Maree Teesson 1, Christina Marel 1, Katherine Mills 1, Shane Darke 2, Joanne Ross 1, Tim Slade 1, Jo White 1, Sonja Memedovic 1, Michael Lynskey 2 and Lucinda Burns 3 1 NHMRC Centre of Research Excellence in Mental Health and Substance Use (CREMS), National Drug and Alcohol Research Centre, UNSW 2 National Addiction Centre, Kings College, London 3 National Drug and Alcohol Research Centre, UNSW
Funders and Collaborators All participants who give their time and experiences. AOD treatment agencies National Health and Medical Research Council (NHMRC) Australian Department of Health National Drug and Alcohol Research Centre NSW Ministry of Health NSW Pharmacy Guild NSW Users and AIDS Association Hepatitis NSW Australia and NSW Needle and Syringe Programs Interstate collaborators: Prof. Alison Ritter, Prof. Robert Ali. ATOS Researchers Sarah Ellis Philippa Ewer Sandra Fairbairn Dr Alys Havard Nicky Henderson Kate Hetherington Dr Christina Marel Sonja Memedovic Dr Louise Mewton A/Prof Katherine Mills Joanne White Evelyn Wilhelm Dr Anna Williamson
Shane Darke Jo Ross Maree Teesson Alys Havard Kath Mills Kate Hetherington & Anna Williamson
A prospective, longitudinal cohort study One of the few studies focusing on heroin dependence N=615 8 not in treatment 535 entering treatment 21 entering MT 21 entering detox 135 entering resi rehab
In 211-12, more people sought treatment for heroin use in Australia than any other illicit drug (n=65,952) (AIHW 212). Heroin dependence How persistent? How chronic? Relapsing condition? When?
Baseline (N=615) % Male 66.2 Mean age 29 % Heroin dependent 97.6 Mean length of heroin use career (years) 9.6 Mean number of drug classes used 4.9 % In current treatment for opiate dependence 87. % Major depression 24.6 % Current PTSD 29.4
Percent 1 8 6 4 2 1.3.8 (n=615) 89 (n=549) 81 (n=495) 1.5 76 (n=469) 2.1 7 (n=429) 6 5 4 3 2 1 Number Baseline 3-mths 1-yr 2-yr 3-yr 11-yr Wave % Interviewed % Deceased N Interviewed
Percent 1 8 6 4 2 1.3.8 (n=615) 89 (n=549) 81 (n=495) 1.5 76 (n=469) 2.1 7 (n=429) N=63 (1.2) 7 (n=431) 6 5 4 3 2 1 Number Baseline 3-mths 1-yr 2-yr 3-yr 11-yr Wave % Interviewed % Deceased N Interviewed
Jo White & Sonja Memedovic & Philipa Ewer Chris Marel
Baseline interviews 21-22 11-year interviews 211-213
We don t know what happens over the longterm for people with heroin dependence Are there indicators that can tell us what is likely to happen over the long-term for their heroin use?
Baseline (N=615) 11-years (N=431) % Male 66.2 64.5 Mean age 29 4 % Crime main source of income 23.9 2.1 % Used heroin in past month 98.7 24.8 % Heroin dependent 97.6 15.1 Mean number of drug classes used 4.9 3.2 % In current treatment for opiate dependence 87. 46.6 % Major depression 24.6 2.9
Percentage 1 8 6 4 2 99 96 25 85 68 42 53 56 48 32 4 4 3 29 18 27 9 2 27 Baseline 11-years
Two stage process: Modelled trajectories of heroin use, and calculated the probability of belonging to a specific group Entering variables into the model to predict those pathways *Finite mixture modelling: specifically group based mixture modelling approach (Jones, et al 21; Nagin 25)
We found 6 pathways of heroin use over 11- years
Trajectory % 1. Early decrease to maintained abstinence 16.2 2. Early decrease with early relapse 7.1 3. Early decrease with late relapse 14.4 4. Gradual decrease to near abstinence 19.1 5. Gradual decrease 21.5 6. No decrease 21.7
1.8 16.2% Probability.6.4.2 1 12 2 3 3 4 4 5 5 6 6 7 78 8 9 1 9 1 11 Early decrease to maintained abstinence (mean) Year Early decrease to maintained abstinence (predicted)
1.8 7.1% Probability.6.4.2 1 12 23 34 45 56 67 78 89 1 9 111 Early decrease with early relapse (mean) Early decrease with early relapse (predicted) Year
