close

Вход

Забыли?

вход по аккаунту

?

Education level and mortality in systemic lupus erythematosus SLEEvidence of underascertainment of deaths due to SLE in ethnic minorities with low education levels.

код для вставкиСкачать
Arthritis & Rheumatism (Arthritis Care & Research)
Vol. 51, No. 4, August 15, 2004, pp 616 – 624
DOI 10.1002/art.20526
© 2004, American College of Rheumatology
ORIGINAL ARTICLE
Education Level and Mortality in Systemic Lupus
Erythematosus (SLE): Evidence of
Underascertainment of Deaths Due to SLE in
Ethnic Minorities With Low Education Levels
MICHAEL M. WARD
Objective. To determine if socioeconomic status, as measured by education level, is associated with mortality due to
systemic lupus erythematosus (SLE), and to determine if these associations differ among ethnic groups.
Methods. Sex- and race-specific mortality rates due to SLE by education level were computed for persons age 25– 64 years
using US Multiple Causes of Death data from 1994 to 1997. SLE-specific mortality rates were compared with all-cause
mortality rates in 1997 to determine if the association between education level and mortality in SLE was similar to that
in other causes of death.
Results. Among whites, the risk of death due to SLE was significantly higher among those with lower levels of education,
and the risk gradient closely paralleled the 1997 all-cause mortality risks by education level. However, in African
American women and men and Asian/Pacific Islander women, the risk of death due to SLE was lower among those with
lower education levels, contrary to the associations between education level and all-cause mortality in these groups.
Comparing the distribution of education levels among deaths due to SLE and all deaths in 1997, persons with lower
education levels were underrepresented among deaths due to SLE in African Americans and Asian/Pacific Islanders.
Conclusion. Among whites, higher education levels are associated with lower mortality due to SLE. These associations
were not present in ethnic minorities, likely due to underascertainment of deaths due to SLE in less-well educated
persons. This underascertainment may be due to underreporting of SLE on death certificates, but may also represent
underdiagnosis of SLE in ethnic minorities with low education levels.
KEY WORDS. Education; Mortality; Socioeconomic status; Systemic lupus erythematosus; Ethnic minorities; Health
disparities.
INTRODUCTION
In epidemiologic studies, higher socioeconomic status has
been consistently associated with lower all-cause mortality and with lower mortality due to cardiovascular and
cerebrovascular disease (1–15). Similar associations have
been found for both whites and African Americans when
race-specific mortality has been examined (2– 4,6 –10,
12,14,15). In studies that used education level as the measure of socioeconomic status, mortality rates among those
Michael M. Ward, MD, MPH: National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, Bethesda, Maryland.
Address correspondence to Michael M. Ward, MD, MPH:
NIH/NIAMS/IRP, Building 10, Room 9S205, 10 Center
Drive, MSC 1828, Bethesda, MD 20892-1828. E-mail:
wardm1@mail.nih.gov.
Submitted for publication June 6, 2003; accepted in revised form September 15, 2003.
616
with ⱕ12 years of education were often 1.5–2.0 times
higher than those of college graduates (1,4,5,8,13). Similar
associations have been reported for patients with rheumatoid arthritis (16 –20).
Few studies have examined the association of socioeconomic status with mortality in systemic lupus erythematosus (SLE) (21–25). Most studies examined clinic-based
cohorts, which may be susceptible to selection bias and
have limited generalizability. Four of the 5 studies used
medical insurance status as the measure of socioeconomic
status, or inferred patients’ socioeconomic status from the
economic characteristics of their area of residence, rather
than using more direct personal measures of socioeconomic status, such as income or education level (21–24).
An ecologic analysis of national data from the United
States found higher mortality rates due to SLE in counties
with higher poverty rates (26). Previous studies also reported higher mortality rates due to SLE among African
Americans and Asian Americans, but these studies did not
Education Level and Mortality in SLE
examine mortality by socioeconomic status (27–29).
Knowing if socioeconomic status is associated with mortality in patients with SLE would indicate whether reports
of higher mortality among ethnic minorities were confounded by differences in socioeconomic status. This
question is important because although the health outcomes of patients with SLE have improved over time,
10 –20% of patients with SLE die within 10 years of its
onset (30).
In this study, I examined the hypothesis that socioeconomic status, measured by education level, was inversely
associated with mortality due to SLE in whites, African
Americans, and Asian/Pacific Islanders in the United
States, using national data on deaths from 1994 to 1997.
Trends in SLE mortality by education level were also
compared with similar trends in all-cause mortality.
