@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Applied Science, Faculty of"@en, "Nursing, School of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Slawson, Gregory"@en ; dcterms:issued "2013-08-19T15:45:04Z"@en, "2013"@en ; vivo:relatedDegree "Master of Science in Nursing - MSN"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Abstract Background: Cannabis is increasingly prescribed clinically and utilized by people living with HIV/AIDS (PLWHA) to address symptoms of HIV disease and to manage side effects of antiretroviral therapy (ART). In light of concerns about the possible deleterious effect of psychoactive drug use on adherence to ART, we sought to determine the relationship between high-intensity cannabis use and adherence to ART among a community-recruited cohort of HIV-positive illicit drug users. In order to identify which PLWHA are accessing prescription cannabis, we also examined prevalence and correlates of those receiving a prescription for cannabis in the past six months. Methods: We used data from the ACCESS study, an ongoing prospective cohort study of HIV-seropositive illicit drug users linked to comprehensive ART dispensation records in a setting of universal no-cost HIV care. We estimated the relationship between at least daily cannabis use in the last six months, measured longitudinally, and the likelihood of optimal adherence to ART during the same period, using a multivariate linear mixed-effects model accounting for relevant socio-demographic, behavioral, clinical and structural factors. Using a cross-sectional design and bivariate statistical methods we also examined the prevalence and correlates of prescribed cannabis. Results: From May 2005 to May 2012, 523 HIV-positive illicit drug users were recruited and contributed 1215 person-years of observation. At baseline, 121 (23.1%) participants reported at least daily cannabis use. In bivariate and multivariate analyses, we did not observe an association between using cannabis at least daily and optimal adherence to prescribed ART (Adjusted Odds Ratio = 1.12, 95% Confidence Interval [95% CI]: 0.76 – 1.64, p-value = 0.555). From November 2011 to December 2012, 519 HIV-positive illicit drug users were surveyed, and in cross-sectional analysis, 81 (15.6%) individuals reported receiving a prescription for cannabis in the past 6 months. We found no significant differences among those who were and were not prescribed cannabis. Conclusions: High-intensity cannabis use was not associated with adherence to ART. A number of PLWHA report receiving a prescription for cannabis use. These findings suggest cannabis continues to be utilized by PLWHA for medicinal and recreational purposes without compromising effective adherence to ART."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/44836?expand=metadata"@en ; skos:note " ANTIRETROVIRAL ADHERENCE AND PRESCRIBED CANNABIS USE IN A POPULATION OF PEOPLE LIVING WITH HIV/AIDS by Gregory Slawson A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN NURSING in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) August 2013 ? GREGORY SLAWSON, 2013 ii Abstract Background: Cannabis is increasingly prescribed clinically and utilized by people living with HIV/AIDS (PLWHA) to address symptoms of HIV disease and to manage side effects of antiretroviral therapy (ART). In light of concerns about the possible deleterious effect of psychoactive drug use on adherence to ART, we sought to determine the relationship between high-intensity cannabis use and adherence to ART among a community-recruited cohort of HIV-positive illicit drug users. In order to identify which PLWHA are accessing prescription cannabis, we also examined prevalence and correlates of those receiving a prescription for cannabis in the past six months. Methods: We used data from the ACCESS study, an ongoing prospective cohort study of HIV-seropositive illicit drug users linked to comprehensive ART dispensation records in a setting of universal no-cost HIV care. We estimated the relationship between at least daily cannabis use in the last six months, measured longitudinally, and the likelihood of optimal adherence to ART during the same period, using a multivariate linear mixed-effects model accounting for relevant socio-demographic, behavioral, clinical and structural factors. Using a cross-sectional design and bivariate statistical methods we also examined the prevalence and correlates of prescribed cannabis. Results: From May 2005 to May 2012, 523 HIV-positive illicit drug users were recruited and contributed 1215 person-years of observation. At baseline, 121 (23.1%) participants reported at least daily cannabis use. In bivariate and multivariate analyses, we did not observe an association between using cannabis at least daily and optimal adherence to prescribed ART (Adjusted Odds Ratio = 1.12, 95% Confidence Interval [95% CI]: 0.76 ? 1.64, p-value = 0.555). From November 2011 to December 2012, 519 HIV-positive illicit drug users were surveyed, and in cross- iii sectional analysis, 81 (15.6%) individuals reported receiving a prescription for cannabis in the past 6 months. We found no significant differences among those who were and were not prescribed cannabis. Conclusions: High-intensity cannabis use was not associated with adherence to ART. A number of PLWHA report receiving a prescription for cannabis use. These findings suggest cannabis continues to be utilized by PLWHA for medicinal and recreational purposes without compromising effective adherence to ART. iv Preface I am extremely grateful for the opportunity and support provided from the entire team at the Urban Health Research Initiative as I worked with data from the ACCESS study. As the primary author I performed the literature review, selected the variables to include in the modeling, defined the type of analysis utilized and interpreted the results. The support of my supervisor Dr. Lynda Balneaves, the ACCESS study Co-Principal Investigator Dr. M-J Milloy and committee member Dr. Thomas Kerr were integral to the iterative process of the construction and explanation of the modeling procedures utilized in the multi-variate analysis, including the selection of confounding variables and also invaluable in providing comments on earlier draft of this thesis. Thesis supervisor Dr. Lynda Balneaves, committee members Dr. Thomas Kerr and Dr. Joy Johnson also provided additional editing and proofreading support during the completion of thesis. It should also be mentioned, additional statistical support was received from Annick Simo, who ran the longitudinal bivariate and multivariate analysis and provided statistical results for the thesis. The candidate, however, performed the interpretation and analysis of the findings. Additional feedback and support during the editing of the thesis was received from the thesis committee members Drs. Balneaves, Kerr and Johnson. Dr. Lynda Balneaves also provided essential feedback in the process, utilizing her knowledge of contemporary cannabis research to add contextual perspective. This research study ?The effects of cannabis use on adherence to HARRT among HIV-positive injection drug users? was reviewed and approved by the UBC-Providence Health Care Research Ethics Board. UBC-PHC REB Number: H05-50233. v Table of Contents Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iv Table of Contents ...........................................................................................................................v List of Tables ............................................................................................................................... vii List of Abbreviations ................................................................................................................. viii Acknowledgements ...................................................................................................................... ix Chapter 1: Introduction ................................................................................................................1 1.1 Literature Review............................................................................................................ 