Assessment of neuropsychiatric symptoms is crucial to the management of dementia. In this context, the Neuropsychiatric Inventory (NPI) is the most widely used comprehensive tool, assessing twelve domains (delusions, hallucinations, dysphoria, anxiety, agitation/aggression, euphoria, disinhibition, irritability/lability, apathy, aberrant motor activity, night-time behaviour and appetite/eating abnormalities). NPI records severity, frequency and caregiver distress. It has been translated and validated into a number of languages and shown to be reliable in cross-cultural studies.

To present preliminary results of the NPI Portuguese validation study.

A cross-sectional study was conducted with a convenience sample of institutionalized elderly (≥ 60 years old) in three Nursing Homes in Portugal. All patients were also assessed with MMSE (cognition) and GDS (depression). NPI was administered to a formal caregiver, usually from the clinical staff. Inter-rater and test-retest reliability were assessed in a subsample of 25 randomly selected elderly.

The sample included 166 elderly, with a mean age of 80.9 (10.2sd). Three out of the NPI behavioural items had significant negative correlations with MMSE: delusions (r=−0.177, p=0.024), disinhibition (r=−0.174, p=0.026) and aberrant motor activity (r=−0.182, p=0.020). The NPI subsection of depression/dysphoria correlated positively with GDS total score (r=0.166, p=0.038). NPI showed good internal consistency (overall α=0.766; frequency α=0.737; severity α=0.734). The inter-rater reliability was excellent (ICC=1.00, CI95% 1.00–1.00), as well as test-retest reliability (ICC=0.91, CI95%0.80–0.96).

These results suggested that the Portuguese version is a valid and reliable instrument for the evaluation of neuropsychiatric symptoms in elderly people.