Outcome in patients perceived as receiving excessive care across different ethical climates: a prospective study in 68 intensive care units in Europe and the USA

Outcome in patients perceived as receiving excessive care across different ethical climates: a... Purpose: Whether the quality of the ethical climate in the intensive care unit (ICU) improves the identification of patients receiving excessive care and affects patient outcomes is unknown. Methods: In this prospective observational study, perceptions of excessive care (PECs) by clinicians working in 68 ICUs in Europe and the USA were collected daily during a 28-day period. The quality of the ethical climate in the ICUs was assessed via a validated questionnaire. We compared the combined endpoint (death, not at home or poor quality of life at 1 year) of patients with PECs and the time from PECs until written treatment-limitation decisions ( TLDs) and death across the four climates defined via cluster analysis. Results: Of the 4747 eligible clinicians, 2992 (63%) evaluated the ethical climate in their ICU. Of the 321 and 623 patients not admitted for monitoring only in ICUs with a good (n = 12, 18%) and poor (n = 24, 35%) climate, 36 (11%) and 74 (12%), respectively were identified with PECs by at least two clinicians. Of the 35 and 71 identified patients with an available combined endpoint, 100% (95% CI 90.0–1.00) and 85.9% (75.4–92.0) (P = 0.02) attained that end- point. The risk of death (HR 1.88, 95% CI 1.20–2.92) or receiving a written TLD (HR 2.32, CI 1.11–4.85) in patients with PECs by at least two clinicians was higher in ICUs with a good climate than in those with a poor one. The differ - ences between ICUs with an average climate, with (n = 12, 18%) or without (n = 20, 29%) nursing involvement at the end of life, and ICUs with a poor climate were less obvious but still in favour of the former. Conclusion: Enhancing the quality of the ethical climate in the ICU may improve both the identification of patients receiving excessive care and the decision-making process at the end of life. Keywords: Perceived excessive care, Ethical climate, Decision-making, Interdisciplinary collaboration, Patient outcomes, Treatment-limitation decisions *Correspondence: dominique.benoit@ugent.be Department of Intensive Care Medicine, Ghent University Hospital, Corneel Heymanslaan 10, Ghent, Belgium Full author information is available at the end of the article The full list of investigators of the DISPROPRICUS study group are listed in the Acknowledgements and in the ESM 3 file. 1040 Introduction Take‑home message Life supporting therapy in intensive care units (ICUs) has been increasingly offered to patients with poor long- Enhancing the quality of the ethical climate in the ICU may improve term prognoses [1, 2], including those with advanced, both the identification of patients receiving excessive care and the end-of-life decision making process. end-stage organ dysfunction or a poor functional status [3–5]. While such therapies should not automatically be considered as non-beneficial, they should be provided Study design and center recruitment only to well-informed patients or relatives in accordance This 28-day observational study was conducted in 12 with their preferences and values, and only if treatment European countries and the United States. National coor- intensity remains proportional to the expected outcome dinators and local investigators were recruited from the [6, 7]. Nevertheless, one in three deaths occurs during or Ethics Section of the European Society of Intensive Care shortly after ICU treatment [2], frequently following dis- Medicine, the APPROPRICUS study group [8] and letters proportionate levels of care [8–13]. sent to experts in communication and end-of-life care in An ethically-based clinical decision-making process the ICU. National coordinators were expected to recruit has to rely on both individual perceptions and objective four centers in their country, translate the questionnaires criteria, followed by interdisciplinary discussions that into their own language using the Brislin method [19], enrich the process for the benefit of the patient. However, obtain ethics committee approval and assist the local expressing a perception of excessive care (PEC) to col- investigators in their data collection and data quality leagues, and more specifically to senior ones, necessitates tasks. Local investigators arranged study initiation meet- a safe climate in which clinicians are empowered to speak ings in their ICUs to enhance clinicians’ participation, up and in which they feel that their opinion is valued and recruited patients after having obtained informed con- subsequently integrated into the decision-making pro- sent and recorded data in a dedicated case-report form cess [14]. In addition to enhancing trust and cohesion on the www.DISPR OPRIC US.be website. in a team, such a climate may also reduce uncertainty in decision-makers by favoring intra- and interdisciplinary Data collection instruments and definition of combined transfer of knowledge, experience and values [14]. Sev- endpoint eral studies have already shown that concordant prognos- Country, hospital, ICU and clinician characteristics are tic estimates [15, 16] or perceptions of inappropriate [17] reported in the ESM 2. Hospital and ICU characteris- or futile care [18] by two clinicians may be considerably tics were collected by the local investigators between more predictive about the patient’s short- and long-term March and May 2014. Country-specific health variables outcomes than usually thought. However, whether the were retrieved from a prior publication [20]. In April and quality of the ethical climate prevailing in a unit further May 2014, clinicians in the participating ICUs completed improves the identification of patients receiving excessive questionnaires on personal characteristics, working con- care, and impacts on patient outcomes and written treat- ditions and the ethical climate prevailing in their units ment-limitation decision (TLD), is unknown. using the ethical decision-making climate questionnaire The objectives of the current multicenter study were (EDMCQ) [14]. This questionnaire consists of 35 items to assess whether the quality of the ethical climate in an with four- or five-point Likert scale options; 11 items ICU is associated with the prognostic value of PEC(s) are on end-of-life care practices; 11 on interdisciplinary with regard to patients’ one-year outcomes and with the reflection, collaboration, and communication and 13 on time from PEC(s) until written TLD during ICU stay or leadership skills of senior doctors. The theoretical frame - death. We hypothesized that the better the ethical cli- work and the validation of this instrument can be found mate, the more the PEC(s) would be predictive about in a previous publication [14]. patients’ one-year outcomes and the shorter the time Daily, during the 28  day study period (between May 4 until written TLD or death. and July 4, 2014), the clinicians anonymously completed a questionnaire about their perceptions of dispropor- Methodology tionate care for each of their patients. Disproportionate This study was approved by the ethics committees of all care was defined as care that is no longer consistent with participating centers and the Danish National Health the expected survival or quality of life (either “too much” Authority. Informed consent was required in all countries or “not enough” care), or that is provided against the to collect the one-year outcomes. The protocol, question - patient’s or relatives’ wishes. Questionnaire completion naires and case-report form are available in the electronic required less than 5  min per patient per day, when care supplementary material (ESM 1). was perceived as disproportionate, and less than 2  min 1041 otherwise. ICU mortality and length of stay were col- ANOVA tests where appropriate) for comparing continu- lected in all patients admitted in the ICU; those already ous variables. Results were expressed as number (%) and admitted prior to the study and those newly admitted median (25–75th percentiles), respectively. during the study period. The characteristics reported in the ESM 2 were collected in patients admitted for rea- Differences in patients’ combined endpoint at one year sons other than monitoring only during the study period. across ethical climates Categorization was left at the discretion of the attend- To simplify the analysis only perceptions of excessive ing physician. Written TLDs were ascertained by chart (“too much”) care were taken into account in the current review. study. As PEC by a clinician alone was only moderately Because staying at home with a good quality of life is predictive of the patient’s combined outcome compared highly valued by patients, the combined patient outcome to no PEC across all climates (ESM 2), and previous pub- in this study was defined as dead, not at home or a utility lications have highlighted the importance of concordance score < 0.5 at 1  year. This endpoint was defined during a between two clinicians [15–18], we compared the prob- study meeting with the national coordinators at the Euro- ability of attaining the combined endpoint for patients pean Society of Intensive Care Medicine congress in Bar- with PECs by at least two clinicians between the ethi- celona on September 30th 2014, approximately one year cal climates. For practical reasons, “PECs by at least two prior to data collection. Patients admitted for reasons clinicians” is referred to as “concordant PECs” through- other than monitoring only who were discharged alive, out the manuscript. Differences in combined endpoint or their families, were contacted by telephone or mail in patients without and with concordant PECs between one year after the ICU stay. The interviewer collected and within climates were compared with a Pearson’s Chi vital status, place of residence, and health-related qual- square and a Fisher’s exact test, respectively. ity of life using the EuroQoL-5D questionnaire [21], with conversion of each health state into a utility index (range Differences in time until death and treatment limitation − 0.1584 to 1.000). This questionnaire measures health in decisions across ethical climates five dimensions: mobility, self-care, usual activities, pain/ Time until identification of patients with concordant discomfort, and anxiety/depression. Each dimension has PECs, and from concordant PECs until written TLD three levels: no problems, moderate problems or severe or death were compared using (cause-specific) hazard problems. Therefore, patients can be classified into 1 of ratios, obtained via Cox regression (accounting for com- 243 possible health states, which is converted into the peting risks) [24]. The cause-specific hazard of an event corresponding utility index (range − 0.1584 to 1.000), expresses the instantaneous risk of that event at a given indicating the preference of being in a health status. A time for patients who are still alive in the ICU at that utility index < 0.5 corresponds with severely compro- time and have not previously experienced that event [24]. mised quality-of-life on at least one of the five dimen - To better explore the so-called “self-fulfilling prophecy sions. Although quality-of-life may be preferentially issue” (prognostication influenced by decision-making), evaluated from the patient, for some older patients prox- we compared the risk of death in patients with concord- ies may provide the most reliable information [22]. ant PECs in different decision-making scenarios (doctor– doctor, doctor–nurse, nurse–nurse) between and within Data analysis climates. Ethical climates: factor and cluster analysis Using the clinicians’ answers to the 35 EDMCQ items, Adjustment for case‑mix, hospital and country characteristics the data were first reduced via exploratory and confirma - To adjust for differential case-mix, hospital and country tory factor analysis to seven latent variables, also called characteristics between climates, we used inverse prob- factors [14]. The average score across clinicians for each ability weighting based on propensity scores [25]. Here, factor in a given ICU was used as input for the cluster the propensity score is the probability of being treated analysis at ICU level (ESM 2). Such analyses seek to mini- in one’s own climate, as obtained using a multinomial mize the similarity of ICUs within climates and maximize model based on patient, hospital and country charac- the dissimilarity of ICUs between climates. In particular, teristics. Adjustment based on propensity scores has we used the partitioning around medoids (PAM) algo- the advantage, relative to other adjustment methods, of rithm to classify the different climates into a pre-speci - preventing model extrapolation, when climates are very fied number of clusters. This algorithm was chosen in different in terms of these characteristics [25]. However, view of its robustness to outliers and noise [23]. Pearson’s one concern about adjustment for case-mix is that it may chi square tests were used for comparing categorical eliminate the effects of potential differences in admis - variables between climates and Kruskal–Wallis tests (or sion policy (which affects case-mix) between climates. 