Piamjariyakul 2015.

Methods	A mixed‐method design with random assignment (pilot study) (Study dates not reported)
Participants	African‐American caregivers of people with heart failure recruited from an outpatient cardiology HF follow‐up clinic in a Midwestern Medical Centre, USA. Caregivers were spouses (65%, N = 13) of patients or were other family members (35%, N = 7), i.e. sister, parent, daughter or granddaughter. Of 20 dyads, 15 (75%) lived in the same household, while 25% (5 dyads) lived separately. Ten caregivers were assigned to the intervention group and 10 to the standard care group. Caregiver age ranged from 40‐78 years with a mean age of 61.4 years (SD 10.0). The majority, n = 17 (85%), were female, 8 (40%) high school or lower, 12 (60%) vocational, college or more. The majority were married, 14 (70%), and employed, 12 (60%). Seven caregivers in the intervention group had vocational or higher education versus 5 in the standard care group. Caregivers reported their chronic health conditions: hypertension (n = 11), myocardial infarction or cardiovascular disease (n = 4), diabetes mellitus (n = 4), osteoarthritis/pain (n = 4), and one caregiver each reported the conditions of depression, thyroid problems, asthma, and HIV.
Interventions	Intervention: The adapted FamHFcare coaching intervention plus standard care (n = 10). Aim: To test whether a culturally‐sensitive telephone coaching intervention could reduce patients’ HF‐related re‐hospitalisation and family caregiver burden and depression, and increase family caregiver confidence, social support, and preparedness to care Interventionist(s): Experienced nurse interventionist Mode of delivery: Telephone Duration: 4 weeks (weekly calls 60‐90 minutes depending on caregiver questions and need for reinforcement). Content: FamHFcare includes 4 weeks of post‐hospital coaching via telephone on specific HF home care skills using teach‐back strategies. FamHFcare aligns with all ACCF/AHA clinical guideline based instructions for daily sodium/fluid restrictions, medication adherence, and symptom monitoring and reporting. Prior to the first telephone session, each family received the coaching program materials by mail: (1) two AHA home caregiving guides (symptoms checklist and staying healthy guidelines for caregivers); (2) a list of local support organizations; (3) the national award winning book Comfort of Home for Chronic Heart Failure: A Guide for Caregivers; (4) low‐sodium booklet, and (5) a plastic daily pill organiser. The nurse interventionist engaged each dyad in four weekly FamHFcare coaching sessions scheduled at their convenience. Standardisation: no details provided Comparison group: Standard care. This included the education and materials routinely given to all HF patients through hospital discharge planning. The standard medical and nursing clinical care in both groups was not changed for this study. Standard care information is not specific to the needs of African‐Americans or to caregivers (n = 10).
Outcomes	1. Burden: a 17‐item five‐point Likert‐type scale in which higher scores indicated more burden or difficulty in providing caregiving. Response options were: 1 = providing caregiving but the task was not difficult to 5 = extremely difficult. Option “N/A = not applicable” was provided and selected by caregivers who did not provide a specific caregiving task. Higher scores indicated greater perceived burden. 2. Psychological health (depression): The Center for Epidemiologic Studies Depression Scale (CES‐D). A higher score indicated higher level of depression. 3. Skill acquisition (preparedness to care): A one‐item Likert type scale (1 = not at all, 4 = very well prepared); higher scores indicating caregivers felt better prepared. Outcome data were collected at 6 months (medium‐term follow‐up > 3 to ≤ 6 month time point).
Notes	Author provided detail via email, which was used in the evaluation of the quality of the intervention, for example, detail on monitoring of delivery of the intervention and adherence to trial protocol. Funding source: Award from Kansas City Life Science Institute, Blue Cross Blue Shield, Kansas City, Kansas (p.466).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Insufficient information to make assessment
Allocation concealment (selection bias)	Unclear risk	Insufficient information to make assessment
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Unable to blind due to nature of the intervention
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"Data collectors were trained research nurses who were blinded to random assignment" (p.468)
Incomplete outcome data (attrition bias) All outcomes	Low risk	Minimal attrition (Intervention: 2 withdrew after completing and evaluating the first two intervention sessions (one was too ill to continue and the other had a busy work schedule); Control group: 1 patient died) (p.470)
Selective reporting (reporting bias)	Low risk	None evident – all caregiver outcomes reported in Table 3 (p.471)
Other bias	Low risk	No statistically significant differences were found for caregivers or patients (p.469).