Increasingly the choice of medicines for patients in healthcare is guided by published national/ regional guidelines. Rivaroxaban (XARELTO®) is a highly selective direct factor Xa inhibitor which inhibits thrombin formation and the development of thrombi. It was recommended as an option for treatment in NICE guidance published in May 2012. The Specialist Cohort Event Monitoring (SCEM) registry study design is a new methodology developed in parallel with the new legislative requirement for pharmaceutical companies to undertake a Risk Management Plan as part of post- authorisation safety monitoring. The Rivaroxaban Observational Safety Evaluation (ROSE) SCEM registry study has been initiated by the DSRU as part of a broader Post-Authorisation Commitment requested by the CHMP to further investigate the safety of rivaroxaban in clinical practice. ROSE aims to monitor short-term (first 3 months) safety and drug utilisation of rivaroxaban prescribed to adult patients for medical conditions requiring anticoagulation (the prevention of stroke and systemic embolism in adults with non- valvular atrial fibrillation (AF) (with ≥1 stroke risk factors), and for the treatment of deep vein thrombosis (DVT) and pulmonary embolism (PE) and prevention of recurrent DVT and PE) by specialists in the secondary care setting in England and Wales. The methodological considerations in choices of comparator cohort for this large pharmacoepidemiological study are presented here Design: A new user observational based cohort design with internal comparator (Figure 1) Study objectives: Primary: to quantify the incidence of haemorrhage (which meets the criteria for major bleed). Secondary: to describe the patient population, off-label use and the incidence of other safety outcomes. Data source and variables: Since this study is conducted under a naturalistic setting, open entry criteria are desirable to maximise external validity, thus there are no specific exclusion criteria. Patients will be identified via specialist networks and data obtained from existing medical records on prognostic/risk factors before and on the start of treatment, NOAC exposure and specific outcomes. Selection Bias The desire is to study rivaroxaban use in a more heterogeneous population than those observed in clinical trials. However, a potential weakness of this (and any observational study) is selection bias arising because of certain patient characteristics which influence the probability of being treated and the non-probability sampling from the accessible target population. In comparison to studies conducted in the primary care setting, by identifying new users within the secondary care setting, risk estimates are likely to be less subject to the influence of: selection bias arising from treatment survivors and inclusion of patients likely to be less complex in terms of • underlying disease, co-morbidities and concomitant medications than in the general disease population immortal time bias arising not only from misclassification of person-time exposed to the new medication but • also from under-ascertainment of events related to the start of treatment. Contextual phenomenon (Clustering) Prescribing of NOACs including rivaroxaban is likely to be determined by how specialists interpret guidelines and realignment of anticoagulant services at the organisational level. Care pathway redesign will extend the multidisciplinary approach to management with further involvement of supplementary and independent prescribers. The implications are that patient health may be correlated with the regions in which they are treated. It is therefore important to explore this clustering to better understand possible disparities in health outcomes and subsequent impact on post-marketing safety evaluation. Thus the design will include an internal contextual cohort comprised of 1:1 ratio of evaluable patients newly prescribed alternative anticoagulant therapy for similar medical conditions requiring anticoagulation in order to characterise the adoption of Rivaroxaban into clinical practice. Multilevel analysis will be employed to assess the collective effects of prescriber, institution and patient factors on the anticoagulant treatment decision to prescribe Rivaroxaban or not. This method is increasingly being used in epidemiology because analysis using common regression without accounting for clustering effects underestimates the standard error and can give biased results. Methodological considerations in evaluating the safety of novel oral anticoagulants (NOACs) in secondary care setting in the UK: defining the contextual comparator cohort Deborah Layton 1,2 , Miranda Davies 1,2 , Alison Evans 1 , Saad Shakir 1,2 . 