A younger man stands near an older woman. A first person VR training scenario for aged care and disability services, seen through the headset.

1 in 5 aged care residents is under a restrictive practice: what 5 frameworks ask of training

In January to March 2026, 20.7% of assessed aged care residents were under a restrictive practice, the national QI Program reports.

By Kim Loza, Head of Product at People Tech Revolution

Published 25 September 2026 Last updated 28 September 2026 Version 2.0

Connect with Kim on LinkedIn Cite this research

20.7%

Aged care residents under a restrictive practice

QI Program, January to March 2026

1,044,851

NDIS reportable incidents notified in a year

NDIS Commission, 2020-21

0.86

Risk ratio for restraint after least-restraint programs

Cochrane review, 2023

24%

Rise in monthly restrictive interventions, England

CQC, 2024/25

The measure

How common is restrictive practice in Australian residential aged care?

About 1 in 5 residents: 20.7% of assessed residents were recorded under a restrictive practice in January to March 2026, chemical restraint excluded (QI Program).

An older woman sits near a younger man on a garden bench, seen through the headset. — A garden bench, seen through the headset.

A simulated aged-care hallway. A man in his mid-eighties stands holding the rail as a carer wearing the headset approaches. — A carer approaches a man at the rail.

An older woman speaks to a seated visitor. A first person VR training scenario for aged care and disability services, seen through the headset. — An older woman speaks to a visitor.

Residents under a restrictive practice, and under a secure area alone, 2021 to 2026

Any restrictive practiceSecure area alone

See the data (percent of assessed residents)

	Sep 2021	Dec 2021	Mar 2022	Jun 2022	Sep 2022	Dec 2022	Mar 2023	Jun 2023	Sep 2023	Dec 2023	Mar 2024	Jun 2024	Sep 2024	Dec 2024	Mar 2025	Jun 2025	Sep 2025	Dec 2025	Mar 2026
Any restrictive practice	23.0%	21.9%	21.4%	21.5%	21.2%	19.8%	19.5%	18.1%	17.4%	17.8%	17.7%	18.8%	19.3%	19.3%	19.8%	20.8%	21.4%	21.8%	20.7%
Secure area alone	17.2%	16.8%	16.7%	16.9%	16.8%	15.7%	15.7%	14.4%	13.8%	14.0%	13.7%	14.6%	15.2%	15.1%	15.3%	15.7%	16.4%	16.5%	15.3%

Percent of assessed residents, quarters ending September 2021 to March 2026. Source: AIHW GEN, QI Program tables, Table 1.1.

The rate fell to 17.4%, then climbed back above 20%

From 23.0% in July to September 2021 the share fell to 17.4% in July to September 2023, then rose to 21.8% in October to December 2025. Across the 19 quarters the Department’s own trend test reads “No change” (risk ratio 0.998).

So what for you: Read your quarterly result against this line before and after any training.

A VR aged care and disability services training scenario in which a care worker walks beside an older man holding a handrail in a corridor. — A care worker walks beside a resident.

44,949 of 206,334 residents in late 2025

That is the October to December 2025 count (QI Program), covering physical, mechanical and environmental restraint and seclusion in a 3-day window.

A facility manager pauses alone on the landing. A VR training scenario for aged care and disability services. — A facility manager pauses on the landing.

Most of it is a secure area

Residents restricted exclusively through a secure area were 15.3% of all residents in January to March 2026, or 73.9% of the restricted group. In July to September 2021 the share was 74.9% (QI Program).

A support worker wearing a headset sits beside an older woman in a sunroom with plants at the window. — A support worker beside an older woman.

Counts before July 2021 are a different measure

Until 30 June 2021 the program counted restraint devices per 1,000 care days: 3.95 in July to September 2019 and 2.92 in April to June 2021. From 1 July 2021 it counts residents, as a percentage, so the 2 series cannot be joined (QI Program).

Reporting

Do incident reports show how often restrictive practices are used?

No. In April to June 2025 providers made 198 notifications of inappropriate use while 42,213 residents were recorded under a restrictive practice, about 213 residents for each notification (ACQSC).

