End-of-life decision-making in respiratory intermediate care units: a European survey

In most Western countries, ∼1% of the population dies annually. Although advances in medicine have greatly improved the ability to treat seriously ill patients and prolong life, there is increasing recognition that extension of life might not always be an appropriate goal.

The 5th International Consensus Conference in Critical Care on challenges in end-of-life care in the intensive care unit (ICU) 1 identified numerous problems, including: 1) variability in practice; 2) inadequate predictive models for death; 3) poor knowledge of patient preferences; 4) poor communication between staff and patients/surrogate decision-makers; 5) insufficient training of healthcare providers; 6) the use of imprecise and insensitive terminology; and 7) incomplete documentation within the medical records. It was, therefore, recommended that research should be conducted to improve end-of-life care.

Chronic obstructive pulmonary disease (COPD), lung cancer, upper and lower respiratory tract infections and restrictive thoracic disorders, including neuromuscular diseases, are the leading causes of death, in that they jointly account for ∼30% of deaths 2; moreover, their prevalence is very likely to increase into the 2020s 3.

Acute-on-chronic respiratory failure is usually the final pathway of these pathologies. A European Respiratory Society (ERS) survey on the epidemiology of respiratory intermediate care units (RICUs) in Europe 4 has clearly shown that the large majority of patients with end-stage chronic respiratory disorders are treated by pulmonologists in those specialised areas. RICUs differ substantially from classical ICUs in terms of patient population, staffing, monitoring systems and, last but not least, use of noninvasive mechanical ventilation (NIMV) as the preferred ventilatory approach where applicable.

In chronically ill respiratory patients, for example, the decision to institute or withdraw acute or chronic mechanical ventilation usually requires the active participation of the patient 5–7. Physicians and educators should target the patient and their caregivers in advance in order to improve education regarding diagnosis and probable disease progress, treatment, prognosis, palliative care options and advance care planning, especially for those who have had a previous episode of hypercapnic respiratory failure requiring NIMV or who are at high risk of ventilatory decompensation.

Unfortunately, all of the largest studies 8, 9, consensus conferences 1, and reviews regarding end-of-life decisions have been performed with reference to acute patients admitted to the ICU, and, therefore, may not necessarily apply to chronic respiratory patients and pulmonary physicians or associated personnel.

The ERS Respiratory Intensive Care Assembly has formed a task force on “ethics and decision-making in end stage lung disease”, with the purpose of evaluating the current epidemiology, practice, behaviours and attitudes towards end-of-life decision-making in respiratory units that treat or monitor patients with end-stage respiratory failure in Europe.

METHODS

The aim of this task force, conducted between May 1, 2005 and October 31, 2005, was to collect data regarding end-of-life decisions in RICUs and high dependency units (HDUs) within Europe by means of a prospective questionnaire. This questionnaire was developed by the two chairmen of the task force (S. Nava and A. Simonds), further developed at the ERS annual congress in September 2004, with the help of task force members, and thereafter sent to the ERS office for formal evaluation by two external reviewers.

Once approved by the ERS office, a formal letter was sent by e-mail to all of the participants in the census on the epidemiology of RICUs in Europe, performed in 2002, and all members of the ERS Respiratory Intensive Care Assembly to invite them to participate in the present study.

Task force members from Italy, UK, Germany, Austria, France, Turkey and Romania were contacted as key informants for their own country in order to identify newly opened units or incomplete recruitment. Three reminder e-mails of invitation were sent over 4 weeks to the potential participants, after which recruitment of units was considered closed.

The questionnaire was developed specifically for the task force (by QBGROUP, Padua, Italy) and was sent by e-mail to each physician who agreed to take part in the study, as well as being available on a dedicated Website for the duration of the study. The questionnaire is shown in the Appendix and includes: 15 questions concerning the nature and epidemiology of end-of-life decisions taken in each unit during the 6-month period; nine questions about communication of these decisions; and nine general questions about the unit’s organisation, the types of patient usually admitted, and the responder’s characteristics. The responders who agreed to participate received a Microsoft Excel file to help them record data in real time and avoid missing information when entering the data on the Website.

After completion of the 6-month data collection, 1 month was allowed for questionnaire completion and data assimilation. Any point of ambiguity was clarified by e-mail and telephone survey. In order to enter the requested data anonymously, each participating centre was able to access the Website questionnaire by means of specific passwords.

At the end of the 6-month period, three e-mails of reminder were sent to those who had agreed to participate, after which time the website was closed. A code generated by computer program was assigned to each unit so that all data were treated anonymously, although they were identifiable by country of origin.

