Oral Presentation Palliative Care Nurses Australia Conference 2020

A multicomponent non-pharmacological intervention to prevent delirium for hospitalised people with advanced cancer: A phase II cluster randomised waitlist controlled trial (the PRESERVE pilot study)  (70345)

Annmarie Hosie 1 , Jane L Phillips 2 , Meera R Agar 2
  1. University of Notre Dame Australia & St Vincent's Health Network Sydney, Darlinghurst, NSW, Australia
  2. IMPACCT, University of Technology Sydney, Ultimo, NSW, Australia


Delirium is common and debilitating during advanced cancer. Multicomponent non-pharmacological interventions targeting fundamental needs have prevented delirium in one-third of older hospitalised people in previous studies. We developed a similar intervention for people with advanced cancer and piloted it in four palliative care units.


  1. To determine feasibility and acceptability of a multicomponent non-pharmacological intervention to prevent delirium in adults with advanced cancer in hospital;
  2. To inform a phase III trial of the intervention.


Phase II cluster randomised waitlisted controlled trial + qualitative sub-study. After two months of interdisciplinary engagement and training, units implemented delirium screening and diagnostic assessment. Sleep, vision and hearing, hydration, orientation, mobility and family domains were implemented at intervention units admission days 1-7. Control units later implemented the intervention (‘waitlist units’). The primary endpoint was adherence, with 60% or more patients receiving at least four completed domains for at least five days considered minimum evidence of feasibility. Secondary outcomes were other care delivery, delirium, and adverse events. Data were analysed descriptively and inferentially.


Sixty-five patients (25 control, 20 intervention, 20 waitlist) had 98% delirium screens and 75% diagnostic assessments completed. Forty patients received the intervention admission days 1-7 (20 at intervention units, 20 at waitlist units), delivered by nurses (67%), physicians (16%), allied health (8.4%), family (7%), patients (1%) and volunteers (0.5%). There was full adherence for 5% patients at intervention units, partial for 25%. Both full and partial adherence was higher at waitlist units: 25% and 45%, respectively. A higher proportion of control unit patients (32%) became delirious within seven days of admission compared to intervention and waitlist units (20%, p=0.5). The intervention caused no adverse events.  Qualitative findings provided insights into implementation of the intervention in inpatient palliative care.


Despite engagement and training and higher adherence at waitlist units, the trial did not achieve its primary feasibility outcome. However, the sample and delirium measurement was achieved, there were no adverse effects, and there was a signal of effect in units implementing the intervention.


Results support a phase III trial with modifications to optimise intervention delivery.