A newly developed self-report instrument, designed to assess variation in the social and psychological impact of a relative's hearing difficulties ('hearing hassles'), was administered to younger relatives of people with hearing impairment. The first aim was to examine the properties (reliability, validity and sensitivity to change) of this instrument, and the second aim was to examine the contributions of a range of characteristics of the older person to the younger relative's appraisal. The analysis was informed by caregiving stress theory (Pearlin et al., 1990). This was a secondary analysis of an existing data set. The original study, which assessed the impact of hearing aid provision on stress-related measures, used a randomized controlled design. In the current analysis (N = 113), group (intervention or control) was controlled in all analyses. The measure of hearing hassles was found to be reliable, valid and sensitive to change. Those with higher hearing hassle scores had relatives with more hearing impairment and hearing disability, and a poorer relationship with the older person, with other characteristics of the older and younger participants controlled. A significant interaction was found between hearing disability and the older person's cognitive function. Implications for theory and intervention were discussed.