The effectiveness and safety of antifibrinolytics in patients with acute intracranial haemorrhage

Statistical analysis plan for an individual patient data meta-analysis [version 2; referees: 2 approved]

Katharine Ker, David Prieto-Merino, Nikola Sprigg, Abda Mahmood, Philip Bath, Zhe Kang Law, Katie Flaherty, Ian Roberts

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Introduction: The Antifibrinolytic Trialists Collaboration aims to increase knowledge about the effectiveness and safety of antifibrinolytic treatment by conducting individual patient data (IPD) meta-analyses of randomised trials. This article presents the statistical analysis plan for an IPD meta-analysis of the effects of antifibrinolytics for acute intracranial haemorrhage. Methods: The protocol for the IPD meta-analysis has been registered with PROSPERO (CRD42016052155). We will conduct an individual patient data meta-analysis of randomised controlled trials with 1000 patients or more assessing the effects of antifibrinolytics in acute intracranial haemorrhage. We will assess the effect on two co-primary outcomes: 1) Death in hospital within 30 days of randomisation, and 2) Death or dependency at final follow-up within 90 days of randomisation. The co-primary outcomes will be limited to patients treated within three hours of injury or stroke onset. We will report treatment effects using odds ratios and 95% confidence intervals. We use logistic regression models to examine how the effect of antifibrinolytics vary by time to treatment, severity of intracranial bleeding, and age. We will also examine the effect of antifibrinolytics on secondary outcomes including death, dependency, vascular occlusive events, seizures, and neurological outcomes. Secondary outcomes will be assessed in all patients irrespective of time of treatment. All analyses will be conducted on an intention-to-treat basis. Conclusions: This IPD meta-analysis will examine important clinical questions about the effects of antifibrinolytic treatment in patients with intracranial haemorrhage that cannot be answered using aggregate data. With IPD we can examine how effects vary by time to treatment, bleeding severity, and age, to gain better understanding of the balance of benefit and harms on which to base recommendations for practice.

Original languageEnglish
Article number120
JournalWellcome Open Research
Volume2
DOIs
Publication statusPublished - 1 Jan 2018
Externally publishedYes

Fingerprint

Antifibrinolytic Agents
Intracranial Hemorrhages
Metadata
Meta-Analysis
Statistical methods
Safety
Random Allocation
Therapeutics
Logistic Models
Hemorrhage
Logistics
Blood Vessels
Seizures
Randomized Controlled Trials
Stroke
Odds Ratio
Confidence Intervals

Keywords

  • Antifibrinolytics
  • Meta-analysis
  • Trials

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Biochemistry, Genetics and Molecular Biology(all)

Cite this

The effectiveness and safety of antifibrinolytics in patients with acute intracranial haemorrhage : Statistical analysis plan for an individual patient data meta-analysis [version 2; referees: 2 approved]. / Ker, Katharine; Prieto-Merino, David; Sprigg, Nikola; Mahmood, Abda; Bath, Philip; Law, Zhe Kang; Flaherty, Katie; Roberts, Ian.

In: Wellcome Open Research, Vol. 2, 120, 01.01.2018.

Research output: Contribution to journalArticle

@article{83eef89fb5d04bc6a8871f83d6942e85,
title = "The effectiveness and safety of antifibrinolytics in patients with acute intracranial haemorrhage: Statistical analysis plan for an individual patient data meta-analysis [version 2; referees: 2 approved]",
abstract = "Introduction: The Antifibrinolytic Trialists Collaboration aims to increase knowledge about the effectiveness and safety of antifibrinolytic treatment by conducting individual patient data (IPD) meta-analyses of randomised trials. This article presents the statistical analysis plan for an IPD meta-analysis of the effects of antifibrinolytics for acute intracranial haemorrhage. Methods: The protocol for the IPD meta-analysis has been registered with PROSPERO (CRD42016052155). We will conduct an individual patient data meta-analysis of randomised controlled trials with 1000 patients or more assessing the effects of antifibrinolytics in acute intracranial haemorrhage. We will assess the effect on two co-primary outcomes: 1) Death in hospital within 30 days of randomisation, and 2) Death or dependency at final follow-up within 90 days of randomisation. The co-primary outcomes will be limited to patients treated within three hours of injury or stroke onset. We will report treatment effects using odds ratios and 95{\%} confidence intervals. We use logistic regression models to examine how the effect of antifibrinolytics vary by time to treatment, severity of intracranial bleeding, and age. We will also examine the effect of antifibrinolytics on secondary outcomes including death, dependency, vascular occlusive events, seizures, and neurological outcomes. Secondary outcomes will be assessed in all patients irrespective of time of treatment. All analyses will be conducted on an intention-to-treat basis. Conclusions: This IPD meta-analysis will examine important clinical questions about the effects of antifibrinolytic treatment in patients with intracranial haemorrhage that cannot be answered using aggregate data. With IPD we can examine how effects vary by time to treatment, bleeding severity, and age, to gain better understanding of the balance of benefit and harms on which to base recommendations for practice.",
keywords = "Antifibrinolytics, Meta-analysis, Trials",
author = "Katharine Ker and David Prieto-Merino and Nikola Sprigg and Abda Mahmood and Philip Bath and Law, {Zhe Kang} and Katie Flaherty and Ian Roberts",
year = "2018",
month = "1",
day = "1",
doi = "10.12688/wellcomeopenres.13262.2",
language = "English",
volume = "2",
journal = "Wellcome Open Research",
issn = "2398-502X",
publisher = "F1000 Research Ltd.",

