Functional Assessment Data Sources: Triangulating Evidence for PBS Formulation in UK Services

February 5, 2026

Functional assessment in Positive Behaviour Support (PBS) stands or falls on evidence. If services rely on one source (for example, incident logs alone), formulations drift into opinion and the resulting plans become inconsistent, hard to teach, and difficult to defend to families, commissioners and regulators. A practical way to strengthen formulation is to treat evidence gathering as a structured, time-limited piece of work with clear roles, agreed data sources and a defined decision point. This article focuses on the day-to-day realities of building a robust evidence base for functional assessment and formulation while staying anchored in PBS principles and values.

What “good evidence” looks like in a PBS functional assessment

In UK regulated services, the aim is not academic perfection. It is a defensible, shared understanding of:

What the behaviour looks like (observable description, not labels)
When it is more or less likely (patterns and triggers)
What seems to maintain it (what changes immediately before/after)
What unmet needs sit underneath (communication, pain, fear, trauma, sensory overload, loss of control)
What has already been tried and why it did or did not work

Evidence should be proportionate to risk. A low-frequency behaviour with limited impact may need basic structured observation plus review of incident records. High-risk behaviour, repeated restrictive practice, or safeguarding concerns justify a more intensive approach, including health checks, more observation time, and formal review through governance routes.

Core data sources to triangulate (and how to gather them realistically)

1) Direct observation (structured, brief, repeated)

Observation is where teams often struggle because it feels time-consuming. The trick is to make it repeatable and brief, using small “slices” of time rather than long shadowing sessions. For example:

3 x 20-minute observations across different shifts
One focused mealtime observation
One focused personal care observation
A short observation during transitions (return from day service, staff handover, medication time)

Use a simple template: setting/context, what happened immediately before, what the person did, what staff did, what changed immediately after, and how the episode ended. The aim is to identify patterns the whole team can recognise.

2) Incident and restraint data (trend, not anecdote)

Incident logs matter, but they must be analysed as a trend: time of day, location, staff mix, activity type, known triggers (noise, crowding, demand), and outcomes (injury, property damage, safeguarding actions). In well-run services, analysis is reviewed at least monthly (and more frequently for high-risk cases), with learning actions tracked to completion.

3) Health, pain and medication information (often the missing link)

Health information is frequently separated from PBS work. In reality, constipation, dental pain, infections, sleep disruption, side effects, withdrawal, epilepsy changes, and sensory issues commonly sit underneath distress. Functional assessment should always include a “health check sweep” appropriate to the person: sleep pattern review, bowel chart where needed, medication review triggers, and escalation routes to GP/specialist teams.

4) Communication and sensory profile information

Many formulations fail because they ignore communication. If a person’s needs are expressed through behaviour, teams must define what communication is available (speech, gesture, AAC, objects of reference) and what support is needed to reduce uncertainty. Sensory needs should be described in practical terms: sound sensitivity, lighting, touch preferences, crowding tolerance, and how staff can adapt environments and routines.

5) Family and carer insight (as evidence, not “nice to have”)

Families often carry the “long view” of what helps and what escalates. Build this in as structured input: what triggers existed historically, what routines were protective, what signs show early distress, and what de-escalation approaches have worked. The key is to record it in a way that can be used operationally (clear cues and actions), not left as narrative notes.

Operational example 1: Night-time incidents and repeated PRN use

Context: A supported living tenant experiences repeated night-time agitation with frequent PRN medication requests and occasional property damage. Staff reports vary: some describe “random behaviour”, others link it to “being unsettled”.

Support approach: The service runs a two-week functional assessment sprint. The senior support worker coordinates brief observations at three different times (late evening, overnight, early morning), while the deputy manager reviews incident data and PRN records. A health review is triggered to consider sleep, pain and medication timing.

Day-to-day delivery detail: Staff introduce a consistent pre-sleep routine (reduced noise, dim lighting, predictable sequence), complete a simple sleep diary, and record early indicators (pacing, checking doors, repeated reassurance-seeking). Overnight staff use a scripted reassurance approach and a “choice board” to reduce repeated verbal demands. Environmental adjustments are trialled (blackout curtain, white noise option, hydration reminder earlier in the evening).