6 1.1.1 Cannabis Use as Medication ....................................................................................... 6 1.1.2 Safety of Cannabis ...................................................................................................... 7 1.1.3 Prevalence of Cannabis Use in PLWHA .................................................................. 10 1.1.4 Symptom Management and Adherence to ART ....................................................... 13 1.1.5 Cannabis, Drug Use and Adherence to ART ............................................................ 15 1.1.6 Barriers and Facilitators to Adherence amongst PLWHA Who Use Illicit Drugs ... 16 1.1.7 Obtaining a Prescription for Cannabis: The Canadian Context ................................ 18 1.2 Purpose of the Study ..................................................................................................... 20 Chapter 2: Research Methods ....................................................................................................22 2.1 Sample........................................................................................................................... 22 2.2 Data Collection ............................................................................................................. 23 2.2.1 Data Measurement for ?High Intensity Cannabis Use and Adherence to ART? ....... 24 2.2.2 Data Measurement for Prevalence and Explanatory Correlates of Receiving a Prescription for Cannabis ...................................................................................................... 26 vi 2.3 Data Analysis for High Intensity Cannabis Use and Adherence to ART ..................... 27 2.4 Data Analysis for Explanatory Correlates of Receiving a Prescription for Cannabis .. 28 Chapter 3: Study Results.............................................................................................................29 3.1 Results of Analysis of High-intensity Cannabis Use and Adherence to ART .............. 29 3.2 Results of Analyses of Explanatory Correlates of Receiving a Prescription for Cannabis .................................................................................................................................... 33 Chapter 4: Conclusion .................................................................................................................36 4.1 Discussion of the Results: Analysis of High-intensity Cannabis Use and Adherence to ART 36 4.1.1 Practice and Policy Implications ............................................................................... 38 4.1.2 Strength and Limitations ........................................................................................... 41 4.2 Discussion of the Results: Analysis of Explanatory Correlates of Receiving a Prescription for Cannabis .......................................................................................................... 42 4.2.1 Strength and Limitations ........................................................................................... 44 4.2.2 Practice, Policy and Future Research........................................................................ 45 4.2.3 Summary ................................................................................................................... 48 References .....................................................................................................................................50 vii List of Tables Table 1: Baseline prevalence of cannabis use of 523 ART-exposed illicit drug users in the last six months ........................................................................................................................................... 29 Table 2: Baseline characteristics of 523 ART-exposed illicit drug users stratified by at least daily cannabis use in the last six months, ACCESS study .................................................................... 30 Table 3: Longitudinal bivariate and multivariate mixed-effects analyses of factors associated with ?95% adherence to ART in the previous six months ........................................................... 32 Table 4: Cross-sectional and bivariate analyses of explanatory correlates of receiving a prescription for cannabis in the previous six months, among 519 participants in the ACCESS study .............................................................................................................................................. 33 viii List of Abbreviations ACCESS- Aids Care Cohort for the Evaluation of Support Services ART- antiretroviral therapy CTP- cannabis use for therapeutic purposes CMA- cannadian medical association CNS- central nervous system CI- confidence interval DTP- drug treatment program HAART- highly active antiretroviral therapy HAND- HIV associated neurocognitive disorders HIV -human immunodeficiency virus MMAP- Medical Marijuana Access program MMAR- Medical Marijuana Access Regulations NRTI- nucleoside reverse transcriptase inhibitors NNRTI- non- nucleoside reverse transcriptase inhibitors OR- odds ratio PI- protease inhibitors PLWHA- people living with HIV/AIDS pVL- plasma HIV-1 RNA viral load PWID- people who use illicit drugs THC- Tetrahydracannibinol ix Acknowledgements I would like to acknowledge the support of my thesis supervisor Dr. Lynda Balneaves, committee members Dr. Thomas Kerr and Dr. Joy Johnson. I would like to thank ACCESS principal investigator Dr. M-J Milloy for his ongoing support and mentorship during this process. I would also like to extend a special thanks to my wife Angie and children, Isabelle and Daniel, for their ongoing support and patience during the completion of the thesis. I wish to thank the ACCESS/VIDUS study participants, as well as current and past researchers and staff for their continued contribution to the research. In addition I would like to thank the pharmacy staff at the SPH Immunodeficiency Clinic and the patients for their insight into antiretroviral therapy medications, adherence to ART and the use of cannabis in the management of HIV. This study was supported by the US National Institutes of Health (R01DA21525). 1 Chapter 1: Introduction Cannabis has been utilized for medicinal purposes for several centuries (Kalant, 2001). In 2001 the Government of Canada amended the Controlled Drug and Substances Act and created Marihuana Medical Access Regulations (MMAR), which allowed patients to legally utilize cannabis to manage the symptoms associated variety of illnesses including HIV/AIDS (Government of Canada, 2011). Symptoms of HIV/AIDS covered in the MMAR include: anorexia, poor appetite, pain, neuropathic pain, nausea/vomiting (Health Canada, 2011). In descriptive studies, people living with HIV/AIDS (PLWHA) have reported cannabis use as an effective medication for managing a number of HIV related symptoms including appetite stimulation, anorexia, stress, nausea/vomiting, and pain (Belle-Isle & Hathaway, 2007; Prentiss, Power, Balmas, Tzuang & Israelski, 2004; Ware, Rueda, Singer & Kilby, 2003). In addition to managing symptoms associated with HIV disease processes, PLWHA have also reported using cannabis to manage the side effects of antiretroviral therapy (ART) medications including: gastro-intestinal upset, insomnia, and peripheral neuropathy (Ware, et al., 2003). Effective symptom management can play a large role in improving the quality of life for patients suffering with HIV. With improvements to ART medications and effective medication adherence, PLWHA are living longer and HIV is now considered a chronic illness. Because of the longer life expectancy and chronic nature of the disease, many PLWHA are exposed to ARV medications and HIV over a long period of time, resulting in a potential increase in adverse effects of ART. As a result, practitioners working with PLWHA must consider prescribing medications and therapies that promote effective symptom management and subsequently improve the quality of life for PLWHA. 