Epidemiology and Public Health Oral Abstract Session #4
Tracks
Track 3
Sunday, April 28, 2024 |
8:45 - 10:15 |
Salon D&D1 |
Overview
Épidémiologie et santé publique séances de présentation orale d’abrégés #4
Speaker
Mélanie Bédard
Graduate Student
McGill University
Predictors of fatal drug poisoning (overdose) in a Canadian HIV-HCV co-infected cohort
Abstract
Background: HIV and Hepatitis C (HCV) co-infection disproportionately affects people who use drugs. Drug poisonings (overdose) in Canada cause a significant number of deaths annually. Investigating potential predictors for drug poisoning could help reduce excess mortality. Our aim is to predict 6-month drug poisoning mortality among people living with HIV-HCV co-infection using data from a national co-infection cohort.
Methods: Data from the Canadian Co-Infection Cohort (CCC), a study which has followed 2,000+ participants from across Canada living with HIV-HCV co-infection since 2003, were used. Participants were eligible for analysis if they ever reported injection or non-injection drug use. A wide array of sociodemographic and clinical covariates from the CCC were included, such as age, sex, ethnicity, income, frequency of drug use, use of harm reduction services, and HIV and HCV clinical measures. We used a weighted random forest model using a classification algorithm. Due to the imbalanced data, weights were added to the model to improve model performance.
Results: Of 2,132 total CCC participants, 1,998 met the eligibility criteria. Of those eligible, 1,799 (90%) reported using injection drugs and 1,879 (94%) reported using non-injection drugs. Drug poisoning was the most frequently recorded known cause of death. Of a total of 94 drug poisoning deaths, 53 drug poisoning deaths occurred within 6 months of a participant’s last visit. Predictors were ranked in order of importance. Predictors that were consistently ranked in the top third of variables were being on prescribed opioids, being on prescribed benzodiazepines, and sex work.
Conclusion: Uncovering important predictors of drug poisoning is the first step towards developing a prediction tool for use in clinical settings. However, at this time, our results cannot be generalized beyond our cohort given the subjectivity of the weighting approach and the small number of outcomes used to build models.
Methods: Data from the Canadian Co-Infection Cohort (CCC), a study which has followed 2,000+ participants from across Canada living with HIV-HCV co-infection since 2003, were used. Participants were eligible for analysis if they ever reported injection or non-injection drug use. A wide array of sociodemographic and clinical covariates from the CCC were included, such as age, sex, ethnicity, income, frequency of drug use, use of harm reduction services, and HIV and HCV clinical measures. We used a weighted random forest model using a classification algorithm. Due to the imbalanced data, weights were added to the model to improve model performance.
Results: Of 2,132 total CCC participants, 1,998 met the eligibility criteria. Of those eligible, 1,799 (90%) reported using injection drugs and 1,879 (94%) reported using non-injection drugs. Drug poisoning was the most frequently recorded known cause of death. Of a total of 94 drug poisoning deaths, 53 drug poisoning deaths occurred within 6 months of a participant’s last visit. Predictors were ranked in order of importance. Predictors that were consistently ranked in the top third of variables were being on prescribed opioids, being on prescribed benzodiazepines, and sex work.
Conclusion: Uncovering important predictors of drug poisoning is the first step towards developing a prediction tool for use in clinical settings. However, at this time, our results cannot be generalized beyond our cohort given the subjectivity of the weighting approach and the small number of outcomes used to build models.
Mariam El Sheikh
Graduate Student
McGill University
Factors associated with not receiving optimal care (NROC) for Hepatitis C (HCV) in the Canadian Co-infection Cohort (CCC)
Abstract
Background: HCV micro-elimination involves targeting priority populations to improve the care cascade. People living with HIV-HCV co-infection are important candidates for micro-elimination due to faster liver disease progression and existing linkages to HIV care. Not receiving optimal care(NROC) remains a major challenge for achieving HCV elimination- different groups potentially face different challenges at each care step.
Methods: The CCC, an open cohort of people with co-infection in care across Canada with bi-annual visits was used to identify factors associated with NROC at various cascade steps (2014-2019). NROC, was defined differently at each cascade step, based on what is considered optimal care from a system perspective:
1) failure to initiate treatment within 1-year of diagnosis (Step1)
2) absence of a negative RNA result within 1-year of treatment completion (Step2)
3) absence of a negative RNA result 1-year post-sustained virologic response (SVR;Step3)
Logistic regression models with a random-intercept term for clinic were used for each step. Results are conditional on completing previous steps.
