1925
Intermittent Claudication, Heart Disease Risk
Factors, and Mortality
The Whitehall Study
George Davey Smith, MB, MSc, MA, Martin J. Shipley, BA, MSc, and Geoffrey Rose, DM
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
In the WVhitehall study, 18,388 subjects aged 40-64 years completed a questionnaire on
intermittent claudication. Of these subjects, 0.8% (147) and 1% (175) were deemed to have
probable intermittent claudication and possible intermittent claudication, respectively. Within
the 17-year follow-up period, 38% and 40% of the probable and possible cases, respectively,
died. Compared with subjects without claudication, the probable cases suffered increased
mortality rates due to coronary heart disease and cerebrovascular disease, but the mortality
rate due to noncardiovascular causes was not increased. Possible cases demonstrated increased
mortality rates due to cardiovascular and noncardiovascular causes. This difference in
mortality pattern may be due to chance. Possible and probable cases still showed increased
cardiovascular and all-cause mortality rates after adjusting for coronary risk factors (cardiac
ischemia at baseline, systolic blood pressure, plasma cholesterol concentration, smoking
behavior, employment grade, and degree of glucose intolerance). Intermittent claudication is
independently related to increased mortality rates. It is not a rare condition, and simple
questionnaires exist for its detection. The latter can be usefully incorporated in cardiovascular
risk assessment and screening programs. (Circulation 1990;82:1925-1931)
ollow-up studies of patients with intermittent
claudication reveal a higher mortality rate
than that in the general population.1-5 Such
patients have a twofold increase in age-specific risk of
death and a loss of 10 years in life expectancy.
Premature cardiovascular disease accounts for most
or all of this increased mortality.
The prevalence of intermittent claudication is increased in people with evidence of other forms of
coronary artery disease, for example, angina, previous myocardial infarction, or electrocardiographic
ischemic changes.6-8 Risk factors for coronary artery
disease are related to the development of intermittent claudication9'10 and are elevated in patients with
intermittent claudication.7"1112 The increased prevalence of other forms of coronary artery disease and
the coexisting unfavorable profile of risk factors may
explain the increased cardiovascular mortality rates
in subjects with intermittent claudication. Previous
F
studies suggest that intermittent claudication uncom-
plicated by evidence of other ischemia is
a
benign
From the Department of Epidemiology and Population Sciences, London School of
Hygiene and Tropical Medicine, Keppel
Street, London.
Address for correspondence: George Davey Smith, MB, MSc,
MA, Department of Epidemiology and Population Sciences, London School of Hygiene and Tropical Medicine, Keppel Street,
London WC1E 7HT, England.
Received November 28, 1989; revision accepted July 17, 1990.
condition,8,13 and after adjusting for existing disease
and other cardiovascular risk factors, there may be
no increased risk of death.7 These studies have been
relatively small, and their power to detect independent effects has not been great. The Whitehall study
See p 2246
of London civil servants14 provides an opportunity to
examine the prognostic importance of intermittent
claudication in a large cohort with long-term follow-up and to determine whether the elevated mortality risk is solely due to the associated ischemia and
coronary artery disease risk factors.
A subsidiary objective of this report concerns the
criteria required for the attribution of intermittent
claudication in questionnaire-based studies. The
Whitehall study used the questionnaire developed by
the London School of Hygiene and Tropical Medicine (LSHTM).15-17 We have, therefore, examined
the effect of varying the criteria for assessing intermittent claudication18 on associations with other risk
factors and on mortality rates.
