Positive Behaviour Support

Which Positive Behaviour Support (PBS) package was actually evaluated, for whom, with what delivery and comparison, and did outcomes reflect the person’s life as well as behaviour?

The evidence needs to distinguish individualised PBS, staff-training models, setting-wide improvement and schoolwide Positive Behavioral Interventions and Supports (PBIS). The adult service-training trial had null comparative clinical findings. The smaller accommodation trial had a favourable setting-level post-intervention result but important nonsignificant individual, engagement and follow-up-change findings. Special-education and schoolwide studies provide different, bounded evidence; they do not override adult null findings.³⁴⁵⁶⁷⁸

The purpose of PBS can include life quality, agency and safety without implying that these outcomes have been demonstrated in every implementation. Staff competence, plan completeness, behaviour ratings, proxy health utility, direct recipient experience and participation must remain separate.

This is targeted evidence reconstruction on 12 September 2026, not a comprehensive systematic review, formal assessment of evidence certainty or clinical publication approval. Earlier draft claims were checked against the retained sources; historical headline ratings are not used as evidence.

Eight entry-level source records are retained: official guidance, a professional identity/definition paper, two original reviews and four reports of three trial cohorts. Hassiotis’s journal and Health Technology Assessment (HTA) reports describe the same trial. Bradshaw is also included in the schoolwide review, and related publications reuse that school cohort. No combined participant, trial or positive-study total is asserted. Across A008/A009 there are 13 records but 12 unique works because National Institute for Health and Care Excellence (NICE) is shared.

Developer/professional materials define the framework and its population boundaries. They do not independently demonstrate benefit. The review did not re-evaluate every primary study within either systematic review. Bradshaw’s original abstract/figures were accessible; its full original methods remain an access limitation, so the article uses only a restrained description of the reported outcomes.

• Adult training, main estimate: Hassiotis’s journal abstract and the HTA report give an adjusted Aberrant Behavior Checklist–Community (ABC-C) difference of −2.14, 95% confidence interval (CI) −8.79 to 4.51. The journal Results prose gives −2.4 with rounded endpoints −8.7 to 4.5. Both support the same null interpretation. The discrepancy remains explicit; the article does not choose a spurious precision.⁴⁵
• Serious-event allocation: Both reports give 45 serious events in 29 people. The journal and HTA Results allocate 26 to intervention and 19 to control; the HTA abstract reverses those arms. Arm-specific rates are not inferred. The events were judged unrelated, which does not establish overall safety.⁴⁵
• Schoolwide review counts: Lee/Gage’s abstract reports 29 studies, 7 randomized and 22 quasi-experimental. Its Results, Table 1 and Discussion report 32, 6 randomized and 26 quasi-experimental. No explanation was established in this review. Neither total is promoted as the resolved number of independent trials.⁸
• Special-education “positive” count: Beqiraj’s abstract’s “significant” wording is broader than its detailed account of heterogeneous single-case outcomes. The article uses the narrative evidence and mixed findings rather than treating 28/30 as a treatment response probability.³
• Follow-up: McGill’s abstract’s maintenance language should not replace the nonsignificant between-group change result. A group difference at a later time and a difference in change from baseline are different analyses.⁶

These are source-reporting issues, not evidence of misconduct. They constrain the claims carried into the working article.

A behavioural outcome can matter greatly while still failing to capture whether a person feels safer, has greater choice or can access a wanted activity. The adult training report includes a positive 12-month proxy-derived quality-adjusted life-year (QALY) signal, an estimate combining length of life and health-related quality of life, and direct interviews with 13 service users; neither cancels the null primary outcome. Proxy utility uses a generic instrument and tariff, while qualitative interview findings concern selected experiences and include absent or difficult support as well as valued relationships.⁵

McGill’s quality-of-life measure was observed engagement during a short observation window, not a full account of life satisfaction or autonomy. Its comparative gain was not significant. Beqiraj found limited formal quality of life (QoL) and direct student social-validity measurement. Thus the article’s insistence on meaningful outcomes is a requirement for practice and review, not a claim those outcomes are already established.³⁶

The revised article preserves genuine refusal and communication access, rejects making ordinary care or connection contingent on compliance, and asks whether a plan narrows participation. These are editorial safeguards informed by person-centred purposes. They do not certify that every study or every service upheld them. Restriction, distress, workload and opportunity costs require explicit local review; missing harms data are not reassurance.