Results: Among 856, 643, and 432 participants, 67%(n=576), 15%(n=97), and 44%(n=191) experienced NROC in Steps 1, 2, and 3, respectively. We observed different patterns across the three steps such as:
1) Indigenous ethnicity, previous incarceration, hazardous-drinking, and low-income were associated with NROC in Step 1.
2) Injection drug use and men who have sex with men(MSM) were associated with NROC in Step 3.
3) Uncontrolled HIV infection, and unknown status of hazardous-drinking were associated with NROC across all steps.
Conclusions: Different groups face different barriers to accessing/staying in care at different stages emphasizing the importance of tailoring interventions to their needs. Markers of vulnerability were associated with NROC in Step 1 but once treatment is initiated, people can be retained in care (Step 2). Ensuring people who inject drugs and MSM remain in care post-SVR (Step 3) is needed.
Methods: The CCC, an open cohort of people with co-infection in care across Canada with bi-annual visits was used to identify factors associated with NROC at various cascade steps (2014-2019). NROC, was defined differently at each cascade step, based on what is considered optimal care from a system perspective:
1) failure to initiate treatment within 1-year of diagnosis (Step1)
2) absence of a negative RNA result within 1-year of treatment completion (Step2)
3) absence of a negative RNA result 1-year post-sustained virologic response (SVR;Step3)
Logistic regression models with a random-intercept term for clinic were used for each step. Results are conditional on completing previous steps.
Results: Among 856, 643, and 432 participants, 67%(n=576), 15%(n=97), and 44%(n=191) experienced NROC in Steps 1, 2, and 3, respectively. We observed different patterns across the three steps such as:
1) Indigenous ethnicity, previous incarceration, hazardous-drinking, and low-income were associated with NROC in Step 1.
2) Injection drug use and men who have sex with men(MSM) were associated with NROC in Step 3.
3) Uncontrolled HIV infection, and unknown status of hazardous-drinking were associated with NROC across all steps.
Conclusions: Different groups face different barriers to accessing/staying in care at different stages emphasizing the importance of tailoring interventions to their needs. Markers of vulnerability were associated with NROC in Step 1 but once treatment is initiated, people can be retained in care (Step 2). Ensuring people who inject drugs and MSM remain in care post-SVR (Step 3) is needed.
Alexander Korzuchowski
Research Coordinator
BC Centre For Excellence In HIV/AIDS
Hepatitis C therapeutic outcomes among people concurrently living with HIV: Examining the impact of modern DAA therapy in British Columbia
Abstract
Objective: Hepatitis C (HCV) therapeutic outcomes have improved after introduction of short-course, pan-genotypic, oral direct-acting antivirals (DAA) with clearance rates of >95%. Despite their effectiveness, DAA treatment access varies among key populations. This study describes DAA treatment outcomes among a population-based cohort of people living with HIV (PLWH) in British Columbia (BC) from 2015 onwards.
Methods: We utilized data from the Seek and Treat for Optimal Prevention of HIV/AIDS (STOP HIV/AIDS) cohort of PLWH between April 1996 and March 2020 in BC, Canada. PLWH aged ≥ 19 years old with an HCV treatment record from January 1 2015 onward and ending on or before March 31 2019 were included with a one-year follow-up for therapeutic outcome and post-treatment measures. First HCV treatment was identified using DAA dispensation from PharmaNet. Therapeutic outcome, SVR (sustained virologic response), was defined using Drug Treatment Program registry laboratory results (10 to 52-week cut-off after treatment end) from the BC Centre for Excellence in HIV/AIDS.
Results: The study included 789 PLWH engaged in HCV treatment in BC from January 2015 to March 2019. Median age at start of treatment was 52 (Q1-Q3: 46-57). Table 1 shows demographic characteristics stratified by SVR achievement. Most common DAA regimens were ledipasvir/sofosbuvir (n=371) and sofosbuvir/velpatasvir (n=274).
Discussion: Findings show descriptive statistics of HCV treatment success among key populations in the STOP/HIV AIDS cohort. Future analysis will model key characteristics associated with achieving SVR. A secondary analysis will test for associations between HCV treatment engagement and HIV treatment adherence.