1042 Therefore, we considered the unweighted results as our severe underlying comorbidities and with greater use principal results. These are expressed as proportions and of advanced and prolonged life-supporting treatments (cause-specific) hazard ratios (HR) along with 95% confi - in the post-operative setting, compared to the other dence intervals (95% CI). Two-sided P values were con- climates. sidered significant at the 0.05 level. Priority was given to comparisons between the good and the poor ethical cli- Differences in patients’ combined endpoint at one year mates (see results) in order to reduce type I errors. We across ethical climates refer to the ESM 2 for a more detailed methodology. Of the 1761 patients admitted for more than only moni- toring with data concerning time until event available Results (Fig.  1), 74 (4.2%) patients were perceived as receiving Ethical climates excessive care by two clinicians, and 107 (6.1%) by more Of 4747 clinicians working in 68 ICUs in Belgium, Czech than two clinicians, resulting in 36 (11.0%), 50 (7.2%), 21 Republic, Denmark, France, Germany, Greece, Hungary, (18.0%) and 74 (12.0%) patients with concordant PECs Italy, Portugal, United Kingdom, Sweden, the Nether- from the good to the poor climate, respectively. Excessive lands and the United States, 2992 (62.6%) completed the care was perceived by these clinicians as being provided EDMCQ (Fig. 1). against the patients’ or relatives’ wishes in 20 (55.5%), The cluster analysis based on the average scores of 25 (50.0%), 11 (52.4%) and 41 (55.4%) (P = 0.94) of these the seven factors identified during the validation of the patients. EDMCQ [14] yielded four different meaningful, mutually The differences in the patients’ combined outcomes exclusive ethical climates. Visual inspection of the scree across ethical climates are reported in Table 1. The prob - plot (ESM 2) revealed that clustering into three clusters abilities of attaining the combined endpoint in patients would drastically increase the total intra-cluster varia- without concordant PECs was 53.5% (95% CI 46.8–60.2), tion (as opposed to using four clusters), while clustering 59.1% (54.6–63.6), 64.0% (53.1–74.9) and 51.8% (47.3– into five clusters would only minimally decrease the total 56.3) from good to poor climate, respectively (P = 0.057, intra-cluster variation [23]. These climates were denomi - difference between good and poor climate, P = 0.74). nated by experts in intensive care (DB, JD), psychology These probabilities increased to 100% (90.0–100), 95.6% (+) (BV, SV) and ethics (RP) as: good, average with and (84.3–98.9), 94.7% (70.6–99.3) and 85.9% (75.4–92.0) in (−) without nurses’ involvement at end-of-life, and poor patients with concordant PECs (P = 0.047, difference (Fig.  2, ESM 2). According to clinicians working in a between good and poor climate, P = 0.020). good climate, leadership by senior doctors is active and facilitates interdisciplinary reflection and decision-mak - Differences in time until death and treatment limitation ing overall. This climate is also characterized by mutual decisions across ethical climates respect, which is pre-requisite to facilitating interdisci- We found no difference in incidence or in time from plinary reflection and ethical awareness [14]. Within the admission until concordant PECs between the good and (+) average climate, clinicians perceive that senior doctors the poor climates; approximately 11% of the patients empower nurses to share interdisciplinary decision-mak- were identified with concordant PECs after 14  days in ing, mainly at end-of-life. Even though clinicians working both climates (Fig. 3a). (−) in an a verage climate believe that their senior doctors The risk of death in patients with concordant PECs are able to make decisions, they do not find them pro - was statistically significantly higher (HR 1.88, 95%CI moting nurse involvement in decision making at end-of- 1.20–2.92) in the good compared to the poor climate. life. Finally, clinicians working in a poor climate perceive The median time until death in patients with concordant a need for improvement in all of these factors. PECs was 5  days (2–18) vs. 14 (6–34) days (P = 0.008), The ICU, clinician, and patient characteristics for each respectively. The difference between the average climates (−) climate are reported in ESM 2. The average and poor and the poor climate was less important, but still in favor climates were more prevalent in Central and South- of the average climates (Fig.  3c). The risk of death in the ern European countries (P < 0.001); however, 10 of the good climate was higher in patients with PECs by two or 24 (41.7%) ICUs with a poor climate were situated in more doctors than in those with PECs by two or more Western Europe and the United States. The ICU experi - nurses (HR 3.13, 95% 1.19–8.23), with the risk of death in ence of clinicians was similar across climates, however, patients with PECs by at least one nurse and one doctor the number of participating doctors was higher in the being intermediate. There was no evidence of such a dif - (−) average and poor, compared to the other two climates. ference in risk of death in the poor climate (HR 0.74, 95% (−) The average and poor climates were also associated 0.29–1.86) (ESM 2). with a slightly higher number of admitted patients with 1043 15 countries 1 country did not participate 1 country unprepaired Phase I 13 countries, 68 ICUs Phase II Ethical climate 2992 clinicians (63%of 4747) analysis (Fig 2) Phase III 3528 patients included of which 2935 patients got 29136 perceptions by 2562 clinicians 1351 admitted during the study period for 353 admitted prior to the study period monitoring only 1824 patients admitted for more than monitoring only during the study period of which 1558 got17703 perceptions by 2244 clinicians 63 (3.5%) missing time until event Analysis of time until 1761 patients ≥ 2PECs(Fig 3a-b) 266withno perceptions 1580 patients 181patients 74 with exact 2 PEC 1126 with no PEC with ≥ 2 PEC without ≥ 2 PEC 107withmore than 2 PEC 188with 1 PEC Analysis of time from ≥ 2 Phase IV PECs to TLD or death (Fig 3c-f) 355 missing 10 missing 171 with ≥ 2 PECwith known 1225 patients without ≥ 2 PECwith 1 year combined outcome: known1 year combined outcome: Analysis of the prognostic value of ≥ 2 PECs with 92.4%(158/171) 55.6% (681/1225) regard to one year outcomes 450 dead, 231 not at home OR utility < 0.5 146 dead, 12 not at home OR utility < 0.5 Fig. 1 Flow chart. Phase I: Recruitment and data collection of hospital and ICU characteristics, Phase II: Ethical climate data collection, Phase III: Daily perceptions of clinicians and collection of patient characteristics during the 28 days study period, Phase IV: Collection of patients’ one year outcomes. PEC(s) perception(s) of excessive care, TLDs treatment-limitation decisions 1044 Fig. 2 Ethical climates. Factor and cluster analysis were used to obtain mutually exclusive climates. Factor analysis attributes and aggregates the 35-item ethical decision-making climate questionnaire into seven factors for each clinician, which describe different aspects of the ethical decision- making climate as perceived by that clinician. These were subsequently averaged across clinicians to obtain seven factor scores per ICU [14]. A (+) (−) cluster analysis based on these averages scores identified four meaningful ethical climates; good, average with and without involvement of nurses at end-of-life (EOL), and poor. The figure visualizes the average factor scores in clinicians per climate. Larger values indicate better agreement with the climate expressed by the corresponding factor. More detailed information can be found in the ESM 2 Table 1 Differences in patients’ one ‑year outcomes across ethical climates in patients with and without concordant PECs Ethical climate P value overall P value good vs. poor climate (+) (−) Good Average Average Poor Patients without concordant PECs (n= 1225) n = 215 n = 464 n = 75 n =471 Combined endpoint 115 (53.5%) 274 (59.1%) 48 (64.0%) 244 (51.8%) 0.057 0.740 Dead 68 (31.6%) 175 (37.8%) 39 (52.0%) 168 (35.7%) Alive not at home or utility < 0.5 47 (21.9%) 99 (21.3%) 9 (12.0%) 76 (16.1%) Patients with concordant PECs (n= 171) n =35 n =46 n = 19 n = 71 Combined endpoint 35 (100%) 44 (95.6%) 18 (94.7%) 61 (85.9%) 0.047 0.020 Dead 33 (94.3%) 41 (89.1%) 18 (94.7%) 54 (76.0%) Alive not at home or utility < 0.5 2 (5.7%) 3 (6.5%) 0 (0.0%) 7 (9.9%) After weighting to adjust for differential case-mix, hospital and country characteristics, the probability of attaining the combined endpoint in patients without and with concordant PECs was 56, 62, 60 and 55% (P = 0.26, difference between good and poor climate, P = 0.82) and 100, 93.9, 93.5 and 86.2% (P = 0.042, difference between the good and the poor climate, P = 0.017) from the good to the poor climate, respectively Patients with concordant PECs had a higher chance Adjustment based on propensity scores of receiving a written TLD in the good compared to the After weighting to adjust for differential case-mix, hos - poor climate (cause-specific HR 2.32, 95%CI 1.11–4.85) pital and country characteristics, the probability of (Fig. 3e). attaining the combined endpoint in patients without 1045 Fig. 3 a–f Competing risk analyses of time from admission until concordant perceptions of excessive care (PECs) by at least two different clinicians, written treatment-limitation-decision ( TLD) and death before and after weighting for country, hospital and patients characteristics using propensity scores. The primary endpoint (dead, not at home or a utility < 0.5 according the EuroQoL-5D questionnaire [21] at one year) is visualized separately in c, d. The sudden increase at day 365 represents the proportion of patients alive with a utility < 0.5 or not living at home. The incidence of the primary endpoint differs from the text because drop-outs are taken into account in competing risk analyses. The results are expressed as (cause- specific) hazard ratios (HR) together with 95% confidence intervals (CI). To avoid type I errors, we gave priority to comparisons between the most extreme (good and poor) climates concordant PECs was 55.8% (48.2–63.1), 62.1% (56.5– 100% (90.0–100), 93.9% (74.3–98.8), 93.5% (64.2–99.1) 67.4), 60.2% (47.4–71.7) and 54.8% (49.4–60.1) from and 86.2% (72.0–93.8), respectively (P = 0.042, differ - good to poor climate, respectively (P = 0.26, difference ence between the good and the poor climate, P = 0.017). between good and poor climate, P = 0.82). These prob - The risk of death in patients with concordant PECs abilities increased in patients with concordant PECs to also remained higher in the good vs. the poor climate 1046 Fig. 3 continued (HR 1.79, 95%CI 1.07–2.98) (Fig.  3d). The median time We preferred to focus on the intuitive-heuristic more until death was 5 (2–18) and 14 (7–30) days (P = 0.026), than the analytic-deductive part of the complex ethical respectively. The risk of death in the good climate decision-making process [26, 27], by asking clinicians remained higher in patients with PECs by two or more whether they felt that the care provided to their patient doctors than in those with PECs by two or more nurses on a specific day was consistent with the expected out - (HR 3.58, 95% 1.42–9.02), with the risk of death in come in terms of survival and quality of life, and whether patients with PECs by at least one nurse and one doctor this amount of care was in line with the patient’s or rela- remaining intermediate. There was no evidence of such a tives’ wishes. We also didn’t focus on futile care, such as difference in