Socioeconomic Status 557disproportionately perceive their environments as threaten-
ing, harmful, or challenging as a result of ethnically speci“c
stimuli (Clark, Tyroler, & Heiss, 2000; S. James, 1993;
Krieger, 1990; Outlaw, 1993; Sears, 1991; Thompson, 1996;
D. Williams, Yu, Jackson, & Anderson, 1997), they may be
required to expend an inordinate amount of •ener gyŽ to cope
with the psychological and physiological stress responses that
follow these perceptions, relative to European Americans.
Over time, the cumulative psychological and physiological
effects associated with these added stressors have the poten-
tial to account for, in part, between- and within-group health
disparities.
SES and Behavioral Risk Factors
The major chronic diseases and disease-speci“c mortality
have common behavioral risk factors that are interrelated in
complex ways. For example, smoking is related to heart
disease and lung cancer; dietary intake (e.g., saturated fat,
cholesterol intake, and sodium intake) and physical inactivity
are related to obesity and hypertension; obesity is related to
hypertension, heart disease, and diabetes; physical inactivity
is related to hypertension; and hypertension is related to heart
disease and cerebrovascular disease (JNC, 1993; NCHS,
1998). Research suggests that smoking, obesity, dietary
intake, and hypertension are inversely related to SES (Harrell
& Gore, 1998; King, Polednak, Bendel et al., 1999; Lowry,
Kann, Collins, & Kolbe, 1996; Luepker et al., 1993;
Winkleby, Robinson, Sundquist, & Kraemer, 1999), and that
statistically adjusting for known behavioral risk factors does
not eliminate the SES-health gradient (Lantz et al., 1998;
Smith, Shipley, & Rose, 1990).
Research has also identi“ed factors that appear to decrease
the probability of disease occurrence. These protective fac-
tors (e.g., physical activity and health knowledge) have been
shown to be positively associated with SES (Jeffrey &
French, 1996; Luepker et al., 1993). Additional research is
needed to delineate why higher disease risk pro“les are over-
represented among persons low in SES (Elman & Myers,
1999; Harrell & Gore, 1998; W. James, Nelson, Ralph, &
Leather, 1997).
SES and Psychosocial Risk Factors
In addition to these more traditional biobehavioral risk and
protective factors, the examination of psychosocial factors
may lead to a more informed understanding of the relation-
ship between SES factors and health outcomes (N. Anderson
& Armstead, 1995; Taylor, Repetti, & Seeman, 1997). That is,
given the plausible mechanistic links between psychosocial
factors and some physical health outcomes and processes (N.
Anderson, McNeilly, & Myers, 1991; Barefoot, Dahlstrom,
& Williams, 1983; Burch“eld, 1985; Cacioppo, 1994; R.
Clark et al., 1999; Everson, Goldberg, Kaplan, Julkunen, &
Solonen, 1998), coupled with the observation that known and
measured risk factors do not account for all of the variability
in SES-health differentials (Lantz et al., 1998; D. Williams,
1996), it is possible that psychosocial factors mitigate the
relationship between SES and health outcomes. These psy-
chosocial factors include anger expression, perceptions of
unfair treatment (e.g., racism and sexism), cynical hostility,
coping styles, and locus of control. For example, S. James,
Strogatz, Wing, and Ramsey (1987) found that the active-
coping style of •John HenryismŽ interacted with SES to in-
crease the risk of hypertension for African American, but not
European American, males. That is, African American males
who were low in active coping and low in SES were nearly
three times more likely to be hypertensive, compared to
African American males who were high in active coping and
high in SES. Subsequent studies have failed to “nd support
for the John Henryism: The ability to assess the degree to
which people feel they can control their environment SES
interaction in females and more af”uent samples (S. James,
Keenan, Strogatz, Browning, & Garrett, 1992; Wiist & Flack,
1992).SES and Prevention and Intervention ApproachesPersons of low SES, regardless of ethnic group, are more
likely to have no health insurance coverage, no physician
contact, greater unmet needs for health care, and more avoid-
able hospitalizations, compared to persons of medium and
high SES (NCHS, 1998). Because access to health care is
generally needed to take advantage of prevention and inter-
vention services, it is reasonable to postulate that SES will be
inversely related to the availability and use of these services.
Also, to the extent that these services are positively related to
health outcomes (Alexander et al., 1999; Fortmann,
Williams, Hulley, Maccoby, & Farquhar, 1982; JNC, 1993),
persons of low SES would be expected to have the poorest
outcomes.
Relative to persons of higher SES, persons of lower SES
are less likely to report ever receiving or being up-to-date on
prevention services such as cholesterol screening, Pap smear,
stress test, mammography, and breast examination (Davis,
Ahn, Fortmann, & Farquhar, 1998; Haywood et al., 1993;
NCHS, 1998; Solberg, Brekke, & Kottke, 1997), but not
blood pressure screening or •neededŽ services (Solberg et al.,
1997). The positive relationship between the receipt of
services and SES has also been observed for intervention