Methods: We utilized data from the Seek and Treat for Optimal Prevention of HIV/AIDS (STOP HIV/AIDS) cohort of PLWH between April 1996 and March 2020 in BC, Canada. PLWH aged ≥ 19 years old with an HCV treatment record from January 1 2015 onward and ending on or before March 31 2019 were included with a one-year follow-up for therapeutic outcome and post-treatment measures. First HCV treatment was identified using DAA dispensation from PharmaNet. Therapeutic outcome, SVR (sustained virologic response), was defined using Drug Treatment Program registry laboratory results (10 to 52-week cut-off after treatment end) from the BC Centre for Excellence in HIV/AIDS.
Results: The study included 789 PLWH engaged in HCV treatment in BC from January 2015 to March 2019. Median age at start of treatment was 52 (Q1-Q3: 46-57). Table 1 shows demographic characteristics stratified by SVR achievement. Most common DAA regimens were ledipasvir/sofosbuvir (n=371) and sofosbuvir/velpatasvir (n=274).
Discussion: Findings show descriptive statistics of HCV treatment success among key populations in the STOP/HIV AIDS cohort. Future analysis will model key characteristics associated with achieving SVR. A secondary analysis will test for associations between HCV treatment engagement and HIV treatment adherence.
Sarah Kelly
Research Coordinator
British Columbia Centre for Excellence in HIV/AIDS
Beyond direct-acting antiviral therapy: Characterizing mental health conditions and depressive symptoms among patients recently treated for hepatitis C virus (HCV)
Abstract
Background: While hepatitis C virus (HCV) treatments have become increasingly tolerable and effective, it is unclear whether patients undergoing HCV treatment are receiving adequate support to engage in healthcare for comorbid conditions. The objective of this study was to characterize psychiatric disorders and depressive symptomatology among a cohort of patients recently treated for HCV.
Methods: Data from the Preservation of Sustained Virologic Response (Per-SVR) study, a prospective cohort of individuals aged 19 or older who achieved SVR following treatment for HCV in British Columbia (BC), were used for this analysis. Participants were enrolled in the study within three months post-HCV treatment and completed an interviewer-administered survey. We used self-reported lifetime psychiatric diagnoses to characterize mental health disorders. Depressive symptoms were assessed through a validated depression scale (CES-D 10). Multivariable logistic regression modelling was used to assess factors associated with significant depressive symptoms.
Results: Of 256 participants, 142 (55%) reported having been diagnosed with at least one psychiatric disorder, including 86 (34%) with depression, and 49 (19%) with anxiety. Using the CES-D 10, we found that 122 (48%) had significant depressive symptoms. Of those, 82 (67%) reported ever being diagnosed with a psychiatric disorder and 64 (52%) reported ever having accessed psychiatric treatment or care. Factors associated with significant depressive symptoms were illicit drug use (odds ratio [OR]: 3.51; 95%CI: 1.36-9.11), intimate relationship dissatisfaction (OR: 3.43; 95%CI: 1.62-7.24), unstable housing (OR: 2.48; 95%CI: 1.37-4.5), experiencing barriers to healthcare (OR: 2.27; 95%CI: 1.02-5.03), and poorer quality of life (OR: 0.03 [per unit]; 95%CI: 0.01-0.17).
Conclusion: We observed a high prevalence of psychiatric disorders among patients recently treated for HCV, with almost half currently experiencing symptoms of depression. Health services that incorporate the provision of mental health care alongside HCV treatment may alleviate barriers to care and improve health outcomes among HCV-affected populations.
Methods: Data from the Preservation of Sustained Virologic Response (Per-SVR) study, a prospective cohort of individuals aged 19 or older who achieved SVR following treatment for HCV in British Columbia (BC), were used for this analysis. Participants were enrolled in the study within three months post-HCV treatment and completed an interviewer-administered survey. We used self-reported lifetime psychiatric diagnoses to characterize mental health disorders. Depressive symptoms were assessed through a validated depression scale (CES-D 10). Multivariable logistic regression modelling was used to assess factors associated with significant depressive symptoms.
