Table 5 Results of the feasibility studies, stratified by study design
Author, year | N (n allocated to intervention, control) | Mean age (years) | Gender (% female) | Intervention effect on social isolation outcomes |
|---|---|---|---|---|
Randomized trial(s) (n = 4) | ||||
Rotondi, 2005 [54•] | 30 PWS (i = 16, c = 14), 21 support persons (i = 11, c = 10) | 37.5 (PWS), 51.52 (support persons) | 70% (PWS), 66.7% (support persons) | - Comparing baseline and 3-month follow up, PWS in the telehealth group showed a trend towards greater social support (F(1, 27) = 3.79, p = .062). For support persons, there were no significant differences for social support (F(1, 18) = 0.36). |
O’Mahen, 2014 [55•] | 83 (i = 41, c = 42) | Not specified | 100% | - At 17-week follow up, there were no between-group differences in perceived support scores (95% CI: − 1.79 to 2.03, p = 0.27), effect size is medium (d = 0.50, 95% CI 1.02 to − 0.02). |
Hanssen, 2020 [56•] | - 64 - Final data analysis: 50 (i = 27, c = 23) | 37.9 (i), 40.3 (c) | 33.3% (i), 39.1% (c) | - There was no group-by-time interaction and no group effect on loneliness - Statistically significant decrease in loneliness scores over time in both groups (b = − 0.004, 95%CI: − 0.007 to − 0.0009, p = 0.01, d = − 0.11). |
Gjerdingen, 2013 [67•] | 39 (doula group = 12, telephone support group = 13, c = 14) | 29.7 overall | 100% | - No significant group differences in social support comparing baseline to 6-month follow up |
Non-randomized trial(s) (n = 2) | ||||
Pfeiffer, 2017 [57•] | 48 (family/friend group = 19, specialist group = 29) | 50.1 overall | 25% overall | - No statistically significant differences between time points were found on perceived social support. - There were also no statistically significant differences between groups at 3 or 6 months for perceived social support |
Lim, 2019 [66•] | 20 (young people with SAD = 9, students with no mental health conditions = 11) | 21 (SAD), 20.36 (students) | 44.44% (SAD), 45.45% (students) | - SAD: loneliness scores decreased in a linear trend from baseline to 3-month follow up - Student: loneliness scores decreased from baseline to post-intervention and 3-month follow up - Across the entire group, loneliness showed mean negative slope (M = − 3.82, 95% CI: − 5.54 to − 2.17). On average, participants’ loneliness scores decreased by 7.64 points by follow-up (d = 0.94). |
Single-group study(s) (n = 10) | ||||
Van Voorhees, 2005 [58•] | 14 | Age range: 18–24 years overall | 42.9% overall | - At post-intervention, there was a trend of increasing social support scores (d = 0.27, CI: − 0.73, 1.24, p = 0.13) |
Rice, 2020 [59•] | 89 | 19.8 overall | 47.2% overall | - Comparing baseline and post-intervention, statistically significant improvements were found on social connectedness (d = 0.63, p < .001), significant decreases in loneliness (d = 0.63, p < .001). Changes in social network were not significant. |
Rice, 2018 [60•] | 42 | 18.5 overall | 50% overall | - Comparing baseline and post-intervention, there was no significant increase in social connectedness (p = 0.711) or social support (p = 0.470) |
Price, 2014, [61•] USA | 31 | 37.1 overall | 45.2% overall | - There was no reported analysis on differences of social support, but from the descriptive statistics table there is around a 2-point decrease on social support comparing baseline to 1-month follow up and baseline to 3-month follow up |
Ludwig, 2020 [62•] | 24 | 25.16 overall | 36.8% overall | - Loneliness scores showed moderate reductions from baseline to mid-treatment (6-week) (d = 0.27). - Changes in participants’ perceived social support increased from baseline to 6-week and post-treatment (12-week) (d = 0.03 and d = 0.10), although modest and not maintained at 16 weeks. |
Lim, 2020 [65•] | 12 | 20.5 overall | 25% overall | - The mean of slopes indicated loneliness scores were more likely to reduce after intervention (M = −  0.34, SD = 0.24). - Participants could be expected to have scores that are about 0.3 standard deviations lower at post-treatment, and about 0.6 standard deviations lower at 3-month follow-up than at baseline |
Dow, 2008 [68•] | 14 | 65.5 overall | 86% overall | - There was a decrease of loneliness scores for 11 participants |
Campbell, 2019 [69•] | 105 | 16.2 overall | 81.9% overall | - Due to the drop-off in response rates between the baseline survey (105/105, 100%) and final survey (8/105, 7.6%), data quality was too low to conduct meaningful analysis. |
Bailey, 2020 [63•] | 20 | 21.7 overall | 55% overall | - Comparing baseline and post-intervention, differences on social connectedness were not statistically significant |
Alvarez-Jimenez, 2018 [64•] | 14 | 20.3 overall | 78% overall | - Statistically significant improvements were found in subscales of social support at post-intervention: attachment (d = 0.70, p = 0.05) and guidance (d = 0.75, p = 0.03). - 33% of participants had a reliable decline on loneliness |