Daniel O'Keefe

From initiating injecting drug use to regular injecting: Retrospective survival analysis of injecting progression within a sample of people who inject drugs regularly

BACKGROUND The initiation of injecting drug use and the commencement of a pattern of regular injecting are key milestones in injecting careers. The progression from initiation to regular injecting is a poorly understood period in these careers. METHODS Cross-sectional baseline data from a sample of people who inject drugs regularly (N=691), recorded the age at which participants initiated injecting drug use and the age they became regular (at least once per month) injectors. Survival analysis compared the rapidity of progression to regular injecting across sub-groups within the sample using bivariate log-rank testing and multivariable Cox regression. RESULTS Half of all participants progressed to regular injecting within 1 year of initiation and by the fourth year post-initiation, 91% had progressed. In bivariate analysis, there were significant differences in equality of hazards by sex (X(2)=7.75, p<0.01), from whom participants learnt to inject (X(2)=22.32, p<0.01) and the drug of injection initiation (X(2)=18.36; p<0.01). In the multivariable Cox model, only initiating injecting with heroin (HR=1.28; 95% CI: 1.09-1.50) compared with other drugs (predominantly methamphetamine) showed a significantly greater hazard, suggesting a faster progression to regular injecting. CONCLUSION This study showed that among our sample of eventual regular injectors, progression from initiation to regular injecting was rapid. By gaining a greater understanding of the dynamics of this progression, the ability to appropriately target interventions and future research is subsequently informed.

Injecting drug use in low and middle income countries: opportunities to improve care and prevent harm

Inadequate response to injecting drug use (IDU) is a significant problem the world over. Low levels of funding, political inaction, poor levels of health service coverage, high prevalence and incidence of IDU‐related blood‐borne viruses (BBVs) and ongoing stigmatization/marginalization affect people who inject drugs (PWID) regardless of the income status of the country they reside in. These barriers and system failings are, however, exacerbated in low and middle‐income countries (LMICs), meaning that the potential consequences of inaction are more pressing. In this narrative review, we describe the levels of IDU and IDU‐specific BBV prevalence in LMICs; levels of harm reduction implementation; the consequences of late or insufficient response, the shortcomings of data collection and dissemination; and the barriers to effective LMIC harm reduction implementation. We also exemplify cases where IDU‐related harms and BBV epidemics have been successfully curtailed in LMICs, showing that effective response, despite the barriers, is possible. In conclusion, we suggest four key priorities on the basis of the review: confirming the presence or absence of IDU in LMICs, improving the collection and dissemination of national IDU‐specific data, increasing the level of harm reduction programme implementation in LMICs, and increasing both national and international advocacy for PWID and attendant public health interventions.

The Association between Intentional Overdose and Same-Sex Sexual Intercourse in a Cohort of People who Inject Drugs in Melbourne, Australia

ABSTRACT Background: People who inject drugs (PWID) are at disproportionately high risk of suicidal behaviors, as are individuals who report same-sex attraction or experience. However, there is little evidence of compounded risk of suicide for individuals who report same-sex sexual intercourse (SSI) and are PWID. Objectives: To explore the associations of lifetime intentional overdose amongst a cohort of PWID, with particular attention to those reporting SSI. Methods: The sample included 529 participants, from an ongoing cohort of 757 PWID. An “ever” SSI variable was created for participants who reported sexual intercourse with a same-sex partner at any longitudinal interview. We explored the adjusted associations between SSI and lifetime intentional overdose using logistic regression. Results: Ninety-one (17%) participants reported ever experiencing an intentional overdose. Forty-one (8%) participants reported SSI at any interview. Three hundred and sixty (68%) participants reported diagnosis of a mental health condition. Diagnosis of a mental health condition (AOR = 2.02, 95% CIs: 1.14, 3.59) and SSI (AOR = 2.58, 95% CIs: 1.22, 5.48) significantly increased the odds of lifetime intentional overdose. Conclusions/Importance: We found a heightened risk of intentional overdose amongst PWID reporting SSI, after controlling for diagnosis of a mental health condition. Services need to be aware of this heightened risk and target interventions appropriately.

Blood-borne virus transmission in an urban, culturally diverse neighbourhood: Results from a cross-sectional bio-behavioural survey using innovative outreach methods in a hard-to-reach population

Background Following a HIV outbreak among Aboriginal people in a culturally diverse inner-city suburb of Melbourne, a blood-borne virus (BBV) screening program was conducted to inform public health interventions to prevent transmission and facilitate timely diagnosis and linkage to care. METHODS In August-September 2014, community health workers recruited people who inject drugs (PWID) from a local needle and syringe program. Participants were tested for hepatitis C virus (HCV), hepatitis B virus (HBV), HIV and syphilis and completed a bio-behavioural questionnaire. RESULTS In total, 128 PWID participated in the study. Serological evidence of exposure to HCV and HBV was detected among 118 (93%) and 57 participants (45%) respectively. Five participants were HIV positive. Independent risk factors for needle sharing were Aboriginality (AOR=6.21, P<0.001), attending health care for mental health problems (AOR=2.79, P=0.023) and inability to access drug treatment in the previous 6 months (AOR=4.34, P=0.023). CONCLUSIONS BBV prevalence in this sample was much higher than reported in other recent Australian studies. This local population is at high risk of further BBV transmission, particularly Aboriginal PWID. Individual and service-related factors associated with risk in the context of a dynamic urban drug culture and HIV outbreak suggest an urgent need for tailored harm-reduction measures.

