Nursing home staff’s perspective on end-of-life care of German nursing home residents: a cross-sectional survey

Several strategies, such as using a short questionnaire, follow-up contact, providing a second copy of the questionnaire at follow-up, personalized letters, and academic origin of the study were applied. These strategies have a positive effect on response as shown by a Cochrane review [35]. All data was collected anonymously.

The data of all questionnaires was entered into a database by one researcher and independently validated by a second researcher. Differences between both researchers were resolved either by discussion or by involving a third researcher.

The four-page questionnaire was developed by a multidisciplinary research team (see Additional file 1). It was based on a study investigating the medical care of nursing home residents as addressed to general practitioners (GPs) within the scope of the HOMERN (Hospitalization and emergency department (ED) visits of nursing home residents) project [36]. Several questions from the original questionnaire, which was designed for GPs, were substituted with questions designed for nursing home management. Most questions and the methods used for this as well as for the study addressed to GPs [36] were similar. Hence, results of both studies will be compared in the discussion.

The questionnaire was subdivided into three different parts. Each part dealt with one of the following issues: a) medical care provision in hospitals and EDs, b) care provision at the end of life, and c) care delivered by physicians and allied health professionals. The present study mainly focuses on aspects dealing with end-of-life care.

Firstly, participants were asked to estimate the proportion of nursing home residents with documented AD. Furthermore, they were requested to judge how many of the ADs of the German nursing home residents contain information about hospital transfers and in how many cases the resident’s wishes documented in the AD were not taken into account. Secondly, respondents were asked to rate five different statements regarding common practices and potential deficits in end-of-life care on a 5-point Likert scale (0 = totally disagree to 4 = totally agree). This also included one question about specialized outpatient palliative care (SAPV) in Germany. SAPV is a nationwide programme available for patients with limited life expectancy being carried out by multidisciplinary and multi-professional teams specialized in palliative care [37].

Thirdly, respondents were requested for estimations of the proportions of nursing home residents dying in hospital and whether these proportions have decreased, increased or have remained unchanged over the past decade. Lastly, participants were asked to indicate their perception of the end-of-life care nursing home residents receive, either as rather poor or rather good. Those who indicated a rather poor quality of end-of-life care were invited to describe what they perceive as the most important measure for improving end-of-life care (free text).

Additionally, sociodemographic data of the respondents, such as age, sex, current position, and the amount of time spent in that position were requested. Respondents were also asked to provide the following information on the associated nursing homes: their sponsorship; number of beds; and location, including details about the federal state, the number of inhabitants (rural ≤20,000; semi-urban > 20,000- ≤100,000; urban > 100,000 inhabitants), and the distance to the next hospital with an ED.

Descriptive statistics were computed. For categorial data frequencies were calculated. Continuous data are presented as mean with standard deviation (SD), median and interquartile range (IQR).

Since not all respondents answered every question in the questionnaire, the analyses were restricted to subjects with no missing values given in the respective questions.

All statistical analyses were carried out using SPSS Version 25.

Free-text answers were summarized and allocated to previously determined categories. Categories were similar to those used in the aforementioned study involving GPs [36], but slightly adjusted to the answers given in this survey. Allocation was conducted by one researcher and independently checked by a second. Discrepancies were either solved by discussion or by involving a third researcher.

The study received a waiver from the local medical ethics committee of the Carl von Ossietzky University of Oldenburg (no. 2018–147).

In total 486 questionnaires were included in the analysis (response: 45.5%). Respondents had a mean age of 48.0 years (SD: 9.8) and were mostly female (71.0%). Almost two thirds (64.7%) were nursing staff managers and 29.9% were nursing home directors. On average, respondents had around 9.7 years (SD: 8.0) of work experience in their current position (Table 1).

The participating nursing homes had on average 89.1 beds (SD: 47.5). The number of beds ranged from 4 to 403 per home. More than half (52.7%) of the nursing homes had a non-profit, 39.2% a private and 8.1% a local sponsorship. More than half (51.6%) of the responding nursing homes were located in rural areas, 28.4% in a semiurban and 20.0% in an urban area. On average, the distance to the next hospital with an ED was 8.5 km (SD: 7.8). More than half (50.4%) of the participating nursing homes were separated from an ED by 5 km or less.