2 To date very little research has described the prevalence and correlates of cannabis use for medical management of HIV and ART-related symptoms. In addition, very little data exists that describes the clinical characteristics and medication regimens of PLWHA who report using cannabis for medical purposes. No long-term prospective studies assessing the demographic, clinical and structural correlates of prescribed cannabis use have been conducted within a setting of universal healthcare, where PLWHA also have legal access to cannabis for therapeutic use. As professionals whose primary goal is to help and support people living with HIV manage their illness and improve the quality of their lives, healthcare providers require more evidence about the characteristics and experiences of HIV-positive individuals who report HIV symptom management with prescribed cannabis. This information will help support decision making, policy development and prescribing practices that involve using cannabis as a potential intervention for effective symptom management of HIV. Since the introduction of highly active ART (HAART), PLWHA have experienced significant gains in HIV-related morbidity and mortality (Hogg et al. 1998; Lima et al., 2007). HAART involves combining several antiretroviral medications that target different areas of HIV reproduction, in order to diminish the virus?s ability to replicate. While a small number of PLWHA remain on single or dual ART options, ART regimes usually include a combination of two nucleoside reverse transcriptase inhibitors (NRTI) with either a non-nucleoside reverse transcription inhibitors (NNRTI) or a protease inhibitors (PI)(British Columbia Center for Excellence in HIV/AIDS, 2009). By combining these medications, clinicians are able decrease the amount of circulating virus in fluids like blood plasma, which diminishes the impact of the virus on immune cells and over time allows the body?s immune system CD4 cells to replenish (Montaner et al., 1998). ART also reduces HIV viral load, or the amount of virus in PLWHA?s 3 blood and circulating fluids, to undetectable levels. This reduction in viral load has benefit to HIV-positive individuals by promoting immune health, and also has the potential secondary public health benefit of reducing the rates of transmission of HIV in the general population (Wood et al., 2012). Recent research has indicated that by effectively adhering to ART medications individuals may protect their non-HIV sexual partners from acquiring the disease (Cohen, et al., 2011). HIV researchers speculate that, through the expansion of access to ART globally, the HIV epidemic could be effectively curtailed (Lima et al., 2008a). With improved ART regimes delivering potential benefits to both individuals and communities, the focus is increasingly turning to supporting adherence to ART in PLWHA and managing side effects associated ART. The PLWHA who are able to maintain optimal adherence to HIV medications have improved life expectancy and reduced morbidity (Zwhalen et al., 2009). Optimal adherence is defined as greater than or equal to 95% adherent to ART therapy (Low-Beer, Yip, O`Shaughnessy, Hogg & Montaner, 2000a). Optimal adherence to ART adherence has been shown to be strongly associated with virological suppression of plasma HIV RNA (pVL), restoration of immunologic function and lengthened survival for PLWHA (Hogg et al. 1998; Low-Beer et al., 2000a; Pallela et al. 1998 ). Unfortunately for PLWHA, in order to be effective, ART requires sustained adherence and patients are expected to be on these medications for life (Lima, et al., 2008b). If optimal adherence is not achieved, HIV virus mutations and drug resistance may occur, limiting ART treatment options and subsequently increasing an individual?s risk of hospitalization and death (Fielden et al., 2008; Hogg, et al., 2006). Sub-optimal adherence and subsequent increase in viral load may also increase the risk of HIV transmission and the transmission of drug resistant strains of HIV in the greater community 4 (Hogg et al., 2006; Wood et al., 2008). As a result, there has been great interest within clinical and research communities regarding strategies, that may enhance adherence to ART. Adherence to HAART can present a number of challenges for individuals. Although dosing regimens have improved in recent years, many regimens can be complex and require multiple doses at various points during the day. Often, effective absorption of ART medications requires that medications are taken with food or with dietary restrictions, which may present difficulty for individuals with food security issues and individuals who lack appetite (Ferguson, Stewart, Funkhouser, Tolson, Westfall & Saag, 2002). Many medications in ART regimes are associated with side effects that need to be managed, to help encourage effective adherence. If these side effects are not effectively managed, they can undermine individual adherence behaviour and can result in the discontinuation of HAART. Lastly, the cost of paying for medications can be prohibitive for individuals of lower socio-economic backgrounds that live in countries where ART are not provided free of charge (McDonnell et.al., 2006). In addition to the barriers to optimal adherence faced by PLWHA, there are additional personal, social and structural factors that can influence effective adherence amongst PLWHA who use illicit drugs (PWID) (Malta, Magnanini, Strathdee & Bastos, 2010). PWID account for a substantial number of existent and newly diagnosed cases of HIV in Canada. PWID often become infected with HIV by sharing contaminated drug equipment and represented 17% of new HIV diagnoses in Canada in 2008 (Public Health Agency of Canada, 2011). Despite the advances in HIV treatment, PWID have frequently been found to exhibit sub-optimal adherence to ART and HIV/AIDS-related morbidity and mortality remains high in this population (Zwahlen et al., 2009; Wood et al. 2003). Amongst PLWHA, sub-optimal adherence has been associated with active illicit drug use (Arnsten et al., 2002; Crisp et al., 2004; Hinklin et 5 al., 2007; Lucas et al., 2001; Malta et al., 2008; Nolan et al., 2011; Palepu et al., 2006; Sharpe et al., 2004). Additional demographic characteristics and individual-level barriers to adherence have been identified amongst PWID populations and include: age; female gender; aboriginal ethnicity, alcohol use; lower self-efficacy; and psychiatric co-morbidities(Arnsten et al., 2007; Bouhnik et al., 2005; Carrieri et al., 2003; Hadland et al., 2012; Kerr et al., 2005a; Miller et al., 2006; Palepu et al., 2004a; Tapp et al., 2011). Recently social and structural barriers to effective adherence have also been identified such as homelessness and incarceration (Bouhnik et al., 2002; Milloy et al., 2011; Milloy et