Kwok 2013.

Study characteristics
Methods	Design: Parallel group randomised study. No. of participant centres: One, Hong Kong, China. Study dates: Recruitment between February 2011 and March 2012.
Participants	Inclusion criteria: Caregivers of care recipients having clinical diagnosis of dementia of any stage, and being the primary caregivers. Exclusion criteria: Caregivers under the age of 18 years and those exhibiting intellectual impairment. No. of participants randomised to interventions: ‐ Telephone‐delivered psychoeducation group (n = 20). ‐ Control group (n = 22). Baseline characteristics: Caregivers: Most caregivers were female (71%), and spouse or child of patients (89%), 60% of caregivers presented an age between 31 and 50 years. Care recipients: Most patients were female (74%).
Interventions	Experimental: A 12‐session psychoeducation program delivered by registered social workers over the telephone (approximately 30 minutes per session, one session per week). Control: DVD with educational information about dementia caregiving. Lenght of intervention: 12 weeks.
Outcomes	Primary outcomes: ‐ Caregiver burden, measured by the Zarit Burden Interview (ZBI), (score range 0 to 88, higher scores indicate greater burden). ‐ Caregiver self‐efficacy, measured by the Revised Scale for Caregiving Self‐efficacy, (continuous scale from 0% to 100%, higher scores indicated greater sense of self‐efficacy). Secondary outcomes: ‐ Care recipients stage of dementia, indexed by the Global Deterioration Scale, (7 stages from 1 = no cognitive decline, to 7 = severe dementia). ‐ Care recipients cognitive functioning, assessed by the Abbreviated Mental Test. ‐ Behavioral problems of care recipients, measured by the participants using the Cohen–Mansfeld Agitation Inventory (scores range 29 to 203, higher scores indicate greater agitated behaviours). Outcomes assessment: At 12 weeks.
Notes	Clinical trial registration code: Not reported. Funding: None declared. Conflicts of interest: None declared.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "The participants were randomly assigned to either a psychoeducation group (intervention group) or a control group using a computerized randomization program."
Allocation concealment (selection bias)	Unclear risk	Comment: There is no information on allocation concealment.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Comment: Interventions were not blinded.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: "Pretests were carried out by a research assistant or by self‐administration of the caregivers. Post‐tests were administered to the participants approximately 3 months after pretest by a research assistant blind to the group assignment of the participants." Comment: Intervention components and procedures were not comparable throughout the study.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Comment: Ther is a low and non differential attrition: two participants of 20 (10%) dropped out from the intervention group, 2 participants of 22 (9.1%) dropped out from the control group.
Selective reporting (reporting bias)	Unclear risk	Comment: There is no trial registration and/or available protocol.
Other bias	Low risk	Comment: None detected.