}

TY - JOUR

T1 - The effectiveness and safety of antifibrinolytics in patients with acute intracranial haemorrhage

T2 - Statistical analysis plan for an individual patient data meta-analysis [version 2; referees: 2 approved]

AU - Ker, Katharine

AU - Prieto-Merino, David

AU - Sprigg, Nikola

AU - Mahmood, Abda

AU - Bath, Philip

AU - Law, Zhe Kang

AU - Flaherty, Katie

AU - Roberts, Ian

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Introduction: The Antifibrinolytic Trialists Collaboration aims to increase knowledge about the effectiveness and safety of antifibrinolytic treatment by conducting individual patient data (IPD) meta-analyses of randomised trials. This article presents the statistical analysis plan for an IPD meta-analysis of the effects of antifibrinolytics for acute intracranial haemorrhage. Methods: The protocol for the IPD meta-analysis has been registered with PROSPERO (CRD42016052155). We will conduct an individual patient data meta-analysis of randomised controlled trials with 1000 patients or more assessing the effects of antifibrinolytics in acute intracranial haemorrhage. We will assess the effect on two co-primary outcomes: 1) Death in hospital within 30 days of randomisation, and 2) Death or dependency at final follow-up within 90 days of randomisation. The co-primary outcomes will be limited to patients treated within three hours of injury or stroke onset. We will report treatment effects using odds ratios and 95% confidence intervals. We use logistic regression models to examine how the effect of antifibrinolytics vary by time to treatment, severity of intracranial bleeding, and age. We will also examine the effect of antifibrinolytics on secondary outcomes including death, dependency, vascular occlusive events, seizures, and neurological outcomes. Secondary outcomes will be assessed in all patients irrespective of time of treatment. All analyses will be conducted on an intention-to-treat basis. Conclusions: This IPD meta-analysis will examine important clinical questions about the effects of antifibrinolytic treatment in patients with intracranial haemorrhage that cannot be answered using aggregate data. With IPD we can examine how effects vary by time to treatment, bleeding severity, and age, to gain better understanding of the balance of benefit and harms on which to base recommendations for practice.

AB - Introduction: The Antifibrinolytic Trialists Collaboration aims to increase knowledge about the effectiveness and safety of antifibrinolytic treatment by conducting individual patient data (IPD) meta-analyses of randomised trials. This article presents the statistical analysis plan for an IPD meta-analysis of the effects of antifibrinolytics for acute intracranial haemorrhage. Methods: The protocol for the IPD meta-analysis has been registered with PROSPERO (CRD42016052155). We will conduct an individual patient data meta-analysis of randomised controlled trials with 1000 patients or more assessing the effects of antifibrinolytics in acute intracranial haemorrhage. We will assess the effect on two co-primary outcomes: 1) Death in hospital within 30 days of randomisation, and 2) Death or dependency at final follow-up within 90 days of randomisation. The co-primary outcomes will be limited to patients treated within three hours of injury or stroke onset. We will report treatment effects using odds ratios and 95% confidence intervals. We use logistic regression models to examine how the effect of antifibrinolytics vary by time to treatment, severity of intracranial bleeding, and age. We will also examine the effect of antifibrinolytics on secondary outcomes including death, dependency, vascular occlusive events, seizures, and neurological outcomes. Secondary outcomes will be assessed in all patients irrespective of time of treatment. All analyses will be conducted on an intention-to-treat basis. Conclusions: This IPD meta-analysis will examine important clinical questions about the effects of antifibrinolytic treatment in patients with intracranial haemorrhage that cannot be answered using aggregate data. With IPD we can examine how effects vary by time to treatment, bleeding severity, and age, to gain better understanding of the balance of benefit and harms on which to base recommendations for practice.

KW - Antifibrinolytics

KW - Meta-analysis

KW - Trials

UR - http://www.scopus.com/inward/record.url?scp=85045508509&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85045508509&partnerID=8YFLogxK

U2 - 10.12688/wellcomeopenres.13262.2

DO - 10.12688/wellcomeopenres.13262.2

M3 - Article

VL - 2

JO - Wellcome Open Research

JF - Wellcome Open Research

SN - 2398-502X

M1 - 120

ER -