UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Risk factors and determinants of HIV and Hepatitis C prevalence and incidence among a cohort of young… Miller, Caroline L. 2002

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2002-0496.pdf [ 4.64MB ]
Metadata
JSON: 831-1.0076915.json
JSON-LD: 831-1.0076915-ld.json
RDF/XML (Pretty): 831-1.0076915-rdf.xml
RDF/JSON: 831-1.0076915-rdf.json
Turtle: 831-1.0076915-turtle.txt
N-Triples: 831-1.0076915-rdf-ntriples.txt
Original Record: 831-1.0076915-source.json
Full Text
831-1.0076915-fulltext.txt
Citation
831-1.0076915.ris

Full Text

RISK,FACTORS A N D DETERMINANTS OF HIV A N D HEPATITIS C PREVALENCE A N D INCIDENCE A M O N G A C O H O R T OF Y O U N G INJECTION D R U G USERS by CAROLINE L. MILLER B . A , Carleton University, 1999  A THESIS SUBMITTED I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R OF S C I E N C E  in T H E F A C U L T Y O F G R A D U A T E STUDIES Department of Individual Interdisciplinary Graduate Studies Program We accept this thesis as conforming to the required standard  ~"Dr. MarMn'T. Sjehechter Dr. Pamela A . Ratner Dr. Patricia Kachuk Dr. Sheila Marshall The University of British Columbia August, 2002 © Caroline L . Miller, 2002  In presenting this thesis in partial fulfilment  of the requirements for an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department  or by his or her  representatives.  It is understood  that copying or  publication of this thesis for financial gain shall not be allowed without my written permission.  The University of British Columbia Vancouver, Canada  DE-6 (2/88)  ABSTRACT Objectives: The primary objectives of this study were to determine the prevalence, incidence and risk factors for H I V and Hepatitis C (HCV) infection among young (aged 13 - 24 years) injection drug users (IDUs) participating i n the Vancouver Injection D r u g Users Study (VIDUS). Methods: This study was a nested sub-study within the V I D U S project. The V I D U S is an open cohort consisting of over 1,400 Vancouver area IDUs. Participants must be aged 13 years and older, have injected illicit drugs at least once i n the previous month and reside i n the greater Vancouver area. Enrolment into the study began i n 1996, and the majority of participants were enrolled between 1996 and 1997. average, 100 new study participants have been enrolled each year.  Since 1997, on A t baseline and  semi-annually thereafter, subjects undergo an interviewer-administered questionnaire and  provide a  venous  blood  sample  for  H I V and  H C V antibody  testing.  Questionnaires are administered i n English, Spanish, Mandarin, French and Cantonese and elicit information on participants' risk factors and living conditions i n the previous six months.  Nonparametric and multivariate regression methods were used w h e n  analyses were conducted on cross-sectional data.  Longitudinal data were analyzed  using person-time techniques.  Results:  Baseline H C V prevalence was 46% (107) and was associated with being  Aboriginal, incarceration i n the previous 6 months, survival sex, <100 lifetime partners, residence i n the D o w n t o w n Eastside, and at least daily heroin, cocaine, or speedball injection. H C V seroconversion occurred among 37 of the youth, an incidence rate of 37.2 per person-years; the median age was 20 years and the median number of years injecting was 1.1. H C V seroconversion was independently associated with having an I D U partner and daily cocaine injection. Baseline H I V prevalence among the youth was ii  10% (23) and was independently associated w i t h being female, daily cocaine injection, older age, and a greater number of years injecting. H I V seroconversion occurred among 16 youth, an incidence rate of 4.4 per person years; the median age was 22 and median number of years injecting was 3.0. H I V incidence was associated with being Aboriginal, at least daily crack or cocaine use, and <100 lifetime sexual partners. Conclusion:  This sub-study  was  the  first  attempt  to  examine  demographic  characteristics and risk factors among young IDUs i n a city coping w i t h epidemic proportions of H C V and H I V among the I D U population. Y o u n g IDUs are of particular importance in terms of prevention of blood-borne infections. Relative to older IDUs, younger IDUs have lower prevalence of H I V and H C V but alarmingly high incidence rates, particularly among female and Aboriginal youth. The w i n d o w of opportunity to act is short, and preventive actions are urgently required. important clues where resources need to be targeted.  iii  This study has provided  TABLE OF CONTENTS Page ABSTRACT  ii  TABLE OF CONTENTS  iv  ACKNOWLEDGEMENTS  vi  1  2.  3.  CHAPTER ONE: INTRODUCTION A N D OVERVIEW  1  The H I V epidemic among youth  1  Hepatitis C among injection drug users  4  Research needs and study justification  5  Study objectives and thesis organization  7  Study setting  8  Methods  10  Limitations.  14  Ethical considerations and confidentiality  15  Summary  16  References  17  CHAPTER TWO: REVIEW OF T H E LITERATURE  23  Introduction  23  H I V prevalence and incidence and associated risk factors  25  H C V prevalence and incidence and associated risk factors  27  Aboriginal people and the H I V epidemic  29  Conclusion  30  References  31  CHAPTER THREE: HIV prevalence and associated risk factors41 Introduction  41  Methods  42  Results  :  43  Discussion  45  References  51  Tables  57  iv  4.  5.  6.  7.  CHAPTER FOUR: HIV incidence  61  Introduction  61  Methods  61  Results  62  Discussion  63  References  65  Table  69  CHAPTER FIVE: H C V prevalence and incidence  65  Introduction  65  Methods  70  Results  72  Discussion  74  References  79  Tables  84  CHAPTER SIX: Aboriginal youth  86  Introduction  86  Methods  87  Results  88  Discussion  89  References  92  Tables  96  CHAPTER SEVEN: General Discussion  97  Summary of study findings  97  Unique contributions  99  Recommendations  100  Conclusions  102  References  103  APPENDICES  106  Statement of authorship Contribution  107  v  ACKNOWLEDGEMENTS I w o u l d like to thank m y committee members, Drs. Schechter, Ratner, Kachuk and Marshall for their mentorship, support, guidance, suggestions and hours of editorial contributions. I w o u l d also very much like to extend m y gratitude to the Michael Smith Foundation for Health Research w h o provided the financial support for this project through a training award.  I also need to thank the staff at V I D U S for numerous  discussions and their contributions of ideas and also the participants without w h o m this study w o u l d not have been possible. I w o u l d also like to thank m y fellow students at the B C Centre for Excellence in H I V / A I D S for numerous enlightening conversions, statistical help, ideas and editorial suggestions; Evan, Paula, Magda, and A m y . I also need to thank the staff at work; Drs. O'Shaughnessy, Montaner and Spittal as well as Peter, L i z , M y r n a , Diane and especially Kathy for their patient help and assistance. I a m also grateful to m y support team Katrina, Carey, Cate, Jodi and m y m o m for the two years of tears and laughter this project produced. A n d finally I a m indebted to m y partner Bob, and to our children, Erin, Samuel and especially Roo for putting up with me during this process. I sincerely hope that this thesis work w i l l give a little back to the community researched and make a contribution towards a safer environment for vulnerable youth.  vi  CHAPTER 1 INTRODUCTION A N D OVERVIEW  1.1  The HIV epidemic among youth Y o u n g people are at the centre of the H I V / A I D S epidemic. A r o u n d the globe an  estimated 11.8 million people aged 15 - 24 years are living with the virus, and one half of all new infections, an estimated 7,000 per day, occur i n this age group[l] . A r o u n d the w o r l d youth are at increased risk for H I V infection for reasons beyond their control. Youth may not receive effective education regarding their sexual health and they may not have access to or knowledge about condom use.  Furthermore, y o u n g  people, especially girls, may be at increased risk for sexual abuse b y familial members, extended family, acquaintances, strangers or sexual partners, increasing their risk for H I V directly or b y limiting their ability to negotiate safer sexual practices[2, 3] . Y o u n g people marginalized by poverty, abuse, and neglect may find themselves living in precarious living arrangements, either on the street or in risky relationships.  These  young people are at increased risk of engaging in injection drug use and sex trade work, thereby significantly increasing their risk of blood-borne infections. It is estimated that globally, there are over 100 million children under the age of 18 living in perilous conditions or on the street[4]  and i n Canada, there are an  estimated 45,000 to 150,000 street youth, most of w h o m live in large urban centres [5] . These children and youth are particularly vulnerable to H I V through their social networks, older contacts, injection drug use - either their o w n or that of their sexual partners - and sex trade work. The relationship and potential transition from street youth to injection drug use remains largely unexplored.  1  However studies of street  youth i n N o r t h America have shown that those at highest risk for H I V are those w h o use injection drugs [6-9]. In Canada, youth aged 24 and under represent a small proportion of the total number of AIDS cases, however they are the group that is most at risk in this shifting epidemic[10] . Furthermore, they are tremendously important i n terms of prevention and fighting the spread of H I V / A I D S . In Canada, the median age of those with new infections has dropped from age 32 (for those infected before 1983) to 23 for those infected between 1985 and 1990[11]. Before the mid-1990's, those most at risk for H I V were men w h o have sex w i t h men (MSM). Post-1995, the epidemic experienced a shift and increasing rates of H I V incidence among IDTJ populations occurred in urban centres across Canada[12] . In 1996, 47% of new infections were attributable to IDUs and i n 1999 that figure dropped to 34%. Currently the modes of H I V transmission are split approximately equally between M S M , IDUs, and heterosexual contact[12]. In Canada, the proportion of females among those w i t h positive H I V test results varies considerably by age and is highest among adolescents and young adults[13] . Y o u n g females are particularly vulnerable to H I V infection due to an increased exposure to sexual violence and participation in sex trade work[14] .  For girls and  women w h o have been exposed to sexual violence or w h o w o r k i n the sex trade, this often means that sexual behaviour occurs i n the context of unequal power.  Condom  use is a sexual behaviour that is often under the control of men, especially when sex is traded or forced, thus condom use becomes problematic for girls and women in sexual relationships where power imbalances exist.  Furthermore, girls and women are at  higher risk for H I V through sexual contact due to increased viral efficacy in sexual transmission from man to women[15] . Of the 7,000 new infections occurring among youth each day, over one half of these occur among young women [4] . Y o u n g women are particularly important in 2  terms of risk for IDUs because they tend to account for a higher proportion of street youth and to be of younger  age when injection drug use is initiated[16  /  17] .  Furthermore, young women who use injection drugs are more likely than their male counterparts to have been sexually abused in childhood and to face further sexual victimization through sex trade w o r k and sexual violence[18] .  A m o n g Canadian  women, about one half of new H I V infections are attributable to injection drug use[12]. The H I V epidemic among Canadian Aboriginal peoples is a growing concern. The term Aboriginal comprises the many First Nations, Inuit, Innu, and Metis people living across the country.  Within these groups there are many nations that are  culturally and linguistically distinct.  According to Statistics Canada, about 3% of the  Canadian population identify as Aboriginal peoples[19] . The proportion of Aboriginal people living i n each province varies, the western provinces have higher proportions; in Manitoba about 14% of the population are Aboriginal and in British Columbia this figure is approximately 4%[19]. The Canadian Aboriginal population is on average ten years younger than the non-Aboriginal population. The majority of aboriginal peoples are under the age of 25, thus the health of Aboriginal youth is of particular importance. In Canada and particularly in the western provinces, Aboriginal people have been disproportionately affected by H I V / A I D S .  The proportion of Aboriginal AIDS  cases increased from 1% before 1990 to 9% in 2000[20] .  Provinces w i t h ethnicity  reporting indicate that Aboriginal people account for approximately 20% of newly diagnosed H I V infections[20] . Governmental policies such as displacement from land to reserve and the residential school system have precipitated cultural, linguistic and familial erosion and have undoubtedly contributed to much higher rates of morbidity and mortality among some Aboriginal people and communities. These policies have placed Aboriginal people at higher risk for living i n unstable housing, poverty and IDUs  3  and these factors have contributed to disproportionately higher rates of H I V and AIDS [21, 22] . The epidemiology of the epidemic in Aboriginal populations is concerning; Aboriginal people infected w i t h H I V are more likely to be i n the I D U transmission group and much higher proportions of females and youth are infected than in the nonAboriginal population.  W h e n Aboriginal Canadians were  compared w i t h non-  Aboriginal Canadians among new H I V infections between 1998 and 2000: females accounted for 47% vs. 20%; i n the age category 20 to 29 years, Aboriginals numbered 33% vs. 20%; and i n the transmission category injection drug use, Aboriginals accounted for 60% vs. 34% [20] . The epidemiology of H I V and AIDS among Aboriginal people have important implications for the prevention and treatment of this population. Moreover, given the historical betrayal characterizing many government policies directed at Aboriginal peoples, intervention and prevention programs will need to come from, and be i n collaboration with, Aboriginal people.  1.2  Hepatitis C among injection drug users The highest risk for hepatitis C ( H C V ) in N o r t h America is the sharing of  contaminated drug equipment between IDUs[23-26]. A m o n g populations of long-term IDUs in N o r t h American urban centres H C V prevalence estimates generally range between 80% and 90% [27-29]. The sexual transmission of H C V is a controversial topic that has neither been proven nor ruled out. However among cohorts of IDUs H C V is highly prevalent and time to infection from initiation into injection drug use is short. Unlike H I V , H C V is highly transmissible through shared injection equipment other than needles, such as cottons , cookers and straws [30,31]. 2  2  3  4  3  4  Typically a cotton ball used to swab the injection area with alcohol. a dish used to heat the powdered from of a drug to reduce it to liquid. a drinking straw that is inserted nasally to snort the powdered form of a drug.  4  Similar to H I V , H C V is a blood-borne virus that is transmitted through contact w i t h infected blood. Unlike H I V , some individuals w h o are infected w i t h H C V clear the virus, however most, approximately 80%, do not and live with chronic H C V infection[32]. Widespread H C V infection i n N o r t h America occurred recently thus the long-term effects of H C V , and particularly co-infection w i t h H I V and H C V , may not be entirely known. What is k n o w n is approximately 80% develop cirrhosis of the liver and 10% - 20% of those go on to die from liver disease and cancer[33] .  The health  consequences of chronic H C V infection are significant. It is estimated that there are between 210,000 and 275,000 people in Canada living w i t h H C V infection. Most of these infections are due to IDUs. Estimates of time to infection from the onset of injection drug use range between 1 and 4 years[27, 34, 35] , thus young IDUs are central to understanding the epidemiology of H C V infection, and preventing the spread of the H C V virus. H C V prevalence estimates among y o u n g IDUs are lower when compared w i t h older injectors, however alarmingly high H C V incidence rates have been documented among young IDUs.  1.3  Research needs and study justification Prior to 1990, the strongest risk factor associated with the spread of H I V in  "developed" nations was unprotected intercourse between men w h o have sex with men (MSM) [10] . Since the mid-1990's the risk categories have shifted and y o u n g women have become the highest risk group for H I V infection worldwide. In the United States teenagers, aged 13 to 21, n o w account for one quarter of all new infections in the United States [36] . In Canada, youth aged 15 - 24, account for one quarter of all new infections [10] .  The H I V epidemic among adolescents is well  5  established and growing [16, 37-40] . Research efforts focusing on risk factors among high-risk youth is urgently needed to stem the tide of new infections. Transmission risk categories among Canadian youth are split fairly evenly between IDUs, heterosexual contact, and M S M . W h e n newly diagnosed H I V data are gender and age stratified, females aged 24 and under account for a greater proportion of new infections than males and IDUs account for one half of those new infections. There have been few studies undertaken in Canada that have examined longitudinally socio-demographic characteristics and risk factors among young IDUs. In Vancouver, British Columbia, much of the I D U population is concentrated in one area of approximately 10 city blocks k n o w n as the D o w n t o w n Eastside (DTES). There are approximately 5,000 I D U residing i n the DTES. Since 1995, Vancouver has experienced an explosive H I V epidemic among IDUs. In 1996 the H I V incidence rate peaked at 19 per 100 person years, and has since declined but still remains unacceptably high. H C V is saturated among the I D U population in Vancouver, where an estimated 85% of IDUs are infected w i t h H C V [ 2 6 ] . Youth living in the DTES are exposed to high rates of poverty, injection drug use and blood-borne infection. Despite being at high risk for blood-borne infections, there have been few studies characterizing the youth in the DTES and there have been no studies to date characterizing the young I D U population who live i n or frequent the DTES. In M a y of 1996 the Vancouver Injection D r u g Users Study (VIDUS) began collecting data. The V I D U S project is an open cohort of Vancouver area IDUs designed to longitudinally examine risk factors for H I V and H C V incidence. Since the project began there have been many published articles and reports detailing the explosive H I V epidemic that occurred in 1996 among Vancouver area IDUs, risk factors for H I V and H C V , and drug use patterns of participants. However, prior to this current study, there had been no investigation of the young V I D U S participants. 6  Despite the growing epidemic that has been documented among youth and particularly among youth w h o use injection drugs, there has been little research undertaken on young IDUs, H I V and H C V prevalence and incidence rates and associated risk factors, particularly in a longitudinal fashion. Y o u n g IDUs are central to preventing the spread of blood-borne infections because of their vulnerability to infection and because it is during this period of life when many behaviour patterns are established  that  will  affect  risk  for  infection throughout  their  adult  years.  Characterizing risk factors and socio-demographics of young IDUs w h o are most at risk will enable communities struggling with these viruses to target resources where they are most needed.  Furthermore, documenting H I V and H C V prevalence and  incidence may help to draw the attention of policy makers w h o can target additional resources for this vulnerable population to prevent the further spread of the viruses.  1.4  Study objectives and thesis organization This thesis is divided into seven chapters. The first chapter provides an overview  of the H I V and H C V epidemic among young people aged 24 and under. This overview sets the stage for the thesis work by providing a brief summary of the current state of knowledge and the impetus for this thesis work. Chapter one gives details of the study site and methodological issues. The second chapter is a review of the current literature on young injection drug users, youth at risk for blood-borne infections i n Canada, and young Aboriginal people in relation to the H I V and H C V epidemics.  Chapters 3  through 6 address the main study objectives as outlined below, and the final chapter 7 summarizes the findings of this thesis w o r k and outlines future study needs and directions for research among young IDUs.  7  The primary objectives of this study were to determine prevalence, incidence and risk factors for H I V and Hepatitis C ( H C V ) infection among young (aged 13 - 24 years) injection drug users participating in the V I D U S project and to address some of the questions regarding the impact that these youth have on the H I V and H C V epidemics i n Canada. The first objective was to estimate the prevalence of H I V at the enrollment questionnaire among the youth participating in the V I D U S project.  