Table 1⇓ reports the mutually exclusive criteria (except for the principle of double effect) adopted in order to define the concepts of end-stage respiratory care and all the end-of-life decisions included in the questionnaire. Do-not-intubate (DNI) and do-not-resuscitate (DNR) orders were placed together. As it has previously been shown that significant differences in end-of-life care occur in Europe depending on geographical location 8, an arbitrary division was applied a priori to the data analysis between North Europe (Germany, UK, Austria, France, Belgium and France; in France, the two centres were located in the northern region) and South Europe (Italy, Spain, Portugal, Turkey and Romania).

The data are presented as a whole and stratified according to geographical area (North and South Europe). All of the variables collected were analysed and tested, where appropriate, to verify the differences in distribution between the two groups (North/South Europe), using univariate analysis. In particular, the Chi-squared test was used to assess the distribution of qualitative variables by area (i.e. religion of participating physicians and types of RICU), whereas an unpaired t-test was used for the analysis of differences in the means of quantitative variables (i.e. percentage of patients receiving withholding, withdrawing or DNI/DNR order or in whom NIMV was the ceiling of ventilatory care and the number of persons involved in the decisions). These variables were considered quantitative since information was obtained at the centre level and not by single patient.

The reasons for withholding or withdrawing decisions were assessed using a rating system of 1–10, where 1 was the most important reason and 10 the least important.

Multivariable ANOVA was performed, inserting one quantitative variable at a time, as dependent variables, versus two independent variables (geographical location and number of cases). This was done in order to verify the impact of the different number of patients enrolled by centre and the geographical location (North versus South Europe) of the unit on the variables analysed.

RESULTS

In total, 141 e-mails were sent to potential participants in the study. Of the responders, 24 stated that another physician from their RICU had already agreed to participate, nine answered that their unit had been closed or transferred, eight declined to participate and 36 never replied; 13 e-mails were returned because they were wrongly addressed. Therefore, 55 out of 95 agreed to participate; the 46 duplicates or wrongly addressed e-mails and units that had shut were excluded. Ultimately, 28 (29.5%) out of 95 questionnaires were returned, despite reminder e-mails. Multiple responses from the same institution were eliminated. The geographical distribution of the institutions was as follows: Italy (11; 37.9%), Germany (6; 20.6%), Turkey (2; 6.8%), the UK (2; 6.8%), France (2; 6.8%), Spain (1; 3.4%), Austria (1; 3.4%), Romania (1; 3.4%), Belgium (1; 3.4%), and Portugal (1; 3.4%).

DISCUSSION

The main finding of the present multicentre study is that 21.5% of patients admitted with respiratory failure to European RICUs or HDUs had an end-of-life decision taken during their hospital stay. The most common practices were withholding treatment, use of NIMV as the therapy ceiling and provision of a DNR/DNI order, this latter practice being employed significantly more frequently in North Europe than in South Europe.

Improvement in the standards of end-of-life care and decision-making has been increasingly recognised as a high priority in present-day society. Interest in the topic is generally high, in part due to increasing attention from the media, particularly in some high-profile cases 10–12, which have generated debate between physicians, lawyers, ethicists, religious groups, patients’ associations and the population as a whole.

The vast majority of the studies assessing the practice of end-of-life care have been performed in general ICUs 8, 9, and so limited data are available regarding the approaches of medical teams in other more specific environments and other subsets of patients.

For example, chronic diseases are the leading cause of death; the World Health Report of 2003 stated that the burden of these diseases will increase dramatically during the next 15 yrs 3. Chronic respiratory disorders are now considered to be one of the five leading causes of death in both Western and developing countries. It is, therefore, not surprising that the large surveys of end-of-life care performed in ICUs report respiratory disorders as the major cause of admission 8, 9.

The final common pathway of patients with chronic respiratory disorders is very often the occurrence of chronic and/or acute respiratory failure. These patients are often followed periodically as outpatients or inpatients by respiratory specialists, who usually direct therapeutic interventions, such as long-term oxygen therapy or long-term mechanical ventilation. The relatively recent growth in the number of RICUs/HDUs in Europe 4 has provided a unique opportunity to assess the approach of respiratory medical and paramedical specialists towards end-of-life care and decisions in this particular population of patients. Despite the fact that the present questionnaire was sent out to the ERS mailing list for RICUs, the geographical distribution of responders is clearly unbalanced, with Italy and Germany accounting for >60% of the RICUs. Interestingly, this distribution is almost identical to that identified by the ERS survey on European RICUs 4, and is, therefore, in keeping with the high prevalence of units in some countries and lack of such facilities in others.