A response to: Self-reported and actual hepatitis C virus status

Sir—We thank our correspondents for their response to our paper, and wish to highlight several points in reply. We agree that the concordance between a person’s actual and perceived hepatitis C virus (HCV) status may be influenced by multiple factors, educational background being one. Concordance may also be affected by the characteristics of different diseases and by the health environments in which those diseases exist. Hepatitis B virus (HBV) is a more complex virus to diagnose and explain than HCV, and so we find it unsurprising that our rates of HCV concordance are higher than the HBV concordance in relatively low-risk first-year Thai medical students. Previous studies have found similarly low levels of concordance for HBV in people who inject drugs (PWID) [1]. We agree with our correspondents that high rates of HCV concordance among PWID should not be used to replace testing with self-report. Our data suggest 20% of Australian PWID do not know their HCV status, and thus are at risk of unknowingly transmitting or acquiring HCV. There is no substitute for widespread availability of free sero-testing and comprehensive and accurate post-test discussions. Our work in the HCV field over more than two decades shows that voluntary blood-borne virus testing outside of traditional settings is acceptable to PWID and that this can improve knowledge and reduce risk taking [2,3].

A response to: Self-reported and actual hepatitis C virus infection status: Letter to the Editor

Sir—We thank our correspondents for their response to our paper, and wish to highlight several points in reply. We agree that the concordance between a person’s actual and perceived hepatitis C virus (HCV) status may be influenced by multiple factors, educational background being one. Concordance may also be affected by the characteristics of different diseases and by the health environments in which those diseases exist. Hepatitis B virus (HBV) is a more complex virus to diagnose and explain than HCV, and so we find it unsurprising that our rates of HCV concordance are higher than the HBV concordance in relatively low-risk first-year Thai medical students. Previous studies have found similarly low levels of concordance for HBV in people who inject drugs (PWID) [1]. We agree with our correspondents that high rates of HCV concordance among PWID should not be used to replace testing with self-report. Our data suggest 20% of Australian PWID do not know their HCV status, and thus are at risk of unknowingly transmitting or acquiring HCV. There is no substitute for widespread availability of free sero-testing and comprehensive and accurate post-test discussions. Our work in the HCV field over more than two decades shows that voluntary blood-borne virus testing outside of traditional settings is acceptable to PWID and that this can improve knowledge and reduce risk taking [2,3].

A response to: Self-reported and actual hepatitis C virus infection status

Sir—We thank our correspondents for their response to our paper, and wish to highlight several points in reply. We agree that the concordance between a person’s actual and perceived hepatitis C virus (HCV) status may be influenced by multiple factors, educational background being one. Concordance may also be affected by the characteristics of different diseases and by the health environments in which those diseases exist. Hepatitis B virus (HBV) is a more complex virus to diagnose and explain than HCV, and so we find it unsurprising that our rates of HCV concordance are higher than the HBV concordance in relatively low-risk first-year Thai medical students. Previous studies have found similarly low levels of concordance for HBV in people who inject drugs (PWID) [1]. We agree with our correspondents that high rates of HCV concordance among PWID should not be used to replace testing with self-report. Our data suggest 20% of Australian PWID do not know their HCV status, and thus are at risk of unknowingly transmitting or acquiring HCV. There is no substitute for widespread availability of free sero-testing and comprehensive and accurate post-test discussions. Our work in the HCV field over more than two decades shows that voluntary blood-borne virus testing outside of traditional settings is acceptable to PWID and that this can improve knowledge and reduce risk taking [2,3].

Understanding drug-related death in recently released prisoners

Rapid advancements in the technology used to study the human genome has led to the identification of a number of genetic variants thought to increase susceptibility to alcohol dependence or influence individual’s treatment responses (e.g. to naltrexone). Optimistic predictions have been made about the potential clinical applications of such knowledge: predictive genetic screening might be used to identify persons at a greater risk of developing alcohol dependence and prevent them from doing so; while pharmacogenetic approaches may allow clinicians to match alcohol dependent persons to more effective treatments. Identifying genes predictive of alcohol dependence or treatment response carries potential harms. An overemphasis of the genetic basis of alcohol dependence could reduce an individual’s belief in their ability to moderate their alcohol consumption or abstain from alcohol. Medicalising alcohol dependence may lead to a greater focus on individual medical treatment at the expense of public health interventions that more broadly reduce the harms of all alcohol consumption, such as increased taxation and raising the minimum legal drinking age. Access to genetic information on alcohol dependence by third parties such as health insurance companies, may also be a significant concern and a potential source of discrimination. This paper critically evaluates the potential benefits and possible harms of clinical applications of genetics research on susceptibility to alcohol dependence. We will review the current evidence regarding the genetics of alcohol dependence; the feasibility of predictive genetic screening and pharmacogenetic approaches to alcohol dependence; and the ethical and social policy issues associated with such uses.