al., 2012; Palepu et al., 2004b; Palepu et al., 2011). Despite the risk active psychoactive drug use can present for non-adherence to ART, few studies have evaluated the relationship between cannabis use and adherence to HIV treatment. Cannabis is used frequently by PLWHA, with one observational study reporting 59% of PLWHA engaged in cannabis use in the past six months (Fogarty et al., 2007). Cannabis use for therapeutic purpose (CTP) has been reported by PLWHA to help manage a number of HIV-related symptoms including; appetite stimulation, anorexia, stress, nausea/vomiting, and pain (Prentiss et al., 2004; Ware, et al., 2003; Woolridge et al., 2005). Despite the evidence to support the efficacy of cannabis in the amelioration of HIV related symptoms (Abrams et al., 2003; Abrams et al., 2007a; Ellis et al., 2009; Haney et al., 2007), the evidence regarding cannabis use and the impact on adherence remains limited and, no long-term prospective studies assessing cannabis use as a factor that may affect adherence have been conducted within a setting of universal healthcare with publically-funded HIV care. The present study was conducted to investigate the role cannabis use among a cohort of PWID living with HIV/AIDS, to evaluate if cannabis use is detrimental to ART adherence. 6 1.1 Literature Review 1.1.1 Cannabis Use as Medication There are a number of psychoactive compounds commonly referred to as cannabanoids found in the cannabis plant. Tetrahydracannibinol (THC) is the cannabinoid that typically accounts for most of the physical and psychological effects seen in patients and cannabis users. THC acts on two types of cannabinoid receptors CB1 and CB2. The CB1 receptors are found mainly in the brain and CNS and account for many of the psychoactive and physiological responses to cannabis (Gali?egue, et al., 1995; Grotenhermen, 2003). Cannabinoids have been identified as potent anti-inflammatories in studies examining their effect on in vitro and in vivo animal samples (Klein et al., 2003) CB2 receptors are located primarily in the immune cells and neurons, which may account for the anti-inflammatory action of THC. In addition to the pharmacodynamics research on cannabis, there is a growing body of scientific data that suggests cannabis can be used for acute management of a variety of symptoms including nausea, vomiting, anorexia(i.e., wasting syndrome), pain, neuropathic pain, muscle spasticity and depression and anxiety (Grotenhermen, 2003; Kalant, 2001).Cannabis has been utilized by individuals to ameliorate symptoms related to a number of diseases including HIV/AIDS, cancer, multiple sclerosis, anorexia, rheumatoid arthritis, arthritis, glaucoma, asthma, and mood disorders (Clark et al., 2004; Grotenhermen, 2003; Ware, Adams, & Guy, 2005). Cannabinoid use for HIV related symptoms and AIDS- related conditions has been examined in a number of observational studies. Early studies highlighted cannabis use as an effective complementary medicine to encourage appetite stimulation, which helped prevent the muscle wasting and anorexia associated with HIV (Beal et al., 1995; Beal et al., 1997). Cannabis use in the treatment of nausea and vomiting related to HIV illness and medications has also been 7 examined in observational studies (de Jong, Prentiss, McFarland, Machekano & Israelski, 2005; Lane et al., 1991). In addition to the descriptive evidence on the use of cannabis in PLWHA populations, randomized control trials have been performed to assess the effectiveness of cannabis in the treatment of symptoms associated with HIV/AIDS. These studies have focused on the use of smoked cannabis to treat HIV- sensory neuropathic pain a condition that impacts an estimated 30% of PLWHA and is the most common symptom of HIV disease (Ellis et al., 2009). In a double-blinded, crossover trial, Ellis et al. (2009) found that smoked cannabis as an adjunct to ?concomitant analgesic use? was effective in treating neuropathic pain as compared to a placebo. The authors found that cannabis use improved neuropathic pain management with minimal side effects. Improvements to mood and daily functioning were also reported by the study participants. Another double-blinded randomized clinical trial (RCT) of the effectiveness of smoked cannabis in the treatment of neuropathic pain found that 34% of the smoked cannabis group reported a decrease in daily pain as compared to 17% in the placebo group (p=0.03) (Abrams et al., 2007a). The authors also found pain reduction greater than 30% was reported in 52% of the smoked cannabis group as compared to 24% in the placebo group (p=0.04). A common criticism of these studies that involve smoked cannabis and the use of a placebo relates to how difficult it is to blind study participants when a smoked substance like cannabis is used. As such, there is a potential for reporting bias if participants have had any previous exposure to cannabis. 1.1.2 Safety of Cannabis There continues to be some debate about the safety and efficacy of prescribing medicinal cannabis within the scientific community (CMA, 2013). Short term adverse effects of cannabis use include: CNS depression, sedation, temporal and spatial distortion, impaired motor function, 8 impaired hand-eye coordination, short term memory impairment, mental confusion, anxiety and disorientation (Kalant, 2004). In addition, while the evidence is unclear related to the impact of long term cannabis use on the respiratory system, high-intensity or daily cannabis smoking has been associated with respiratory dysfunction in cannabis smokers (Hall & Degenhardt, 2009). Prescribing physicians may have concerns about safely prescribing cannabis when side effects like mental confusion and memory impairment may impact the patient?s ability to safely adhere to their medications. HIV-associated neurocognitive disorders (HAND), created when the HIV virus enters the CNS and damages nerves cells, is frequently found in PLWHA with more advanced HIV disease and can impact memory loss. One study examining memory impairment among PLWHA who use cannabis, compared subjects with symptoms associated with more advanced HIV, with those who did not display symptoms (Cristiani, Pukay-Martin & Bornstein, 2004). The authors found participants experiencing HIV symptoms experienced an increase in memory impairment compared to participants who did not report experiencing HIV symptoms. The study relied on small samples sizes and failed to control for the confounding effect of HAND within the study participants. There are also concerns related to cannabis being utilized for recreational or non-intended purposes and the potential for psychological addiction with chronic use (Bonn-Miller, Oser, Bucossi, & Traffton, 2012; Zvolensky et al., 2011). In addition to the acute side effects, chronic effects might also include bronchitis, emphysema, tachycardia, postural hypotension and decreased sperm counts (Crowford, 2003). A recent systematic review of safety studies of medicinal cannabis over the past 40 years examined 23 RCTs and 8 observational studies and concluded that with short term use the rates of non-serious adverse events were higher among medical cannabis users then controls (rate ratio = 1.86, 95% Confidence Interval (95% CI): 1.52-2.21) and the rates of serious adverse events did 9 not differ between groups (rate ratio= 1.04, 95% CI: 0.78-1.39) (Wang, Collet, Shapiro & Ware, 2008). The authors also identified that risks associated with long term use were poorly characterized and further research is required to help guide health policy and medical professionals. There are concerns that cannabis use may diminish the immune response