This initial  investigation set the stage for characterizing w h o may be most at risk for the H I V virus among young injection drug users i n the Vancouver area. The second objective was to estimate the H I V incidence rate among the V I D U S youth and to use the longitudinal data available to ascertain time relevant risk factors for seroincident youth. The third objective was to investigate risk factors for H C V prevalence and incidence among the youth. The incident analysis allowed for the identification of time relevant risk associations for H C V among the youth. The fourth objective  was  to  characterize  the  young  Aboriginal  VIDUS  participants. This fourth objective was added following the previous three analyses where Aboriginal youth were consistently identified at high risk for both H I V and HCV. Each of the four objectives is addressed in a chapter, consisting of a stand alone manuscript that has either been published, accepted for publication or under review w i t h an international peer-reviewed scholarly journal. The final chapter, chapter seven, provides a review of the study findings and discusses the implications of the results. The unique contributions and their relevance to the current literature are discussed. Finally recommendations are put forth for future directions in research and resources that may be useful i n caring for high-risk youth and preventing further transmission of 8  blood-borne infections. In summary, the goals of this study were to contribute to the body of knowledge of youth at risk for blood-borne infections by characterizing the V I D U S youth and establishing w h i c h risk factors were most important. 1.5  Study setting, overview of methods, and limitations"  1.5.1  Study setting The Vancouver Injection D r u g Users Study (VIDUS) office is located i n a  storefront office in the D o w n t o w n Eastside of Vancouver.  V I D U S began enrolling  participants i n M a y of 1996 following the identification of an outbreak of H I V infection among Vancouver area IDUs. The V I D U S project was designed as an open cohort, where current injection drug users (used injection drugs within the previous month) can enroll at any time.  Subjects were recruited through self-referral and street  outreach. Subjects were eligible if they had injected illicit drugs at least once during the month prior to enrollment, resided in the Greater Vancouver region, and were able to provide written informed consent.  Follow-up was every six months, at which time  participants were eligible to return for follow-up interviews and blood tests. Every six months, participants were asked to complete a detailed socio-demographic and health status survey administered by trained nurses and interviewers and to provide a venous blood sample that was tested for H I V and H C V .  Since the study began, over 1,400  Vancouver area injection drug users have been enrolled, the majority - approximately 1,100 - were enrolled during 1996 and 1997, however V I D U S continues to enroll new participants.  There are a variety of methods employed to ensure optimal follow-up of this hard to reach population. Some of the strategies include: contacting agencies (social services or personal contacts) that the subject frequents to remind the participant to 9  attend  the V I D U S clinic, contacting the family physician, providing monetary  compensation ($20.00) to return, contacting police and conducting interviews in jail when the participant is incarcerated, and conducting interviews in hospital w h e n appropriate. These methods have proven highly successful - V I D U S currently has a 75% follow-up rate. The V I D U S study has received ethical approval from the St. Paul's Hospital committee on human experimentation.  2.5.2  Methods: The VIDUS questionnaires When participants first enroll in the V I D U S project they are administered the  baseline questionnaire by one of the trained interviewers of the project. Semi-annually thereafter participants are eligible to complete a follow-up questionnaire.  The V I D U S  project, as of June 2002, is currently administering follow-up questionnaire number 11. A l l follow-up questionnaires ask standard socio-demographic, sexual and drug risk characteristics, however some questions have been changed, deleted, or added in some of the follow-up questionnaires.  Following enrollment, participants complete the  current follow-up questionnaire for that time period. For example, a participant w h o was enrolled in January 2002 and returned for follow-up in June 2002 w o u l d have completed the baseline questionnaire and follow-up number 11.  2.5.3  Methods: Definition of youth Youth in this study were defined as aged 24 years and younger at the time of  recruitment.  Definitions of youth vary greatly and there seems to be no consistent  means to define youth or adolescence.  Some studies of risk factors for H I V among  homeless or street youth have used the age cutoff as 19 years and younger[41] while 10  others have used age 24 years and younger[9] . United States based cohort studies of young IDUs have defined youth as age 29 years and younger[42,43]. The rationale for using aged 24 and younger as the cutoff for this study is based on the age criterion for youth a n d / o r adolescence as defined i n reports generated by the United Nations, the centres for disease control i n the United States and Canada, as well as Statistics Canada[4, 10, 44] .  This age group represents those most at risk for blood-borne  infections among IDUs, given their young age and risky drug and sexual patterns. Finally, established or older injectors are likely to be already H C V positive and it m a y be more difficult to ascertain risk factors for this virus if older injectors were included in these analyses. During the study period there were 232 to 235 V I D U S participants w h o were enrolled and aged 24 years and younger. These youth are the focus of this thesis.  1.5.4  Variables of interest The V I D U S Questionnaire asks participants questions about their  current  socioeconomic and demographic profile, and injection and sexual behaviour.  All  continuous variables are asked i n reference to the previous six months at the time of interview. For this study, the following variables were operationalized either i n some or all of the analyses: Sociodemographics: Sex at birth (female/male), Aboriginal (are y o u Aboriginal, First Nations, Inuit, or Metis? [yes/no]), age (used mostly as a continuous variable or using cut-points [please see methods used i n chapter 5]), education (ever attended high school [yes/no]), incarceration (yes/no), and unstable housing (lived in a shelter, hotel, jail, treatment centre, or on the street [yes/no]).  11  Health Status: H a v e y o u been diagnosed b y a physician with an STD i n the past 6 months? (yes/no). Sexual Abuse: Has someone i n your past ever sexually abused you? (yes/no). Drug use behaviour: Years injecting, (current age minus how old were y o u when y o u first start injecting?), Current age, (mostly used as a continuous variable, however it was used with cut-points [please see 5]), >1 daily heroin, cocaine, speedball (participants were asked about their drug use frequency i n the past six months for each drug, all responses answered yes for once or greater than once per day were included), >1 daily crack use (same as above, however it is used as opposed to injected crack), b o r r o w i n g or lending (yes/no), help injecting (have y o u ever needed help injecting i n the past six months? [yes/no]), alcohol use (in the past six months have y o u regularly had alcohol to drink? [yes/no]), methadone ever (have y o u ever been on methadone maintenance therapy [yes/no]). Sexual behaviours: Sex trade (in the past six months have y o u traded sex for money, drugs or shelter? [yes/no]), use of condoms (always use condoms w i t h client, casual, or regular sexual partners? [yes/no]), number of lifetime sexual partners  (in earlier  analyses <20 lifetime sexual partners was used and i n subsequent analyses <100 lifetime sexual partners was used), I D U partner (participants were asked if their current sexual partner is an injection drug user [yes/no]).  1.5.5  Study analyses There are four sub-studies each comprising a chapter of this document. Specific  methods for each of the studies are described i n the methods section i n each of the relevant manuscript chapters.  Below I have provided a brief description of the  methods employed for each of the sub-studies i n this document. In each of the sub12  studies, the same sample was used, participants in V I D U S w h o were 24 years of age and younger at the time of recruitment. In the first two studies, there were 232 youth who met the age criteria and i n the second two studies there were 234 and 235 youth respectively due to additional enrollment. The first sub-study is investigating H I V prevalence among the youth. In order to examine the relationship between H I V infection, and the independent baseline variables noted above (sociodemographics, drug use and sexual behaviours) bivariate analysis was undertaken comparing risk factors between young and older injectors and between HIV-positive and negative youth. To assess independent predictors of the primary outcome measure, H I V prevalence, variables significant i n the bivariate analysis (P<0.05) were considered for inclusion in multivariate logistic regression modeling. These models assumed a linear relationship between independent baseline explanatory variables and log odds of H I V infection. Corresponding odds ratios and test-based 95% confidence intervals were calculated. To calculate the incidence of H I V infection among the V I D U S youth, a timedependent case-control analysis was undertaken to characterize differences between youth w h o became sero-positive and those w h o remained negative. Risk factors were examined from the questionnaire eliciting information i n the six months prior to the seroconversion occurring. Controls were randomly selected from the available H I V negative youth i n that same questionnaire time period. To ascertain risk factors for H C V prevalence among the youth, bivariate comparisons between youth w h o were H C V positive and negative at baseline were undertaken.  Multivariate regression models were used to identify  independent  predictors of H C V positivity among the youth. Predictors of H C V incidence among the youth were calculated using person-time of observation techniques, specifically C o x regression modeling. Youth w h o dropped out of the study contributed person-time to 13  the date of their last visit, or last k n o w n negative test date.  Temporal trends in  incidence were stratified by demographic variables (e.g. gender, age, ethnicity) and variables of interest (e.g. types of drug use, sex trade). In the final sub-analysis a bivariate comparison between Aboriginal and nonAboriginal youth was undertaken.  Comparisons were made between H I V and H C V  sero-status, socio-demographic characteristics, and drug and sexual risk factors. Due to statistical power limitations that precluded formal statistical analysis to investigate temporal trends for HIV-positivity among Aboriginal youth, a profile of the youth w h o became H I V seropositve was undertaken to ascertain similarities and  differences  between Aboriginal and non-Aboriginal youth.  1.6 Limitations of the study There are a number of limitations associated w i t h this study that should be mentioned.  1.6.1 Poor validity and reliability of self-reported behavioural data among IDU populations, have been raised as issues in previous studies. There have been several studies undertaken that have raised the issue of social desirability biases in the responses of I D U populations.  While Latkin and Ameijdan concluded that social  desirability had minimal effect i n earlier studies, a more recent study by Latkin suggested that social desirability did lead to under-reporting  of syringe sharing  behaviour [45,46]. The V I D U S project has w o r k e d diligently at minimizing the effects of social desirability by using interviewer-administered questionnaires.  The staff w h o  have been hired to do the interviews are long-time workers within the D o w n t o w n Eastside community and are familiar to many of the residents and youth participating in the project. 14  1.6.2  Generalizability of the results may be a limitation.  The V I D U S sample of  young injection drug users may be particularly at risk cohort of youth given the high percentage of the cohort residing in the D o w n t o w n Eastside and i n unstable housing. While the V I D U S sample of youth may not be representative of other y o u n g I D U populations, the investigators of the V I D U S project k n o w of no recruitment methods that w o u l d have selected for particular youth characteristics over others.  VIDUS  youth are likely representative of other populations of young IDUs living i n urban centres where there are high concentrations of IDUs l i v i n g i n one small urban area.  1.6.3  Loss to follow-up among the youth may present an additional concern for  generalizability.  Given the transient nature of I D U populations and in particular  young IDUs, loss to follow-up may affect the generalizability of the study. There were 37 (16%) youth w h o were lost to follow-up among the V I D U S participants. W h e n we compared the youth lost to follow-up with those w h o were followed, we found no differences with respect to socio-demographic characteristics and risk factors except that Aboriginal youth were more likely to be followed.  1.7 Ethical considerations and confidentiality A l l participants in this study provided informed consent and signed copies are kept on file. A l l participants are given detailed explanations of their rights as human subjects complying with guidelines stated by the U B C Office of Research, Clinical Research Ethics Board. The V I D U S study consent form carefully outlines the purpose of the research, the objectives and duration of the subjects' participation in the study, and procedures employed to ensure confidentiality i n lay language. N o personal identifiers 15  are used i n the study questionnaires and data are reported i n aggregate form only. In addition to ethics approval being granted for the V I D U S study, approval was sought for this sub-study from the St. Paul's Hospital Committee on H u m a n Experimentation. In addition, a great deal of care was taken to protect V I D U S participants from further harm that could occur from participating. Issues of confidentiality are taken v e r y seriously and every attempt is undertaken  to protect  participants' anonymity.  Furthermore, issues of a sensitive nature that are asked i n the questionnaire and m a y cause emotional stress or pain for the participant, such as sexual abuse, are handled w i t h care and counsellors and outreach workers are available if necessary.  1.8 Summary This study has provided data for policy makers i n the Greater Vancouver Region to use when allocating resources toward the prevention of blood-borne viruses among a highly vulnerable group - y o u n g IDUs. The data are also made available to other Canadian cities coping with blood-borne infections within their I D U populations and may be useful for providing reference points for young IDUs at highest risk. There is an urgent need to start addressing the vulnerability of youth to blood-borne infections. The V I D U S cohort provided an opportunity to establish contact w i t h this high-risk and hard-to-reach population and to analyze the wealth of data available through this project.  16  1.9 References cited  1.  U N A I D S . A I D S epidemic update. Geneva: U N A I D S , 2001:1 - 29  2.  U N D P . Adolescent Sexuality, Gender and the H I V Epidemic. London: United Nations Development Program, 2000:1 - 16  3.  Denenberg  R. Childhood sexual abuse as an H I V risk factor in w o m e n .  Treatment Issues 1997;11:22-24 4.  U N A I D S . Children and young people in a w o r l d of AIDS. Geneva: U N A I D S , 2001:1 -13  5.  Society  CP.  Bringing street  youth  out  of  the  shadows.  CPS  News  1998;May/June:5-6 6.  Roy E, Lemire N , Haley N , Boivin J-F, Frappier J-Y, Claessens C . Injection drug use among street youth: A dynamic process. Canadian Journal of Public Health 1998;89:239-240  7.  van den Hoek A . STD control in drug users and street youth. Genitourinary Medicine 1997;73:240-244  8.  Roy E, Haley N , Leclerc P, Lemire N , Boivin J-F, Frappier J-Y, Claessens C . Prevalence of H I V infection and risk behaviours among Montreal street youth. International Journal of STD and AIDS; i n press  9.  Kipke M D , O'Connor S, Palmer R, MacKenzie R G . Street youth in Los Angeles: Profile of a group at high risk for human immunodeficiency virus infection. Archives of Pediatrics and Adolescent Medicine 1995;149:513-9  10.  Health Canada. H I V and AIDS A m o n g Youth in Canada. Ottawa: Laboratory Centre for Disease control, 2000:1 - 4 17  11.  Anonymous. A g e of Canadians infected w i t h H I V dropping. C M A J . V o l . 156, 1997  12.  Health Canada. H I V / A I D S Epi Update. Ottawa: Laboratory Centres for Disease Control, 2002:1 - 76  13.  Health Canada. H I V and AIDS A m o n g W o m e n in Canada. Ottawa: Laboratory Centres for Disease Control, 2000:1 - 3  14.  Potterat TJ, Rothenberg  RB, M u t h SQ, Darrow W W , Phillips-Plummer L.  Pathways to prostitution: The chronology of sexual and drug abuse milestones. Journal of Sex Research 1998;35:333-340 15.  van Ameijden EJC, van den Hoek J A R , Hartgers C , Coutinho R A . Risk factors for the transition from noninjection to injection drug use and accompanying A I D S risk behavior in a cohort of drug users. American Journal of Epidemiology 1994;139:1153-1163  16.  Bowler S, Sheon A R , D'Angelo LJ, V e r m u n d S H . H I V and AIDS among adolescents i n the United States: increasing risk in the  1990s. Journal of  Adolescence 1992;15:345-71 17.  H u b a GJ, Melchior L A , Panter A T , Trevithick L , Woods ER, Wright E, Feudo R, Tierney S, Schneir A , Tenner A , Remafedi G , Greenberg B, Sturdevant M , Goodman E, Hodgins A , Wallace M , Brady R E , Singer B. Risk factors  and  characteristics of youth living with, or at high risk for, H I V . AIDS Education & Prevention 2000;12:557-75 18.  Glod C A . Long-term consequences of childhood physical and sexual abuse. Archives of Psychiatric Nursing 1993;8:163-173  19.  Statistics Canada . 1996: Aboriginal data. Ottawa: Statistics Canada, 1998:1 -13  20.  Health Canada. H I V / A I D S A m o n g Aboriginal Persons i n Canada Remains a Pressing Issue. Ottawa: Centre for Infectious Disease Prevention and Control, 2001:1-8  21.  Network C A A L . Discrimination, HrV/AIDS and Aboriginal People. Montreal: The Canadian Aboriginal A I D S Network, 1999  22.  Force B A H A T . The Red Road: Pathways to Wholeness: A n Aboriginal Strategy for H I V and A I D S i n BC. Vancouver, 1998  23.  Alter M J . Hepatitis C virus infection in the United States. Journal of Hepatology 1999;31:88-91  24.  Alter M J , Moyer L A . The importance of preventing hepatitis C virus infection among injection drug users in the United States. Journal of Acquired Immune Deficiency Syndromes 1998;18:S6-10  25.  C D C . Recommendations for Prevention and Control of Infection and H C V Related Chronic Disease. Atlanta: Centers for Disease Control, 1998  26.  Patrick D M , Tyndall M W , Cornelisse P G , L i K , Sherlock C H , Rekart M L , Strathdee SA, Currie SL, Schechter M T , O'Shaughnessy  M V . Incidence of  hepatitis C virus infection among injection drug users during an outbreak of H I V infection. C M A J (Canadian Medical Association Journal) 2001;165:889-95 27.  Garfein RS, Doherty M C , Monterroso ER, Thomas D L , Nelson K E , Vlahov D . Prevalence and incidence of hepatitis C virus infection among young adult injection drug users. JAIDS: Journal of Acquired Immune Deficiency Syndromes 1998;18:Sll-9  28.  MacDonald M A , Wodak A D , Dolan K A , van Beek I, Cunningham P H , Kaldor J M . Hepatitis C virus antibody prevalence among injecting drug users at selected 19  needle and syringe programs i n Australia, 1995-1997. Collaboration of Australian NSPs. Medical Journal of Australia 2000;172:57-61 29.  Thorpe L E , Ouellet LJ, Levy JR, Williams IT, Monterroso ER. Hepatitis C virus infection: prevalence, risk factors, and prevention opportunities among y o u n g injection drug users in Chicago, 1997-1999. Journal of Infectious  Diseases  2000;182:1588-94 30.  Hagan H , Thiede H , Weiss N S , Hopkins SG, Duchin JS, Alexander ER. Sharing of drug preparation equipment as a risk factor for hepatitis C . American Journal of Public Health 2001;91:42-6  31.  Thorpe L , Ouellet L, Hershow R, Bailey S, Williams I, Williamson J, Monterroso E, Garfein R. Risk of Hepatitis C Virus Infection among Y o u n g A d u l t Injection D r u g Users W h o Share Injection Equipment. American Journal of Epidemiology 2002;155:645 - 653  32.  B C C D C . Hepatitis i n British Columbia. Vancouver: B C C D C , 2001:1 - 2  33.  C D C . Recommendations for Prevention and Control of H C V Infection and HCV-Related Chronic Disease. Atlanta: Centers for Disease Control, 2002  34.  Diaz T, Des Jarlais D C , Vlahov D , Perlis TE, Edwards V , Friedman SR, Rockwell R, Hoover D , Williams IT, Monterroso ER. Factors associated w i t h prevalent hepatitis C : differences among young adult injection drug users i n lower and upper Manhattan, N e w York City. American Journal of Public Health 2001;91:2330  35.  H a h n JA, Page-Shafer K , L u m PJ, Ochoa K , Moss A R . Hepatitis C virus infection and needle exchange use among young injection drug users i n San Francisco. Hepatology 2001;34:180-7  20  36.  U N I C E F . The Progress of Nations 1999. N e w York: United Nations Children's Fund, 1999  37.  H e i n K , Dell R, Futterman D , Rotheram-Borus M J , Shaffer N . Comparison of HIV+ and H I V - adolescents:  risk factors  and  psychosocial  determinants.  Pediatrics 1995;95:96-104 38.  Fennema JS, V a n Ameijden EJ, V a n Den Hoek A , Coutinho R A . Y o u n g and recent-onset injecting drug users are  at higher risk for  H I V . Addiction  1997;92:1457-65 39.  DiClemente RJ. The emergence of adolescents as a risk group for  human  immunodeficiency virus infection. Journal of Adolescent Research 1990;5:7-17 40.  