Accessing alcohol and drug treatment services after an assessment

Web report

Last updated: 04 Jun 2026

Topic: Alcohol & other drug treatment services

AIHW (Australian Institute of Health and Welfare) (2026) Accessing alcohol and drug treatment services after an assessment, AIHW, Australian Government, accessed 26 June 2026.

PDF | 2.5MB

Save web report

Scope and methodology

What data is included in this report?

The Alcohol and Other Drug Treatment Services National Minimum Data Set (AODTS NMDS) is a service based (not demand based) data set collected by the Australian Institute of Health and Welfare. The AODTS NMDS contains information about closed treatment episodes provided to clients by specialist alcohol and other drug (AOD) treatment services across Australia who receive funding through the commonwealth Drug and Alcohol Program or are solely funded via a Primary Health Network (including government and non-government services). Data is collected annually, with collection periods beginning 1 July and ending 30 June of the following year (financial years). This report includes data on:

closed treatment episodes provided between 2018–19 and 2023–24
valid Statistical Linkage Key 581 (SLK-581) data from 2018–19 to 2023–24.

For further information see Technical notes: AODTS NMDS coverage and data quality.

AODTS NMDS data can be counted using 2 measures: clients or closed treatment episodes provided to clients.

A typical AOD client can receive multiple treatment episodes. Based on the number and frequency of all episodes received between 2018–19 and 2023–24, clients were assigned to one of 2 cohorts reflecting their treatment pathway.

Data on both client cohorts and treatment episodes are included in this report to capture how clients interact with AOD services over multiple episodes, over the scope of this study. Due to changes in data quality over time, cohort-level and episode-level treatment data varies in coverage.

Table 1 and 2 summarises the differences for treatment cohort-level and treatment episode-level data. For further information see Defining ‘Assessment only’ treatment cohorts and Overview of analysis variables and inclusion criteria.

Table 1: Measures included from the AODTS NMDS data
Measure	Description
Treatment cohort	Includes all closed treatment episodes received by a client: a person who receives an AOD treatment episode for their own drug use. Most clients undergo multiple treatment episodes over the course of their treatment; each episode is matched to a client and counted towards the treatment cohort the client is assigned to.
Treatment episode	The period of contact between a client and treatment provider. Each closed treatment episode contains a single principal drug of concern and one main treatment type.

Table 2: Time coverage of the AODTS NMDS data
Time coverage	Description
Treatment cohort	Clients who received treatment between 2018–19 and 2023–24
Treatment episode	Closed treatment episodes provided between 2018–19 and 2023–24

Longitudinal ‘Assessment only’ client cohort data

This report presents longitudinal data from the AODTS NMDS between 2018–19 and 2023–24 to describe the characteristics of clients who at any point received a main treatment type of ‘Assessment only’, and their subsequent treatment episode. Clients can only receive treatment for one main treatment type per closed treatment episode, but clients may undergo multiple treatment episodes with different main treatment types.

In this report, clients are included if they received at least one closed treatment episode in at least one collection period (financial year) between 2018–19 and 2023–24, where ‘Assessment only’ was the main treatment type. The client enters the study from the first ‘Assessment only’ episode recorded within the study period.

Between, 2018–19 and 2023–24, a total of 157,108 clients received an ‘Assessment only’ episode at any point.

Table 3: Clients who received a main treatment type of 'Assessment only', by state/territory, between 2018–19 and 2023–24
State/territory	Number of clients	Percentage of clients
New South Wales	22,138	14%
Victoria	62,689	40%
Queensland	42,741	27%
Western Australia	3,378	2.2%
South Australia	12,181	7.8%
Tasmania	2,546	1.6%
Australian Capital Territory	3,339	2.1%
Northern Territory	8,096	5.2%
Australia	157,108	100%

Note: A client's state or territory is assigned at the start of their first assessment only treatment episode between 2018–19 and 2023–24. It is based on where the client receives treatment.

Methodology

This report uses data from the Alcohol and Other Drug Treatment Services National Minimum Data Set (AODTS NMDS), focusing on clients accessing AODTS who received an ‘Assessment only’ episode for their own drug use at any point between 1 July 2018 and 30 June 2024.