in PLWHA. Cannabinoids (CB2) potent anti-inflammatory action may impair immune response by impacting inflammatory mechanisms, which help mediate immune response to viruses (Klein et al., 2003). Previous in vitro and in vivo studies examining the influence of cannabinoids on the immune system to viruses, have found that cannabinoids diminish the immune response to HIV (Reiss, 2010). The research on smoked cannabis and cannabinoids in human subjects is limited, however one observational study of medical and recreational use of cannabis amongst PLWHA, so found no significant differences in CD4+ cell counts amongst cannabis users (Furler et al., 2004). A few studies have examined the impact of cannabis use on ART drug levels and particularly PI medications. A study examining drug levels of atazanavir amongst 32 PLWHA cannabis users, found the median atazanavir trough concentrations were lower amongst cannabis users in comparison to non-users and that 50% of cannabis users had trough concentrations below the therapeutic range (Ma et al., 2009). The lower drug levels, however, did not have an impact on the clinical markers viral load or CD4 count in the sample. An additional RCT of the effects of smoked cannabis and dronabinol use on PLWHA receiving the PI?s indinavir and nelfinavir found the maximum blood plasma concentration of these drugs to be slightly lower amongst cannabis smokers (Kosel et al., 2002). However the author?s did not find concurrent dronabinol and cannabis use to have any clinically significant short-term effects or adverse impacts on plasma viral load and CD4 counts. In a short-term RCT of patients on the PI?s 10 indinavir and nelfinavir, where the authors examined the impact of cannabinoids on immune response, the results indicated there were no statistically significant decreases in CD4 and CD8 cell levels or plasma drug levels in patients using smoked cannabis or drabinol as compared to a placebo group (Abrams et al., 2003). Both groups using smoked cannabis or oral cannabanoid medications, showed an increase in CD4 and CD8 cell counts as compared to the placebo group. 1.1.3 Prevalence of Cannabis Use in PLWHA To date a number of observational studies have focused on the prevalence of cannabis use for symptom management in HIV populations. In a cross sectional study based on survey data from participants enrolled in the British Columbia Center for Excellence Drug Treatment Program (DTP) between October 1998 and September 1999, 14% of respondents reported using cannabis for medicinal purposes (Braitstein et al., 2001). The DTP dispenses medications for the majority of PLWHA in the province of British Columbia, so the sample reflects a good representation of PLWHA living in the province and on medications. A multivariate analysis of factors associated with cannabis use found that experiencing gastrointestinal side effects and peripheral neuropathy were positively and independently associated with cannabis use. Medicinal cannabis users were more likely to be male and younger in age. The data is based on figures from 1998-1999, shortly before the MMAR was introduced and may be out of date with current prevalence patterns. The authors also conclude that PLWHA are using cannabis to manage gastrointestinal side effects and peripheral neuropathy. These HIV-related side effects are frequently reported by PLWHA (Nicholas et al., 2007; Obrien, Clark, Besch, Lynn-Myers & Patricia, 2003). The study did not examine the recreational use of cannabis, cannabis use and adherence or the clinical characteristics of those who report using (CTP). 11 An additional cross-sectional study based in eastern Canada surveyed 160 participants recruited from HIV clinics in Toronto, Montreal and Ottawa and 19 participants from one compassion club about their use of cannabis and the synthetic cannabinoid dronabinol which is available in pill form. In the sample 59 participants (37%) reported current use of cannabis. One hundred and fourteen (70%) participants reported experiencing adverse effects related to ART, with 23 participants (14%) indicating these adverse effects negatively impacted their level of adherence. Just over a quarter of participants (28%) reported the use of cannabis or dronabinol to manage the side effects of ART. In addition to these findings, 7% of participants reported the use of cannabis or dronabinol improved their adherence to ART. Up to 37% of the participants also reported cannabis use for HIV symptom management (Ware et al., 2003). Participants reported using cannabis to effectively manage symptoms including stress relief, loss of appetite, weight loss, pain, and nausea and vomiting. Although study design limitations prevented causal relationships between cannabis use and improved adherence to HAART from being inferred, the potential supportive role of cannabis in symptom management and medication adherence in PLWHA is intriguing. The possible benefit, however, must be balanced against the side effects reported by study participants including euphoria (87%), dry mouth (63%), drowsiness (45%), paranoia and heart palpitations (26%) and anxiety (25%). A cross-sectional survey and retrospective chart review conducted in Ontario, Canada between 1999 and 2001 examined the prevalence and predictors of medicinal cannabis use amongst a convenience sample of 104 PLWHA (Furler, Einarson, Millson, Walmsley & Bendayan, 2004). In the sample, 43% of participants reported any cannabis use and 29% reported CTP. Of the CTP users the most common reasons provided for use included: appetite stimulation (70%); sleep/relaxation (37%); nausea/vomiting (33%); pain management (20%) and 12 anxiety/depression (20%). The recreational cannabis users were more likely to be male and have a history of intravenous drug use. In a multivariate analysis, only having a mean income less than $20,000 was found to be predictive of medicinal cannabis use. The authors also did not examine negative side effects of cannabis use. Another cross-sectional study conducted in Northern California, United States (US), documented the prevalence of cannabis use for symptom management amongst a convenience sample of 252 PLWHA recruited from 3 public health clinics, and found that 23% of participants reported using cannabis in the past month. Of those who reported recent cannabis use, 44% participants reported using cannabis to manage symptoms including nausea, anorexia, pain, anxiety and depression (Prentiss et al., 2004). Bivariate analysis identified participants experiencing moderate to severe nausea as being more likely to have used cannabis (Odds Ratio [OR] = 3.1, 95% CI: 1.6?5.9). The researchers did not examine whether the source of nausea was related to ART medications or symptoms of HIV disease. The authors also did not examine negative side effects of cannabis use. An additional study based in California, US examined a convenience sample of 1746 CTP users, and reported on therapeutic benefits of cannabis (Reinarman, Nunberg, Lanthier & Heddleston, 2011). The most common benefits reported were: pain relief (83%); improved sleep (71%); reduced muscle spasms (41%) and headaches (41%); anxiety relief (38%); improved appetite (38%); and decreased nausea and vomiting (28%). Another interesting reported benefit, related to the lack of universal medical coverage in the US, was the use of CTP as a substitute for prescription medication, by 51% of the participants. The author?s suggested that participants might be utilizing CTP because they lacked sufficient medical insurance to cover the cost of prescribed medications. The authors also discussed the potential for recreational users to 13 fabricate