Thomas P A , Weisfuse IB, Greenberg A E , Bernard G A , Tytun A , Stellman SD. Trends in the first ten years of AIDS in N e w York City. The N e w York C i t y Department  of  Health  AIDS  Surveillance  Team.  American Journal  of  Epidemiology 1993;137:121-33. 41.  Clements K , Gleghorn A , Garcia D , Katz M , Marx R. A risk profile of street youth in  Northern  California:  Implications  for  gender-specific  human  immunodeficiency virus prevention. Journal of Adolescent Health 1997;20:343353 42.  Doherty M C , Garfein RS, Monterroso E, Brown D , Vlahov D . Correlates of H I V infection among young adult short-term injection drug users. AIDS 2000;14:71726  43.  Thorpe L E , Bailey SL, H u o D , Monterroso ER, Ouellet LJ. Injection-related risk behaviors i n young urban and suburban injection drug users in Chicago (19971999). Journal of Acquired Immune Deficiency Syndromes 2001;27:71-8  21  44.  C D C . C D C Fact Sheet: Y o u n g People at Risk. U S A : Centers for Disease Control and Prevention, 2000  45.  Latkin C A . Comments on "The moral economies of homeless heroin addicts: confronting ethnography, H I V risk, and everyday violence in San Francisco shooting encampments,"  by Phillippe Bourgois. Substance Use & Misuse  1998;33:2375-82 46.  van Ameijden EJ, van den Hoek A R , Coutinho R A . Injecting risk behavior among drug users i n Amsterdam, 1986 to 1992, and its relationship to AIDS prevention programs. American Journal of Public Health 1994;84:275-81.  22  CHAPTER 2 REVIEW OF T H E LITERATURE  2.1  Author's note  The following is a brief overview of the H I V and H C V literature as it pertains to this thesis topic.  A more specific review of relevant literature has been done in the  introduction to the four stand-alone manuscripts (chapters 3 through 6).  2.1  Introduction A n estimated 11 million people aged 15 - 24 are living with H I V / A I D S , and one  half of all new infections, over 7,000 daily, are occurring among y o u n g people. SubSaharan Africa is the hardest hit, where over 70% of the young people living with H I V / A I D S reside[l]. Youth are at the centre of the epidemic, they are the generation coming of age i n a w o r l d where AIDS is robbing individuals, families, communities and nations of human security. M a n y youth are at increased risk for H I V infection for a number of reasons beyond their control.  Youth may not have access to H I V  education or condoms. Y o u n g people, especially girls, may be the victims of sexual abuse. Furthermore, there are an estimated 100 million youth under age 18 living or working on the streets increasing their risk for sex trade work, injection drug use, and vulnerable relationships[l]. Young people at the heart of the H I V epidemic in Sub-Saharan Africa are largely vulnerable to infection through sexual transmission[2]. In N o r t h America, risk factors for transmission among young people are split about equally between heterosexual contact, men w h o have sex with men (MSM) and IDUs[3-6] . The links  23  between IDUs and heterosexual transmission of the H I V virus through sexual contact between I D U and non-IDU have been documented [7, 8] .  Youth, especially street  youth, may be particularly at risk for heterosexual transmission through sex with an I D U [ 9 ] . H I V is systemically patterned so as to render some young people more likely to become infected than others.  Gender, ethnicity, socio-economic status, sexuality,  and age are important factors i n structuring vulnerability to infection. In N o r t h America, since 1996, w o m e n account for 20% - 35% of H I V infections[8, 10] .  The number of w o m e n infected per year has been steadily  increasing since 1995. Infection rates among females vary greatly by age, the highest proportion of females infected are among female youth[10] . Since 1999 females have accounted for between 40% and 50% of new H I V positive tests among youth.  IDUs  accounts for approximately 35% to 45% of H I V infections among w o m e n overall and in some studies of IDUs, females are at greater risk for infection than m a l e s [ l l , 12] . Increasing proportions of young females becoming H I V infected are concerning and may have implications on their life expectancy and also on their children's quality of life. IDUs i n N o r t h America continue to contribute significantly to the H I V epidemic. In Canada in 1999, 34% of the new infections were due to IDUs, a drop from 1996 wherein 47% of new infections were attributable to IDUs[3,13]. The drop i n infections is likely due to expanded needle exchange programs (NEPs), however high H I V incidence rates among some groups of IDUs such as w o m e n and youth indicate that other interventions are clearly required.  In the United States, IDUs accounts for  approximately one quarter of new infections per year[14] .  M a n y IDUs are also  involved i n sex trade work and may engage i n unprotected sex w i t h clients[7,15, 16] . Furthermore among IDUs condom use w i t h casual and regular sexual partners is low[17]. 24  The major route of Hepatitis C ( H C V ) infection i n N o r t h America is sharing drug equipment between IDUs[18-21] .  Cohort investigations of H C V infection  among IDUs suggest that H C V is widespread, up to 90% of IDUs are infected and incidence rates are high, ranging between 11 to 29 per 100 person years[22-28] . Time to infection from first initiating injection drug use is estimated to range between 1 and 4 y e a r s [ l l , 24, 29, 30]. H C V prevalence estimates among y o u n g IDUs range between 25% and 65%, greater age and number of years injecting are important predictors of H C V positivity[26, 31, 32]. While sexual transmission of H C V remains a controversial topic, it has been postulated that there is risk of lateral transmission of H C V between sexual partners [33, 34] . Furthermore, risk factors for H C V may provide an indication of the potential for the spread of H I V among IDUs. 2.2 HIV prevalence and incidence and associated risk factors H I V seroprevalence among IDUs i n N o r t h  America varies markedly b y  geographic region, and by proximity to major urban centres. range from 12% to 41% [35-41] .  Prevalence estimates  H I V seroprevalence among I D U i n Vancouver  increased from 4% i n 1992-93 to 23% i n 1996-97 [42] and again to 30% in 1998 [35] . A s of A p r i l 2002, H I V seroprevalence among all V I D U S participants is approximately 32%. A t the beginning of the V I D U S study (May to November, 1997) seroincidence of H I V infection was 19.4 per 100 person years, but has subsequently declined to 2-5 per 100 person years after the first year of the study[11,17]. There have been few studies specifically documenting H I V seroprevalence in younger IDUs. Results from I D U cohort studies have suggested that younger IDUs have an increased risk ratio for H I V (range: 2.2 - 3.3)[43-45] . In a study that looked at H I V seroprevalence among IDUs aged 16 - 24 i n Puerto Rico, prevalence was 24%[46] . In the Puerto Rican sample risk factors associated with H I V were increased age, initiating injection at a younger age, lower level of education, having a history of 25  incarceration, and reporting a history of sexually transmitted infections. Another study from the United States found H I V prevalence among young I D U aged 13 to 29 to be 14.4% [47] .  In this later study factors  most  strongly  associated  with H I V  seroprevalence were trading anal sex, injecting cocaine or speedballs, smoking crack daily, and having two or more "trainers" before self-injection. In Canada, among Winnipeg IDUs, HIV-prevalence rates for youth between the ages of 20 and 24 years was 10.5% [48] . A Vancouver study of female and male y o u n g offenders aged 12 to 19 years of age revealed an H I V prevalence rate of 0.25% [49] . Results from the Enhanced Surveillance of Canadian Street Youth Study showed that overall 20% of participants had ever injected drugs and i n the same study, of street youth i n Vancouver, 35% reported ever injecting drugs[3]. A m o n g street youth in San Francisco aged 19 years and younger, 43% reported ever injecting drugs[50] . Several investigations of street youth have found IDUs to be significantly associated with H I V seroprevalence[9, 51-53]. The links between homeless youth and injection drug use are clear, and intervention programs for street youth that focus on reducing the incidence of IDUs or educating youth on safe injection practices may help prevent blood-borne viruses. The centres for disease control in the United States and Canada have both indicated that w o m e n are at increased risk for H I V infection among IDUs [8, 54] . In Canada an estimated 54% of new H I V infections among w o m e n were attributable to IDUs in 1999.  In the United States during 1999, 42% of AIDS diagnoses were  attributable to IDUs among women. A m o n g V I D U S participants, higher proportions of females were baseline H I V prevalent than males, 35% vs. 26% respectively[11, 17] . Several studies have found that younger female IDUs are at greater risk for H I V infection[44,45, 55].  26  Some of the excess risk for H I V among w o m e n could be attributed to sexual behaviours other than injection risk, in part because HTV infection is more effective from male to female than it is from female to male through sexual contact[16, 56] . Furthermore, females are more likely than males to engage i n survival sex [57-59] , to report multiple sex partners and to have partners w h o are injection drug users and use condoms less frequently [48, 57, 60] . Females may be more likely than males to be influenced or coerced into engaging in behaviour such as high-risk sexual practices or risky injection behaviour than are males [61-65] . Another important group w i t h respect to the H I V epidemic identified by the centres for disease control in the United States and Canada are Amer-Indian Nations and Aboriginal people[66, 67] .  Governmental policies directed at Canadian Aboriginal  people such as removal from land to reserve and the residential school system have precipitated the loss of cultural and linguistic identity among some Aboriginal Nations. These policies have resulted in disparities between Aboriginal and non-Aboriginal Canadians i n terms of health and economic status [68-72] . Consequentially, Aboriginal people are over-represented among the I D U population and among Canadians infected with H I V . A m o n g V I D U S participants, Aboriginal people account for 25% and have been found to be at increased risk for H I V seroprevalence[17].  2.3 H C V prevalence and incidence and associated risk factors It has been suggested that the high transmission rate of H C V among IDUs is due to a combination of factors including high infectiousness per syringe-sharing contact, infectiousness throughout the course of the disease, and the frequency of needle and equipment sharing contacts i n a given community [73] . Researchers have estimated that parenteral transmission of H C V is 10-fold more efficient than that of HIV[74] .  27  IDUs are the leading risk category for H C V infection;. Prevalence estimates of H C V infection among IDUs range from 38-95% [24] [31] [75] [76] . In the V I D U S cohort, overall H C V prevalence is 87%. Approximately 85% of persons infected w i t h H C V develop persistent viremia and this generates a large reservoir of infected IDUs with potential for spreading the virus. There are approximately 40,000 cases of H C V in British Columbia, four times the national average[34]. Injecting w i t h someone else's potentially contaminated needle and syringe ("needle borrowing") is the principal route of transmission, however cohort studies of IDUs i n Chicago and Seattle noted that sharing of other injection equipment including cottons, cookers, and straws also facilitated parenteral transmission of the H C V virus[74, 77] . Studies of y o u n g IDUs have determined that time to H C V infection following initiation into injection drug use is 1 - 4 years [23, 29, 30, 78]. Other studies have also suggested that young IDUs are more likely to borrow needles and share injection equipment than older IDUs[47, 79]. M a n y youth w h o inject drugs also engage i n risky sexual activity[51, 80, 81]. Therefore i n addition to parenteral risk, y o u n g IDUs may engage i n sexual behaviour that has been postulated as risk factors for HCV[82] . Thus young IDUs are an important group w i t h respect to the virus. Consistent factors associated with H C V seroprevalence among IDUs include at least daily cocaine and speedball injection[26, 31, 83] , sharing needles[ll, 22] , and sharing other injection equipment[74, 77]. Some studies have found female gender and sex trade work to be factors associated with H C V seroincident infection[78, 83] . In other studies of H C V transmission, associations have included high numbers of sexual partners and l o w condom use[26, 75]. In an earlier study of the V I D U S cohort, overall H C V seropositivity at baseline occurred among 82% of the cohort and was associated with older age, incarceration, higher number of sexual partners, greater number of years injecting, sex trade work, and borrowing syringes [83] . In this earlier study, the 28  overall incidence density rate was 29 per 100 person-years and was associated with female gender, Aboriginal ethnicity, participants engaged i n sex trade w o r k , b o r r o w i n g needles, and at least daily cocaine or speedball injection. 2.4 Aboriginal People and the HIV epidemic The annual proportion of AIDS cases among Canadian Aboriginal people has risen from less than 1% before 1990 to 10% i n 1999[67] .  Aboriginal people are over-  represented among new H I V infections (i.e.19% in 1998, 24% in 1999, 22% i n 2000 and 26% in 2001). Females represent one half of all positive H I V tests among Aboriginal people but only 20% among non-Aboriginal people. Aboriginal people are younger when diagnosed w i t h H I V than non-Aboriginal people; 31% vs. 20% are among the age group 20 to 29.  H I V infected Aboriginal people are more likely to be in the  transmission group IDUs than non-Aboriginal people, 60% vs. 29%. These figures are worrisome especially because current surveillance data are incomplete, the majority of positive H I V tests reported to the Centres for Disease Control have no ethnic information[3] . Aboriginal people are the fastest growing population in Canada, w i t h 32% of the population under the age of 14 [84] . A Vancouver study conducted on incarcerated youth, researchers noted that Aboriginal youth were five times more likely to have engaged in injection drug use than non-Aboriginal youth [49].  There is little data  specifically on Aboriginal youth, however available data suggest that Aboriginal persons are infected w i t h H I V at a younger age than non-Aboriginal persons, that injection drug use is the most important mode of transmission, and that the epidemic shows no sign of abating[67] .  2.5  Conclusion 29  The future of the H I V epidemic is i n the hands of young people. M a n y of the behaviours they adopt now w i l l be those they will maintain throughout the course of their lives and w i l l determine the course of the epidemic for years to come. There are many examples of successful interventions that show when young people  are  provided w i t h effective education, access to condoms and sterile injecting equipment it positively affects the H I V epidemic. Learning to better protect children and youth from sexual victimization, poverty, and the streets will be the greatest challenge, however it w i l l be necessary to decrease transmission of the H I V virus and to protect youth from the H I V epidemic sweeping the globe.  30  2.6 References cited 1.  U N A I D S . Children and young people in a w o r l d of AIDS. Geneva: U N A I D S , 2001:1 - 13  2.  U N A I D S . A I D S epidemic update. Geneva: U N A I D S , 2001:1 - 29  3.  Health Canada. H I V / A I D S Epi Update. Ottawa: Laboratory Centres for Disease Control, 2002:1 - 76  4.  Health Canada. H I V and AIDS A m o n g Youth in Canada. Ottawa: Laboratory Centre for Disease control, 2000:1 - 4  5.  C D C . Youth Risk Behaviour Surveillance System: Centres for Disease Control and Prevention, 1998  6.  C D C . C D C Fact Sheet: Y o u n g People at Risk. U S A : Centers for Disease C o n t r o l and Prevention, 2000  7.  Chiasson M A , Stoneburner R L , Hildebrandt DS, E w i n g W E , Telzak EE, Jaffe H W . Heterosexual transmission of HIV-1 associated w i t h the use of  smokable  freebase cocaine (crack). A I D S 1991;5:1121-6 8.  C D C . H I V / A I D S among U S women: minority and young w o m e n at continuing risk: Centers for Disease Control, 2001  9.  Roy E, Lemire N , Haley N , Boivin J-F, Frappier J-Y, Claessens C . Injection drug use among street youth: A dynamic process. Canadian Journal of Public Health 1998;89:239-240  10.  Health Canada. H I V and A I D S A m o n g W o m e n in Canada: Centre for Infectious Disease Control and Prevention, 2000 31  11.  Strathdee S A , Patrick D M , Currie SL, Cornelisse P G , Rekart M L , Montaner JS, Schechter M T , O'Shaughnessy M V . Needle exchange is not enough: lessons from the Vancouver injecting drug use study. A I D S 1997;ll:F59-65  12.  Strathdee S A , Galai N , Safaiean M , Celentano D D , Vlahov D , Johnson L , Nelson K E . Sex differences i n risk factors for hiv seroconversion among injection drug users: a 10-year perspective. Archives of Internal Medicine 2001;161:1281-8  13.  Health Canada H A E U . H I V / A I D S A m o n g Injection D r u g Users in Canada. Ottawa: Bureau of H I V / A I D S , STD and TB Update Series, Laboratory Centre for Disease Control, 1999  14.  C D C . A Glance at the H I V Epidemic. Atlanta: C D C , 2001:1 - 3  15.  A k l i l u M , Messele T, Tsegaye A , Biru T, M a r i a m D H , van Ben them B, Coutinho R, Rinke de Wit T, Fontanet A . Factors associated w i t h HIV-1 infection among sex workers of A d d i s Ababa, Ethiopia. A I D S 2001;15:87-96  16.  van Ameijden EJ, v a n den Hoek A J , van Haastrecht H J , Coutinho R A . Trends in sexual behavior and the incidence of sexually transmitted diseases and H I V among drug-using prostitutes, Amsterdam 1986-1992. A I D S 1994;8:213-21  17.  Spittal P, Craib K , W o o d E, Laliberte N , L i K , Tyndall M , O'Shaughnessy M , Schechter M . Risk factors for elevated H I V incidence rates among injection drug users i n Vancouver. C M A J 2002;166:894 - 899  18.  Alter M J . Epidemiology of hepatitis C . Hepatology 1997;26:62S-65S  32  female  19.  Alter M J , Kruszon-Moran D , Nainan O V , McQuillan G M , Gao F, M o y e r L A , Kaslow R A , Margolis H S . The prevalence of hepatitis C virus infection i n the United  States,  1988  through  1994. N e w  England Journal  of  Medicine  1999;341:556-62 20.  C D C . Recommendations for Prevention and Control of Infection and H C V Related Chronic Disease. Atlanta: Centers for Disease Control, 1998  21.  Crofts N . G o i n g where the epidemic is. Epidemiology and control of hepatitis C among injecting drug users. Australian Family Physician 2001;30:420-5  22.  Brunton C , K e m p R, Raynel P, Harte D , Baker M . Cumulative incidence of hepatitis C seroconversion i n a cohort of seronegative injecting drug users. N e w Zealand Medical Journal 2000;113:98-101  23.  Alter M J , Moyer L A . The importance of preventing hepatitis C virus infection among injection drug users in the United States. Journal of Acquired Immune Deficiency Syndromes 1998;18:S6-10  24.  Galeazzi B, Tufano A , Barbierato E, Bortolotti F. Hepatitis C virus infection in Italian intravenous  drug users: epidemiological and clinical aspects.  Liver  1995;15:209-12 25.  Gombas W , Fischer G , Jagsch R, Eder H , Okamoto I, Schindler S, Muller C , Ferenci P, Kasper S. Prevalence and distribution of hepatitis C subtypes in patients w i t h opioid dependence. European Addiction Research 2000;6:198-204  26.  H a h n JA, Page-Shafer K , L u m PJ, Ochoa K , Moss A R . Hepatitis C virus infection and needle exchange use among young injection drug users i n San Francisco. Hepatology 2001;34:180-7  27.  MacDonald M A , Wodak A D , Dolan K A , van Beek I, Cunningham P H , Kaldor J M . Hepatitis C virus antibody prevalence among injecting drug users at selected needle and syringe programs i n Australia, 1995-1997. Collaboration of Australian NSPs. Medical Journal of Australia 2000;172:57-61  28.  Williams I. Epidemiology of hepatitis C i n the United States. American Journal of Medicine 1999;107:2S-9S  29.  Taylor A , Goldberg D , Hutchinson S, Cameron S, Gore S M , M c M e n a m i n J, Green S, Pithie A , Fox R. Prevalence of hepatitis C virus infection among injecting drug users in Glasgow 1990-1996: are current harm reduction strategies working? Journal of Infection 2000;40:176-83  30.  Hagan H , M c G o u g h JP, Thiede H , Weiss N S , H o p k i n s S, Alexander ER. Syringe exchange and risk of infection with hepatitis B and C viruses. American Journal of Epidemiology 1999;149:203-13  31.  Garfein RS, Doherty M C , Monterroso ER, Thomas D L , Nelson K E , Vlahov D . Prevalence and incidence of hepatitis C virus infection among young adult injection drug users. JAIDS: Journal of Acquired Immune Deficiency Syndromes 1998;18:Sll-9  32.  Ogilvie E L , Veit F, Crofts N , Thompson SC. Hepatitis infection  among  adolescents resident in Melbourne Juvenile Justice Centre: risk factors  and  challenges. Journal of Adolescent Health 1999;25:46-51 33.  C D C . Recommendations for Prevention and Control of H C V Infection and HCV-Related Chronic Disease. Atlanta: Center for Disease Control, 2002  34.  C D C Hepatitis i n British Columbia. Vancouver: B C C D C , 2001:1-2  35.  Schechter M T , Strathdee S A , Cornelisse P G , Currie S, Patrick D M , Rekart M L , O'Shaughnessy M V . D o needle exchange programmes increase the spread of H I V among injection drug users?: an investigation of the Vancouver outbreak. A I D S 1999;13:F45-51  36.  Strathdee S A , van Ameijden EJ, Mesquita F, Wodak A , Rana S, Vlahov D . C a n H I V epidemics among injection drug users be prevented? A i d s 1998;12:S71-9  37.  