80 aging, mental health, and well being

mental health and subjective well being are
powerful indicators of how well societies serve
their members both individually and collec
tively. That is, effective societies not only
meet the basic needs of their members, but
also provide the conditions and opportunities
that sustain emotional health and perceptions
that life is good.
Three topics regarding aging, mental health,
and well being are reviewed here: descriptive
information about the distribution and
dynamics of mental health and subjective
well being in late life, evidence about the social
antecedents of mental health and subjective
well being in late life, and the role of social
factors in the course and outcome of late life
depression.

THE EPIDEMIOLOGY OF MENTAL
HEALTH AND WELL BEING IN
LATE LIFE

The vast majority of Americans are relatively
free of psychiatric or emotional symptoms and
are generally satisfied with their lives. This
pattern is at least as strong for older adults
as for young and middle aged adults. It is
important to define the terms mental health
and subjective well being in both concep
tual and empirical terms. Subjective well
being is the more straightforward of the two
and is generally conceptualized as perceptions
that life is satisfying and meaningful. Typical
measurement strategies include a global self
assessment of life satisfaction (e.g., as unsatisfy
ing, somewhat satisfying, and very satisfying),
multi item life satisfaction scales, and, more
recently, multidimensional scales that tap sev
eral aspects of life quality (e.g., life satisfaction,
purpose in life, self acceptance). Each measure
ment strategy has characteristic strengths and
weaknesses. The global rating is easily and
quickly administered, but generates limited
variability. Life satisfaction scales generate
more variability than global self ratings, but
often include items that arguably measure the
conditions that generate satisfaction with life
in addition to subjective well being. The con
ceptual and empirical clarity of multidimen
sional scales is even more problematic. For
example, high quality social relationships
is one of the subscales of the most commonly
used multidimensional scale. Most sociolo
gists, however, view social bonds as a predic
tor of subjective well being rather than an
element of it.
Defining and measuring mental health is
even more problematic. Although the label
mental health is typically used, in fact
investigators define and measure emotional
distress and dysfunction rather than mental
health. Two distinctions are sources of contro
versy among researchers. The first quandary is
whether to measure overall psychological dis
tress, regardless of the types of symptoms indi
viduals experience, or whether to measure
specific psychiatric syndromes such as depres
sion and anxiety. At this point, both approaches
are used, although the latter is more common.
The second controversy is whether to use diag
nostic measures of the presence or absence of
mental illness or to use symptom scales that are
used in continuous form. Again, there are
countervailing advantages and disadvantages.
Diagnostic measures have the advantage of
identifying severe cases of mental illness, ren
dering findings of interest to clinicians and
policymakers, as well as to sociologists. The
disadvantage of diagnostic measures is that
they have limited variability and ignore much
of the significant distress caused by emotional
symptoms that do not meet the criteria for a
full blown psychiatric diagnosis. In contrast,
the advantage of symptom scales is that they
capture the full range of psychiatric symptoms
in the population, but focus on a distribution
in which most symptomatic individuals suf
fer few if any functional consequences from
their symptoms. Although discussion of these
issues is often heated, empirical evidence
suggests that the relationships between social
factors and diagnostic vs. symptom scales of a
specific syndrome vs. psychological distress
are highly similar (Kessler 2002; Mirowsky
& Ross 1989).
Sociologists initially hypothesized that older
adults would be disadvantaged in life satisfac
tion relative to their younger counterparts as a
result of the social and physical losses char
acteristic of late life. Contrary to this hypoth
esis, older adults report significantly higher