PALLIATIVE ASSESSMENT COHORT IN THE GERIATRIC EMERGENCY SETTING

In Denmark, there is a lack of systematic workflows for basic palliative care and relief. There is evidence that conversations about palliative needs during hospitalization should be initiated by healthcare professionals, as patients and relatives do not necessarily bring up the need themselves due to doubts about the relevance of the hospital contact, but also ambiguity and ambivalence regarding wishes and options for medical treatment in the last part of life are experienced as barriers.

PALMA is an implementation research project and classified as T3 on the translational scale. Inspired by The Medical Research Council (MRC), the intervention is developed through the involvement of relevant stakeholders, including patients and relatives, to ensure relevance. In addition, the feasibility is tested in a comprehensive feasibility study with a view to adapting to already established workflows.

The PALMA cohort is aimed at the part of the intervention development that will develop a stratification tool for identifying older people at risk of dying within 180 days, as well as the algorithm that will define the weight of the cross-sectoral communication regarding identified palliative needs and hospital-outgoing activities for relief.

Establishing a cohort enables to measure the EORTC-C15 PAL and Clinical Frailty Scale (CFS) applicability to the geriatric population of interest. Data are intended to inform risk stratification in a following implementation feasibility study. In PALMA we intend to identify elderly with high risk of dying within three months.

By potentially adding a time-consuming process of palliative assessment and management to a considerable workload at the geriatric emergency department is likely to act as a barrier for implementation. Hence, a non-time-consuming risk stratification process must be identified, which forms the objective of the PALMA cohort study.

In previous studies we have shown frailty, measures by CFS to be a moderate predictor of 90-day mortality and may along with immobilization status serve as quick and reliable identifier of patients at risk of dying within 90 or 180 days. The EORTC-C15 PAL is a key validated European instrument to measure patients' palliative needs. It has mostly been tested in patients in Specialised Palliative Care units and thus, we do not know if the instrument is meaningful to uncover palliative care needs in geriatric patients.

Likewise, we do not know whether the EORTC-C15 PAL corresponds with the degree of frailty or severe immobility at an acute hospital admission at the emergency department. But across various common diseases, we do know that underlying sociodemographic inequality factors and premises may mediate increased mortality and health care utilization considerably.

Cohort methodology: To answer the research question of whether frailty and/or immobility may significantly identify patients with palliative care needs the association (regression coefficient and/or relative risk) and the area under receiver operating characteristic (AUROC) curve between the assumed risk assessment tools and high score of EORTC-C15 PAL is investigated in a cross-sectional cohort design. Further, the cohort is followed for 90-day mortality to test the association between EORTC-C15 PAL scores and mortality in acutely admitted geriatric patients. (See table I for more specific question asked).

Question asked:

1. What is the association between EORTC-C15 PAL scores, CFS and immobility, and how are these associations mediated by sociodemographic characteristics?
2. May CFS/immobility assessment be used as quick screening tool for high symptom risk stratification?

With a total sample size of 198 patients (99 frail and 99 non-frail), the study will have a power of 80% at an alpha of 0.05% to detect a minimum clinically relevant difference of 10 points on PRO-PAL symptom scales. To meet 10% dropout, 10% of PRO-PAL possibly being filled out inappropriately due to severe cognitive impairment and securing 99 frail patients in a consecutive inclusion design, the total sample size of the cohort is set at 300 patients.