1 DSRU, Southampton, United Kingdom; 2 University of Portsmouth, Portsmouth, United Kingdom. www.dsru.org Drug Safety Research Unit Methodological considerations Disclosure The Drug Safety Research Trust is a registered independent charity (No. 327206) operating in association with the University of Portsmouth and is the sponsor of the study. For this study, the DSRU receives support from Bayer, the manufacturer of XARELTO®. The Comprehensive Clinical Research Network (CCRN): Non-malignant Haematology Specialty Group has adopted the study, supported by the Stroke Research Network and the CCRN: Cardiovascular Specialty Group. They will collaborate in multi-site enrolment of investigative sites and patient recruitment, plus maintain specialist engagement. Thus potential obstacles affecting recruitment are likely to be minimised (Figure 2). A positive ethics opinion was received Nov 2012. Identification of investigators is ongoing. Background Aims and Objectives Methods Results The SCEM design provides a framework suitable to evaluate the safety of newly marketed medicines in the secondary care setting. By capturing data on a contextual cohort we hope to gain better understanding of the variability of, and influence on, treatment decisions and prescribing of novel treatments which appear to have some advantages but for which there are significant differences within the health care community about recommended use. Conclusions Figure 2. The National Institute for Health Research (NIHR) 16 22 8 18 19 24 1 23 3 20 10 21 13 6 9 2 12 11 17 25 7 5 14 4 15 Birmingham & the Black Country 1 Central & East London 2 Cheshire & Merseyside 3 County Durham & Tees Valley 4 Cumbria & Lancashire 5 Essex &Hertfordshire 6 Greater Manchester 7 Hampshire & Isle of Wight 8 Kent and Medway 9 Leiscestershire, Northamptonshire & Rutland 10 London (North West) 11 London (South) 12 Norfolk and Suffolk 13 North and East Yorkshire & Northern Lincolnshire 14 Northumberland & Tyne and Wear 15 Peninsula 16 South Yorkshire 17 Surrey & Essex 18 Thames Valley 19 Trent 20 West Anglia 21 Western 22 West Midlands (North) 23 West Midlands (South) 24 West Yorkshire 25 Figure 1. ROSE Registry study online Rivaroxaban approved for study indications, DSRU decides to monitor drug Specialist prescribers enrolled to study, consent obtained from patients started treatment with rivaroxaban or treated via best practice standard care within secondary care setting and simple baseline questionnaire completed 12 week questionnaires completed by prescribers irrespective of whether the patient stopped. Questionnaires returned, data entered. If discharged to primary care, GPs contacted to capture relevant outcome and exposure data. Data reviewed by Research Fellow. Suspected ADRs and events of interest followed up. Confidentiality & security carefully maintained. Comprehensive Clinical Research Network (CCRN) and Non-Malignant Haematology Speciality group Members. Adapted from: http://www.crncc.nihr.ac.uk/about_us/ccrn. Excludes Scotland and Northern Ireland
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Increasingly the choice of medicines for patients in healthcare is guided by published national/regional guidelines. Rivaroxaban (XARELTO®) is a highly selective direct factor Xa inhibitor which inhibits thrombin formation and the development of thrombi. It was recommended as an option for treatment in NICE guidance published in May 2012.
The Specialist Cohort Event Monitoring (SCEM) registry study design is a new methodology developed in parallel with the new legislative requirement for pharmaceutical companies to undertake a Risk Management Plan as part of post-authorisation safety monitoring.
The Rivaroxaban Observational Safety Evaluation (ROSE) SCEM registry study has been initiated by the DSRU as part of a broader Post-Authorisation Commitment requested by the CHMP to further investigate the safety of rivaroxaban in clinical practice.
ROSE aims to monitor short-term (first 3 months) safety and drug utilisation of rivaroxaban prescribed to adult patients for medical conditions requiring anticoagulation (the prevention of stroke and systemic embolism in adults with non-valvular atrial fibrillation (AF) (with ≥1 stroke risk factors), and for the treatment of deep vein thrombosis (DVT) and pulmonary embolism (PE) and prevention of recurrent DVT and PE) by specialists in the secondary care setting in England and Wales. The methodological considerations in choices of comparator cohort for this large pharmacoepidemiological study are presented here
Design: A new user observational based cohort design with internal comparator (Figure 1)
Study objectives: Primary: to quantify the incidence of haemorrhage (which meets the criteria for major bleed). Secondary: to describe the patient population, off-label use and the incidence of other safety outcomes.
Data source and variables: Since this study is conducted under a naturalistic setting, open entry criteria are desirable to maximise external validity, thus there are no specific exclusion criteria. Patients will be identified via specialist networks and data obtained from existing medical records on prognostic/risk factors before and on the start of treatment, NOAC exposure and specific outcomes.
Selection BiasThe desire is to study rivaroxaban use in a more heterogeneous population than those observed in clinical trials. However, a potential weakness of this (and any observational study) is selection bias arising because of certain patient characteristics which influence the probability of being treated and the non-probability sampling from the accessible target population.