medical conditions or symptoms in order to obtain legally sanctioned access to cannabis. This would allow recreational consumers a way to avoid legal penalty. The authors suggested it is difficult to assess the magnitude of this phenomenon, but it is plausible given the harsh legal penalties associated with illegal use in the US. A cross-sectional study based in the United Kingdom (UK), where cannabis is not legal for medicinal purposes, found a significant number of PLWHA were using CTP. The sample consisted of 143 participants who attended a large outpatient clinic for HIV care and reported CTP to treat symptoms associated with HIV (Woolridge et al., 2005). Study participants reported the following symptoms were effectively treated via CTP: improved appetite, muscle pain, nausea, anxiety, nerve pain, and depression. The authors did not assess for negative side effects of cannabis use, but found 47% of respondents reported ?memory deterioration? due to cannabis use (Woolridge et al., 2005, p.361). The authors concluded that patient use of cannabis for symptom management is highly prevalent despite significant legal barriers in the U.K. 1.1.4 Symptom Management and Adherence to ART Given the importance of adherence to HAART medication in increasing the effectiveness of treatment and decreasing morbidity and mortality in PLWHA, identifying medications like cannabis that can ameliorate the side effects of HAART may help reduce the number of PLWHA who fail to achieve optimal adherence. As described earlier, side effects of ART medications, particularly nausea and vomiting, are frequently reported as a reason for poor adherence rates and even discontinuation of medication regimes altogether (O?Brien et al., 2003; Prentiss et al., 2004). To date, the evidence on cannabis use and adherence to ART is limited and contradictory. In the two Canadian studies previously discussed, researchers identified that PLWHA may be using cannabis to manage symptoms of ART medications. In Ware and colleagues? study (2003) 14 a small portion of PLWHA (7%) reported cannabis use helped to improve their medication adherence. A recent cross sectional study based in San Francisco, California compared cannabis use intensity, incidence of HIV symptoms and side effects to ART and adherence to ART medications (Bonn-Miller, Oser, Bucossi, & Traffton, 2012). The authors compared three groups, participants who used cannabis daily, less then daily and those who did not use cannabis. They found participants who reported daily or greater cannabis use had worse symptoms and poorer adherence than those who used cannabis weekly and those who did not use cannabis. The authors concluded it was unclear whether the intensity of cannabis use or the greater number of reported side effects and symptoms were the cause of ineffective adherence in the daily cannabis user group. The study was limited by its cross-sectional design and inconsistencies in ART adherence measurements between groups. The study also failed to include a number of potential confounders that have been found in previous studies of ART adherence to have a significant impact on effective adherence. A cross-sectional study of 178 PLWHA in Northern California found that 24% of the participants reported using cannabis in the previous month (de Jong et al., 2005). The authors did not find a statistically significant association between cannabis use and adherence (OR = 0.90, 95% CI: 0.4?1.9, p = 0.83), however adherence was negatively associated with illicit drug use and alcohol use. The authors further stratified cannabis use based on symptoms including moderate to severe nausea. Among those who reported moderate to severe nausea, cannabis users were more likely than non-cannabis users to report effective adherence to ART (OR 3.30, p= 0.07). The authors also found participants without nausea who used cannabis, reported lower rates of adherence when compared to non-cannabis users. The sample sizes for the regression analysis in this study were quite small possibly limiting the inclusion of factors associated with 15 adherence, including housing and incarceration. In addition the authors relied on a weak measurement of adherence, specifically self-reported adherence over the past week, with those who missed one dose considered non-adherent. This type of measurement has the potential to overestimate non-adherence and adherence rates (Kerr, Walsh, Lloyd-Smith & Wood, 2005). The authors also failed to stratify illicit drug use by specific drugs, which may have confounded the study findings. 1.1.5 Cannabis, Drug Use and Adherence to ART There is some concern within the HIV medical community that cannabis use may impede effective adherence to ART, by creating memory deficit and CNS disturbance. To date the evidence analysing the impact of cannabis use on adherence is limited and contradictory. In a cross-sectional study of adherence and sexual behaviour involving 255 PLWHA in the South Eastern U.S., participants who reported cannabis use in the past week reported poor adherence to ART (Kalichman & Rompa, 2003). The study did not include many personal and structural factors, which may have confounded the association between cannabis use and adherence. In addition, the authors did not find statistically significant differences in HIV clinical markers CD4+count and viral load between adherent and non-adherent groups, suggesting the ?self- report? adherence measurement used may not have resulted in accurate measurements. Another cross-sectional study of 2484 individuals based in France focused on drug use and adherence to HAART in a general population of PLWHA (Perreti-Watel, Spire, Lert & Obadia, 2006). The authors utilized data clustering to analyze multiple drug use patterns and found that cannabis use was associated with poor adherence when embedded with other drug use including heroin and alcohol use. The study relied on a poor measurement of adherence (i.e., 16 based on self-report data) and also failed to measure clinical factors such as CD4+ and viral load, which help support adherence data conclusions. An additional cross-sectional study involving 1910 participants receiving HIV-related care in US hospitals focused on non-adherence, mental health correlates and drug use (Tucker et al 2003). In a multivariate analysis of drug use variables and adherence the authors found cannabis use to be associated with non-adherence (OR 1.71, 95% CI: 1.22-2.31). In an additional analysis in which they examined any drug use, alcohol use, mental health variables and adherence to ART, the authors did not find illicit drug use to be statistically significant with poor adherence. A further study of 764 PWID in Baltimore, US, found no association between cannabis use and non-adherence (Lucas et al., 2001). Currently no studies have examined cannabis use and adherence in exclusively PWID populations in settings like Canada where access to ART is free and medically prescribed cannabis exists. In addition to this, the existing research has relied on cross-sectional observation, poor definitions and measures of adherence and have not controlled for other potential confounding variables, including various structural factors, which may be impacting ART adherence. 1.1.6 Barriers and Facilitators to Adherence amongst PLWHA Who Use Illicit Drugs A number of studies have identified barriers to optimal ART adherence in PWID populations. Active use of heroin, crack, and cocaine has been frequently but not consistently associated with poor adherence to ART among PWID participants (Crisp et al., 2004; Lucas et al.; 2001; Palepu et al., 2003; Peretti-Watel et al., 2006; Tucker et al., 2003). In addition to active drug use, alcohol use and depression in PWID subjects have been identified as factors associated with poor adherence (Bouhnik et al., 2005; Lima et al., 2007; Palepu et al., 2003). Many of these studies did not include cannabis as a potential predictor variable in their analyses. 17 Furthermore, many of the studies have taken place in environments where access to ART and HIV care can be negatively impacted by the cost of medications and healthcare, unlike the setting of this study. Further research is required to determine if an association between effective ART adherence and cannabis exists in PWID populations. A further limitation of studies that have examined ART adherence amongst PWID was the failure to include structural factors that may confound the association between drug use and adherence to ART. Structural factors including; homelessness, incarceration and limited access to primary care and HIV services, have been identified as barriers that influence adherence and HIV care amongst PLWHA populations and are frequently experienced by PWID. Recent studies in PWID populations have included structural factors that may influence adherence to ART. Homelessness has been identified as a structural factor that negatively influences ART adherence (Knowlton et al., 2006; Milloy, et al, 2012; Palepu et al., 2011) Recent incarceration was also identified as a factor that can negatively impact HAART adherence and clinical outcomes for HIV-positive PWID in studies in Canada and the U.S. (Milloy, et al. 2011; Palepu et al., 2003; Small, et al., 2009; Waldrop-Valverde & Valverde, 2005). In addition to this, PWID who were able to access drug treatment through methadone maintenance therapy (MMT) achieved higher rates of adherence (Malta et al., 2008; Palepu et al., 2006). These findings highlight how drug using populations can benefit greatly from comprehensive medical services that include addiction support programs, social support and housing support services and address the complexity of clients suffering from addiction and other comorbidities like HIV. While these studies have examined active drug use and structural factors that may influence adherence, they have not included cannabis as a potential variable of interest in their analyses. 18 In summary, current research on cannabis use in PLWHA has focused primarily on cannabis? role in symptom management. The majority of these studies have assessed self-reported cannabis use, in cross-sectional analyses and found PLWHA to report cannabis to be an effective medication for treatment of HIV-illness related symptoms. A few studies have examined the impact of cannabis use on adherence to ART, with some researchers reporting improved adherence and others identifying cannabis use associated with sub-optimal adherence. These studies have been limited by cross-sectional analyses, a limited number of confounding variables, poor measures of adherence, and a lack of clinical measures to support the validity of their findings (i.e., viral load). While practitioners are interested in identifying medications that support symptom management in PLWHA, the safety of prescribing cannabis and the impact on adherence is not well established. To date, no studies have examined the association between cannabis use and effective adherence to HARRT in a longitudinal cohort of PLWHA PWID. 1.1.7 Obtaining a Prescription for Cannabis: The Canadian Context In 2001 the Government of Canada created the Medical Marijuana Access Regulations that permitted seriously ill individuals, including PLWHA to access cannabis for medical purposes. The MMAR permits the use CTP by PLWHA for the following HIV-related symptoms: pain, nausea/vomiting, appetite and anorexia (Government of Canada, 2011). Surveys of PLWHA based in Canada have found reported medicinal cannabis use to range between 17 to 37% amongst PLWHA populations (Brainstein et al., 2001; Furler et al., 2005; Ware et al., 2003). Since its inception, the MMAR and its administrative arm the Medical Marijuana Access program (MMAP) have been criticized by physicians and CTP users. CTP users have been critical of the MMAP?s inability to provide a safe and good quality cannabis product and suggest the application process was inefficient and created a barrier for many CTP applicants (Lucas et 19 al., 2012). In surveys of PLWHA reporting CTP, a limited number have obtained a prescription for legal cannabis use and a majority continue to rely on illegal sources to obtain cannabis (Belle-Isle & Hathaway, 2007). In June 2013, the Government of Canada outlined a number of proposed changes to the MMAR including: the removal of the government`s role in the production and distribution of cannabis; the removal of the government`s role in providing authorization to individuals with healthcare practitioners would becoming the sole decision makers regarding eligibility for CTP; promoting the production of dried cannabis by licensed producers; and promoting the sale and distribution by specific regulated parties (Government of Canada, 2013).While a number of doctors have been supportive of PLWHA seeking legally sanctioned access to medicinal cannabis, a small number of PLWHA still face difficulties in obtaining support from their physicians and report having to visit a number of physicians before being granted access to CTP (Belle-Isle & Hathaway, 2007). Physician organizations have been critical of the MMAR and the proposed changes, suggesting until there is sufficient scientific evidence outlining the benefits, efficacy and potential long-term adverse effects of CTP it is improper for physicians to be the primary ?gatekeepers? of CTP access (CMA, 2013). CTP user groups have also been critical of the proposed changes, suggesting they will increase cannabis costs for CTP users and limit availability to dried cannabis, while potentially less harmful sources of cannabis including tinctures and edible forms will no longer be accessible (Canadian Association of Medical Cannabis Dispensaries, 2013). Given the historical reluctance of some physicians to authorize PLWHA to access CTP, reported barriers to access MMAP by CTP users and current policy shifts to alter the Government of Canada?s role in CTP, we sought to determine explanatory correlates of receiving 20 access to CTP in a population of PLWHA. While surveys of PLWHA based on relatively small sample sizes have outlined some of the demographic and symptom management details about medicinal cannabis users, to date few studies have examined the demographic, clinical and contextual factors associated with receiving a prescription for medical cannabis. In addition, very little research has examined the prevalence of being prescribed cannabis, explanatory correlates of prescribed cannabis use of PLWHA. As such, more research is required to determine what clinical and demographic characteristics? are associated with prescribed cannabis use, and the safety of cannabis in this population. 