risk of death in the poor climate (HR 1.58, in the studies of Neville et al. [18], because this terminol- 95% 0.45–5.55) (ESM 2). ogy presupposes a high degree of certainty concerning However, we no longer found evidence of a differ - the final fatal prognosis, whereas nowadays technologi - ence in time until TLD between the good and the poor cal innovation frequently excludes patients’ spontane- climates (cause-specific HR 1.76, 95%CI 0.73–3.92) ous death in ICU [6, 7]. By doing so, we acknowledged (Fig. 3f ). uncertainty [26] (benefit vs. harm) and patient and fam - ily autonomy, as an integral part of the complex ethical Discussion decision-making process at the bedside [28]. Neverthe- In this large, multicenter, prospective, ICU study, we less, PEC was highly predictive about patients’ one-year found that concordant PECs by at least two clinicians outcomes, more specifically when expressed by two or were far more predictive about the primary composite more than two clinicians. endpoint of death, not living at home, or having poor Concordant PECs by at least two different clini - quality of life one year after ICU admission, compared cians were more predictive about the combined end- to absence of PEC. We found evidence of a difference in point in the good compared to the poor ethical climate one-year outcomes, time until death and written TLD (P = 0.028). Patients with concordant PECs also had in patients with concordant PECs across the four ethical a higher risk of death and of receiving a written TLD climates identified by our questionnaire. The evidence of in the good compared to the poor climate. The differ - a difference in time until written TLD disappeared after ence in endpoints between the average and the poor cli- adjusting for differential case-mix, hospital and country mates was less obvious, but still in favor of the former characteristics. compared to the latter, thus objectively validating our In contrast to the study by Detsky et  al. [16], clini- EDMCQ instrument [14]. Unfortunately, we can neither cians in our study were not explicitly expected to pro- exclude nor confirm self-fulfilling prophecy in the good vide prognostic estimates about the patients’ outcomes. climate. However, it is of note that it took about 14 days 1047 to identify all patients with concordant PECs in both cli- collaboration [6, 8, 14, 32, 38, 39], and early involvement mates and, for half of these patients, another 5 days to die of palliative care [30, 37, 40]. Our EDMCQ instrument in the good vs. 14  days in the poor climate (P = 0.002). may be used for that purpose [14, 32]. In line with the results of the EDMCQ, this suggests Our study has several limitations. First, the participat- that the decision to forgo life sustaining treatment in the ing ICUs were not selected at random, which may have good climate was not premature, and once excessive care affected the external validity of our results. Second, inclu - was perceived by at least two clinicians, it occurred in a sion of patients was left at the discretion of the attending (−) timely fashion. Furthermore, in a sub-analysis, we found doctor. However, except in the average climate (ESM no difference in risk of death between patients with con - 2), we found no evidence of a difference in ICU mor - cordant PECs by different professionals in the poor cli - tality rates or length of stay in the subgroup of patients mate, as opposed to the good climate. This indicates that admitted for monitoring only across climates, indicating identification of patients with excessive care by doctors that the attending doctors included patients in a similar in the poor climate was not followed by active decision- way. We further minimized confounding bias by account- making. In addition to, respectively, increasing the risk ing for differences in case-mix, using inverse probability of prolonged suffering and complicated grief in patients weighting based on propensity scores. Third, we did not and relatives [29, 30], decision-paralysis as a strategy to use classical severity-of-illness scores in our analysis. cope with prognostic uncertainty [8, 12, 31] may also However, in line with our primary objective, we preferred induce moral distress and increase intention to leave in to include short- and long-term prognostic factors [4, 5] clinicians [6, 32–34]; a fact that is even more pertinent that are commonly used by clinicians during decision- considering the high number of concordant PEC records making, rather than classical severity-of-illness scores perceived as violating the patient’s or relatives’ wishes which have never been validated for predicting long- in this study. After weighting for the specific case-mix term outcomes. Fourth, one has to keep in mind that the within a hospital and country, only the risk of receiving incidence of patients with concordant PECs is probably a written TLD in the good compared to the poor climate underestimated, as patients admitted prior to the study was no longer significantly different. This may suggest period and those who remained in ICU for longer than that the quality of the ethical climate in an ICU is impor- the study period (and were expected to reach more cli- tant in identifying patients receiving excessive care and in nician concordance with time) were excluded from the subsequently triggering the decision-making process at analysis. Finally, although the ICU experience of clini- end-of-life, whereas formalizing that process via a writ- cians was similar, we cannot exclude that the lower num- ten TLD seems more case-mix and culture dependent. ber of participating doctors in the good compared to the This is in line with previous studies showing a huge vari - poor climate may have biased our results in favor of the ability in written TLDs between countries and ICUs [35]. latter, concealing even larger differences between the two. The probability of dying or surviving with a poor qual - ity of life at one year in patients without concordant PECs Conclusion was 53.5, 59.1, 64.0 and 51.8% from good to poor climate, Our results suggest that improving the quality of the ethi- respectively, largely exceeding that of many malignancies cal climate in ICU may favor the identification of patients [36]. Therefore, in line with the definition of dispropor - receiving excessive care and the subsequent decision- tionate care [6, 8, 9], clinicians did not find poor prog - making process at end-of-life. This may benefit the qual - nosis sufficient by itself to lead to a PEC. Concordant ity of the dying process in ICUs. PECs by at least two clinicians increased the probability Electronic supplementary material of reaching the combined endpoint to 100% in the good, The online version of this article (https ://doi.org/10.1007/s0013 4-018-5231-8) (+) (−) 95.6% in the average and 94.7% in the average cli- contains supplementary material, which is available to authorized users. mate, compared to 85.9% in the poor. Despite the poor prognosis we found a relatively low incidence of writ- Author details ten TLDs within the 14  days in these patients; ranging Department of Intensive Care Medicine, Ghent University Hospital, Corneel Heymanslaan 10, Ghent, Belgium. Department of Intensive Care Medicine, from 20% in the poor to only 35% in the good and about Vejle Hospital, Vejle, Denmark. Institute of Regional Research, University 45% in average climates (P = 0.011). Although caution 4 of Southern Denmark, Odense C, Denmark. Department of Anaesthesiology in interpreting this result is required due to small sam- and Intensive Care, Sahlgrenska University Hospital, Gothenburg, Sweden. 5 6 King’s College Hospital, London, UK. Department of Medical Oncology, ple size, these probabilities highlight the urgent need for University of Groningen, University Medical Center Groningen, Groningen, improving advance-care planning before ICU admission 7 The Netherlands. Hôpital Saint-Louis and University, Paris-7, Paris, France. [37], as well as triage and decision-making at end-of-life Department of Anesthesiology and Intensive Care, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Prague, in ICU. This should more specifically be achieved via Czech Republic. Department of Anesthesia, Critical Care, and Pain Medicine, ethical climates that favor interdisciplinary reflection and Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, 1048 MA, USA. Service des soins intensifs et urgences oncologiques, Institut Jules Gaia (Paula Fernandes, Ana Isabel Paixão), Instituto Português de Oncologia, Bordet, ULB, Brussels, Belgium. SCDU Anestesia e Rianimazione, Azienda Porto (Filomena Faria), Sweden: Sahlgrenska University Hospital, Gothenburg and Ospedaliero Universitaria, “Maggiore della Carità”, Novara, Italy. Semmel- (Johan A. Malmgren), Sahlgrenska University Hospital/Östra, Gothenburg weis University Budapest, Budapest, Hungary. Intensive Care Department, (Bertil Andersson), Skåne University Hospital, Malmö (Eva Åkerman), Karolinska Hospital S.António, Porto, Portugal. Tettnang Hospital, Tettnang, Germany. University Hospital, Karolinska (Andreas Hvarfner), The Hospital of Norrköping, Department of Psycho-analysis and Clinical Consulting, Faculty of Psychol- Norrköping (Robert Svensson), United Kingdom: King’s College Hospital, Lon- ogy and Educational Sciences, Ghent University, Ghent, Belgium. Depart- don ( Victoria Metaxa), USA: Beth Israel Deaconess Medical Center and Harvard ment of Intensive Care Medicine, Erasmus MC University Medical Center Medical School, Boston MA (Daniel Talmor, Ariel Mueller, Valerie Banner-Good- Rotterdam, Rotterdam, The Netherlands. Department of Applied Mathemat- speed), Henry Mayo Newhall Memorial Hospital, Valencia, CA (Dee Rickett), ics, Computer Science and Statistics, Faculty of Sciences, Ghent University, Mayo Clinic, Rochester, MN (Michael E. Wilson, Richard Hinds). Ghent, Belgium. London School of Hygiene and Tropical Medicine, London, UK. Department of Geriatric Medicine, Ghent University Hospital, Ghent, Author Contributions Belgium. Study concept and design: DDB, BVB, RDP. Design of the questionnaire: DDB, HIJ, JM, SV, EJOK, JD, BVB, EA, RDP. Coordination of the translation of the Acknowledgements questionnaire: HIJ, JM, VM, AKR, MD, KR, DT, APM, LC, LZ, PM, AM. Acquisition of This study was supported by a European Society of Intensive Care Medi- data: DDB, HIJ, JM, VM, AKR, MD, KR, DT, APM, LC, LZ, PM, AM, BG. Analysis and cine/European Critical Care Research Network clinical research award and a interpretation of data: DDB, SV, SV, SV, BVB, EA, RDP. Drafting of the manuscript: Fonds voor Wetenschappelijk Onderzoek senior clinical investigators grant DDB, VM, DT, SV, BVB, EA, RDP. Critical revision of the manuscript for important (1800513N) obtained in 2012 by DB. We are grateful to Ariella Van Sompel for intellectual content: DDB, HIJ, JM, VM, AKR, MD, KR, DT, APM, LC, LZ, PM, AM, SV, having performed the factor and cluster analysis together with VDB and RP EJOK, JD, SV, SV, BG, BVB, EA, RDP. Statistical expertise: SV, SV. Obtained funding: (under supervision of SVH and SVS) and Jolien Roels for having performed the DDB, JD. Administrative, technical, or material support: DDB, JD, BG. Steering data cleaning and the univariate analysis (under supervision of DB, SVB and committee: DDB, SV, EJOK, JD, SV, BG, BVB, EA, RDP. SVS). Participating centers and local investigators: Belgium: University Hospital, Vrije Universiteit Brussel, Brussels (Herbert Spapen, Marie-Claire Van Malderen, Compliance with ethical standards Godelieve Opdenacker), Leuven University Hospital, Leuven (Geert Meyfroidt, Dieter Mesotten, Joost Wauters, Marie Van Laer and Alexander Wilmer, Joost Conflicts of interest Wauters, Helga Ceunen), ZNA Stuivenberg, Antwerpen (Inneke E De Laet, DB reports grants from Gilead, Astellas, Fisher-Paykel, Baxter, Alexion and Anita Jans), Ghent University Hospital, Gent (Dominique Benoit, Sandra Oeyen, Fresenius Kabi outside the submitted work. KR reports honoraria from Alexion, Ingrid Herck, Stephanie Bracke, Charlotte Clauwaert), Institut Jules Bordet, outside the submitted work. MD reports grant from MSD and Jazz Pharma, Bruxelles (Meert Anne-Pascale, Leclercq Nathalie), CHU-Brugmann, Bruxelles personal fees from Astellas and Bristol-Myers Squibb, and non-financial sup - (Devriendt Jacques), CHU Saint Pierre, Bruxelles (Dechamps Philippe), Czech port from Astellas, Bristol-Myers Squibb, Astute Medical, and Sanofi Aventis. Republic: Liberec