A randomised controlled trial (RCT) allocates participants or settings by chance. A quasi-experimental design (QED) compares groups without random allocation. The comparisons and limitations of each study still need separate examination.

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 A008-S01; count once across both entries.

2. Gore NJ et al. (2022), Positive Behavioural Support in the UK: A State of the Nation Report; IJPBS 12 Supplement 1²

• Role and design: Professional identity/version and values source
• Design: Professional framework/updated definition, not an efficacy study
• Population and setting: UK learning-disability context; explicit caution about transfer to autism without learning disability and other populations
• Intended and actual delivery: 2022 definition integrates rights/values, theory/evidence and process/strategy; twelve components, person-centred assessment and personalized multicomponent support
• Comparison and other support: Not applicable
• Outcomes, reporters and time points: Life quality, self-determination and reduced harmful/restrictive practices are framework goals; no effect assigned
• Fidelity and adaptation: Co-production and implementation are defining expectations, not proof of their delivery
• Burden and unwanted effects: Restrictive/aversive practice rejected as framework aim; cost/harm rates not estimated
• Limits: Authors include PBS proponents and researchers; identity source cannot independently establish efficacy or universal acceptance
• Access: Full original text; relevant methods, results and limitations read
• Exact locators: Printed pp 9, 12–14; 2022 definition Table 1 and person-centred foundation
• Overlap: No independent replication inferred from this record.

3. Beqiraj L et al. (2022), Positive behavioural support for children and young people with developmental disabilities in special education settings: A systematic review³

• Role and design: Original evidence synthesis
• Design: Original systematic narrative review; predominantly single-case studies, one uncontrolled group pre/post study
• Population and setting: 30 studies; eligibility 3–25 years with intellectual disability/autism in special education; aggregate participant count not independently extracted
• Intended and actual delivery: Interventions coded against ≥7/9 PBS characteristics; only 2 explicitly called themselves PBS; components/doses differed
• Comparison and other support: Single-case baseline/control conditions or pre/post; no common comparator
• Outcomes, reporters and time points: 28/30 reported reductions, four mixed across all outcomes; formal QoL indicators in 2 studies; direct student social-validity input in 1
• Fidelity and adaptation: Fidelity high-risk in 15/30; participant/personnel masking uncertain; 10 reported all 9 characteristics
• Burden and unwanted effects: Some behaviour increases; workload/recipient experience incompletely studied; no uniform harms estimate
• Limits: Counts are not pooled effect or response probability; retrospectively framework-aligned packages are not one named programme; no efficacy transfer to adult service training
• Access: Full original text; relevant methods, results and limitations read
• Exact locators: Sections 2.2, 3.2, 3.4.3–3.4.4, 3.6 and Discussion
• Overlap: No independent replication inferred from this record.

4. Hassiotis A et al. (2018), Clinical outcomes of staff training in positive behaviour support to reduce challenging behaviour in adults with intellectual disability: cluster randomised controlled trial⁴

• Role and design: Direct empirical evidence
• Design: Pragmatic cluster RCT
• Population and setting: 23 English community intellectual-disability services; 246 adults allocated 109/137; 233 in primary repeated-measures analysis; 225 at 12 months
• Intended and actual delivery: Six training days for local clinicians plus mentoring; clinicians intended to assess/plan for recipients; only 33/108 intervention records contained all components, 47 partial, 28 no paperwork
• Comparison and other support: Training+usual care versus usual care; medication and other routine psychosocial care continued
• Outcomes, reporters and time points: Carer ABC-C over 12 months: abstract −2.14, 95% CI −8.79 to 4.51, p=.528; no significant clinical secondary differences
• Fidelity and adaptation: Independent review of 33 available plans judged weak; staff turnover and low reach
• Burden and unwanted effects: 45 serious events in 29 people, judged unrelated; mostly physical-health admissions; one death. Arm counts conflict with report abstract
• Limits: Unblinded proxy reporters, incomplete delivery, small carer subgroups; null package-as-delivered result neither proves all PBS ineffective nor proves ideal-fidelity benefit
• Access: Full original text; relevant methods, results and limitations read
• Exact locators: Methods; Primary outcome; Secondary outcomes; Intervention fidelity; Discussion
• Overlap: Same trial as A009-S05, NCT01680276; count as one cohort. Journal body says −2.4 (CI −8.7, 4.5), unlike abstract/report −2.14; retain discrepancy, not silently reconcile.

