Do guidelines on euthanasia and physician-assisted suicide in Dutch hospitals and nursing homes reflect the law? A content analysis
- B A M Hesselink1,
- B D Onwuteaka-Philipsen1,
- A J G M Janssen2,
- H M Buiting1,
- M Kollau1,
- J A C Rietjens3,
- H R W Pasman1
- 1VU University Medical Center, EMGO Institute for Health and Care Research, Department of Public and Occupational Health, Amsterdam, The Netherlands
- 2VU University Medical Center, Department of Medical Affairs, Amsterdam, The Netherlands
- 3Erasmus MC, Department of Public Health, Rotterdam, The Netherlands
- Correspondence to B A M Hesselink, VU University Medical Center, EMGO Institute for Health and Care Research, Van der Boechorststraat 7, Amsterdam 1081 BT, The Netherlands;
- Received 29 October 2010
- Revised 5 May 2011
- Accepted 24 May 2011
- Published Online First 27 June 2011
To describe the content of practice guidelines on euthanasia and assisted suicide (EAS) and to compare differences between settings and guidelines developed before or after enactment of the euthanasia law in 2002 by means of a content analysis. Most guidelines stated that the attending physician is responsible for the decision to grant or refuse an EAS request. Due care criteria were described in the majority of guidelines, but aspects relevant for assessing these criteria were not always described. Half of the guidelines described the role of the nurse in the performance of euthanasia. Compared with hospital guidelines, nursing home guidelines were more often stricter than the law in excluding patients with dementia (30% vs 4%) and incompetent patients (25% vs 4%). As from 2002, the guidelines were less strict in categorically excluding patients groups (32% vs 64%) and in particular incompetent patients (10% vs 29%). Healthcare institutions should accurately state the boundaries of the law, also when they prefer to set stricter boundaries for their own institution. Only then can guidelines provide adequate support for physicians and nurses in the difficult EAS decision-making process.
- Care of the dying patient
- living wills/advance directives
- policy guidelines/inst. review boards/review cttes.
- suicide/assisted suicide
In 2002, the Dutch euthanasia law was enacted.1 This law states that euthanasia and physician-assisted suicide (EAS) is not punishable if the attending physician acts in accordance with due care criteria stated in the law. Of all deaths in The Netherlands in 2005, 1.7% were the result of euthanasia and 0.1% were the result of physician-assisted suicide.2 To promote careful decision-making and performance of EAS, institutional practice guidelines can be useful. EAS requires careful decision-making, because it is a difficult process that includes clinical, legal, ethical and personal emotional aspects.3 Physicians do not always have very much experience in this respect. Practice guidelines can support them in this complex decision-making process. Above that, institutional practice guidelines can describe responsibilities of healthcare professionals and prevent illegal practices.4 Guidelines should therefore also include information that accurately states the boundaries of the law.5 The results of a Dutch study among clinical physicians and nursing home physicians who indicated that there were no practice guidelines on EAS in their institution showed that half of them felt a need for such guidelines.6
The usefulness of practice guidelines depends, among other things, on their content. For guidelines on EAS it is relevant that the specific phases of the euthanasia decision-making process, the participants in the decision-making process, and that at least the six due care criteria as laid down in the Dutch law are described. In 1994, the content of euthanasia guidelines was investigated at institutional level in Dutch nursing homes, before the enactment of the euthanasia law. The results showed that only 65% of the guidelines described all due care criteria.7 Euthanasia has also been legalised in Belgium in 2002.8 The content of the euthanasia guidelines in nursing homes and hospitals was studied in Belgium in 2005 and 2006, respectively. These guidelines described several phases of the euthanasia care process, including the involvement of caregivers, patients and relatives; and also addressed ethical issues.9 10
In 2005, the existence of practice guidelines on EAS at institutional level was investigated in Dutch hospitals, nursing homes, psychiatric hospitals, hospices and institutions for the mentally disabled. The results showed that 62% of the institutions had practice guidelines for EAS. In this paper we focus on the content of EAS guidelines of Dutch hospitals and nursing homes, as EAS guidelines do most exist in these institutions (hospitals 89% and nursing homes 79%).11 Furthermore, we know that approximately half of all Dutch hospital physicians and 74% of all Dutch nursing home physicians have received explicit requests for EAS.5 In addition, hospitals and nursing homes do differ in their patient population. There is a high percentage of incompetent patients in nursing homes, and hospitals more often have younger patient with incurable diseases. It would be interesting to see if guidelines differ in their attention for specific patient groups.
