Quality of life in dementia: impact of cognition and insight on applicability of the SF-36.

Objectives: Comparability of measures of quality of life in dementia and in other diagnostic groups, such as mild cognitive impairment, normal aging, or other diseases, is highly desirable. However, the impact of cognitive deficits and impaired insight on applicability and validity of generic instruments is sparsely studied.

Participants And Measurements: Sixty patients with dementia [38 women; age: mean (SD) = 78.7 (6.4) years; Mini-Mental State Examination (MMSE): mean (SD) = 20.2 (6.0)] recruited as part of the start-modem study, a multicenter care research study in Germany, completed the generic instrument SF-36 and the specific instrument Quality of Life-Alzheimer's Disease (QOL-AD).

Results: QOL-AD self-rating scores [mean (SD) = 32.8 (5.9)] and SF-36 subscales indicated moderate to good quality of life in the total group. Reliability and validity of five subdomains of the SF-36 were poor in subgroups of patients with impaired insight or with MMSE scores less than 17 (Cronbach's α <0.7, no significant correlation to the QOL-AD). In contrast, for patients with both adequate insight and MMSE score greater than 16 (n = 33; 55%) Cronbach's α of the subdomains of the SF-36 ranged between 0.920 and 0.676. Seven of the eight subdomains correlated significantly with the QOL-AD self-rating and composite score in this group of patients (0.355 ≤ r ≤ 0.709).

Conclusions: Despite the impact of insight and cognition on self-rated quality of life, we found reliable and valid data for a broad spectrum of patients with dementia. According to the present data, the SF-36 is suitable for dementia patients with both insight into their deficits and an MMSE score greater than 16.