V a n Ameijden EJ, Langendam M W , Notenboom J, Coutinho R A . Continuing injecting risk behaviour: results from the Amsterdam Cohort Study of drug users. Addiction 1999;94:1051-61  38.  Patrick D M , Strathdee S A , Archibald C P , Ofner M , Craib K J , Cornelisse P G , Schechter  M T , Rekart  M L , O'Shaughnessy  M V . Determinants  of  HIV  seroconversion i n injection drug users during a period of rising prevalence in Vancouver. International Journal of STD & A I D S 1997;8:437-45 39.  Bruneau J, Lamothe F, Franco E , Lachance N , Desy M , Soto J, Vincelette J. H i g h rates of H I V infection among injection drug users participating in needle exchange programs in Montreal: results of a cohort study. American Journal of Epidemiology 1997;146:994-1002  40.  Moss A R , Vranizan K . Charting the epidemic: the case study of H I V screening of injecting drug users i n San Francisco, 1985-1990. British Journal of Addiction 1992;87:467-71  41.  Coates TJ, Aggleton P, Gutzwiller F, Des Jarlais D , Kihara M , Kippax S, Schechter M , van den  Hoek  JA. H I V prevention  1996;348:1143-8  35  in developed  countries.  Lancet  42.  Health Canada H A E U . Risk Behaviour A m o n g Injection D r u g Users in Canada. Ottawa: Bureau of H I V / A I D S , STD, and TB Update Series, Laboratory Centre for Disease Control, 1999  43.  Nelson K E , Vlahov D , Solomon L , C o h n S, M u n o z A . Temporal trends of incident human immunodeficiency virus infection i n a cohort of injecting drug users i n Baltimore, M d . Archives of Internal Medicine 1995;155:1305-11  44.  Zaccarelli M , Gattari P, Spizzichino L , Portaleone  A , Venzia S, Rezza G .  Understanding the Dynamics of H I V infection among injection drug users in Rome in the 1990's: the combined use of cross-sectional, longitudinal and behavioural data. In: XI International Conference on A I D S . Vancouver, July 1996: abstract Mo.C.461,1996 45.  Fennema JS, V a n Ameijden EJ, V a n Den Hoek A , Coutinho R A . Y o u n g and recent-onset injecting drug users are  at higher  risk for  H I V . Addiction  1997;92:1457-65 46.  Marrero Rodriguez C A , Robles RR, C o l o n H M , Freeman D H , Matos T D , Reyes JC. H I V risk behaviors and H I V seropositivity among young injection drug users. Puerto Rico Health Sciences Journal 1993;12:7-12  47.  Doherty M C , Garfein RS, Monterroso E, Brown D , Vlahov D . Correlates of H I V infection among young adult short-term injection drug users. Aids 2000;14:71726  48.  Health Canada H A E U . H I V and AIDS A m o n g Youth in Canada. Ottawa: Population and Public Health Branch, Health Canada, 2000  36  49.  Rothon D A , Strathdee S A , C o o k D , Cornelisse P G . Determinants of HlV-related high risk behaviours  among young offenders:  a w i n d o w of  Canadian Journal of Public Health. Revue Canadienne  opportunity.  de Sante Publique  1997;88:14-7 50.  Krai A H , Molnar BE, Booth R E , Watters JK. Prevalence of sexual risk behaviour and substance use among runaway and homeless adolescents in San Francisco, Denver and N e w York City. International Journal of STD & A I D S 1997;8:109-117  51.  Caffaro A , M a h JK, Lang R A , Joffres M R . Prevalence of sexually transmitted diseases in juvenile prostitutes and street youth. Canadian Journal of Infectious Diseases 1994;5:21-27  52.  van den Hoek A . STD control in drug users and street youth. Genitourinary Medicine 1997;73:240-244  53.  Clements K , Gleghorn A , Garcia D , Katz M , Marx R. A risk profile of street youth in  Northern  California:  Implications  for  gender-specific  human  immunodeficiency virus prevention. Journal of Adolescent Health 1997;20:343353 54.  Health Canada. H I V and AIDS A m o n g W o m e n i n Canada. Ottawa: Laboratory Centre for Disease Control, 2000:1 - 3  55.  Levine OS, Vlahov D , Brookmeyer R, C o h n S, Nelson K E . Differences i n the incidence of hepatitis B and human immunodeficiency virus infections among injecting drug users. Journal of Infectious Diseases 1996;173:579-83  37  56.  Padian N S , Shiboski S C , Glass SO, Vittinghoff E. Heterosexual transmission of human immunodeficiency virus (HrV) i n northern California: results from a tenyear study. American Journal of Epidemiology 1997;146:350-7  57.  Futterman D , H e i n K , Reuben N , Dell R, Shaffer N . H u m a n immunodeficiency virus-infected adolescents: the first 50 patients i n a N e w York City program. Pediatrics 1993;91:730-5  58.  Greene J M , Ennett ST, Ringwalt C L . Prevalence and correlates of survival sex among runaway and homeless youth. American Journal of Public Health 1999;89:1406-1409  59.  H e i n K , Dell R, Futterman D , Rotheram-Borus M J , Shaffer N . Comparison of HIV+ and H I V - adolescents:  risk factors  and  psychosocial  determinants.  Pediatrics 1995;95:96-104 60.  Krai A H , Lorvick J, Edlin BR. Sex- and drug-related risk among populations of younger and older injection drug users in adjacent neighborhoods  in San  Francisco. Journal of Acquired Immune Deficiency Syndromes 2000;24:162-7 61.  Beitchman J H , Zucker K J , H o o d JE, daCoasta G A , A k m a n D , Cassavia E. A review of the long-term effects of child sexual abuse. C h i l d Abuse and Neglect 1992;16:101-118  62.  Denenberg R. Childhood sexual abuse as an H I V risk factor i n w o m e n . Treatment Issues 1997;11:22-24  63.  G l o d C A . Long-term consequences of childhood physical and sexual abuse. Archives of Psychiatric N u r s i n g 1993;8:163-173  38  64.  Holland J, Scott S, Thompson R. Sex, gender and power: young  women's  sexuality i n the shadow of AIDS. Sociology of Health and Illness 1990;12:336 - 50 65.  Latkin C A , Mandell W , Knowlton A R , Doherty M C , Vlahov D , Suh T, Celentano D D . Gender differences i n injection-related behaviors among injection drug users i n Baltimore, Maryland. A I D S Education & Prevention 1998;10:257-63  66.  C D C . Surgeon General Calls for Action on H I V / A I D S Crisis i n American Indian and Alaska Native Communities: Centers for Disease Control, 2000  67.  Health Canada. H I V / A I D S A m o n g Aboriginal Persons in Canada Remains a Pressing Issue. Ottawa: Centre for Infectious Disease Prevention and Control, 2001:1 -8  68.  Coulehan JL, Arena V , Welty T. Determinants of Blood Pressure in Navajo Adolescents. American Indian and Alaskan Native Mental Health Resource .1990;3:27 - 36  69.  Culhane Speck D . A n Error in Judgement: Medical Care in an Indian/White Community: Talon Books, 1987  70.  Fisher A . Alcoholism and Race: the Misapplication of Both Concepts to N o r t h American Indians. Canadian Review of Sociology and Anthropology 1987;24:81 98  71.  Hagey N J , McBride C . Highlights of Aboriginal Conditions 1981 - 2001: Part 1Demographic Trends; Part 2-Social Conditions; Part 3-Economic Conditions. Ottawa: Department of Indian Affairs and Northern Development, 1989  72.  Culhane Speck D . The Pleasure of the C r o w n : Anthropology, L a w and First Nations: Talon Books, 1998 39  73.  Crofts N , Aitken C K , Kaldor J M . The force of numbers: w h y hepatitis C is spreading among Australian injecting drug users while H I V is not. Medical Journal of Australia 1999;170:220-1  74.  Thorpe L , Ouellet L, Hershow R, Bailey S, Williams I, Williamson J, Monterroso E, Garfein R. Risk of Hepatitis C virus infection among young adult injection drug users who share injection equipment.  American Journal of Epidemiology  2002;155:645 - 653 75.  Romanowski  B,  Campbell  PJ,  Preiksaitis  JK,  Fonseca  K.  Human  immunodeficiency virus seroprevalence and risk behaviors in patients attending sexually transmitted disease clinics in Alberta. Sexually Transmitted Diseases 1997;24:487-94 76.  Krook A , Albert J, Andersson S, Biberfeld G , Blomberg J, Eklund I, Engstrom A , Julander I, K a i l K , Martin C , Stendahl P, Struve J, Sonnerborg A . Prevalence and risk factors for H T L V - I I infection in 913 injecting drug users in Stockholm, 1994. JAIDS: Journal of Acquired Immune Deficiency Syndromes 1997;15:381-6  77.  Hagan H , Thiede H , Weiss N S , Hopkins SG, Duchin JS, Alexander ER. Sharing of drug preparation equipment as a risk factor for hepatitis C . American Journal of Public Health 2001;91:42-6  78.  van den Hoek JA, van Haastrecht H J , Goudsmit J, de Wolf F, Coutinho R A . Prevalence, incidence, and risk factors of hepatitis C virus infection among drug users i n Amsterdam. Journal of Infectious Diseases 1990;162:823-6  79.  Friedman S, Desjarlais D . Risk Factors for H I V infection among new drug injectors. In: IX International Conference on AIDS. Berlin, 1993  40  80.  Roy E, Haley N , Leclerc P, Boivin JF, Cedras L , Vincelette J. Risk factors for hepatitis C virus infection among street youths. C M A J 2001;165:557-60  81.  Clements K , Gleghorn A , Garcia D , Katz M , Marx R. A risk profile of street youth in northern California: implications for gender-specific human immunodeficiency virus prevention. Journal of Adolescent Health 1997;20:343-53  82.  Conry-Cantilena C , VanRaden M , Gibble J, Melpolder J, Shakil A O , Viladomiu L , Cheung L , DiBisceglie A , Hoofnagle J, Shih JW. Routes of infection, viremia, and liver disease in blood donors found to have hepatitis C virus infection. N e w England Journal of Medicine 1996;334:1691-6  83.  Patrick D M , Tyndall M W , Cornelisse P G , L i K , Sherlock C H , Rekart M L , Strathdee S A , Currie SL, Schechter  M T , O'Shaughnessy  M V . Incidence  of  hepatitis C virus infection among injection drug users during an outbreak of H I V infection. C M A J 2001;165:889-95  84.  Canadian AIDS Legal Network. Discrimination, H I V / A I D S and Aboriginal People. Montreal: The Canadian Aboriginal A I D S Network, 1999  41  CHAPTER 3 HIV PREVALENCE A N D ASSOCIATED RISK FACTORS  3.1 FOREWORD This manuscript has been accepted for publication as "Females experiencing sexual and drug related vulnerabilities are at elevated risk for H I V infection among youth w h o use injection drugs" w i t h the Journal of Acquired Immune Deficiency Syndromes, July 2002, Vol.30, No.3. 3.2  Introduction Youth are of particular importance with respect to H I V / A I D S , not only because  they are at risk for infection, but it is during this period of life w h e n many behaviour patterns are established that will affect their future risk of H I V infection [1]. The H I V epidemic among youth is well established and growing[2-4].  The United Nations  estimates there are 6,400,000 females and 3,900,000 males aged 15 - 24 living with H I V around the globe[5]. In Canada, age at infection has decreased from age 32 before 1983 to 23 i n 1990[6], and in the United States rates of H I V infection among adolescent women have tripled between 1985 and 1999[7]. Youth at higher risk for H I V infection are those w h o use injection drugs[l, 8]. Some cohort studies of injection drug users (IDUs) have reported that younger age and more recent initiation into injection are associated with increased risk for HIV[9-11]. This is troubling, particularly i n light of recent school-based surveys suggesting that between 3% and 9% of high school students report use of injection drugs[12, 13]. Other studies have found that young female IDUs are particularly vulnerable to H I V infection, likely due to sexual and drug related risk factors[3,14]. We undertook this study to characterize socio-demographic characteristics and sexual and drug risk factors among young IDUs in a city where an ongoing H I V -  42  epidemic among IDUs has occurred[15].  The objectives were to determine socio-  demographic, drug and sexual differences between younger and older IDUs and to investigate risk factors for H I V infection among the young injectors.  3.3 Methods Data were collected within a prospective open cohort study of injection drug users, the Vancouver Injection D r u g Users Study (VIDUS). A description of this study has been previously published[15]. V I D U S has recruited over 1,400 Vancouver area IDUs since M a y 1996. The study office is located in the D o w n t o w n Eastside (DTES) of Vancouver. The DTES is Vancouver's poorest neighbourhood and many of the city's IDUs reside in an area of approximately ten city blocks where inexpensive housing in the form of hotels and single room occupancies (SROs), abound. Eligibility criteria for the study included residing i n the city of Vancouver and surrounding municipalities and having injected in the previous month. There were no age restrictions.  Participants were administered a questionnaire by trained nurse-  interviewers and were eligible to return for follow-up every six months. A t each visit, eligible participants were tested for H I V and H C V through venous blood samples. This analysis is based on data from the enrolment questionnaire, administered between M a y 1996 and November 2000.  This study was approved by the St. Paul's Hospital  Committee on H u m a n Experimentation. For the purposes of this study, young injectors were defined as those aged 24 or less at the time of recruitment.  Y o u n g and older injectors were compared using  contingency table analysis. Similarly, HIV-positive young and older participants and H I V positive and H I V negative young female injectors were compared to determine risk factors for HIV-positivity among the youth.  A l l risk factor variables refer to  behaviours and circumstance i n the 6 months prior to the questionnaire. 43  Selected  variables of interest are defined as follows: survival sex refers to trading sex for money drugs or shelter; condom use is always; help injecting is any; and sexual abuse is ever. Chi-square and Fisher's exact tests were used to compare categorical variables, whereas Wilcoxon rank sum tests were used to compare continuous variables.  Logistic  regression models were used to identify independent predictors of HlV-positivity among the young female IDUs. Forward stepwise regression was used involving all variables that were significant (p<0.10) i n bivariate analyses. Those significant at p<0.05 were retained i n the final model. A l l reported p-values are two-sided.  3.4  Results There were a total of 1,437 participants eligible for this analysis. A total of 232  were aged 13 - 24 years (median 21) and 1,205 were older than 24 years (median 36) at baseline. A s w o u l d be expected, the duration of injection career was much shorter in the young injectors compared with the older injectors (median 3 years vs 13 years; p<0.001). A g e at first injection was also much lower among the youth (median 17 vs. 20; p=<0.001). The prevalence of H I V at baseline was also lower in the young injectors (10.0% vs. 23.5%). The variables i n Table 1 are comparing baseline socio-demographic, sexual and drug risk variables between the young and older V I D U S participants. Y o u n g injectors were more likely to be female, engage in sex trade work, have casual sexual partners, report using condoms w i t h casual sexual partners, inject heroin daily, use crack cocaine daily, and require help w h e n injecting. Young injectors were less likely to have: any high school education, greater than 20 lifetime sexual partners, and inject cocaine on a daily basis.  There was no statistical difference between  the young and older  participants w i t h respect to being Aboriginal, having a history of incarceration,  44  injecting speedballs daily, needle sharing behaviours, and having a history of sexual abuse. Table 2 compares socio-demographic and risk variables between young and older V I D U S participants w h o were H I V positive at the baseline questionnaire.  As  w o u l d be expected, the median age between the younger and older HIV-positive participants differed (median 22 vs. 35; p=<0.001), as did the number of years injecting (median 7 vs. 13; p=<0.001), and age at first injection  (median 16 vs. 20; p=<0.001).  Y o u n g seropositive participants were more likely to: be female, w o r k in the sex trade, have experienced sexual abuse, have greater than 20 lifetime partners, inject heroin and speedballs at least daily, and use crack cocaine at least daily. Y o u n g participants were marginally more likely to be Aboriginal and less likely to be married or living common law. There was no difference between the two groups w i t h respect to: casual sexual partners; condom use w i t h casual partners; daily cocaine injection; requiring help to inject; and needle sharing. In a multivariate model (data not shown) y o u n g injectors were significantly more likely to be female and inject speed on an at least daily basis.  Of the 232 young injectors, 23 were H I V positive at baseline and 209 were H I V negative. Of the 23 seropositive youth, 20 (87%) were female.  Due to the very high  proportion of females affected and to control for collinearity, Table 3 compares risk factors between young HIV-positive and negative females.  HIV-positive y o u n g  females were older (median 22 vs. 20; p=<0.001), had a greater number of years injecting (median 6.5 vs. 2; p=<0.001), however, age at first injection did not differ (median 16 vs. 16; p=<0.213). The HIV-positive female youth were more likely to w o r k in the sex trade, have greater than 20 lifetime partners, and inject speedballs daily. H I V positive female youth were less likely to have a regular sexual partner, to always use condoms w i t h casual sexual partners, and marginally less likely to have casual sexual 45  partners. There was no statistical difference between the two groups w i t h respect to using condoms w i t h casual sexual partners, injecting heroin daily, and smoking crack at least daily. Table 4 presents the results of our logistic regression modeling for independent predictors of seropositivity among the young female injectors. The final model shows that increased age, injecting for a greater period of time, and daily speedball injection were independently associated w i t h HIV-positivity. H a v i n g a regular sexual partner was a protective factor among the young females. Finally, Table 5 presents a profile of the 23 (10%) HIV-positive youth mcluding socio-demographic, sexual and drug risk factors. This profile shows the highest burden of infection is in young females.  The three young males who were seropositive at  baseline were all Aboriginal, all of w h o m injected cocaine daily, and one of the y o u n g men engaged i n sex trade work. Most of the youth engaged i n polydrug use on a daily basis, including heroin and cocaine injection and i n some cases, speedballs and crack. A low proportion of the youth reported always using condoms w i t h regular and casual sexual partnerships but interestingly, many of the youth reported always using a condom w i t h clients. 3.5  Discussion A m o n g young injectors in Vancouver, female youth are at highest risk for H I V  infection. Typically i n I D U cohorts the ratio of females to males is approximately one third to two thirds[15-17]. between females and males.  However, among V I D U S youth the split is about 50% Another study of young IDUs, the R E A C H cohort in  Baltimore Maryland, had a similar ratio of young female and male participants[10]. Whether these findings reflect a selection bias or are indicative of young  females  entering into injection drug use earlier than their male counterparts, requires further investigation. 46  In our study we found the H I V baseline prevalence was much higher among young females than young males (17 vs. 3). Doherty et al. also found higher baseline H I V prevalence, although not significant (17 vs. 11), among young female participants in the R E A C H cohort[10]. In both cases, the female I D U had unique risk profiles of combined sexual and drug related risk[18]. Other studies of young IDUs have similarly found that female injectors have higher risk profiles [19, 20].  These findings have  important implications for prevention and addiction treatment for y o u n g IDUs and warrant further explorations into the unique characteristics and risk factors for y o u n g female injectors. We found similar proportions of Aboriginal youth compared w i t h older injectors in VIDUS, however Aboriginal people are far over-represented population in the V I D U S cohort and across Canada [21].  among the I D U  Half of the HIV-positive  female youth and all of the young males were First Nations participants. First Nation populations in N o r t h America have shouldered the burden of increased rates of morbidity  and  communities.  mortality, and  HIV/AIDS  has  devastating  potential  i n these  The respective national centres for disease control i n Canada and the  United States have identified the Aboriginal and Amer-Indian nations as increasingly vulnerable to the epidemic. Collaborative research i n these populations is urgently needed to stem the tide of new infections[22, 23]. When we considered sexual risk variables, we found that the youth were significantly more likely to engage i n survival sex when compared to older participants. Furthermore there was an overwhelming association between HIV-positivity and survival sex among  the female youth w h e n  participants and HIV-negative female youth.  compared  to  older  HIV-positive  Sex trade work has been shown to  increase both risk for H I V infection as well as vulnerability to increased drug use and violence[24-26]. The high number of female youth and the strong relational association between survival sex and H I V positivity, suggests an urgent need for further research 47  and intervention programs among young females engaged in survival sex[27].  The  overall rates of condom use fell below 30% and the youth engaged i n more frequent sexual activity through survival sex and w i t h casual sexual partners. The risks posed b y the sexual transmission of H I V among y o u n g females i n this cohort cannot be ruled out. We found high rates of sexual abuse i n both younger and older V I D U S participants.  However, sexual abuse among the young female participants was  reported by a staggering 70% of the HIV-positive and 60% of the HIV-negative youth. Childhood sexual abuse has been linked to subsequent w o r k i n the sex trade and a reduced ability to negotiate condom use[28, 29]. Furthermore, requiring help to inject was consistently reported combined  more frequently among the young participants.  