District Hospital, Liberec (Ivana Zykova), Masaryk University, EA reports grants and personal fees from Gilead, Alexion, MSD, Cubist and Brno and University Hospital, Brno (Jan Malaska), Third Faculty of Medicine, personal fees from Baxter, outside the submitted work. All other authors have Charles University, Prague (Matous Schmidt), Hospital and Polyclinic Havirov, no conflict of interest to report. Havirov (Igor Satinsky), Institute for Experimental and Clinical Medicine, Prague (Eva Kieslichova), 3rd Medical Department, First Faculty of Medicine, Charles Open Access University in Prague and General University Hospital, Prague (Jarmila Krizova), This article is distributed under the terms of the Creative Commons Attribu- Karlovy Vary District Hospital, Karlovy Vary (Robert Janda), Pardubice District tion-NonCommercial 4.0 International License (http://creativecommons.org/ Hospital, Pardubice (Magdalena Fortova, Jiri Matyas), First Faculty of Medicine, licenses/by-nc/4.0/), which permits any noncommercial use, distribution, and Charles University and General University Hospital, Prague (Katerina Rusinova, reproduction in any medium, provided you give appropriate credit to the Ondrej Kopecky), Denmark: Herning Hospital, Herning (Christian Alves Køhler original author(s) and the source, provide a link to the Creative Commons Pedersen), Kolding Hospital, Kolding (Stine Hebsgaard), Vejle Hospital, Vejle license, and indicate if changes were made. (Rikke Frank Aagaard Johnsen), Holbæk Hospital, Holbæk ( Tina Charlotte Bitsch Hansen), France: Saint-Etienne University Hospital and Jacques Lisfranc Medical School, Saint-Etienne (Michael Darmon), Saint-Louis University Received: 10 February 2018 Accepted: 14 May 2018 Hospital, APHP, Université Paris-7, Paris (Danielle Reuter, Elie Azoulay), Institut Published online: 28 May 2018 Paoli Calmette, Marseilles (Djamel Mokart), Montfermeil Hospital, Montfermeil (François Vincent), Germany: University Hospital Jena, Jena (Christiane S. Har- tog), Viersen General Hospital, Viersen (Peter Gretenkort), Tettnang Hospital, Tettnang (Andrej Michalsen), Greece: Agia Olga Hospital, Athens (Aikaterini References Kounougeri), Evangelismos Hospital, Athens (Serafim Nanas), Agios Pavlos 1. Wunsch H, Linde-Zwirble WT, Harrison DA, Barnato AE, Rowan KM, Angus Hospital, Thessaloniki (Despina Papachristou), AHEPA University Hospital, DC (2009) Use of intensive care services during terminal hospitalizations Thessaloniki, (Ioanna Soultati), G.Gennimatas Hospital, Thessaloniki (Dimitrios in England and the United States. Am J Respir Crit Care Med 180:875–880 Lathyris), Hippokratio General Hospital, Thessaloniki (Marili Pasakiotou), Papa- 2. Teno JM, Gozalo PL, Bynum JPW, Leland NE, Miller SC, Morden NE, Scupp georgiou General Hospital, Thessaloniki (Marina Oikonomou), Hungary: Sem- T, Goodman DC, Mor V (2013) Change in end-of-life care for medicare melweis University Budapest, Budapest (Gábor Élő, Orsolya Szűcs), Kaposi Mór beneficiaries. Site of death, place of care, and health care transtitions in Teaching Hospital, Kaposvár University, Kaposvár (János Fogas), St. Stephen 2001 2005, and 2009. JAMA 309:470–477 and St. Leslie Metropolitan Hospital, Budapest (Ilona Bobek), Italy: Azienda 3. Ho TH, Barbera L, Saskin R, Lu H, Neville BA, Earle CC (2011) Trends in Ospedaliero Universitaria, “Maggiore della Carità”, Novara, and Department of aggressiveness of end of life cancer care in the universal health care Translational Medicine, Università del Piemonte Orientale, Novara (Francesco system of Ontario, Canada. J Clin Oncol 29:1587–1591 Della Corte, Carlo Olivieri, Rosanna Vaschetto, Laura Cancelliere), Ospedale 4. Wang CY, Calfee CS, Paul DW, Janz DR, May AK, Zhuo H, Bernard GR, Civile San Salvatore, and Department of Life, Health and Environmental Matthay MA, Ware LB, Neudoerer K ff angelaris K (2014) One-year mortality Sciences (MeSVA), University of L’Aquila and Department of Emergency, San and predictors of death among hospital survivors of acute respiratory Salvatore Hospital, L’Aquila (Franco Marinangeli, Tullio Pozone, Alessandra distress syndrome. Intensive Care Med 40:388–396 Ciccozzi), The Netherlands: Canisius Wilhelmina Ziekenhuis, Nijmegen (A. 5. Garland A, Olafson K, Ramsey CD, Yogendran M, Fransoo R (2014) Distinct Schouten, Monique Bruns), Medical Center Leeuwarden, Leeuwarden (Rik T. determinants of long-term and short-term survival in critical illness. Gerritsen, Matty Koopmans), Erasmus University Hospital of Rotterdam (Erwin Intensive Care Med 40:1097–1105 Kompanje, Ditty van Duijn), University of Groningen and University Medical 6. Kompanje EJ, Piers RD, Benoit DD (2013) Causes and consequences of Center Groningen, Groningen (Jan G. Zijlstra, Anne KL Reyners), Wilhelmina disproportionate care in intensive care medicine. Curr Opin Crit Care Ziekenhuis Assen, Assen (Johan G. Lutisan), Portugal: Hospital S.António, Porto 19:630–635 (Raquel Monte, José António Pinho, Pedro Pimenta), CHVNG, Vila Nova de 1049 7. Bosslet GT, Pope TM, Rubenfeld GD, Lo B, Truog RD, Rushton CH, Curtis 22. Oeyen SG, Vandijck D, Benoit DD, Annemans L, Decruyenaere J (2010) JR, ford DW, Osborne M, Misak C, Au DH, Azoulay E, Brody B, Fahy BG, Hall Quality of life after intensive care: a systematic review of the literature. JB, Kesecioglu J, Kon AA, Lindell KO, White DB, on behalf of the American Crit Care Med 38:386–400 Thoracic Society ad hoc committee on Futile and Potentially Inappropri- 23. Everitt BS, Landau S, Leese M, Stahl D (2011) Cluster analyis, 5th edition. ate Treatment (2015) An official ATS/AACN/ACCP/ESICM/SCCM policy Wiley. ISBN: 978-0-470-74991-3 statement: responding to requests for potentially inappropriate treat- 24. Putter H, Fiocco M, Geskus RB (2007) Tutorial in biostatistics: competing ments in intensive care units. Am J Respir Crit Care Med 191:1318–1330 risks and multi-state models. Statist Med 26:2389–2430 8. Piers R, Azoulay E, Ricou B, Dekeyser Ganz F, Decruyenaere J, Max A, 25. Rubin DB (1997) Estimating causal effects from large data sets using Michalsen A, Azevedo Maia P, Owczuk R, Rubulotta F, Depuydt P, Meert propensity scores. Ann Intern Med 127:757–763 AP, Reyners A, Aquilina A, Bekaert M, Van den Noorgate N, Schrauwen W, 26. Lighthall GK, Vazquez-Guillamet C (2015) Understanding decision making Benoit D (2011) APPROPRICUS study group of the ESICM. Perception of in critical care. Clin Med Res 13:156–168 appropriateness of care among European and Israeli intensive care unit 27. Zollo L, Pellegrini MM, Ciappei C (2017) What sparks ethical decision mak- nurses and doctors. JAMA 306:2694–2703 ing? The interplay between moral intuition and moral reasoning: lessons 9. Piers R, Azoulay E, Ricou B, Dekeyser Ganz F, Max A, Michalsen A, from the scholastic doctrine. J Bus Ethics 145:681–700 Azevedo Maia P, Owczuk R, Rubulotta F, Meert AP, Reyners A, Decruy- 28. Luce JM, White DB (2009) A history of ethics and law in the intensive care enaere J, Benoit DD (2014) Inappropriate care in European ICUs. unit. Crit Care Clin 25:221–237 Confronting views from nurses and junior and senior physicians. Chest 29. Wright AA, Zhang B, Ray A, Mack JW, Trice E, Balboni T, Michell SL, Jackson 146:267–275 VA, Block SD, Maciejewski PK, Prigerson HG (2008) Associations between 10. Anstey MH, Adams JL, McGlynn EA (2015) Perceptions of the appropriate- end-of-life discussions, patient mental health, medical care near death, ness of care in California adult intensive care units. Crit Care 19:51 and caregiver bereavement adjustment. JAMA 300:1665–1673 11. Vincent JL (1999) Forgoing life support in western European inten- 30. Curtis JR, Vincent JL (2010) Ethics and end-of-life care for adults in the sive care units: the results of an ethical questionnaires. Crit Care Med intensive care unit. The Lancet 376:1347–1353 27:1626–1633 31. Kimbell B, Murray SA, Macpherson S, Boyd K (2016) Embracing inherent 12. Palda VA, Bowman KW, Mclean RF, Chapman MG (2005) “Futile” care: do uncertainty in advanced illness. BMJ 354:i3802 we provide it? Why? A semistructured, Canada-wide survey of intensive 32. Dzeng E, Curtis JR (2018) Understanding ethical climate, moral distress, care unit doctors and nurses. J Crit Care 20:207–213 and burnout: a novel tool and a conceptual framework. BMJ Qual Saf:1–5. 13. Giannini A, Consonni D (2006) Physicians’ perceptions and attitudes https ://doi.org/10.1136/bmjqs -2018-00790 5 (Epub ahead of print) regarding inappropriate admissions and resource allocation in the inten- 33. Hamric AB, Blackhall LJ (2007) Nurse-physician perspective on the care of sive care setting. Br J Anaesth 96:57–62 dying patients in intensive care units: collaboration, moral distress, and 14. Vandenbulcke B, Piers R, Jenssen HI, Malmgrem J, Metaxa V, Reyners AK, ethical climate. Crit Care Med 35(2):422–429 Darmon M, Rusinova K, Talmor D, Meert AP, Cancellierre L, Zubek L, Maia 34. Schwarzkopf D, Rüddel H, Thomas-Rüddel DO, Felfe J, Poidinger B, P, Michalsen A, Decruyenaere J, Kompanje EJO, Azoulay E, Vlerick P, Van Matthäus-Krämer CT, Hartog CS, Bloos F (2017) Perceived nonbeneficial den Sompel A, Vansteelandt S, Vanheule S, Benoit DD (2018) On behalf treatment of patients, burnout, and intention to leave among ICU nurses the DISPROPRICUS study group of the Ethics Section of the ESICM. Ethical and junior and senior physicians. Crit Care Med 45:e265–e273 decision-making climate in the ICU: theoretical framework and validation 35. Mark NM, Rayner SG, Lee NJ, Curtis JR (2015) Global variability in with- of self-assessment tool. BMJ Qual Saf. https ://doi.org/10.1136/bmjqs holding and withdrawal of life-sustaining treatment in the intensive care -2017-00739 0 ([Epub ahead of print]) unit: a systematic review. Intensive Care Med 41:1572–1585 15. Meadow W, Pohlman A, Reynolds D, Rand L, Correia C, Christoph E, Hall 36. De Angelis R, Sant M, Coleman MP, Francisci S, Baili P, Pierannunzio D, J (2014) Power and limitations of daily prognostications of death in the Trama A, Visser O, Brenner H, Ardanaz E, Bielska-Lasota M, Engholm G, medical ICU for outcomes in the following 6 months. Crit Care Med Nennecke A, Siesling S, Berrino F, Capocaccia R, Eurocare-5 Working 42:2387–2392 Groups (2014) Cancer survival in Europe 1999–2007 by country and 16. Detsky ME, Harhay MO, Bayard DF, Delman AM, Buehler AE, Kent SA, age: results of the Eurocare-5. A population-based study. Lancet Oncol Cuiffetelli IV, Cooney E, Gabler NB, Ratcliffe SJ, Mikkelsen ME, Halpern SD 15:23–34 (2017) Discriminative accuracy of physician and nurse predictions for 37. Khandelwal N, Kross EK, Engelberg RA, Coe NB, Long AC, Curtis JR (2015) survival and functional outcomes 6 months after ICU admission. JAMA Estimating the effect of palliative care interventions and advance care 317:2187–2195 planning on the ICU utilization: a systematic review. Crit Care Med 17. Singal RK, Sibbald R, Morgan B, Quinlan M, Parry N, Radford M, Martin CM 43:1102–1111 (2014) A prospective determination of the incidence of perceived inap- 38. Puntillo KA, McAdam JL (2006) Communication between physicians propriate care in critically ill patients. Can Respir J 21:165–170 and nurses as a target for improving end-of-life care in intensive care 18. Neville TH, Wiley JF, Yamamoto MC, Flitcraft M, Anderson B, Curtis JR unit: challenges and opportunities for moving forward. Crit Care Med (2015) Concordance of nurses and physicians on whether patients are 34(Suppl.):S332–S340 receiving futile treatment. Am J Crit Care 24:403–411 39. Jensen HI, Ammentrop J, Erlandsen M, Ording H (2011) Witholding or 19. Brislin RW (1976) Translation: application and research. Gardner Press Inc, withdrawing therapy in intensive care units: an analysis of collaboration New York among healthcare professionals. Intensive Care Med 37:1696–1705 20. Rhodes A, Ferdinande P, Flaatten H, Guidet B, Metnitz PG, Moreno RP 40. Aslakson RA, Curtis JR, Nelson JE (2014) The changing role of palliative (2012) The variability of critical care bed numbers in Europe. Intensive care in the ICU. Crit Care Med 42:2418–2428 Care Med 38:1647–1653 21. Rabin R, de Charro F (2001) EQ-5D: a measure of health status from the EuroQol Group. Ann Med 33:337–343 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Intensive Care Medicine Springer Journals

Outcome in patients perceived as receiving excessive care across different ethical climates: a prospective study in 68 intensive care units in Europe and the USA

Free
11 pages

Loading next page...
