Development and psychometric evaluation of an observational coding system measuring person-centered care in spouses of people with dementia

Background: The notion of person-centered care has been important in investigating relationships between people with dementia and paid carers, and measures are available to assess this. It has been suggested that person-centered care may be a useful construct to apply to understand family-care relationships. However, no measures of person-centered care in this context exist. The study aimed to develop an observational measure of person-centered care for this purpose.

Method: First, a coding system incorporating a range of behaviors that could be considered person-centered or non-person-centered was constructed. Examples included a code relating to whether the person with dementia was involved in planning a task, and a code relating to how the spouse responded to confusion/distress. Second, 11 couples, where one partner had a dementia, were recruited and videotaped cooperating on an everyday task. The system was applied to the care-giving spouse's behaviors, labeling examples of behavior as person-centered or non-person-centered. The final step involved assessing the inter-rater reliability of the system.

Results: The system captured nine categories of behavior, which were each divided into person-centered and non-person-centered types. The system had good reliability (Cohen's κ coefficients were: 0.65 for category and whether behaviors needed to be placed in a category; 0.81 for category excluding the decision about whether behaviors needed to be placed in a category; and 0.79 in relation to whether behaviors were person-centered or non-person-centered.)

Conclusions: Although the small sample size limits the implications of the results, the system is a promising quantitative measure of spousal person-centered care.