disempowerment  associated  with  childhood sexual abuse,  The  continued  vulnerability to sexual predators through sex trade work and requiring help to inject is undeniably fueling the epidemic, specifically among young women[30, 31]. Treatment for drug addiction will need to consider lifetime sexual violence along w i t h treatment for drug dependency, especially for young female IDUs. Issues around empowerment, self-care including safe injection practices, and condom negotiation need  to  be  addressed i n order for young women w h o have a history of sexual abuse and sex trade work, to protect themselves from H I V / A I D S . D r u g risk variables unique to the youth included greater use of daily heroin, whereas older participants were more likely to inject cocaine daily. The finding that daily heroin injection is greater among the young participants is counter-intuitive to the image of drug use among youth as a party enhancer. While the pharmacological effects of cocaine have been linked to elevated self-esteem, heroin has been described as having a numbing effect[32]. Increased heroin use may be explained b y older contacts and drug use networks a n d / o r b y a lifetime characterized b y violence through  48  childhood sexual abuse, sex trade work, and early entry into drug use and street life[30]. The youth were also more likely to smoke crack cocaine daily, a behaviour that has previously been shown to increase risk for HIV-infection among high-risk youth in Baltimore and San Francisco [10, 33].  Doherty et al. suggested that the association  between increased risk for H I V and daily crack use may be linked to higher risk sexual behaviours among young crack users. Our findings of higher proportions of female, sex trade work, casual partners, l o w condom use and daily crack use among the y o u n g IDUs suggest confirmation of this assertion. The drug risk variable that was associated w i t h HIV-positivity, in univariate and multivariate analysis, among the female youth was daily speedball use.  The use of  injection speedballs by IDUs has been similarly associated w i t h increased risk for H I V in other cities such as Baltimore[9,10] and N e w York[34]. However, speedball use a m o n g young injectors, particularly females, warrants further investigation, particularly i n light of the recent overdose deaths in Houston, Texas that were attributed to speedball use[35].  Whether female youth are using speedballs for their combined effects of  alleviating pain associated w i t h heroin withdrawal and cocaine's fast pace properties that may enable sex trade work, or due to drug use networks remains to be clarified. However, education around safe mixing and injection practices may help to prevent overdose deaths and H I V infection. While daily heroin and speedball injection appear to be characteristic of y o u n g female IDUs, daily cocaine injection occurred in the three young HIV-positive males as shown in the profile in Table 5.  The profile also illustrates the high rates of daily  polydrug use that most of the HIV-positive youth were engaged in. Rather than focussing on one specific drug that may be increasing risk among the youth, it may be that the association between polydrug use and HIV-seroprevalence is mediated b y  49  increased sex trade work and sexual activity, particularly for y o u n g females with histories of childhood sexual abuse[36]. When we compared the younger and older participants we found the youth were more likely to require help when injecting i n the preceding six months.  This  behaviour, especially among youth, may increase vulnerability to both needle sharing and social networks and partnerships characterized by inequities in power[31]. Needle borrowing and lending were not statistically different from older injectors, however almost half of the youth reported needle sharing behaviors. This is cause for concern as the association between HIV-positivity and needle sharing behaviors has been well documented [9, 37, 38].  The potential link between needle sharing and those w h o  require help when injecting is something that requires further investigation and m a y shed some light on high-risk networks. There are several limitations with respect to this study. There may be sampling bias with respect to generalizabiltiy to other young injectors.  A s i n most studies of  IDUs, our study was based on self-reported behaviours and caution should be exercised in the interpretation of such data.  However, other studies of IDUs have found that  socially desirable responses had only a negligible effect on the responses concerning risk behaviours for H I V infection[39]. Another limitation is the risk behaviors identified in this study were cross-sectional i n nature, and d i d not account for subsequent changes in behaviours.  Nevertheless, these data provide information on a vulnerable  population i n w h i c h there has been little previous research conducted. In summary, targeted interventions are required among young female and Aboriginal injectors. In our study, among the youth, females with multi-factoral drug and sexual vulnerabilities were at highest risk for H I V . Furthermore, we found that over 50% of the HIV-positive youth were Aboriginal. develop youth specific programs that are gender  50  There is an urgent need to  and culturally appropriate  and  incorporate lifetime sexual violence survived through childhood sexual abuse and trade work.  51  3.6  References cited  1.  Health Canada H A E U : H I V and AIDS A m o n g Youth i n Canada. Ottawa: Population and Public Health Branch, Health Canada, 2000.  2.  H e i n K , Dell R, Futterman D , Rotheram-Borus M J , Shaffer N : Comparison of HIV+ and H I V - adolescents:  risk factors  and  psychosocial determinants.  Pediatrics 1995; 95(1): 96-104. 3.  Fennema JS, V a n Ameijden EJ, V a n Den Hoek A , Coutinho R A : Y o u n g and recent-onset injecting drug users are at higher risk for H I V . Addiction 1997; 92(11): 1457-65.  4.  Futterman D , Chabon B, Hoffman N D : H I V and AIDS in adolescents. Pediatric Clinics of N o r t h America 2000; 47(1): 171-88.  5.  U N I C E F : Progress of Nations. United Nations Joint Programme  on AIDS  (UNAIDS), 2000. 6.  Canadian Centres for Disease Control: A g e of Canadians infected w i t h H I V dropping. Canadian Medical Association Journal 1997; 156(6): 755.  7.  CDC: HIY/AIDS  A m o n g US women:  Minority  and  Young  Women  at  Continuing R i s k . : Center for Disease Control, 2001. 8.  C D C : C D C Fact Sheet: Y o u n g People at Risk. U S A : Centers for Disease C o n t r o l and Prevention, 2000.  9.  Garfein RS, Doherty M C , Monterroso ER, Thomas D L , Nelson K E , Vlahov D : Prevalence and incidence of hepatitis C virus infection among young adult injection drug users. JAIDS: Journal of Acquired Immune Deficiency Syndromes 1998; 18(Suppl 1): S l l - 9 .  10.  Doherty M C , Garfein RS, Monterroso E, Brown D , Vlahov D : Correlates of H I V infection among young adult short-term injection drug users. AIDS 2000; 14(6): 717-26.  11.  Nelson K E , Vlahov D , Solomon L , C o h n S, M u n o z A : Temporal trends of incident human immunodeficiency virus infection i n a cohort of injecting drug users i n Baltimore, M d . Archives of Internal Medicine 1995; 155(12): 1305-11.  12.  C D C : Youth Risk Behavior Surveillance System. : Centers for Disease C o n t r o l and Prevention, 1998.  13.  Health Canada: Canada's D r u g Strategy. Ottawa: The Office of Alcohol, Drugs and Dependency Issues, 1998.  14.  Levine OS, Vlahov D , Brookmeyer R, C o h n S, Nelson K E : Differences in the incidence of hepatitis B and human immunodeficiency virus infections among injecting drug users. Journal of Infectious Diseases 1996; 173(3): 579-83.  15.  Strathdee S A , Patrick D M , Currie SL, et al.: Needle exchange is not enough: lessons from the Vancouver injecting drug use study. A I D S 1997; 11(8): F59-65.  16.  V a n Ameijden EJ, Langendam M W , Notenboom J, Coutinho R A : Continuing injecting risk behaviour: results from the Amsterdam Cohort Study of drug users. Addiction 1999; 94(7): 1051-61.  17.  Hagan H , M c G o u g h JP, Thiede H , Weiss N S , Hopkins S, Alexander ER: Syringe exchange and risk of infection w i t h hepatitis B and C viruses. American Journal of Epidemiology 1999; 149(3): 203-13.  18.  Doherty M C , Garfein RS, Monterroso E , Latkin C , Vlahov D : Gender differences in the initiation of injection drug use among young adults. Journal of Urban Health 2000; 77(3): 396-414. 53  19.  Mullen L , Barry J: A n analysis of 15-19-year-old first attenders at the Dublin Needle Exchange, 1990-97. Addiction 2001; 96(2): 251-8.  20.  Rosenberg PS, Biggar RJ: Trends i n H I V incidence among young adults i n the United States. J A M A 1998; 279(23): 1894-9.  21.  Canadian AIDS Legal Network: Discrimination, H I V / A I D S and  Aboriginal  People. The Canadian Aboriginal A I D S Network, 1999. 22.  C D C : Surgeon General Calls for Action on H I V / A I D S Crisis in American Indian and Alaska Native Communities: Centers for Disease Control, 2000.  23.  Health Canada: H I V and AIDS A m o n g Aboriginal People in Canada. Bureau of H I V / A I D S , STD and TB, 2000.  24.  Spittal P M , Schechter M T : Injection drug use and despair through the lens of gender. C M A J 2001; 164(6): 802-3.  25.  Nemoto T, Brown LS, Jr., Foster K , C h u A : Behavioral risk factors of human immunodeficiency  virus  infection  among  intravenous  drug  users  and  implications for preventive interventions. AIDS Education & Prevention 1990; 2(2): 116-26. 26.  Potterat JJ, Rothenberg  RB, M u t h SQ, Darrow W W , Phillips-Plummer L:  Pathways to prostitution: The chronology of sexual and drug abuse milestones. Journal of Sex Research 1998; 35(4): 333-340. 27.  Tortu S, Beardsley M , Deren S, et al.: H I V infection and patterns of risk among women drug injectors and crack users in l o w and high sero-prevalence sites. A I D S Care 2000; 12(1): 65-76.  28.  Brannigan A , V a n Brunschot E G : Youthful prostitution and child sexual trauma. International Journal of L a w and Psychiatry 1997; 20(3): 337-354.  29.  Spatz W i d o m C , Kuhns JB: Childhood victimization and subsequent risk for promiscuity, prostitution,  and  teenage  pregnancy:  A  prospective  study.  American Journal of Public Health 1996; 86(11): 1607-1612. 30.  Miller M : A model to explain the relationship between sexual abuse and H I V risk among women. A I D S Care 1999; 11(1): 3-20.  31.  Krai A H , Bluthenthal R N , Erringer E A , Lorvick J, Edlin BR: Risk factors a m o n g IDUs who give injections to or receive injections from other drug users. Addiction 1999; 94(5): 675-83.  32.  Heffernan K , Cloitre M , Tardiff K , M a r z u k P M , Portera L , Leon A C : C h i l d h o o d trauma as a correlate of lifetime opiate use in psychiatric patients. Addictive Behaviors 2000; 25(5): 797-803.  33.  Des Jarlais D C , Abdul-Quader A , Minkoff H , Hoegsberg B, Landesman S, Tross S: Crack use and multiple AIDS risk behaviors. Journal of Acquired Immune Deficiency Syndromes 1991; 4(4): 446-7.  34.  Futterman D , H e i n K , Reuben N , Dell R, Shaffer N : H u m a n immunodeficiency virus-infected adolescents: the first 50 patients i n a N e w York City program. Pediatrics 1993; 91(4): 730-5.  35.  Yardley J: U p To 18 M a y Have Died of Overdoses In Houston. N e w York Times 2001 August 15, 2001.  36.  Spittal P, Tyndall M , L i K , Laliberte N , O'Shaughnessy M , Schechter M : Risk factors for H I V seroconversion among female and male injectors. Canadian Journal of Infectious Disease 2001; 12 (Supplement B).  55  37.  Krai A H , Lorvick J, Edlin BR: Sex- and drug-related risk among populations of younger and older injection drug users in adjacent neighborhoods  i n San  Francisco. Journal of Acquired Immune Deficiency Syndromes 2000; 24(2): 162-7. 38.  Bluthenthal R N , K r a i A H , Gee L , Erringer E A , Edlin BR: The effect of syringe exchange use on high-risk injection drug users: a cohort study. AIDS 2000; 14(5): 605-11.  39.  Latkin C A , Vlahov D , Anthony JC: Socially desirable responding and selfreported H I V infection risk behaviors among intravenous drug users. Addiction 1993; 88(4): 517-26.  56  Table 1:  A comparison of sociodemographic and sexual and drug risk variables between young IDUs (<24 years) and older IDUs (>24 years) in the VIDUS cohort Young  Older IDUs  Odds Ratios  •IDUs  (1205, 84%)  [95% CI]  p-value  (232,16%) Female  117 (50)  388 (32)  2.1 [1.6,2.8]  0.001  Aboriginal  56 (24)  306 (25)  0.9 [0.7,1.3]  0.686  Incarceration  72 (31)  407(34)  0.9 [0.7,1.2]  0.417  Any High School  162 (70)  1006 (83)  0.5 [0.3, 0.6]  0.001  Sex Trade  96 (41)  292 (24)  2.2 [1.7,2.9]  0.001  Casual Partners  106 (46)  363(30)  2.0 [1.5, 2.6]  0.001  Condom w/ Casual  61 (26)  186 (15)  2.0 [1.4, 2.7]  0.001  >20 Lifetime Partners  140 (60)  808 (67)  0.7 [0.6,1,0]  0.048  Sexual Abuse  89 (38)  429(36)  1.1 [0.8,1.5]  0.423  >_1 per day Heroin  119 (51)  478 (40)  1.6 [1.2, 2.1]  0.001  >_1 per day Cocaine  77 (33)  559 (46)  0.6 [0.4,0.8]  0.001  >_1 per day Speedball  33 (14)  205 (17)  0.8 [0.5,1.2]  0.295  >_1 per day Crack  40 (17)  115(10)  2.0 [1.3,2.9]  0.001  Needle Borrowing  97 (42)  454 (38)  1.2 [0.9,1.6]  0.236  Needle Lending  85 (37)  440 (37)  1.0 [0.8,1.3]  0.971  Help Injecting  123 (53)  470 (39)  1.8 [1.3,2.3]  0.001  57  Table 2:  A comparison of sociodemographic and risk variables between HIV  positive young and older VIDUS participants  HIV+Young  HIV+ Older  (23,10%)  (285,24%)  Female Gender  20(87)  111 (39)  10.5 [3.7,29.2]  0.001  Aboriginal  12 (52)  94 (33)  2.2 [0.9,5.1]  0.062  1(4)  76 (27)  0.1 [0.0, 0.7]  0.017  Sex Trade  20 (87)  98 (34)  12.7 [4.7, 34.6]  0.001  Sexual Abuse  16 (70)  116 (41)  3.3 [1.4, 8.0]  0.007  Casual Partners  7(30)  79 (28)  1.1 [0.5, 2.9]  0.780  Condom w/ Casual  3(13)  58 (28)  0.4 [0.1,1.3]  0.128  >20Lifetime Partners  22 (96)  177 (62)  13.4 [2.8, 64.9]  0.001  Heroin Frequency  12 (52)  85 (30)  2.6 [1.1, 5.9]  0.026  Cocaine Frequency  13 (57)  161 (56)  1.0 [0.4, 2.4]  0.998  Speedball  11(48)  65 (23)  3.1 [1.4, 7.1]  0.007  Crack Frequency  6(26)  28 (10)  3.2 [1.2, 8.5]  0.017  Needle Borrowing  8(35)  109 (38)  0.9 [0.4,2.1]  0.742  Needle Lending  8(35)  93 (33)  1.1 [0.5, 2.7]  0.833  Help Injecting  13 (57)  115 (40)  1.9 [0.8,4.5]  0.130  Married  Odds Ratios  p-value  [95% CI]  Frequency  58  Table 3:  A comparison of sociodemographic and risk variables between HIVpositive and HIV-negative young female VIDUS participants  HIV+ female  H I V - female  Odds Ratios  (20,17%)  (97, 83%)  [95% CI]  Aboriginal  9(45)  29 (30)  1.9 [0.7,5.1]  0.189  Sex Trade  19 (95)  57 (59)  13.3 [2.6,69.3]  0.002  Sexual Abuse  15 (75)  58 (60)  2.0 [0.7, 5.9]  0.201  Any High School  11 (55)  67 (69)  0.5 [0.2,1.5]  0.224  Regular Partner  6(30)  58 (60)  0.3 [0.1, 0.8]  0.015  Casual Partners  4(20)  42 (43)  0.3 [0.1,1.0]  0.052  1(5)  25 (26)  0.2 [0.0, 0.9]  0.042  20 (100)  64 (66)  21.3 [1.2,363.1]  0.002  Heroin Frequency  11 (55)  57 (59)  0.9 [0.3,2.3]  0.756  Cocaine Frequency  10 (50)  32 (33)  2.0 [0.8,5.3]  0.149  Speedball  10 (50)  10 (10)  8.7 [3.2, 23.5]  0.001  Crack Frequency  6(30)  17 (18)  2.0 [0.7, 5.9]  0.201  Needle Borrowing  7(35)  42 (43)  0.7 [0.3,1.9]  0.493  Needle Lending  8(40)  38 (39)  1.0 [0.4, 2.8]  0.945  Help Injecting  13 (65)  64 (66)  1.0 [0.3,2.6]  0.933  Condom w/ Casual >20 Lifetime Partners  p-value  Frequency  59  Table 4  Logistic regression analysis in which the outcome was HIV-positivity among the female youth (<24 years)  Unadjusted  Adjusted  Odds Ratios  Odds Ratios  [95% CI]  [95% CI]  7.7 [2.6, 22.8]  1.7 [1.3, 2.3]  1.3 [1.1,1.6]  7.5 [1.9, 30.0]  7.8 [3.4,17.9]  0.2 [0.0, 0.6]  1.3 [1.2,1.5]  0.3 [0.1,0.9]  Increased Age Per year Speedball >_1 per day Regular Partner Yes vs. no Education Yes vs. no  60  Table 5  Profile of socio-demographic, sexual and drug risk factors among the youth (<24 years)  N N Y Y Y Y Y N N Y N N N Y N Y Y Y N Y Y Y Y  61  ii  Y Y N N Y Y N Y Y Y Y Y Y Y N Y N Y Y Y N Y N  N N Y N N N N N N N N N N N Y N N N N N N N N  3  Y N N N N N N Y N N N N N N N N N N N N N N Y  n  Condom W/  Y Y Y Y Y Y Y Y Y Y Y Y Y Y N Y N Y Y Y N Y Y  33  Condom W/  Daily Speedball  Y N Y Y Y Y Y N Y Y Y N Y Y Y Y N Y N Y N Y Y  Daily Cocaine  Y Y Y Y Y Y Y N Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y  Daily  >20 Partners  Sex Trade  Age (years)  Ethnicity*  Gender  1 F 22 O Y 2 F 22 Y O F 3 O 24 Y 4 F Y O 20 F 22 5 O Y 6 F 21 Y . O 7 F 17 Y O 8 M • A 24 N 9 F 21 Y O F A 10 24 Y 11 F A Y 23 12 F A 21 N 13 M A N 23 14 F. O 22 Y F 15 Y O 23 F 16 A 22 Y 17 F 24 Y O 18 F A 24 Y 19 F A 21 Y F A 20 24 Y 21 F A Y 20 22 F A 22 Y 23 M A Y 19 *0=Other, A=Aboriginal  ) a*  Y Y Y N Y N N N Y Y Y N N Y Y Y Y N Y Y Y Y M  CHAPTER 4 HIV INCIDENCE A N D ASSOCIATED RISK FACTORS  4.1  Foreword This letter was a letter published as " H I V Incidence and Associated Risk Factors  A m o n g Young Injection D r u g Users" i n AIDS 2002, volume 16, Number 3. The brevity of this report was due to l o w statistical power or statistically small numbers for H I V incidence among the youth.  4.2  Introduction  Youth are of particular importance w i t h regard to the H I V epidemic because of the w i n d o w of opportunity  available to  prevent  disease transmission.  However,  researchers have noted that H I V infection among injection drug users (IDUs) is associated w i t h younger age and a fewer number of years injecting(l, 2). Studies have found that young female IDUs are at higher risk for H I V incidence than young male IDUs(3, 4).  Other studies have noted that important predictors of HlV-positivity  among young injectors include: cocaine and speedball injection, smoking crack cocaine, and younger age at first injection(5, 6). In a recent study of young injectors, Doherty et al. found that H I V incidence was associated w i t h injection cocaine and smoking crack cocaine daily although there were very few cases of incident H I V during their study period(6).  4.3  Methods We undertook this study to determine rates and risk factors for incident H I V  infection among young injectors i n a city where an explosive and ongoing H I V epidemic has occurred among IDUs.  Data were collected through an open cohort  study, the Vancouver Injection D r u g Users Study (VIDUS), since M a y 1996. Participants are eligible to enroll if they have injected i n the previous month and are residents of the 62  Lower Mainland of British Columbia.  A t baseline and every six months participants  were eligible to complete an interviewer-administered questionnaire eliciting questions concerning socio-demographics, drug use, and sexual patterns as well as testing through venous blood sample for H I V and Hepatitis C . For the current analysis, youth were defined as those participants i n the cohort w h o were <24 years at enrolment. A nested case-control using prospective data was undertaken to compare socio-demographic and risk factors for H I V incident youth cases versus HIV-negative youth controls. Risk factors for the cases were measured using behaviour data from the questionnaire administered in the six months prior to the seroconversion occurring. Controls were matched with cases based on age and questionnaire time period. The data were compared using contingency table analysis. A l l reported p-values are two-sided.  4.4  Results A m o n g the 232 participants aged <24 years, the H I V incidence rate was 4.37 per  100 person years.  In total there were 16 incident cases over a 5-year study period.  Despite limited statistical power, a number of strong associations were detected. Cases were more likely to be Aboriginal (56% vs 16%; p=0.004), have > 20 lifetime sexual partners (81% vs 47%; p=0.023), inject cocaine daily (63% vs 19%; p=0.002), and to use crack cocaine daily (63% vs 19%; p=0.002).  There was a marginal difference between  cases and controls with respect to daily speedball injection (31% vs 9%; p = 0.055). Differences between the two groups with respect to engaging i n sex trade w o r k (50% vs. 28%; p = 0.134), residing in unstable housing (69% vs. 47%; p=0.152), female gender (63% vs. 41%; p = 0.153), and any condom use w i t h casual sexual partners (31% vs 25%; p = 0.653) d i d not reach statistical significance. In a logistic regression analysis we found that Aboriginal status (AOR= 6.9, 95% CI= 1.8, 27.4, p=0.05) remained significantly associated with H I V infection [data not shown]. variable is highly important, other  However we feel that while this  variables may not have 63  reached  statistical  significance i n multivariate models due to l o w statistical power and should not discount the importance of the variables detected i n the case-control analysis.  4.5  Discussion To our knowledge this is one of the few studies to examine time relevant risk  factors associated w i t h H I V infection i n a young and vulnerable population in longitudinal fashion.  The incidence rate of 4.37 per 100 person years in a y o u n g  population w i t h short injection careers is alarming. Furthermore, the numerous drug and sexual vulnerabilities identified bodes poorly for the future of the epidemic unless immediate preventive action takes place. Our w o r k raises a number of important issues. First, young Aboriginal people were at markedly elevated risk. Second, similar to other studies of y o u n g IDUs, we found that smoking crack daily increased the risk for H I V infection(7-10).  Third, we  also confirmed that daily use of injection cocaine and speedballs was associated w i t h H I V infection, as was a high number of lifetime sexual partners(5, 6, 9,11,12). Of the H I V seroconverters, 63% occurred i n female participants and 50% were among those engaging i n sex trade work. Use of condoms w i t h casual sexual partners was very l o w for both positive and negative youth and over 50% reported living in unstable housing. The lack of statistical association should not negate the risks. Sex trade work has been previously shown to increase H I V vulnerability among y o u n g IDUs(7, 13, 14). Furthermore, a high number of sexual partners combined w i t h low rates of condom use is cause for concern.  