 
/lp/springer_journal/outcome-in-patients-perceived-as-receiving-excessive-care-across-6MH9WdI6Qi
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Medicine & Public Health; Intensive / Critical Care Medicine; Anesthesiology; Emergency Medicine; Pneumology/Respiratory System; Pain Medicine; Pediatrics
ISSN
0342-4642
eISSN
1432-1238
D.O.I.
10.1007/s00134-018-5231-8
Publisher site
See Article on Publisher Site

Abstract

Purpose: Whether the quality of the ethical climate in the intensive care unit (ICU) improves the identification of patients receiving excessive care and affects patient outcomes is unknown. Methods: In this prospective observational study, perceptions of excessive care (PECs) by clinicians working in 68 ICUs in Europe and the USA were collected daily during a 28-day period. The quality of the ethical climate in the ICUs was assessed via a validated questionnaire. We compared the combined endpoint (death, not at home or poor quality of life at 1 year) of patients with PECs and the time from PECs until written treatment-limitation decisions ( TLDs) and death across the four climates defined via cluster analysis. Results: Of the 4747 eligible clinicians, 2992 (63%) evaluated the ethical climate in their ICU. Of the 321 and 623 patients not admitted for monitoring only in ICUs with a good (n = 12, 18%) and poor (n = 24, 35%) climate, 36 (11%) and 74 (12%), respectively were identified with PECs by at least two clinicians. Of the 35 and 71 identified patients with an available combined endpoint, 100% (95% CI 90.0–1.00) and 85.9% (75.4–92.0) (P = 0.02) attained that end- point. The risk of death (HR 1.88, 95% CI 1.20–2.92) or receiving a written TLD (HR 2.32, CI 1.11–4.85) in patients with PECs by at least two clinicians was higher in ICUs with a good climate than in those with a poor one. The differ - ences between ICUs with an average climate, with (n = 12, 18%) or without (n = 20, 29%) nursing involvement at the end of life, and ICUs with a poor climate were less obvious but still in favour of the former. Conclusion: Enhancing the quality of the ethical climate in the ICU may improve both the identification of patients receiving excessive care and the decision-making process at the end of life. Keywords: Perceived excessive care, Ethical climate, Decision-making, Interdisciplinary collaboration, Patient outcomes, Treatment-limitation decisions *Correspondence: dominique.benoit@ugent.be Department of Intensive Care Medicine, Ghent University Hospital, Corneel Heymanslaan 10, Ghent, Belgium Full author information is available at the end of the article The full list of investigators of the DISPROPRICUS study group are listed in the Acknowledgements and in the ESM 3 file. 1040 Introduction Take‑home message Life supporting therapy in intensive care units (ICUs) has been increasingly offered to patients with poor long- Enhancing the quality of the ethical climate in the ICU may improve term prognoses [1, 2], including those with advanced, both the identification of patients receiving excessive care and the end-of-life decision making process. end-stage organ dysfunction or a poor functional status [3–5]. While such therapies should not automatically be considered as non-beneficial, they should be provided Study design and center recruitment only to well-informed patients or relatives in accordance This 28-day observational study was conducted in 12 with their preferences and values, and only if treatment European countries and the United States. National coor- intensity remains proportional to the expected outcome dinators and local investigators were recruited from the [6, 7]. Nevertheless, one in three deaths occurs during or Ethics Section of the European Society of Intensive Care shortly after ICU treatment [2], frequently following dis- Medicine, the APPROPRICUS study group [8] and letters proportionate levels of care [8–13]. sent to experts in communication and end-of-life care in An ethically-based clinical decision-making process the ICU. National coordinators were expected to recruit has to rely on both individual perceptions and objective four centers in their country, translate the questionnaires criteria, followed by interdisciplinary discussions that into their own language using the Brislin method [19], enrich the process for the benefit of the patient. However, obtain ethics committee approval and assist the local expressing a perception of excessive care (PEC) to col- investigators in their data collection and data quality leagues, and more specifically to senior ones, necessitates tasks. Local investigators arranged study initiation meet- a safe climate in which clinicians are empowered to speak ings in their ICUs to enhance clinicians’ participation, up and in which they feel that their opinion is valued and recruited patients after having obtained informed con- subsequently integrated into the decision-making pro- sent and recorded data in a dedicated case-report form cess [14]. In addition to enhancing trust and cohesion on the www.DISPR OPRIC US.be website. in a team, such a climate may also reduce uncertainty in decision-makers by favoring intra- and interdisciplinary Data collection instruments and definition of combined transfer of knowledge, experience and values [14]. Sev- endpoint eral studies have already shown that concordant prognos- Country, hospital, ICU and clinician characteristics are tic estimates [15, 16] or perceptions of inappropriate [17] reported in the ESM 2. Hospital and ICU characteris- or futile care [18] by two clinicians may be considerably tics were collected by the local investigators between more predictive about the patient’s short- and long-term March and May 2014. Country-specific health variables outcomes than usually thought. However, whether the were retrieved from a prior publication [20]. In April and quality of the ethical climate prevailing in a unit further May 2014, clinicians in the participating ICUs completed improves the identification of patients receiving excessive questionnaires on personal characteristics, working con- care, and impacts on patient outcomes and written treat- ditions and the ethical climate prevailing in their units ment-limitation decision (TLD), is unknown. using the ethical decision-making climate questionnaire The objectives of the current multicenter study were (EDMCQ) [14]. This questionnaire consists of 35 items to assess whether the quality of the ethical climate in an with four- or five-point Likert scale options; 11 items ICU is associated with the prognostic value of PEC(s) are on end-of-life care practices; 11 on interdisciplinary with regard to patients’ one-year outcomes and with the reflection, collaboration, and communication and 13 on time from PEC(s) until written TLD during ICU stay or leadership skills of senior doctors. The theoretical frame - death. We hypothesized that the better the ethical cli- work and the validation of this instrument can be found mate, the more the PEC(s) would be predictive about in a previous publication [14]. patients’ one-year outcomes and the shorter the time Daily, during the 28  day study period (between May 4 until written TLD or death. and July 4, 2014), the clinicians anonymously completed a questionnaire about their perceptions of dispropor- Methodology tionate care for each of their patients. Disproportionate This study was approved by the ethics committees of all care was defined as care that is no longer consistent with participating centers and the Danish National Health the expected survival or quality of life (either “too much” Authority. Informed consent was required in all countries or “not enough” care), or that is provided against the to collect the one-year outcomes. The protocol, question - patient’s or relatives’ wishes. Questionnaire completion naires and case-report form are available in the electronic required less than 5  min per patient per day, when care supplementary material (ESM 1). was perceived as disproportionate, and less than 2  min 1041 otherwise. ICU mortality and length of stay were col- ANOVA tests where appropriate) for comparing continu- lected in all patients admitted in the ICU; those already ous variables. Results were expressed as number (%) and admitted prior to the study and those newly admitted median (25–75th percentiles), respectively. during the study period. The characteristics reported in the ESM 2 were collected in patients admitted for rea- Differences in patients’ combined endpoint at one year sons other than monitoring only during the study period. across ethical climates Categorization was left at the discretion of the attend- To simplify the analysis only perceptions of excessive ing physician. Written TLDs were ascertained by chart (“too much”) care were taken into account in the current review. study. As PEC by a clinician alone was only moderately Because staying at home with a good quality of life is predictive of the patient’s combined outcome compared highly valued by patients, the combined patient outcome to no PEC across all climates (ESM 2), and previous pub- in this study was defined as dead, not at home or a utility lications have highlighted the importance of concordance score < 0.5 at 1  year. This endpoint was defined during a between two clinicians [15–18], we compared the prob- study meeting with the national coordinators at the Euro- ability of attaining the combined endpoint for patients pean Society of Intensive Care Medicine congress in Bar- with PECs by at least two clinicians between the ethi- celona on September 30th 2014, approximately one year cal climates. For practical reasons, “PECs by at least two prior to data collection. Patients admitted for reasons clinicians” is referred to as “concordant PECs” through- other than monitoring only who were discharged alive, out the manuscript. Differences in combined endpoint or their families, were contacted by telephone or mail in patients without and with concordant PECs between one year after the ICU stay. The interviewer collected and within climates were compared with a Pearson’s Chi vital status, place of residence, and health-related qual- square and a Fisher’s exact test, respectively. ity of life using the EuroQoL-5D questionnaire [21], with conversion of each health state into a utility index (range Differences in time until death and treatment limitation − 0.1584 to 1.000). This questionnaire measures health in decisions across ethical climates five dimensions: mobility, self-care, usual activities, pain/ Time until identification of patients with concordant discomfort, and anxiety/depression. Each dimension has PECs, and from concordant PECs until written TLD three levels: no problems, moderate problems or severe or death were compared using (cause-specific) hazard problems. Therefore, patients can be classified into 1 of ratios, obtained via Cox regression (accounting for com- 243 possible health states, which is converted into the peting risks) [24]. The cause-specific hazard of an event corresponding utility index (range − 0.1584 to 1.000), expresses the instantaneous risk of that event at a given indicating the preference of being in a health status. A time for patients who are still alive in the ICU at that utility index < 0.5 corresponds with severely compro- time and have not previously experienced that event [24]. mised quality-of-life on at least one of the five dimen - To better explore the so-called “self-fulfilling prophecy sions. Although quality-of-life may be preferentially issue” (prognostication influenced by decision-making), evaluated from the patient, for some older patients prox- we compared the risk of death in patients with concord- ies may provide the most reliable information [22]. ant PECs in different decision-making scenarios (doctor– doctor, doctor–nurse, nurse–nurse) between and within Data analysis climates. Ethical climates: factor and cluster analysis Using the clinicians’ answers to the 35 EDMCQ items, Adjustment for case‑mix, hospital and country characteristics the data were first reduced via exploratory and confirma - To adjust for differential case-mix, hospital and country tory factor analysis to seven latent variables, also called characteristics between climates, we used inverse prob- factors [14]. The average score across clinicians for each ability weighting based on propensity scores [25]. Here, factor in a given ICU was used as input for the cluster the propensity score is the probability of being treated analysis at ICU level (ESM 2). Such analyses seek to mini- in one’s own climate, as obtained using a multinomial mize the similarity of ICUs within climates and maximize model based on patient, hospital and country charac- the dissimilarity of ICUs between climates. In particular, teristics. Adjustment based on propensity scores has we used the partitioning around medoids (PAM) algo- the advantage, relative to other adjustment methods, of rithm to classify the different climates into a pre-speci - preventing model extrapolation, when climates are very fied number of clusters. This algorithm was chosen in different in terms of these characteristics [25]. However, view of its robustness to outliers and noise [23]. Pearson’s one concern about adjustment for case-mix is that it may chi square tests were used for comparing categorical eliminate the effects of potential differences in admis - variables between climates and Kruskal–Wallis tests (or sion policy (which affects case-mix) between climates. 