End-of-life Practices in 2019 vs. 2014

Background and Rationale Recently published ethical practice guidelines focus on shared decision-making, patient-centred care, family-centred care, and prevention of patient distress after withdrawal of life-sustaining treatments (LSTs). Furthermore, several, recent randomized controlled trials (RCTs) have provided a robust evidence-base for complex and resource-demanding interventions such as advance care planning (ACP). ACP is also recommended by authors of major opinion articles. In addition, new laws, governmental statements, and educational activities across Europe support an autonomy-oriented end-of-life care.

In 2014, the authors conducted a survey of key opinion leaders on ethical resuscitation practices in 31 European Countries. The authors administered a comprehensive questionnaire to 1-2 "experts" from each country; subjectivity-related bias could not be excluded; however, the questionnaire was actually administered twice over a 6-month period to all participants, in order to confirm results' reproducibility. The questionnaire spanned across the following 4 domains: A: ethical practices (41 questions); B: access to best available care (39 questions); C: death diagnosis and organ donation (22 questions); and D: emergency care organization (40 questions). Accordingly, a 142-point scoring system of the responses of the participants was developed. Country-specific scores varied widely [e.g. score range of 1-41 for the ethical practices (domain A), and of 9-32 for emergency care organization (domain D)]. We also found a significant association between domain A and domain D scores (r2 = 0.42, P < 0.001).

The results of the 2014 survey highlighted variability across European countries in their approach to the ethics of resuscitation/end-of-life care. Results also indicated the presence of substantial need for improvements in all the aforementioned domains of practice and emergency care organization On the other hand, such evolution should be substantially augmented and accelerated by the above-described combination of new guidelines, RCT-based support of ACP, legislation / governmental policies, and educational activities.

With this study the authors undertake a methodologically improved version of the 2014 survey, in order to test the following hypotheses: 1) compared to 2014, there may be significant improvements in overall domain A to D scores for 2019, reflecting improved quality of ethical practice in the field of resuscitation/end-of-life care; 2) such progress, may be more marked in countries with "low" (i.e. below-average) domain A to D scores for 2014.

METHODS Ethics Committee Approval The study protocol will be submitted for approval to the Regional Ethics and Scientific Committee of Evaggelismos General Hospital of Athens, the Athens Eye Clinic, and the Athens Polyclinic.

An eventually obtained approval will be used to support the conduct of the survey study in at least 31 European countries, which are listed as follows: Austria, Belgium, Bosnia & Herzegovina, Croatia, Cyprus, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Italy, Luxembourg, Malta, Norway, Poland, Portugal, Republic of Northern Macedonia, Romania, Russian Federation, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, The Netherlands, Turkey, and the United Kingdom. Respondents from additional European countries (e.g. Ireland) may also be invited to participate.

The informed consent procedure

The following text for potential study participants is proposed (text will be presented in informed consent format):

According to the Helsinki declaration], participation in research has the essential prerequisite of free and voluntary informed consent of the participant.

According to Regulation 679/2016 {or General Data Protection regulation (GDPR) of the European Parliament and of the Council}, ""consent" of the data subject means any freely given, specific, informed and unambiguous indication of the data subject's wishes by which he or she, by a statement or by a clear affirmative action, signifies agreement to the processing of personal data relating to him or her." Participation in the current survey study means that the participant accepts to share his/her knowledge on ethical practices (domain A), on access to best-quality care (domain B), on death diagnosis and organ donation (domain C), and on emergency care organization (domain D) in their country.

According to GDPR, "pseudonymization means the processing of personal data in such a manner that the personal data can no longer be attributed to a specific data subject without the use of additional information, provided that such additional information is kept separately and is subject to technical and organizational measures to ensure that the personal data are not attributed to an identified or identifiable natural person." A set of 50 consecutive random numbers (range, 50-99) will be generated using "Research Randomizer (www.randomizer.org). Each one of the aforementioned numbers will be assigned to a European country (country code) drawn according to alphabetic order from the list of participating European countries. Next to the 2 digits of the country code, a third digit corresponding to the participant's temporal order of participation relative to other participants from the same country; participant's temporal order will be determined according to the exact time of completion of the study questionnaire. This will form the participant's personal, 3-digit, code.

Data on ethical practices and emergency care organization will be stored next to the participant's personal code in Microsoft Excel format. This electronic Masterfile will be stored onto the password-protected hard disk of the principal investigator's personal computer. In addition, next to the participant's personal code, the investigators will enter the following personal data 1) name; professional qualification(s) (e.g. physician, nurse, paramedic, other); 2) highest academic or professional rank (e.g. professor, lead clinician, lead nurse, etc.); 3) professional occupation-more than one possible (e.g. academic teaching, academic research; hospital emergency department, intensive care unit, emergency medical service, etc.); and 4) years of professional experience as healthcare provider, academic, or both. Subsequently, personal codes and names will be copied in a second Excel file, the Decoding file. Finally, all names will be removed from the Masterfile, whereas the personal, 3-digit codes will be retained. Consequently, the Masterfile will fulfill the criterion of "pseudonymization." The Masterfile data will be used for statistical analyses aimed at testing the study hypotheses. In this context, and within the next 3-4 months, the authors will resend to the participants the data corresponding to their responses in Excel format so that they may be able to confirm data accuracy. In addition, the authors will also send to participants the data corresponding to other respondents from the same country so that the participant may be able to state whether he/she agrees with others' responses. In the event of disagreement(s) the authors may request that a participant attempts to reach consensus with the other respondents from his/her country. However, participants should note that the reaching of such consensus should not in anyway be considered as "mandatory".