Other research efforts have shown that  females, particularly young females, are at increased risk through sexual contact due to the increased efficacy of viral transmission from male to female sex partners(12, 15). The frequency and type of drug use engaged i n by this group of young IDUs demands response from policy makers and community workers. In summary, our study found a high rate of new infections among young IDUs engaged i n dual drug and sexual risks. The very high proportion of young Aboriginals 64  seroconverting is cause for alarm. Canadian and increasingly American First Nations people are witnessing the rapid spread of the virus and urgent action is required to stem the tide of new infections(16, 17).  In conclusion, targeted interventions  are  required among young IDUs that focus on young Aboriginal and female populations and that take dual drug and sexual vulnerability into account.  65  4.6 References cited  1.  Fennema JS, V a n Ameijden EJ, V a n Den Hoek A , Coutinho R A . Y o u n g and recent-onset injecting drug users are  at higher  risk for  H I V . Addiction  1997;92(ll):1457-65. 2.  Garfein RS, Vlahov D , Galai N , Doherty M C , Nelson K E . V i r a l infections in shortterm injection drug users: the prevalence of the hepatitis C , hepatitis B, human immunodeficiency, and human T-lymphotropic viruses. American Journal of Public Health 1996;86(5):655-61.  3.  Levine OS, Vlahov D , Brookmeyer R, C o h n S, Nelson K E . Differences in the incidence of hepatitis B and human immunodeficiency virus infections among injecting drug users. Journal of Infectious Diseases 1996;173(3):579-83.  4.  Zaccarelli M , Rezza G , Girardi E , Puro V , Pezzotti P, Lelli V , et al. Monitoring H I V trends i n injecting drug users: an Italian experience. A I D S 1990;4(10):1007-10.  5.  Marrero Rodriquez C A , C o l o n H M , Freeman D H , Matos T D , Reyes JC. H I V risk behaviour and H I V seropositivity among young injection drug users. Public Research Health Science Journal 1993;12:7 -12.  6.  Doherty M C , Garfein RS, Monterroso E, Brown D , Vlahov D . Correlates of H I V infection  among  young  adult  short-term  injection  drug  users.  AIDS  2000;14(6):717-26. 7.  Chiasson M A , Stoneburner R L , Hildebrandt DS, E w i n g W E , Telzak EE, Jaffe H W . Heterosexual transmission of H I V - 1 associated w i t h the use of freebase cocaine (crack). A I D S 1991;5(9):1121-6.  66  smokable  8.  M u r p h y S, Rosenbaum M . W o m e n who use cocaine too much: smoking crack vs. snorting cocaine. Journal of Psychoactive Drugs 1992;24(4):381-8.  9.  K r a i A H , Lorvick J, Edlin BR. Sex- and drug-related risk among populations of younger and older injection drug users i n adjacent neighborhoods  i n San  Francisco. Journal of Acquired Immune Deficiency Syndromes 2000;24(2):162-7. 10.  Des Jarlais D C , Abdul-Quader A , Minkoff H , Hoegsberg B, Landesman S, Tross S. Crack use and multiple AIDS risk behaviors. Journal of Acquired Immune Deficiency Syndromes 1991;4(4):446-7.  11.  E d l i n BR, Irwin K L , Faruque S, M c C o y C B , W o r d C , Serrano Y , et al. Intersecting epidemics—crack cocaine use and H I V infection among inner-city y o u n g adults. Multicenter Crack Cocaine and H I V Infection Study Team. N e w England Journal of Medicine 1994;331(21):1422-7.  12.  D'Angelo LJ. Adolescents and H I V infection: a clinician's perspective. Acta Paediatrica. Supplement 1994;400:88-94.  13.  van Ameijden EJ, van den Hoek AJ, van Haastrecht H J , Coutinho R A . Trends in sexual behaviour and the incidence of sexually transmitted diseases and H I V among drug-using prostitutes, Amsterdam 1986-1992. A i d s 1994;8(2):213-21.  14.  Spittal P M , Schechter M T . Injection drug use and despair through the lens of gender. C M A J 2001;164(6):802-3.  15.  Futterman D , Chabon B, Hoffman N D . H I V and AIDS in adolescents. Pediatric Clinics of N o r t h America 2000;47(l):171-88.  16.  CDC. HIV/AIDS  Among  US women:  Minority  Continuing Risk: Centers for Disease Control; 2001.  67  and  Young  Women  at  Health Canada. H I V and AIDS A m o n g Aboriginal People in Canada: Bureau H I V / A I D S , STD and TB; A p r i l 2000.  68  Table 1:  Comparison of sociodemographic characteristics, drug and sexual risk variables between seroincident youth cases (N=16) and HIV-negative youth controls (N=32)  HIV+ Youth  H I V - Youth  Odds Ratios  p-value  (n = 16)  (n =32)  Crack Frequency  10 (63)  6(19)  7,2 [2.0, 26.3]  0.002  Cocaine Frequency  10(63)  6(19)  7.2 [2.0, 26.3]  0.002  Aboriginal  9(56)  5(16)  6.9 [1.9, 25.9]  0.004  >20 Lifetime Partners  13 (81)  15 (47)  4.9 [1.2,19.6]  0.023  Speedball Frequency  5(31)  3(9)  4.4 [1.0,20.3]  0.055  Sex Trade  8(50)  9(28)  2.6 [0.7, 8.9]  0.135  Unstable H o u s i n g  11(69)  15 (47)  2.5 [0.7, 8.8]  0.152  Female  10 (63)  13 (41)  2.4 [0.7, 8.4]  0.153  Heroin Frequency  27 (59)  38 (41)  2.3 [0.6,8.8]  0.206  Help Injecting  7(44)  9(28)  2.0 [0.6, 7.0]  0.279  Sexual Abuse  7(44)  10(31)  1.7 [0.5, 6.0]  0.393  C o n d o m w / Casual  10 (31)  4 (25)  0.7 [0.2,2.3]  0.653  69  [95% CI]  CHAPTER 5 HEPATITIS C PREVALENCE A N D INCIDENCE A N D ASSOCIATED RISK FACTORS  5.1 Foreword This manuscript has been  accepted for publication as "Opportunities for  Prevention: Hepatitis C prevalence and incidence among young injection drug users" in Hematology, September 2002 Vol.36, No.3.  5.2 Introduction In N o r t h America the leading cause of Hepatitis C infection ( H C V ) is the sharing of contaminated equipment between injection drug users (IDUs)[l] . The incidence of H C V infection among populations of IDUs ranges from 4.2 to 22.0[2-6] per 100 personyears and prevalence estimates fall between 30% and 90% [2-4, 7, 8] . Due to the rapid acquisition of H C V infection following initiation into injection drug use, young IDUs or recent injection initiates represent an important group for H C V prevention. This m a y be particularly important in urban areas experiencing high prevalence of blood-borne infections among I D U populations. Studies have shown that younger IDUs engage in high-risk behaviours to a greater extent than established users, which increases their vulnerability to blood-borne infections[9,10]. The principal route for H C V infection among IDUs is using someone  else's  contaminated needle a n d / o r other drug use equipment [1,11]. Y o u n g IDUs may have a lower a n t i - H C V seroprevalence than older IDUs due to more recent initiation into injection drug u s e [ l l ] . However, younger IDUs may share needles and other injection equipment more so than older IDUs [12, 13] . Furthermore, younger IDUs may be more likely to have casual sexual partners, engage i n sex trade work, and use condoms 70  inconsistently [13-15]. Thus, i n addition to parenteral risk, y o u n g IDUs may engage in sexual behaviours that have been presumed to be risk factors for infection [16]. Vancouver, Canada has experienced an explosive epidemic of H I V and extensive spread of H C V among IDUs. Little is k n o w n about recent initiates and young IDUs. Identifying risk factors for H C V infection among young IDUs may provide important prevention clues for cities coping w i t h outbreaks of blood-borne infections among I D U populations. We undertook this study to estimate the prevalence and incidence of H C V infection among IDUs aged 24 and under and to characterize socio-demographic characteristics and risk factors for H C V among the young IDUs.  5.3 Methods Data were collected within a prospective open cohort study of injection drug users, the Vancouver Injection D r u g Users Study (VIDUS). A description of this study has been previously published[17] . V I D U S has recruited over 1,400 Vancouver area IDUs since M a y 1996. The study office is located in a storefront in the D o w n t o w n Eastside (DTES) of Vancouver.  The DTES is Vancouver's poorest neighbourhood  where an estimated 5,000 IDUs reside in an area of approximately ten city blocks, inexpensive housing i n the form of hotels and single room occupancies (SROs) abound. Eligibility criteria included residing in the city of Vancouver and surrounding municipalities, having injected i n the previous month, and aged 13 years and older. Participants were administered a questionnaire by trained nurse-interviewers and were eligible to return for follow-up every six months.  A t each visit, eligible participants  were tested for H I V and H C V antibodies through venous blood sample. This study was approved by the St. Paul's Hospital Committee on H u m a n Experimentation.  71  5.3.1  Instrument The V I D U S questionnaire is administered by trained nurses and interviewers and  elicits information regarding socio-demographic, sexual and drug risk characteristics. Continuous variables were elicited in reference to the previous six-months prior to the interview.  This includes the variables sex trade work, frequency  of injection,  incarceration, housing situation, sexual behaviours, needle and equipment sharing, and help injecting. Sex trade was defined as trading sex for money, drugs, or shelter. Needing help injecting was defined as ever requiring someone's help to inject drugs in the previous six months.  The variable other equipment includes cookers, cottons,  spoons and all other injection equipment aside from needles.  Unstable housing was  defined as living on the street, shelter, jail, hostel, or hotel.  5.3.2  Statistical analyses For the purposes of this study, young injectors were defined as those aged 24  years or less at the time of recruitment. The rationale for the age cutoff is based on the age criterion for youth a n d / o r adolescence used i n reports on H I V / A I D S generated b y the United Nations and the centres for disease control in the United States and Canada[18-20] . Young injectors identified at baseline as H C V antibody positive were compared with baseline a n t i - H C V negative youth using contingency table analysis. Chi-square and Fisher's exact tests were used to compare categorical variables and the Wilcoxon rank sum tests were used to compare continuous variables.  Logistic  regression models were used to identify independent predictors of baseline H C V positivity. Variables significant at p<0.05 in univariate analyses were included in an unconditional logistic regression model. A l l reported p-values were two-sided. Youth w h o became H C V antibody positive during the study period were compared to youth w h o remained H C V negative and had at least one follow-up visit. 72  The date of seroconversion was estimated using the midpoint between the last negative and the first positive antibody test result. Cumulative incidence rates of H C V infection were calculated using Kaplan-Meier methods. In these analyses, time zero was defined as the date of enrolment.  Participants who consistently remained seronegative were  considered to be right censored at the time of their most recent test result. A n n u a l rates of H C V seroconversion were calculated with actuarial methods. Relative risks and 95 percent confidence intervals were obtained for risk factors of interest. unadjusted  time dependent Cox regression models were  Adjusted and  used to identify risk  associations with H C V seroconversion among the y o u n g participants.  A l l p-values  reported were two-sided.  5.4  Results Of the 1,437 participants enrolled i n V I D U S , 232 were aged 13 - 24 years (median  21) at baseline. Almost half of the young participants, 107 (46%) were H C V positive at baseline, their median age was 22 (IQR: 20 - 23) years, and they had been injecting for a median duration of 4 years (IQR: 2 - 7 ) . The median age of HCV-negative youth was 20 (IQR: 18 - 22), and they had been injecting for a median of 1.3 years (IQR: 0.3 - 3). Table 1 compares baseline socio-demographic, drug and sexual risk variables between anti-HCV positive and anti-HCV negative youth.  Baseline positive youth  were older and their risk of being H C V positive more than doubled per 2-year increase i n the number of years the youth had been injecting. A s w o u l d be expected, H C V positive youth were more likely to be infected w i t h H I V than H C V negative youth. H C V positive youth were more likely to be of Aboriginal ethnicity (37% vs. 13%; p=0.001), have been incarcerated i n the previous six months (40% vs. 24%; p=0.006), engaged in sex trade w o r k (53% vs. 32%; p=0.001) have greater than 100 /  lifetime sexual partners (40% vs. 26%; p=0.021), have had previous STD(s) (43% vs. 73  28%; p=0.013) reside i n the I D U epicentre (53% vs. 33%; p=0.002), and inject heroin /  (52% vs. 35%; p=0.009), cocaine (40% vs. 17%; p=0.001), and speedballs daily (19% vs. 65; p=0.004). There was no difference between H C V anti-body positive and negative youth w i t h respect to always using condoms w i t h regular (14% vs. 20%; p=0.196), casual (23% vs. 28%; p=0.387) and client sexual partners (37% vs. 30%; p=0.228), using crack cocaine daily (11% vs. 11%; p=0.963), being female (51% vs. 50%; p=0.784), borrowing needles (42% vs. 41%; p=0.864), or sharing other drug equipment (79% vs. 79%;p=0.896).  Table 2 presents the results of our logistic regression modeling. Baseline antiH C V positivity among the youth was independently associated w i t h older age (OR: 1.29 [CI: 1.11,1.49]), a greater number of years injecting (OR: 1.27 [CI: 1.12, 1.42]), daily cocaine injection (OR: 2.58 [CI: 1.30, 5.15]), and sex trade work (OR: 2.30 [CI: 1.21, 4.37]). There were 76 youth w h o had follow-up data and 51 w h o were lost to follow-up. There was no difference between the youth w h o returned for follow-up and those w h o did not with respect to: female gender, those engaged i n sex trade work, age, years injecting, incarceration i n the previous six months, or types of drugs used. Aboriginal youth were significantly more likely than non-Aboriginal youth to return for follow-up visits. Of the 76 H C V negative youth w h o had follow-up data, 37 (49%) became antiH C V positive over the study period for an incidence rate of 37.3 (CI: 26.2, 51.4) per 100 person years. There was no difference between those w h o became H C V seropositive and those w h o remained negative w i t h respect to: female gender (62% vs. 51%; p=0.339), Aboriginal ethnicity (19% vs. 18%; p=0.913), and median age (20 [IQR: 19 - 22] vs. 20 [IQR: 18 - 22]). However number of years injecting was higher among the seroconverters, (2 [IQR: 0.5 - 4] vs. 1 [IQR: 0.5 - 3]) for those w h o remained negative.  74  Table 3 presents the results of our C o x regression model identifying risk associations w i t h H C V seroconversion among the youth. Only variables that reached marginal significance or significance are shown in table 3. Unadjusted risk associations for H C V seroconversion among the youth were; daily heroin (RR: 2.07 [CI: 1.03, 4.15]), cocaine (RR: 4.52 [CI: 2.23, 9.18]), speedball injection (RR: 2.76 [CI: 1.05, 7.31], having a partner w h o uses injection drugs (RR: 2.71 [CI: 1.35, 5.57]), requiring help w h e n injecting (RR: 2.24 [CI: 1.09,4.60]), and borrowing needles (RR: 2.57 [CI: 1.27, 5.21]). In the adjusted model, the variables that remained significant were; cocaine (RR: 3.04 [CI: 1.20, 7.70]), having a partner w h o uses injection drugs (RR: 2.48 [CI: 1.08, 5.66]), and requiring help when injecting (RR: 2.48 [CI: 1.08, 5.66]). 5.5 Discussion  The H C V incidence rate of 37.3 per person years among these young IDUs is alarming. Half of those w h o became H C V positive did so within the first 2 years of their injection career.  This finding underscores the need for targeted  interventions  among new initiates i n the early stage of injection drug use. Recent findings indicating the potential reversibility of H C V infection post exposure using interferon therapy are encouraging [21] .  Treatment opportunities for H C V infection will be particularly  important for y o u n g users given the rapid rates of H C V seroconversion that occur among I D U populations [3,11]. In recent studies of young IDUs, H a h n et al. identified an H C V incidence rate of 11 per 100 person years i n San Francisco, while i n Baltimore incidence rates of 16 and 23 per 100 person years were found [9, 11] . In all these studies, "young" was defined as age 30 and under.  We found an a n t i - H C V prevalence of 46% at baseline i n VIDUS,  similar to the San Francisco cohort when restricted to 24 years and younger, however our incident rate was much higher. For every two years the youth continue to inject, their risk of being H C V positive doubles so that those w h o have been injecting for six 75  or more years are ten times more likely to be H C V positive. In the San Francisco, Baltimore and the Vancouver studies, greater number of years injecting was associated w i t h HCV-positivity. We found a significant association between H C V positivity and H I V infection. A m o n g H I V positive youth, the prevalence of H C V at baseline was 88%, suggesting that co-infection w i l l be an almost universal problem among these y o u n g HIV-infected IDUs. It may be the case that interventions such as safe injection rooms that have shown success in reducing risk behaviours may help reduce the incidence of H I V and H C V among young injection drug users i n endemic cities[22] . Females accounted for half of the baseline H C V prevalent cases, and 63% of the incident cases. In unadjusted and adjusted analyses, associations w i t h seroconversion included requiring help to inject i n the previous 6 months and having an I D U sex partner. Furthermore, while condom use did not reach statistical significance in our analyses, use of condoms in all sexual partnerships is well below 40% in this y o u n g cohort and sexual transmission should not be ruled out.  While other studies have  suggested that intimate partnerships may be protective against blood-borne infections [11] , our findings suggest that young female IDUs may be at increased risk for H C V infection both through sexual and parenteral risk from their intimate partners. This may be particularly important for young IDUs as sexual partnerships may be less stable. Education and intervention programs among high-risk youth that address both sexual and drug risks may contribute to reducing blood-borne infections among youth w h o use injection drugs. Frequency of injection, or injecting at least daily, was associated with H C V infection i n Vancouver, Baltimore, and San Francisco among y o u n g IDUs [9, 11] . Cocaine has been a persistent risk factor for blood-borne infection among the V I D U S cohort, likely due to the increased frequency of injection that is characteristic of injection cocaine use [17, 23, 24] . However, many of the youth were also using heroin and 76  speedballs along with cocaine on a daily basis. occurring  among  these  young  injectors  The amount of daily polydrug use  is concerning  and  warrants  further  investigation. In previous studies of these y o u n g injectors we have found high rates of sexual abuse and participation in sex trade w o r k [25] . Sex trade work was again associated w i t h baseline H C V prevalence. The links between childhood victimization, violence in the sex trade and high-risk drug using behaviours such as frequent polydrug use should not be ignored in considering intervention and treatment programs for y o u n g IDUs. H i g h and increasing rates of H C V have been found in Canadian prison populations[26] . We found an association between H C V prevalence and incarceration in the previous six months.  Incarceration did not come out as a predictor of H C V  seroconversion, therefore we were unable to directly associate incarceration and H C V seropositivity.  However these data raise serious concerns about the potential for  blood-borne infection transmission among young IDUs within the prison system. In our study almost 80% of the youth reported  sharing ancillary drug  equipment. This risk factor d i d not reach statistical significance, most likely because this behaviour was saturated among these young IDUs, however transmission through shared ancillary equipment should not be ruled out. The study b y Hagan et al. showed a strong  association between  H C V seroconversion and sharing  ancillary drug  equipment [6]. For IDUs recently initiating injection use, the message that blood-borne infections, particularly H C V , can be transmitted through ancillary equipment may not have been heard, possibly because y o u n g IDUs are less likely to access services and therefore receive education and prevention information.  Another reason for high  proportions of sharing drug equipment among young IDUs could be the nature of their social network systems which may be characterized by trust and sharing. Given the high-risk nature of sharing drug preparation equipment and the high proportion of 77  youth reporting this behaviour, youth specific services such as peer group intervention programs, may help reduce sharing equipment among this vulnerable population. H C V prevalence was associated w i t h living in the downtown eastside of Vancouver, the I D U epicentre.  It has been argued that public policies which have  concentrated IDUs into one small area within an urban setting, have contributed to the rapid spread of blood-borne infections among I D U populations [23] . It is concerning that many of the youth reported residing within the I D U epicentre. It is likely that the lack of affordable housing options for high-risk youth, outside of the I D U epicentre, has contributed to this high incidence rate of H C V infection. Needle borrowing i n the six months prior to interview was reported among 40% of the youth, a finding similar to that of young IDUs i n other cities with well supported needle exchange programs (NEPs)[9, 13] . While it may be encouraging that over half of the young IDUs reported not sharing needles, for those w h o continue to share, other interventions are clearly required.  