Many people with dementia are cared for by their spouses. Poor quality of the dyadic relationship is a known risk factor for carer burden and mistreatment of the person with dementia. This study aimed to design and cross validate the Tavistock Relationships Observational Scale (TROS) for assessment of the quality of the relationship in such couples. The study was conducted within the context of the therapeutic intervention with n = 11 couples. Couples' video-recorded interactions were assessed independently by three psychologists using the 13-item TROS. Cronbach alphas of the scale computed for the three raters were in the range 0.76-0.88 (on average 0.83), thus demonstrating high internal consistency of the scale. Fleiss Kappa coefficient for the three rates computed for 22 assessed videos ranged between 0.45 (moderate agreement) and 0.85 (almost perfect agreement) with 50% of coefficients being above 0.50, suggesting the good performance of the TROS in terms of yielding a satisfactory inter-rater agreement. The TROS also showed good external validity when using as validation criteria caregiving partner's burden (measured with the Zarid Burden Interview: the scales Social burden [Rho = -0.65, p = 0.021] and Role strain [Rho = -0.68, p = 0.014]), depression of the person with dementia (measured with the Cornell Scale for Depression in Dementia: Rho = -0.78, p = 0.020), and relationship satisfaction of the caregiving partner (measured with the dyadic adjustment scale: Rho = 0.52, p = 0.038). The TROS also showed a good capacity to detect the change in scores from time 1 to time 2; moreover, power analysis showed that the TROS can detect the change in scores even in very small samples (n = 5 couples). The TROS may be a useful tool for a standardized assessment of the quality of dyadic relationships in couples with dementia in research, nursing, or clinical settings to assist clinical psychologist, nurses, or social workers in a preliminary evaluation of the situation of couples with dementia, to choose a suitable care or intervention, to assess the effectiveness of interventions focusing on improvement of dyadic relationships, or to assess the risk of spousal abuse or neglect of a person with dementia.
Read full abstract