Article Text
PDF
Abstract
Background Little is known in Europe about end-of-life (EOL) decisions and advance directives (AD), particularly in patients with severe advanced disease. Switzerland is a multicultural and multilingual federal country and has the particularity of being divided into four linguistic and cultural regions
Objective To understand better in different regions of Switzerland which specific patient's characteristics could have an impact on their decision to complete AD or not.
Design/setting/participants Prospective study conducted in four palliative care units. Patients with an advanced oncological disease, fluent in French, German or Italian and with a Mini-Mental State Examination >20 were included. Demographic data, symptom burden (Edmonton Symptom Assessment System, ESAS; Hospital Anxiety and Depression Scale, HADS) and spiritual well-being (Functional Assessment of Chronic Illness Therapy-Spiritual well-being, FACIT-sp) have been assessed. A structured questionnaire has been completed by patients, their relatives and health professionals.
Results 143 patients were included (mean age 68.3 years; 62 male). 41 completed ADs. No particular features were associated with the completion of ADs. Most patients were satisfied with the medical information received. A third of them were not worrying about their future, especially those living in the German-speaking part. Should they become unable to communicate, 87 expected their relative to transmit their own wishes, but only 38 had spoken recently with them about what they wanted. 23 of the 69 included relatives would like to play a more active role in decision-making.
Conclusions These results illustrate the fact that terminally ill patients wish to be active in decision-making, but only seldom transmit their wishes to their relative or complete a written document. The discussion about ACP should be defined according to the particularity of each region and the role of healthcare professionals’ attitudes towards ADs, but we should also be creative and find other ways to promote shared decision-making.
- Clinical decisions
Statistics from Altmetric.com
Introduction
Asking patients about their expectations and preferences regarding treatment is increasingly recognised as an important domain in end-of-life care (EOL). The preferences of the patients can be documented with advance care planning (ACP) or written advance directives (ADs). ACP is defined as the discussion with the patients and/or the relatives about the goals and preferences of care. ADs are written documents in which patients outline the treatment they want to receive or not; in this case, they will no longer be competent to communicate their wishes. Although studies have demonstrated that a majority of patients are keen to discuss the subject with their physician,1–8 only a minority has done so or completed ADs. The influence of ethnic cultures, the presence of religious and spiritual beliefs and some healthcare factors have all been suggested to play a prominent role on the discussion about ACP and ADs completion rates.9–17
Little is known in Europe about EOL decisions and ADs, particularly in patients with severe advanced disease.18 ,19 Switzerland is a multicultural and multilingual federal country and has the particularity of being divided into four linguistic and cultural regions. Some regions, such as Geneva, have implemented a specific legislation on ADs for nearly 20 years, but most had no specific legislation until 2013, the date of their introduction in the Swiss civil code.20 ,21
Within this context, the main aim of our study was to better understand if certain patients’ characteristics played a role in their decision whether to complete ADs or not.
Our secondary aims were to better understand preferences for information and for involvement in decisions of patients and relatives as well as to better describe experiences and opinions of palliative care professionals about ADs issues.
Methods
This is a prospective study conducted in four palliative care units in Switzerland from January 2010 to June 2012. The different palliative care units are located in three regions, each with a different language and cultural background: French (Geneva and Neuchâtel), German (Graubünden) and Italian (Ticino).
Hospitalised patients with advanced cancer were included at about the fifth day after admission, if they met the following criteria: an estimated life expectancy <6 months, fluency in the local language and a Mini Mental State Examination (MMSE) >20. Patients were not included if they were in their last days of life.
At inclusion, patients were asked to designate a close relative who could participate in the study. After inclusion, patients were informed about the possibility of completing their ADs during hospitalisation. Patients who then wanted to go further on with the process and complete their ADs received more extensive information and were provided with help in writing the document.
Collected data
Demographic information, religion, nationality, origin, educational level and main cancer diagnosis were collected.
ADs that were completed before hospitalisation as well as those completed while in hospital were collected.
Assessment tools
MMSE assessed the cognitive status at admission.22 We decided to use a lower cut-off of 20 rather than the usual score of 24 to include patients with mild cognitive impairment who could understand and complete the questionnaire.