Methods
In the Whitehall study, 18,403 men aged 40-64
years were examined between 1967 and 1969. Clinical
measurements included height, weight, blood pressure, forced expiratory volume in 1 second (FEV1 0),
1926
Circulation Vol 82, No 6, December 1990
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
forced vital capacity (FVC), and a limb lead electrocardiogram. Subjects were studied the morning after
an overnight fast; 50 g oral glucose was administered;
and 2 hours after this, a capillary blood sample was
drawn for the measurement of glucose and cholesterol concentrations. A questionnaire on age, civil
service employment grade, and smoking habits was
completed. Full details of procedures used have been
previously reported.14
The electrocardiogram was coded according to the
Minnesota system19 and was regarded as positive for
ischemia if Q/QS items (codes 1.1-3), ST/T items
(codes 4.1-4 or 5.1-3), or left bundle branch block
(code 7.1) was present. A regression analysis of
FEV1.o and FVC against height was performed, and
the coefficients were used to adjust the measurements to a uniform height of 175 cm. Subjects with a
plasma glucose concentration of 11.1 mmol/l or more
(.200 mg/100 ml) or with previously diagnosed diabetes constituted the diabetic group; nondiabetic
subjects with glucose concentrations greater than the
95th centile point (5.4-11.0 mmol/l, 96-199 mg/100
ml) formed the group with impaired glucose tolerance, and other subjects were designated as being
normoglycemic. Civil service employment grade is in
four levels: administrators, professionals and executives, clerical, and other (mainly unskilled manual)
grades. "Low work grade" refers to the clerical and
other grades. For 873 subjects from the Diplomatic
Service and British Council, employment grade was
not comparable to the rest of the sample. These
subjects have been categorized as a separate group in
the analyses that involve grade. Subjects who smoke
cigarettes have been categorized as "current smoker," "exsmoker," and "never smoker." In addition,
adjustment for smoking habits has included a term
for the number of cigarettes per day smoked by
current smokers. The 640 men who smoked pipes or
cigars only have been categorized as a separate group
in the analyses that involve smoking status.
The LSHTM chest pain and intermittent claudication questionnaire was administered by an interviewer to 938 men and was self-administered by the
other subjects. The prevalence of positive response to
the intermittent claudication section was similar in
these two groups.17 Data regarding angina and possible myocardial infarction have been taken from this
questionnaire for the present analysis.
To be classified as having intermittent claudication, subjects had to report that they developed calf
pain while walking, that they had not developed such
pain when standing still or sitting, that they stopped
walking or slowed down when pain developed, and
that the pain had then usually disappeared within 10
minutes. In addition, if subjects reported that the
pain had never disappeared while walking, they were
classified as having "probable" intermittent claudication. If they reported that the pain had disappeared
while they were walking (which may have been at a
slow pace), they were classified as having "possible"
intermittent claudication. In previous reports from
this study,14,17 only probable intermittent claudication has been discussed; however, in the present
report we have studied both categories.
Fifteen subjects failed to complete the intermittent
claudication questionnaire and have been excluded
from all analyses. In addition, for other variables,
data were missing for the following number of subjects: blood pressure, five; plasma cholesterol, 684;
glucose tolerance, 129; lung function, 18; body mass
index, three; smoking status, seven; angina, five;
possible myocardial infarction, 76; and electrocardiogram, 78. Subjects were only excluded from analyses
for which they were missing specific data. When full
adjustment for all risk factors was performed, 17,592
participants were included in the analyses.
Records from more than 99% of subjects were
flagged at the National Health Service Central Registry. Death certificates were coded according to the
eighth revision of the Intemational Classification of
Diseases (ICD), and this almost-complete mortality
follow-up to January 31, 1985 provides the basis for
this analysis. Death has been classified as being due
to coronary heart disease (ICD codes 410-414),
cerebrovascular disease (ICD codes 430-438), cardiovascular disease (ICD codes 390-458), lung cancer (ICD code 162), or any neoplasm (ICD codes
140-239). All-cause and noncardiovascular mortality
have also been examined.
Mortality rates have been calculated using personyears at risk. These rates and also all means and
proportions have been standardized for age by the
direct method, using the total population as the
standard. Tests of significance and confidence intervals for the age-adjusted rate ratios have been calculated by fitting models using the statistical package
GLIM.20,21 Adjustment for other major risk factors and
calculation of confidence intervals was done using
Cox's proportional hazards regression model.22
Results
Probable and possible cases of intermittent claudication were reported by 0.8% (147) and 1% (175) of
subjects, respectively. Each rate more than doubled
from the age groups 40-44 to 60-64 years. The
relations between reported symptoms and coronary
heart disease risk factors are shown in Tables 1 and
2, which show slightly elevated average levels of
plasma cholesterol concentration and body mass
index and show lower average ventilatory measures.
