Core Outcome Measures in Effectiveness Trials

Defining and Measuring successful extubation and reintubation as part of the COVenT Core Outcome Set; a protocol for a systematic review

General Information

Successful extubation and reintubation are important clinical endpoints that are frequently recorded in critical care clinical trials. They have also been selected, by international consensus through the COVenT (Developing a Core Outcome Set in Ventilation Trials) Study, as 2 of the 6 core outcomes that should be measured in the core outcome set (COS) for trials with an intervention aiming to reduce the duration of mechanical ventilation. This COS is the first of its kind in the field of critical care medicine.
It is clear that development of this COS represents a valuable step forward in the promotion of transparent results reporting, clearer data collection strategy and more robust clinical trials within critical care medicine. However, to confer maximum benefit from the application of this COS, it is not sufficient to recommend a list of outcomes which should be measured in each trial of this nature. In order for a COS to be successfully applied to trials the definition and optimum measurement process for an outcome must also be established. Only partial benefit can be conferred by recording and measuring the same ‘what’ of an outcome; it is when this is combined with the same ‘how’ that effective comparison of interventions across different trials can truly occur.
This review aims to determine how successful extubation and reintubation are defined and measured in trials aiming to reduce the duration of mechanical ventilation. It will also explore how these outcomes are handled with respect to those patients who undergo tracheostomy.
This review will inform a consensus meeting to determine how successful extubation and reintubation should be defined and measured within the context of the recently developed COS.

This study will be undertaken by Dr Suzanne Ringrow as part of a PhD programme at the School of Medicine, Dentistry and Biomedical Sciences in Queens University Belfast.
Principal Supervisor Dr Bronagh Blackwood – Queens University Belfast
Dr Dave Liu – University of Queensland, Brisbane
Prof Michael Reade – University of Queensland, Brisbane
Prof Danny McAuley – Queens University Belfast
Prof Mike Clarke - Queens University Belfast
Prof Tim Walsh – The University of Edinburgh

Further Study Information

Current Stage:
May 2017 - August 2017
Funding source(s):
Department of Employment and Learning Northern Ireland

Health Area

Disease Category
Lungs & airways

Disease Name

Target Population

Age Range
0 - 100


Nature / type of Intervention


Consensus meeting
Systematic review


Eligibility Criteria:
Study Type: Randomized Controlled Clinical Trials
Population: Adults and Children (excluding neonates) who are mechanically ventilated in intensive care units.
Intervention: Any trials with an intervention that may reduce the duration of mechanical ventilation and who measured duration of mechanical ventilation, either alone or within a composite measure (those trials that could have used the COVenT COS if it had been available at the time the trial was performed). For inclusion these trials must also measure one or both of the outcomes of interest.
Outcomes: Duration of Mechanical Ventilation
Reintubation and/or Successful extubation

Search Strategy (incl. study selection and data collection):
The trials for this review will be from 3 sources;
1) A Pubmed search of top ranked critical care and general medicine journals
2) The Cochrane Library
3) The World Health Organization Registry of Trials database (

1) A Pubmed search of critical care and general medicine journals
SR and DL will both search the top ranked critical care journals and general medicine journals, as listed by Journal Citation Reports 2015, published between 2012 and 2017 independently. These include: Am J Respir Crit Care Med, CCM, ICM, Lancet Resp Med, Chest, Resuscitation, Crit Care, Ann Intensive Care, J Neurotrauma, Crit Care Resus and Pediatric CCM, NEJM, Lancet, JAMA and BMJ. We will include RCTs that meet the eligibility criteria and will also search for systematic reviews that meet the eligibility criteria to extract further RCTs from. Both searches will be compared to check for inter-rater reliability. BB will act as arbitrator if there is disagreement.

2) The Cochrane Library
Using the eligibility criteria, SR will search the Cochrane Library to source relevant systematic reviews, published between 2012 and 2017, to extract relevant RCTs from. A 10% sample of the systematic reviews will be reviewed by DL to check for inter-reviewer reliability. If there is agreement, SR will proceed to extract trials, of which 10% sample will be reviewed independently by DL. If there is disagreement SR and DL will discuss and increase the sampling until agreement is reached. Once the trials are extracted SR will send a 10% sample to DL and the same process will be followed. BB will act as arbitrator if there is disagreement.

3) The World Health Organization Registry of Trials database (
SR will search the World Health Organisation Database for trials ongoing or pending that meet the eligibility criteria; registered between 2012 and 2017. For any trial registered but already published, the publication will be sought and analysed preferentially to the registration or protocol. A 10% sample will be cross-checked by DL in the same process as above.

Data Collection:
Data extraction will begin when the list of eligible trials is agreed. SR will arrange the trials in reverse chronological order from 2017 until 2000. If saturation is not reached by this publication date or there are insufficient papers we will continue until the team is agreed that saturation has been reached. We will not limit the search by country or language. Data will be extracted independently from the first 10% of the trials by SR and the second 10% by DL, using pre-piloted data extraction forms. This will then be swapped and checked for consistency. If there is good consistency the remainder of the trials will be divided between SR and DL who will continue to work independently to extract the data. Once all data are extracted a random 10% sample will be swapped and verified.

Data from published trials will be compiled and presented separately from the results of the trials that are ongoing or pending. This will allow us to capture how these outcomes may be reported and defined in the future meaning that we can be confident that any subsequent recommendations made following the close of this study will be relevant to the current cohort of researchers.

Data Items and Summary Measures:
We will collect 2 tiers of data in this review which will be presented as summary tables using descriptive statistics:
1. Operational definitions of 'duration of mechanical ventilation',‘successful extubation’ and ‘reintubation’ in 4 key areas in line with the SPIRIT checklist for reporting outcome measures in trials; the specific measurement variable, analysis metric, methods of aggregation and time points.
2. How the trial reports ‘successful extubation’ and ‘reintubation’ in the context of high flow oxygen, NIV and tracheostomy.
Following completion of this review the results will be presented to a patient representative group and their feedback will be recorded. Expert panel consensus meetings will then be held. The panel experts will be invited from those who participated in the original Delphi portion of the study for COVenT – this includes clinicians, clinical trialists and research methodology experts.
The consensus meetings will take the form of 2 webinar meetings from which different panels (with similar demographics) will discuss the findings of the review and come to a consensus. Following this a recommendation will be made about how ‘successful extubation’ and ‘reintubation’ should be defined, measured and reported for the COVenT COS.

Any intervention that may modify the duration of invasive mechanical ventilation will be included. This may include extra corporeal devices; protocols to assist discontinuation from ventilation (e.g. ventilator and sedation related); ventilator modes (including automated weaning modes); medications/ strategies for Acute Respiratory Distress Syndrome.

Stakeholders Involved

Clinical experts
Consumers (caregivers)
Consumers (patients)

Study Type

COS for clinical trials or clinical research

The site uses cookies, some may have been set already. Please refer to our privacy policy & cookie usage statement.
If you continue to use the site we'll assume you're happy to accept the cookies.