Psychometric Testing of the Revised 15-item Bakas Caregiving Outcomes Scale

Abstract

Family caregivers of stroke survivors experience a variety of negative social, emotional, and health-related outcomes as a result of providing care. : To psychometrically test the revised 15-item Bakas Caregiving Outcomes Scale (BCOS) measuring life changes specifically resulting from providing care. The original 10-item BCOS was improved by adding five items addressing financial well-being, level of energy, role functioning, physical functioning, and general health. Psychometric testing of the revised 15-item BCOS using a sample of 147 family caregivers of stroke survivors approximately 4 months after stroke was conducted to determine the quality of the items, internal consistency reliability, test-retest reliability, construct validity, and criterion-related validity. Most caregivers were women (78.9%), White (68.0%) or African American (29.9%), and either spouses (60.1%) or adult children (31.3%). Satisfactory evidence of internal consistency (alpha = .90) and 2-week test-retest reliability (intraclass coefficient [ICC] = .66; 95% confidence interval [CI] = 0.42-0.81) was provided, with item-to-total correlations ranging from .41 to .74. Unidimensionality was supported by confirmatory factor analysis with indices, indicating a good fit. Using hierarchical multiple regression, 36% of the BCOS variance was explained by constructs in the conceptual model [F(11,132) = 6.72, p < .001]. Criterion-related validity was supported by correlations with the 36-item Short Form (SF-36) General Health Subscale (r = .32, p < .001) and a criterion variable measuring how caregivers' lives had changed overall (r = .67, p < .001). The revised 15-item BCOS has evidence of satisfactory reliability and validity in family caregivers of stroke survivors. The BCOS is a valuable measure in research and can be used to identify priority areas for nursing interventions designed to improve caregivers' outcomes.