Methods
A 2-phased approach was used in a systematic search of the literature using the following electronic databases: PubMed, CINAHL, and PsycINFO. Grey literature was included from ProQuest for dissertations and theses. Conference abstracts of the American Academy of Hospice and Palliative Medicine/Hospice and Palliative Nurses Association Annual Assembly were also searched in the Web of Science electronic database.
Using a standard population-intervention-comparison-outcome format and selecting terms guided by Medical Subject Headings terms and the PsycINFO thesaurus, keywords were initially grouped into 3 categories (Table 1). Terms were combined using Boolean logic OR within each category and AND across categories to purposefully conduct a broad search. Relevant articles known to the authors were noted to be missing from the initial catchment, and a second search of the literature was conducted with the assistance of a health sciences librarian. Medical Subject Headings terms were revised, and keywords were added to enhance a second search. The following additional terms were used: end of life, life limiting illness, family meeting, family conference, terminal care, professional family relations, terminally ill patients, and family communication. Articles retrieved by these search strategies were subjected sequentially to title, abstract, and full-text screening. Each citation was independently screened by at least 2 researchers.
Inclusion criteria for the systematic search included original primary research, published after the Patient Self-determination Act of 1991, that evaluated structured communication interventions directed toward patients and families. Studies were limited to adult participants (>18 years old), those with a life-limiting illness, and their adult family members. Articles published in English, French, and Portuguese were included in the review. These languages were included as at least 1 team member had sufficient reading skill and the team hoped their inclusion would facilitate a review with a broader multicultural perspective. Unpublished literature in the form of theses and dissertations were considered. Studies about euthanasia and physician-assisted suicide were excluded as they do not meet the objective of this systematic review.
Quality Appraisal
Quality appraisal was addressed in 2 sequential processes. Initially, the research reports to be included were divided among the team members for review and relevant data were extracted into a standardized matrix. The matrix included explicit identification of strengths and weakness of each study's design and methods. To further evaluate the quality of the included articles, a standardized quality assessment tool was used. No articles were excluded for poor quality.