A client is captured in this analysis from the first ‘Assessment only’ treatment episode they received during the study period. This is not necessarily their first engagement (treatment episode) with specialist AOD services. The 3 analysis approaches used to examine this study population included:

Descriptive analysis: was used to compare the client and treatment episode characteristics of clients who re-engaged within one year after an ‘Assessment only’ episode and those who did not.
Statistical modelling methods: were used to examine the factors associated with receiving further treatment within one year (365 days) after an ‘Assessment only’ episode.
Survival analysis: was used to estimate the time between an ‘Assessment only’ episode and any subsequent treatment episode received at any point between 2018–19 and 2023–24.

Defining ‘Assessment only’ treatment cohorts

‘Assessment only’ study population definitions

Clients who received a treatment episode of ‘Assessment only’ during the 6-year period 2018–19 to 2023–24, were identified as clients who have ‘ever’ received ‘Assessment only’ as a main treatment type. This group of clients were identified as the ‘Assessment only’ study population. For more information see Summary of study.

A 6-year study period of 2018–19 to 2023–24 was chosen to understand recent trends in the longitudinal dataset due to state and territory changes in policies, programs, service delivery and planning.

Clients were excluded from the ‘Assessment only’ study population analysis if they:

Did not receive a treatment episode of ‘Assessment only’ within the study period.
Received treatment for another person’s AOD use.
Received their first recorded ‘Assessment only’ closed treatment episode for the study period between 1 July 2023 and 30 June 2024, since future episodes could not be examined for these clients.

Episodes which occurred before a client’s first ‘Assessment only’ episode or after the client’s subsequent treatment episode were also excluded.

These criteria ensure that the initial cohort received treatment for their own drug use and that there was enough time for clients to have received further treatment within one financial year. However, it is important to note that clients may have received treatment before 1 July 2018, and/or continued to receive treatment beyond 30 June 2024. Services accessed in these periods are outside the scope of this report.

‘Assessment only’ treatment cohort definitions

Two mutually exclusive cohorts were defined based on client’s treatment outcomes following their first ‘Assessment only’ episode within the 6-year defined study period.

To better understand the factors that led to further treatment after an ‘Assessment only’ episode, statistical modelling methods were used. ‘Assessment only’ clients were followed up for one year (365 days) from the time they first received their ‘Assessment only’ episode. The 2 cohorts were defined for the model based on whether they received an outcome of further treatment as followed:

Clients who received further treatment within one year – this includes clients who received any subsequent treatment episode within one year after receiving their first ‘Assessment only’ episode.
Clients who did not receive further treatment within one year – this includes clients who received a subsequent treatment episode, but this occurred more than a year after receiving their first ‘Assessment only’ episode, and those who did not receive further treatment within the study period

The criterion above was used to define the treatment cohorts for comparison in Table 4: Definition of AODTS NMDS treatment cohorts.

Table 4: Definition of AODTS NMDS treatment cohorts
Treatment cohort	Study entry	Study exit
Clients who received further treatment within one year	First treatment episode in the period is ‘Assessment only’.	At least one additional treatment episode recorded after the initial assessment within one year (365 days).
Clients who did not receive further treatment within one year	First treatment episode in the period is ‘Assessment only’.	No further treatment episodes recorded after this initial assessment within one year (365 days).

Counting methodology

Treatment episodes were ordered chronologically by episode start date. Each client was included in only one cohort, based on their first ‘Assessment only’ episode and subsequent treatment history. Only treatment episodes recorded within the 6-year window (1 July 2018 to 30 June 2024) were considered. Episodes outside this range were excluded from study population classification. For more information see Figure 7: Examples of client pathways through specialist AOD treatment, 2018–19 to 2023–24.

Summary of study

Figure 6: Analysis criteria of clients who received a main treatment type of ‘Assessment only’, between 2018–19 and 2023–24

See extended description following this image.

For more information, see Overview of analysis variables and inclusion criteria.