5. Hassiotis A et al. (2018), Positive behaviour support training for staff for treating challenging behaviour in people with intellectual disabilities: a cluster RCT; HTA 22(15)⁵

• Role and design: Direct empirical evidence
• Design: Full trial report with economic, process and naturalistic 36-month follow-up
• Population and setting: Same 246 adults/23 services; 184 at 36 months; qualitative sample included 13 service users, 11 family carers, 10 paid carers, 12 managers plus 12 therapists/4 trainers
• Intended and actual delivery: Same six-day package; intended 12.5 clinical hours per participant excluding travel/paperwork, mentoring offered; actual delivery incomplete
• Comparison and other support: Same usual care; follow-up naturalistic
• Outcomes, reporters and time points: 36-month ABC difference −3.70 (CI −9.25, 1.85), p=.191. Proxy EuroQol five-dimension youth questionnaire (EQ-5D-Y) derived 12-month QALY difference +.076 (CI .011, .140); economic advantage dissipated later. Interviews mixed
• Fidelity and adaptation: Weak plans and limited reach documented; interviews distinguish positive, absent and difficult engagement
• Burden and unwanted effects: QALY complete data 96 intervention/108 control; workload/turnover and disappointment described; harms total same as journal
• Limits: Proxy EQ-5D-Y with adult tariff is not direct autonomy/QoL testimony; selected interviews not randomized effect; report abstract reverses serious-event arm counts relative to Results
• Access: Full original text; relevant methods, results and limitations read
• Exact locators: Ch 2 Methods: training, QALYs, qualitative procedure; Ch 3 Results: Tables 13/17, long-term follow-up, stakeholder themes; abstract
• Overlap: Companion of A009-S04; no additional trial. Abstract 46 interviews counts service users/family/paid carers/managers; additional therapist/trainer interviews separately reported.

6. McGill P et al. (2018), Reducing challenging behaviour of adults with intellectual disabilities in supported accommodation: a cluster randomized controlled trial of setting-wide positive behaviour support⁶

• Role and design: Direct empirical evidence
• Design: Setting-cluster trial; per-protocol analyses
• Population and setting: 24 supported-accommodation settings, 81 adults aged 19–84, one English provider; 11 experimental/13 control; 21 settings at post-test
• Intended and actual delivery: Setting quality review, improvement standards, manager/staff coaching and monthly monitoring 8–11 months; both arms could access individual PBS
• Comparison and other support: Usual supported accommodation plus access to individual PBS; added setting-wide package in experimental arm
• Outcomes, reporters and time points: Setting-level ABC change favoured intervention at 3–6 months post-package(p=.04); individual sensitivity p=.07; observed engagement/QoL proxy p=.46. Later between-group change p=.13
• Fidelity and adaptation: Standards monitored; primary informants unblinded; post-observers initially blind but usually guessed allocation
• Burden and unwanted effects: Substantial organizational effort; attrition, setting disruption and deaths noted; no comprehensive adverse-event comparison
• Limits: Small one-provider sample; per-protocol/unequal cluster weighting; engagement is narrow QoL indicator; follow-up group difference is not significant change advantage
• Access: Full original text; relevant methods, results and limitations read
• Exact locators: Methods 2.4–2.8; Results 3.2–3.5; Tables 4/5
• Overlap: No independent replication inferred from this record.

7. Bradshaw CP, Waasdorp TE & Leaf PJ (2012), Effects of school-wide positive behavioral interventions and supports on child behavior problems⁷

• Role and design: Direct empirical evidence
• Design: Cluster randomized schoolwide effectiveness trial
• Population and setting: 37 elementary schools; 12,344 pupils; 12.9% special education at baseline; Maryland
• Intended and actual delivery: Schoolwide framework; exact training dose/coaching not verified in original abstract
• Comparison and other support: Comparison schools; detailed concurrent provision not verified in abstract
• Outcomes, reporters and time points: Teacher-rated behaviour, concentration, social-emotional/prosocial outcomes over 5 assessments/4 school years; selected benefits; 33% lower likelihood of office referral reported
• Fidelity and adaptation: School-wide Evaluation Tool (SET) figure visible; full fidelity/missing-data methods not reviewed
• Burden and unwanted effects: Workload, child experience and harms not established from abstract
• Limits: No inference about individualized PBS/adult disability services, all special educational needs and disabilities (SEND) pupils, suspensions or direct emotion-regulation ability
• Access: Original abstract and figures; PMC full-text access returned verification page; original full methods not retrieved in bounded review
• Exact locators: Original abstract and figure captions, PMID 23071207
• Overlap: Included in Lee/Gage review; related Waasdorp 2012 report uses same 37-school cohort, not another trial.