According to the above, the authors request the participant's free and voluntary consent for

1. Collection and storing of the above-mentioned data under password protection
2. Inclusion of the above-mentioned data (at least in part) in an Acknowledgment section of a published article reporting on the results of the survey study

As essential prerequisites for consent validity we ask you to respond to the following questions:

A) Has the participant read / understood the above-presented information? B) Has the participant had adequate time to consider granting their consent for the use of their personal data? C) If the participant asked any questions about any aspect of the survey study, did they receive satisfactory replies from the study investigator? D) Does the participant understand that he/she may withdraw their consent at any time until the publication of the survey study results? E) Does the participant understand that after publication of the study results and dissemination of their personal data, it may become not possible to reverse such dissemination? F) Does the participant agree with the aforementioned procedure of pseudonymization and subsequent dissemination of their personal data?

Methods Potential study participants from the European countries will be contacted via e-mail and invited for an investigators meeting during the 2019 Resuscitation Congress, September 19-21, Ljubljana, Slovenia.

Participant inclusion criteria are presented in the relevant section To reduce the risk of subjectivity-related bias, there will be a minimum of 3 and a maximum of 5 participants from each one of the aforementioned European countries; responses will not be excluded from the analyses if originating from less than 3 participants. Following obtainment of consent, participants will be able to electronically access the questionnaire through Typeform (www.typeform.com). As an example, the 2014 questionnaire is still accessible at https://erc.typeform.com/to/dZjrOQ. The study questionnaire is summarized in Table 1.

As in the authors' 2014 survey study, respondents will have to choose either among 4 options, i.e. never, sometimes, usually and always or between no and yes. Respondents will also be asked to comment on domain subsections. Subsequently, responses of never/sometimes and usually/always will be respectively grouped as no and yes, because never/sometimes does not and usually/always does reflect common/everyday practice. All data will be entered in an Excel Masterfile. Participants from each country will be asked to reconfirm their answers after a period of approximately 3-4 months (procedure detailed above); they will also be asked whether they agree or disagree with answers provided by other participants of the same country. In cases of disagreement, the authors will encourage resolution through consensus. For data analysis, the authors will use a dichotomous quantitizing approach by grading a positive response with 1 and a negative response with 0.

As stated above, the authors will encourage consensus-based resolution of discrepancies among respondents from each country. Full details of each case of discrepant responses will be recorded. Subsequently, for each survey question and country, mean response grades will be calculated and used in the data analyses. In the absence of discrepant responses, or following a consensus, response grades will either have the value of 1 or 0. In cases of failure to reach consensus, response grades will have non-integer, positive values below the value of 1.

Study outcomes For the 2019 survey, Domain A, B, C, and D subcomponent and total scores will by calculated by summing up the corresponding response grades; this specific, summation procedure will not take into account the grades of responses to "new" questions included only in the 2019 questionnaire (Table 1).

The primary outcome will be the presence / absence of statistically significant differences between 2014 and 2019 in Domain A to D total scores of all participating countries.

The secondary outcome will be the presence / absence of statistically significant differences between 2014 and 2019 in Domain A to D total scores of countries with "low" Domain total scores in 2014. The latter "low" 2014 scores will be defined as Domain total scores lower than the corresponding, overall mean score values of 2014; more specifically, "low" 2014 total scores for Domain A, B, C, and D will be those not exceeding the values of 18, 7, 12, and 23, respectively.

According to the above, "high" (or above-average) 2014 total scores for Domains A, B, C, and D will be those exceeding the values of 18, 7, 12, and 23, respectively.

The tertiary outcome will be the presence / absence of statistically significant differences between changes in "low" 2014 Domain total scores from 2014 to 2019, and changes in "high" 2014 Domain total scores from 2014 to 2019.

Plan of Statistical analyses As previously, the normalities of the distributions of domain and subcomponent scores will be determined by Kolmogorov-Smirnov test. Data will be reported as number, number (percentage), and mean±SD or median (interquartile range), as appropriate. Comparisons pertaining to the primary, secondary, and tertiary outcome will be conducted using an independent samples t-test or the Mann Whitney exact U test as appropriate.

As previously, bivariate linear regression will be performed to explore possible associations between the 2019 Domain A and D total scores. These analyses will include 1) the 2019 total scores calculated as reported above; and 2) the 2019 total scores after the addition of the grades of the responses to the "new - 2019-only" questions. All analyses will be performed using SPSS version 24.