Guidelines for the Psychosocial and Bereavement Support of Family Caregivers of Palliative Care Patients

The literature review was carried out to evaluate information published in the last decade related to family caregivers of palliative care patients. The objectives were (1) to outline the experiences, impact, and needs of family caregivers during care provision for their relative and into bereavement; (2) to review evidence-based psychosocial and bereavement support interventions for family caregivers of palliative care patients; and (3) to use the review as a reference for evaluating the levels of evidence for the guideline development process. The specific search strategy, inclusion criteria, and results are outlined elsewhere.^33,39

A meeting with the project investigators was arranged to review and further develop the first draft of the clinical guidelines. The project team consisted of a multidisciplinary group of academics, clinicians, a policy maker, and manager who had significant expertise in palliative and/or bereavement care. Prior to the meeting, project investigators were sent a copy of the literature review and a draft of the guidelines for review.

The evidence to support the guidelines was rated in accordance with the National Health and Medical Research Council (NHMRC) (Australia).⁴⁰ The NHMRC sets out the basis for rating key components of the body of evidence for each recommendation (guideline). Due to the paucity of high-level evidence in this field, and the subsequent need to draw on evidence outside of the gold-standard randomized controlled trial, two further levels of evidence were added. Qualitative studies were acknowledged, and expert opinion was included to support some of the guidelines. The process incorporated one reviewer who rated literature based on the above mentioned criteria. Rated literature was then randomly selected (50%) for further review by two of the authors who were blinded to the level of evidence rating. Any discrepancies were discussed amongst the group until consensus was achieved.

Ethics approval to conduct this phase of the project was gained from St Vincent's Hospital (Melbourne). A panel of state and national key stakeholders from Australia were identified by the project team (via purposive sampling) and invited by letter to participate in either a focus group or structured interview that was designed to obtain opinion and feedback on the second draft of the guidelines. Stakeholders included representatives from multiple health care disciplines currently involved in palliative or bereavement care, policy makers, and consumers. Five stakeholders participated in a focus group, nine participated in an interview, and two provided written feedback. Focus groups participants included a clinical psychologist, pastoral care worker, a palliative care nurse with specialisation in aged care, a multicultural health representative and a palliative care volunteer. The focus group was audio taped for the purpose of research analysis. Individual interviews were conducted with a palliative care physician, oncology social workers (n=2), caregiver support advisor (Carers' Australia), psychiatrist, palliative care nurse practitioner, palliative care service manager, psychooncology service manager, and support service manager (Cancer Council New South Wales (NSW). Further written feedback was provided by a consumer representative and a support service manager (Cancer Council NSW).

The majority of respondents agreed that the guidelines would be helpful to patients, families, and professionals working in clinical palliative care. The points exemplifying the major findings from the focus group and interviews are outlined in Table 1.

A summary of the findings from the focus group and interviews were presented to the project investigators. The investigators and project officers discussed the findings and further revised the guidelines based on the comments and recommendations made, and a third draft of the guidelines was developed. This version included a set of principles to underpin the guidelines.

A modified Delphi process was conducted to obtain consensus on the guidelines from national and international experts in the field of palliative care. Ethics approval was provided by St Vincent's Hospital (Melbourne). Invitations to participate in the process were sent to national and international experts (n=82) in the fields of palliative and bereavement care. Participants were selected by members of the project team on the basis that they were recognized as an authority or expert in palliative and/or bereavement care.

The guidelines and a set of principles to underpin them developed by the research team were accompanied by a questionnaire comprising four parts. Part A was designed to collect participant information including gender; country; region (metropolitan, regional city, or rural area); current occupation; primary profession; and main area of speciality of participants. Part B consisted of one qualitative question inviting participants to review the principles (n=13) and comment on the clarity and comprehensiveness of the principles. Part C invited participants to rate their level of agreement with each guideline and to provide feedback on each guideline. Part C questions were answered using a Likert scale of 1-5 (1=strongly disagree and 5=strongly agree), with a sixth answer option of “don't know.” Four qualitative questions were included to ascertain further feedback from participants on each guideline and to encourage participants to provide further examples of good clinical practice and the source from which the example was generated (e.g., literature or personal experience). Part D provided participants with the opportunity to add general comments regarding the guidelines. The Delphi process was administered via the Internet using LimeSurvey® software.

There is no standard level of consensus that should be attained by an expert panel involved in a Delphi process.⁴¹ The research group made an a priori decision that a consensus required 75% or more of the members of the expert group to agree or strongly agree with the proposed guideline. Participant responses were evaluated and summarized to identify those guidelines with a high level of agreement and those for which there was no agreement or clear disagreement. To determine support or otherwise for each guideline, scores for each question were totalled, with a higher score indicating a more positive response to the guideline.

Demographic data were subject to descriptive analysis (frequencies and percentages). Data were analyzed for each guideline separately (frequency, medians, means, standard deviations) to enable decisions to be made as to whether consensus had or had not been reached. Written comments were thematically analysed and data circulated to the research group for consideration. When consensus had been reached (greater or equal to 75% in agreement), the guideline was considered to be completed.

Fifty-two experts registered to participate in the survey (recruitment rate (52/82=63%) and 26 completed the survey, corresponding to a response rate of 50%. The majority of participants were female (n=19, 73%) and worked in a major metropolitan area (n=20, 77%). Many participants were international (n=16, 62%), with 10 participants based in Australia (38%). Generally, participants were academics (n=20, 77%), with five (19%) in clinical roles, and one participant in clinical management. Palliative care was the most common specialization (n=17, 65%), followed by bereavement (n=5, 18%), oncology, aged care, and psychooncology (all n=1, 4%).

Feedback on the principles was very positive with most participants (n=24) agreeing with them and providing only minor feedback in terms of phraseology modifications. Five participants commented that the use of “primary family caregiver” could be exclusionary to other family members and caregivers/support persons who may not be recognized as family; alternate terms were suggested.

Analysis of guideline rating data revealed consensus was reached for all 20 guidelines, with more than 75% (M=87%) of the 26 expert panel members rating each guideline positively.

Overall, participants agreed that the guidelines were important, practical, realistic, relevant, and applicable to both specialist palliative care and general health care providers. Qualitative feedback provided generally focused on the details associated with the guidelines. For example, respondents agreed with Guideline 10, Determine the current state of and risk for poor psychological health and/or prolonged grief and plan relevant intervention(s), with a consensus level of 92.3%. However, many respondents stated that the screening tools utilized should be reliable, valid, and administered by appropriately trained staff. All qualitative feedback was reviewed by the research team and relevant alterations were made. As consensus was reached in the first round of the Delphi process, a second review round by the expert panel was not required.

Official endorsement of the guidelines was provided by Beyondblue, the national depression initiative; Caregivers Australia; Caregivers Victoria; Palliative Care Australia; Palliative Care Victoria; Australian Centre for Grief and Bereavement; St Christopher's Hospice, London (United Kingdom); and the International Observatory of End of Life Care (United Kingdom).