Core Outcome Sets for Trials to Prevent and/or Treat Delirium: The Del-COrS Collaboration

Contributed by Louise Rose, BN, MN, PhD, Professor in Critical Care Nursing, Florence Nightingale Faculty of Nursing, Midwifery and Palliative Care, King’s College London, London, UK

Delirium is a common, costly and potentially preventable syndrome associated with short- and long-term adverse outcomes for patients, family members, and healthcare organizations. Reassuringly, the numbers of clinical trials evaluating delirium interventions continue to increase. Historically, selection of outcomes for such trials has been at the discretion of the investigators devising the study protocol. This has unfortunately led to disparate outcomes with differing definitions as well as heterogeneous measurement instruments and time-points. The lack of standardisation impairs the ability to pool results from trials evaluating the same intervention, thus reducing the precision of meta-analyses, and precluding evidence-based clinical or health policy decisions based on high-quality evidence.1,2

Core outcome sets (COS) help address this heterogeneity in outcome and measures. A COS is a minimum set of outcomes, identified through rigorous consensus building methods and widespread consultation, that should be measured and reported in all studies related to a specific health condition.3 Consultation includes health service users (i.e., patients and their family), policy makers, health technology assessors, research and health system funders, and healthcare professionals, as well as researchers. Various research groups have emphasized the need for COS for trials of delirium interventions in recent years. Indeed, the 2019 Scientific NIDUS Think Tank identified development and uptake of a COS, and use of uniform, standardized measurement tools for delirium case identification and severity rating as key priority areas.

To address the need for COS in delirium research, the Del-COrS international collaborative, led by Professor Louise Rose and Dr Valerie Page, was formed. This collaboration comprises delirium experts representing the four populations described below and diverse healthcare professions. The aim of the Del-COrS international collaboration, funded by the Canadian Institutes of Health Research (CIHR), is to develop consensus on COS appropriate for future trials of interventions designed to prevent or treat (or both) delirium in four patient groups at high risk of delirium:

  1. patients requiring intensive care unit (ICU) admission
  2. patients requiring hospitalisation in an acute care setting (without ICU admission)
  3. patients receiving palliative care, and
  4. older adults living in residential care or the community

To achieve this aim, the Del-COrS international collaboration is using methods outlined in the Core Outcome Measures in Effectiveness Trials (COMET) Handbook version 1.0 to conduct a four-stage COS item generation and item reduction process.4 The first step is a series of systematic reviews which are essential to generate a list of potential outcomes (as well as how they are measured) from published trials and registered trial protocols. We are pleased to share that the first of these systematic reviews, focused on delirium trials in the ICU, is now available in Critical Care Medicine journal (

For this systematic review, we searched electronic databases, systematic review repositories, and trial registries (1980 to March 2019); assessed quality of outcome reporting using the MOMENT study scoring system; and assessed risk of bias using the Cochrane tool or SIGN checklists depending on study design. We categorized reported outcomes using COMET taxonomy. Of the 195 studies recruiting 74,632 participants, 67% evaluated an intervention to prevent delirium, 10% an intervention to treat delirium, and 23% an intervention to prevent or treat delirium.

Most studies (55%) evaluated pharmacological agents, with dexmedetomidine (39 studies) and melatonin/melatonin receptor agonists (13 studies) being most commonly studied, and 30 different drugs evaluated. A protocol or bundle (e.g., ABCDE bundle) was evaluated in 38 studies. These data, alone, indicate the large volume of delirium research conducted in the critically ill to-date, however only one trial was conducted in a pediatric ICU population, indicating a pressing need for further research in this field.

Included studies described a mean (standard deviation) of 10 (6.2) outcomes overall, and 2.2 (1.3) delirium-specific outcomes. Delirium incidence (89% of prevention studies; 69% of dual aim, i.e., prevention or treatment studies) and delirium duration (30% of incidence studies; 36% of dual aim studies) were the delirium-specific outcomes reported most frequently. Delirium duration was the most common outcome (60%) reported in delirium treatment trials. The Confusion Assessment Method (CAM)-ICU, with Richmond Agitation-Sedation Scale (RASS) to assess sedation level prior to the CAM-ICU, was the most common measure combination used to ascertain delirium, reported in 35% of studies. Measurement generally began at study randomization or ICU admission, but timing of discontinuation and measurement frequency were highly heterogeneous. Other non-delirium specific outcomes commonly reported included ICU length of stay (69%), mortality (50%), hospital length of stay (47%), and ventilation duration (43%).

Through this systematic review the Del-COrS international collaboration confirmed the heterogeneity and multiplicity of outcomes and measures, underscoring the need for a COS for deliriuim trials in critically ill patients. Subsequent steps of COS development used by the Del-COrS team include:

  • qualitative research to identify outcomes of importance to delirium survivors and family members;
  • a web-based Delphi consensus process; and
  • a face-to-face meeting using modified nominal group consensus technique.

These steps have now been completed for both the ICU and acute hospitalization (without ICU) patient groups. They were presented at the European Delirium Association meeting last year and will be presented at the American Delirium Society meeting in June 2020. We hope to publish the findings of these two COS in an upcoming NIDUS blog, so watch this space!

The Del-COrS international collaboration is endorsed by the American Delirium Society, the Australasian Delirium Association, and the European Delirium Association and includes Louise Rose (co-lead), Valerie Page (co-lead), Meera Agar, Bronagh Blackwood, Lisa Burry, Noll Campbell, Mike Clarke, Philip Devendra, John Devlin, Imogen Featherstone, Leslie Fornear, Annmarie Hosie, Miriam Johnson, Rich Jones (NIDUS investigator), Jacques Lee, John Marshall, Dale Needham (NIDUS investigator), Jordan Rivera, Nadine Schofield, Najma Siddiqui, and Angela Sung.

  1. Clarke M, Williamson P. Core outcome sets and systematic reviews. Syst Rev 2016;5(11).
  2. Moloney R, Messner D, Tunis S. The increasing complexity of the core outcomes landscape. J Clin Epidemiol 2019;116:150-154.
  3. Williamson P, Altman D, Blazeby J, et al. Developing core outcome sets for clinical trials: issues to consider. Trials 2012;13:132.
  4. Williamson P, Altman D, Bagley H, et al. The COMET Handbook: version 1.0. Trials 2017;18(Suppl 3):280.
Posted in Delirium Research, NIDUS Blog.

Leave a Reply