Functional assessment

What does functional assessment tell a team, how do its methods differ, and what can be inferred about the benefit of subsequent support?

The material evidence supports disciplined uncertainty. Indirect reports and descriptive sequences can help frame an assessment, but their agreement with experimental results is variable. Experimental functional analysis investigates controlled relations under particular conditions; it does not read intention or explain all of a person’s life. A study that changes assessment methods and then delivers treatment addresses an assessment-plus-treatment pathway, not the effect of assessment alone.²³⁴⁵

A008 therefore retains no universal direct Encyclopaedia category. The assessment question determines whether emotional, communication, sensory/environmental or safety/participation reading is relevant. These categories are not competing diagnoses or confirmed functions.

This is targeted evidence reconstruction on 12 September 2026, prompted by the complete captured article and an earlier unverified profile. Searches pursued primary method-comparison papers, an original correspondence review, a randomized comparison of assessment sequences and official guidance. Search paths and access limitations are retained in the structured register. No exhaustive database search, preregistered selection, independent screening, pooled analysis or formal assessment of evidence certainty is claimed.

Five material sources are retained: three original empirical papers, one original review available at abstract level, and official guidance. These are not five intervention trials. National Institute for Health and Care Excellence (NICE) is shared with A009. The Functional Analysis Screening Tool (FAST) validation subset overlaps its reliability sample; Thompson reuses earlier data; Call reports a subset of a larger telehealth project. Review counts are not added to direct-study participants.

Looking for barriers can help a team plan support without assigning blame. The reviewed evidence does not establish that this process always produces more sustainable change than generic consequences. Observed patterns identify possible contributors to investigate; they do not prove triggers. Greater team consistency is a process aim, not a promised effect.

For the Call trial, a null difference cannot prove equivalence. Both arms had more assessment and expertise than ordinary antecedent–behaviour–consequence (ABC) recording. The high reduction among completers is particularly unsuitable as a general success rate because treatment continued until the criterion was reached or participation stopped. The article retains the result as a boundary on assuming that more experimental assessment necessarily improves every subsequent outcome.⁵

People’s communication and accounts remain necessary even when they do not identify a measurable contingency. A formulation must distinguish observed events, proxy interpretation and first-person experience. Health, pain, trauma and safeguarding questions cannot be settled by a behaviour questionnaire. NICE provides a proportionate assessment/planning framework within learning disability; its guidance is not proof of intervention efficacy for all special educational needs and disabilities (SEND) populations.¹

The original experimental papers were read to establish distinctions and limits, not to create a procedural guide. They do not authorize staff to provoke dangerous behaviour, deny basic support or reproduce experimental conditions. Appropriate expertise, consent/assent, risk assessment, and a defensible balance of information against burden are required. The article does not operationalize experimental testing.

No source here establishes universal acceptability, a comprehensive harm rate or improved quality of life from assessment alone. The person’s ability to refuse, the privacy and burden of repeated records, communication access and immediate support are explicit editorial review criteria. Absence of reported harms is not evidence of safety.

The entries below retain source role, actual comparison and access limits. Unverified or unreported details are not filled from general knowledge. The linked source register is the structured counterpart.

1. NICE NG11 (2015), Challenging behaviour and learning disabilities: prevention and interventions; full guideline recommendation summary¹

• Role and design: Official guidance; scope and practice rationale
• Design: Guideline, not a trial
• Population and setting: Children, young people and adults with learning disabilities; care and educational settings
• Intended and actual delivery: Phased assessment and a written, reviewed support plan; assess communication, health, environment and resources; no fixed dose
• Comparison and other support: Not applicable
• Outcomes, reporters and time points: Guidance specifies behaviour and quality-of-life baselines and review; no effect estimate extracted
• Fidelity and adaptation: Competence and sustainable delivery considered; adaptations individualized
• Burden and unwanted effects: Assessment effort, staff/family resources and risk management are explicit; no study harm rate
• Limits: Scope is learning disability, not every pupil or every behaviour; recommendation is not causal proof.
• Access: Full original text; relevant methods, results and limitations read
• Exact locators: Recommendations 14.5.1–14.5.12; 14.6.1 (full guideline numbering)
• Overlap: Same work as A009-S01. Count once across the linked entries.

2. Thompson RH & Iwata BA (2007), A comparison of outcomes from descriptive and functional analyses of problem behavior²

• Role and design: Direct empirical evidence
• Design: Within-person method comparison using existing descriptive data
• Population and setting: 12 adults aged 30–52 with severe/profound intellectual disability; selected from an earlier 27-person sample; day programme and assessment settings
• Intended and actual delivery: Naturalistic observations compared with later specialist experimental assessment within one year
• Comparison and other support: Methods differed in setting, timing and controlled conditions; no intervention comparison
• Outcomes, reporters and time points: Most frequent observed consequence matched experimental findings in 3/12 cases; attention was most frequent for 8, confirmed for 2 of those 8
• Fidelity and adaptation: Mean observational agreement 95% in sampled experimental sessions
• Burden and unwanted effects: Specialist assessment time and potentially distressing procedures; no systematic harm endpoint identified
• Limits: Small selected adult sample; observational associations are not causes; not a universal 25% accuracy estimate
• Access: Full original text; relevant methods, results and limitations read
• Exact locators: pp 333–337; Methods, Table 1 and Results/Discussion
• Overlap: Reuses Thompson/Iwata 2001 descriptive data and some previously published experimental data; these are not additional replications.

