DoLS, Restrictions and Dementia: How to Evidence Least Restrictive Practice Day to Day

January 23, 2026

Restrictions in dementia care rarely start as “a deprivation of liberty”. They start as small, practical decisions: a keypad code, a bedroom sensor, a “don’t let her go outside alone”, a medication kept out of reach, a door that is routinely locked. Over time, these decisions can become embedded, poorly reviewed and difficult to justify—especially when staff teams change. A defensible approach is to treat restrictions as an explicit governance issue, not an informal care preference, and to evidence least restrictive practice in daily routines and records. This article builds on the wider Safeguarding, capacity, consent and human rights knowledge set and aligns with core dementia service models that commissioners will expect you to apply consistently.

What counts as a restriction in dementia services?

A restriction is anything that limits a person’s freedom of movement, choice or access to everyday activities, beyond what they would reasonably accept if they had full capacity and no coercion. In practice, restrictions often include:

Locked doors / coded exits or preventing access to outdoor space without supervision.
One-to-one observation used primarily to prevent leaving, rather than to meet a positive support goal.
Bed rails, sensors, pressure mats and other measures that change a person’s ability to move freely.
Withholding personal items (phone, money, keys, lighters) without a clear review pathway.
Routine sedation or “as required” medication used to manage behaviour rather than treat a clinical need.

The defensibility test is simple: can the service explain the specific risk, the alternatives tried, and the review process that keeps the restriction time-limited and proportionate?

From “restriction” to “lawful restriction”: making the pathway operational

Teams can feel overwhelmed by legal terminology, but the operational work is practical:

Identify: what restriction is in place and when it applies (not just “DoLS in place”).
Justify: what risk it addresses, and how it supports the person’s wellbeing or safety.
Minimise: what least restrictive options were tried first, and what is still being tried now.
Review: set a cadence (weekly initially, then monthly) with clear outcomes and exit criteria.

Where authorisation routes apply, the key is that authorisation supports a rights-based framework: it is not “permission to restrict”, but a structured way to ensure restrictions are necessary, proportionate and monitored.

Designing a least restrictive review cycle that actually works on shift

In dementia services, restrictions drift when review is too complex, too infrequent, or reliant on one manager. A workable review cycle includes:

Shift-level prompts: a quick check during handover: what restrictions are active today, what is being trialled, what needs escalation.
Weekly restriction review (short-form): what changed, what incidents occurred, what alternatives were tried, and whether restriction can be reduced.
Monthly governance review: themes across people (staffing patterns, environment issues, activity gaps), and whether restrictions correlate with distress.
Outcome evidence: not just “no incidents”, but wellbeing measures: agitation frequency, sleep quality, engagement, falls, and relational safety.

Operational Example 1: Locked door “for safety” becoming the default

Context: A person with dementia repeatedly attempts to leave the service to “go home”. The team locks external doors and tells staff to redirect, but the person becomes more distressed and attempts to tailgate others.

Support approach: The service defines the specific risks (traffic, getting lost, hypothermia) and tests least restrictive alternatives: planned accompanied walks at predictable times, an “outdoor access routine” after meals, and meaningful roles that reduce exit-seeking triggers (helping set tables, short tasks that provide purpose). The team also introduces environmental supports (clear signage, calm waiting area by the door) and communication techniques that reduce confrontation.

Day-to-day delivery detail: Staff do not simply record “tried to abscond”. They record triggers (time, noise, unmet need), the de-escalation used, whether the person accepted an alternative, and how long it took to settle. The service uses a simple daily log to track exit-seeking frequency and response effectiveness. If the door lock remains necessary, the plan sets clear review points and describes what would need to improve for restriction to be reduced.

How effectiveness or change is evidenced: Evidence includes a reduction in repeated attempts, shorter distress episodes, improved engagement, and fewer tailgating incidents. Review notes show how the service continually tried alternatives rather than relying solely on restriction.

Operational Example 2: Sensor use to reduce falls risk without increasing distress

Context: A person experiences nighttime falls. The team installs a bed sensor that alerts staff, but the noise and repeated staff entry disrupt sleep and increases agitation.

Support approach: The service reframes the goal: reduce falls while protecting sleep and dignity. They trial a package of least restrictive interventions first: medication review for sedatives, toileting schedule, improved lighting and contrast, a low bed, decluttered routes, and a quiet reassurance approach rather than repeated checks.

Day-to-day delivery detail: Night staff document not only sensor activations, but whether the person needed the toilet, appeared in pain, was disoriented, or was seeking comfort. The sensor settings are reviewed (volume, escalation thresholds) and staff are coached to approach quietly and predictably. If the sensor remains in place, the care plan defines it as time-limited and linked to measurable outcomes, with a review date and specific criteria for removal.

How effectiveness or change is evidenced: Falls data, sleep quality notes, agitation frequency, and incident reports show whether the combined approach reduced falls without increasing distress. The review demonstrates proportionality and active minimisation of restriction.

Operational Example 3: “PRN for agitation” drifting into routine sedation

Context: A person becomes distressed in late afternoon. Staff increasingly use PRN medication to prevent escalation, but there is limited documentation of non-pharmacological approaches and the person appears withdrawn.

Support approach: The service treats medication as one option within a restrictive practice framework. They complete a structured review with clinical input and build a non-pharmacological plan: predictable routines, sensory support, a calming activity at the time distress usually peaks, reduced noise, and a consistent staff approach. They define when PRN is appropriate and when it is not.

Day-to-day delivery detail: Staff use an ABC-style prompt (antecedent, behaviour, consequence) to record triggers and what was attempted before PRN. A senior reviews PRN use weekly, looking for patterns and whether alternatives were attempted. The service links PRN decisions to capacity/best interests where relevant and ensures any restrictive element is reviewed and reduced when possible.

How effectiveness or change is evidenced: Reduced PRN frequency, improved engagement, and documented successful de-escalation attempts show the service is minimising restriction. Governance minutes show clinical review outcomes and learning actions (training, staffing adjustments, activity redesign).

Expectations to evidence

Commissioner expectation

Commissioners expect a clear restrictive practice framework: identification of restrictions, documented rationales, evidence of least restrictive alternatives, and a repeatable review cycle that reduces restrictions over time where possible. They will look for auditability and consistency across teams and shifts, not reliance on individual judgement.

Regulator / Inspector expectation (CQC)

CQC will look for person-centred practice that protects rights and dignity: staff who can explain why a restriction exists, how the person’s wishes are considered, how alternatives are tried, and how restrictions are reviewed and stepped down. Inspectors will also test whether records show outcomes and the person’s experience, not only incidents and controls.

Governance controls that prevent “restriction drift”

Strong services use simple, repeatable controls:

Restriction register: what restrictions exist, why, authorisation status where relevant, and next review date.
Quality sampling: monthly checks of restriction records for rationale, alternatives, and review evidence.
Escalation rules: when restrictions must be reviewed by a senior or clinician (e.g., repeated PRN use, repeated exit-seeking incidents, multiple falls).
Learning loop: themes feed into environment changes, staffing patterns, training, and activity design.

The aim is not paperwork—it is a service culture where restriction is always questioned, always justified, and always reviewed with a clear route to reduce it.