As mentioned above, in 2002 the euthanasia law was enacted in The Netherlands. Besides differences between hospitals and nursing homes, we also focus in this paper on differences between guidelines that were written/revised before the enactment of the law (until 2001) and guidelines that were written/revised after the enactment of the law (as from 2002), to see if and in what way the law affects the content of institutional EAS guidelines and in what way the law is implemented in these guidelines?
Therefore the aims of this study were to analyse the content of practice guidelines for EAS in hospitals and nursing homes, and to compare differences between these settings and guidelines developed before or after enactment of the euthanasia law.
Study population and design
The present study was part of the evaluation study of the Euthanasia Act.2 Data were collected from October 2005 to March 2006. Questionnaires were sent to all Dutch hospitals and nursing homes. The management of these institutions (119 hospitals and 218 nursing homes) were asked whether they had a guideline on EAS, and if so to provide a copy of this guideline. Of these institutions, 56 (19 hospitals and 37 nursing homes) had to be excluded because they did not meet the inclusion criteria (7×24-h inpatient nursing care), or they had merged with another institution. Of the 281 remaining institutions, 192 returned the questionnaire (68%). A total of 154 of these institutions indicated that they had an EAS guideline, 150 of them provided a copy of the guideline. Of the 150 received guidelines, 99 guidelines (46 hospitals and 53 nursing home guidelines) were analysed. We made a distinction between guidelines that were written/revised before the Dutch euthanasia law was enacted (before 2002) and still in use without changes at the time and guidelines that were written/revised after this law was enacted (after 2002). The other 51 guidelines were excluded, because they did not meet the definition of an EAS guideline used in this study: a written protocol to guide caregivers in approaching a problem that includes a decision-making process and/or a phased care plan and at least describe the due care criteria to some extent. Of the institutions in this study (ie, hospitals and nursing homes with a EAS guideline), 91% of the hospitals and 34% of the nursing homes had an ethics committee, 41% of the hospitals and 34% of the nursing homes had a religious affiliation. Whether or not an institution had a religious affiliation did not seem to be related to the existence of practice guidelines on EAS.11
The guidelines were analysed using a checklist of items, based on the study carried out in 1994,7 the Dutch euthanasia law1 and literature.12–14 We piloted this initial checklist in five guidelines to see if the checklist was complete and no important issues forgotten, and at the same time to see if the different reviewers interpreted the items of the checklists in the same way. The final checklist included the following main topics: (1) general characteristics of practice guidelines: format document, formulation and categorically excluding specific patient groups, i.e., dementia patients, coma patients, incompetent patients (categorically excluding patient groups is more strict than the law); (2) request: involved parties and their roles, objections of conscience and advance euthanasia directives; (3) decision-making: due care criteria, involved parties and their roles, refusal of euthanasia requests; (4) performance: involved parties and their roles; report and aftercare. The following six Dutch due care criteria for EAS are laid down in the law: (1) the patient's request should be voluntary and well considered; (2) the patient's suffering should be unbearable and without prospect of improvement; (3) the patient should be informed about their situation and prospects; (4) there are no reasonable alternatives; (5) another independent physician should be consulted and (6) the termination of life should be performed with due medical care and attention.1
We assigned the aspects of handling a patient's request for EAS, advance directives, conscientious objections, decision-making, due care criteria, categorical exclusion of specific patient groups and the role of the physician and nurse in the performance of EAS, to one of three main categories: (1) required by law; (2) more strict than law and (3) not stated in the law. In this case the term ‘law’ encompasses a broad definition, namely the euthanasia law itself, the intention of the legislator as stipulated down in documents of the House of Representatives and the Senate, the judgements of Regional Euthanasia Review Committees (annual reports) and professional guidelines formulated by (national) professional organisations.
