A) Trained or experienced facilitators
In all interventions the ACP discussions were facilitated by a trained health care professional (mostly nurses or allied health workers) [9–11, 43–46] or by health care professionals already experienced in counselling and communicating with patients about ACP, such as social workers or palliative care physicians [7, 47]. The facilitator trainings ranged from half a day to two days and used a competency-based educational approach, comprising interactive discussions about the key components of ACP, role play exercises, reading materials, and learning to assess a person’s capacity to engage in ACP [9–11, 43, 45, 46]. A minority of studies did not provide training to the health care professionals involved [8, 48, 49].
B) Identification of patients
In most interventions, patient selection was focused on those with a serious advanced life-limiting illness, such as advanced cancer, COPD, end-stage renal disease or end-stage congestive heart failure [7, 8, 10, 43, 44, 47, 50, 51]. Other indicators used to select patients with whom ACP was initiated were age (e.g. all patients ≥65 years) [9, 45, 47, 48, 50, 51], admission to a health care facility (hospital or nursing home) [9, 44–46, 48] and the expectation of serious complications or death within the next year [10, 43, 44].
A number of interventions used specific tools such as individualized patient-specific questionnaires about the patient’s preferences for discussing ACP , a question prompt list about end-of-life care provided to patients before their consultation to stimulate conversations , or educational material about ACP mailed to patients in advance [47, 51] to prepare them and to facilitate patient-centred discussions. Structured preference-elicitation and decision aids for ACP to help patients consider their health care options were also implemented as tools during the ACP discussions [9, 10].
D) Structured discussions
All interventions included structured discussion of the patient’s values, goals and beliefs [9, 44], an assessment of their understanding of their illness [10, 43, 44, 46], discussion of their future treatment preferences [9–11, 43, 44, 46], the assessment of their surrogates’ understanding of their illness and treatment preferences and their role as health care agents [9–11, 43–45, 48] and the opportunity to complete ADs [9, 10, 43, 45–48]. The discussions reportedly lasted between one and one and a half hours [9, 10, 45]. In all studies, the patient was encouraged to include their family.
E) Completion of ACP documents
Most ACP interventions provided the opportunity to complete ACP documents (e.g. documentation of treatment preferences, appointment of a health care proxy, appointment of a surrogate decision-maker) [9–11, 44–51]. A number of studies reported that the completed documents were filed in the patient’s medical records or charts [9, 11, 46]. In two studies conducted in an inpatient setting, extra time was dedicated during the interdisciplinary team meeting after the ACP intervention to discussion of the care wishes of the patient [11, 44].