Residential serious incident notifications by type, April to June 2025

Unreasonable use of force7,221

Inappropriate use of restrictive practices198

All other incident types7,405

All residential notifications14,824

See the data (notifications)

Measure	Notifications
Unreasonable use of force	7,221
Inappropriate use of restrictive practices	198
All other incident types	7,405
All residential notifications	14,824

Notifications, April to June 2025. Source: ACQSC, Sector Performance Report, SIRS.

Force dominates the reports; restrictive practice sits near 200

Unreasonable use of force accounted for 7,221 notifications, against 14,824 in all. Inappropriate use of restrictive practices ranged from 160 to 232 a quarter from July to September 2023 to July to September 2025.

So what for you: An incident count describes the reporting rule, not how often a practice is used.

A VR aged care and disability services training scenario in which a support coordinator leans forward at a desk while a young participant answers. — A support coordinator reviews a plan.

Disability counts every unauthorised use

NDIS providers notified 1,044,851 reportable incidents in 2020-21. 98.7% concerned unauthorised restrictive practices, and 93% of those were chemical or environmental restraint (NDIS Commission).

A simulated community-services meeting room. A man in his wheelchair leads the conversation, his mother beside him, a planner wearing the headset opposite. — A man in a wheelchair leads a meeting.

Regulated use is reported every month

In April to June 2025, 15,746 participants, 2.1% of all participants, were subject to a regulated restrictive practice (NDIS Commission). Unauthorised use fell from 0.78% to 0.75% of participants between mid 2022 and mid 2023 (Annual Report).

Other markets

What does a sustained fall in restraint look like?

In US nursing homes, long-stay residents physically restrained fell from 0.33% in the January 2019 release to 0.13% in December 2025, about 60% lower (CMS).

US long-stay nursing home residents physically restrained, releases 2019 to 2026

Jan 2019 release0.33%

Jan 2020 release0.25%

Jan 2021 release0.21%

Jan 2022 release0.17%

Jan 2023 release0.15%

Jan 2024 release0.13%

Feb 2025 release0.14%

Dec 2025 release0.13%

Aug 2026 release0.13%

See the data (percent physically restrained)

Measure	Percent physically restrained
Jan 2019 release	0.33%
Jan 2020 release	0.25%
Jan 2021 release	0.21%
Jan 2022 release	0.17%
Jan 2023 release	0.15%
Jan 2024 release	0.13%
Feb 2025 release	0.14%
Dec 2025 release	0.13%
Aug 2026 release	0.13%

Percent of long-stay residents, national, CMS measure 409, releases January 2019 to August 2026. Source: CMS provider data, State and US averages.

A 60% fall across 7 years, then a plateau

It has held near 0.13% since the January 2024 release, across 14,690 certified nursing homes. Definitions differ from Australia’s, so read the shape, not the level.

So what for you: A reduction program needs a measure held steady for years.

A simulated community mental health room. A man in his mid-twenties sits forward, elbows on his knees; a clinician in the headset sits back, giving him room, holding the silence. — A clinician listens to a young man.

England is moving the other way

Average restrictive interventions a month in mental health services rose 24%, from 13,240 to 16,462, between 2023/24 and 2024/25; people subject to them rose 22%, from 2,929 to 3,563 (CQC). CQC says the rise could in part reflect better reporting.

An older man sits at a craft table. A first person VR training scenario for aged care and disability services, seen through the headset. — An older man at a craft table.

Settings differ sharply

In English care homes where over half of residents are autistic or have a learning disability, restraint incidence was nearly 12 times that of homes with none, and around 15% to 20% more residents had restraint in their care plan (CQC).

A community nurse greets an older woman. A VR training scenario for healthcare. — A community nurse greets an older woman.

US surveyors cite restraint

Physical restraint citations (tag F604) numbered 446 during 2024. Chemical restraint citations (F605) went from 25 for 2024 to 759 for 2025, after the tag text widened to unnecessary psychotropic medication, which points to a scope change rather than a change in practice (CMS).

