POS0521 RISKS OF SEVERE INFECTION AFTER THE INTRODUCTION OF bDMARDs IN NEWLY DIAGNOSED RHEUMATOID ARTHRITIS PATIENTS: A POPULATION-BASED INTERRUPTED TIME-SERIES ANALYSIS

Background

Biological disease-modifying anti-rheumatic drugs (bDMARDs) are effective in suppressing inflammation and preventing joint damage. But bDMARDs may be associated with increased risk of severe infection. Evidence on this is contradictory with some studies showing increased risk, whereas others reporting no significant changes.

Objectives

To determine the impact of the introduction of bDMARDs on severe infection among patients newly diagnosed with RA compared with non-RA individuals.

Methods

In this age- and gender-matched cohort study using administrative health data for the population of BC, Canada, all incident RA patients diagnosed between 1995–2007 were identified. Non-RA individuals were randomly selected from the general control population to match with RA. Incident RA/non-RA individuals were then divided into quarterly cohorts according to their diagnosis date. Two outcomes were examined: (1) first severe infection (FSI) after RA onset necessitating hospitalization or occurring during hospitalization; and (2) all severe infections (ASI) after RA onset. We calculated the 8-year FSI and ASI rate for each cohort. We conducted interrupted time-series analyses to compare levels and trends of FSI and ASI in RA and non-RA individuals diagnosed during pre-bDMARDs (1995–2001) and post-bDMARDs (2003–2007) periods. Adjusted 8-year FSI and ASI rates for RA and non-RA cohorts diagnosed five years after bDMARDs introduction were compared with expected rates assuming no bDMARDs introduction, based on extrapolation of pre-bDMARDs trends.

Results

A total of 60,226 and 588,499 incident RA/non-RA individuals were identified. We identified 8,954 FSI and 14,245 ASI in RA, and 56,153 FSI and 79,819 ASI in non-RA. The 8-year FSI rates among RA patients diagnosed in the pre-bDMARDs period decreased over time but leveled off among those diagnosed in the post-period (Figure 1). The adjusted difference between the post- and pre-bDMARDs secular trends of 8-year FSI rates was 0.68 (p=0.03) in RA and 0.03 (p=0.67) in non-RA (Table 1). The 8-year ASI rates among RA patients diagnosed in the pre-bDMARDs period decreased over time but increased significantly among those diagnosed in the post-period (Figure 1). The adjusted difference between the post- and pre-bDMARDs secular trends of 8-year ASI rates was 1.85 (p=0.001) in RA and 0.12 (p=0.29) in non-RA (Table 1). For RA cohort diagnosed 5 years after bDMARDs introduction, ASI rate increased by 20.4% than expected rates assuming no bDMARDs introduction. In contrast, ASI rate in non-RA increased by only 10.9%.

Table 1.

Results of interrupted time-series analysis of FSI/ASI rates, adjusting for age, gender, chronic obstructive pulmonary disease, Romano Charlson Comorbidity Index, diabetes, chronic kidney diseases, alcoholism, cancer, prior hospitalization with infection and socio-economic status at disease diagnosis year, using stepwise model selection

OutcomeParameterRANon-RAUnadj. Diff (95% CI)Adj. Diff (95% CI)Unadj. Diff (95% CI)Adj. Diff (95% CI)p-valuep-valuep-valuep-valueFSITrend0.63 (0.03, 1.22) 0.04410.68 (0.09, 1.27) 0.02920.08 (-0.08, 0.25) 0.32370.03 (-0.12, 0.19) 0.6728Level (1 year post-intervention)0.50 (-2.00, 2.99) 0.69890.31 (-1.88, 2.49) 0.78470.41 (-0.21, 1.03) 0.20410.26 (-0.24, 0.75) 0.31035 years post-intervention3.01 (-0.85, 6.87) 0.13313.02 (-0.48, 6.52) 0.09860.75 (-0.24, 1.73) 0.14330.39 (-0.46, 1.25) 0.3721ASITrend1.84 (0.83, 2.84) 0.00091.85 (0.81, 2.89) 0.00110.28 (0.04, 0.53) 0.03050.12 (-0.10, 0.34) 0.2877Level (1 year post-intervention)-1.21 (-5.41, 3.00) 0.5763-1.44 (-5.44, 2.56) 0.48501.46 (0.42, 2.49) 0.00851.20 (0.38, 2.02) 0.00645 years post-intervention6.14 (0.26, 12.01) 0.04665.97 (0.02, 11.93) 0.05602.60 (1.08, 4.12) 0.00171.69 (0.45, 2.92) 0.0109Figure 1.