Results: Of 256 participants, 142 (55%) reported having been diagnosed with at least one psychiatric disorder, including 86 (34%) with depression, and 49 (19%) with anxiety. Using the CES-D 10, we found that 122 (48%) had significant depressive symptoms. Of those, 82 (67%) reported ever being diagnosed with a psychiatric disorder and 64 (52%) reported ever having accessed psychiatric treatment or care. Factors associated with significant depressive symptoms were illicit drug use (odds ratio [OR]: 3.51; 95%CI: 1.36-9.11), intimate relationship dissatisfaction (OR: 3.43; 95%CI: 1.62-7.24), unstable housing (OR: 2.48; 95%CI: 1.37-4.5), experiencing barriers to healthcare (OR: 2.27; 95%CI: 1.02-5.03), and poorer quality of life (OR: 0.03 [per unit]; 95%CI: 0.01-0.17).
Conclusion: We observed a high prevalence of psychiatric disorders among patients recently treated for HCV, with almost half currently experiencing symptoms of depression. Health services that incorporate the provision of mental health care alongside HCV treatment may alleviate barriers to care and improve health outcomes among HCV-affected populations.
David Moore
Research Scientist
Bc Centre For Excellence In Hiv/aids
Hepatitis C virus incidence among gay, bisexual and other men who have sex with men in Vancouver, Toronto and Montreal: 2017 - 2023.
Abstract
Introduction: Gay, bisexual and other men who have sex with men (GBM) are a priority population for microelimination of hepatitis C virus (HCV) in Canada. Previous research has demonstrated high levels of treatment uptake among HCV-positive GBM in Montreal, Toronto and Vancouver which should limit HCV transmission risk. We calculated HCV incidence and related correlates among HCV antibody negative GBM recruited in all three cities.
Methods: Sexually active GBM, aged ≥16 years, were recruited through respondent-driven sampling from 02/2017 to 08/2019. Participants completed a computer-assisted self-interview and tests for HIV, HCV, and other sexually transmitted infections at enrolment and every 6–12 months until Feb 28, 2023. Using pooled three-city data, we calculated HCV seroincidence and used multivariable generalized linear mixed modelling to examine associations with new HCV seropositivity.
Results: We recruited 941 HCV antibody-negative participants in Montreal, 378 in Toronto, and 575 in Vancouver with at least one follow-up visit. A total of 19 participants developed HCV seropositivity over a median of 4.21 years (Q1-Q3 3.14–4.61) for an incidence of 0.27 per 100 person-years. New HCV seropositivity was associated with living with HIV (Adjusted Incident Rate Ratio [aIRR]= 5.13; 95% CI 2.01-13.1) older age (aIRR= 1.05; 95% CI 1.02-1.09 per year) and number of anal sex partners in the previous six months(P6M) (aIRR=1.01 per partner; 95% CI 1.01-1.02) but not P6M injection drug use (IRR 0.19; 95% CI 0.03-1.36). We found no differences in seroincidence by city: IRR=0.20 for Toronto (95% CI 0.03-1.31) and IRR=0.46 for Vancouver (95% CI 0.15-1.44) compared to Montreal.
Conclusion: Among urban GBM, we found very low HCV seroincidence. However, older GBM, those living with HIV and those with more anal sex partners are at higher risk. The latter finding and the lack of association with injection drug use suggests that transmission is likely through sex.
Methods: Sexually active GBM, aged ≥16 years, were recruited through respondent-driven sampling from 02/2017 to 08/2019. Participants completed a computer-assisted self-interview and tests for HIV, HCV, and other sexually transmitted infections at enrolment and every 6–12 months until Feb 28, 2023. Using pooled three-city data, we calculated HCV seroincidence and used multivariable generalized linear mixed modelling to examine associations with new HCV seropositivity.
Results: We recruited 941 HCV antibody-negative participants in Montreal, 378 in Toronto, and 575 in Vancouver with at least one follow-up visit. A total of 19 participants developed HCV seropositivity over a median of 4.21 years (Q1-Q3 3.14–4.61) for an incidence of 0.27 per 100 person-years. New HCV seropositivity was associated with living with HIV (Adjusted Incident Rate Ratio [aIRR]= 5.13; 95% CI 2.01-13.1) older age (aIRR= 1.05; 95% CI 1.02-1.09 per year) and number of anal sex partners in the previous six months(P6M) (aIRR=1.01 per partner; 95% CI 1.01-1.02) but not P6M injection drug use (IRR 0.19; 95% CI 0.03-1.36). We found no differences in seroincidence by city: IRR=0.20 for Toronto (95% CI 0.03-1.31) and IRR=0.46 for Vancouver (95% CI 0.15-1.44) compared to Montreal.