The rules

Where does each framework put the training duty?

England puts it in statute, US hospital rules require documented competence, and Australia places it in standards and quality indicators beside binding rules on consent, plans and reporting.

A support worker talks with a young man in a plain room, seen through the headset, a floating panel of replies beside them. — A support worker talks with a young man.

A clinician wearing a headset stands at a hospital bedside facing a simulated patient propped up on pillows, practising the bedside conversation in a ward that looks like the ward. — A clinician at a hospital bedside.

A physiotherapist wearing a headset leans toward a seated older woman in a recliner. — A physiotherapist beside a seated woman.

A VR aged care and disability services training scenario in which a job coach stands beside a worker resting her hands on a crate in a workshop. — A job coach stands beside a worker.

NDIS: training sits in quality indicators

The Rules list the regulated practices (s 6) and require monthly reports (s 14); unauthorised use is a reportable incident (s 16). The Commission’s module lists as a quality indicator: “Workers receive training in the safe use of restrictive practices.”

A case manager leans toward an older man. A VR training scenario for aged care and disability services. — A case manager leans toward an older man.

Aged care: last resort, consent and a date

The Aged Care Act 2024 protects a provider from liability for use before 1 December 2026 where consent was given and the rules were met (s 163). Rule 162-15 requires last resort and that “best practice alternative strategies are used before the restrictive practice is used”.

A receptionist wearing a headset leans over the desk while an older man waits at the counter. — A receptionist greets an older man.

Training sits in Outcome 2.9

The Aged Care Rules 2025 say: “The provider must provide aged care workers with training and supervision to enable them to effectively perform their roles.” Rule 162-30 asks for monitoring for distress, side effects and changes in behaviour while a practice is in use.

A charge nurse in green scrubs briefs nurses at a ward station, a headset visible among them. — A charge nurse briefs the ward.

Hospitals: Actions 5.35 and 5.36

Where restraint or seclusion is clinically necessary, the NSQHS Comprehensive Care Standard asks for systems that minimise and, where possible, eliminate it, govern it under legislation and report it to the governing body.

A man waits at the triage bay counter. A first person VR training scenario for healthcare, seen through the headset. — A man waits at the triage counter.

United States: documented competence

Hospital staff “must be trained and able to demonstrate competency” in restraint and seclusion (482.13). Nursing facilities must use “the least restrictive alternative for the least amount of time” (483.12) and give nurse aides at least 12 hours of training a year (483.95).

A clinician wearing a headset sits knee to knee with a simulated client who is talking and gesturing, in a plain consulting room. — A clinician sits knee to knee with a client.

England: training in statute

The Use of Force Act 2018 requires training on topics from avoiding force to the impact of trauma, refresher training (s 5) and a record of each use (s 6). CQC expects certified training to Restraint Reduction Network standards (statutory guidance), and regulation 13 bars disproportionate restraint in regulated care.

The evidence

Does training on its own reduce restrictive practice?

Not on current evidence: organisational least-restraint programs probably reduce restraint in long-term care (risk ratio 0.86), while the effect of simple education is uncertain (Cochrane).

A wheelchair volunteer talks with a coordinator. A VR training scenario for aged care and disability services. — A volunteer talks with a coordinator.

Programs, not courses

A Cochrane review of 11 studies and 19,003 participants found moderate-certainty evidence that least-restraint policy programs probably reduce residents with at least 1 physical restraint (risk ratio 0.86, confidence interval 0.78 to 0.94; 3,849 participants, 4 studies).

A dietitian leans toward a client. A VR training scenario for healthcare. — A dietitian leans toward a client.

Education alone, and hospitals

The same review is uncertain whether simple educational interventions reduce physical restraint. In general hospitals the evidence on organisational programs is very low certainty (Abraham et al. 2022).

A clinician sits with a young woman outside. A VR training scenario for healthcare. — A clinician sits with a young woman outside.

A VR feasibility trial

A randomised feasibility study on 3 UK psychiatric wards reported a reduction in restrictive practice rates in the VR training group, with no difference in self-efficacy, anxiety or burnout scores (Phiri et al. 2024).