1.2 Purpose of the Study The primary purpose of this study was to determine whether an association exists between high-intensity or greater than daily cannabis use and adherence to ART in a population of PWID who are HIV-positive. Despite the evidence to support the efficacy of cannabis in the amelioration of HIV related symptoms, the evidence regarding cannabis use and the impact on adherence remains limited. To date, no long-term prospective studies assessing high?intensity cannabis use as a factor that may affect adherence have been conducted within a setting of universal healthcare where access to ART and HIV treatment is without user fees. The present study investigated the role of high-intensity or greater than daily cannabis use among a cohort of PWID, to determine if daily cannabis use was detrimental to ART adherence. A secondary goal of this study was to identify the prevalence and correlates of being prescribed cannabis among a cohort of HIV-positive PWID living in Vancouver, British Columbia. Further, this study examined the rates of cannabis use for HIV-related symptoms among those who are and who are not prescribed cannabis. The study aims to add to the growing body of research on the prevalence, correlates and impact of cannabis use in PLWHA. In 21 particular, the study aims to describe high?intensity cannabis use and the association with adherence to HAART in a population of PWID where access to ART is free and cannabis is accessible with a prescription from a healthcare practitioner. The study was guided by the following questions: 1. What is the prevalence and demographic, clinical and structural correlates of high-intensity cannabis use in PLWHA PWID? 2. What is the longitudinal association between high intensity cannabis use and HAART adherence in PLWHA who are PWID? 3. What are the demographic, clinical and structural explanatory correlates associated with receiving a prescription for cannabis within PLWHA population of PWID? 22 Chapter 2: Research Methods The following chapter describes the research methods utilized for both analyses. 2.1 Sample Data for this study was obtained from the AIDS Care Cohort to evaluate Exposure to Survival Services (ACCESS), an ongoing observational prospective cohort of over 900 HIV-positive illicit drug users. The study participants were recruited using community-based methods including snowball sampling and extensive outreach in Vancouver?s Downtown Eastside (DTES) and among local HIV/AIDS service organizations. Following recruitment and the provision of informed consent, participants complete an extensive interviewer-administered questionnaire, provide a blood sample to determine CD4 count and HIV-1 viral load, participate in a nurse-administered questionnaire and physical examination. Follow-up occurs semi-annually. ACCESS recruitment is ongoing and includes the following criteria: individuals must be HIV-infected, aged 18 years or older, English speaking, and having used illicit drugs other than cannabinoids in the month prior to enrollment. ACCESS has been approved by the University of British Columbia and Providence Healthcare Research Ethics Board. Prior to the baseline interview, participants were asked to read through the consent form and then a research staff member reviewed the form and explained each detail of the consent to participants, to ensure informed consent took place. Participants are also given the option of withdrawing their participation from the study at any point. Due to the sensitive subject matter discussed in the questionnaire, and to help mitigate the potential for discomfort and stress to participants during the interview, it is explained to participants they have the option of not answering questions if they create undue stress during the interview. 23 2.2 Data Collection At baseline and semi-annually, participants consent to participate in self-report interviews with trained interviewers and provide blood samples for CD4+ count, viral load, resistance and genotype testing. Two standardized interviewer-administered questionnaires are provided to the participants. One instrument elicits demographic data, information about drug use and sexual practices, exposures to the criminal justice and healthcare systems. A second interview, performed by a registered nurse, focuses on health-related outcomes including healthcare access and utilization, HIV management, and additional questions related to cannabis use. Most behavioural and health related outcomes recorded refer to the previous six month time period. In addition to the interviews, participants are examined by the nurse and provide blood samples for serological analysis. Participants are given a stipend ($20 CDN) at each study visit. Data from the semi-structured interviews is augmented with data on HIV care, ART medication exposure, and treatment/clinical outcomes obtained via a confidential linkage to the British Columbia Centre for Excellence in HIV/AIDS? Drug Treatment Program (BC-CfE), as described elsewhere (Wood et al., 2003a). The BC-CfE runs a centralized ART dispensary for the province of British Columbia, and provides ART dispensing information on each participant in the ACCESS study. In addition to this, the program also contains an HIV/AIDS monitoring lab that provides full retrospective and prospective CD4+ cell count and plasma HIV-1 RNA viral load observations for each participant in the study. Of note is the fact that all HIV care is provided free of charge to every PLWHA living in British Columbia, allowing analysis of adherence to treatment free of the confounding influence of the cost of medications and HIV care. 24 2.2.1 Data Measurement for ?High Intensity Cannabis Use and Adherence to ART? The main study outcome, adherence to prescribed ART, was based on pharmacy refill data and was defined as the number of days ART was dispensed divided by the number of days the participant was eligible for ART in the previous six months. The outcome was further dichotomized into optimal adherence versus sub-optimal adherence (?95% vs.<95%). This validated measure using pharmacy refill data has been previously shown to be strongly associated with both virological suppression and survival (Palepu et al. 2001; Wood et al., 2003b; Wood et al., 2003c). This study included all ACCESS participants who were exposed to ART, had a baseline CD4+ and viral load within ?180 days of recruitment, and contributed at least one follow-up interview after the baseline interview. Individuals who initiated ART following recruitment were added and the date of the first interview following initiation was the baseline date. The primary explanatory variable of interest was high-intensity cannabis use, defined as the self-report of at least daily cannabis use, in the previous six month period preceding the follow-up interview. In order to estimate the relationship between high intensity cannabis use and adherence to ART, a number of explanatory variables were identified as possibly confounding this relationship, including; age, gender (female/male), Aboriginal ancestry (yes/no), educational attainment (< high school diploma/?high school diploma or greater) and formal employment in the previous six months (yes/no). The formal employment variable was defined, as in previous studies (Richardson et al., 2010), as having a regular job with a salary or temporary work in the six-month period prior to the interview. Individual and illicit drug use variables included; methadone maintenance therapy (yes/no), frequent cocaine injection (?daily/4 drinks per day (yes/no)) and binge drug use (yes/no). The Center for Epidemiological studies Depression scale (CES-D) was used to measure depression. The variable was dichotomized (scores ?16/<16) with individuals who scored greater than 16 considered to be depressed. The CES-D has been shown to be a reliable and valid indicator of mild to severe depression with a reliability rating of Cronbach?s Alpha = 0.85 (Low-Beer et al., 2000b; Radloff, 1977). Clinical variables identified as potential confounders included CD4+ cell count (per 100 cells/ml), plasma HIV-1 RNA level (copies/ml, per log10 increase), time elapsed since HIV diagnosis (measured in months), ART regimen contains a protease inhibitor (PI) (yes/ no), and HIV experience of the prescribing physician (