Borrowing needles was associated with H C V  seroconversion univariately, however it did not remain i n the final model. Borrowing needles is likely the most efficient means for contracting H C V among IDUs, however socially desirable responses may explain w h y it was not a direct measure of risk i n this cohort.  H C V seroconversion may also be explained b y the high number of youth  reporting sharing ancillary drug equipment. Interventions such as safe injection rooms may be particularly important for young IDUs in order to provide education on safe injection practices and clean equipment. This study has several limitations that should be noted. W e have included crosssectional data of the baseline H C V prevalent youth i n order to characterize y o u n g IDUs w h o may be most at risk and the types of behaviours they may be engaged in, however causal inference based on cross-sectional data needs to be interpreted with care. Furthermore, the study instrument d i d not address the practices of tattooing and body piercing, which have been shown to be associated with HCV-infection[27] . 78  Another important limitation concerns generalizability. A s w i t h other observational studies with voluntary recruitment, our study may have limited potential for generalization. While our results are not applicable to general populations, they likely reflect urban centres experiencing high prevalence of IDUs and H C V infections among these populations. W e know of no difference in sampling for the V I D U S project that may have selected for any specific youth characteristics that may differ from other young IDUs. Nevertheless, as this is not a probability sample, bias in recruitment is always possible. A s in most studies of IDUs, our study was based on self-reported behaviours and caution should be exercised in the interpretation of such data.  Other  studies of IDUs have found that socially desirable responses had only a negligible effect on the responses concerning risk behaviours for infection [28] . For young IDUs living in endemic cities, the w i n d o w of opportunity for H C V prevention is narrow. The links between high-risk p o l y d r u g use, unstable housing, sex trade w o r k and sexual partnership  vulnerability should not be ignored.  complexities of these risk associations require a complex response.  The  While access to  sterile syringes has reduced the transmission of blood-borne infections, we need to complement this service w i t h others that address housing, vulnerability among peer group networks, condom use in intimate and casual partnerships, and the dynamic drug using relationships of those w h o require help w h e n injecting.  79  5.6  References cited  1.  C D C . Recommendations for Prevention and Control of Infection and H C V Related Chronic Disease. Atlanta: Centers for Disease Control, 1998  2.  V a n Ameijden EJ, Langendam M W , Notenboom J, Coutinho R A . Continuing injecting risk behaviour: results from the Amsterdam Cohort Study of drug users. Addiction 1999;94:1051-61  3.  Garfein RS, Vlahov D , Galai N , Doherty M C , Nelson K E . V i r a l infections in shortterm injection drug users: the prevalence of the hepatitis C , hepatitis B, human immunodeficiency, and human T-lymphotropic viruses. American Journal of Public Health 1996;86:655-61  4.  Broers B, Junet C , Bourquin M , Deglon JJ, Perrin L , Hirschel B. Prevalence and incidence rate of H I V , hepatitis B and C among drug users on methadone maintenance treatment i n Geneva between 1988 and 1995. A I D S 1998;12:2059-66  5.  Thorpe L E , Ouellet LJ, Levy JR, Williams IT, Monterroso ER. Hepatitis C virus infection: prevalence, risk factors, and prevention opportunities among y o u n g injection drug users in Chicago, 1997-1999. Journal of Infectious  Diseases  2000;182:1588-94 6.  Hagan H , Thiede H , Weiss N S , Hopkins SG, Duchin JS, Alexander ER. Sharing of drug preparation equipment as a risk factor for hepatitis C . American Journal of Public Health 2001;91:42-6  7.  Galeazzi B, Tufano A , Barbierato E, Bortolotti F. Hepatitis C virus infection in Italian intravenous  drug users: epidemiological and clinical aspects. Liver  1995;15:209-12  80  8.  Alter M J , Kruszon-Moran D , Nainan O V , McQuillan G M , Gao F, Moyer L A , Kaslow R A , Margolis H S . The prevalence of hepatitis C virus infection in the United  States,  1988  through  1994. N e w  England Journal  of  Medicine  1999;341:556-62 9.  Garfein RS, Doherty M C , Monterroso ER, Thomas D L , Nelson K E , Vlahov D . Prevalence and incidence of hepatitis C virus infection among y o u n g adult injection drug users. Journal of Acquired Immune  Deficiency  Syndromes  1998;18:Sll-9 10.  Doherty M C , Garfein RS, Monterroso E, Brown D , Vlahov D . Correlates of H I V infection among young adult short-term injection drug users. AIDS 2000;14:71726  11.  H a h n JA, Page-Shafer K , L u m PJ, Ochoa K , Moss A R . Hepatitis C virus infection and needle exchange use among young injection drug users in San Francisco. Hepatology 2001;34:180-7  12.  Seal K H , Ochoa K C , H a h n JA, Tulsky JP, Edlin BR, Moss A R . Risk of hepatitis B infection among young injection drug users i n San Francisco: opportunities for intervention. Western Journal of Medicine 2000;172:16-20  13.  Krai A H , Molnar BE, Booth RE, Watters JK. Prevalence of sexual risk behaviour and substance use among runaway and homeless adolescents i n San Francisco, Denver and N e w York City. International Journal of STD & A I D S 1997;8:109-17  14.  Clements K , Gleghorn A , Garcia D , Katz M , Marx R. A risk profile of street youth in northern California: implications for gender-specific human immunodeficiency virus prevention. Journal of Adolescent Health 1997;20:343-53  15.  Martinez TE, Gleghorn A , Marx R, Clements K , Boman M , Katz M H . Psychosocial histories, social environment, and H I V risk behaviors of injection 81  and noninjection drug using homeless youths. Journal of Psychoactive Drugs 1998;30:1-10 16.  Conry-Cantilena C , VanRaden M , Gibble J, Melpolder J, Shakil A O , Viladomiu L , Cheung L , DiBisceglie A , Hoofnagle J, Shih JW. Routes of infection, viremia, and liver disease in blood donors found to have hepatitis C virus infection. N e w England Journal of Medicine 1996;334:1691-6  17.  Strathdee SA, Patrick D M , Currie SL, Cornelisse P G , Rekart M L , Montaner JS, Schechter M T , O'Shaughnessy M V . Needle exchange is not enough: lessons from the Vancouver injecting drug use study. A i d s 1997;ll:F59-65  18.  C D C . Y o u n g People at Risk: H I V / A I D S A m o n g America's Youth. Atlanta: Centers for Disease Control, 2000  19.  Bureau of H I V / A I D S . H I V and AIDS A m o n g Youth i n Canada. Ottawa: Health Canada, 2002  20.  U N A I D S . Children and young people in a w o r l d of AIDS. Geneva: United Nations, 2001  21.  Jaeckel E, Cornberg M , Wedemeyer H , Santantonio T, Mayer J, Zankel M , Pastore G , Dietrich M , Trautwein C , Manns M . Treatment of Acute Hepatitis C w i t h Interferon Alfa-2b. The N e w England Journal of Medicine . October 1,2001  22.  Ronco C , Spuhler G , Coda P, Schopfer R. Evaluation for alley-rooms I, II, and III i n Basel. Social Preventive Medicine 1996;41:58 - 68  23.  Schechter M T , Strathdee S A , Cornelisse P G , Currie S, Patrick D M , Rekart M L , O'Shaughnessy M V . D o needle exchange programmes increase the spread of H I V among injection drug users?: an investigation of the Vancouver outbreak. A i d s 1999;13:F45-51  24.  Patrick D M , Tyndall M W , Cornelisse P G , L i K , Sherlock C H , Rekart M L , Strathdee SA, Currie SL, Schechter M T , O'Shaughnessy M V . Incidence of hepatitis C virus infection among injection drug users during an outbreak of H I V infection. C M A J 2001;165:889-95  25.  Miller C, Tyndall M , Spittal P, L i K , Laliberte N , Schechter M . H I V Incidence and Associated Risk Factors A m o n g Y o u n g Injection D r u g Users. AIDS 2002;16:491 493  26.  Ford P, Pearson M , Sankar-Mistry P, Stevenson T, Bell D , Austin J. H I V , hepatitis C and risk behaviour i n a Canadian medium-security federal penitentiary: Queen's University H I V Prison Study Group. Queen's Journal of Medicine 2000;93:113 - 9  27.  Haley R W , Fischer R P . Commercial tattooing as a potentially important source of hepatitis C infection. Clinical epidemiology of 626 consecutive patients unaware of their hepatitis C serologic status. Medicine 2001;80:134-51  28.  Latkin C A , Vlahov D . Socially desirable response tendency as a correlate of accuracy of self-reported H I V serostatus for H I V seropositive injection drug users. Addiction 1998;93:1191-7  83  Table 1:  A comparison of sociodemographic and risk variables between baseline H C \ positive and negative young IDUs (<24 years, N=234)  Age <19 20-24 Years Injecting 0-1 2-3 4-5 >6 Female Aboriginal HIV-Positive Incarceration > 1 per day H e r o i n > 1 per day Cocaine > 1 per day Speedball > 1 per day Crack Needle B o r r o w i n g Sharing Equipment H e l p Injecting Sex Trade >100 LifetimePartners C o n d o m w/ Regular C o n d o m w / Casual C o n d o m w / Client Previous S T D Residing i n D T E S  HCV-positive (107,46%)  HCVnegati ve (127, 54%)  Odds Ratios [95% CI]  p-value  15 (14) 92 (86)  46 (36) 81 (64)  1 3.5 [1.8, 6.6]  0.001  21 (20) 20 (19) 26 (24) 40 (37) 55 (51) 40 (37) 21 (9) 43 (40) 56 (52) 43 (40) 20 (19) 12(11) 45 (42) 85 (79) 50 (47) 57 (53) 43 (40) 15 (14) 25 (23) 40 (37) 46 (43) 57 (53)  65 (51) 31 (24) 19 (15) 12 (9) 63 (50) 17 (13) 4(2) 30 (24) 45 (35) 22 (17) 8(6) 14 (11) 52 (41) 100 (79) 75 (59) 40 (32) 33 (26) 26 (20) 36 (28) 38 (30) 35 (28) 42 (33)  1 2.0 [0.9,4.2] 4.2 [2.0,9.0] 10.3 [4.8, 22.1] 1.1 [0.6,1.8] 3.9 [2.1, 7.2] 7.5 [2.8,19.9] 2.2 [1.2, 3.8] 2.0 [1.2, 3.4] 3.2 [1.8, 5.8] 3.4 [1.5, 7.8] 1.0 [0.4, 2.3] 1.0 [0.6,1.8] 1.0 [0.6, 2.0] 0.6 [0.4,1.0] 2.5 [1.5, 4.2] 1.9 [1.1, 3.3] 0.6 [0.3,1.3] 0.8 [0.4,1.4] 1.4 [0.8, 2.4] 2.0 [1.6,3.4] 2.3 [1.4, 3.9]  84  0.001  0.784 0.001 0.001 0.006 0.009 0.001 0.004 0.963 0.864 0.896 0.060 0.001 0.021 0.196 0.387 0.228 0.013 0.002  2:  Logistic  regression  determining  associations  of  anti-  seroprevalence among the young (<24) IDUs in the VIDUS cohort  Variable  Adjusted Odds Ratios [95% CI]  Age 1.29 [1.11,1.49]  Per year Years Injecting  1.27 [1.12,1.42]  Per year Frequent Cocaine Injection >1 per day  2.58 [1.30,5.15]  Sex Trade 2.30 [1.21,4.37] '  Yes vs. no  L  v  85  J  Table 3:  Risk associations for the young IDUs (<24) who tested antiH C V positive during the study period Risk Ratio 195% CI]  Adjusted Risk Ratio [95% CI]  Unstable Housing  1.88 [0.93,3.81]  1.86 [0.85,4.03]  > 1 Daily Heroin  2.07 [1.03,4.15]  1.39 [0.63, 3.04]  > 1 Daily Cocaine  4.52 [2.23, 9.18]  3.04 [1.20, 7.70]  > 1 Daily Speedball  2.76 [1.05, 7.31]  1.01 [0.29,3.37]  IDU Partner  2.71 [1.35, 5.57]  2.48 [1.08, 5.66]  Previous STD  1.91 [0.95, 3.85]  1.42 [0.60,3.38]  Need Help Injecting  2.24 [1.09, 4.60]  2.48 [1.08,5.66]  Borrowing Needles  2.57 [1.27, 5.21]  1.13 [0.47, 2.73]  Characteristic  *A11 continuous variables are elicited i n reference to the previous six months at the interview prior to seroconversion. **Only the variables that reached marginal significance or significance are reported in the table.  86  CHAPTER 6 C O M P A R I N G SOCIO-DEMOGRAPHIC A N D RISK FACTORS BETWEEN ABORIGINAL A N D NON-ABORIGINAL Y O U T H  6.1  Foreword This short communication is currently under review as "Higher prevalence and  incidence of H I V and hepatitis C and associated risk factors among young indigenous injection drug users".  6.2  Introduction In Canada and the United States, the respective centers for disease control have  identified Aboriginal and Amer-Indian Nations at high risk for H I V / A I D S [ l - 3 ] . Aboriginal people i n Canada are over represented among marginalized groups such as IDUs and street youth, especially in provinces such as British Columbia where higher proportions of Aboriginal people reside[4] . Historical factors such as the residential school system, displacement from land to reserve, and governmental policies that have precipitated cultural, language, and familial erosion have facilitated vulnerability to high-risk behaviour and blood-borne infections among Aboriginal people.  Despite  higher risk profiles, there is little documentation of the H I V / A I D S epidemic among Aboriginal peoples. In Canada, the  term Aboriginal includes the  many  different  nations of  indigenous peoples including First Nations, Inuit, and Metis, all groups having diverse languages and cultural traditions. In Canada, approximately 3% of the total population identify themselves as Aboriginal and in some provinces the proportion increases to between 4% and 12%[5] . In 1998 of the newly diagnosed H I V infections i n Canada, Aboriginal peoples numbered 26% and i n 1999 this figure was 18%[2] . The proportion  87  of A I D S cases attributable to Aboriginal peoples increased from 1% before 1990 to 11% in 1999[2]. The epidemiology of the H I V epidemic among Aboriginal peoples differs from non-Aboriginal peoples.  Between 1998 and 2000 females accounted for 47% of new  H I V infections among the Aboriginal populations and only 20% among non-Aboriginal people[6]. N e w l y infected Aboriginal people were significantly younger w i t h one third of new infections occurring among individuals under 29 years of age[6] .  Among  Aboriginal peoples, over half of the new infections were attributable to IDUs[7] . W e undertook this study to compare young (aged 24 and under) Aboriginal IDUs and nonAboriginal IDUs in a city where explosive and ongoing H I V and H C V epidemics have occurred and where approximately 25% of the I D U population are Aboriginal. 6.3  Methods Data were collected within a prospective open cohort study of injection drug  users, the Vancouver Injection D r u g Users Study (VIDUS).  This study has been  previously described [8] . V I D U S has recruited over 1,400 Vancouver area IDUs since M a y 1996, 25% are Aboriginal.  The study office is located in a storefront  D o w n t o w n Eastside (DTES) of Vancouver.  The DTES is Vancouver's  i n the poorest  neighbourhood where an estimated 5,000 IDUs and 1,000 street youth reside i n an area of approximately ten city blocks; inexpensive housing i n the form of hotels and single room occupancies (SROs) abound. Eligibility criteria included residing in the city of Vancouver and surrounding municipalities and having injected i n the previous month. restrictions.  There were no  age  Participants were administered a questionnaire by trained nurses and  interviewers and were eligible to return for follow-up every six months. A t each visit, eligible participants were tested for H I V and H C V antibodies through venous blood sample.  This study was approved by the St. Paul's Hospital Committee on H u m a n  Experimentation.  88  6.3.1  Instrument The V I D U S questionnaire is administered by trained nurses and interviewers and  elicits information regarding socio-demographic, sexual and drug risk characteristics. Continuous variables were elicited i n reference to the previous six months prior to the interview. This includes the variables sex trade work, frequency of injection, housing situation, sexual behaviours, and needle sharing. Sex trade was defined as trading sex for money, drugs, or shelter.  Unstable housing was defined as living on the street,  shelter, jail, hostel, or hotel. Aboriginal youth were identified through the question: A r e you of First Nations, Aboriginal, Inuit, or Metis origin?  6.3.1  Statistical analyses For the purposes of this study, young injectors were defined as those aged 24  years or less at the time of recruitment. W e used contingency table analysis to compare socio-demographic and risk factor variables from  the baseline questionnaire for  Aboriginal and non-Aboriginal youth. Chi-square and Fischer's exact tests were used to identify differences i n risk factors. There were 18 subjects w h o seroconverted under study. Since this l o w number precluded formal statistical analyses, we undertook a profile of the youth w h o became H I V seropositive during the study period to illustrate risk factors associated w i t h seroconversion.  The variables shown in the profile were from the questionnaire  answered in reference to the six months prior to the seroconversion being identified. A l l reported p-values are two-sided.  6.4 Results There were 235 participants aged <24 years, 57 (24%) were Aboriginal and 178 (76%) were Non-Aboriginal.  Aboriginal youth were more likely to be HIV-positive  (OR: 4.9 CI; [2.5 - 9.7]), HCV-positive (OR: 3.8 CI; [1.8 - 7.7]), female (OR: 2.7 CI; [1.4 4.9]), work i n the sex trade (OR: 2.6 CI; [1.4 - 4.8]), use condoms w i t h clients (OR: 2.9 C I ; 89  [1.6 - 5.4]), and inject cocaine (OR: 2.7 C I ; [1.5 - 5.0]), speedballs (OR: 2.3 C I ; [1.1 - 5.0]) and use crack (OR: 2.5 CI; [1.2 - 5.1]) on a daily basis. There was a marginal association between a history of sexual abuse and Aboriginal ethnicity (OR: 1.8 CI; [0.9 - 3.2]). Aboriginal youth were less likely to borrow needles (OR: 0.4 CI; [0.2 - 0.7]), and use condoms with casual sexual partners.  There were no differences between the two  groups w i t h respect to unstable housing (OR: 1.0 CI; [0.6 - 1.9]), alcohol use (OR: 1.1 C I ; [0.6 - 2.0]), or ever accessing methadone maintenance therapy (OR: 0.0 CI; [0.0 - 0.0]). Table 2 presents a profile of the young participants w h o seroconverted during the study period.  Over half of the youth w h o became H I V seropositive were  Aboriginal, and 70% of the Aboriginal youth w h o seroconverted were female.  The  majority of the seropositive Aboriginal youth were engaged i n sex trade work, had a high number of lifetime sexual partners and used condoms inconsistently w i t h casual and regular sexual partners.  The majority of these Aboriginal youth were injecting  cocaine on a daily basis.  6.5 Interpretation Aboriginal youth were almost 5 times more likely to be H I V positive and almost 4 times more likely to be H C V positive than non-Aboriginal youth. While there has been much literature documenting the explosive H I V outbreak  that occurred in  Vancouver, it is only recently that discussion has begun to take place regarding the epidemic's concentration among Aboriginal people[9] .  The significantly higher  proportion of Aboriginal youth with H I V and H C V infection should be an urgent warning to policy makers that these communities require resources to stem the tide of new infections. Furthermore, there is a need for more research to understand the full scope of the epidemic within Aboriginal communities. Almost 70% of the young Aboriginal participants were female. This differs from other populations of young IDUs in which females account for 30 - 50% of the cohort[10-13]. The link between young Aboriginal females, injection drug use and sex 90  trade work requires further investigation. The concentration of new H I V infections among young Aboriginal females is concerning and underscores the need for targeted interventions. Furthermore, the prevalence of HIV-positive young Aboriginal w o m e n i n prime child bearing years, highlights the need for medical services designed to meet the needs of these young women and potentially their children. Almost all the Aboriginal youth w h o became seropositive were engaged i n sex trade work as were well over half of the HIV-negative Aboriginal youth.  Sex trade  work has been previously shown to increase H I V vulnerability among IDUs as well as vulnerability to predators and continued sexual violence[14, 15] . Furthermore, half of the Aboriginal youth reported a history of sexual abuse.  The links between sexual  victimization and sexual violence experienced through sex trade work should not be ignored. While it is encouraging that Aboriginal youth were more likely to  report  condom use w i t h clients, overall condom use, especially among intimate sexual partners, was low. Aboriginal youth were significantly less likely to use condoms w i t h casual sexual partners, in fact only 15% reported condom use with casual partners. Sexual transmission may play an important role i n the higher prevalence and incidence of H I V among this population of IDUs. Aboriginal youth were much more likely to inject cocaine on a daily basis. Injection cocaine has been a persistent and strong risk factor for H I V and H C V infection among IDUs[8,16,17]. Aboriginal youth were more likely to use crack and speedballs on a daily basis, risk factors that have also been shown to increase vulnerability to blood-borne infection[10,18]. A further 42% of Aboriginal youth were using heroin on a daily basis.  The amount of daily poly-drug use occurring among these y o u n g  injectors is concerning. Associations between crack use and increased vulnerability in the sex trade has been previously documented[19] . The multiple sexual and drug risk factors identified in this young population may make intervention and prevention programs challenging. However these factors will likely need to be considered w h e n 91  developing programs to reduce the risk of blood-borne infections among y o u n g Aboriginal IDUs. Despite the high prevalence of H I V , H C V and associated risk factors, none of these y o u n g injectors have ever accessed methadone maintenance therapy.  This is  concerning as methadone has been shown to decrease risky behaviours [20, 21]. Access issues to opiate treatment in this population may need to be carefully examined. Furthermore, other harm reduction methods such as safe injection sites may reduce blood-borne transmission within the Aboriginal community, especially if developed in collaboration w i t h Aboriginal peoples. Almost 40% of the young Aboriginal people were HIV-positive and 82% were H C V positive. population.  These prevalence findings are exceedingly high for such a y o u n g  Furthermore, the H I V incidence rate of 6.8 and the concentration of  seroconversions among young Aboriginal females suggests that the H I V epidemic among Aboriginal injection drug users is far from abating.  