The Horizons Project: An empirical analysis of alcohol and drug treatment funding, purchasing and workforce mechanisms

image - Three Glasses Original
Date Commenced:
April 2017
Expected Date of Completion:
March 2020
Project Supporters:

NHMRC Project Grant (GNT1128100)

Drug Type:
Project Members: 
image - 1313976712 Alison Ritter 005
Director, Drug Policy Modelling Program
Ph 02 9385 0236
image - Timothy Dobbins Square
Associate Professor
Ph +61 (2) 9385 0333
image - 1314146187 Chalmers Jenny 06
Dr Jenny Chalmers
Conjoint Senior Lecturer
Project Main Description: 

Alcohol and other drug (AOD) treatment is key for improving health and reducing the social impact of AOD use. However, the treatment itself is not the only variable that impacts on whether health outcomes are improved. The way in which treatment is funded, purchased, and staffed is likewise important. While these variables are important determinants of treatment outcomes, no Australian research to date has examined how the funding, purchasing, and staffing of treatment impacts on client treatment outcomes. The key structural features associated with treatment outcomes are:

  1. Funding arrangements
  2. Purchasing mechanisms
  3. Provider type
  4. Workforce characteristics

These structural features are the focus for this project, which has been funded by the NHMRC (2017 to 2019). The project will provide an empirical foundation to guide decisions about how to fund and purchase AOD treatment services. 

Project Collaborators: External: 

Dr Michael Livingston
La Trobe University

Dr Lynda Berends
Trace Research

Professor Harvey Whiteford
The University of Queensland

Mr Sam Biondo
CEO, Victorian Alcohol And other Drug Association (VAADA)

Ms Helene Delany
Manager, Alcohol and other Drug Policy, ACT Health

Associate Professor Adrian Dunlop
Addiction Medicine Specialist, NSW Ministry of Health

Dr Moira Hewitt
Head, Tobacco Alcohol and Other Drugs Unit. Australian Institute of Health and Welfare

Ms Rebecca Lang
CEO, Queensland Network of Alcohol and Drug Agencies (QNADA)

Professor Ann Roche
Director, National Centre Education and Training on Addictions (NCETA)

Dr Kerstin Stenius
Department of Mental Health & Substance Abuse Services, National Inst for Health & Welfare, Finland

Associate Professor Jessica Storbjork
Stockholm University, Sweden


This project has the potential to be of direct benefit to policymakers, AOD treatment providers and AOD researchers in a number of ways. Having accurate information on the funding arrangements, purchasing mechanisms and workforce characteristics of services will help, amongst other things, to:

  • Better plan the ways in which treatment services are funded;
  • Understand the implications of different types of purchasing models;
  • Assist with the allocation of resources to maximise health and wellbeing outcomes;
  • Provide data for new research questions of interest to funders and treatment providers about the AOD workforce characteristics; and
  • Gain insight in funding and workforce characteristics in relation to treatment outcomes to enhance the capacity of the AOD sector.

This study has two aims:

Aim 1. to establish a database of Australian alcohol and other drug (AOD) treatment service system characteristics which includes detailed data on funding, purchasing, workforce and organisation/provider characteristics, alongside episode, person and state characteristics;

Aim 2. Using multi-level modelling, to test hypotheses regarding relationships between service system characteristics and treatment outcomes (as measured by retention and reason for cessation). There are any number of things that could be tested including open competitive tendering; block funding; workforce characteristics; and provider type.

Design and Method: 

This project is premised on the establishment of a comprehensive, Australian AOD treatment service system database that contains information about treatment episodes, agencies and funding. The data needed for this project includes:

  • De-identified data on treatment episodes and people receiving treatment (through the AIHW AODTS-NMDS);
  • Funding data – the sources and amounts of funding to each AOD treatment agency;
  • Purchasing data – the ways in which each agency is contracted to provide the AOD services (e.g. through competitive tendering, block grants and so on); and
  • The AOD workforce – the key characteristics of the AOD workforce in each agency.

There are two phases to the project:

Phase 1: Establish the database (2017-2018), consistent with Aim 1

The AOD treatment service system database will comprise the following data:

  • Episode data – regarding an individual episode of care received (AODTS-NMDS)
  • Person data – regarding the demographic characteristics of the individual receiving treatment (AODTS-NMDS)
  • Organisation/provider data – the key features of the organisation which provides the treatment: purchasing, workforce and funding
  • State data: including treatment utilisation rate and prevalence rates of substance misuse.

Phase 2: Conduct analyses to test relationships between key treatment service system characteristics and treatment outcomes (2019), consistent with Aim 2

The two treatment outcomes are: 1) retention (time-in-treatment) and 2) reason for treatment cessation. Multiple studies have shown significant associations between both of these outcomes and post-treatment substance use for both alcohol dependence treatment and drug (largely opioid) treatment.

We will use generalised linear mixed models to fit multi-level models accounting for the hierarchical nature of episodes (level 1), persons (level 2) and organisations (level 3), with state as a fixed effect.


Consistent with an evidence-based policy paradigm, this project seeks to generate evidence to enable better decisions by policy makers regarding funding and purchasing AOD treatment. While there is a wealth of research evidence about treatment effectiveness, there is no Australian research to inform considerations of structural features of the system.

Project Research Area: 
Project Status: