Subjective Memory Complaints at Midlife and Across
the Life Span
SMC Assessed by a Single or Few Questions with Face
Validity
Two reports from the Baltimore NIMH Epidemiological Catchment Area (ECA)
study provide insights into the relationship between SMC and objective memory
performance in quite large, community-based samples. Here, SMC were measured
by a single question about the presence of memory problems, for example,“Do you
find that you have trouble with your memory?”(Bassett and Folstein 1993 )or
“How is your memory compared to that of other people of your age?”(Podewils
et al. 2003 ). Thefirst of these studies (Bassett and Folstein 1993 ) used a 3-word
recall measure from the Mini-Mental State Exam to measure objective memory in
810 adults (age range 18–85+ years; 72% under age 65). Across this broad age
range, the prevalence of poor memory performance (11%) was about half the
prevalence of reported memory problems (22%), but people with memory com-
plaints were twice as likely to show impairments in objective performance. Overall,
there was 68% agreement between memory complaints and poor memory perfor-
mance. The relationship between memory complaints and memory performance
varied with age; SMC were significantly related to objective performance in young,
but not middle-aged and older adults (Bassett and Folstein 1993 ).
The second ECA study (Podewils et al. 2003 ) involved a population-based
cohort of 1488 subjects across the adult life span (30–80+ years; 65% under age
60). In addition to the SMC question, this study assessed the amount of distress
caused by SMC, by including an item,“How often do youworryabout forgetting
things?”The objective memory task was a 20-word recall task, which was more
difficult and more sensitive than the 3-word recall task used in the earlier study
(Bassett and Folstein 1993 ). Notably, 36% of the sample reported worry about their
memory, with 11% of those worrying often. As in the earlier study, ratings of
memory and the frequency of worry about memory were significantly associated
with performance on an objective memory task after controlling for age, education,
race, and sex. In comparison with worry about memory, however, self-reports of
memory function were a better indicator of memory performance. There were also
notable discrepancies between self-ratings of worry and self-ratings of memory
function. Worry was higher among middle-aged adults (40–50 years) than older
individuals, but older adults (especially those over 80) rated memory as lower than
middle-aged individuals. Worry was higher in women compared with men, but
women ranked their memory performance similarly to men. The authors concluded
that“The present findings should reassure younger people in the baby-boom
generation that memory worry is frequent but generally unfounded on the basis of
objective testing.”
It is interesting to note that recent studies have reported a subtle but significant
decline in word recall and psychomotor speed in midlife women as they transition
280 M.T. Weber and P.M. Maki