and non-model-specific special education. Source.

   ↩1
6. 6.

   FPG Snapshot 70 (July 2013) — LEAP and TEACCH comprehensive treatment models. Source.

   ↩1
7. 7.

   Turner-Brown et al. (2019; online 2016) — Preliminary efficacy of Family Implemented TEACCH for Toddlers. Source.

   ↩1
8. 8.

   Zhu et al. (2025) — Efficacy of structured teaching program for rehabilitation of children with ASD. Source.

   ↩1
9. 9.

   Li et al. (2025) — Effectiveness of TEACCH-based interventions in improving adaptive skills. Source.

   ↩1
10. 10.

    Hume et al. (2021) — Evidence-Based Practices: Third Generation Review. Source.

    ↩1