CONCLUSIONS

Despite the relatively low rate of response to the present survey, this study showed that, in European respiratory intermediate care units and high dependency units, an end-of-life decision is taken for ∼30% of the patients admitted. The most common practices were withholding treatment, the use of noninvasive mechanical ventilation as a ceiling therapy and provision of a do-not-resuscitate/do-not-intubate order, the latter occurring significantly more frequently in North Europe compared with South Europe. Patients, when competent, and their families are often involved, together with nurses, in reaching these key decisions.

Appendix

SECTION 1

1) How many patients did you admit to your Unit during the study period?

2) How many patients died in your Unit during the study period?

3) How many patients were discharged from your Unit in a terminal phase of their disease, after withdrawing and/or withholding any therapy?

4) How often did you give a “do not resuscitate” or “do not intubate” order in the group of patients admitted to your Unit?

5) How often did you practice withholding of therapy (eg. inotropic support, renal replacement therapy, invasive mechanical ventilation) in the group of patients admitted to your Unit?

6) Give the reasons more likely to influence your withholding decision, by ranking 1-10 in order of importance (most important first):

|__|__| Prediction of low probability of hospital survival

|__|__| Prediction of poor functional status after hospital discharge

|__|__| Perception of patient’s preference

|__|__| Patient’s estimate of his/her quality of life

|__|__| Decision of the patient

|__|__| Decision of the family

|__|__| Age

|__|__| Presence of co-morbidities

|__|__| Use of inotropes or vasopressor

|__|__| Use of haemodialysis

7) How often did you practice withdrawing treatment in the group of patients admitted to your Unit?

8) Give the reasons more likely to influence your withdrawing decision, by ranking 1-9 in order of importance (most important first):

|__| Prediction of poor likelihood of hospital survival

|__| Prediction of poor functional status after hospital

|__| Perception of patient’s preference (verbal or advance directive)

|__| Decision of the patient

|__| Decision of the family

|__| Age

|__| Presence of co-morbidities

|__| Use of inotropes or vasopressor

|__| Use of haemodialysis

9) How many patients receive NIV as the ceiling of ventilatory care (i.e. elective decision they will not receive invasive ventilation) if deterioration occurs?

10) How many patients received NIV solely as a palliative care strategy e.g. to reduce dyspnoea?

11) How many times did you refuse RICU/HDU admission in a patient with end-stage respiratory disorder and a poor life expectancy?

12) How often did you use the “principle of double effect” in patients admitted to your Unit?

(i.e. use of opiates or sedatives to reduce symptoms of dyspnoea, pain or distress in the knowledge this same treatment may also shorten life)

13) How often did you practice euthanasia in patients admitted to your Unit?

14) How many patients who received an end-of-life decision died in your Unit during the study period?

15) How many patients who received an end-of-life decision were affected by:

□ Spinal muscular atrophy Type II

□ Spinal muscular atrophy Type I

□ Motor neurone disease (ALS)

□ Duchenne muscular dystrophy

SECTION 2

16) How many of your patients who received an end-of-life decision were able to take an autonomous decision?

17) How often did you ask those patients about the “end-of-life” care?

18) How often did you ask the patient’s family about the “end of life” care?

19) How often did you seek “external” help in end-of-life decision?

20) If you were seeking external help, please specify

□ Ethical Committee

□ Other physician

□ Psychologist/psychiatrist

□ Clergyman, priest, imam or similar

□ Nurses

□ Respiratory therapist/physiotherapist

□ Other: please specify: _________________________________

21) How often did you involve the nurses in the end-of-life decisions?

22) How many times did you have a formal discussion with the patient and relatives at the time of admission to the Unit (or soon after) regarding end-of-life decisions?

23) How many patients and/or relatives did seek religious advice and support?

24) How many patients did employ written Advance Directives?

SECTION 3

25) Where are you from?

26) Where is your RICU located?

□ University hospital

□ University-affiliated hospital

□ Community hospital

□ Rehabilitation centre

□ Other - please specify: ___________________

27) How many beds has your Unit?

28) What is your religion?

□ None

□ Catholic

□ Muslim

□ Jewish

□ Protestant

□ Other - please specify: __________________

29) How old are you? |__|__| years

30) Gender: □ M □ F

31) How many scientific articles about end-of-life decision did you read in the study period?

32) How many symposia about end-of-life decision did you attend in the study period in the major national and international meetings?

33) Please list any national or international guidelines you observe on end of life care and withholding/withdrawing therapy.