Psychometric assessment of depression in an elderly general medical population. Over- or underassessment?

A total of 247 consecutively evaluated geriatric medical patients was administered a battery of neuropsychological and psychological tests as part of their diagnostic workup for unexplained deterioration in their functioning. Depression was assessed with a short form of the MMPI, the Brief Symptom Inventory, and the Geriatric Depression Scale. By Research Diagnostic Criteria, most suffered from major (59%) or minor (21%) depressions; some degree of cognitive impairment was seen in 80% of the patients, defining a population of "vulnerable" geriatric patients typical of referrals to a general medical hospital setting. Using both conventional score cutoff criteria and discriminant analyses, false-negative rates up to 53% for major depression and 100% for minor depression were found. Psychometric misrecognition of depression was not related to degree of dementia or education but on some measures was positively associated with verbal intelligence level and patient age. Contrary to previous suggestions that psychometric measures overestimate depression in the elderly, these findings suggest that there may be a subgroup of elderly in which treatable affective distress is not appreciated.