1042 Therefore, we considered the unweighted results as our severe underlying comorbidities and with greater use principal results. These are expressed as proportions and of advanced and prolonged life-supporting treatments (cause-specific) hazard ratios (HR) along with 95% confi - in the post-operative setting, compared to the other dence intervals (95% CI). Two-sided P values were con- climates. sidered significant at the 0.05 level. Priority was given to comparisons between the good and the poor ethical cli- Differences in patients’ combined endpoint at one year mates (see results) in order to reduce type I errors. We across ethical climates refer to the ESM 2 for a more detailed methodology. Of the 1761 patients admitted for more than only moni- toring with data concerning time until event available Results (Fig.  1), 74 (4.2%) patients were perceived as receiving Ethical climates excessive care by two clinicians, and 107 (6.1%) by more Of 4747 clinicians working in 68 ICUs in Belgium, Czech than two clinicians, resulting in 36 (11.0%), 50 (7.2%), 21 Republic, Denmark, France, Germany, Greece, Hungary, (18.0%) and 74 (12.0%) patients with concordant PECs Italy, Portugal, United Kingdom, Sweden, the Nether- from the good to the poor climate, respectively. Excessive lands and the United States, 2992 (62.6%) completed the care was perceived by these clinicians as being provided EDMCQ (Fig. 1). against the patients’ or relatives’ wishes in 20 (55.5%), The cluster analysis based on the average scores of 25 (50.0%), 11 (52.4%) and 41 (55.4%) (P = 0.94) of these the seven factors identified during the validation of the patients. EDMCQ [14] yielded four different meaningful, mutually The differences in the patients’ combined outcomes exclusive ethical climates. Visual inspection of the scree across ethical climates are reported in Table 1. The prob - plot (ESM 2) revealed that clustering into three clusters abilities of attaining the combined endpoint in patients would drastically increase the total intra-cluster varia- without concordant PECs was 53.5% (95% CI 46.8–60.2), tion (as opposed to using four clusters), while clustering 59.1% (54.6–63.6), 64.0% (53.1–74.9) and 51.8% (47.3– into five clusters would only minimally decrease the total 56.3) from good to poor climate, respectively (P = 0.057, intra-cluster variation [23]. These climates were denomi - difference between good and poor climate, P = 0.74). nated by experts in intensive care (DB, JD), psychology These probabilities increased to 100% (90.0–100), 95.6% (+) (BV, SV) and ethics (RP) as: good, average with and (84.3–98.9), 94.7% (70.6–99.3) and 85.9% (75.4–92.0) in (−) without nurses’ involvement at end-of-life, and poor patients with concordant PECs (P = 0.047, difference (Fig.  2, ESM 2). According to clinicians working in a between good and poor climate, P = 0.020). good climate, leadership by senior doctors is active and facilitates interdisciplinary reflection and decision-mak - Differences in time until death and treatment limitation ing overall. This climate is also characterized by mutual decisions across ethical climates respect, which is pre-requisite to facilitating interdisci- We found no difference in incidence or in time from plinary reflection and ethical awareness [14]. Within the admission until concordant PECs between the good and (+) average climate, clinicians perceive that senior doctors the poor climates; approximately 11% of the patients empower nurses to share interdisciplinary decision-mak- were identified with concordant PECs after 14  days in ing, mainly at end-of-life. Even though clinicians working both climates (Fig. 3a). (−) in an a verage climate believe that their senior doctors The risk of death in patients with concordant PECs are able to make decisions, they do not find them pro - was statistically significantly higher (HR 1.88, 95%CI moting nurse involvement in decision making at end-of- 1.20–2.92) in the good compared to the poor climate. life. Finally, clinicians working in a poor climate perceive The median time until death in patients with concordant a need for improvement in all of these factors. PECs was 5  days (2–18) vs. 14 (6–34) days (P = 0.008), The ICU, clinician, and patient characteristics for each respectively. The difference between the average climates (−) climate are reported in ESM 2. The average and poor and the poor climate was less important, but still in favor climates were more prevalent in Central and South- of the average climates (Fig.  3c). The risk of death in the ern European countries (P < 0.001); however, 10 of the good climate was higher in patients with PECs by two or 24 (41.7%) ICUs with a poor climate were situated in more doctors than in those with PECs by two or more Western Europe and the United States. The ICU experi - nurses (HR 3.13, 95% 1.19–8.23), with the risk of death in ence of clinicians was similar across climates, however, patients with PECs by at least one nurse and one doctor the number of participating doctors was higher in the being intermediate. There was no evidence of such a dif - (−) average and poor, compared to the other two climates. ference in risk of death in the poor climate (HR 0.74, 95% (−) The average and poor climates were also associated 0.29–1.86) (ESM 2). with a slightly higher number of admitted patients with 1043 15 countries 1 country did not participate 1 country unprepaired Phase I 13 countries, 68 ICUs Phase II Ethical climate 2992 clinicians (63%of 4747) analysis (Fig 2) Phase III 3528 patients included of which 2935 patients got 29136 perceptions by 2562 clinicians 1351 admitted during the study period for 353 admitted prior to the study period monitoring only 1824 patients admitted for more than monitoring only during the study period of which 1558 got17703 perceptions by 2244 clinicians 63 (3.5%) missing time until event Analysis of time until 1761 patients ≥ 2PECs(Fig 3a-b) 266withno perceptions 1580 patients 181patients 74 with exact 2 PEC 1126 with no PEC with ≥ 2 PEC without ≥ 2 PEC 107withmore than 2 PEC 188with 1 PEC Analysis of time from ≥ 2 Phase IV PECs to TLD or death (Fig 3c-f) 355 missing 10 missing 171 with ≥ 2 PECwith known 1225 patients without ≥ 2 PECwith 1 year combined outcome: known1 year combined outcome: Analysis of the prognostic value of ≥ 2 PECs with 92.4%(158/171) 55.6% (681/1225) regard to one year outcomes 450 dead, 231 not at home OR utility < 0.5 146 dead, 12 not at home OR utility < 0.5 Fig. 1 Flow chart. Phase I: Recruitment and data collection of hospital and ICU characteristics, Phase II: Ethical climate data collection, Phase III: Daily perceptions of clinicians and collection of patient characteristics during the 28 days study period, Phase IV: Collection of patients’ one year outcomes. PEC(s) perception(s) of excessive care, TLDs treatment-limitation decisions 1044 Fig. 2 Ethical climates. Factor and cluster analysis were used to obtain mutually exclusive climates. Factor analysis attributes and aggregates the 35-item ethical decision-making climate questionnaire into seven factors for each clinician, which describe different aspects of the ethical decision- making climate as perceived by that clinician. These were subsequently averaged across clinicians to obtain seven factor scores per ICU [14]. A (+) (−) cluster analysis based on these averages scores identified four meaningful ethical climates; good, average with and without involvement of nurses at end-of-life (EOL), and poor. The figure visualizes the average factor scores in clinicians per climate. Larger values indicate better agreement with the climate expressed by the corresponding factor. More detailed information can be found in the ESM 2 Table 1 Differences in patients’ one ‑year outcomes across ethical climates in patients with and without concordant PECs Ethical climate P value overall P value good vs. poor climate (+) (−) Good Average Average Poor Patients without concordant PECs (n= 1225) n = 215 n = 464 n = 75 n =471 Combined endpoint 115 (53.5%) 274 (59.1%) 48 (64.0%) 244 (51.8%) 0.057 0.740 Dead 68 (31.6%) 175 (37.8%) 39 (52.0%) 168 (35.7%) Alive not at home or utility < 0.5 47 (21.9%) 99 (21.3%) 9 (12.0%) 76 (16.1%) Patients with concordant PECs (n= 171) n =35 n =46 n = 19 n = 71 Combined endpoint 35 (100%) 44 (95.6%) 18 (94.7%) 61 (85.9%) 0.047 0.020 Dead 33 (94.3%) 41 (89.1%) 18 (94.7%) 54 (76.0%) Alive not at home or utility < 0.5 2 (5.7%) 3 (6.5%) 0 (0.0%) 7 (9.9%) After weighting to adjust for differential case-mix, hospital and country characteristics, the probability of attaining the combined endpoint in patients without and with concordant PECs was 56, 62, 60 and 55% (P = 0.26, difference between good and poor climate, P = 0.82) and 100, 93.9, 93.5 and 86.2% (P = 0.042, difference between the good and the poor climate, P = 0.017) from the good to the poor climate, respectively Patients with concordant PECs had a higher chance Adjustment based on propensity scores of receiving a written TLD in the good compared to the After weighting to adjust for differential case-mix, hos - poor climate (cause-specific HR 2.32, 95%CI 1.11–4.85) pital and country characteristics, the probability of (Fig. 3e). attaining the combined endpoint in patients without 1045 Fig. 3 a–f Competing risk analyses of time from admission until concordant perceptions of excessive care (PECs) by at least two different clinicians, written treatment-limitation-decision ( TLD) and death before and after weighting for country, hospital and patients characteristics using propensity scores. The primary endpoint (dead, not at home or a utility < 0.5 according the EuroQoL-5D questionnaire [21] at one year) is visualized separately in c, d. The sudden increase at day 365 represents the proportion of patients alive with a utility < 0.5 or not living at home. The incidence of the primary endpoint differs from the text because drop-outs are taken into account in competing risk analyses. The results are expressed as (cause- specific) hazard ratios (HR) together with 95% confidence intervals (CI). To avoid type I errors, we gave priority to comparisons between the most extreme (good and poor) climates concordant PECs was 55.8% (48.2–63.1), 62.1% (56.5– 100% (90.0–100), 93.9% (74.3–98.8), 93.5% (64.2–99.1) 67.4), 60.2% (47.4–71.7) and 54.8% (49.4–60.1) from and 86.2% (72.0–93.8), respectively (P = 0.042, differ - good to poor climate, respectively (P = 0.26, difference ence between the good and the poor climate, P = 0.017). between good and poor climate, P = 0.82). These prob - The risk of death in patients with concordant PECs abilities increased in patients with concordant PECs to also remained higher in the good vs. the poor climate 1046 Fig. 3 continued (HR 1.79, 95%CI 1.07–2.98) (Fig.  3d). The median time We preferred to focus on the intuitive-heuristic more until death was 5 (2–18) and 14 (7–30) days (P = 0.026), than the analytic-deductive part of the complex ethical respectively. The risk of death in the good climate decision-making process [26, 27], by asking clinicians remained higher in patients with PECs by two or more whether they felt that the care provided to their patient doctors than in those with PECs by two or more nurses on a specific day was consistent with the expected out - (HR 3.58, 95% 1.42–9.02), with the risk of death in come in terms of survival and quality of life, and whether patients with PECs by at least one nurse and one doctor this amount of care