Outcome selection underpins clinical trial interpretation. Inconsistency in outcome selection and reporting hinders comparison of different trials' results, reducing the utility of research findings.
We conducted an iterative consensus process to develop a set of Core Outcome Measures for Perioperative and Anaesthetic Care (COMPAC), following the established Core Outcome Measures for Effectiveness Trials (COMET) methodology. First, we undertook a systematic review of RCTs in high-impact journals to describe current outcome reporting trends. We then surveyed patients, carers, researchers, and perioperative clinicians about important outcomes after surgery. Finally, a purposive stakeholder sample participated in a modified Delphi process to develop a core outcome set for perioperative and anaesthesia trials.
Our systematic review revealed widespread inconsistency in outcome reporting, with variable or absent definitions, levels of detail, and temporal criteria. In the survey, almost all patients, carers, and clinicians rated clinical outcome measures critically important, but clinicians rated patient-centred outcomes less highly than patients and carers. The final core outcome set was: (i) mortality/survival (postoperative mortality, long-term survival); (ii) perioperative complications (major postoperative complications/adverse events; complications/adverse events causing permanent harm); (iii) resource use (length of hospital stay, unplanned readmission within 30 days); (iv) short-term recovery (discharge destination, level of dependence, or both); and (v) longer-term recovery (overall health-related quality of life).
This core set, incorporating important outcomes for both clinicians and patients, should guide outcome selection in future perioperative medicine or anaesthesia trials. Mapping these alongside standardised endpoint definitions will yield a comprehensive perioperative outcome framework.
Oliver Boney, S. Ramani Moonesinghe, Paul S. Myles, Michael P. W. Grocott, and the StEP-COMPAC group
- COS for clinical trials or clinical research
- Systematic review of outcomes measured in trials
- Delphi process
- Systematic review
The process of developing a Core Outcome Set for anaesthesia and perioperative medicine will involve the following stages:
3.1 Identifying relevant stakeholders (patients, healthcare professionals, academic researchers, and health service managers)
3.2 Identifying all outcomes of anaesthesia, surgery and perioperative care of potential interest to relevant stakeholders
3.3 Using iterative Delphi methodology to achieve consensus among stakeholders regarding what outcomes should be included in the eventual Core Outcome Set
3.4 Publicising and encouraging the adoption of the Core Outcome Set in future anaesthetic and perioperative medicine research
‘Core Outcomes’ definition:
All generic perioperative outcomes, i.e. any that may be considered relevant to all patients undergoing major surgery, will be included as potential ‘core outcomes’. Hence generic outcomes (e.g. mortality) will be included; surgery-specific outcomes (e.g. bile leak following hepatobiliary surgery) will not. Outcomes that, though clearly associated with certain types of surgery (e.g. arrhythmias following cardiothoracic surgery) may affect any patient after surgery, will be included for consideration in subsequent stages of the Delphi consensus process.
Identifying potential core outcomes:
1) We will undertake systematic reviews of recent randomised controlled trials and prospective longitudinal trials in anaesthesia, surgery and perioperative medicine which report any generic outcomes that could be considered potential ‘core’ outcomes. Database searches of EMBASE, Medline, and the Cochrane Library will be employed to retrieve relevant trials from a fifteen month period from January 2013 to March 2014, and generic outcomes will be extracted. These outcomes will be entered into a database. Two systematic reviews will be conducted:
- one describing the range of generic clinical outcomes reported in the recent anaesthetic and surgical literature, and how those outcomes are variously measured
- one describing the range of generic patient-reported outcomes reported in the recent literature
2) We will seek views from patients and the public regarding what constitute important outcomes following surgery and anaesthesia. Views will be sought via:
- an online survey
- interviews of patients in outpatient pre-assessment clinics (i.e. pre-operatively)
- interviews of patients in outpatient follow-up clinics (i.e. postoperatively)
3) We will seek views from healthcare professionals, and from healthcare managers and policy makers, on important outcomes from the process of surgery. This will likely involve free-text surveys or face-to-face interviews.
From these sources, we will generate a comprehensive database of all commonly used perioperative outcome measures – both clinical and patient-reported – classified by domain.