Kerryn Butler

The hidden costs of drug and alcohol use in hospital emergency departments

INTRODUCTION AND AIMS This study estimates the burden of drug and alcohol morbidity on hospitals in New South Wales (NSW) by observing a multi-site collective sample utilising survey information and data linkage. Specifically we aimed to determine the prevalence of alcohol and other drug (AOD) problems and to estimate patterns of utilisation of hospital services, costs of presentations, and admissions for patients with AOD problems. DESIGN AND METHODS Patients were recruited from eight NSW public hospitals presenting to the hospital emergency department over a 10 day period. Participants completed a self-administered survey with demographic characteristics and questions about substance use. More than two-thirds (68%) of participants consented to provide access to their NSW Health medical data for a period spanning 2.5 years. RESULTS One-third (35%) of the total sample were identified as having problematic AOD use with one in five of these patients requiring a high level of intervention. Those patients requiring a high level of intervention present more often and cost more per presentation. If admitted they were more likely to have longer stays and were also more likely to be admitted to a psychiatric ward and have a longer stay in the ward. DISCUSSION This study demonstrates a need for AOD interventions in the emergency department setting, both because it represents an opportunity for intervention in a population in which problems with substance use is highly prevalent, and because there is evidence that AOD imposes additional costs on the health system. [Butler K, Reeve R, Arora S, Viney R, Goodall S, van Gool K, Burns L. The hidden costs of drug and alcohol use in hospital emergency departments. Drug Alcohol Rev 2016;35:359-366].

Driving under the influence among frequent ecstasy consumers in Australia: Trends over time and the role of risk perceptions

BACKGROUND Driving under the influence (DUI) of alcohol and illicit drugs is a serious road safety concern. This research aimed to examine trends in DUI across time and changes in attitudes towards the risks (crash and legal) associated with DUI among regular ecstasy users (REU) interviewed in Australia. METHODS Participants were regular (at least monthly) ecstasy users surveyed in 2007 (n=573) or 2011 (n=429) who had driven a car in the last six months. Face to face interviews comprised questions about recent engagement of DUI and roadside breath (alcohol) and saliva (drug) testing. Participants also reported the risk of crash and of being apprehended by police if DUI of alcohol, cannabis, ecstasy, and methamphetamine. RESULTS There were significant reductions in DUI of psychostimulants (ecstasy, methamphetamine, cocaine, LSD) but not alcohol or cannabis between 2007 and 2011. This was accompanied by increased experience of roadside saliva testing and increases in crash and legal risk perceptions for ecstasy and methamphetamine, but not alcohol or cannabis. When the relationship between DUI and risk variables was examined, low crash risk perceptions were associated with DUI of all substances and low legal risk perceptions were associated with DUI of ecstasy. CONCLUSIONS The observed reduction in DUI of psychostimulants among frequent ecstasy consumers may be related to increased risk awareness stemming from educational campaigns and the introduction of saliva testing on Australian roads. Such countermeasures may be less effective in relation to deterring or changing attitudes towards DUI of cannabis and alcohol among this group.

How does the use of multiple needles/syringes per injecting episode impact on the measurement of individual level needle and syringe program coverage?

BACKGROUND Recent work by McCormack et al. (2016) showed that the inclusion of syringe stockpiling improves the measurement of individual-level syringe coverage. We explored whether including the use of a new parameter, multiple sterile syringes per injecting episode, further improves coverage measures. METHODS Data comes from 838 people who inject drugs, interviewed as part of the 2015 Illicit Drug Reporting System. Along with syringe coverage questions, the survey recorded the number of sterile syringes used on average per injecting episode. We constructed three measures of coverage: one adapted from Bluthenthal et al. (2007), the McCormack et al. measure, and a new coverage measure that included use of multiple syringes. Predictors of multiple syringe use and insufficient coverage (<100% of injecting episodes using a sterile syringe) using the new measure, were tested in logistic regression and the ability of the measures to discriminate key risk behaviours was compared using ROC curve analysis. RESULTS 134 (16%) participants reported needing multiple syringes per injecting episode. Women showed significantly increased odds of multiple syringe use, as did those reporting injection related injuries/diseases and injecting of opioid substitution drugs or pharmaceutical opioids. Levels of insufficient coverage across the three measures were substantial (20%-28%). ROC curve analysis suggested that our new measure was no better at discriminating injecting risk behaviours than the existing measures. CONCLUSION Based on our findings, there appears to be little need for adding a multiple syringe use parameter to existing coverage formulae. Hence, we recommend that multiple syringe use is not included in the measurement of individual-level syringe coverage.