Extended description for Figure 6 Expand

The tree diagram illustrates a hierarchical structure of the process of selecting the cohorts presented in this analysis. It starts with a total population at the top and branches into different groups based on assessment and treatment criteria. The chart uses color-coded boxes to represent different stages and analyses.

There are 4 colour schemes to highlight the different numbers displayed:

White box with grey border: AODTS NMDS episode data
Green boxes: Assessment only (AO) population
Teal boxes: Survival analysis
Blue boxes: Statistical modelling methods
Grey boxes: Episodes excluded from the analysis

At the top of the diagram:

White box with grey border: 1,282,499 episodes were provided to 493,310 clients with a valid Statistical Linkage Key (SLK) for their own drug use between 2018–19 and 2023–24.
Grey box 1: 595,907 (46%) of episodes were excluded where 317,373 (64%) clients did not receive an ‘Assessment only’ episode between 2018–19 and 2023–24 (an assessment may have been part of client’s treatment outside the defined period).

This then flows down to the next level as follows:

Green box 1: 178,937 (36%) clients received ‘Assessment only’ ever over 686,542 (54%) of episodes provided.
Grey box 2: 433,888 episodes were excluded because clients the episodes occurred before a client’s first ‘Assessment only’ episode or after the client’s subsequent episode or they received their first ‘Assessment only’ episode in 2023–24.

This then flows down to the next level as follows:

Green box 2: 252,654 episodes were provided to 157,108 clients for their first ‘Assessment only’ episode. This group forms the ‘Assessment only’ study population.

The green box then splits into two streams:

Left stream 1 – Clients who received further treatment at any point between 2018–19 and 2023–24 (Dark teal).
Right stream 1 – Clients who did not receive further treatment at any point between 2018–19 and 2023–24 (Teal).

For clients who did not receive further treatment at any point between 2018–19 and 2023–24 (Right stream 1):

Teal box 1 and 2: 61,562 episodes (24%) were provided to 61,562 (39%) clients did not receive further treatment at any point between 2018–19 and 2023–24. These values were censored from the survival analysis.

For clients who did receive further treatment at any point between 2018–19 and 2023–24 (Left stream 1):

Dark teal box 1 and 2: the 191,092 episodes (76%) were provided for 95,546 (61%) clients who received further treatment at any point between 2018–19 and 2023–24.

Left stream 1 is then split further into two groups:

Left stream 2 – ‘Assessment only’ clients who received further treatment within one year (365 days).
Right stream 2 – ‘Assessment only' clients who received further treatment after one year (365 days).

For ‘Assessment only’ clients who received further treatment within one year (365 days) (Left stream 2):

Dark teal box 3 and 4: 160,620 episodes (84%) were provided to 80,310 clients (84%) who received further treatment within one year (365 days). These values flow into:
1. Dark blue box 1 and 2 160,620 episodes (51%) were provided to 80,310 clients (64%) who received further treatment within one year whose data were used in for statistical modelling methods.

For ‘Assessment only' clients who received further treatment after one year (365 days) (Right stream 2):

Light teal box 1 and 2: 30,472 (16%) episodes were provided to 15,236 (16%) clients who received further treatment after 365 days whose data were used in the survival analysis.
Grey box 3: 15,236 subsequent episodes (6%) provided to clients after one year were excluded for statistical modelling methods.

These values from right stream 2 and values from Teal box 1 and 2 flow into:

Light blue box 1 and 2: consists of 15,236 clients (10%) and episodes (6%) who received treatment after 365 days (right stream 2, light teal box 1 and 2) and 61,562 (39%) clients and episodes (24%) who did not receive a subsequent treatment at any point between 2018–19 and 2023–24 (right stream 1, teal box 1 and 2) for a total of 76,798 clients (49%) and episodes (3%) who did not receive further treatment within 365 days. These data were used in the statistical modelling methods.

Analysis technical information

To better understand the factors that led to clients receiving further treatment within one year (365 days), 2 treatment cohorts were set up. The follow up period of one year was chosen because:

provides a reasonable window for clients to receive further treatment
and is informed by the survival analysis presented in ‘Time to further treatment following an ‘Assessment only’ episode’. It was found, 84% of clients who received further treatment did so within one year (365 days) This was used as a threshold to define the treatment cohorts.