The respondents stated that, in their estimation, approximately 45.9% (SD: 21.6) of the residents held an AD. They also estimated that in 29.4% (SD: 27.2) of the cases these AD contain information regarding hospital transfers at the end of life. According to the respondents’ perception, in more than a quarter (28.4%; SD: 26.8) of all residents with an AD the wishes documented in these directives are not taken into account (Table 2).

Respondents estimated that 31.5% (SD: 19.9) of the nursing home residents die in hospital. More than half (55.4%) of the respondents think that the proportion of nursing home residents dying in hospital has decreased over the past 10 years.

About a third (35.6%) of the respondents agreed that nursing home residents are treated in hospitals too often at the end of life, while approximately another third (34.2%) of the respondents was convinced of the opposite. Nearly two thirds (65.5%) answered that residents should be enrolled in SAPV more frequently. Around two thirds (63.1%) of the respondents see a need for further education of nurses working in end-of-life care, while 35.9% think that GPs are already well-trained regarding end-of-life care. More than three quarters (76.9%) of the respondents agreed that GPs should be available after office hours in the case of end-of-life care-issues (Fig. 1).

Overall, approximately one third (35.4%) rated end-of-life care rather poor.

Overall, 183 participants elaborated on what they see as the most important measure for improving end-of-life care mentioning a total of 334 items. For 320 of these items a total of 15 categories were built, the remaining 14 items were so diverse that they could not be allocated to one of the 15 categories. The four most frequently mentioned categories were “more nursing staff” (28.4%), “better integration and better availability of palliative and hospice care in nursing homes” (17.4%), as well as “better qualification of the nursing staff” (12.9%), and “closer involvement of the GPs” (9.3%). The remaining categories were mentioned in 0.9–5.1% of the cases and included, for example, measures like better communication and cooperation, better availability of medication, strengthening the residents’ wills/ advance care planning (ACP), less hospital treatment, and better and clearer legislation regarding both, conducting and forgoing medical treatment at the end of life (Table 3).

In general, end-of-life medical care of nursing home residents was perceived as rather poor by more than one third of the respondents (35.4%). The prevalence of ADs was estimated at 45.9%. Nearly two thirds (63.1%) of the respondents think that nursing home staff should be educated better regarding end-of-life care and about a third (33.0%) suggest more training for end-of-life care for GPs. The proportion of residents dying in hospital was estimated at 31.5%. More than half (55.4%) assume a decrease in in-hospital deaths over the past 10 years.

In fact, a German study published in 2006 found that 28.9% of the 792 nursing home residents included in the study died in hospital [33], while another German study, published in 2019, found a proportion of 29.5% in-hospital deaths among around 67,000 German nursing home residents [34]. These figures suggest that the proportions of in-hospital deaths have remained more or less constantly high over the past decades, while the international literature shows inconsistent findings [32]. End-of-life hospitalization, though, is often linked to adverse health effects [20‐29]. Hence, many older people prefer dying at home or in the nursing home [38].

A possible way of fostering a decrease in end-of-life hospitalization could be the further promotion and development of ACP. A systematic review has already shown that hospitalization can be decreased through ACP [39]. ADs are usually part of ACP. Respondents in this study and GPs from the above mentioned study [36] estimated that around 45.9 and 36.8% of the residents, respectively, hold an AD. These estimations are similar to previous studies which have shown that around 12–37% of the German nursing home population hold an AD [38, 40, 41]. Respondents also estimated that only 29.4% of the ADs contained information on end-of-life hospital transfers. Literature from the United States and Canada suggests that around 4–23% of the nursing home residents have do-not-hospitalize orders [42‐45]. However, the American and the Canadian health care systems can hardly be compared to the German one. When asked for the most important measure to improve end-of-life care, only around 5.0% of the respondents of this study and 2.1% of the GPs [36] mentioned strengthening the resident’s will or supporting the use of ADs. This could be due to the fact that ADs are often perceived as recommendations rather than binding instructions [46]. Moreover, a study from Germany has shown that most ADs either contained only very little information or were invalid [40]. Results from a Dutch study show, though, that even with only around 5% of nursing home residents holding ADs, end-of-life hospitalization was only around just 8% during the last month of life [47]. A possible reason may be that all care homes included in the aforementioned study have palliative care goals [47]. The sole use of ACP and ADs may have positive effects on the quality of end-of-life care; however, when embedded into palliative care concepts the positive effects of measures such as ACP and ADs could be manifold.