Conclusion: Among urban GBM, we found very low HCV seroincidence. However, older GBM, those living with HIV and those with more anal sex partners are at higher risk. The latter finding and the lack of association with injection drug use suggests that transmission is likely through sex.
Alannah Hannigan
Clinical Research Coordinator
British Columbia Centre For Excellence in HIV/AIDS
Reducing Gaps In The HCV Cascade-Of-Care Among People Who Use Drugs: Evaluation And Application Of Lessons Learned From A Health Services Implementation Pilot Study
Abstract
Background:
Enhancing access to hepatitis C virus (HCV) care and treatment for people who use drugs (PWUD) is critical to eliminating HCV as a public health threat. However, the actions required to improve treatment uptake remain unclear. The Hep C Connect (HCC) study is a nurse-led HCV testing and linkage-to-care intervention developed for clients of a supervised consumption site (SCS) in Vancouver. In this study, we examine opportunities to improve the HCV cascade-of-care among PWUD.
Methods:
The HCC pilot ran from November 2021-October 2023. Participants (SCS clients ages 19+) completed point-of-care HCV antibody (Ab) testing and an interviewer-administered survey. The survey captured experiences with healthcare, drug-use, HCV testing and treatment history, and HCV knowledge. In this analysis, we characterize HCV knowledge (validated 8-item questionnaire) and progress across the HCV cascade-of-care.
Results:
The HCC study engaged 188 participants (median age of 42 (34, 50); 31% identifying as women); 35% reported not having a current primary healthcare provider. Knowledge of HCV was high among participants (median score of 7/8 correct responses). We found a substantial proportion of participants were HCV Ab+: 111/188 (59%) participants had a reactive HCV Ab test. Of those, 62 (55.9%) chose to engage in confirmatory RNA testing, and 30 (27%) returned a positive HCV RNA result. Despite 95% of participants knowing that untreated HCV can result in liver failure and 92% knowing there is a curative treatment for HCV, 44% of participants chose not to engage in RNA confirmatory testing following an HCV Ab+ result.
Conclusion:
The HCC intervention successfully engaged a population with high HCV sero-prevalence and found high HCV knowledge. However, almost half of the HCV Ab+ participants chose not to engage in confirmatory testing. Our results suggest that areas for further intervention should focus on facilitating linkages-to-care and reducing barriers to testing and clinical engagement.
Enhancing access to hepatitis C virus (HCV) care and treatment for people who use drugs (PWUD) is critical to eliminating HCV as a public health threat. However, the actions required to improve treatment uptake remain unclear. The Hep C Connect (HCC) study is a nurse-led HCV testing and linkage-to-care intervention developed for clients of a supervised consumption site (SCS) in Vancouver. In this study, we examine opportunities to improve the HCV cascade-of-care among PWUD.
Methods:
The HCC pilot ran from November 2021-October 2023. Participants (SCS clients ages 19+) completed point-of-care HCV antibody (Ab) testing and an interviewer-administered survey. The survey captured experiences with healthcare, drug-use, HCV testing and treatment history, and HCV knowledge. In this analysis, we characterize HCV knowledge (validated 8-item questionnaire) and progress across the HCV cascade-of-care.
Results:
The HCC study engaged 188 participants (median age of 42 (34, 50); 31% identifying as women); 35% reported not having a current primary healthcare provider. Knowledge of HCV was high among participants (median score of 7/8 correct responses). We found a substantial proportion of participants were HCV Ab+: 111/188 (59%) participants had a reactive HCV Ab test. Of those, 62 (55.9%) chose to engage in confirmatory RNA testing, and 30 (27%) returned a positive HCV RNA result. Despite 95% of participants knowing that untreated HCV can result in liver failure and 92% knowing there is a curative treatment for HCV, 44% of participants chose not to engage in RNA confirmatory testing following an HCV Ab+ result.
Conclusion:
The HCC intervention successfully engaged a population with high HCV sero-prevalence and found high HCV knowledge. However, almost half of the HCV Ab+ participants chose not to engage in confirmatory testing. Our results suggest that areas for further intervention should focus on facilitating linkages-to-care and reducing barriers to testing and clinical engagement.