1.They were group interventions with a focused educational input, or they provided targeted support activit
2.They targeted specific groups, such as women, care-givers, the widowed, the physically inactive, or people with serious mental health problems . The majority of studies judged to be partially effective also targeted specific groups.
3.They stated that the experimental samples were representative of the intended target group .
4.They enabled some level of participant and/or facilitator control or consulted with the intended target group before the intervention .
5.They evaluated an existing service or activity (demonstration study) or were developed and conducted within an existing service.
6.Participants were identified from agency lists (GPs, social services, service waiting lists), obituaries, or through mass-media solicitation. Three studies acknowledged a problem of self-selection.
7.The studies included some form of process evaluation and their quality
was judged to be high (6 of 10).
8.The only identified studies evaluating the effectiveness of physical ac- tivity (2 of 10) were effective in reducing loneliness, although in one this was reversed after 12 months.