The last phase of life with dementia in Swiss nursing homes: the study protocol of the longitudinal and prospective ZULIDAD study

In the year 2010, 35 million people were estimated to live with dementia worldwide and it is predicted that this number will double every 20 years until 2050 [1]. Even though the percentages differ between countries, the majority of people with advanced dementia die in long-term care facilities (67 % in the U.S. [2]; 50.2 % in Wales, 92 % in Netherlands [3]; 30.1 % in Germany [4]). The last months of life of nursing home residents with advanced dementia (RAD) are frequently accompanied by distressing symptoms such as dyspnea, pain, pressure ulcers and eating problems, and these symptoms increase with the proximity of death [5, 6]. Due to behavioral problems, unidentified pain, inappropriate medication or other factors, quality of life can be impaired [7]. Even though there is little difference between those people dying with dementia and those dying without cognitive impairments regarding burdensome symptoms in the last phase of life [8] and causes of death [9], RAD are frequently not perceived as having a terminal condition and mostly do not receive optimal palliative care [5, 7, 10]. In order to emphasize the appropriateness and necessity of palliative care in dementia, the European Association of Palliative Care issued a white paper which attempts to define high quality palliative care in dementia [11]. However, only some of the white paper’s recommendations have been included in national dementia strategies so far [12].

The increasing amount of studies in the field motivated van der Steen and Goodman [13] to point out what kind of research is needed in order to advance in research as well as in practical care regarding dementia at the end of life. Besides the demand for theory-driven research, they advocate for study designs which allow for comparisons either with other patient groups, health care systems or countries and for analyses of disease trajectories. Furthermore they advocate for multidisciplinary research. The latter is important not only because palliative care is per se multidisciplinary, but also because RAD are unable to reliably communicate their wellbeing, discomfort or care preferences. The conflation of different proxy estimations, e.g., from relatives (REL), primary nurses (PN) or physicians regarding central outcome measures such as pain, suffering, quality of life is essential in order to provide optimal individualized care. So far, most studies regarding palliative care in dementia have been small and only few have applied a prospective design that allows the determination of disease trajectories and the concomitant proxy estimations [14] (Table 1). Furthermore, the four published prospective studies often used different outcome measures in different proxy groups and this hampers the direct comparison of proxy estimations. As a consequence, further multi-perspective prospective studies on RAD are needed, especially in Switzerland, where data is completely lacking. Exploratory research adapted to the Swiss cultural context and the local structures of the health care system can shed light on the current state of dementia palliative care in Switzerland. Furthermore, in accordance with current recommendations on research strategies in the field [13], for the present study data is prospectively collected from multiple perspectives. To facilitate comparability of data with other countries, the study methodology is closely modeled after existing high-quality studies. This report presents the methodology established in the ZULIDAD study.

The importance of dementia as a life-limiting condition is increasing. As a consequence, providing end-of-life care in dementia will become one of the most challenging tasks for health care systems. Yet, existing knowledge about palliative care in dementia is scarce. With a prospective cohort design the ZULIDAD study examines the last phase of life of RAD. It is one of the few existing prospective studies on the last phase of life in dementia and it is the first prospective study to describe the situation in the greater Zurich area, Switzerland. The combination of data from REL, PN and RAI-MDS allows a comprehensive description and an approximation to the actual quality of life and quality of dying of the RAD. Furthermore, the multi-perspective approach will help to close knowledge gaps regarding differences in REL’s and PN’s perceptions of central outcome measures at the end of life in dementia. Systematic differences in proxy estimations (e.g., REL and PN) can provide a basis for interventions and interventional studies aiming at optimizing palliative care.

The alignment of the ZULDAD study design, inclusion criteria and applied measures to related studies allows for comparisons between European countries and between Europe and U.S. However, due to the geographical limitation to the greater Zurich area and the limitation to municipal NH and NH specialized in dementia care the study results will neither be generalizable to the French- and Italian-speaking parts of Switzerland nor to all nursing homes in Zurich. The right-censoring of some of the data (because not all RAD will be followed until death) is another bias that needs to be considered when interpreting the data. A unique characteristic of the ZULIDAD study is that it is performed as participatory research project with stakeholder involvement throughout the entire study process. The ZULIDAD RT promotes the practical relevance and the effective knowledge transfer of study results.

In conclusion, the ZULIDAD study provides prospective multi-perspective data on the last phase of life of RAD. This is an important pre-requisite to improve palliative care in advanced dementia in Switzerland and internationally.