Corry 2015.

Methods	Randomised controlled trial (pilot study) (September 2009‐September 2015)
Participants	Caregivers of people with multiple sclerosis (PwMS) recruited from 3 neurological sites (n = 70) in the Republic of Ireland. Caregiver mean age was 51.3 years (SD 13.4), range 22‐84 years; 47.1% were male and 52.9% were female. The majority had at least secondary level education (51.5%) or tertiary education (41.4%). The remaining 7.1% had primary education. Twelve (17.1%) were single, 50 (71.4%) were married and 2 (2.9%) were living as married. Three (4.3%) were separated or divorced and three (4.3%) widowed. An average of 8.8 hours (SD 9.04) were spent caring in a 24‐hour period and the average number of years caring for the PwMS was 11 (SD 7.69). Most (60%) were living with the care‐recipient. Twenty‐five (35.3%) were in paid employment with hours ranging from 6 to 90 hours per week. Thirty‐six (51.4%) were not employed.
Interventions	Intervention: Nurse‐led pro‐active telephone support (n = 33) Aim: To enable nurse specialists in multiple sclerosis (NSMS) help family members and caregivers of PwMS learn problem management skills in order to be better prepared for their role in supporting a person with MS Interventionist(s): Three NSMS who had completed a postgraduate diploma in clinical practice along with a certificate in MS nursing Mode of delivery: Telephone Duration:3 months (four calls; two in month one, one in month two and one in month three) Content: In advance of receiving the calls, the support persons received the support person guidebook. During the calls, the NSMS referred to the guidebook. The support person guidebook was designed to facilitate the process and enable nominated support persons prepare for the calls from the nurse specialists. Scripted interviews were designed to provide a focus for the telephone contacts and help the NSMS and support persons structure their interaction using a problem management approach. Standardisation: Standardisation of interventionist training was maximised through the inclusion of a training section in the intervention manual. Training for delivery of the intervention was provided in accordance with the intervention manual. Each NSMS received a minimum of two hours one‐to‐one training. Comparison group: Usual care (n = 38).
Outcomes	1. Quality of life: WHOQoL BREF instrument ‐ 26 questions covering four domains (physical, psychological, social relations, and environmental). Higher scores indicated better quality of life. 2. Burden: Caregiver Reaction Assessment (CRA), a 24‐item scale, which assesses how caregivers react to caring for an ill person in 5 domains: how caring affects caregivers health, daily schedule (schedule disruption), finances, their sense of self‐worth (self‐esteem), and their family. Higher scores indicated greater burden. 3. Skill acquisition (preparedness to care): using the support person preparedness scale, an 8‐item subscale with a five point rating scale developed as part of The Family Care Inventory in the early 1980s (Archbold 1990). Higher scores indicated greater perceived preparedness to care. 4. Health status and well‐being (self‐efficacy): Self‐Efficacy for Problem‐solving scale, a 4‐item caregiver self‐efficacy in problem management scale and Self‐Efficacy for Obtaining Respite subscale. Higher scores indicated greater self‐efficacy. 5. Satisfaction with the intervention: Client Satisfaction Questionnaire‐8 (CSQ‐8), an 8‐item questionnaire with a four option response ranging from 1‐4, with higher scores indicating greater satisfaction. Data were collected at the 4‐week time point which was prior to completion of the intervention, and at the 3‐month time point (approximately week 12), which was the end of the intervention time point.
Notes	The term support person (SP) was used for caregivers. Standard deviation data for burden and preparedness to care were obtained from the author. Following email communication with the originators of the burden instrument (CRA), the subscale result for 'schedule disruption' was used for 'Burden' in the review. Funding source: Fellowship from the Health Research Board Ireland
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	A random list of numbers for the control and intervention groups at each site was generated by a statistician independent of the study (p.174).
Allocation concealment (selection bias)	Low risk	Allocation held by a person independent of the trial and sent via email on enrolment to the trial (p.174)
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	High risk	Not blinded
Incomplete outcome data (attrition bias) All outcomes	Low risk	Overall attrition minimal (2 per group) and reasons provided (p.238)
Selective reporting (reporting bias)	Low risk	All prespecified outcomes were reported.
Other bias	High risk	The interventionists who delivered the intervention delivered care to both groups.