How effectiveness is evidenced: PRN requests are tracked weekly, night-time incident frequency is plotted, and sleep diary data shows longer settled periods. The GP review identifies medication timing changes and constipation management, which correspond with a reduction in agitation.

Operational example 2: “Aggression” during personal care and staffing assumptions

Context: In a registered care setting, a person frequently hits out during personal care. Some staff attribute it to “refusal” or “attention seeking”. Incidents trigger occasional restraint, raising restrictive practice concerns.

Support approach: The PBS lead runs structured observations during personal care with two different staff members, and gathers input from family about past trauma and preferred approaches. The team reviews whether capacity and consent are being handled consistently under the Mental Capacity Act, and whether the plan describes least restrictive options.

Day-to-day delivery detail: Staff agree a consistent “ask–pause–support” approach: explain, offer choice, wait, and proceed only with clear consent indicators (or best-interests rationale where required). The service changes sequencing (wash face/hands first, then pause), offers towel coverage for dignity, uses the same two staff for a settling period, and introduces a visual “now/next” prompt to reduce unpredictability. The team logs early distress signs (tensing, turning head away) and stops at the first sign, switching to de-escalation rather than pushing through.

How effectiveness is evidenced: The service tracks reduction in restraint use, notes improved participation, and audits Mental Capacity documentation quality (capacity assessments, best-interests notes, and evidence of least restrictive practice). Family feedback is recorded after two weeks and again at six weeks.

Operational example 3: Community access incidents and safeguarding thresholds

Context: A person supported in the community becomes distressed in shops, leading to shouting, throwing items, and occasional absconding risk. Staff begin avoiding community activities, reducing quality of life and raising safeguarding concerns about isolation and rights restriction.

Support approach: The team reviews antecedent patterns (time, environment, crowding), sensory triggers, and communication needs. They also review risk assessments to ensure “positive risk-taking” is balanced with safety and does not default to blanket restrictions.

Day-to-day delivery detail: The service introduces graded exposure: short visits at quieter times, pre-visit planning with the person using pictures, clear exit plans, and a consistent de-escalation script. Staff carry agreed sensory supports (noise-reduction option, preferred object), and use a “break card” to allow the person to step away without conflict. Incidents are reviewed within 24–48 hours with the same template so learning is consistent across staff.

How effectiveness is evidenced: Community access frequency increases safely, incident severity reduces, and the person shows increased tolerance. The service tracks time-in-community, number of successful visits, and incidents requiring additional support, and reviews the plan monthly in governance.

Commissioner expectation: evidence-led formulation that can be audited

Commissioner expectation: Commissioners typically expect providers to demonstrate that PBS assessments are structured, proportionate to risk, and produce measurable outcomes. Practically, this means you can show:

What data sources were used and over what time period
How family/carer input was gathered and reflected in the plan
How health factors were considered and escalated
How restrictive practice is monitored and reduced over time
How learning is recorded, actioned and reviewed

Regulator expectation: CQC-ready reasoning for least restrictive practice

Regulator / Inspector expectation (CQC): CQC will look for clear reasoning that links assessment evidence to day-to-day support and demonstrates the least restrictive approach. Teams should be able to explain:

How staff understand the person’s distress and respond consistently
How capacity, consent and best-interests decisions are evidenced where relevant
How incidents and restrictive interventions are reviewed and reduced
How governance oversight works (who reviews, how often, and what changes)

Governance: making sure the assessment doesn’t sit in a drawer

Evidence gathering only matters if it leads to a plan that is implemented, reviewed and improved. Strong governance is simple and repetitive:

Named owner for the formulation and plan (often PBS lead or service manager)
Implementation check within 7–14 days (are staff using it as written?)
Outcome review at 4–6 weeks (are incidents reducing? is quality of life improving?)
Restrictive practice oversight (restraint/PRN/environmental restrictions tracked and challenged)
Learning loop from incidents into plan amendments, with version control

When services can show that loop, functional assessment becomes a live operational tool rather than paperwork.