Providing high quality care for dying patients and their families is very important. One way one can assess the care provided is to ask bereaved relatives to complete a questionnaire after their family member has died. The questionnaire can ask about their experiences and their thoughts about the care provided to their family member.

One such questionnaire is the 'Care Of the Dying Evaluation' (or CODE). CODE has been developed with the help and support of bereaved relatives and has been used extensively within the United Kingdom. In this project the investigators want to use the CODE questionnaire to look at bereaved relatives' views about care provided in seven different countries within Europe and Latin America. In the first part of the project CODE was translated into the main language of each country. Volunteers and bereaved relatives in each country were asked to give feedback about whether CODE was easy to understand, sensitive, and easy to complete. Based on the feedback a common version of CODE that is suitable for use across all the countries was developed.

In the next phase of the project, relatives who have recently experienced a bereavement where one of their family members has died from cancer in a hospital, will be invited to complete the CODE questionnaire about two months after the patient's death. The relatives may complete CODE on paper, using a computer, or by interview. The aim is to have 100 completed CODE questionnaires from each of the seven countries. The data from the questionnaires will be used to make a report on the current quality of care for dying cancer patients in hospitals across the seven countries. It will also be possible to compare the care between the countries and identify areas needing improvement.

In the next phase of the project, health care professionals, researchers and bereaved relatives together will use their knowledge and experience to find effective ways to improve the weak areas identified, and assess the results of putting these changes into practice.

ABSTRACT

Background: In order to ensure the highest quality of care is provided for dying patients and their families, one needs to first be able to robustly evaluate the current quality of care. One identified method is to assess this from the user-perspective by conducting bereaved relatives' surveys. 'Care Of the Dying Evaluation' (CODE) is a recognised, validated post-bereavement questionnaire which has been extensively used within the United Kingdom. The ERANet-LAC CODE project aims to use CODE in a wider international context. In Work Package 1 of the project, CODE will be translated and pilot tested in volunteers and bereaved relatives in the partner countries. Using a consensus procedure, a final version of the international CODE (i-CODE) questionnaire will be developed.

Aims: The present project (Work Package 2) aims to advance the international evidence in care for dying cancer patients by undertaking an observational study of bereaved relatives' views across seven participating countries. The overall aims of Work Package 2 (WP2) are to:

• Conduct an international survey of bereaved relatives of cancer patients dying in hospitals, using the CODE questionnaire
• Use the CODE data to provide feedback about the quality of care and level of family support at an international and national level, allowing for cross-country comparisons
• Conduct a web survey of all participating institutions to be able to interpret the survey data in the context of information about each organisation and the level of specialist palliative care provision at each site / country

Methods: Bereaved adult relatives, to adult cancer patients who had an 'expected' death in hospital, will be approached face to face, by telephone or in writing and invited to complete the CODE questionnaire. The CODE questionnaire pack will be sent out 6-8 weeks after bereavement. CODE may be completed via self-completion paper questionnaire, via on-line questionnaire, or facilitated via a researcher using a tablet, telephone or face-to-face interview, according to what is feasible and acceptable in each country. The aim is 100 completed questionnaires per country. Basic demographics will be recorded for all potential participants and for the patients who died.

Analysis of results: Data will be analysed using SPSS and according to the CODE user guide, to present a common international report, individual country reports and cross-country comparisons. Data will be interpreted in the context of knowledge about the individual sites/countries.

Conclusions: The expected outcomes from WP2 are an international survey and cross-country comparisons about the current quality of care for dying cancer patients as perceived by bereaved people, including key areas where care needs to be improved. WP3 of the project will use the results of the international survey to implement changes to improve the care.

INTRODUCTION

The present project was proposed in response to the ERANet-LAC 2nd Joint Call on Research and Innovation and approved for funding under the European Commission's 7th framework program. The aim of the project is to inform and develop an evidence based approach to systematically standardize assessment and care of dying cancer patients, using relative generated outcomes. The project will be conducted in four European and three South American countries three years starting from January 2017. The project contains three Work Packages (parts). This protocol describes the work to be undertaken in Work Package 2 (CODE International Survey).

BACKGROUND

The delivery of appropriate care for dying cancer patients remains a key medical, social, economic and political issue. However, the quality of care for the dying is diverse both within and between EU-LAC countries.

Based on the international evidence base, a set of core principles for care of the dying has been defined. These principles are applicable to the care of dying cancer patients, as well as other patients, worldwide.

In order to ensure the highest quality of care provisions, one needs to first be able to robustly evaluate the current quality of care. One method is to assess this from the user-perspective by conducting bereaved relatives' surveys.

'Care Of the Dying Evaluation' Questionnaire

One post-bereavement questionnaire is 'Care Of the Dying Evaluation' (CODE). This is a 42-item self-completion questionnaire, developed within the UK, and focused on the quality of care and the level of support provided to individuals and their families in the last days of life and the immediate post-bereavement period.

CODE has been used and validated with bereaved relatives, including undertaking cognitive 'think aloud' interviews to assess face and content validity; assessing CODE's stability over time by conducting test-retest reliability analysis; and assessing the construct validity and internal consistency of CODE. Subsequently, CODE has been used across hospices, hospitals and community settings.

In Work Package 1 of the present project, CODE was translated from English into the main language of each partner country according to established international procedures, and pilot tested in volunteers and bereaved relatives to assure understanding and appropriate cultural adaptation. Based on feedback from the testing in each country, a common, international version of CODE (iCODE) has been established through a consensus procedure.

