Family and Peer Support for Alzheimer's Caregivers: Psychometric Reevaluation of the Perceived Social Support Subscales

Purpose: Published information on the shortened, 20-item version of the Perceived Social Support Scale (PSSS20) is limited. The purpose of this study was to evaluate the psychometric properties of the PSSS20 family (PSSS20-Fa) and friends (PSSS20-Fr) subscales. The population of focus was informal, Alzheimer's disease (AD) caregivers, who provide help and support with daily activities for persons diagnosed with AD to avoid institutionalization and minimize costly in-home care for the care recipient. The relevance of this population in this study is tied to their commonly reported, coping method of seeking social support.

Method: The study involved analysis of cross-sectional data from 229 attendees of two, large Alzheimer's caregiver conferences in the eastern United States. A self-report questionnaire included demographic items and content from standardized measures of stress, psychological resilience, and the two types of social support. Entire sample data were analyzed for sample characteristics; descriptives on the standardized measures; and evidence of construct validity with the PSSS20-Fa and PSSS20-Fr via their correlations with caregivers' perceived stress and resilience. Odds ratios with Cochran–Mantel–Haenszel test for significance were calculated to corroborate correlational evidence of validity. Sample data were randomly halved (n₁=115 for PSSS20-Fa; n₂=114 for PSSS20-Fr) to conduct principal axis factor analysis on each support measure. The number of factors retained for final solution was established by a minimum eigenvalue of 1.0. Subsamples 1 and 2 data were also used to examine internal consistency among PSSS20-Fa and PSSS20-Fr responses.

Results: The typical participant in the study was identified as 45 years old, married, Caucasian female who is a child of the care recipient. No significant, mean differences in overall PSSS20-Fa and PSSS20-Fr scores existed among demographic characteristics. Caregivers reported moderately high levels family and friend support, moderate level of stress, and a high level of resilience. Preliminary construct validity for PSSS20-Fa and PSSS20-Fr was suggested by their mutual, significant (p < .05) correlation, along with their significant convergence with perceived stress and resilience. Odds ratios corroborated validity, indicating likelihood of higher support given lower stress and higher resilience. Factor analyses indicated a three factor solution with all items loaded for each social support measure. Each measure evidenced good, internal consistency: for PSSS20-Fa, Cronbach's a = .89 and Guttman split-half coefficient = .92; for PSSS20-Fr, Cronbach's a = .90 and Guttman split-half coefficient = .94.

Implications: Findings from this study offered an empirical perspective on supportive assistance from family members and friends of AD caregivers. The PSSS20-Fa and PSSS20-Fr appear to be psychometrically sound measures of social support, more convenient in brevity than their original predecessors. The PSSS20 is a concise, easy-to-administer instrument that can be used to appraise perceptions of social support; in this study, used among persons caring for those with AD. Such appraisal offers social work professionals some insight regarding the caregiver's perceived environment of support or isolation, which in turn can lead to promotion of supportive interactions that encourage feelings of social belonging.