Skip to main content
Register

Dashboards

Areas of work for Seclusion

Note: Northland has low or no data since April 2025. Tairāwhiti have lower inpatient volumes from March 2025. West Coast have lower inpatient volumes from July 2025.

Seclusion dashboards

The number of seclusion events, total duration of seclusion per 1,000 bed nights and seclusion events per 100,000 of population. All dashboards can be filtered by financial year quarter and demographic information.

Note: you must be registered and sign in to use the KPI Programme data dashboards

New to the KPI Programme? Get tips on how to use the dashboards.

Why seclusion matters

Growing evidence over the past decade supports the need to reduce the use of seclusion and other restrictive practices. Seclusion has adverse physical and psychological impacts on both the people and staff involved (Askew, Fisher, & Beazley, 2019; Cusack et al., 2018; Hawsawi et al., 2020; Mellow, Tickle, & Rennoldson, 2017). The use of seclusion contradicts contemporary recovery-focused and trauma informed approaches and constitutes a human rights violation (Chieze et al., 2019; Mellow, Tickle, & Rennoldson, 2017; World Health Organization, 2019).

Read our seclusion evidence review.

Technical Details

Data source

The data in this dashboard is sourced from PRIMHD and StatsNZ population

Technical notes and shared definitions

seclusion activity has:

  • ActivityTypeCode = T33
  • ActivityUnitType = “SECLUSION”

and is recorded against a referral with:

  • TeamSetting = “Inpatient based”
  • TeamType of
    • Inpatient
    • Forensic
    • Intellectual Disability Dual Diagnosis
    • Speciality

bednight activity (for the purposes of the seclusion KPI) has:

  • ActivityTypeCode not TCR
  • ActivityUnitType = “BEDNIGHT”
  • ActivityUnitCount > 0
  • And where there does not exist a LEAVE record for the same date
  • Same team restriction as above
  • Note that there are no checks on ActivityTypeCode against the TeamType

bednight is:

  • A single midnight (00:00) that is crossed by a bednight activity
    For example, a bednight activity with start date 2020-01-20 17:30 and end date 2020-01-21 09:00 would have one bednight; a bednight with start date 2020-01-20 00:00 and end date 2020-01-21 09:30 would also have one bednight.
  • Where bednight activities have been recorded with both start and end times of 00:00, we count only one of those.
    For example, a bednight with start date 2020-01-20 00:00 and end date 2020-01-21 00:00 counts as only a single bednight. This is in line with the ActivityUnitCount calculated by MOH.

Total # bednights (sometimes seen as beddays) is the distinct count of bednights between the reporting start and end date. Where a single bednight activity crosses multiple reporting periods, only the individual bednights within the reporting period are counted.
For example, a bednight activity with start date of 2019-12-20 18:00 and end date of 2020-01-13 09:00 would have 11 bednights counted in the Oct-Dec19 quarter and 13 bednights counted in the Jan-Mar20 quarter.

Population = sourced from StatsNZ, DHB projection for the reporting period

Seclusion activities are sometimes recorded differently due to varying business processes. To accommodate this, we create seclusion events or seclusion episodes in line with Te Whatu Ora recommendations.

Where there are fewer than 60 minutes between seclusion activities (for the same person and the same referral), these activities are rolled up into a single seclusion event.

seclusion event combines overlapping or adjacent seclusion activities for an individual tangata whai ora within a single referral.

Indicator definitions

Total # seclusion events = distinct count of seclusion events where the seclusion event start date falls within the reporting period.

# tāngata whai ora secluded = distinct count of tāngata whai ora with a seclusion event where the seclusion event start date falls within the reporting period.

Seclusion events per 1000 bednights = Total # seclusion events / (Total # bednights / 1,000)

Seclusion events per 100k population = Total # seclusion events / (Population / 100,000)

Seclusion tāngata whai ora per 100k population = # tāngata whai ora secluded / (Population / 100,000)

Hours seclusion = sum of the hours of T33 seclusion activity that occur within the reporting period. When a single seclusion event crosses multiple reporting periods, only the hours within the reporting period are counted.
For example, a seclusion event with start date of 2019-12-28 18:00 and end date of 2020-01-02 09:00 would have 78 hours counted in the Oct-Dec19 quarter and 33 hours counted in the Jan-Mar20 quarter.

Additional notes

Seclusion events are allocated to a shift based on the seclusion event start time:

  • 00:00 – 06:59 night
  • 07:00 – 14:59 morning
  • 15:00 – 23:59 afternoon

Likewise, seclusion events are allocated to a weekday based on the seclusion event start date.

Note the fundamental distinction between seclusion events, which are allocated to a reporting period based on their start date, and seclusion hours and bednights, which are broken apart when they cross reporting boundaries and then apportioned pro rata. There are variations in these definition across the sector, so if you observe seclusion numbers that are close but not identical, this date logic may be a good place to check first.

A note about forensic data

Forensic seclusion events are summarised in a separate report. All forensic seclusion events and bednights are excluded from the national and Te Whatu Ora division overviews.

MH&A KPI Programme is part of the Wise Group Copyright ©2026

Sign in