All guidelines were analysed by couples of two reviewers (BAMH, BDOP, HMB, MK, JACR, HRWP), according to the checklist. Differences of opinion were resolved in consensus meetings. If no consensus could be reached, a third reviewer made the final decision. The percentages of agreement between the pairs of two reviewers ranged between 74% and 100% (90% on average) per aspect.
Data were analysed with descriptive statistics. The statistical significance of differences between hospitals and nursing homes and guidelines until 2001 and guidelines as from 2002 were calculated using χ2 tests. Fisher's exact test was used if cells had an expected frequency of less than five.
Receiving a patient's request for EAS
Table 1 shows that the majority of the practice guidelines (94%) contained a description of how to react to requests for EAS. In 28% of the guidelines it was stated that the physician only informs the nurse/nursing aid after a request for EAS when the patient has given informed consent. Four per cent of the practice guidelines stated that the family should always be informed after a patient has made a request for EAS, and 39% stated that this should happen only after the patient had given informed consent. Almost half of the guidelines (48%) stated that in the case of incompetent patients, advance euthanasia directives can be considered as a euthanasia request. Whether or not an institution had a religious affiliation was not related to more reluctance to perform euthanasia on the basis of an advance directive (not in table 1). In the majority of guidelines (95%) the subject of conscientious objections was discussed.
There were several differences between hospital guidelines and nursing home guidelines. Nursing home guidelines more frequently described in detail the role of the nurse after receiving a request for EAS. Hospitals guidelines provided more details with regard to advance euthanasia directives.
There were some significant differences between guidelines developed or revised before and after the euthanasia law. As from 2002, it was more frequently stated that an advance euthanasia directive is to be seen as a request for euthanasia if it applies to the patient's present situation (23% vs 0%). Furthermore, to adhere to the request, all other due care criteria must be met (37% vs 7%).
The decision-making process
Table 2 shows that the majority of the guidelines (81%) stated that the attending physician is responsible for the decision-making regarding EAS. The most frequently mentioned group of people, other than the attending physician, as possible participants in the EAS decision-making process were the nurse/nursing aid (86%), followed by the family (56%). In 88% of the guidelines, the role of the management was described; it was most frequently stated that ‘the management must be informed about the planned administration of euthanasia’ (67%). In 60% of the guidelines there was a description of how to handle in the case of refusal of a request for euthanasia, mainly stating that the patient should be informed about the opportunity of referral to another physician.
There were several differences between hospital guidelines and nursing home guidelines. Nursing home guidelines more frequently described the patient's family as possible participants (66% vs 43%), and less frequently described the general practitioner as a possible participant in the decision-making process (36% vs 65%). Only the hospital guidelines (9%) stated that the management must give permission for the actual performance of euthanasia.
There was one difference between guidelines developed or revised before and after the euthanasia law. As from 2002, it was less frequently stated that the management of the institution must be informed before the administration of euthanasia is planned (58% vs 89%).
Mentioned due care criteria
Table 3 shows that all five due care criteria (the sixth due care criterion, ie, the termination of life, should be performed with due medical care and attention, is described in table 4) were mentioned in most of the practice guidelines. The most frequently described due care criterion was consultation (99%), followed by voluntary and well-considered request (98%), hopeless and unbearable suffering (95%), informing the patient (84%) and no reasonable alternatives available (82%). With regard to ‘voluntary and well-considered request’, 79% of the guidelines stated that a written advance directive is always necessary (stricter than the law). With regard to ‘hopeless and unbearable suffering’, 19% of the guidelines included the aspect ‘life expectancy’ (which is not mentioned in the law).
There were several differences between hospital guidelines and nursing home guidelines. Hospital guidelines more frequently described the due care criteria ‘no reasonable alternatives available’ (93% vs 72%), and more frequently provided details with regard to the due care criterion ‘consultation’, such as the consultant should not be a co-attending physician of the patient (85% vs 55%), or a trainee (43% vs 13%).