Demand

Is demand for restrictive practices capability rising?

Yes. Australian search interest in “restrictive practices” rose from an index of 34.2 for 2020 to 83.9 for 2025 (Google Trends).

Google Trends interest in Australia, annual mean, 2020 to 2025

Restrictive practicesBehaviour support plan

See the data (relative index, 100 = peak month)

	2020	2021	2022	2023	2024	2025
Restrictive practices	34.2	44.9	52.3	67.4	84.7	83.9
Behaviour support plan	23.6	28.8	33.7	34.2	37.9	52.0

Relative index, annual mean of monthly values, 2020 to 2025. Source: Google Trends, Australia, captured 26 September 2026.

Search interest more than doubled

Interest peaked in May 2024 and held through 2025; “behaviour support plan” rose from 23.6 to 52.0. The index shows direction, not volume.

So what for you: Buyers search the rule, not a course name.

Behaviour support plans lodged with the NDIS Commission, 2019-20 to 2022-23

2019-206,003

2020-2110,109

2021-2213,091

2022-23 (derived)15,768

See the data (plans lodged)

Measure	Plans lodged
2019-20	6,003
2020-21	10,109
2021-22	13,091
2022-23 (derived)	15,768

Plans lodged per financial year. Source: NDIS Commission annual reports, 2019-20 to 2022-23.

Plans lodged grew 2.6 times in 3 years

The 2022-23 figure adds the reported rise of 2,677 to the 2021-22 count. Each plan sets strategies workers carry out.

So what for you: Training has to connect to the plan a worker implements.

A VR aged care and disability services training scenario in which a young support worker stands alone in a staff alcove. — A support worker in a staff alcove.

The workforce these duties reach

Aged care employed 414,000 direct care staff in 2023, 217,000 in residential care (workforce survey), and 2,590 residential services ran on 30 June 2025. Disability had 17,834 registered providers at 30 June 2021 and 2,529 suitable behaviour support practitioners in 2021-22.

An oncology nurse talks with a reclining man. A VR training scenario for healthcare. — A nurse talks with a reclining man.

In a front room at dusk, a woman with a bag over her arm talks to a worker wearing a headset and holding a notebook, an older man watching from his armchair. — A worker with a notebook in a front room.

A midwife leans toward a new mother. A VR training scenario for healthcare. — A midwife leans toward a new mother.

Sources in brief

Which datasets and studies does this rest on?

A row per source, with its design and sample.

Datasets, reports and studies behind the findings
Study	Design	Sample	What was measured	Result
AIHW GEN (2026), QI Program tables	National indicator	Assessed residents	Restrictive practice share	20.7%; trend no change
ACQSC (2025), Sector Performance Report	Regulator notifications	Residential care	SIRS by type	198 against 14,824
NDIS Commission (2021), Annual Report 2020-21	Regulator notifications	NDIS providers	Reportable incidents	1,044,851; 98.7% unauthorised
NDIS Commission (2025), Quarterly Performance Report	Regulator data	NDIS participants	Regulated practices	15,746 (2.1%)
CMS (2026), provider data, State and US averages	National measure	Long-stay residents	Physical restraint	0.33% to 0.13%
CQC (2025), Monitoring the Mental Health Act	Regulator data	Mental health services	Interventions a month	up 24%
CQC (2024), State of Care 2023/24	Provider returns analysis	English care homes	Restraint incidence	nearly 12 times
Möhler et al. (2023), Cochrane Database of Systematic Reviews	Systematic review, RCTs	19,003 participants	Physical restraint	RR 0.86
Abraham et al. (2022), Cochrane Database of Systematic Reviews	Systematic review	General hospitals	Physical restraint	Very low certainty
Phiri et al. (2024), World Journal of Psychiatry	Randomised feasibility study	Psychiatric wards	Restrictive practice rates	Reduction in VR group

Decisions

What does this mean for a buyer?

4 decisions, in the order a program meets them.