The dynamics between  disempowerment associated with sexual victimization, sex trade work, daily polydrug use and sexual vulnerability in intimate partner relationships will need to be addressed in order to prevent further  transmission. The historical and continued betrayal  characterizing many governmental policies directed at Aboriginal peoples demands that the response to these epidemics comes in collaboration and from the Aboriginal communities; otherwise the effects may be minimal.  92  6.6  References cited  1.  Metier R, C o n w a y G A , Stehr-Green J. AIDS surveillance among American Indians and Alaska natives. American Journal of Public Health 1991;81:1469-71  2.  Health Canada. H I V and AIDS A m o n g Aboriginal People i n Canada: Bureau of H I V / A I D S , STD and TB, 2000  3.  C D C . Surgeon General Calls for Action on HrV/AIDS Crisis i n American Indian and Alaska Native Communities: Centers for Disease Control, 2000  4.  Gfellner B, Hundleby J. Patterns of D r u g Use among Native and White Adolescents: 1990 - 1993. Canadian Journal of Public Health 1995;86:95 - 97  5.  Statistics Canada. 1996 Census: Aboriginal data. Ottawa: Statistics Canada, 1998:1 -13  6.  Health Canada. H I V and AIDS i n Canada: Division of H I V / A I D S Epidemiology and Surveillance, 2001  7.  Health Canada. H I V / A I D S A m o n g Aboriginal Persons i n Canada Remains a Pressing Issue. Ottawa: Centre for Infectious Disease Prevention and Control, 2001:1 -8  8.  Strathdee S A , Patrick D M , Currie SL, Cornelisse P G , Rekart M L , Montaner JS, Schechter M T , O'Shaughnessy M V . Needle exchange is not enough: lessons from the Vancouver injecting drug use study. A I D S 1997;ll:F59-65  9.  Spittal P, Craib K , W o o d E, Laliberte N , L i K , Tyndall M , O'Shaughnessy M , Schechter M . Risk factors for elevated H I V incidence rates among injection drug users i n Vancouver. C M A J 2002;166:894 - 899  93  female  10.  K r a i A H , Lorvick J, Edlin BR. Sex- and drug-related risk among populations of younger and older injection drug users in adjacent neighborhoods in San Francisco. Journal of Acquired Immune Deficiency Syndromes 2000;24:162-7  11.  Garfein RS, Doherty M C , Monterroso ER, Thomas D L , Nelson K E , Vlahov D . Prevalence and incidence of hepatitis C virus infection among young adult injection drug users. JAIDS: Journal of Acquired Immune Deficiency Syndromes 1998;18:Sll-9  12.  Thorpe L E , Bailey SL, H u o D , Monterroso ER, Ouellet LJ. Injection-related risk behaviors in young urban and suburban injection drug users i n Chicago (19971999). Journal of Acquired Immune Deficiency Syndromes 2001;27:71-8  13.  Doherty M C , Garfein RS, Monterroso E , Latkin C , Vlahov D . Gender differences in the initiation of injection drug use among young adults. Journal of U r b a n Health 2000;77:396-414  14.  Spittal P M , Schechter M T . Injection drug use and despair through the lens of gender. C M A J 2001;164:802-3  15.  v a n Ameijden EJ, van den H o e k AJ, van Haastrecht H J , Coutinho R A . Trends in sexual behaviour and the incidence of sexually transmitted diseases and H I V among drug-using prostitutes, Amsterdam 1986-1992. A I D S 1994;8:213-21  16.  Schechter M T , Strathdee S A , Cornelisse P G , Currie S, Patrick D M , Rekart M L , O'Shaughnessy M V . D o needle exchange programmes increase the spread of H I V among injection drug users? A n investigation of the Vancouver outbreak. A I D S 1999;13:F45-51  94  17.  Patrick D M , Tyndall M W , Cornelisse P G , L i K , Sherlock C H , Rekart M L , Strathdee S A , Currie SL, Schechter M T , O'Shaughnessy  M V . Incidence of  hepatitis C virus infection among injection drug users during an outbreak of H I V infection. C M A J 2001;165:889-95 18.  Edlin BR, Irwin K L , Faruque S, M c C o y C B , W o r d C , Serrano Y , Inciardi JA, Bowser BP, Schilling RF, Holmberg SD. Intersecting epidemics—crack cocaine use and H I V infection among inner-city young adults. Multicenter Crack Cocaine and H I V Infection Study Team. N e w England Journal of Medicine 1994;331:14227  19.  Chiasson M A , Stoneburner R L , Hildebrandt DS, E w i n g W E , Telzak EE, Jaffe H W . Heterosexual transmission of HIV-1 associated w i t h the use of  smokable  freebase cocaine (crack). A I D S 1991;5:1121-6 20.  Langendam M W , van Brussel G H , Coutinho R A , van Ameijden EJ. Methadone maintenance and cessation of injecting drug use: results from the Amsterdam Cohort Study. Addiction 2000;95:591-600  21.  Colon H M , Robles RR, Deren S, Sahai H , Finlinson H A , Andia J, C r u z M A , K a n g SY, Oliver-Velez D . Between-city variation in frequency of injection among Puerto Rican injection drug users: East Harlem, N e w York, and Bayamon, Puerto Rico. Journal of Acquired Immune Deficiency Syndromes 2001;27:405-13  95  Table 1:  Comparison of sociodemographic characteristics, drug and sexual risk variables between Aboriginal youth (N=57) and Non-Aboriginal youth (N=178) i n the V I D U S project  Aboriginal  Non-Aboriginal  Youth  Youth  (n = 57)  ( n = 178)  O d d s Ratios  *p-value  [95% C I ]  HIV-positive  22 (39%)  20 (11%)  4.9 [2.5, 9.7]  0.001  HCV-positive  47 (82%)  99 (56%)  3.8 [1.8, 7.7]  0.001  Female  39 (68)  80 (45)  2.7 [1.4, 4.9]  0.002  Unstable Housing  33 (58)  102 (57)  1.0 [0.6,1.9]  0.937  Sex Trade  34 (60)  64 (36)  2.6 [1.4,4.8]  0.002  Sexual Abuse  28 (49)  63 (35)  1.8 [0.9,3.2]  0.064  Condoms w/ Casual  9(16)  52 (29)  2.2 [1.0,4.8]  0.045  Condoms w/ Clients  30 (53)  49 (28)  2.9 [ 1.6, 3.2]  0.001  Alcohol Use  25 (44)  75 (42)  1.1 [0.6,2.0]  0.819  < 1 Daily Heroin  24 (42)  96 (54)  0.6 [0.3,1.1]  0.120  < 1 Daily Cocaine  29 (51)  49 (28)  2.7 [1.5, 5.0]  0.001  < 1 Daily Crack  16 (28)  24 (13)  2.5 [1.2, 5.1]  0.011  < 1 Daily Speedball  13 (23)  20 (11)  2.3 [1.1,5.0]  0.029  Borrow Needles  14 (25)  84 (47)  0.4 [0.2, 0.7]  ,0.003  Methadone Ever  0(0)  5(3)  0.0 [0.0, 0.0]  0.201  96  Table 2  Profile of socio-demographic, sexual and drug risk factors among the young (<24 years) VIDUS participants who have HIV seroconverted during the study period  N  Y  2  M  O  .25  N  N  Y  Y  Y  Y  N  Y  Y  3  M  O  1  N  Y  N  N  N  N  N  Y  N-  4  F  O  3  Y  Y  Y  Y  N  N  N  N  Y  5  M  O  2  N  N  Y  N  Y  N  N  N  N  6  F  O  1  Y  Y  Y  N  N  N  Y  N  N  7  F  O  8  Y  Y  Y  N  N  N  N  N  N  8  F  O  2  Y  Y  Y  N  N  Y  N  N  Y  9  M  A  3  Y  Y  Y  Y  Y  N  N  Y  N  10  M  A  4  N  Y  N  N  N  N  N  N  N  11  F  A  4  Y  Y  N  Y  N  N  N  N  Y  12  M  A  2  N  Y  N  Y  N  N  N  N  N  13  F  A  3  Y  Y  N  Y  Y  N  N  N  Y  14  F  A  3  Y  N  N  Y  N  N  Y  N  Y  15  F  A  7  Y  Y  Y  Y  N  N  Y  N  Y  16  F  A  2  Y  . Y  N  N  N  Y  Y  N  Y  17  F  A  4  Y  Y  N  N  N  N  N  N  Y  18  F  A  .25  Y  Y  Y  Y  Y  Y  N  N  Y  *0=Other, A=Aboriginal  97  .  Condom W/  N  Clients  N  Condom W/  N  Casual  Y  Condom W/  N  Regular  Y  Daily Crack  Y  Daily  2  Speedball  O  Daily  F  Cocaine  Daily  Heroin  >100 Partner  Sex Trade  Ethnicity*  Years Fixing  Gender 1  CHAPTER 7 GENERAL DISCUSSION  7.1  Summary of study findings Data from the V I D U S project have provided insight into a hidden and little  k n o w n population i n Canada that is at high risk for transmission of blood-borne infections such as H I V and Hepatitis C . This preliminary investigation suggests that young IDUs w h o are most at risk for these infections are female and Aboriginal youth. Through this investigation, several risk factors of youth who are most vulnerable to infection were characterized. In this population H I V infection is concentrated among females. In fact 87% of the baseline H I V positive youth were female. A further 63% of the seroincident youth cases were female. This differs from older populations of IDUs where typically there are more males w h o inject and who therefore comprise a higher proportion of the affected population [1-3] . In the studies that have identified females at higher risk for H I V , overall presence of females is lower. A m o n g the youth, females account for over one half of the population and the majority of H I V infections is concentrated among the young females.  To m y knowledge the finding that H I V infection is concentrated  among females i n a young I D U population is a unique contribution to the literature that has not been documented i n a developed nation. Health Canada has documented the increasing impact that H I V and AIDS is having on Aboriginal communities. However there has been little epidemiologic investigation  into  the  impact  that H I V is having  within  these  communities.  Furthermore, given the younger overall age of the Aboriginal population and the younger age w h e n H I V infection is occurring among this group, examining the epidemic from within a youth cohort may be a useful contribution to understanding the epidemic. There was a much higher prevalence of both H I V and H C V among the young Aboriginal IDUs and especially among young Aboriginal females. 98  Risk factors associated with H I V prevalence among the female youth were increased age, sex trade work, not using condoms with casual partners, a high number of lifetime sexual partners, and greater than once daily speedball injection. There were a number of sexual risk factors identified among these young IDUs. Risk factors that remained in the model when adjusted were increased age and greater than once daily speedball injection. Daily speedball injection being an important predictor differs from other analyses done on the older V I D U S participants. While it is difficult to determine w h y this may be the case, it may be worthwhile to explore the link between higher proportions of young females engaged in sex trade w o r k injecting heroin and speedballs on a daily basis. The only independent predictor of H I V incidence among the y o u n g IDUs was Aboriginal ethnicity. Other associations detected univariately were at least daily crack use and cocaine injection, and a high number of lifetime partners. While daily cocaine injection is a consistent risk factor for H I V among the V I D U S cohort as well as other cohorts of IDUs, it may be informative to explore cocaine use among Aboriginal youth[4-6] . Furthermore, among the youth in V I D U S , a greater number of lifetime sexual partners is an important predictor of HIV-positivity. Finally, similar to other cohorts of high-risk youth, frequent or daily crack use was linked to H I V among these young IDUs[7, 8 ] . H C V seroprevalence among the youth was independently associated with increased age, a greater number of years injecting, at least daily cocaine injection, and sex trade work. A greater number of years injecting and increased age are consistent risk factors for H C V in other studies of IDUs.  The consistency w i t h which these  variables are shown to be associated with H C V infection should send an urgent message to policy makers i n Vancouver and in other urban settings that resources are needed to prevent the spread of this virus among young IDUs. Daily cocaine injection was again confirmed as an important predictor of H C V seroprevalence.  99  The chaotic  nature and frequency of injection associated w i t h cocaine use is likely fueling the H C V epidemics. Independent associations w i t h H C V seroconversion were daily cocaine injection, having a partner who uses injection drugs, and requiring help w h e n injecting. latter  two variables are  interesting and an addition to  the  current  The  literature;  furthermore they indicate the potential importance of sexual partnership vulnerability among young IDUs. These two variables did not reach statistical significance in other analyses when the cohort was considered as a whole. Thus these variables may be important i n terms of prevention among young IDUs. When young Aboriginal participants were compared w i t h young non-Aboriginal participants, much higher H I V and H C V prevalence rates were found.  In fact  Aboriginal youth had almost a five-fold increase of being H I V positive and a four-fold increase of being H C V positive. The large disparity between Aboriginal youth and non-Aboriginal youth i n terms of seroprevalence of both viruses is concerning. In most cohorts of IDUs, females account for approximately 30% of the population[9, 10] . A m o n g young IDUs, the number of females participating i n I D U cohorts increases to roughly half the population[ll, 12] . A m o n g Aboriginal youth, females accounted for 68% of the group vs. 45% of non-Aboriginal youth. This finding warrants further investigation as to w h y young Aboriginal females may be particularly vulnerable to engaging i n injection drug use.  Furthermore, given the higher risk  associations found among young females, urgent intervention programs are required to meet the needs of young Aboriginal females who use injection drugs.  7.2 Unique contributions, impact and implications The studies comprising this thesis have contributed to our understanding of the impact that H I V and H C V are having on youth in Canada and w h o is most at risk for these viruses. These studies compliment the few that have characterized young IDUs in North America, particularly i n a longitudinal fashion. Furthermore, because this thesis 100  was undertaken from an interdisciplinary perspective, it is hoped that the contributions made from the data analyses attempted  to go beyond a mere reporting of the  numbers, and to theorize about w h y some youth are at higher risk than others. It is also hoped that this w o r k will generate discussion within health boards and among policy makers to advocate for more research, resources and better policy development related to harm reduction and treatment initiatives for this vulnerable population. Through this work, we now k n o w young females and particularly y o u n g Aboriginal females are becoming infected at an earlier age than young males.  We  k n o w from these studies that those most at risk are engaging in dual drug and sexual related high-risk behaviours; many of the youth worked i n the sex trade, and there was low condom use especially in intimate partner relationships.  Furthermore, there  appears to be considerable poly-drug use among the youth, which has implications for prevention, treatment, and intervention programs. In a recent (April 2002) H I V / A I D S Epi Update report generated by Health Canada, the studies described here were mentioned several times, thereby contributing to how we think about risk i n Canada and among youth. Perhaps, through publication of these manuscripts, there will be greater dissemination of the findings.  This  knowledge w i l l allow public access within the community and among researchers alike to advocate for services and targeted interventions. Finally, these initial findings m a y lead to other studies of high-risk youth i n Canada.  7.3  Recommendations In  general,  the  research  described  here  gives  rise  to  two  primary  recommendations. The first is for more research to be conducted on young high-risk youth i n Canada, including both IDUs and non-IDUs w h o are at risk for H I V infection. The second is a call for intervention studies among this vulnerable population of y o u n g IDUs.  101  7.3.2  Research It is clear that youth w h o use injection drugs are at high risk for blood-borne  infections. In the case of H C V , time to infection is short. More research is needed to better understand the transition from non-injection to injection drug use so that appropriate interventions can occur that either help stop this transition or provide education on safer injection practices. Given the established risk for H I V and H C V among street youth w h o use injection drugs, researching the translation between street life and injection drug use may facilitate better prevention i n this area. Y o u n g IDUs are sharing needles and other drug preparation equipment. Investigating what form of education may better reach this population is important. C o n d o m use was l o w among the youth studied. Exploring ways to encourage and support Canadian youth to use condoms will be important to reduce the incidence of blood-borne infections. The high prevalence of sex trade w o r k among this y o u n g population is concerning. Research is needed on h o w best to intervene and prevent sex trade w o r k among youth since clearly there is an association between sex trade w o r k and increased vulnerability to blood-borne viruses.  A history of sexual violence was  highly prevalent among the high-risk young females. Researching ways to intervene among high-risk youth w h o have been the victims of sexual abuse may be important in understanding drug use and sexual vulnerability i n this population.  7.3.2  Intervention There are many examples of successful interventions that have occurred among  high-risk populations particularly from parts of Africa and Thailand[13] .  There are  examples of peer based interventions that have reduced the practice of sharing needles and drug preparation equipment among young IDUs[14, 15] . Interventions such as these may be worth exploring i n this context given the high prevalence of sharing among these young IDUs.  There are several studies that have shown success in  102  increasing condom use among young populations[16, 17] .  Increasing the use of  condoms or better facilitating their use among girls w o u l d also be useful i n this setting.  7.4  Conclusions The aim of the research presented here was to describe the prevalence and  incidence of H I V and H C V among  a previously u n k n o w n population and  to  characterize their risk for blood-borne infection. This w o r k has made contributions to our understanding of this vulnerable population. The studies summarized here and those of others strongly suggest a need for further research and intervention studies to prevent the spread of these viruses.  The combined sexual and drug vulnerabilities  identified i n these studies, particularly among young females, will make intervening challenging. However, an investment i n these youth w h o are "just starting out" will be an important investment i n terms of the health of the Canadian population.  103  7.5 References cited  1.  Bluthenthal R N , Krai A H , Erringer E A , Edlin BR. Use of an illegal syringe  exchange and injection-related risk behaviors among street-recruited injection drug users in Oakland, California, 1992 to 1995. JAIDS: Journal of Acquired  Immune  Deficiency Syndromes 1998;18:505-11 2.  Bruneau J, Lamothe F, Franco E, Lachance N , Desy M , Soto J, Vincelette J. H i g h  rates of H I V infection among injection drug users participating i n needle exchange programs in Montreal: results of a cohort study. American Journal of Epidemiology 1997;146:994-1002 3.  van Ameijden EJ, van den Hoek A R , Coutinho R A . Injecting risk behavior  among drug users i n Amsterdam, 1986 to 1992, and its relationship to AIDS prevention programs. American Journal of Public Health 1994;84:275-81 4.  Schechter M T , Strathdee S A , Cornelisse P G , Currie S, Patrick D M , Rekart M L ,  O'Shaughnessy M V . D o needle exchange programmes increase the spread of H I V among injection drug users?: an investigation of the Vancouver outbreak.  AIDS  1999;13:F45-51 5.  Paone D , Des Jarlais D C , Caloir S, Jose B, Shi Q, Friedman SR. Continued risky  injection subsequent to syringe exchange use among injection drug users i n N e w Y o r k City. A I D S Education & Prevention 1997;9:505-10 6.  Des Jarlais D C , Abdul-Quader A , Minkoff H , Hoegsberg B, Landesman S, Tross  S. Crack use and multiple AIDS risk behaviors. Journal of Acquired Immune Deficiency Syndromes 1991;4:446-7 7.  Chiasson M A , Stoneburner R L , Hildebrandt DS, E w i n g W E , Telzak EE, Jaffe H W .  Heterosexual transmission of HIV-1 associated w i t h the use of smokable freebase cocaine (crack). A I D S 1991;5:1121-6 8.  Edlin BR, Irwin K L , Faruque S, M c C o y C B , W o r d C , Serrano Y , Inciardi JA,  Bowser BP, Schilling RF, Holmberg SD. Intersecting epidemics—crack cocaine use and 104  H I V infection among inner-city young adults. Multicenter Crack Cocaine and H I V Infection Study Team. N e w England Journal of Medicine 1994;331:1422-7 9.  Strathdee SA, Patrick D M , Currie SL, Cornelisse P G , Rekart M L , Montaner JS,  Schechter M T , O'Shaughnessy M V . Needle exchange is not enough: lessons from the Vancouver injecting drug use study. A I D S 1997;ll:F59-65 10.  Moss A R , Vranizan K . Charting the epidemic: the case study of H I V screening of  injecting drug users in San Francisco, 1985-1990.  British Journal  of  Addiction  1992;87:467-71 11.  Fennema JS, V a n Ameijden EJ, V a n Den Hoek A , Coutinho R A . Y o u n g and  recent-onset injecting drug users are at higher risk for H I V . A d d i c t i o n 1997;92:1457-65 12.  H a h n JA, Page-Shafer K , L u m PJ, Ochoa K , Moss A R . Hepatitis C virus infection  and needle exchange  use among young injection drug users i n San Francisco.  Hepatology 2001;34:180-7 13.  U N A I D S . Children and young people in a w o r l d of AIDS. Geneva: U N A I D S ,  2001:1 - 13 14.  van Ameijden EJ, Coutinho R A . Large decline i n injecting drug use in  Amsterdam,  1986-1998: explanatory  mechanisms  and  determinants of  injecting  transitions. Journal of Epidemiology & Community Health 2001;55:356-63 15.  Rietmeijer C A , Kane M S , Simons P Z , Corby N H , Wolitski RJ, Higgins D L , Judson  F N , C o h n D L . Increasing the use of bleach and condoms among injecting drug users in Denver: outcomes of a targeted, community-level H I V prevention program. AIDS 1996;10:291-8 16.  Phoolcharoen W . H I V / A I D S prevention in Thailand: success and challenges.  Science 1998;280:1873-4 17.  Jamner M S , Wolitski RJ, C o r b y N H . Impact of a longitudinal community H I V  intervention targeting injecting drug users' stage of change for condom and bleach use. American Journal of Health Promotion 1997;12:15-24  105  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0076915/manifest

Comment

Related Items