Assessing Social Isolation: A Pilot Study of Different Methods

Purpose: This pilot study has two purposes: 1) to determine if standard recruitment methods reach isolated older adults; and 2) to compare self-report measures of social isolation to senior housing staff ratings of residents’ social isolation.

Methods: Older adults living in a senior housing community (SHC) in St. Louis, MO were invited to participate in a survey using standard recruitment methods (announcements, mailed letters, and flyers). Each resident who completed the survey was given a $20 research incentive. The Lubben Social Network Scale 6 (LSNS-6) was used to determine the residents’ self-reported social isolation. The two subscales from the LSNS-6 (family subscale and friends subscale) were also compared to staff ratings of isolation. The senior housing manager rated each resident based on the extent of their social isolation using the following four categories: not at all, a little, some, or a lot. Residents were marked as missing if the housing manager did not know the extent of a resident’s isolation.

Findings: Thirty-five percent of the 135 residents (n=46) completed the survey. The housing staff rated 55.3% (83) of residents as not isolated at all, 24% (36) as a little isolated, 8% (12) as having some isolation, and 2.7% (4) of residents as having a lot of isolation. Residents who participated in the survey had lower staff ratings of isolation (mean staff isolation score=1.18) in comparison to residents who did not participate (mean staff isolation score=1.70).

Forty-four residents had self-report isolation data and staff ratings of isolation. The LSNS-6 found 29.5% of residents (13) were socially isolated. The family subscale of the LSNS-6 found 50% of residents (22) were isolated from their families. The friend subscale of the LSNS-6 found 31.8% of residents (14) were isolated from their friends. Staff rated 82% of residents (36) as having no isolation, and 18% of residents (8) as having a little isolation; staff did not rate any residents as having some isolation or a lot of isolation. Twenty-two percent of the residents (8) who staff rated as having no isolation were deemed socially isolated based upon their LSNS-6 scores.

Conclusions/Implications for Future Studies: Senior housing staff ratings indicate standard recruitment methods were not effective methods to survey the most isolated residents. It is important to be aware of this potential sampling bias when conducting social isolation research. Furthermore, there are discrepancies between the self-report measures of social isolation and the staff ratings of residents’ isolation; the self-report measures of social isolation found a higher prevalence of isolation in comparison to staff ratings. Future studies should consider utilizing alternative methods for surveying isolated residents in senior housing, including targeted outreach and higher research incentives. Additional studies using larger samples and more senior housing communities are needed to determine if these results are consistent. Social isolation is an important area of study for social workers, as indicated by the fact that it is proposed as a Grand Challenge by the American Academy of Social Work and Social Welfare.