Symptoms of the patients were assessed with the Edmonton Symptom Assessment System (ESAS) on the day of inclusion.23 The ESAS consists of nine 10 cm visual analogue scales (VAS) to assess the intensity of pain, tiredness, nausea, depression, anxiety, drowsiness, appetite, well-being and shortness of breath Scores range from 0 (best possible) to 10 (worst possible). The sum of the intensity of the nine VAS was calculated, which gives a range from minimum 0 to maximum 90 points.
To better assess anxiety and depression, we used the Hospital Anxiety and Depression Scale.24 It consists of 14 questions, 7 for anxiety and 7 for depression. Each item was answered by the patient on a four point (0–3) response scale with a score range from 0 to 21 for anxiety, and from 0 to 21 for depression. A score of 0–7 for either subscale was regarded as being within the normal range.
Functional Assessment of Chronic Illness Therapy-Spiritual well-being (FACIT-sp) was completed to assess the spiritual well-being of each patient.25 The FACIT-sp is part of the larger FACIT measurement system of which the Functional Assessment of Cancer Therapy–General (FACT-G) is the core instrument.26 We then calculated the sum of the results of the items.
Questionnaire to evaluate the preferences for information and for involvement in decision-making
The questionnaire was created by the main investigator (SP) because of the lack of a short, simple and validated scale suitable for this hospitalised population.27 ,28 We decided to include one validated question about the decision-making preferences (active decision-making vs shared decision-making or passive decision-making) drawn from an existing tool validated in patients with cancer.29 ,30 The questionnaire was tested on five patients. The original French questionnaire was then translated by the co-investigators into the local language (German and Italian). The questionnaire was administered by the co-investigators (CC, CG, GG, YP).
A second and third questionnaire, similar to the patient's one, were respectively created for the designated relative and for physicians and nurses working in the four palliative care units. This questionnaire for the healthcare professionals was only completed at the very end of the study in June 2012.
Statistics
Bivariate association between each of the variables and the completion of ADs was assessed using Fisher’s exact test for categorical variables and either by the Mann-Whitney U test or the Student t test for continuous ones. All scores have been compared between those who completed their ADs and those who did not complete their AD either by the Mann-Whitney U test or the Student t test when the distribution was Gaussian. Bivariate association between each of the variables and the completion of ADs was assessed using the Fisher’s exact test for categorical variables and either by the Mann-Whitney U test or the Student t test for continuous ones. When planning the use of multiple regression technique, the following empirical rule applies when performing the power analysis: it is necessary to count approximately 10 observations for each variable included in the model.
Thus, to accommodate the full regression model, we need to enrol a minimum of 100 participants.
The study was approved by the local ethics committees. Written consent was completed by each patient and relative.
Results
One thousand three-hundred and eighty-four patients with advanced cancer were hospitalised during the study period in the different palliative care units. One hundred and forty-three patients were included after a mean of 9±2 days after admission (100 in Geneva, 2 in Neuchâtel, 28 in Graubünden and 13 in Ticino). Results from Geneva and Neuchâtel, both in the French-speaking part of Switzerland, were collapsed for analysis. Characteristics and reasons for exclusion of the study were only available for Geneva (figure 1).
Study flow diagram. AD, advance directives.
Completion of ADs
Before admission to the palliative care units, 11/102 (11%) patients had completed their ADs in Geneva, 5/28 (18%) in Graubünden and 0/13 in Ticino. During hospitalisation in each region, 15/102 (15%) completed ADs in Geneva, 10 /28 (36%) in Graubünden and 0/13 in Ticino) (p=0.028). ADs completed during hospitalisation were finalised (mean±SD) 14±3 days after admission.
Patient characteristics
The main patient characteristics are described in tables 1 and 2. Patients who had completed their ADs were diagnosed significantly more often with breast cancer (8/41; 19%; p=0.043) and designated relatively more frequently (35/41; 85%; p=0.032).
Main characteristics of included patients
Assessment of the patients
Characteristics of the patients were relatively consistent in the three regions. However, more patients were married in Graubünden (21/28; 75%), compared to Geneva (44/102; 43%) and Ticino (7/13; 54%) (p=0.021). Geneva had a lower proportion of patients of Swiss nationality (43/102; 43%), compared to Graubünden (27/ 28; 96%) and Ticino (11/13; 85%) (p<0.001). The main foreign nationalities in Geneva were Italian and French. More patients were catholic in Ticino (8/13; 62%), compared to Graubünden (14/28; 50%) and Geneva (49/102; 48%) (p=0.674). More patients were protestant in Graubünden (12/28; 43%) than in Geneva (31/102; 31%) and Ticino (3/13; 23%).
Patient's questionnaire
Most patients (115/143; 80%) affirmed that they were satisfied with the medical information they had received until now (see table 3). However, 21/143 (15%) felt that they did not get enough information and in contrast 5% (7/143) estimated the information received as too detailed. Forty-seven of 143 patients (33%) were not worried about their future, especially those living in Graubünden (22/28; 79%).
Patients questionnaire
Globally, 27 (20%) of 143 patients preferred that the physician take the medical decisions alone (passive decision-making), though regional differences were striking with 10/13 (77%) of the Ticino patients saying that the physicians alone should take the decisions.
Most of the patients talked with their relatives about their disease, but only 13 of 143 (19%) had a discussion about potential complications. Three patients out of 11 (27%) in the Italian part discussed complications with their relatives. Forty-six of 143 (67%) felt that their relative should inform the physician about their wishes and take decisions jointly.
Motivations of patients to complete their ADs were to ensure their autonomy, to avoid aggressive treatment and also to decrease the burden of the decision-making task on their family. Two-thirds of the patients who had not written their ADs at inclusion did not know that the possibility to write their ADs existed.
Association between completion of ADs and patient characteristics
In the univariate analysis, the answer to the validated question for distinguishing between the different decision-making preferences was statistically significant in distinguishing patients who completed their ADs from those who did not (p=0.033). However, after adjusting for different characteristics of the patients, this association was not significant anymore (p=0.056). No other characteristics were associated with the completion of ADs (see table 4).
Univariate and multivariate logistic regression to predict the completion of ADs
Relative's questionnaire
One hundred and sixteen patients designated a relative who could participate in the study. Sixty-nine of 116 relatives (59%) agreed to actually participate (see table 5). The results showed that most of the relatives were satisfied with the information they had received until now (66/69; 96%). Fifty-four of 69 (78%) relatives talked about the future with the patient. In the case when the patient could not communicate anymore, 23 of 69 (33%) of all relatives would prefer to have a more active role in decision-making than only convey the patient's wishes. Twenty-four of 69 relatives (35%) did not know about ADs. There were no differences across the three regions.
Relatives questionnaire
Palliative care team questionnaire
Forty-three nurses and eight physicians completed the survey (completion rate 85%) at the end of the study (see table 6). Their mean age was 46±8 year (±SD). They had a general working experience of (mean±SD) 17±8 years and a working experience in palliative care of (mean±SD) 8±6 years. They were working in the palliative care unit since (mean±SD) 4.8±4 years. There were no differences in the demographic characteristics of the healthcare professionals according to the regions, but physicians were younger (mean age 38.8±8) and had less professional experience (11±7 years) than nurses, p=0.012. The main results of the questionnaire are described in table 6. Thirty-nine of 52 (75%) healthcare professionals, except in Ticino (2/7), felt that they should always discuss with the patient to anticipate potential complications. On the other hand, the Italian-speaking healthcare professionals feared to increase the patient's anxiety by evoking possible future complications. The main obstacles encountered by patients in completing their ADs were being in disagreement with the relatives about healthcare decisions. Other remarks cited by the nurses were difficulties in involving the physician in the process of ADs and the challenge to help the patient in finding the right terms to put down in the ADs. Concerning the decision-making process, most of the healthcare professionals felt that the patient's relative should communicate his wishes to the physician and finally make the decision jointly with him.
Questionnaire for the health professionals
Discussion
Our study confirms the difficulties of promoting ADs in terminally ill patients in Switzerland. Only 11% of the patients had completed their ADs before hospitalisation, although Geneva has had a specific law emphasising the legal value of ADs for over 17 years now. This is a very low rate, compared to many other international studies where completion rates from 20% up to almost 70% were found in a population suffering from metastatic cancer.31–34 Our results confirm that there is an urgent need to better inform the patients about their right to complete their ADs, since two-thirds of our participants did not even know what an AD is.35–37 This information concerns even the general population since even the relatives had not previously heard about ADs. After inclusion in the study, a higher proportion of patients completed their AD. Their AD increased, but was still disappointing especially in Geneva (27%). This is even more surprising since the patients we included in our study were severely sick and close to the end of their life and therefore could be especially exposed to such reflections. In Graubünden, however, almost half of the included patients decided after receiving detailed information to complete their ADs. The opposite is found in Ticino where none of the participants decided to write their ADs.