There were differences between guidelines developed or revised before and after the euthanasia law. As from 2002, the due care criteria that were more frequently described were: ‘no reasonable alternatives available’ (89% vs 64%) and aspects of the due care criterion ‘consultation’, such as discuss the request with the patient (73% vs 50%), a written report should be made (85% vs 64%) and the possibility of support and consultation on euthanasia in The Netherlands (42% vs 4%).
Categorically excluding patient groups
Table 5 shows that 41% of the guidelines were stricter than the euthanasia law in categorically excluding patient groups. Whether or not an institution had a religious affiliation was not related to more objections of conscience in the case of specific patient groups (not in table 5). There were several differences between hospital guidelines and nursing home guidelines. Nursing home guidelines were more frequently stricter than the law in excluding patients with dementia (30% vs 4%) and incompetent patients (25% vs 4%).
There were differences between guidelines developed or revised before and after the euthanasia law. As from 2002, guidelines less often categorically excluded patient groups (32% vs 64%) and in particular incompetent patients (10% vs 29%).
Role of the physician and nurse
Table 4 shows that the majority of guidelines (91%) described the role of the physician in the performance of euthanasia, and 33% of the guidelines described the role of the physician in assisted suicide. Fifty-one per cent of the guidelines contained a description of the role of the nurse in the performance of euthanasia. Little attention was paid to adherence to ‘the method, substance and dosage according to the recommendations of the Royal Dutch Pharmaceutical Society (25% for euthanasia vs 9% assisted suicide)’.
There were several differences between hospital guidelines and nursing homes guidelines. Hospital guidelines more frequently stated that only the physician is allowed to administer the drugs for euthanasia (83% vs 62%) and more frequently described the physician's role in physician-assisted suicide (46% vs 23%).
There was one difference between guidelines developed or revised before and after the euthanasia law. As from 2002, it was less frequently stated that the physician should be present when the drugs are administered (31% vs 57%) and also in the period between the administration of the drugs and the patient's death (31% vs 54%).
We found that most guidelines stated that the attending physician is responsible for the decision to grant or refuse a request for EAS. The due care criteria were mentioned in the majority of guidelines, but aspects relevant for assessing these criteria were not always described. Half of the guidelines described the role of the nurse in the performance of euthanasia. Compared with the hospital guidelines, the nursing home guidelines were more frequently stricter than the law in excluding patients with dementia (30% vs 4%) and incompetent patients (25% vs 4%). As from 2002, the guidelines were less strict than before 2002 in categorically excluding patient groups (32% vs 64%), in particular incompetent patients (10% vs 29%).
A strength of this study is that all Dutch hospitals and nursing homes were invited to participate, and approximately two thirds did so. The possible subjectivity of content analysis can be considered a limitation of the study. However, all the guidelines were assessed according to a checklist by two trained researchers, and the agreement between the assessors was high. Another limitation could be that we know now which topics and relevant aspects of these topic are mentioned in the guidelines, but not how exactly these are stated in the guidelines, and more importantly if these descriptions are helpful for users. Mentioning a topic in a guideline does not necessarily mean that it is a good guideline that supports users. However, that was not the scope of this part of the study, but of another part of our study about awareness and the use of practice guidelines on medical end-of-life decisions. We found that hospital physicians (of hospitals of which the guidelines are evaluated in this study) do feel supported by the EAS guideline of their institution.15 Physicians who had used the practice guideline and felt supported by it most often mentioned the ‘clear procedure/decision-making’ as a reason for the support.15
New, compared with the earlier study on the content of the guidelines,7 is that we could subdivide the information we analysed into aspects that were required by law, aspects that were stricter than the law and aspects that were not stated in the law but can be of practical value in guidelines.