Unadjusted rates.

Conclusion

Arthritis onset after bDMARDs introduction is associated with an elevated risk of severe infection in RA patients, compared with matched non-RA individuals.

Acknowledgements

We would like to thank the Ministry of Health of British Columbia and Population Data BC for providing access to the administrative data. All inferences, opinions, and conclusions drawn in this publication are those of the authors, and do not reflect the opinions or policies of the Data Stewards or the [British Columbia] Ministry of Health. No personal identifying information was made available as part of this study. Procedures used were in compliance with British Columbia’s Freedom in Information and Privacy Protection Act. Ethics approval was obtained from the University of British Columbia’s Behavioral Research Ethics Board (H15-00887).

Disclosure of Interests

None declared.

POS0503 RISKS OF CARDIOVASCULAR EVENTS AFTER THE INTRODUCTION OF bDMARDs IN NEWLY DIAGNOSED RHEUMATOID ARTHRITIS PATIENTS: A POPULATION-BASED INTERRUPTED TIME-SERIES ANALYSIS

Background

Rheumatoid arthritis (RA) is associated with increased risk of cardiovascular (CV) events. Biological disease-modifying anti-rheumatic drugs (bDMARDs) are effective in suppressing inflammation and preventing joint damage and may help lower the risk of CV events. However, recent epidemiological studies have shown mixed results with some suggesting a lower risk of CV events, while others reporting no significant differences.

Objectives

To determine the impact of the introduction of bDMARDs on incident cardiovascular disease (CVD) among patients newly diagnosed with RA compared with matched non-RA individuals.

Methods

In this age- and gender-matched cohort study using administrative health data for the population of BC, Canada, all incident RA patients diagnosed between 1995–2007 were identified. Non-RA individuals were randomly selected from the general control population to match with RA. Incident RA and non-RA individuals were then divided into quarterly cohorts according to their diagnosis date. The outcome of interest was incident CVD event after RA onset, which include acute myocardial infarction, cerebrovascular accident, and venous thromboembolism. We calculated the 8-year incident CVD rate for each cohort. We conducted interrupted time-series analyses to compare levels and trends of CVD in RA and non-RA individuals diagnosed during pre-bDMARDs (1995–2001) and post-bDMARDs (2003–2007) periods with intervention time set at year of 2002. Adjusted 8-year CVD rates for RA and non-RA cohorts diagnosed five years after bDMARDs introduction were compared with expected rates assuming no bDMARDs introduction, based on extrapolation of pre-bDMARDs trends.

Results

A total of 60,226 and 588,499 incident RA and non-RA individuals were identified. We identified 6,740 and 48,653 incident CVD events in total in RA and non-RA individuals, respectively. We observe no change in the secular trends of the 8-year CVD rates in both RA and non-RA individuals diagnosed in pre- and post-bDMARDS periods (Figure 1): the adjusted difference between the post- and pre-bDMARDs secular trends of 8-year CVD rates was 0.23 (p=0.26) for RA patients and -0.07 (p=0.33) for non-RA individuals (Table 1). However, we observed a reduction in the level of CVD rates among RA patients diagnosed in the post-bDMARDs period and no change in non-RA (Figure 1): the adjusted difference in level comparing points immediately before and after the intervention, and accounting for pre-intervention trend was -1.61 (p=0.03) in RA, while it was -0.02 (p=0.93) in non-RA (Table 1).

Table 1.