Our main hypothesis in this study was that different characteristics may influence a person's decision to write ADs, as previously described in studies conducted in the USA. 17 ,34 ,38–40 The better predictor was the location of the palliative care units across Switzerland. The trend of our results is similar to the results of a recent survey conducted in 503 oncological and haematological outpatients in Germany that demonstrated that 31% (n=157) indicated having an AD.41 More south, on the contrary, a recent review of EOL care across Southern Europe demonstrated that in Italy and Spain, there was still a low incidence of all EOL decisions and a low use of ADs compared to northern Europe.42 These results could be explained by the catholic tradition still promoting the medical paternalism, gradual disclosure and negative attitudes towards withholding and withdrawal of treatment.42 In our study, however, different religious beliefs or spiritual well-being did not predict if a patient would complete his ADs or not. Research about the role of cultural similarities and differences across Europe in EOL decision-making still needs to be developed.43
Most patients affirmed that they were satisfied with the medical information they had until now. However, 15% felt that they did not get enough information and on the contrary 5% estimated the received information as too detailed. Patients living in Graubünden, compared to those from the other regions, were less worried about their future.
The majority of the included patients preferred for an active role in decision-making. These results can be explained with the promotion of the patient's autonomy during these last years.29 Still, about one-fifth of the patients did not want to take part in the medical decisions. The separate analysis of each region again showed a north-south difference with 75% of the Italian-speaking patients preferring the physician to take decisions by himself. These results could be interpreted in the sense of a deep trust in healthcare professionals, together with the wish of the patient to not overloading their relatives with such important decisions. The fact that patients felt that relatives should inform physicians about their wishes and take decisions jointly may reflect patients’ wishes of not leaving relatives alone during possibly ethically and morally challenging decision-making.
This point is important as the new legal framework in Switzerland empowers the role of the informal caregiver in decision-making if the patient becomes incompetent. However, only a minority of these sick patients had discussed the potential complications with their relatives. Our results that the role of the designated relative is not perceived the same as the patient and the relative's perspective. Patients want their relative to represent the patients’ wishes only but not to add their own opinion. However, one-third of the relatives prefer a more active role in decision-making and not being just a transmitter of the patient’s wishes.
Finally, a possible reflection of cultural influences was found in the answers given by the healthcare professional. The Italian-speaking palliative care professionals feared to increase the patient's anxiety by talking about possible future complications. Although several data invalidate these beliefs, healthcare professionals’ attitudes certainly influence patients’ decisions about completing ADs or not. This might explain why none of the patients in the Italian-speaking part of Switzerland have written ADs.
The main limitation of our study is the use of a not prospectively validated questionnaire, because we could not find a questionnaire suitable for this population. However, we insert a validated question in the questionnaire concerning decision-making. The second limitation is that the sample of patients from the German and Italian parts of Switzerland is limited; however, the results are different across regions. Owing to the small values of the sample size in these regions, this study is an exploratory one.
Conclusion
Patients globally trusted their physicians to make good medical decisions and they also judged the information received as adequate. In general, patients tended to opt for a shared decision-making, even if 20% preferred the physician to assume total responsibility for the decisions.
These results illustrate the paradox that terminally ill patients wish to be active in decision-making, but only seldom transmit their wishes to their relative or complete a written document. The discussion about ACP should be defined according to the particularity of each region and the role of healthcare professionals’ attitudes towards ADs, but we should also be creative and find other ways to promote shared decision-making.44
Acknowledgments
The authors thank all patients and relatives from the three regions who took their time to complete the questionnaires. They are also grateful to the palliative care team for participating in this study and to Dr Gilbert Zulian for providing useful comments on the manuscript.
References
Footnotes
Contributors SP, CG, GG, CC and PV-B were responsible for the overall content. YP was responsible for conducting the questionnaires. FH was responsible for the planning and reporting of the study.
Funding The study was supported by the grant “Kathe Zingg-Schwichtenberg” of the Swiss Medical Academy. The funding source was not involved in the conduct of the study or development of the submission.
Competing interests None declared.
Patient consent Obtained.
Ethics approval Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.