was in line with the patient’s or rela- remaining intermediate. There was no evidence of such a tives’ wishes. We also didn’t focus on futile care, such as difference in risk of death in the poor climate (HR 1.58, in the studies of Neville et al. [18], because this terminol- 95% 0.45–5.55) (ESM 2). ogy presupposes a high degree of certainty concerning However, we no longer found evidence of a differ - the final fatal prognosis, whereas nowadays technologi - ence in time until TLD between the good and the poor cal innovation frequently excludes patients’ spontane- climates (cause-specific HR 1.76, 95%CI 0.73–3.92) ous death in ICU [6, 7]. By doing so, we acknowledged (Fig. 3f ). uncertainty [26] (benefit vs. harm) and patient and fam - ily autonomy, as an integral part of the complex ethical Discussion decision-making process at the bedside [28]. Neverthe- In this large, multicenter, prospective, ICU study, we less, PEC was highly predictive about patients’ one-year found that concordant PECs by at least two clinicians outcomes, more specifically when expressed by two or were far more predictive about the primary composite more than two clinicians. endpoint of death, not living at home, or having poor Concordant PECs by at least two different clini - quality of life one year after ICU admission, compared cians were more predictive about the combined end- to absence of PEC. We found evidence of a difference in point in the good compared to the poor ethical climate one-year outcomes, time until death and written TLD (P = 0.028). Patients with concordant PECs also had in patients with concordant PECs across the four ethical a higher risk of death and of receiving a written TLD climates identified by our questionnaire. The evidence of in the good compared to the poor climate. The differ - a difference in time until written TLD disappeared after ence in endpoints between the average and the poor cli- adjusting for differential case-mix, hospital and country mates was less obvious, but still in favor of the former characteristics. compared to the latter, thus objectively validating our In contrast to the study by Detsky et  al. [16], clini- EDMCQ instrument [14]. Unfortunately, we can neither cians in our study were not explicitly expected to pro- exclude nor confirm self-fulfilling prophecy in the good vide prognostic estimates about the patients’ outcomes. climate. However, it is of note that it took about 14 days 1047 to identify all patients with concordant PECs in both cli- collaboration [6, 8, 14, 32, 38, 39], and early involvement mates and, for half of these patients, another 5 days to die of palliative care [30, 37, 40]. Our EDMCQ instrument in the good vs. 14  days in the poor climate (P = 0.002). may be used for that purpose [14, 32]. In line with the results of the EDMCQ, this suggests Our study has several limitations. First, the participat- that the decision to forgo life sustaining treatment in the ing ICUs were not selected at random, which may have good climate was not premature, and once excessive care affected the external validity of our results. Second, inclu - was perceived by at least two clinicians, it occurred in a sion of patients was left at the discretion of the attending (−) timely fashion. Furthermore, in a sub-analysis, we found doctor. However, except in the average climate (ESM no difference in risk of death between patients with con - 2), we found no evidence of a difference in ICU mor - cordant PECs by different professionals in the poor cli - tality rates or length of stay in the subgroup of patients mate, as opposed to the good climate. This indicates that admitted for monitoring only across climates, indicating identification of patients with excessive care by doctors that the attending doctors included patients in a similar in the poor climate was not followed by active decision- way. We further minimized confounding bias by account- making. In addition to, respectively, increasing the risk ing for differences in case-mix, using inverse probability of prolonged suffering and complicated grief in patients weighting based on propensity scores. Third, we did not and relatives [29, 30], decision-paralysis as a strategy to use classical severity-of-illness scores in our analysis. cope with prognostic uncertainty [8, 12, 31] may also However, in line with our primary objective, we preferred induce moral distress and increase intention to leave in to include short- and long-term prognostic factors [4, 5] clinicians [6, 32–34]; a fact that is even more pertinent that are commonly used by clinicians during decision- considering the high number of concordant PEC records making, rather than classical severity-of-illness scores perceived as violating the patient’s or relatives’ wishes which have never been validated for predicting long- in this study. After weighting for the specific case-mix term outcomes. Fourth, one has to keep in mind that the within a hospital and country, only the risk of receiving incidence of patients with concordant PECs is probably a written TLD in the good compared to the poor climate underestimated, as patients admitted prior to the study was no longer significantly different. This may suggest period and those who remained in ICU for longer than that the quality of the ethical climate in an ICU is impor- the study period (and were expected to reach more cli- tant in identifying patients receiving excessive care and in nician concordance with time) were excluded from the subsequently triggering the decision-making process at analysis. Finally, although the ICU experience of clini- end-of-life, whereas formalizing that process via a writ- cians was similar, we cannot exclude that the lower num- ten TLD seems more case-mix and culture dependent. ber of participating doctors in the good compared to the This is in line with previous studies showing a huge vari - poor climate may have biased our results in favor of the ability in written TLDs between countries and ICUs [35]. latter, concealing even larger differences between the two. The probability of dying or surviving with a poor qual - ity of life at one year in patients without concordant PECs Conclusion was 53.5, 59.1, 64.0 and 51.8% from good to poor climate, Our results suggest that improving the quality of the ethi- respectively, largely exceeding that of many malignancies cal climate in ICU may favor the identification of patients [36]. Therefore, in line with the definition of dispropor - receiving excessive care and the subsequent decision- tionate care [6, 8, 9], clinicians did not find poor prog - making process at end-of-life. This may benefit the qual - nosis sufficient by itself to lead to a PEC. Concordant ity of the dying process in ICUs. PECs by at least two clinicians increased the probability Electronic supplementary material of reaching the combined endpoint to 100% in the good, The online version of this article (https ://doi.org/10.1007/s0013 4-018-5231-8) (+) (−) 95.6% in the average and 94.7% in the average cli- contains supplementary material, which is available to authorized users. mate, compared to 85.9% in the poor. Despite the poor prognosis we found a relatively low incidence of writ- Author details ten TLDs within the 14  days in these patients; ranging Department of Intensive Care Medicine, Ghent University Hospital, Corneel Heymanslaan 10, Ghent, Belgium. Department of Intensive Care Medicine, from 20% in the poor to only 35% in the good and about Vejle Hospital, Vejle, Denmark. Institute of Regional Research, University 45% in average climates (P = 0.011). Although caution 4 of Southern Denmark, Odense C, Denmark. Department of Anaesthesiology in interpreting this result is required due to small sam- and Intensive Care, Sahlgrenska University Hospital, Gothenburg, Sweden. 5 6 King’s College Hospital, London, UK. Department of Medical Oncology, ple size, these probabilities highlight the urgent need for University of Groningen, University Medical Center Groningen, Groningen, improving advance-care planning before ICU admission 7 The Netherlands. Hôpital Saint-Louis and University, Paris-7, Paris, France. [37], as well as triage and decision-making at end-of-life Department of Anesthesiology and Intensive Care, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Prague, in ICU. This should more specifically be achieved via Czech Republic. Department of Anesthesia, Critical Care, and Pain Medicine, ethical climates that favor interdisciplinary reflection and Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, 1048 MA, USA. Service des soins intensifs et urgences oncologiques, Institut Jules Gaia (Paula Fernandes, Ana Isabel Paixão), Instituto Português de Oncologia, Bordet, ULB, Brussels, Belgium. SCDU Anestesia e Rianimazione, Azienda Porto (Filomena Faria), Sweden: Sahlgrenska University Hospital, Gothenburg and Ospedaliero Universitaria, “Maggiore della Carità”, Novara, Italy. Semmel- (Johan A. Malmgren), Sahlgrenska University Hospital/Östra, Gothenburg weis University Budapest, Budapest, Hungary. Intensive Care Department, (Bertil Andersson), Skåne University Hospital, Malmö (Eva Åkerman), Karolinska Hospital S.António, Porto, Portugal. Tettnang Hospital, Tettnang, Germany. University Hospital, Karolinska (Andreas Hvarfner), The Hospital of Norrköping, Department of Psycho-analysis and Clinical Consulting, Faculty of Psychol- Norrköping (Robert Svensson), United Kingdom: King’s College Hospital, Lon- ogy and Educational Sciences, Ghent University, Ghent, Belgium. Depart- don ( Victoria Metaxa), USA: Beth Israel Deaconess Medical Center and Harvard ment of Intensive Care Medicine, Erasmus MC University Medical Center Medical School, Boston MA (Daniel Talmor, Ariel Mueller, Valerie Banner-Good- Rotterdam, Rotterdam, The Netherlands. Department of Applied Mathemat- speed), Henry Mayo Newhall Memorial Hospital, Valencia, CA (Dee Rickett), ics, Computer Science and Statistics, Faculty of Sciences, Ghent University, Mayo Clinic, Rochester, MN (Michael E. Wilson, Richard Hinds). Ghent, Belgium. London School of Hygiene and Tropical Medicine, London, UK. Department of Geriatric Medicine, Ghent University Hospital, Ghent, Author Contributions Belgium. Study concept and design: DDB, BVB, RDP. Design of the questionnaire: DDB, HIJ, JM, SV, EJOK, JD, BVB, EA, RDP. Coordination of the translation of the Acknowledgements questionnaire: HIJ, JM, VM, AKR, MD, KR, DT, APM, LC, LZ, PM, AM. Acquisition of This study was supported by a European Society of Intensive Care Medi- data: DDB, HIJ, JM, VM, AKR, MD, KR, DT, APM, LC, LZ, PM, AM, BG. Analysis and cine/European Critical Care Research Network clinical research award and a interpretation of data: DDB, SV, SV, SV, BVB, EA, RDP. Drafting of the manuscript: Fonds voor Wetenschappelijk Onderzoek senior clinical investigators grant DDB, VM, DT, SV, BVB, EA, RDP. Critical revision of the manuscript for important (1800513N) obtained in 2012 by DB. We are grateful to Ariella Van Sompel for intellectual content: DDB, HIJ, JM, VM, AKR, MD, KR, DT, APM, LC, LZ, PM, AM, SV, having performed the factor and cluster analysis together with VDB and RP EJOK, JD, SV, SV, BG, BVB, EA, RDP. Statistical expertise: SV, SV. Obtained funding: (under supervision of SVH and SVS) and Jolien Roels for having performed the DDB, JD. Administrative, technical, or material support: DDB, JD, BG. Steering data cleaning and the univariate analysis (under supervision of DB, SVB and committee: DDB, SV, EJOK, JD, SV, BG, BVB, EA, RDP. SVS). Participating centers and local investigators: Belgium: University Hospital, Vrije Universiteit Brussel, Brussels (Herbert Spapen, Marie-Claire Van Malderen, Compliance with ethical standards Godelieve Opdenacker), Leuven University Hospital, Leuven (Geert Meyfroidt, Dieter Mesotten, Joost Wauters, Marie Van Laer and Alexander Wilmer, Joost Conflicts of interest Wauters, Helga Ceunen), ZNA Stuivenberg, Antwerpen (Inneke E De Laet, DB reports grants from Gilead, Astellas, Fisher-Paykel, Baxter, Alexion and Anita Jans), Ghent University Hospital, Gent (Dominique Benoit, Sandra Oeyen, Fresenius Kabi outside the submitted work. KR reports honoraria from Alexion, Ingrid Herck, Stephanie Bracke, Charlotte Clauwaert), Institut Jules Bordet, outside the submitted work. MD reports grant from MSD and Jazz Pharma, Bruxelles (Meert Anne-Pascale, Leclercq Nathalie), CHU-Brugmann, Bruxelles personal fees from Astellas and Bristol-Myers Squibb, and non-financial sup - (Devriendt