METHODS
Data on the number of deaths occurring in the United
States in persons with SLE were abstracted from Multiple
Causes of Death files of the National Center for Health
Statistics (31). These files include information on all recorded deaths occurring in the United States each year,
based on death certificates filed in each state. Records for
each death include the underlying cause of death and up
to 20 additional conditions listed as associated or contributing causes of death, by International Classification of
Diseases Ninth Revision (ICD-9) codes (32). Each record
also includes the decedents’ age, sex, race (white, black,
Native American, Asian/Pacific Islander), Hispanic ethnicity, and place of residence, as well as information on the
time and place of death.
Information on the decedents’ education level was first
included on the Standard Certificate of Death in 1989, but
this item was not immediately adopted by all states. In
1993, information on education level was not available for
deaths in New York City, Georgia, Oklahoma, Rhode Island, and South Dakota, was missing for ⬎20% of deaths
in Kentucky and West Virginia (and therefore considered
unreliable), and was missing for 9% of deaths in other
states. Beginning in 1994, information on education level
was available for deaths in 46 states, New York City, and
the District of Columbia. Georgia, Oklahoma, Rhode Island, and South Dakota did not include education level on
their death certificates. Data on education level were missing for ⬎20% of deaths in Kentucky. In the remaining
states, data on education level were missing for ⱕ4.2% of
deaths since 1994.
For this study, all deaths occurring in 45 states and the
District of Columbia (excluding Georgia, Oklahoma, Rhode
Island, South Dakota, and Kentucky) from 1994 to 1997 in
which SLE was included as either the underlying cause,
immediate cause, or contributing cause of death were included. The study was limited to persons age 25– 64 years
because SLE most often occurs in young adulthood or
middle age, because most people have completed their
formal education by age 25, and because death certificate
data on education level may be less accurate among persons age 65 and older (33). For each death, data were
617
abstracted on age, sex, race, years of education, and place
of death.
Nephritis and renal failure are markers of severe SLE
and major risk factors for death (23,34 –37), and likely
identify a subset of persons in whom the diagnosis of SLE
is made with greater specificity. Therefore, decedents
whose records included acute or chronic glomerulonephritis (ICD-9 codes 580 –583), acute renal failure (ICD-9
code 584), chronic renal failure (ICD-9 code 585), or renal
failure, unspecified (ICD-9 code 586) as one of the multiple
causes of death along with SLE were also identified.
Mortality rates by education level for each sex and race
subgroup were computed using census counts as the denominators. Counts by age (in 5-year age groups), sex, race,
education level, and state from the 1990 census were projected to 1994 –1997, and the projections were added to
provide the population at risk of death in these years (38).
Counts for Georgia, Oklahoma, Rhode Island, South Dakota, and Kentucky were omitted from these totals. Although information on deaths by both race and Hispanic
ethnicity was available, mortality by Hispanic ethnicity
could not be analyzed because census data on education
level by both race and Hispanic ethnicity were not available for each state. Therefore, Hispanics were not analyzed
as a separate category, but were included in either the
white, African American, or Native American racial
groups. For white and African American women, mortality
rates were computed for education levels of 0 – 8 years,
9 –11 years, 12 years, 13–15 years, and 16 or more years.
Because fewer deaths due to SLE occurred in men and
Asian/Pacific Islander women, mortality rates for these
groups were computed for education levels of 0 –11 years,
12 years, and ⱖ13 years. Because validation studies of
death certificate information have indicated that ethnic
minorities with ⬍12 years of education were more likely to
be recorded as being high school graduates than recorded
with lower levels of educational attainment, comparisons
were also performed using only 2 categories of education
(ⱕ12 versus ⱖ13 years) (33).
Sex- and race-specific mortality rates were age adjusted
using direct standardization, with the 1990 United States
population as the standard. For strata with ⬍100 deaths,
95% confidence intervals (95% CIs) for the mortality rates
were computed based on Poisson distribution. For strata
with ⱖ100 deaths, 95% CIs for the mortality rates were
computed based on z distribution (39). Poisson regression
models were used to estimate rate ratios for mortality due
to SLE among education groups, with the highest education category as the reference group. P values ⱕ 0.05 were
considered statistically significant. Analyses were performed separately for each sex and race, and for persons
with SLE included as any category of cause of death (underlying, immediate, or contributing), for those with SLE
as the underlying cause of death, and for those with SLE
who had renal disease included among the causes of
death. Analyses were performed using SAS programs (version 8; SAS Institute, Cary, NC).
To determine if the association between education level
and mortality due to SLE was similar to that of other
causes of death, these analyses were repeated using all
deaths in 1997. Consistency in the education-related risks
618
Ward
Table 1. Number of deaths due to SLE in 1994 –1997 and all deaths in 1997 by sex-race subgroup and years of education (<12
years versus >12 years)*
White
women
SLE deaths
ⱕ12 years’
⬎12 years’
All deaths
ⱕ12 years’
⬎12 years’
White
men
African American
women
African American
men
Asian/Pacific Islander
women
education
education
1,456 (65.9)
754 (34.1)
290 (66.1)
149 (33.9)
944 (61.3)
597 (38.7)
137 (64.0)
77 (36.0)
73 (44.8)
90 (55.2)
education
education
91,138 (68.2)
42,461 (31.8)
150,474 (67.1)
73,690 (32.9)
25,885 (73.3)
9,408 (26.7)
42,446 (79.6)
10,837 (20.4)
2,007 (58.5)
1,426 (41.5)
* Data presented as no (%). SLE ⫽ systemic lupus erythematosus.
of mortality between deaths due to SLE and all-cause
mortality would support the validity of this association,
whereas inconsistency would suggest either that the education-mortality association was modified in the setting of
SLE, or that there were deficiencies in the identification of
deaths due to SLE.
RESULTS
From 1994 to 1997, there were 4,779 deaths among persons age 25– 64 years for which SLE was recorded as the
underlying, immediate, or contributing cause. Data on education level were missing for 160 persons (3.3%), and
these people were excluded from analysis. These persons
did not differ from the 4,619 persons who had information
on education level in age, sex, race, or the likelihood that
SLE was the underlying cause of death. Persons with missing data on education level were more likely to have died
in 1994 (4.2%) than in 1997 (2.2%). Of the 4,619 persons
studied, 3,939 (85.3%) were women, 2,649 (57.3%) were
white, 1,755 were African American (38.0%), 187 (4.0%)
were Asian/Pacific Islander, and 28 (0.6%) were Native
American. Three hundred forty-seven persons (7.5%)
completed ⱕ8 years of education, 583 (12.6%) completed
9 –11 years, 2,003 (43.4%) completed 12 years, 981
(21.2%) completed 13–15 years, and 705 (15.3%) completed ⱖ16 years. The mean age of death was 46.1 years.
SLE was the underlying cause of death for 2,852 persons
(61.7%). Renal disease was listed as one of the multiple
causes of death for 1,049 persons (22.7%), with the most
common diagnoses being unspecified renal failure (n ⫽
564) and chronic renal failure (n ⫽ 293). Table 1 shows the
distribution of deaths due to SLE in 1994 –1997 and all
deaths in 1997 by education level (ⱕ12 years versus ⬎12
years). There were too few deaths due to SLE among
Asian/Pacific Islander men (n ⫽ 24) and Native Americans
for meaningful analysis, and these groups were not analyzed further.
Among white women, the age-adjusted mortality rate
due to SLE (as an underlying, immediate, or contributing
cause of death) decreased from 20.9/million in those with
0 – 8 years of education to 7.4/million in those with ⱖ16
years of education (Table 2). White women with 0 – 8 years
of education were twice as likely to have died of SLE than
those with ⱖ16 years of education, and the risks decreased
progressively with increasing education level. Results
were similar for analyses of SLE as the underlying cause of
death and for the subgroup with renal disease, with somewhat stronger risk gradients for these outcomes. Findings
in white men were similar to those in women, with mortality rates due to SLE as any cause of death or as the
underlying cause significantly higher among those with
ⱕ12 years of education (Table 3). For both white women
and white men, the education gradient in risk of mortality
due to SLE closely paralleled the education gradient in
risk in all-cause mortality (Figure 1).
Among African American women, the age-adjusted mortality rates due to SLE were similar among those with 12
years, 13–15 years, and ⱖ16 years of education (Table 4).
However, the mortality rates due to SLE were lower among
those with 0 – 8 years or 9 –11 years of education. African
American women with 0 – 8 years of education were only
48% as likely to die of SLE as those with a college education, and those with 9 –11 years of education were only
76% as likely to die of SLE as those with a college education. Results were similar in analyses of SLE as the underlying cause of death and for the subgroup with renal disease as a cause of death. These decreased risks of mortality
due to SLE contrasted with the increased risk of all-cause
mortality among less-well educated African American
women (Figure 1).
Results for African American men and Asian/Pacific
Islander women demonstrated a pattern similar to that of
African American women. Among African American men,
the risk of SLE-related mortality was 30% lower among
those with 0 –11 years of education than among those with
ⱖ13 years of education (adjusted odds ratio [OR] 0.70),
whereas all-cause mortality was approximately twice as
high in those with 0 –11 years of education (Table 5 and
Figure 1). Among Asian/Pacific Islander women, the risk
of SLE-related mortality among those with 0 –11 years of
education was only 72% of the risk of those with ⱖ13
years of education, whereas the risk of all-cause mortality
was slightly increased among those with 0 –11 years of
education (Table 6 and Figure 1).
To limit the effects of any upgrading of education levels
to high school graduate, the data were also analyzed by
comparing those with ⬎12 years of education with those
reported as having ⱕ12 years of education (Table 7). White
women with ⱕ12 years of education were 1.64 times more
likely to die of SLE than white women with ⬎12 years of
education. This risk was similar to the all-cause mortality
Education Level and Mortality in SLE
619
Table 2. Mortality due to SLE among white women, by education level*
Education
SLE as underlying,
immediate, or
contributing cause of
death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
SLE as underlying cause of
death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
SLE with renal disease among
causes of death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
0–8 years
9–11 years
12 years
13–15 years
16 years or more
191
20.9 (17.5–24.4)
2.05 (1.40–3.00)
0.0002
260
14.1 (12.2–16.0)
1.64 (1.16–2.31)
0.005
1,005
12.1 (11.3–12.8)
1.57 (1.20–2.04)
0.0008
433
7.7 (6.9–8.4)
1.03 (0.72–1.33)
0.90
321
7.4 (6.4–8.4)
1.00
–
122
14.0 (11.1–16.9)
2.37 (1.61–3.47)
⬍0.0001
155
8.9 (7.4–10.4)
1.72 (1.21–2.45)
0.003
595
7.2 (6.6–7.8)
1.59 (1.21–2.09)
0.0007
269
4.7 (4.1–5.3)
1.00 (0.74–1.37)
0.97
196
4.7 (4.0–5.4)
1.00
–
43
4.7 (3.1–6.2)
2.63 (1.31–5.26)
0.007
51
3.0 (2.1–3.9)
1.79 (0.92–3.46)
0.09
201
2.5 (2.1–2.8)
1.72 (1.04–2.86)
0.04
83
1.5 (1.1–1.8)
0.97 (0.56–1.80)
0.98
60
1.6 (1.2–2.1)
1.00
–
* Rates are deaths per million. SLE ⫽ systemic lupus erythematosus; RR ⫽ rate ratio; 95% CI ⫽ 95% confidence interval.
risk of 1.51 associated with lower education levels in
white women. Among white men, the risk of mortality due
to SLE in those with ⱕ12 years of education (OR 2.12) was
almost identical to the all-cause mortality risk (OR 2.10).
However, among African American women (OR 1.03), African American men (OR 0.92), and Asian/Pacific Islander
women (OR 0.90), there was no association between lower
education levels and mortality due to SLE, even though in
each group those with lower education levels had significantly higher risks of all-cause mortality (ORs of 1.38, 1.76,
and 1.22, respectively). Results were similar for the subgroups with mortality due to SLE as the underlying cause
and mortality due to SLE with renal disease.
A possible explanation for the divergent association between education level and mortality risk due to SLE and
its association with all-cause mortality in ethnic minorities, and for the differences in this association between
ethnic minorities and whites, may be found in the distribution of education levels in these groups. In white
women and white men, two-thirds of deaths due to SLE
Table 3. Mortality due to SLE among white men, by education level*
Education
SLE as underlying, immediate, or contributing
cause of death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
SLE as underlying cause of death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
SLE with renal disease among causes of death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
0–11 years
12 years
13 years or more
102
3.2 (2.4–3.8)
2.13 (1.42–3.19)
0.0002
188
2.9 (2.4–3.3)
2.11 (1.51–2.96)
⬍0.0001
149
1.4 (1.1–1.7)
1.00
–
54
1.8 (1.2–2.3)
2.31 (1.40–3.81)
0.0009
96
1.5 (1.1–1.8)
2.03 (1.33–3.09)
0.0009
82
0.7 (0.6–0.9)
1.00
–
25
0.8 (0.5–1.2)
1.81 (0.83–3.95)
0.14
42
0.6 (0.4–0.8)
1.59 (0.82–3.07)
0.17
45
0.4 (0.3–0.6)
1.00
–
* Rates are deaths per million. SLE ⫽ systemic lupus erythematosus; RR ⫽ rate ratio; 95% CI ⫽ 95% confidence interval.
620
Ward
Figure 1. Relative odds of mortality due to systemic lupus erythematosus as the underlying, immediate, or contributing cause of death
(hatched bars) and all-cause mortality (open bars) in persons age 25– 64 years, by education level, in each sex–race subgroup. Odds ratios
were adjusted for age. Persons in the highest education category were the reference group.
and two-thirds of deaths due to any cause occurred in
persons with ⱕ12 years of education (Table 1). However,
among African American women, African American men,
and Asian/Pacific Islander women, there were relatively
fewer deaths due to SLE occurring in those with ⱕ12 years
of education than deaths due to any cause. For example,
among African American women, 73.3% of all deaths occurred in persons with ⱕ12 years of education, but only
61.3% of deaths due to SLE occurred in this education
group, representing an absolute decrease of 12% and a
relative decrease of 20%. This suggests a selective absence
of deaths due to SLE recorded among ethnic minorities
with ⱕ12 years of education.
DISCUSSION
This national, population-based study indicates that education level has an important association with mortality
due to SLE among whites. Compared with those with ⱖ16
years of education, the risk of mortality due to SLE was
twice as high among white women with ⱕ8 years of education. Mortality risks decreased with increasing education level so that risks were similar between those with
13–15 years and ⱖ16 years of education. Among white
women, the mortality risk gradient with education level
was higher among those with renal disease, as might be
expected if renal disease was a marker of more severe SLE,
Education Level and Mortality in SLE
621
Table 4. Mortality due to SLE among African American women, by education level*
Education
SLE as underlying, immediate,
or contributing cause of
death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
SLE as underlying cause of
death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
SLE with renal disease among
causes of death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
0–8 years
9–11 years
12 years
13–15 years
16 years or more
68
33.4 (23.0–43.8)
0.48 (0.31–0.73)
0.0006
221
46.5 (40.0–53.1)
0.76 (0.57–1.02)
0.07
655
59.2 (54.6–63.8)
1.04 (0.82–1.31)
0.76
380
46.5 (41.5–51.5)
0.82 (0.63–1.06)
0.12
217
59.6 (50.9–68.3)
1.00
–
37
19.1 (10.1–26.8)
0.44 (0.24–0.81)
0.008
138
30.4 (25.0–35.7)
0.78 (0.53–1.15)
0.21
450
40.7 (36.9–44.5)
1.12 (0.82–1.53)
0.46
241
28.2 (24.4–31.9)
0.80 (0.57–1.12)
0.19
140
39.2 (32.0–46.4)
1.00
–
14
4.9 (1.4–8.4)
0.34 (0.19–0.62)
0.0003
59
12.2 (8.8–15.5)
0.72 (0.50–1.03)
0.08
155
13.9 (11.6–16.1)
0.87 (0.65–1.18)
0.39
89
11.1 (8.6–13.6)
0.68 (0.49–0.95)
0.03
60
16.0 (11.5–20.4)
1.00
–
* Rates are deaths per million. SLE ⫽ systemic lupus erythematosus; RR ⫽ rate ratio; 95% CI ⫽ 95% confidence interval.
and if socioeconomic status was associated with the severity of SLE. Among white men, risks of mortality due to SLE
were twice as high among those with ⱕ12 years of education, compared with those with ⬎12 years of education.
The association between education level and mortality
due to SLE in whites paralleled that found for all-cause
mortality, and is consistent with many previous studies
that demonstrate that persons of higher socioeconomic
status have lower mortality rates (1–15). These associations support the validity of the analytic approach used
here, which merged information on education level from
death certificates with census data to compute mortality
rates by education level.
In contrast to the findings in whites, and in contrast to
the associations between socioeconomic status and allcause mortality in many prior studies, higher education
levels were not associated with lower mortality due to SLE
among African Americans or Asian/Pacific Islander
women. In fact, lower education levels appeared protective of mortality due to SLE among African American
Table 5. Mortality due to SLE among African American men, by education level*
Education
SLE as underlying, immediate,
or contributing cause of
death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
SLE as underlying cause of
death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
SLE with renal disease among
causes of death
n
Age-adjusted rate (95% CI)
RR (95% CI)
P
0–11 years
12 years
13 years or more
49
5.9 (4.0–7.7)
0.70 (0.55–0.90)
0.004
88
9.3 (7.3–11.3)
1.07 (0.87–1.29)
0.53
77
8.8 (6.6–11.1)
1.00
–
28
3.6 (2.1–5.2)
0.71 (0.48–1.05)
0.09
59
6.1 (4.5–7.7)
1.16 (0.85–1.58)
0.36
49
5.4 (3.7–7.2)
1.00
–
14
1.6 (0.6–2.7)
0.60 (0.32–1.12)
0.11
26
2.7 (1.6–3.8)
0.91 (0.55–1.50)
0.71
26
2.9 (1.6–4.1)
1.00
–
* Rates are deaths per million. SLE ⫽ systemic lupus erythematosus; RR ⫽ rate ratio; 95% CI ⫽ 95% confidence interval.
622
Ward
Table 6. Mortality due to SLE among Asian/Pacific Islander women, by education level*
Education
0–11 years
SLE as underlying, immediate,
or contributing cause of
death
n
24
Age-adjusted rate (95% CI)
14.4 (8.0–20.9)
RR (95% CI)
0.72 (0.43–1.20)
P
0.21
SLE as underlying cause of
death
n
13
Age-adjusted rate (95% CI)
8.3 (3.1–13.4)
RR (95% CI)
0.64 (0.35–1.16)
P
0.15
SLE with renal disease among
causes of death
n
5
Age-adjusted rate (95% CI)
2.6 (0–5.0)
RR (95% CI)
0.49 (0.23–1.04)
P
0.07
12 years
13 years or more
49
22.0 (15.5–28.4)
1.02 (0.69–1.52)
0.92
90
20.0 (15.7–24.3)
1.00
–
30
13.7 (8.6–18.9)
0.97 (0.62–1.50)
0.90
61
13.2 (9.7–16.6)
1.00
–
10
3.6 (1.3–5.9)
0.69 (0.39–1.22)
0.21
26
5.1 (3.0–7.2)
1.00
–
* Rates are deaths per million. SLE ⫽ systemic lupus erythematosus; RR ⫽ rate ratio; 95% CI ⫽ 95%
confidence interval.