3. Iwata BA, DeLeon IG & Roscoe EM (2013), Reliability and validity of the Functional Analysis Screening Tool³

• Role and design: Direct empirical evidence
• Design: Two linked reliability/correspondence studies of the 16-item FAST
• Population and setting: 151 people with intellectual disability or autism, ages 5–53; 196 behaviours; three specialist populations. Experimental comparison: 59-person subset, 69 analyses
• Intended and actual delivery: Two familiar informants independently completed questionnaires, typically 15–20 min; specialist assessment for subset
• Comparison and other support: Informant agreement and agreement against experimental results, not treatment efficacy
• Outcomes, reporters and time points: Item agreement 71.5%; outcome agreement 64.8%; FAST predicted highest-rate experimental condition in 44/69 analyses (63.8%)
• Fidelity and adaptation: Experimental observations checked; questionnaire reporter backgrounds varied
• Burden and unwanted effects: Informant recall and assessment burden; no systematic recipient safety/acceptability outcome
• Limits: Multiple behaviours per person; agreement rules permit tied predictions; FAST insufficient as a stand-alone basis for treatment; not a proof of intention
• Access: Full original text; relevant methods, results and limitations read
• Exact locators: pp 275–282; Studies 1/2, Tables 1/5, General Discussion
• Overlap: Study 2 subset of Study 1; do not add 151+59 as independent participants.

4. Contreras BP, Tate SA, Morris SL & Kahng S (2023; online 2022), A systematic review of the correspondence between descriptive assessment and functional analysis⁴

• Role and design: Original evidence synthesis; limited access
• Design: Original systematic review
• Population and setting: 48 studies; aggregate participant count/ages not verified from accessible abstract
• Intended and actual delivery: Descriptive methods varied; structured and unstructured approaches compared with experimental assessment
• Comparison and other support: Method correspondence, not intervention effectiveness
• Outcomes, reporters and time points: Abstract: exact correspondence in 50% of comparisons; structured descriptions performed better; absence predicted better than presence
• Fidelity and adaptation: Full coding and study-level methods not accessible in this review
• Burden and unwanted effects: Burden/harms not extractable from abstract
• Limits: Comparison denominator and population breakdown unavailable; cannot use 50% as individual diagnostic probability
• Access: Original abstract and bibliographic record only; full methods not verified
• Exact locators: Original abstract, PMID 36409837
• Overlap: No independent replication inferred from this record.

5. Call NA et al. (2024; online 4 December 2023), A comparative effectiveness trial of functional behavioral assessment methods⁵

• Role and design: Direct empirical evidence
• Design: Stratified randomized comparison of assessment sequences followed by functional communication training (FCT)
• Population and setting: 57 young autistic children, non-FA 31/functional analysis (FA) 26; specialist telehealth work with caregivers; entry 18–83 months, automatic reinforcement and complicating medical conditions screened out
• Intended and actual delivery: Both groups received indirect+ABC and structured descriptive assessment; FA group also experimental analysis; then caregiver FCT with expert coaching until criterion or discontinuation
• Comparison and other support: Added FA versus no FA within otherwise intensive assessment/FCT pathway; not ordinary school ABC versus specialist care
• Outcomes, reporters and time points: Observed FCT success criterion reached by 25/31 vs 22/26 allocated participants (80.6% vs 84.6%; p=.69); 47 completers all reached criterion. No significant treatment-session advantage
• Fidelity and adaptation: Fidelity available 36/57; means structured descriptive assessment (SDA) 96.7%, FA 92.1%, FCT 96.4%
• Burden and unwanted effects: FA added about 22 assessment sessions on average; caregiver time/technology; no systematic safety comparison established
• Limits: Selected socially mediated targets, small groups, expert clinicians, incomplete fidelity; nonsignificance is not equivalence. Success-stopping design makes completer reductions unsuitable as general response rates
• Access: Full original text; relevant methods, results and limitations read
• Exact locators: Original PDF pp 2–5, 11–14; Tables 1–3 and treatment comparison
• Overlap: Subset of a larger telehealth project; related reports are not independent replications. Intervention outcomes belong to the FCT pathway, not assessment alone.

• 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.

This common key applies to the retained source summaries below. Shorter headings such as design, dose or findings are read in that context. The key supplies no missing source data, converts no proxy report into recipient testimony and assigns no new certainty grade.

1. 1.

   NICE NG11 (2015), Challenging behaviour and learning disabilities: prevention and interventions; full guideline recommendation summary. Original source; Additional original access 1. Reviewed 12 September 2026. Access: Full original text; relevant methods, results and limitations read.

   ↩1 ↩2
2. 2.

   Thompson RH & Iwata BA (2007), A comparison of outcomes from descriptive and functional analyses of problem behavior. Original source. DOI: 10.1901/jaba.2007.56-06; Additional original access 1. Reviewed 12 September 2026. Access: Full original text; relevant methods, results and limitations read.

   ↩1 ↩2
3. 3.

   Iwata BA, DeLeon IG & Roscoe EM (2013), Reliability and validity of the Functional Analysis Screening Tool. Original source. DOI: 10.1002/jaba.31. Reviewed 12 September 2026. Access: Full original text; relevant methods, results and limitations read.

   ↩1 ↩2
4. 4.

   Contreras BP, Tate SA, Morris SL & Kahng S (2023; online 2022), A systematic review of the correspondence between descriptive assessment and functional analysis. Original source. DOI: 10.1002/jaba.958. Reviewed 12 September 2026. Access: Original abstract and bibliographic record only; full methods not verified.

   ↩1 ↩2
5. 5.

   Call NA et al. (2024; online 4 December 2023), A comparative effectiveness trial of functional behavioral assessment methods. Original source. DOI: 10.1002/jaba.1045; Additional original access 1. Reviewed 12 September 2026. Access: Full original text; relevant methods, results and limitations read.

   ↩1 ↩2 ↩3