1.8 14.4% Probability.6.4.2 1 12 23 34 45 56 67 78 89 1 9 111 Early decrease with late relapse (mean) Year Early decrease with late relapse (predicted)
1.8 19.1% Probability.6.4.2 1 12 23 34 45 56 67 78 89 1 9 111 Year Gradual decrease to near abstinence (mean) Gradual decrease to near abstinence (predicted)
1 Probability.8.6.4.2 21.5% 1 12 23 34 45 56 67 78 89 1 9 111 Year Gradual decrease (mean) Gradual decrease (predicted)
1 Probability.8.6.4.2 21.7% 1 12 23 34 45 56 67 78 89 1 9 111 Year No decrease (mean) No decrease (predicted)
1.8 Probability.6.4.2 1 2 3 4 5 6 7 8 9 1 11 1 2 3 4 5 6 7 8 9 1 Early decrease to maintained abstinence (mean) Early decrease with early relapse (mean) Early decrease with late relapse (mean) Gradual decrease to near abstinence (mean) Gradual decrease (mean) No decrease (mean) Year
1.8 Probability.6.4.2 1 2 3 4 5 6 7 8 9 1 11 1 2 3 4 5 6 7 8 9 1 Early decrease to maintained abstinence (mean) Early decrease with early relapse (mean) Early decrease with late relapse (mean) Gradual decrease to near abstinence (mean) Gradual decrease (mean) No decrease (mean) Year
Aim: to examine factors associated with each trajectory A series of multinomial logistic regressions were conducted on covariates hypothesised to be associated with heroin use trajectories
Demographic characteristics Sex, education Main source of income Homelessness Prison history Past month criminal involvement at baseline Physical and mental health General and physical and mental health ASPD BPD Current and lifetime PTSD Major depression Ever attempted suicide No. traumas Drug use history No. drug classes used last month Types of drugs used last month History of IDU Treatment history In current treatment Current treatment category Index treatment category No. previous rx episodes
Aged more than 3 Been using heroin for less than 1 years Ever overdosed Severe mental health disability Had experienced rape
Significantly more likely to be in the early decrease to maintained abstinence group. 1.8 Probability.6.4.2 1 2 3 4 5 6 7 8 9 1 11 1 2 3 4 5 6 7 8 9 1 Year
Significantly more likely to be in the early decrease with early relapse group. 1.8 Probability.6.4.2 1 2 3 4 5 6 7 8 9 1 11 1 2 3 4 5 6 7 8 9 1 Year
Significantly more likely to be in the gradual decrease to near abstinence group. 1.8 Probability.6.4.2 1 2 3 4 5 6 7 8 9 1 11 1 2 3 4 5 6 7 8 9 1 Year
Significantly more likely to be in the early decrease with early relapse group. 1.8 Probability.6.4.2 1 2 3 4 5 6 7 8 9 1 11 1 2 3 4 5 6 7 8 9 1 Year
Significantly more likely to be in the early decrease with early relapse group, early decrease with late relapse group, or gradual decrease to near abstinence group. 1.8 Probability.6.4.2 1 2 3 4 5 6 7 8 9 1 11 1 2 3 4 5 6 7 8 9 1 Year
1 Probability.8.6.4.2 21.7% 1 12 23 34 45 56 67 78 89 1 9 111 Year No decrease (mean) No decrease (predicted)
Heroin dependence is often a chronic, recurring condition, associated with complex trajectories. More than 8% of the cohort achieved some level of reduction in their heroin use. It is difficult to predict who will follow a chronic pathway The critical factors guiding trajectories are: Aged more than 3 Used heroin for less than 1 years History of overdose Ever been raped Severe mental health disability
Examining how treatment across the 11-year follow-up impacts upon a person s trajectory Continued follow-up of the ATOS cohort to examine the pathways of the group as they continue to age 1.8.6.4.2 1 2 3 4 5 6 7 8 9 1 11 12 13 14 15 1 2 3 4 5 6 7 8 9 1 11 12 13 14??????
m.teesson@unsw.edu.au Mental Health and Substance Use Thank you!