Which is the best way to measure loneliness in people with dementia? A comparison of single and multi-item measures
Isla Rippon, Christina Victor
Brunel University London, Uxbridge, United Kingdom

One area where there is potential for studies of loneliness in later life to be more inclusive is by encompassing people with dementia. Few loneliness studies include this important group or their carers. One explanation offered for this exclusion is the ability of people with dementia to answer questions about loneliness. This study looks at the ability of people with dementia to answer questions about loneliness by looking at responses to two loneliness measures: a single-item and a multi-item measure. Data are drawn from baseline wave of the Improving the Experience of Dementia and Enhancing Active Life (IDEAL) study, a five-year (2014-2018) cohort study of 1,550 people with dementia and 1,286 family members or friends who provide support. Loneliness was measured using both a single-item self-report measure and the six-item De Jong Gierveld (DJG) scale. Comparisons are drawn with responses to these questions by carers to establish if there are issues specific to people with dementia or problems with these questions more generally. 10.1% of participants with dementia reported loneliness using the single-item measure and 35.3% using the DJG scale (14.6% and 61.5% respectively for carers). Fewer respondents had missing data on the single-item measure in comparison to the DJG scale (0.9% vs 6.6% for participants with dementia and 4.1% vs 7.0% for carers). Participants with dementia who did not complete the DJG scale had significantly lower levels of cognition (ACE-III) and increased neuropsychiatric symptoms and carers who did not complete the measure had higher levels of wellbeing.