There are no significant relations with blood pressure
or glucose intolerance. Smoking is more prevalent
among subjects reporting intermittent claudication,
as are angina and possible myocardial infarction
according to questionnaire reports but not ischemia
according to electrocardiography. The prevalence of
any suspect ischemia, that is, angina, possible myocardial infarction, or electorcardiographic evidence
of ischemia, is raised in both intermittent claudication groups. Any suspect ischemia is more prevalent
among probable than possible cases; however, this
latter difference is not significant (p=0.12). Analysis
Davey Smith et al Intermittent Claudication and Mortality
1927
TABLE 1. Mean and SEM of Age-Adjusted Risk Factors in Men With Probable, Possible, or No Intermittent Claudication
No
(n=18,066)
Intermittent claudication
Possible
(n = 175)
Mean
SEM
137.5
1.81
Probable
(n = 147)
Risk factor
Mean
SEM
Mean
SEM
Systolic blood pressure (mm Hg)
135.9
0.15
136.1
1.59
Diastolic blood pressure
(mm Hg)
84.5
0.10
83.5
1.07
84.9
1.14
Plasma cholesterol concentration
(mmol/l)
5.11
0.01
5.27
0.11
0.10
5.34*
Body mass index (kg/M2)
24.7
0.02
25.3*
0.23
25.2
0.24
3.13
0.004
3.02*
0.042
0.044
3.00t
FEV1.0 (1)
0.005
FVC (1)
4.03
3.84t
0.050
3.84t
0.059
FEV1o0, forced expiratory volume in 1 second; FVC, forced vital capacity.
Test for differences between possible or probable intermittent claudication and no intermittent claudication groups: *p<0.05, tp<0.01,
tp<0.001.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
of covariance revealed that the poorer ventilatory
function shown by probable and possible cases is not
solely due to differences in smoking behavior.
Of the participants with intermittent claudication,
14 (8%) of the possible and 15 (10%) of the probable
cases were under treatment by their physicians at the
time of the examination.
Age-adjusted mortality rates for coronary heart
disease, stroke, noncardiovascular disease, lung cancer, all neoplasms, and all causes are shown in Table
3. Cardiovascular mortality is much higher among
those with both probable and possible intermittent
claudication. For possible, but not probable, cases of
intermittent claudication, mortality rates due to noncardiovascular causes are also elevated. The largest
difference in mortality rates between probable and
possible cases is the higher rate due to noncardiovascular causes in the latter (p=0.02).
Table 3 also shows that the proportions of deaths
due to cardiovascular or noncardiovascular causes
among the possible cases of intermittent claudication
is little different from the proportions shown by the
nonintermittent claudication group (X2=1.16,
p>0.2). However, for the probable cases, there is a
clear increased likelihood of deaths due to cardiovascular causes (x2=1755, pp<.00Ol).
To examine whether the increased mortality rates
among cases reflected only a short-term risk, we
analyzed the data excluding deaths within 5 years of
examination. For both the probable and the possible
intermittent claudication groups, mortality rates due
to all causes and cardiovascular causes were elevated
by factors of approximately two and three times,
respectively. For the possible, but not the probable,
cases, the mortality rate due to noncardiovascular
causes was elevated.
The relative mortality rates throughout the whole
follow-up period for the probable and possible intermittent claudication groups are given in Table 4. The
increases could be due to the higher prevalence of
cardiac ischemia or to elevated coronary heart disease
risk factors. Relative mortality rates adjusted for systolic blood pressure, plasma cholesterol concentration, smoking habits (including number of cigarettes
TABLE 2. Prevalence of Age-Adjusted Risk Factors in Men With Probable, Possible, or No Intermittent Claudication
Intermittent claudication
Possible
Probable
n
n
n
Risk factor
%
%
%
p
p
1
1
93.3
94.2
139
89.6
154
Normoglycemic
16,745
, 0.09
.
976
8.7
17
4
0.17
5.5
3.1
Glucose intolerant
4
1.2
217
1.7
2.7
3
J
Diabetic
J
1
1
19
Never smokers
25
19.5
16.0
16.0
3,409
.
.