End-of-life care was rated as rather poor by 35.4% of the respondents of this study as well as by 53.8% of the GPs of the recently conducted and comparable study [36] suggesting that there still is room for improvement. On the one hand the respondents of this study see potential for improvement in increasing the number of residents being enrolled in SAPV. These programmes have been available in Germany since 2007 and a study has already proven their effectiveness in reducing end-of-life hospital transfers and emergency physician consultations [48]. However, their utilization still seems to lack practicability, especially in rural areas [48]. Moreover, palliative care in Germany is so far mostly addressed to people with malignant diseases. The proportion of cancer patients among those receiving palliative care is more than 90% [49], but in nursing home residents dementia, cardiovascular diseases, fractures, and infections are far more common [50, 51]. This further supports the respondents’ perception of residents being enrolled in palliative care programmes too infrequently.

In our recently conducted study on GPs, especially GPs with palliative training rated end-of-life care as rather poor [36]. This, on the other hand, suggests that strengthening the ties between palliative care and nursing home care is an important measure to improve end-of-life care. Studies from the US have already shown that palliative care consultations can contribute to less end-of-life hospitalization [52, 53]. Furthermore, countries fostering palliative care concepts in nursing homes have considerably lower prevalences of end-of-life hospitalization [47, 54‐56]. However, even within Europe, there is a broad range of health care systems and the integration of palliative care in nursing homes is very diverse [57]. Thus, ‘good practice’ concepts of palliative care integration in nursing homes from other countries cannot be applied without changes. Every country has to develop solutions to fit their own, specific health care system framework.

The key to further implementation and use of palliative care concepts seems to be educational training as literature has shown that nurses in German nursing homes considerably lack education in palliative care [58]. Respondents of this study as well as GPs [36] have also suggested that a better qualification and education of the nursing staff could significantly improve end-of-life care. A better understanding of the different processes that occur at the end of life and around dying could lead to a more adequate response to residents’ needs and concerns at this particularly vulnerable stage of life. Additionally, further qualification could lead to an expansion of responsibilities for the nursing staff. With better education and broader responsibilities, pain and symptom management, for example, could be improved. Studies have already investigated ways to better educate those involved in end-of-life care, but a systematic review has shown that these studies and interventions, unfortunately, lack quality [59]. Hence, further research needs to be conducted in order to develop and evaluate high-quality training for end-of-life care.

The aim of this study was to gain an overview of perceptions of German nursing home staff on end-of-life care. Therefore, some of the data have to be interpreted cautiously, especially the estimations the participants were requested for as it is not clear how valid these estimations are. As the majority of respondents were either nursing staff managers or nursing home directors, their answers may have been biased in order to represent their own nursing home more positively. However, it was emphasized that all data was collected anonymously and the proportion of estimated in-hospital deaths fits very well into the literature. Furthermore, for most of the information we were interested in, to date hardly any robust data and literature from Germany can be found. In order to get more detailed and robust data supporting the overview gained here, further prospective studies should be conducted.

The argumentation of this study implies that end-of-life hospitalization is a negative event. There may be situations in which nursing home care is too limited to adequately deal with residents’ health issues and where hospital treatment is needed. However, literature suggests that end-of-life hospitalization often results in adverse effects [20‐29] and that the majority of nursing home residents prefers to die in the nursing home rather than in a hospital [38]. Moreover, as many as two out of three of all hospitalizations in nursing home residents are considered potentially avoidable [19, 60‐63]. Therefore, not every hospital transfer, but certainly inappropriate and unnecessary ones can be considered negative events. The study itself did not distinguish appropriate from inappropriate or avoidable hospitalization.

A major strength of this study is the large sample of nearly 500 participants and a high response (45.5%).

Questionnaires were, in most cases, not answered by those directly involved in end-of-life care of the nursing home residents, but by nursing staff managers or nursing home directors. However, nursing staff managers and nursing home directors are more likely to have a broader and more comprehensive overview of the common practice and of the potential deficits of the end-of-life care in their nursing homes.