8. Lee A & Gage NA (2020), Updating and expanding systematic reviews and meta-analyses on the effects of school-wide positive behavior interventions and supports⁸

• Role and design: Original evidence synthesis
• Design: Original systematic review/meta-analysis; robust variance estimation for dependent outcomes
• Population and setting: Abstract 29 studies (7 RCT/22 quasi-experimental design (QED)); Results/Table 1 32 (6 RCT/26 QED), 8,781 schools. Internal discrepancy unresolved; mainly US elementary, one Norway study
• Intended and actual delivery: Mostly universal Tier 1; mean 3.4 years implementation; packages not identical
• Comparison and other support: Non-SWPBIS comparison groups; some co-interventions and quasi-experimental designs
• Outcomes, reporters and time points: Selected aggregate school discipline, academic and organizational gains; individual study outcomes mixed; behavioural effect 0.26 standard deviation (SD) (17 studies/45 outcomes)
• Fidelity and adaptation: 10 studies did not report fidelity scores; subgroup/individual-level reporting limited
• Burden and unwanted effects: Burden/harms and recipient experience not synthesized comprehensively
• Limits: Unresolved abstract/body counts; multiple reports can share cohorts; school-level effects not individual PBS efficacy; heterogeneous outcomes/settings
• Access: Full original text; relevant methods, results and limitations read
• Exact locators: Abstract; Results 3.1, Table 1, Table 3; Discussion 4/Limitations 4.1
• Overlap: Includes A009-S07 and companion cohort reports; do not add review counts to direct trials.

• 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.

   Gore NJ et al. (2022), Positive Behavioural Support in the UK: A State of the Nation Report; IJPBS 12 Supplement 1. Original source. Reviewed 12 September 2026. Access: Full original text; relevant methods, results and limitations read.

   ↩1
3. 3.

   Beqiraj L et al. (2022), Positive behavioural support for children and young people with developmental disabilities in special education settings: A systematic review. Original source. DOI: 10.1111/jar.12989. Reviewed 12 September 2026. Access: Full original text; relevant methods, results and limitations read.

   ↩1 ↩2 ↩3 ↩4
4. 4.

   Hassiotis A et al. (2018), Clinical outcomes of staff training in positive behaviour support to reduce challenging behaviour in adults with intellectual disability: cluster randomised controlled trial. Original source. DOI: 10.1192/bjp.2017.34. Reviewed 12 September 2026. Access: Full original text; relevant methods, results and limitations read.

   ↩1 ↩2 ↩3 ↩4
5. 5.

   Hassiotis A et al. (2018), Positive behaviour support training for staff for treating challenging behaviour in people with intellectual disabilities: a cluster RCT; HTA 22(15). Original source. DOI: 10.3310/hta22150; Additional original access 1; Additional original access 2. Reviewed 12 September 2026. Access: Full original text; relevant methods, results and limitations read.

   ↩1 ↩2 ↩3 ↩4 ↩5
6. 6.

   McGill P et al. (2018), Reducing challenging behaviour of adults with intellectual disabilities in supported accommodation: a cluster randomized controlled trial of setting-wide positive behaviour support. Original source. DOI: 10.1016/j.ridd.2018.04.020. Reviewed 12 September 2026. Access: Full original text; relevant methods, results and limitations read.

   ↩1 ↩2 ↩3 ↩4
7. 7.

   Bradshaw CP, Waasdorp TE & Leaf PJ (2012), Effects of school-wide positive behavioral interventions and supports on child behavior problems. Original source. DOI: 10.1542/peds.2012-0243. Reviewed 12 September 2026. Access: Original abstract and figures; PMC full-text access returned verification page; original full methods not retrieved in bounded review.

   ↩1 ↩2
8. 8.

   Lee A & Gage NA (2020), Updating and expanding systematic reviews and meta-analyses on the effects of school-wide positive behavior interventions and supports. Original source. DOI: 10.1002/pits.22336. Reviewed 12 September 2026. Access: Full original text; relevant methods, results and limitations read.

   ↩1 ↩2 ↩3