44
37.9
26.3
50
0.004
30.5
0.009
Exsmokers
6,595
42.6
57.7
96
J
53.5
81
Current smokers
J
7,422
24
27
<0.001
4.6
837
12.9
<0.001
16.5
Angina
20.1
30
<0.001
Possible myocardial infarction
6.6
14.0
25
<0.001
1,181
4.8
9
>0.5
6.3
5.8
13
>0.5
Abnormal electrocardiogram
1,127
52
<0.001
<0.001
34.0
25.0
48
15.1
Any suspect ischemia
2,696
0.07
57
0.11
29.7
50
30.4
Low work grade
24.8
4,278
in
p indicates whether the proportions in the possible or probable intermittent claudication groups are different from those the group with
no intermittent claudication.
No
1928
Circulation Vol 82, No 6, December 1990
TABLE 3. Age-Adjusted Mortality Rates and Number of Deaths in Men With Probable, Possible, or No Intermittent Claudication
Intermittent claudication
No
Possible
Probable
Cause of death
Rate
n
Percent
Rate
n
Percent
Rate
n
Percent
5.1
Coronary heart disease
39
13.3*
32
1,326
46
16.1*
35
63
Cerebrovascular disease
0.8
204
6
6
2.5t
9
5
2.3t
9
Cardiovascular disease
6.9
53
42
60
17.1*
20.6*
45
1,788
80
Noncardiovascular disease
6.1
47
11.0:
1,600
28
40
11
4.2
20
Lung cancer
1.5
380
11
8
11
2
2.8t
0.8
4
All cancers
4.1
31
1,071
17
24
3.2
6.6t
8
14
All causes
13.0
100
70
3,401
28.1*
100
24.9*
56
100
Rates are for 1,000 person-years.
All causes mortality includes 13 deaths in the groups without intermittent claudication in which the specific cause of death was unknown.
Percent, percentage of deaths attributed to each cause or group of causes.
Tests for differences between possible or probable intermittent claudication and no intermittent claudication groups: *p<0.001; tp<0.05;
tp<0.01.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
smoked per day by current smokers), employment
grade, and degree of glucose intolerance have, therefore, been calculated for all subjects (Table 4) and for
the group with no evidence of other ischemia at baseline (Table 5). The same pattern of elevated mortality
rates is seen after these exclusions and adjustments.
For comparison, the adjusted relative rates for
subjects with angina compared with those without
angina were 3.01 for coronary heart disease and 1.86
for all-cause mortality. The corresponding values for
possible myocardial infarction were 2.57 and 1.71.
early retirement, and the consequent "healthy
worker effect" may explain the higher prevalences
seen in studies based on general populations.112327
However, several studies of working populations
have also revealed higher prevalence rates.7 25,26 The
issue is complicated by the fact that small changes in
the LSHTM questionnaire may produce large effects
on responses.16 Several of the studies referred to
above used translations of the questionnaire, which
may change its sense. Indeed, the agreement between probable intermittent claudication and physician diagnosis has varied from 30% in Sweden23 to
Discussion
The prevalence of probable intermittent claudication among these civil servants is lower than that seen
in most studies using the LSHTM questionnaire.7,1123-27 Intermittent claudication may lead to
more than 60% in France24 and Britain.'1 This might
contribute to the higher prevalence estimates in
Scandinavian studies.722527
The association between intermittent claudication
and other evidence of cardiovascular disease that was
TABLE 4. Relative Mortality Rates and 95% Confidence Intervals for Men With Probable or Possible Intermittent Claudication Versus No
Intermittent Claudication
Intermittent claudication
Possible
Cause of death
Coronary heart disease
Cerebrovascular disease
Cardiovascular disease
Noncardiovascular disease
Lung cancer
All cancers
All causes
Adjustments
Age only
Full
Age only
Full
Age only
Full
Age only
Full
Age only
Full
Age only
Full
Age only
Full
RR
2.54
2.12
3.00
2.44
2.46
2.05
1.83
1.70
2.07
1.73
1.66
1.58
2.15
1.86
Probable
95% CI
1.7,
1.4,
1.3,
1. 1,
1.8,
1.5,
1.3,
1.2,
1.0,
0.8.