women, and similar trends were evident among African
American men and Asian/Pacific Islander women. This
effect appeared to be specific for mortality due to SLE,
because all-cause mortality in these groups was higher
among those with lower education levels. There are several potential explanations for the difference. It is possible
that SLE is more prevalent in ethnic minorities with higher
education levels and more prevalent in whites with low
education levels, and these differences in prevalence
translated into the differences in mortality rates that were
observed. However, this explanation seems unlikely, as
there is no evidence that the prevalence of SLE varies by
socioeconomic status differently in different ethnic
groups. Postponement of mortality past age 65 may also
influence these results, but for this explanation to account
for the ethnic differences detected, one would need to
postulate that high education levels lead to preferential
postponement of mortality in whites with SLE, whereas
low education levels lead to preferential postponement of
mortality in ethnic minorities with SLE, which seems improbable. Similarly, the findings could be explained by
high mortality before age 25 in ethnic minorities with ⬍12
years of education, but low mortality before age 25 in
whites with ⬍12 years of education, in those with onset of
SLE in adolescence or young adulthood. However, the
generally low mortality in this age group makes this explanation unlikely (24,40,41).
It also is possible that the severity of SLE differs among
ethnic groups. The interpretation that ethnic differences in
severity account for the results would need to explain why
low education levels would be associated with more severe SLE and higher mortality rates in whites but with less
severe SLE and lower mortality rates in ethnic minorities.
Similar associations were also present in the subgroup of
patients with SLE and renal disease as a cause of death.
The associations in this subgroup, which controls in part
for the severity of SLE, suggest that differences in severity
are unlikely to account for the findings. A differential
Table 7. Relative risks of mortality due to SLE and all-cause mortality in persons with <12 years of education, compared with
those with >12 years of education, by sex-race subgroup*
White women
All-cause mortality
SLE as underlying, immediate,
or contributing cause
SLE as underlying cause
SLE with renal disease
White men
African American African American
women
men
Asian/Pacific
Islander women
1.51 (1.24–1.83)† 2.10 (1.62–2.73)†
1.64 (1.34–2.01)† 2.12 (1.52–2.96)†
1.38 (1.20–1.57)†
1.03 (0.85–1.25)
1.76 (1.29–2.40)‡
0.92 (0.78–1.08)
1.22 (1.10–1.40)§
0.90 (0.66–1.23)
1.68 (1.36–2.07)† 2.11 (1.44–3.12)†
1.81 (1.12–2.90)§ 1.66 (0.91–3.01)
1.12 (0.89–1.40)
0.99 (0.83–1.18)
0.98 (0.72–1.33)
0.78 (0.49–1.25)
0.84 (0.56–1.25)
0.60 (0.32–1.12)
* Data are presented as age-adjusted rate ratios (95% confidence intervals). SLE ⫽ systemic lupus erythematosus
† P ⬍ 0.0001.
‡ P ⬍ 0.001.
§ P ⬍ 0.05.
Education Level and Mortality in SLE
association between education level and mortality by race
would also conflict with the results of a cohort study that
demonstrated similar socioeconomic effects on mortality
in whites and African Americans with SLE (24).
A more likely explanation for the findings in African
Americans and Asian/Pacific Islanders is underascertainment of deaths due to SLE in those with ⱕ12 years of
education. This group was underrepresented among
deaths due to SLE, with relative decreases of 20 –25%
compared with the proportion of deaths from all causes
among persons of similar education levels in these ethnic
groups. Such underrepresentation was not present among
whites, for which the proportions with ⱕ12 years of education were identical in deaths due to SLE and deaths due
to any cause. Because of this underascertainment, it may
not be possible to report valid mortality rates due to SLE in
African Americans and Asian/Pacific Islanders from this
source.
Bias in the reporting of education level on death certificates results in more people with low education levels
being classified as having completed high school (33).
Although more common among decedents 65 years or
older, this misclassification also occurs in younger age
groups and is more common among African Americans.
Although this upgrading may have impacted our examination of risk gradients by education level in ethnic minorities, racial differences in the associations between education level and SLE-related mortality on one hand, and
between education level and all-cause mortality on the
other hand, persisted when results were compared between those with some college and those with high school
educations or less.
This study indicates that mortality due to SLE is higher
among whites with lower socioeconomic status, as measured by educational attainment. It is difficult to draw
conclusions about the association between education level
and mortality due to SLE in other ethnic groups because of
evidence suggesting possible underascertainment of
deaths due to SLE in ethnic minorities with lower levels of
education. This underascertainment may be due to more
frequent misreporting of causes of death among ethnic
minorities with low socioeconomic status, and specifically
underreporting of SLE among these patients. However, of
more concern is that this underascertainment could reflect
the underdiagnosis of SLE in persons of low socioeconomic status, particularly ethnic minorities. Underdiagnosis would likely be associated with inadequate treatment.
Population-based prospective cohort studies with active
case findings are needed to investigate these possibilities,
and to investigate the role of limited access to medical care
in the diagnosis and treatment of SLE in ethnic minorities
of lower socioeconomic status. These findings also suggest
that SLE may be underrepresented in United States mortality statistics (29).
623
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
REFERENCES
1. Feldman JJ, Makuc DM, Kleinman JC, Cornoni-Huntley J.
National trends in educational differentials in mortality. Am J
Epidemiol 1989;129:919 –33.
2. Sorlie P, Rogot E, Anderson R, Johnson NJ, Backlund E. Black-
24.
25.
white mortality differences by family income. Lancet 1992;
340:346 –50.
Keil JE, Sutherland SE, Knapp RG, Tyroler HA. Does equal
socioeconomic status in black and white men mean equal risk
of mortality? Am J Public Health 1992;82:1133– 6.
Pappas G, Queen S, Hadden W, Fisher G. The increasing
disparity in mortality between socioeconomic groups in the
United States, 1960 and 1986. N Engl J Med 1993;329:103–9.
Sorlie PD, Backlund E, Keller JB. U.S. mortality by economic,
demographic, and social characteristics: the National Longitudinal Mortality Study. Am J Public Health 1995;85:949 –56.
Kaplan GA, Keil JE. Socioeconomic factors and cardiovascular disease: a review of the literature. Circulation 1993;88:
1973–98.
Wei M, Valdez RA, Mitchell BD, Haffner SM, Stern MP, Hazuda HP. Migration status, socioeconomic status, and mortality rates in Mexican Americans and non-Hispanic whites: the
San Antonio Heart Study. Ann Epidemiol 1996;6:307–13.
Elo IT, Preston SH. Educational differentials in mortality:
United States, 1979-85. Soc Sci Med 1996;42:47–57.
Kaufman JS, Long AE, Liao Y, Cooper RS, McGee DL. The
relation between income and mortality in U.S. blacks and
whites. Epidemiology 1998;9:147–55.
Schalick LM, Hadden WC, Pamuk E, Navarro V, Pappas G.
The widening gap in death rates among income groups in the
United States from 1967 to 1986. Int J Health Serv 2000;30:
13–26.
Comstock GW, Tonascia JA. Education and mortality in
Washington County, Maryland. J Health Social Behav 1977;
18:54 – 61.
Modan B, Wagener DK. Some epidemiological aspects of
stroke: mortality/morbidity trends, age, sex, race, socioeconomic status. Stroke 1992;23:1230 – 6.
Howard G, Russell GB, Anderson R, Evans GW, Morgan T,
Howard VJ, et al. Role of social class in excess black stroke
mortality. Stroke 1995;26:1759 – 63.
Escobedo LG, Giles WH, Anda RF. Socioeconomic status,
race, and death from coronary heart disease. Am J Prev Med
1997;13:123–30.
Casper ML, Barnett EB, Armstrong DL, Giles WH, Blanton CJ.
Social class and race disparities in premature stroke mortality
among men in North Carolina. Ann Epidemiol 1997;7:146 –
53.
Pincus T, Callahan LF. Formal education as a marker for
increased mortality and morbidity in rheumatoid arthritis.
J Chronic Dis 1985;38:973– 84.
Pincus T, Brooks RH, Callahan LF. Prediction of long-term
mortality in patients with rheumatoid arthritis according to
simple questionnaire and joint count measures. Ann Intern
Med 1994;120:26 –34.
Leigh JP, Fries JF. Mortality predictors among 263 patients
with rheumatoid arthritis. J Rheumatol 1991;18:1307–12.
Wolfe F, Mitchell DM, Sibley JT, Fries JF, Bloch DA, Williams
CA, et al. The mortality of rheumatoid arthritis. Arthritis
Rheum 1994;37:481–94.
Maiden N, Capell HA, Madhok R, Hampson R, Thomson EA.
Does social disadvantage contribute to the excess mortality in
rheumatoid arthritis patients? Ann Rheum Dis 1999;58:
525–9.
Reveille JD, Bartolucci A, Alarcon GS. Prognosis in systemic
lupus erythematosus: negative impact of increasing age at
onset, black race, and thrombocytopenia, as well as causes of
death. Arthritis Rheum 1990;33:37– 48.
