Research paper
Opiate substitution treatment to reduce in-prison drug injection: A natural experiment

https://doi.org/10.1016/j.drugpo.2013.01.004Get rights and content

Abstract

Background

There is emerging evidence that opiate substitution treatment (OST) in prison is associated with reduced injecting drug use (IDU). In Australia OST is available in prison for men and women in all jurisdictions except Queensland, where it is available only for women. The aim of this study was to examine the association between in-prison OST and in-prison IDU in New South Wales (NSW) and Queensland.

Methods

Cross-sectional survey of Australian prisoners in NSW (N = 1128) and Queensland (N = 1325). Lifetime IDU and in-prison IDU measured by self-report.

Results

Lifetime history of IDU was significantly more common among females than males in both jurisdictions. Among those with a lifetime history of IDU, the lifetime prevalence of in-prison IDU was significantly higher for males than females in both jurisdictions, however the magnitude of this sex difference was considerably greater in Queensland than in NSW.

Conclusion

Provision of OST in prison is associated with a reduced lifetime prevalence of in-prison drug injection, among those with a lifetime history of IDU. Irrespective of OST policies, women with a history of IDU are less likely than their male counterparts to inject in prison; reasons for this novel and replicable sex difference require further examination.

Introduction

A history of injecting drug use (IDU) is common among prisoners in many countries around the world including Australia (Butler et al., 2003, Carpentier et al., 2012), where the most recent national census estimated that 55% of prison receptions had a lifetime history of IDU (AIHW, 2011). Despite concerted efforts to restrict the flow of drugs into prisons, a substantial proportion of those with a history of IDU continue to inject in prison (Carpentier et al., 2012, Kinner et al., 2012). Although intended primarily as a treatment for opiate dependence, there is emerging evidence that opiate substitution treatment (OST) in prison is associated with a range of positive outcomes including reduced IDU and syringe sharing in prison (Larney, 2010, Stallwitz and Stöver, 2007), although at least one study has found a positive association between in-prison OST and in-prison IDU (Teutsch et al., 2010).

In Australia, correctional services are administered at the state level and policies with regard to in-prison OST vary between jurisdictions. OST is currently available to eligible male and female adult prisoners in all Australian jurisdictions except Queensland, where treatment is (with very rare exceptions) not available for men but has been available for women who enter custody pregnant since 2001, and more recently for women who enter custody on an OST program and are expected to spend less than 12 months in custody (Rodas, Bode, & Dolan, 2012). OST coverage in Queensland remains poor with 37 of around 480 female prisoners and 2 of more than 5100 male prisoners receiving OST as at 30th November 2012 (M. Wilson, personal communication 30/11/12).

A recent cross-sectional study of more than 1300 adult prisoners in Queensland found that 56% had a lifetime history of IDU and 22% had a lifetime history of in-prison IDU (Kinner et al., 2012). Although there was some evidence that OST was associated with a reduced likelihood of recent in-prison IDU, this difference was not statistically significant, possibly because only a subset of females in the study (and therefore only a small minority of the sample) were able to access OST in prison. Consistent with this, although females were more likely than males in the study to have a lifetime history of IDU, they were less likely to have injected in prison. Although these findings are consistent with the view that OST reduces IDU in prison, OST availability in the Queensland study was confounded with sex and an equally plausible interpretation of the findings is that women with a history of IDU are less likely than their male counterparts to inject in prison, for reasons unrelated to in-prison OST policies.

Whereas OST availability is confounded with sex in Queensland, OST has been available for eligible men and women in NSW prisons since 1987. Coverage is good with an average of 16% of the population (1411 males and 228 females) receiving methadone or (more rarely) buprenorphine in 2009 (S. Sunjic, personal communication, 11/12/12). With an estimated 20% of the NSW prisoner population using heroin or other opiates on a daily or almost daily basis in the year before prison (Indig et al., 2010), coverage appears to be good.

This striking jurisdictional policy difference provides a unique opportunity for a natural experiment: If OST in prison reduces IDU in prison, one might expect similar proportions of male and female IDU to report in-prison IDU in NSW, but a larger proportion of males than females to report in-prison IDU in Queensland (as has been observed). Conversely, if OST in prison does not reduce IDU in prison, one would expect any sex difference in the proportion of IDU reporting in-prison IDU to be the same in NSW and Queensland.

We hypothesised that among those with a lifetime history of IDU (1) the lifetime prevalence of in-prison IDU would be higher among prisoners in Queensland than in NSW and (2) women would be less likely than men to report a lifetime history of in-prison IDU in Queensland, but this sex difference would not be observed in NSW.

Section snippets

Methods

Data for the present study come from the 2009 NSW Inmate Health Survey (IHS) and the baseline interview from the Passports study in Queensland. The NSW IHS included a sample of 1128 prisoners stratified by age (<25, 25–44, 45+ years), sex (women were over-sampled to 20%) and Aboriginality and was generally representative of the NSW prisoner population (Indig et al., 2010). The survey was conducted using Computer Assisted Telephone Interviewing across all prisons in NSW and had an 85% response

Hypothesis 1: jurisdictional difference

In Queensland, of 1322 participants with valid responses on all variables of interest 737 (56%) reported a lifetime history of IDU and among these, 286 (39%) reported having ever injected in prison. In NSW, of 1106 participants with valid responses on all variables of interest, 504 (46%) reported a lifetime history of IDU and among these, 211 (42%) reported having ever injected in prison (Table 1). Among those with a lifetime history of IDU, there was no significant difference in the lifetime

Discussion

We hypothesised that if in-prison OST was associated with reduced in-prison drug injection, then the lifetime prevalence of in-prison IDU would be higher in Queensland (where OST coverage in prison is poor for women and near zero for men) than in NSW (where OST has been available in prison for women and men since 1987). This hypothesis was not supported, with around two-fifths of those with a lifetime history of IDU reporting in-prison IDU in both jurisdictions.

Noting that male prisoners are

Conclusions

Although by no means definitive, our findings are consistent with those of previous studies (Larney, 2010) and provide further evidence that the provision of OST in closed settings can reduce drug injection among those with a history of IDU. The factors influencing injecting practices in closed settings are complex, and may vary in important ways between men and women. Although not a panacea, appropriately targeted OST, in combination with evidence-based infection control measures such as NSP (

Acknowledgements

In Queensland, the authors wish to acknowledge the Passports participants for sharing their stories, Queensland Corrective Services for supporting the project and facilitating data collection, and the Passports project team. Thanks also to Michael Wilson for providing OST coverage data. The Passports study was funded by NHMRC grant #409966. Stuart Kinner is supported by NHMRC grant #1004765.

In NSW, we wish to acknowledge the funding agencies including NSW Department of Health (Mental Health and

References (15)

There are more references available in the full text version of this article.

Cited by (9)

View all citing articles on Scopus
View full text