The Potential Reach of Opioid Substitution Settings to Deliver HCV Care to People Who Inject Drugs in Australia.

BACKGROUND Recent efforts in Australia to engage people who inject drugs (PWID) in hepatitis C (HCV) care have focussed on opioid substitution treatment (OST) services as a delivery site. This approach potentially excludes non-opioid using PWID. This study aimed to determine differences between those currently receiving OST and those not among a sample of PWID. METHODS Additional questions on HCV testing were included in the 2013 Illicit Drug Reporting System (IDRS), an annual sentinel surveillance system. The IDRS recruits PWID in all Australian capital cities from a range of sources, predominantly needle and syringe programs. All participants are reimbursed AUD

Estimating prevalence of drug and alcohol presentations to hospital emergency departments in NSW, Australia: impact of hospital consultation liaison services

40 for a ~45 minute interview-administered survey. RESULTS Current OST was reported by 44% of the total sample, and two-thirds reported an opioid as their drug of choice. Those participants who reported current OST were significantly more likely than those not in OST to report heroin as their drug of choice (65% vs. 43%, p<0.0.001) and the drug injected most often (53% vs. 30%, p<0.001). Compared to those in OST, those not in OST were more likely to report methamphetamine as their drug of choice (29% vs. 14%, p<0.001) or drug injected most often (33% vs. 17%, p<0.001). Current OST clients were more likely to have been tested for HCV antibodies (anti-HCV) than those not in OST (96% vs. 93%, p<0.05) and to report an anti-HCV positive result (75% vs. 64%, p<0.001). Those receiving OST were no more likely than those not to undergo further HCV-related testing (e.g. RNA) (62% vs. 56%, p=0.136). Both groups reported further HCV-related testing was undertaken primarily at a community general practice. DISCUSSION Despite a large proportion of current PWID receiving OST, there remains a substantial minority who are neither seeking nor eligible for treatment. Efforts to improve access to HCV care for PWID in Australia therefore need to be expanded beyond OST, especially given the large proportion of participants who reported inadequate HCV testing.

Evaluating the Impact of Hospital Based Drug and Alcohol Consultation Liaison Services

The impact of drug and alcohol misuse has been the subject of widespread media discussion in the past year, particularly in the context of restricted alcohol trading hours in an effort to reduce alcohol-fuelled violence. A recent study evaluating NSW Healths drug and alcohol consultation liaison (CL) services1 demonstrates how pervasive drug and alcohol problems are, and the impact they have on the health system. This paper highlights how expanding CL services to fill current unmet need could deliver a range of benefits to patients and hospitals.

Corrigendum to “Characterising dark net marketplace purchasers in a sample of regular psychostimulant users” [International Journal of Drug Policy 35 (2016) 32–37]

Consultation liaison (CL) services provide direct access to specialist services for support, treatment advice and assistance with the management of a given condition. Alcohol and other drugs (AOD) CL services aim to improve identification and treatment of patients with AOD morbidity. Our objective was to evaluate the costs and consequences of AOD CL services in hospitals in New South Wales, Australia. Patients were surveyed at eight hospitals and problematic AOD use was identified using the Alcohol, Smoking and Substance Involvement Screening Test (n=1615). For consenting participants, medical record data were obtained from 18 months pre- to 12 months post-survey. We used interrupted time series analyses to compare utilization and costs for patients with and without AOD problems and changes over time between those who received AOD CL and similar patients. Approximately 35% of patients surveyed had AOD problems (excluding tobacco) with 7% requiring intensive treatment. Only 24% of patients requiring intensive treatment were treated by AOD CL. Those treated had relative improvements over time in the cost of presentations to emergency departments, emergency admission performance and increased uptake of appropriate pharmaceuticals. The estimated net benefit of AOD CL services was at least AUD

Characterising dark net marketplace purchasers in a sample of regular psychostimulant users

100,000 savings per hospital per year. Expanding AOD CL services to address current unmet need may lead to even greater cost savings for hospitals.

New psychoactive substance use among regular psychostimulant users in Australia, 2010–2015

The authors regret that there is an inconsistency between figures presented in the abstract and those reported in the text of the manuscript. Specifically, the abstract states that 68 out of 800 participants in the sample reported recent darknet use, while the manuscript states that 66 of the 745 participants reported recent darknet use. The latter numbers are correct for the analyses performed, that is, 66 of the 745 interviewed participants had purchased from the darknet in the preceding 12 months, and comprise the sample included in the regression model. While the former numbers are correct for the entire sample of recruited participants, they are not appropriate for the analyses performed. The authors would like to apologise for any inconvenience caused.