1.0,
1.0,
1.7,
1.5,
3.6
3.0
6.8
5.6
3.3
2.8
2.7
2.5
4.2
3.7
2.7
2.6
2.7
2.4
RR
3.08
2.90
2.79
2.53
2.92
2.69
0.80
0.71
0.58
0.46
0.87
0.73
1.91
1.72
95% CI
2.2. 4.3
2.0, 4.1
1.1, 6.8
0.9, 7.3
2.2, 3.9
2.0, 3.7
0.4, 1.4
0.4, 1.3
0.1, 2.3
0.1, 1.8
0.4, 1.7
0.3, 1.5
1.5, 2.5
1.3, 2.3
RR, relative rates.
Full, fully adjusted relative rates are adjusted for age, systolic blood pressure, cholesterol, smoking habits, employment grade, and degree
of glucose intolerance.
Davey Smith et al Intermittent Claudication and Mortality
1929
TABLE 5. Relative Mortality Rates and 95% Confidence Intervals for Probable or Possible Intermittent Claudication
Versus No Intermittent Claudication in Men With No Suspected Ischemia at Baseline
Intermittent claudication
Probable
Possible
RR
95% CI
CI
95%
RR
Cause of death
1.8, 4.5
2.84
1.8, 4.5
2.82
Coronary heart disease
1.0, 10.5
3.31
2.1, 12.8
5.22
Cerebrovascular disease
1.7, 4.0
2.59
1.9, 4.2
2.81
Cardiovascular disease
0.2, 1.3
0.56
1.3, 3.2
2.06
Noncardiovascular disease
0.41
1.0, 5.2
2.31
0.1, 3.0
Lung cancer
1.8
0.2,
0.66
1.1, 3.2
1.87
All neoplasms
2.3
1.1,
1.55
1.8, 3.2
2.41
All causes
RR, relative rate.
Relative rates are adjusted for age, systolic blood pressure, cholesterol, smoking habits, employment grade, and
degree of glucose intolerance.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
shown in the present study is well established.6 8,13
The absence of a relation between intermittent claudication and electrocardiographic evidence of ischemia is surprising in the light of other studies, but it
is based on relatively small numbers of subjects with
both. Smoking has been consistently related to intermittent claudication in prevalence studies71' to a
similar degree shown here. This relation is less strong
than that seen in the clinical situation.
The relation between intermittent claudication and
coronary heart disease risk factors other than smoking
has been inconsistent in previous prevalence studies.
Plasma cholesterol was elevated in some,727 but not
all," studies. Similarly, blood pressure has been found
to be elevated"1,27 or else no higher.7 Negative findings
have been reported for body mass index27 and both
fasting glucose" and glucose tolerance.7 The present
results-small elevations in plasma cholesterol and
body mass index, and no significant relation with blood
pressure or glucose tolerance add to the inconsistent
set of previous findings.
The failure to detect a relation between claudication and diabetes may be a chance finding, or it may
reflect selection out of employment in the civil service by participants who were both diabetic and had
intermittent claudication. We cannot separate these
two possibilities.
In the Framingham study,9'10.3 the presence of
intermittent claudication was determined at each
biennial examination. Smoking, glucose intolerance,
and blood pressure were powerful independent predictors of intermittent claudication; serum cholesterol was a weak predictor, and relative weight was
inversely associated with occurrence of intermittent
claudication. The results from comparing the incidence data from Framingham with the data from
prevalence studies suggests that in the latter a selective survival of individuals at lower risk may have led
to underestimation of the importance of glucose
tolerance and blood pressure.
There are no large differences between the two
intermittent claudication groups in either their crosssectional relations or their association with cardio-
vascular mortality. This rather unexpected finding
suggests that the category of possible intermittent
claudication has acceptable validity. In previous studies, the percentages of deaths from cardiovascular
causes in subjects with intermittent claudication has
ranged from 60-65%4 to more than 80%.2,3 Thus, the
percentages of cardiovascular deaths in both the
probable intermittent claudication group and in the
possible intermittent claudication group lie within
the previous boundaries.
It appears, however, that whereas probable and
possible intermittent claudication cases have an increased risk of cardiovascular death, only possible
cases have an increased risk of dying from other
causes. This nonspecific mortality excess is little
changed by excluding deaths within 5 years of the
examination and, therefore, does not seem to be due
to the inclusion of subjects with severe illnesses in
this group. It could be that participants in the possible intermittent claudication group received lessaggressive treatment and risk factor intervention
than did the probable group, leading to elevated
rates of mortality from lung and other cancers among
the former. In fact, the possible group did have
marginally higher rates of smoking than did the
probable group. However, the finding that only a
small proportion of participants were under treatment by their physicians does not add support to the
notion that differing degrees of intervention lead to
the mortality pattern seen.