In comparison to studies conducted in the primary care setting, by identifying new users within the secondary care setting, risk estimates are likely to be less subject to the influence of:
selection bias arising from treatment survivors and inclusion of patients likely to be less complex in terms of •underlying disease, co-morbidities and concomitant medications than in the general disease population
immortal time bias arising not only from misclassification of person-time exposed to the new medication but •also from under-ascertainment of events related to the start of treatment.
Contextual phenomenon (Clustering)Prescribing of NOACs including rivaroxaban is likely to be determined by how specialists interpret guidelines and realignment of anticoagulant services at the organisational level. Care pathway redesign will extend the multidisciplinary approach to management with further involvement of supplementary and independent prescribers. The implications are that patient health may be correlated with the regions in which they are treated.
It is therefore important to explore this clustering to better understand possible disparities in health outcomes and subsequent impact on post-marketing safety evaluation. Thus the design will include an internal contextual cohort comprised of 1:1 ratio of evaluable patients newly prescribed alternative anticoagulant therapy for similar medical conditions requiring anticoagulation in order to characterise the adoption of Rivaroxaban into clinical practice.
Multilevel analysis will be employed to assess the collective effects of prescriber, institution and patient factors on the anticoagulant treatment decision to prescribe Rivaroxaban or not. This method is increasingly being used in epidemiology because analysis using common regression without accounting for clustering effects underestimates the standard error and can give biased results.
Methodological considerations in evaluating the safety of novel oral anticoagulants (NOACs) in secondary care setting in the UK: defining the contextual comparator cohortDeborah Layton1,2, Miranda Davies1,2, Alison Evans1, Saad Shakir1,2. 1DSRU, Southampton, United Kingdom; 2University of Portsmouth, Portsmouth, United Kingdom.
www.dsru.orgDrug Safety Research Unit
Methodological considerations
DisclosureThe Drug Safety Research Trust is a registered independent charity (No. 327206) operating in association with the University of Portsmouth and is the sponsor of the study. For this study, the DSRU receives support from Bayer, the manufacturer of XARELTO®.
The Comprehensive Clinical Research Network (CCRN): Non-malignant Haematology Specialty Group has adopted the study, supported by the Stroke Research Network and the CCRN: Cardiovascular Specialty Group. They will collaborate in multi-site enrolment of investigative sites and patient recruitment, plus maintain specialist engagement. Thus potential obstacles affecting recruitment are likely to be minimised (Figure 2). A positive ethics opinion was received Nov 2012. Identification of investigators is ongoing.
Background
Aims and Objectives
Methods
Results
The SCEM design provides a framework suitable to evaluate the safety of newly marketed medicines in the secondary care setting.
By capturing data on a contextual cohort we hope to gain better understanding of the variability of, and influence on, treatment decisions and prescribing of novel treatments which appear to have some advantages but for which there are significant differences within the health care community about recommended use.
Conclusions
Figure 2. The National Institute for Health Research (NIHR)
16
22 8 18
19
24
123
3 20
1021
13
6
9
212
11
17
257
5
14
4
15
Birmingham & the Black Country 1Central & East London 2Cheshire & Merseyside 3County Durham & Tees Valley 4Cumbria & Lancashire 5Essex &Hertfordshire 6Greater Manchester 7Hampshire & Isle of Wight 8Kent and Medway 9Leiscestershire, Northamptonshire & Rutland 10London (North West) 11London (South) 12Norfolk and Suffolk 13North and East Yorkshire & Northern Lincolnshire 14Northumberland & Tyne and Wear 15Peninsula 16South Yorkshire 17Surrey & Essex 18Thames Valley 19Trent 20West Anglia 21Western 22West Midlands (North) 23West Midlands (South) 24West Yorkshire 25
Figure 1. ROSE Registry study online
Rivaroxaban approved for study indications, DSRU decides to monitor drug
Specialist prescribers enrolled to study, consent obtained from patients started treatment with rivaroxaban or treated via best practice standard care within secondary care setting and simple
baseline questionnaire completed
12 week questionnaires completed by prescribers irrespective of whether the patient stopped. Questionnaires returned, data entered. If discharged to primary care, GPs contacted to capture relevant outcome and exposure data.
Data reviewed by Research Fellow.
Suspected ADRs and events of interest followed up.
Confidentiality & security carefully maintained.
Comprehensive Clinical Research Network (CCRN) and Non-Malignant Haematology Speciality group Members. Adapted from: http://www.crncc.nihr.ac.uk/about_us/ccrn. Excludes Scotland and Northern Ireland
Related Documents