Results of interrupted time-series analysis of incident CVD rates, adjusting for age, gender, chronic obstructive pulmonary disease, Romano Charlson Comorbidity Index, diabetes, angina, hypertension, chronic kidney disease, peripheral vascular disease, atrial fibrillation, glucocorticoid, non-steroidal anti-inflammatory drugs, CVD medications, fibrates, contraceptives, and aspirin use at disease diagnosis year, using stepwise model selection

OutcomeParameterRANon-RAUnadj. Diff (95% CI)Adj. Diff (95% CI)Unadj. Diff (95% CI)Adj. Diff (95% CI)p-valuep-valuep-valuep-valueCVDTrend-0.15 (-0.72, 0.42) 0.60860.23 (-0.17, 0.64) 0.2620-0.09 (-0.27, 0.08) 0.3084-0.07 (-0.20, 0.07) 0.3290Level (1 year post-intervention)-1.36 (-3.17, 0.45) 0.1474-1.61 (-2.96, -0.25) 0.02510.22 (-0.41, 0.84) 0.5011-0.02 (-0.44, 0.40) 0.93455 years post-intervention-1.96 (-4.48, 0.55) 0.1332-0.67 (-2.80, 1.46) 0.5418-0.15 (-1.06, 0.75) 0.7421-0.29 (-0.97, 0.39) 0.4102Figure 1.

Unadjusted rates.

Conclusion

Arthritis onset after bDMARDs introduction is associated with a significant reduction in the risk of incident CVD events among RA patients, but not in the matched non-RA individuals.

Acknowledgements

We would like to thank the Ministry of Health of British Columbia and Population Data BC for providing access to the administrative data. All inferences, opinions, and conclusions drawn in this publication are those of the authors, and do not reflect the opinions or policies of the Data Stewards or the [British Columbia] Ministry of Health. No personal identifying information was made available as part of this study. Procedures used were in compliance with British Columbia’s Freedom in Information and Privacy Protection Act. Ethics approval was obtained from the University of British Columbia’s Behavioral Research Ethics Board (H15-00887).

Disclosure of Interests

None declared.

Development and validation of health system performance measures for opioid use disorder in British Columbia, Canada

BACKGROUND

Performance measurement provides an evidence-based means to inform development of interventions to improve the quality of care for people who use opioids. We aimed to develop and assess the predictive validity of health system performance measures for opioid use disorder (OUD) in British Columbia (BC), Canada.

METHODS

Performance measures were generated using retrospective population-level administrative datasets (both provincial and regional) and publicly-reported retrospective data according to four domains (care engagement, clinical guideline compliance, integration, and healthcare utilization). The adjusted odds ratio was estimated via generalized linear mixed models to determine predictive validity for all-cause hospitalization or mortality within 6 months of measurement.

FINDINGS

A total of 102 performance measures were constructed. We identified 55,470 diagnosed PWOUD, and 39,456 ever engaged in opioid agonist treatment (OAT). We found divergent rates of treatment for concurrent conditions (7.4% for alcohol use disorder to 80.1% for HIV/AIDS), low levels of linkage to OAT and other outpatient care following acute care, and increasing levels of service provision, including increases in OAT prescribers and pharmacies, naloxone kit distribution and overdose prevention site visitation. Our analyses on the predictive validity measures largely supported a priori hypotheses on the direction of effect on the outcome.

CONCLUSIONS

We identified a range of priorities to improve the quality of care for PWOUD, with critical gaps in linkage to care through acute care settings and long-term engagement in OAT. The proposed measures can be derived for geographic and clinical subgroups and updated over time, providing a basis to monitor and evaluate efforts to address the public health burden of OUD.

Determinants of selection into buprenorphine/naloxone among people initiating opioid agonist treatment in British Columbia

BACKGROUND

Studies assessing the comparative effectiveness of methadone versus buprenorphine/naloxone for opioid use disorder in real-world settings are rare - challenged by structural differences in delivery across settings and factors influencing treatment selection. We identified determinants of selection into buprenorphine/naloxone and quantified contributions of individual and provider-level covariates in a setting delivering both medications within the same healthcare settings.

METHODS

Utilizing linked health administrative datasets, we conducted a retrospective cohort study of people with opioid use disorder (PWOUD) receiving opioid agonist treatment (OAT) in British Columbia, Canada, from 2008-2017. Determinants of buprenorphine/naloxone selection were identified using a generalized linear mixed model with random intercept terms for providers and individuals. We determined the influence of individual demographics, clinical history, measures of provider experience and preference, and dates of key policy changes.