The difference in the criteria for classifying subjects as having possible or probable intermittent
claudication is slight: in the former, the leg pain may
have disappeared while subjects were walking; in the
latter, pain could not have disappeared. While this
small change approximately doubled the number of
positive responses, it probably could not select
groups of very different characteristics. We cannot
explain the difference; possibly, it was due to chance.
Data from other studies using the LSHTM questionnaire would help to answer this.
It has been reported that relaxing the criteria for
intermittent claudication when using the LSHTM
1930
Circulation Vol 82, No 6, December 1990
questionnaire leads to increased sensitivity and reduced specificity, when measured against clinical assessments of peripheral arterial disease.18 The group
selected with relaxed criteria would be expected to
contain more false-positive cases, that is, would contain a lower percentage of "real" cases and would,
thus, demonstrate lower mortality rates. In fact, the
possible intermittent claudication group did show a
lower mortality rate due to cardiovascular causes than
did the probable intermittent claudication group, but
this was more than compensated for by a higher
mortality rate due to noncardiovascular causes. It
could be that an effect of relaxing the criteria for
intermittent claudication was disguised by what may
have been a chance increase in noncardiovascular
mortality. Even if this is so, however, it is evident that
the "possible" group contains many genuine cases.
The doubling of the mortality rate in those with
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
intermittent claudication was similar to that shown in
other population-based studies using interview or
questionnaire methods of diagnosis.713 This elevation of mortality risk is less than that seen with
peripheral arterial disease defined by more stringent
criteria.428 The elevated mortality risk seen with
large vessel peripheral arterial disease appears to be
independent of coexisting cardiovascular disease and
risk factors,28 as was the elevated mortality risk seen
with intermittent claudication in the present study.
Questionnaire-defined intermittent claudication and
large-vessel peripheral arterial disease measured
through more detailed assessments clearly overlap to
a large degree18 and both appear to be related to
diffuse atherosclerosis, which leads to a high risk of
mortality.
Apparently, in population-based studies, intermittent claudication carries with it about the same risk
as does angina or possible myocardial infarction
assessed by questionnaire. Some, but not all, of this
increased risk can be attributed to the associated
levels of coronary risk factors. Intermittent claudication is not a rare disease; in men older than 60 years,
it is two to three times as prevalent as diabetes
mellitus.5 In the present sample, only 10% of subjects
with probable intermittent claudication were under
medical care for their condition. The LSHTM questionnaire can, therefore, identify a group of subjects,
many of whom are not receiving medical care, whose
increased risk of death may be partially preventable.
The questionnaire does not require a physician to
administer it, but it may be self-completed or -administered by an interviewer after brief training. Thus, it
could be usefully added to cardiovascular risk assessment and screening programs.
References
1. Jelnes R, Gaardsting 0, Hougaard Jensen K, Baekgaard N,
Tonnesen KH, Schroeder T: Fate in intermittent claudication:
Outcome and risk factors. Br Med J 1986;293:1137-1140
2. Bloor K: Natural history of arteriosclerosis of the lower
extremities. Am R Coll Surg Engl 1961;28:36-52
3. Hughson WG, Mann JI, Tibbs DJ, Woods HF, Walton 1:
Intermittent claudication: Factors determining outcome. Br
Med J 1978;1:377-379
4. Kallero KS: Mortality and morbidity in patients with intermittent claudication as defined by venous occlusion plethysmography: A ten-year follow-up study. J Chronic Dis 1981;34:
455-462
5. Dormandy J, Mahir M, Ascady G, Balsano F, De Leeuw P,
Blombery P, Bousser MG, Clement D, Coffman J, Deutshinoff
A, Bletry 0, Hampton J, Mahler F, Ohlin P, Rieger H,
Stranden E, Turpie AGG, Urai L, Verstraete M: Fate of the
patient with chronic leg ischaemia. J Cardiovasc Surg 1989;30:
50-57
6. Reid DD, Holland WW, Humerfelt S, Rose G: A cardiovascular survey of British postal workers. Lancet 1966;1:
614-618
7. Reunanen A, Takkunen H, Aromaa A: Prevalence of intermittent claudication and its effect on mortality. Acta Med
Scand 1982;211:249-256
8. Tillgren C: Obliterative arterial disease of the lower limbs: III.
Prognostic influence of concomitant coronary heart disease.