Pick the measure your regulator uses before you set a target. QI prevalence (20.7%), SIRS notifications (198 a quarter) and NDIS notifications (1,044,851 a year) count different things, so never benchmark across them.
Combine training with an organisational least-restraint program. The Cochrane risk ratio of 0.86 came from policy programs; the effect of education alone is uncertain.
Rehearse the consent and plan pathway before 1 December 2026. The Aged Care Act 2024 limits liability protection to use before that date, and 73.9% of restricted residents are held by a secure area, which the same consent rules cover.
Keep a competence record where the rule asks for it. US hospitals must document demonstrated competence (482.13), and England’s mental health units must record each use of force as monthly interventions rise 24%.

Board paper lines, ready to paste into a business case

In January to March 2026, 20.7% of assessed Australian aged care residents were recorded under a restrictive practice, and the Department’s trend test on 19 quarters from July 2021 reads no change (National Aged Care Mandatory Quality Indicator Program). A Cochrane review found that organisational least-restraint programs probably reduce the number of long-term care residents physically restrained (RR 0.86, 3,849 participants, 4 studies), while the effect of simple education is uncertain (Möhler and colleagues, 2023). In England, average restrictive interventions a month in mental health services rose 24% between 2023/24 and 2024/25 (Care Quality Commission, 2025).

A carer wearing a headset stands by an open car door as an older man approaches across a car park. — A carer at an open car door.

A therapist wearing a headset walks behind a white cane user along a suburban footpath. — A therapist walks behind a cane user.

A palliative care nurse talks with a visitor. A VR training scenario for healthcare. — A nurse talks with a visitor.

Limits and method

Each dataset counts a different thing, so the figures describe reporting as much as practice.

The QI Program moved on 1 July 2021 from restraint devices per 1,000 care days to a percentage of residents, adding environmental restraint and seclusion; the series cannot be joined.
The indicator was renamed Restrictive practices from April to June 2025; the method text reads the same, not checked against the QI Manual. Chemical restraint is excluded.
Denominators differ: QI assesses residents in a 3-day window, SIRS and NDIS count notifications, the US measure is a share of long-stay residents, England reports a monthly average.
SIRS counts inappropriate use; lawful use is not notified. Sector Performance Reports paused after 1 November 2025, so the series ends at October 2025.
NDIS annual reports do not give incident totals on a common basis each year, and the 2024-25 report could not be read in full.
CMS releases each cover the prior 4 quarters, so years are release years; the date of the F605 tag change was not confirmed.
CQC gives 2 figures for 2023/24: 13,240 a month in its 2024/25 report, over 14,200 in its 2023/24 report.
Google Trends is relative within its query set and country; each annual value averages the monthly values.
Evidence certainty is moderate in long-term care and very low in hospitals; the VR study is a feasibility trial.
Every source was read on 26 September 2026.

Where PTR fits

PTR builds VR and AI roleplay programs where staff rehearse de-escalation and the conversations before a restrictive practice, with a baseline and follow-up measure planned into a pilot. Start with training for aged care educators or de-escalation training, see de-escalation training for healthcare, and plan with how a pilot runs and business case and evaluation.

Related guides

Cite this research

Free to quote and cite with a link to this page. Name the version you read, so readers can see what has changed since.

APA 7

Loza, K. (2026, September 27). 1 in 5 aged care residents is under a restrictive practice: what 5 frameworks ask of training (Version 2.0). People Tech Revolution. https://peopletechrevolution.com/insights/restrictive-practices-training-requirements

Data: AIHW GEN QI Program: January to March 2026 supplementary tables (Table 1.1) and quarterly reports (Table 2); ACQSC Sector Performance Reports (SIRS); NDIS Commission annual reports 2019-20 to 2022-23 and Quarterly Performance Report Q4 2024-25; CMS provider data; CQC; Google Trends; the rules and studies in Sources. Read 26 September 2026.

Version history

Version	Date	What changed
Version 2.0	27 September 2026	Rebuilt on national data: QI Program restrictive practice series, SIRS, NDIS, CMS and CQC data, Google Trends, the duty wording and the Cochrane evidence.
Version 1.0	25 September 2026	First published.