The due care criteria
The majority of the guidelines described the due care criteria, ranging from 82% for ‘no reasonable alternative available’ to 99% for ‘consultation of another physician’. Describing the due care criteria, laid down in the euthanasia law, can be seen as a minimum requirement for high-quality EAS guidelines, but are not of much value if not operationalised in enough detail to be useful in practice. Therefore, they should describe which aspects must be taken into consideration in assessing a due care criterion in order to be really helpful for the users. However, this is especially difficult for the two more subjective due care criteria ‘hopeless and unbearable suffering’ and ‘voluntary and well considered request’. It is known from other research that one quarter of Dutch physicians who had received a request for EAS had experienced problems in the decision-making, mainly with regard to these due care criteria (hopeless and unbearable suffering (79%) and whether the request was voluntary or well considered (58%)).16 With respect to the more procedural due care criteria, there is still room for improvement in many guidelines. For instance, knowing how to determine whether a possible alternative is reasonable or realistic is important when assessing whether there are no more reasonable alternatives available. However, this was only included in 19% of the guidelines.
EAS guidelines and the euthanasia law
There are two ways in which the EAS guidelines do not always reflect the euthanasia law. First, the guidelines are not always complete in providing information about the law. For example, only approximately two thirds of the hospital guidelines and one third of the nursing home guidelines stated that an advance euthanasia directive can be considered as a request for euthanasia. A content analysis of euthanasia policies of nursing homes in Flanders (Belgium) also showed that only 31% of the guidelines described a procedure for handling advance directives. In that study, it was found that the religious affiliation of the institution influenced describing such a procedure in the guidelines. Nursing homes with no religious affiliation more often described this procedure in their guidelines compared with Catholic nursing homes.10 In our study it was found that religious affiliation did not influenced describing such a procedure in the EAS guidelines.
However, it is also possible that it is not left out on purpose, but because of lack of knowledge. Research among Dutch physicians, medical students and euthanasia consultants revealed that the majority of physicians and medical students in The Netherlands were not aware of the fact that euthanasia on the basis of an advance directive is legal (under certain conditions).5 17 18
Second, some guidelines provide information that can be considered stricter than law. An example is that approximately two thirds of the guidelines state that a written advance directive is always necessary, while according to the law it is not necessary that a patient has written down the request, although it is considered desirable. An important way in which approximately six out of 10 institutions were stricter than the law was in categorically excluding specific patient groups, such as dementia patients, stating that it was illegal to grant a euthanasia request in these groups. Although it is clear that it is probably more difficult to meet the due care criteria in these patient groups, they are not excluded by law. However, institutions have the possibility to limit the possibilities for EAS in their institution as a recent European resolution (2010) states: ‘No person, hospital or institution shall be coerced, held liable or discriminated against in any manner because of a refusal to perform, accommodate, assist or submit to (among other things) euthanasia, for any reason.’19
However, it is not clear whether it is a deliberate choice of institutions to have EAS guidelines that are stricter than law or whether they are not aware of the boundaries of the law. If the latter is the case, improved knowledge of the law should lead to adjustment of the guidelines. If it is a deliberate decision, it is important that patients and professionals are aware of this stricter policy, so they can take this into account in their choice of a healthcare institution.
Finally, it should be acknowledged that practice EAS guidelines should also address aspects that are not described in the law, but are necessary to ensure practical feasibility. For instance, it is very useful to address how to act if a request for euthanasia is refused, because it is known that approximately eight out of 10 requests made in hospitals and nine out of 10 requests made in nursing homes do not result in euthanasia.20 There was little (60%) attention for the aspect how to act if a request is refused in the guidelines. A study on content analysis of euthanasia policies of nursing homes in Belgium also showed that there was little attention to this aspect.10 However, as mentioned before, from another part of the study we know that physicians felt supported by the EAS guideline of their institution, especially with regard to the ‘clear procedure/decision-making’.15
Roles of nurse
In recent years increasing attention has been paid to the role of the nurse in euthanasia, among other things in the development of Dutch EAS guidelines for nurses and physicians.14 In line with this, almost nine out of 10 guidelines mention the nurse as a possible participant in the decision-making. This is a substantial increase, compared with 1994 when it was found that three out of 10 EAS guidelines mentioned involvement of the nurse in decision-making.7 However, we also found that only half of the guidelines described the role of the nurse in the performance of euthanasia. The aspects that were described in some guidelines were accurate, indicating that nurses are not allowed to perform euthanasia, but can be involved in the preparation, and can assist a physician in multiple ways. Other studies have reported that nurses are often involved in the performance of euthanasia, and perform tasks that are illegal and beyond their professional responsibilities.21 22 More attention should be paid to nurses in institutional practice guidelines, in order to clarify their role in the euthanasia process.
