Table 4.

Reporting and analysis of delirium outcomes in delirium prevention and treatment trials: problems identified and considerations for future research

Problem identified in the review	Considerations for future research
There is considerable heterogeneity in the design of delirium RCTs; including variation in the duration of follow-up, frequency of delirium assessments, whether delirium is assessed after ICU discharge and patient population being evaluated (e.g., cardiac surgery vs. critically ill patients)	Delirium outcome definitions should be explicitly defined: The definition should include the maximum duration of follow-up, the frequency of delirium assessments, whether delirium is assessed after ICU discharge, and how patient mortality is incorporated, or accounted for, in the outcome definition. The definition for delirium composite outcomes should include how mortality is incorporated and how delirium and coma status is defined after ICU discharge, if delirium assessments are terminated at ICU discharge. For example, a delirium RCT conducted among MV/ARF patients may define delirium incidence as whether a patient screens positive for delirium during at least one assessment while alive in the ICU within 14 days of randomization. The consensus among key stakeholders (patients, families, and clinicians) for primary and secondary delirium outcome definitions is warranted.
Delirium incidence and duration of delirium are most often compared across intervention groups using two-sample tests for proportions or means. When delirium assessments are terminated at ICU discharge with risk for post-discharge delirium or mortality rates are high, as expected in delirium RCTs conducted among critically ill ICU patients, comparisons of proportions or means may be misleading.	Statistical analysis methods for delirium outcomes should summarize censoring due to ICU discharge and the competing risk of death: Comparisons of ICU discharge or death across the interventions should be provided and alternative survival analysis methods should be considered, but have yet to be fully evaluated for delirium RCTs. Coma or deep sedation may be considered an additional censoring event; the impact of which has not been evaluated. Recurrent event survival methods may offer increased power to detect differences in delirium incidence across intervention groups in delirium prevention trials conducted among critically ill patients where delirium episodes may be recurring.
Delirium composite outcomes are common outcomes in RCTs targeting both prevention and treatment of delirium. In such RCTs, mortality may be ranked as the worst state and assigned a value of zero and if delirium assessments are terminated at ICU discharge, it may be assumed that patients are free of delirium after ICU discharge.	In general, the average of a rank-based delirium composite outcome is not directly interpretable: Non-parametric tests that focus on the ranking of the numerical values of the composite outcome measure should be used to make comparisons across intervention groups. Further evaluation of composite outcomes is warranted in delirium RCTs that terminate assessments at ICU discharge to understand the behavior of these outcomes (i.e., type I error rate), when interventions may impact both onset and duration of delirium, as well as the length of ICU stay and mortality.
Only 6 (9%) of 65 primary delirium outcomes were analyzed using methods that adjusted for baseline variables.	Adjusting for prognostic baseline variables for delirium may improve the precision of statistical comparisons of delirium outcomes across intervention groups (i.e., increase statistical power). Statistical approaches accounting for prognostic baseline variables have been developed for a wide range of outcome types (e.g., binary, time-to-event, and rank-based composites). The potential precision gains from these approaches have not been evaluated within delirium RCTs despite the availability of known risk factors commonly collected in delirium RCTs, including age, APACHE severity of illness score, and use of sedatives.