Interventions improved both, moderately: objective social contact by Hedges’ g = 0.43 and perceived quality of connection by g = 0.33. Different things worked for each. Giving people access to other people was the most effective way to raise contact (g = 0.67). Teaching skills for the biased beliefs, fear-driven avoidance and practical barriers that get in the way was the most effective way to improve how connected people felt (g = 0.53).
How it was done
A systematic review and meta-analysis of 58 randomised controlled trials of interventions for social isolation, loneliness, social participation or social connectedness in adults (mean age 62). It separates two outcomes that earlier reviews had merged: how much contact people had, and how good their connections felt.
What it means
The split is the useful part. Earlier reviews had mixed contact and quality into one outcome and got mixed results. Kept apart, the picture is clear: put people in a room and they will have more contact, but it does not follow that they will feel more connected. That took a different intervention, aimed at what people believed about social situations and at the avoidance those beliefs produce.
It matches the prediction-error cluster. If the barrier to reaching out is a wrong forecast of how it will land, an intervention aimed at the forecast should do more for how connected people feel than one that only adds opportunities, and here it did.
The quality effects appear as negative numbers in the paper because those measures run the other way: lower loneliness and distress is the improvement. The sizes are given here without the sign.