RESULTS

A total of 39,605 individuals experienced 178,976 OAT episodes (methadone:139,439(77.9 %);buprenorphine/naloxone:39,537(22.1 %)). Male sex, less OAT experience, younger age, mental health conditions and chronic pain were associated with higher odds of buprenorphine/naloxone prescription. For providers, higher client-attachment, more complex OAT case-mixes, and higher buprenorphine/naloxone prescribing-preference were also associated with higher odds of buprenorphine/naloxone prescription. Observed individual-level covariates explained 9.7 % of variance in odds of buprenorphine/naloxone selection, while observed provider-level covariates explained 20.0 %. Controlling for covariates, residual unmeasured between-individual variance accounted for 18.5 % of the explained variation in the odds of buprenorphine/naloxone selection, while unmeasured between-provider variance accounted for 28.4 %.

CONCLUSION

Provider characteristics were more influential in selection of buprenorphine/naloxone over methadone informing subsequent analyses of comparative effectiveness of these regimens.

Short term health-related quality of life improvement during opioid agonist treatment

BACKGROUND

Opioid dependence is associated with high levels of morbidity, yet sparse data exists regarding the health-related quality of life (HRQoL) of individuals with opioid dependence, particularly following treatment initiation. To inform cost-effectiveness analyses of treatment modalities, this study investigates short-term changes in HRQoL following enrollment into opioid agonist treatment (OAT), across treatment modalities and patient subgroups.

METHODS

Data was analyzed from the Starting Treatment with Agonist Replacement Therapies (START) and Prescription Opioid Addiction Treatment Studies (POATS) randomized controlled trials. Participants included individuals dependent on prescription opioids (POs) or heroin, receiving limited-term or time-unlimited treatment. PO- or heroin-users in START received buprenorphine/naloxone (BUP/NX) or methadone (MET) over 24 weeks. PO-users in POATS received psychosocial care and short-term (4-week) taper with BUP/NX, with non-responders offered subsequent extended (12-week) stabilization and taper. HRQoL was assessed using the short-form SF-6D while in and out of OAT, with distinction between MMT and BUP/NX in START. Linear mixed effects regression models were fitted to determine the independent effects of OAT on HRQoL and characterize HRQoL trajectories.

RESULTS

Treatment had a similar immediate and modest positive association with HRQoL in each patient subgroup. The association of OAT on HRQoL was statistically significant in each model, with effect sizes between 0.039 (heroin-users receiving BUP/NX) and 0.071 (PO-users receiving MET). After initial improvement, HRQoL decreased slightly, or increased at a diminished rate.

CONCLUSIONS

OAT, whether delivered in time-limited or unlimited form, using BUP/NX or MET, is associated with modest immediate HRQoL improvements, with diminishing benefits thereafter.

Gender differences in mortality among treated opioid dependent patients

AIMS

To assess gender differences in characteristics, mortality rates, and the causes and predictors of death among treated opioid-dependent individuals.

METHODS

Linked vital statistics data were obtained for all individuals first enrolled in publicly funded pharmacological treatment for opioid dependence in California from 2006 to 2010. Standardized mortality ratios (SMR) were calculated by gender. Cox proportional hazards models with time-varying covariates were fitted to determine the effect of gender on the hazard of all-cause mortality, controlling for covariates.

RESULTS

Over a median 2.6 years (interquartile range: 1.4-3.7), 1.031 deaths were observed, including 2.2% (259/11,564) of women and 3.7% (772/20,758) of men. Women had a greater increased risk of mortality compared to the general population (SMR 5.1 95% CI: 4.5, 5.7) than men (SMR 4.3 95% CI: 4.0, 4.6). The relative risk of death for women compared with men was 1.18 (95% CI: 1.02, 1.36). Women had a lower instantaneous hazard of all-cause mortality than men (HR 0.58, 95% CI 0.50, 0.68), controlling for other factors. Significant interaction effects indicated that among men, mortality risk was decreased by full-time employment and increased by non-daily heroin use (relative to daily use) and medical problems. Concurrent opioid and methamphetamine/cocaine use increased mortality risk among women and decreased it among men.

CONCLUSIONS

Treatment for opioid dependence is likely to reduce mortality risk among men by addressing employment and medical problems, and via interventions to reduce overdose risk after heroin abstinence, and among women by attending to the concurrent use of methamphetamine/cocaine and opioids.