Jacques), CHU Saint Pierre, Bruxelles (Dechamps Philippe), Czech port from Astellas, Bristol-Myers Squibb, Astute Medical, and Sanofi Aventis. Republic: Liberec District Hospital, Liberec (Ivana Zykova), Masaryk University, EA reports grants and personal fees from Gilead, Alexion, MSD, Cubist and Brno and University Hospital, Brno (Jan Malaska), Third Faculty of Medicine, personal fees from Baxter, outside the submitted work. All other authors have Charles University, Prague (Matous Schmidt), Hospital and Polyclinic Havirov, no conflict of interest to report. Havirov (Igor Satinsky), Institute for Experimental and Clinical Medicine, Prague (Eva Kieslichova), 3rd Medical Department, First Faculty of Medicine, Charles Open Access University in Prague and General University Hospital, Prague (Jarmila Krizova), This article is distributed under the terms of the Creative Commons Attribu- Karlovy Vary District Hospital, Karlovy Vary (Robert Janda), Pardubice District tion-NonCommercial 4.0 International License (http://creativecommons.org/ Hospital, Pardubice (Magdalena Fortova, Jiri Matyas), First Faculty of Medicine, licenses/by-nc/4.0/), which permits any noncommercial use, distribution, and Charles University and General University Hospital, Prague (Katerina Rusinova, reproduction in any medium, provided you give appropriate credit to the Ondrej Kopecky), Denmark: Herning Hospital, Herning (Christian Alves Køhler original author(s) and the source, provide a link to the Creative Commons Pedersen), Kolding Hospital, Kolding (Stine Hebsgaard), Vejle Hospital, Vejle license, and indicate if changes were made. (Rikke Frank Aagaard Johnsen), Holbæk Hospital, Holbæk ( Tina Charlotte Bitsch Hansen), France: Saint-Etienne University Hospital and Jacques Lisfranc Medical School, Saint-Etienne (Michael Darmon), Saint-Louis University Received: 10 February 2018 Accepted: 14 May 2018 Hospital, APHP, Université Paris-7, Paris (Danielle Reuter, Elie Azoulay), Institut Published online: 28 May 2018 Paoli Calmette, Marseilles (Djamel Mokart), Montfermeil Hospital, Montfermeil (François Vincent), Germany: University Hospital Jena, Jena (Christiane S. Har- tog), Viersen General Hospital, Viersen (Peter Gretenkort), Tettnang Hospital, Tettnang (Andrej Michalsen), Greece: Agia Olga Hospital, Athens (Aikaterini References Kounougeri), Evangelismos Hospital, Athens (Serafim Nanas), Agios Pavlos 1. Wunsch H, Linde-Zwirble WT, Harrison DA, Barnato AE, Rowan KM, Angus Hospital, Thessaloniki (Despina Papachristou), AHEPA University Hospital, DC (2009) Use of intensive care services during terminal hospitalizations Thessaloniki, (Ioanna Soultati), G.Gennimatas Hospital, Thessaloniki (Dimitrios in England and the United States. Am J Respir Crit Care Med 180:875–880 Lathyris), Hippokratio General Hospital, Thessaloniki (Marili Pasakiotou), Papa- 2. Teno JM, Gozalo PL, Bynum JPW, Leland NE, Miller SC, Morden NE, Scupp georgiou General Hospital, Thessaloniki (Marina Oikonomou), Hungary: Sem- T, Goodman DC, Mor V (2013) Change in end-of-life care for medicare melweis University Budapest, Budapest (Gábor Élő, Orsolya Szűcs), Kaposi Mór beneficiaries. Site of death, place of care, and health care transtitions in Teaching Hospital, Kaposvár University, Kaposvár (János Fogas), St. Stephen 2001 2005, and 2009. JAMA 309:470–477 and St. Leslie Metropolitan Hospital, Budapest (Ilona Bobek), Italy: Azienda 3. Ho TH, Barbera L, Saskin R, Lu H, Neville BA, Earle CC (2011) Trends in Ospedaliero Universitaria, “Maggiore della Carità”, Novara, and Department of aggressiveness of end of life cancer care in the universal health care Translational Medicine, Università del Piemonte Orientale, Novara (Francesco system of Ontario, Canada. J Clin Oncol 29:1587–1591 Della Corte, Carlo Olivieri, Rosanna Vaschetto, Laura Cancelliere), Ospedale 4. Wang CY, Calfee CS, Paul DW, Janz DR, May AK, Zhuo H, Bernard GR, Civile San Salvatore, and Department of Life, Health and Environmental Matthay MA, Ware LB, Neudoerer K ff angelaris K (2014) One-year mortality Sciences (MeSVA), University of L’Aquila and Department of Emergency, San and predictors of death among hospital survivors of acute respiratory Salvatore Hospital, L’Aquila (Franco Marinangeli, Tullio Pozone, Alessandra distress syndrome. Intensive Care Med 40:388–396 Ciccozzi), The Netherlands: Canisius Wilhelmina Ziekenhuis, Nijmegen (A. 5. Garland A, Olafson K, Ramsey CD, Yogendran M, Fransoo R (2014) Distinct Schouten, Monique Bruns), Medical Center Leeuwarden, Leeuwarden (Rik T. determinants of long-term and short-term survival in critical illness. Gerritsen, Matty Koopmans), Erasmus University Hospital of Rotterdam (Erwin Intensive Care Med 40:1097–1105 Kompanje, Ditty van Duijn), University of Groningen and University Medical 6. Kompanje EJ, Piers RD, Benoit DD (2013) Causes and consequences of Center Groningen, Groningen (Jan G. Zijlstra, Anne KL Reyners), Wilhelmina disproportionate care in intensive care medicine. Curr Opin Crit Care Ziekenhuis Assen, Assen (Johan G. Lutisan), Portugal: Hospital S.António, Porto 19:630–635 (Raquel Monte, José António Pinho, Pedro Pimenta), CHVNG, Vila Nova de 1049 7. Bosslet GT, Pope TM, Rubenfeld GD, Lo B, Truog RD, Rushton CH, Curtis 22. Oeyen SG, Vandijck D, Benoit DD, Annemans L, Decruyenaere J (2010) JR, ford DW, Osborne M, Misak C, Au DH, Azoulay E, Brody B, Fahy BG, Hall Quality of life after intensive care: a systematic review of the literature. JB, Kesecioglu J, Kon AA, Lindell KO, White DB, on behalf of the American Crit Care Med 38:386–400 Thoracic Society ad hoc committee on Futile and Potentially Inappropri- 23. Everitt BS, Landau S, Leese M, Stahl D (2011) Cluster analyis, 5th edition. ate Treatment (2015) An official ATS/AACN/ACCP/ESICM/SCCM policy Wiley. ISBN: 978-0-470-74991-3 statement: responding to requests for potentially inappropriate treat- 24. Putter H, Fiocco M, Geskus RB (2007) Tutorial in biostatistics: competing ments in intensive care units. Am J Respir Crit Care Med 191:1318–1330 risks and multi-state models. Statist Med 26:2389–2430 8. Piers R, Azoulay E, Ricou B, Dekeyser Ganz F, Decruyenaere J, Max A, 25. Rubin DB (1997) Estimating causal effects from large data sets using Michalsen A, Azevedo Maia P, Owczuk R, Rubulotta F, Depuydt P, Meert propensity scores. Ann Intern Med 127:757–763 AP, Reyners A, Aquilina A, Bekaert M, Van den Noorgate N, Schrauwen W, 26. Lighthall GK, Vazquez-Guillamet C (2015) Understanding decision making Benoit D (2011) APPROPRICUS study group of the ESICM. Perception of in critical care. Clin Med Res 13:156–168 appropriateness of care among European and Israeli intensive care unit 27. Zollo L, Pellegrini MM, Ciappei C (2017) What sparks ethical decision mak- nurses and doctors. JAMA 306:2694–2703 ing? The interplay between moral intuition and moral reasoning: lessons 9. Piers R, Azoulay E, Ricou B, Dekeyser Ganz F, Max A, Michalsen A, from the scholastic doctrine. J Bus Ethics 145:681–700 Azevedo Maia P, Owczuk R, Rubulotta F, Meert AP, Reyners A, Decruy- 28. Luce JM, White DB (2009) A history of ethics and law in the intensive care enaere J, Benoit DD (2014) Inappropriate care in European ICUs. unit. Crit Care Clin 25:221–237 Confronting views from nurses and junior and senior physicians. Chest 29. Wright AA, Zhang B, Ray A, Mack JW, Trice E, Balboni T, Michell SL, Jackson 146:267–275 VA, Block SD, Maciejewski PK, Prigerson HG (2008) Associations between 10. Anstey MH, Adams JL, McGlynn EA (2015) Perceptions of the appropriate- end-of-life discussions, patient mental health, medical care near death, ness of care in California adult intensive care units. Crit Care 19:51 and caregiver bereavement adjustment. JAMA 300:1665–1673 11. Vincent JL (1999) Forgoing life support in western European inten- 30. Curtis JR, Vincent JL (2010) Ethics and end-of-life care for adults in the sive care units: the results of an ethical questionnaires. Crit Care Med intensive care unit. The Lancet 376:1347–1353 27:1626–1633 31. Kimbell B, Murray SA, Macpherson S, Boyd K (2016) Embracing inherent 12. Palda VA, Bowman KW, Mclean RF, Chapman MG (2005) “Futile” care: do uncertainty in advanced illness. BMJ 354:i3802 we provide it? Why? A semistructured, Canada-wide survey of intensive 32. Dzeng E, Curtis JR (2018) Understanding ethical climate, moral distress, care unit doctors and nurses. J Crit Care 20:207–213 and burnout: a novel tool and a conceptual framework. BMJ Qual Saf:1–5. 13. Giannini A, Consonni D (2006) Physicians’ perceptions and attitudes https ://doi.org/10.1136/bmjqs -2018-00790 5 (Epub ahead of print) regarding inappropriate admissions and resource allocation in the inten- 33. Hamric AB, Blackhall LJ (2007) Nurse-physician perspective on the care of sive care setting. Br J Anaesth 96:57–62 dying patients in intensive care units: collaboration, moral distress, and 14. Vandenbulcke B, Piers R, Jenssen HI, Malmgrem J, Metaxa V, Reyners AK, ethical climate. Crit Care Med 35(2):422–429 Darmon M, Rusinova K, Talmor D, Meert AP, Cancellierre L, Zubek L, Maia 34. Schwarzkopf D, Rüddel H, Thomas-Rüddel DO, Felfe J, Poidinger B, P, Michalsen A, Decruyenaere J, Kompanje EJO, Azoulay E, Vlerick P, Van Matthäus-Krämer CT, Hartog CS, Bloos F (2017) Perceived nonbeneficial den Sompel A, Vansteelandt S, Vanheule S, Benoit DD (2018) On behalf treatment of patients, burnout, and intention to leave among ICU nurses the DISPROPRICUS study group of the Ethics Section of the ESICM. Ethical and junior and senior physicians. Crit Care Med 45:e265–e273 decision-making climate in the ICU: theoretical framework and validation 35. Mark NM, Rayner SG, Lee NJ, Curtis JR (2015) Global variability in with- of self-assessment tool. BMJ Qual Saf. https ://doi.org/10.1136/bmjqs holding and withdrawal of life-sustaining treatment in the intensive care -2017-00739 0 ([Epub ahead of print]) unit: a systematic review. Intensive Care Med 41:1572–1585 15. Meadow W, Pohlman A, Reynolds D, Rand L, Correia C, Christoph E, Hall 36. De Angelis R, Sant M, Coleman MP, Francisci S, Baili P, Pierannunzio D, J (2014) Power and limitations of daily prognostications of death in the Trama A, Visser O, Brenner H, Ardanaz E, Bielska-Lasota M, Engholm G, medical ICU for outcomes in the following 6 months. Crit Care Med Nennecke A, Siesling S, Berrino F, Capocaccia R, Eurocare-5 Working 42:2387–2392 Groups (2014) Cancer survival in Europe 1999–2007 by country and 16. Detsky ME, Harhay MO, Bayard DF, Delman AM, Buehler AE, Kent SA, age: results of the Eurocare-5. A population-based study. Lancet Oncol Cuiffetelli IV, Cooney E, Gabler NB, Ratcliffe SJ, Mikkelsen ME, Halpern SD 15:23–34 (2017) Discriminative accuracy of physician and nurse predictions for 37. Khandelwal N, Kross EK, Engelberg RA, Coe NB, Long AC, Curtis JR (2015) survival and functional outcomes 6 months after ICU admission. JAMA Estimating the effect of palliative care interventions and advance care 317:2187–2195 planning on the ICU utilization: a systematic review. Crit Care Med 17. Singal RK, Sibbald R, Morgan B, Quinlan M, Parry N, Radford M, Martin CM 43:1102–1111 (2014) A prospective determination of the incidence of perceived inap- 38. Puntillo KA, McAdam JL (2006) Communication between physicians propriate care in critically ill patients. Can Respir J 21:165–170 and nurses as a target for improving end-of-life care in intensive care 18. Neville TH, Wiley JF, Yamamoto MC, Flitcraft M, Anderson B, Curtis JR unit: challenges and opportunities for moving forward. Crit Care Med (2015) Concordance of nurses and physicians on whether patients are 34(Suppl.):S332–S340 receiving futile treatment. Am J Crit Care 24:403–411 39. Jensen HI, Ammentrop J, Erlandsen M, Ording H (2011) Witholding or 19. Brislin RW (1976) Translation: application and research. Gardner Press Inc, withdrawing therapy in intensive care units: an analysis of collaboration New York among healthcare professionals. Intensive Care Med 37:1696–1705 20. Rhodes A, Ferdinande P, Flaatten H, Guidet B, Metnitz PG, Moreno RP 40. Aslakson RA, Curtis JR, Nelson JE (2014) The changing role of palliative (2012) The variability of critical care bed numbers in Europe. Intensive care in the ICU. Crit Care Med 42:2418–2428 Care Med 38:1647–1653 21. Rabin R, de Charro F (2001) EQ-5D: a measure of health status from the EuroQol Group. Ann Med 33:337–343

Journal

Intensive Care MedicineSpringer Journals

Published: May 28, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off