AIMS & OBJECTIVES

The present project (Work Package 2) aims to advance the international evidence in care for dying patients by undertaking an observational study of bereaved relatives' views across seven participating countries, by use of the CODE questionnaire.

The overall aims of Work Package 2 (WP2) are:

• Conduct an international survey of bereaved relatives of cancer patients dying in hospitals, using the CODE questionnaire
• Conduct a web survey of all participating institutions to aid in the interpretation of the survey data
• Use the CODE data to provide feedback about the quality of care and level of family support at an international and national level, allowing for cross-country comparisons

METHODS

Study sites

The necessary number of hospitals caring for adult cancer patients will be recruited as study sites. The study will only be performed in institutions defined as hospitals

Recruitment

The relative / next-of-kin must be approached after the patient's death. The investigators will approach the person recorded as next-of-kin in the patient's hospital record.

Only one completed questionnaire will be included per deceased patient.

Recruitment will be done prospectively. The method of recruitment will be adapted as to what is feasible and in the individual country.

DATA COLLECTION

CODE International Survey

Data to be collected The CODE questionnaire is the common data collection tool of the study. CODE also includes demographic data. Information about the hospital stay will be included for all cases, as part of the inclusion procedure.

Method of data collection

The CODE questionnaire will be sent to the participants 6-8 weeks post bereavement, with on reminder. CODE International Survey was intended as a postal survey, but other data collection methods will also be accepted.

Sample estimation:

Minimum acceptable number of questionnaires per country is 100 (with this number, a 95% confidence interval will give a margin of error of +/- 10% of the primary outcome).

Data transfer

The electronic version and database for CODE for each country/language will be developed in Norway using the Corporate Surveyor software. The databases will be merged in Norway for the final common analyses.

A web survey of study sites will also be conducted.

DATA ANALYSIS

Data will be analysed using the R statistical package, using descriptive statistics (proportions (%) for categorical data; means and standard deviations for continuous parametric data and medians and interquartile range for continuous non-parametric data).

The dataset, anonymised at patient, site and country level, will be aggregated at international level to enable an international report. Results will be summarised in tabular and graphical format, as appropriate.

Free text comments will be analysed using qualitative methods (text condensation).

PROJECT MANAGEMENT AND WORK PLAN

All partners have signed the consortium agreement. The PIs are responsible for the CODE international survey in their respective countries. The group of PIs constitute the project General Assembly and decision-making body.

WP1 (Preparation and Piloting) ran from month 1 to month 7 of the project, i.e. from January to August 2017.

WP2 is designed to run from month 8 to month 22 of the project, i.e. from September 2017 to November 2018.

Time line and deliverables

Confirmation of ethical approval for CODE International Survey in each country: month 6 (July 2017) Web survey of institutions ready for use: month 6 (July 2017) Fully developed, tested and web-accessible data collection tools: month 6 (July 2017) i-CODE ready for use: month 7 (Aug 2017) Completed web survey of inclusion sites: month 10 (Nov 2017) Completed international survey with at least 600 completed CODE questionnaires: month 22 (Nov 2018) Report on CODE International Survey data: month 22 (Nov 2018)

SPONSORSHIP AND BUDGET

Financial support for the project was granted from national research funding agencies in the participating countries, except in the UK. The University of Liverpool, UK, takes part in the project as self-financed partner.

No commercial interests are involved in the study. No conflicts of interest have been declared. Participants receive no financial benefit for participation.

ETHICAL APPROVAL AND GOVERNANCE

The Project Lead (PI) for each country will seek the necessary approvals for CODE International Survey, from their local/regional ethics committee, and from their institutional review board as needed. All countries will need ethical approval for WP2.

All potential participants will be given a Participant Information Sheet detailing the aims of the study and informing that participation is fully voluntary.

PATIENT AND PUBLIC INVOLVEMENT Both volunteers and bereaved relatives will be deeply involved in developing i-CODE to be used for the international survey.

After the survey has been performed, bereaved relatives will be involved in the action planning on how to address the main areas of concern identified.

CONFIDENTIALITY AND DATA MANAGEMENT

The Project Lead for each country will preserve the confidentiality of participants taking part in the study. All data will be anonymised and individual participants will not be identifiable from published data.

All data will be stored on a secure, password protected research server. Data will be stored for up to five years after the completion of the project before being confidentially shredded or deleted.

DISSEMINATION AND PUBLICATIONS

A website for the project will be established.

Authorship will be defined according to the Vancouver Guidelines. National data may be published by the study group in each country.

The last project meeting with presentations and discussions of the project findings will be arranged as an open international conference.

Inclusion Criteria:

• Next-of-kin to a patient who died an 'expected' death from cancer in hospital
• Has been present at the hospital together with the patient at least some of the time during the patient's last two days. (This might not always be known or recorded, therefore the participant may have the option to pass the questionnaire on to somebody better placed to complete it.)
• Patient was ≥ 18 years of age at the time of Death
• Patient had been admitted to the hospital (not any specific ward) at least three calendar days (e.g., admission August 1st, died August 3rd)
• Able to give written informed consent, which might be implied when the participant completes and returns the questionnaire, in keeping with the ethical stipulations for each country

Exclusion Criteria:

• Patient had a sudden and unexpected death
• Unable to complete the questionnaire due to language abilities or reduced cognitive functioning (in some countries the offer of a translator would be provided if someone wanted to complete the questionnaire but had difficulty due to languages)

Next-of-kin to cancer patient who died an expected death in hospital

The plan only includes sharing of data within the consortium.

One year, during the data collection.