Hospitals versus nursing homes
Several differences were found between hospitals and nursing homes. Some are possibly related to the way work is organised in the different institutions, ie, a description of the role of a team in dealing with a request, and the possible participation of family in decision-making. The latter was more frequently found in nursing homes guidelines than in guidelines from hospitals. The family might play a more important role in nursing homes, because many patients are or become incompetent and mostly stay in a nursing home for a long period (until death) compared with patients in hospitals.
It is noticeable that, while especially in nursing homes many patients are incompetent, mainly due to dementia, hospitals more frequently state that an advance euthanasia directive can be considered as a euthanasia request in an incompetent patient, which is in line with the law. Above that, nursing homes guidelines were more frequently stricter than the law in excluding patients with dementia and incompetent patients, compared with hospital guidelines. This is probably related to a reluctance to perform euthanasia on the basis of an advance euthanasia directive. This is noticeable as patients with dementia are a substantial part of the nursing home patient population.
It is known that nursing home physicians more frequently consider euthanasia in an incompetent patient unacceptable than clinical specialists (55% vs 16%).5 Furthermore, data for The Netherlands showed that in practice it never or hardly ever occurs that physicians follow an advance euthanasia directive in a patient who has become incompetent.23
Also in Belgium it seems that nursing homes do have somewhat restrictive policies towards euthanasia. This may be explained by the fact that the majority of Belgian nursing homes had a restrictive stance on euthanasia.10
Before and after the enactment of the law
In The Netherlands, with the enactment of the euthanasia law the already existing EAS regulations were formalised in the law. However, our study shows that guidelines that were developed or revised after the enactment of the law more frequently adhered to the law. For instance, there was a substantial decrease in the percentage of guidelines that categorically excluded specific patient groups, approximately six out of 10 before 2002 and three out of 10 after 2002. This could either be because the law clarified the regulations, or because the regulations became more readily accepted by the developers of the guidelines. In line with our findings, in Belgium it was found that the due care criteria as stated in their euthanasia law had a positive impact on attention paid to these criteria in guidelines.9 10
Especially in Flemish Catholic nursing home guidelines, attention was given to palliative care and interdisciplinary cooperation, both important elements of the palliative filter. However, the absence of a palliative filter in The Netherlands does not mean that that there is no attention for palliative care in Dutch nursing homes. More than half of the Dutch nursing homes guidelines mentioned that it must be clear if there is realistic palliative care available for the patient.
The results of our study show that the content of institutional practice guidelines on EAS are not always an optimal source of information for physicians. Information with regard to advance euthanasia directives and due care criteria is not always complete. The role of the nurse also deserves more attention in many guidelines. Finally, we recommend that healthcare institutions accurately state the boundaries of the law, also if they prefer to set stricter boundaries for their own institution. Only guidelines that provide adequate information and sufficient practical details with regard to the procedure that must be followed after a request for EAS can provide adequate support for physicians and nurses in the difficult EAS decision-making process. However, guidelines can only be supportive in the process, but have to be interpreted for each individual patient and request. It does not solve the ethical decision-making.
Funding This study was supported by a grant from the Dutch Ministry of Health.The study sponsors did not have any role in the design of the study, the data collection, analysis, or interpretation of data, in the writing of the report; or in the decision to submit the article for publication. The researchers are independent from the funders.
Competing interests None to declare.
Provenance and peer review Not commissioned; externally peer reviewed.