Table 1 Principle objectives, research questions and outcomes

Process: To provide an estimate for the number of care homes needed for the main trial

What are the potential recruitment and attrition rates for enrolment of residents?

Resources: To investigate the acceptability of the proposed measurements to residents and the most appropriate informant

Which type of data collection diaries (care home resident level), designed by the team to capture health and social care resource use, are most acceptable to residents and staff (>75% completed)? And are these diaries valid (contain information that can be used to calculate costs)?

Activities of daily living Number of falls during previous 6 months

Mortality during previous 6 months

Hospital admissions during previous 6 months including total length of stay; number of admissions; time to first admission

Health-related quality of life

Which health-related quality of life measure, the EuroQol (EQ-5D 3 L) [23], ASCOT Social Care-Related Quality of Life (SCRQoL) [24] or the DEMQOL-V4 [25] is more acceptable to residents (greater % of questionnaires completed)? And are resident and proxy reports of HRQoL comparable?

Are baseline and outcomes measures outcome rates adequate (achieving >75% complete data for each)? And what resources are required in determining these outcomes?

Management: To assess the adherence to, and sustainability of Stop Delirium!

What are the adherence rates for the various components of the intervention? And what are the facilitators and barriers for sustainability and integration into routine care, after the intervention?

Total number of medications Health related quality of life Health and social care costs as measured by health related quality of life measures and health and social care resource use

What are the costs from the perspective of the service provider (health and social care) associated with delirium and with Stop Delirium?

Scientific: To determine the most appropriate method to capture resource use in this setting and population and costs of delirium and of delivering the intervention

Can a high (>90%) coverage for delirium screening be achieved using the short Confusion Assessment Method (CAM) [26]? And can a high completion rate (>90% complete) be achieved for the Delirium Rating Scale-Revised (DRS-R-98) [27] in those positive for CAM? Is delirium be determined reliably (>90% inter-rater reliability) using these instruments?

Delirium severity: proportion of residents with severe delirium during a one-month period

Delirium duration (days positive for delirium) during a one month period

Delirium incidence on any on any day during a one-month post-intervention period

Mortality during previous 6 months

Hospital admissions during previous 6 months including total length of stay, number of admissions, time to first admission

What are the rates of delirium and admission to hospital in residents in intervention homes compared to control homes post intervention? And which measure yields more complete hospital admission data: length of stay, number of admissions or time to admission?

To estimate the rates of delirium and hospital admissions in intervention and control homes

What is the intraclass coefficient (ICC) for the proposed primary outcome, delirium occurrence?

The primary outcome for this study is the presence of delirium on any day during a one-month post intervention period determined by screening with short version CAM [26] on alternate days (except Sundays) and confirmed for those positive or borderline using the Delirium Rating Scale-98 (DRSR98) [27].