For more information on the treatment cohort criteria, see Defining ‘Assessment only’ treatment cohorts.

Descriptive analysis was conducted to examine the client and treatment episode characteristics between clients who received further treatment within one year (365 days) after receiving an ‘Assessment only’ episode and those who did not.

Statistical modelling methods were used to examine the factors associated with the likelihood of receiving further treatment within one year (365 days) after an ‘Assessment only’ episode. Separate multivariate models were fitted for each state/territory controlling for all factors included in the model. A national model is not presented as it was not statistically valid due to variations in service structures, and state and territory programs. Missing values were excluded from the model and due to the differences of the state and territory, different groupings of variables were used for each stratified model. See below for more information on the statistical modelling methods.

Statistical modelling methods Expand

To examine factors associated with receiving further treatment within one year (365 days) following an ‘Assessment only’ episode, the modified Poisson regression modelling method was used.

This approach estimates probability of the outcome and provides relative risk (risk ratios) rather than odds ratios, offering a more interpretable measure of association. See table 5 for the overview of variables and analysis criteria included and excluded for each stratified state and territory model.

Table 5: Overview of variables and analysis criteria included and excluded for each stratified state and territory model
State/territory stratified models	Variable	Analysis criteria
For all stratified models	Sex	‘Another term’ and ‘Not stated’ was coded as missing due to small numbers.
For all stratified models	Age	Clients aged less than 10 years and aged over 100 excluded.
For all stratified models	Indigenous status	Variable excluded.
For all stratified models	Preferred language	Data item was grouped as ‘English’ and ‘All other preferred languages except for English’.
For all stratified models	Country of birth	Data item was grouped as ‘Australia and ‘Overseas born’.
For all stratified models	Source of referral	‘Other’ was coded as missing.
For all stratified models	Principal drug of concern	‘Alcohol’, ‘Cannabis’, ‘Methamphetamine’, Heroin’, and ‘Pharmaceuticals’ were used in the model and ‘All other drugs’ was grouped together
For all stratified models	Delivery setting	‘Other’ was coded as missing. Exclusions were applied depending on the state or territory due to small numbers.
For all stratified models	Reason for cessation	‘Other’ was coded as missing.
For all stratified models	Previous treatment	No exclusions.
For all stratified models	Remoteness area	Exclusions were applied depending on the state or territory.
Western Australia	Delivery setting	Variable excluded.
South Australia	Delivery setting	Variable excluded,
Tasmania	Remoteness area	Tasmania only has a remoteness area of Inner regional and Outer regional.
Australian Capital Territory	Remoteness area	Variable excluded, the Australian Capital Territory only has one remoteness area.
Northern Territory	Remoteness area	Northern Territory only has a remoteness area of Outer regional, Remote, and very remote.

Survival analysis Expand

Survival analysis examines time-to-event data, considering both if and when an event occurs. It accounts for incomplete observations (censoring) making it suitable for this analysis as not all clients have the same amount of follow up time.

The Kaplan-Meier method estimates the probabilities of receiving further treatment at different time points and presents them as a survival curve, enabling comparison of time to further treatment after an ‘Assessment only’ episode over a defined follow up period.

For more information on the methodology see Figure 7 on how survival analysis censoring works.

Accessing alcohol and drug treatment services after an assessment

Citation

Scope and methodology

On this page In this section

What data is included in this report?

Longitudinal ‘Assessment only’ client cohort data

Methodology

Defining ‘Assessment only’ treatment cohorts

‘Assessment only’ study population definitions

‘Assessment only’ treatment cohort definitions

Counting methodology

Summary of study

Figure 6: Analysis criteria of clients who received a main treatment type of ‘Assessment only’, between 2018–19 and 2023–24

Analysis technical information

Figure 7: Examples of client pathways through specialist AOD treatment, by analysis techniques, 2018–19 to 2023–24

Connect with us

Search

Other ways to browse

Feedback