The 'Expanded HIV care in opioid substitution treatment' (EHOST) cluster-randomized, stepped-wedge trial: A study protocol

The public health response to HIV/AIDS has turned its focus onto optimizing health care system delivery to maximize case identification, access and sustained engagement in antiretroviral treatment (ART). Opioid Agonist Treatment (OAT) provides a critical opportunity for HIV testing and linkage to ART. The EHOST study is a cluster-randomized, stepped-wedge trial to evaluate a prescriber-focused intervention to increase HIV testing rates, and optimize ART engagement and retention outcomes among individuals engaged in OAT. The study will encompass all drug treatment clinics currently admitting patients for the treatment of opioid use disorder across the province of British Columbia, encompassing an estimated 90% of the OAT caseload. The trial will be executed over a 24-month period, with groups of clinics receiving the intervention in 6-month intervals. Evaluation of the proposed intervention's effectiveness will focus on three primary outcomes: (i) the HIV testing rate among those not known to be HIV positive; (ii) the rate of ART initiation among those not on ART; and (iii) the rate of ART continuation among those on ART. A difference-in-differences analytical framework will be applied to estimate the intervention's effect. This approach will assess site-specific changes in primary outcomes across clusters while adjusting for potential residual heterogeneity in patient case mix, volume, and quality of care across clinics. Statistical analysis of outcomes will be conducted entirely with linked population-level administrative health datasets. Facilitated by established collaborations between key stakeholders across the province, the EHOST intervention promises to optimize HIV testing and care within a marginalized and hard-to-reach population.

Characterizing longitudinal health state transitions among heroin, cocaine, and methamphetamine users

AIMS

Characterize longitudinal patterns of drug use careers and identify determinants of drug use frequency across cohorts of primary heroin, methamphetamine (MA) and cocaine users.

DESIGN

Pooled analysis of prospective cohort studies.

SETTINGS

Illicit drug users recruited from community, criminal justice and drug treatment settings in California, USA.

PARTICIPANTS

We used longitudinal data on from five observational cohort studies featuring primary users of heroin (N=629), cocaine (N=694) and methamphetamine (N=474). The mean duration of follow-up was 20.9 years.

MEASUREMENTS

Monthly longitudinal data was arranged according to five health states (incarceration, drug treatment, abstinence, non-daily and daily use). We fitted proportional hazards (PH) frailty models to determine independent differences in successive episode durations. We then executed multi-state Markov (MSM) models to estimate probabilities of transitioning between health states, and the determinants of these transitions.

FINDINGS

Across primary drug use types, PH frailty models demonstrated durations of daily use diminished in successive episodes over time. MSM models revealed primary stimulant users had more erratic longitudinal patterns of drug use, transitioning more rapidly between periods of treatment, abstinence, non-daily and daily use. MA users exhibited relatively longer durations of high-frequency use. Criminal engagement had a destabilizing effect on health state durations across drug types. Longer incarceration histories were associated with delayed transitions toward cessation.

CONCLUSIONS

PH frailty and MSM modeling techniques provided complementary information on longitudinal patterns of drug abuse. This information can inform clinical practice and policy, and otherwise be used in health economic simulation models, designed to inform resource allocation decisions.

The effect of homelessness on hospitalisation among patients with HIV/AIDS

The purpose of this study was to determine the effect of homelessness on the costs and patterns of hospitalisation in patients with HIV/AIDS. A retrospective longitudinal study design, based on medical records data covering 2,768 person-years of observation between 1997 and 2003 on patients with HIV/AIDS, was employed. A contextual measure of neighbourhood socioeconomic status (SES) was also used to uncover differences among low- and high-SES neighbourhood dwellers. The association of homelessness and neighbourhood SES with total annual hospitalisation costs, length of stay, numbers of hospital and emergency department admissions and the probability of an operating room procedure, controlling for other covariates, was assessed using multivariate regression analysis. Our results suggest that the homeless and low-SES neighbourhood residents had a large proportion of total costs attributable to admissions for acute events related to the progression of disease. Hospitalisations for planned operating room procedures comprised a relatively larger proportion of hospitalisation costs for high-SES neighbourhood residents. One implication of our findings is that improvements in the continuity of care and cost savings on inpatient care may be realised through further development of social assistance programs aimed at reaching the homeless and residents of low-SES neighbourhoods.