Acta Med Scand 1965;178:121-128
9. Kannel WB, Skinner JJ Jr, Schwartz M, Shurtleff D: Intermittent claudication: Incidence in the Framingham Study. Circulation 1970;41:875-883
10. Kannel WB, McGee DL: Update on some epidemiologic
features of intermittent claudication: The Framingham study.
JAm Geriatrics Soc 1985;33:13 -18
11. Hughson WG, Mann JI, Garrod A; Intermittent claudication;
prevalence and risk factors. Br Med J 1978;1:379-381
12. Juergens JL, Barker NW, Hines EA Jr: Arteriosclerosis obliterans: Review of 520 cases with special reference to pathogenic and prognostic factors. Circulation 1960;21:188-195
13. Kannel WB, Schurtleff D: The natural history of arteriosclerosis obliterans. Cardiovasc Clin 1971;3:37-52
14. Reid DD, Brett GZ, Hamilton PJS, Jarrett RJ, Keen H, Rose
G: Cardiorespiratory disease and diabetes among middle-aged
male civil servants. Lancet 1974;1:469-473
15. Rose GA: The diagnosis of ischaemic heart pain and intermittent claudication in field surveys: Bull WHO 1962;27:
645-658
16. Rose GA: Ischemic heart disease: Chest pain questionnaire.
Milbank Memorial Fund Q 1965;43:32-39
17. Rose G, McCartney P, Reid DD: Self-administration of a
questionnaire on chest pain and intermittent claudication. Br
JPrev Soc Med 1977;31:42-48
18. Criqui MH, Fronek A, Klauber MR, Barrett-Connor E, Gabriel S: The sensitivity, specificity and predictive value of
traditional clinical evaluation of peripheral arterial disease:
Results from noninvasive testing in a defined population.
Circulation 1985;71:516-522
19. Rose GA, Blackurn H: Cardiovascular Survey Methods.
Geneva, World Health Organization, 1968
20. Baker RJ, Nelder JA: The GLIM System Release 3; Generalised
Linear Interactive Modelling. Oxford, Numerical Algorithms
Group, 1978
21. Berry G: The analysis of mortality by the subject-years
method. Biometrics 1983;39:178-184
22. Cox DR: Regression models and life-tables. J R Stat Soc
1972;34(series B):187-220
23. Isacsson S: Venous occlusion plethysmography in 55-year old
men: A population study in Malmo, Sweden. Acta Med Scand
Suppl 1972;537
24. Richard JL, Ducimetiere P, Elgrishi I, Gelin J: Depistage par
questionnaire de l'insuffisance coronarienne et de la claudication intermittente. Rev Epidemiol Med Soc Sante Publique
1972;20:735-755
25. Gyntelberg F: Physical fitness and coronary heart disease:
Male residents in Copenhagen aged 40-59. Dan Med Bull
1973;1-4
26. De Backer IG, Kornitzer M, Sobolski J, Denolin H: Intermittent claudication-Epidemiology and natural history. Acta
Cardiol 1979;34:115-124
Davey Smith et al Intermittent Claudication and Mortality
27. Schroll M, Munck 0: Estimation of peripheral arteriosclerotic
disease by ankle blood pressures: Measurements in a population study of 60-year-old men and women. J Chronic Dis
1981;34:261-269
28. Criqui MH, Coughlin SS, Fronek A: Noninvasively diagnosed peripheral arterial disease as a predictor of mortality:
Results from
768-773
a
1931
prospective study. Circulation 1985;72:
KEY WORDS * questionnaire * peripheral vascular disease
epidemiology
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Intermittent claudication, heart disease risk factors, and mortality. The Whitehall Study.
G D Smith, M J Shipley and G Rose
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Circulation. 1990;82:1925-1931
doi: 10.1161/01.CIR.82.6.1925
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1990 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/82/6/1925
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/