An evaluation of the predictive validity and reliability of two measures of person-centered caregiving with nursing home residents with dementia

Abstract

The goal of this dissertation was to test the predictive validity of two measurements of person-centered care (PCC), the Person-Centered Behavioral Inventory (PCBI), and the Global behavior scale (GBS). These measures were developed as part of a previous study to assess certified nurse aide students’ person-centered skills following an educational training program to increase person-centered behaviors (Grosch, Medvene, & Wolcott, 2008). This study tested whether the person-centered behavioral inventory and the global behavior scale could significantly predict scores found using the Resistiveness to Care (RTC) scale. A negative linear relationship was expected based on the theory of Need-Driven Dementia-Compromised Behavior (NDB) that specified that resident’s resistiveness to care was a response to needs of theirs that had not been met. The person-centered behavioral inventory and global behavior scale measured the caregiver’s person-centered skills. Based on the need-driven dementia-compromised theory of behavior, it was expected that caregivers who engaged in person-centered behaviors would be more likely to meet residents’ needs and thus reduce their resistiveness to care. Therefore, person-centered caregiving, as measured by the person-centered behavioral inventory and the global behavior scale, were expected to be negatively associated with residents’ resistiveness to care as measured by the resistiveness to care scale. This study was conducted in collaboration with Kristine Williams, Ph.D. and the University of Kansas School of Nursing using an existing data set consisting of 80 caregiving interactions between nursing staff and residents with dementia. Caregivers’ person-centeredness was coded using the person-centered behavioral inventory and the global behavior scale. vi Residents’ resistiveness to care (e.g. kicking, pulling away, yelling, etc.) during the caregiving interaction was coded using the Resistiveness to scale. Results of correlational analyses showed that nursing staffs’ scores on the global behavior scale marginally predicted residents’ scores for the resistiveness to care scale. The personcentered behavioral inventory was negatively correlated with the resistiveness to care scale, but not significantly. An emergent finding was that nursing staff members’ age was the biggest predictor among variables of residents’ resistive behaviors with age and resistiveness to care scores significantly negatively correlating. The person-centered behavior measures showed strong concurrent validity across residents’ varying behaviors and tasks. The measures also showed good inter-rater agreement and the global behavior scale showed high internal consistency.