Dubois EL, Wierzchowiecki M, Cox MB, Weiner JM. Duration
and death in systemic lupus erythematosus: an analysis of
249 cases. JAMA 1974;227:1399 – 402.
Ginzler EM, Diamond HS, Weiner M, Schlesinger M, Fries JF,
Wasner C, et al. A multicenter study of outcomes in systemic
lupus erythematosus. I. Entry variables as predictors of prognosis. Arthritis Rheum 1982;25:601–11.
Ward MM, Pyun E, Studenski S. Long-term survival in systemic lupus erythematosus: patient characteristics associated
with poorer outcomes. Arthritis Rheum 1995;38:274 – 83.
Alarcon GS, McGwin G Jr, Bastian HM, Roseman J, Lisse J,
624
26.
27.
28.
29.
30.
31.
32.
33.
34.
Ward
Fessler BJ, et al. Systemic lupus erythematosus in three ethnic
groups. VIII. Predictors of early mortality in the LUMINA
cohort. Arthritis Rheum 2001;45:191–202.
Walsh SJ, DeChello LM. Geographical variation in mortality
from systemic lupus erythematosus in the United States. Lupus 2001;10:637– 46.
Kaslow RA, Masi AT. Age, sex, and race effects on mortality
from systemic lupus erythematosus in the United States. Arthritis Rheum 1978;21:473–9.
Kaslow RA. High rate of death caused by systemic lupus
erythematosus among U.S. residents of Asian descent. Arthritis Rheum 1982;25:414 – 8.
Trends in deaths from systemic lupus erythematosus: United
States, 1979-1998. MMWR Morbid Mortal Wkly Rep 2002;51:
371– 4.
Trager J, Ward MM. Mortality and causes of death in systemic
lupus erythematosus. Curr Opin Rheumatol 2001;13:345–51.
U.S. Department of Health and Human Services, National
Center for Health Statistics. Multiple causes of death, 1994
(computer file), 1995 (computer file), 1996 (computer file),
1997 (computer file). Hyattsville (MD): U.S. Department of
Health and Human Services, National Center for Health Statistics (producer); 1998. Ann Arbor, MI: Inter-University Consortium for Political and Social Research (distributor).
World Health Organization. Manual of the international statistical classification of diseases, injuries, and causes of death,
based on the recommendations of the ninth revision conference, 1975. Geneva: World Health Organization; 1977.
Sorlie PD, Johnson NJ. Validity of education information on
the death certificate. Epidemiology 1996;7:437–9.
Abu-Shakra M, Urowitz MB, Gladman DD, Gough J. Mortality
studies in systemic lupus erythematosus: results from a single
35.
36.
37.
38.
39.
40.
41.
center. II. Predictor variables for mortality. J Rheumatol 1995;
22:1265–70.
Ward MM, Pyun E, Studenski S. Mortality risks associated
with specific clinical manifestations of systemic lupus erythematosus. Arch Intern Med 1996;156:1337– 44.
Jacobsen S, Petersen J, Ullman S, Junker P, Voss A, Rasmussen JM, et al. A multicenter study of 513 Danish patients with
systemic lupus erythematosus. II. Disease mortality and clinical factors of prognostic value. Clin Rheumatol 1998;17:478 –
84.
Cervera R, Khamashta MA, Font J, Sebastiani GD, Gil A,
Lavilla P, et al. Morbidity and mortality in systemic lupus
erythematosus during a 5-year period: a multicenter prospective study of 1000 patients. Medicine (Baltimore) 1999;78:
167–75.
United States Census Web site. Accessed August 2, 2002.
URL: www.census.gov/prod/cen1990/cp3/cp-3-4.pdf.
U.S. Department of Health and Human Services, National
Center for Health Statistics. Technical appendix for vital statistics of the United States, 1995 mortality. Hyattsville (MD):
U.S. Department of Health and Human Services, National
Center for Health Statistics; 1999.
Jacobsen S, Petersen J, Ullman S, Junker P, Voss A, Rasmussen JM, et al. Mortality and causes of death of 513 Danish
patients with systemic lupus erythematosus. Scand J Rheumatol 1999;28:75– 80.
Manger K, Manger B, Repp R, Geisselbrecht M, Geiger A,
Pfahlberg A, et al. Definition of risk factors for death, end
stage renal disease, and thromboembolic events in a monocentric cohort of 338 patients with systemic lupus erythematosus. Ann Rheum Dis 2002;61:1065–70.
Документ
Категория
Без категории
Просмотров
0
Размер файла
173 Кб
Теги
death, level, ethnic, minorities, low, due, systemic, erythematosus, sle, education, lupus, underascertainment, sleevidence, mortality
1/--страниц
Пожаловаться на содержимое документа