Summary of findings for the main comparison. Telephone support intervention compared to Usual care for providing education and psychosocial support for informal caregivers of adults with diagnosed illnesses.

Telephone intervention compared to Usual care for providing education and psychosocial support for informal caregivers of adults with diagnosed illnesses
Patient or population: Informal caregivers of adults with diagnosed illnesses Setting: Community Intervention: Education or psychosocial telephone support Comparison: Usual care
Outcomes	Anticipated absolute effects* (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with Telephone intervention
Quality of Life End of intervention Assessed with: WHOQoL Brief instrument (26‐item), SF‐36 (0‐100 scale), SF‐12, and Adapted BKOS (15‐item, 0‐7 scale) For all scales, higher scores indicated higher QoL.	The mean score for QoL in the intervention group was 0.02 standard deviations lower (0.24 lower to 0.19 higher)	364 (4 RCTs)	⊕⊕⊕⊝ MODERATE 1,2	One further study reported that caregivers receiving a telephone intervention may have slightly higher QoL at end of intervention, compared with usual care. Overall, at the end of intervention, telephone interventions probably have little or no effect on caregiver QoL
Burden End of intervention Assessed with: Revised Memory and Behaviour Problem Checklist (0‐24 range), 24‐ and 26‐item Caregiver Reaction Assessment, Family Appraisal of Caregiving Questionnaire (Caregiver Strain subscale – Palliative Care), Burden Interview (22‐ and 12‐item inventories; 0‐4 scale), Modified BKOS scale (22‐item, 5‐point scale) For all scales, higher scores indicated higher burden.	The mean score for Burden in the telephone group was 0.11 standard deviations lower (0.3 lower to 0.07 higher)	788 (9 RCTs)	⊕⊕⊝⊝ LOW 2,3	Two further studies reported caregiver burden. One reported that telephone interventions may decrease burden; the other reported no change in the intervention group, compared with usual care. Overall, at the end of intervention, telephone interventions may have little or no effect on caregiver burden.
Psychological health: Depression End of intervention Assessed with: Center for Epidemiologic Studies Depression Scale 11‐item SF, 10‐item and, 20‐item measures (including German version) (0‐3 scales), Brief Symptom Inventory (18‐item, 5‐point scale), and the Geriatric Depression Scale (30‐item, score range 0‐30) For all scales, higher scores were associated with increased depression/symptoms of depression.	The mean score for depression in the telephone group was 0.37 standard deviations lower (0.7 to 0.05 lower )	792 (9 RCTs)	⊕⊝⊝⊝ VERY LOW 2,4	Three further studies reported caregiver depression. One reported that telephone interventions may decrease depression; the other two reported no change in the intervention group, compared with usual care. Overall, we are uncertain of the effects of telephone interventions on caregiver depression at the end of intervention.
Satisfaction with the intervention End of intervention	See comment	‐	‐	No study was found that assessed this outcome comparatively. Six studies measured satisfaction with the intervention in the intervention group only. All six reported high levels of satisfaction with the intervention (i.e. 'mostly', 'very much so', 'good' or 'excellent').
Adverse events including suicide and suicide ideation	See comment	‐	‐	No studies measured these outcomes.
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

¹ Although participant numbers were relatively few at n = 364, they were deemed sufficient for studies evaluating these types of intervention in a population of informal caregivers; we therefore did not downgrade on sample size (imprecision) for this reason.

² Most information was from studies at low or unclear risk of bias on most items in the 'risk of bias' tool, although in some studies, one or two risk of bias criteria were assessed as having high risk of bias. We therefore downgraded by 1 level for plausible risk of bias that could seriously alter the results.

³ Some variation in the effect estimates and moderate heterogeneity; evidence downgraded by 1 level (serious inconsistency).

⁴ Variation in the effect estimates across studies and substantial heterogeneity; evidence downgraded by 2 levels (very serious inconsistency).

^{BKOS: Bakas Caregiver Outcomes Scale
 QoL: Quality of life
 SF: Short Form
 SF‐12: Short Form ‐12 items
 SF‐36: Short Form ‐ 36 items
 WHOQoL: World Healthcare Organisation Quality of Life}