Download The WHO Multinational Study of Vascular Disease

T
he WHO Multinational Study
of Vascular Disease in Diabetes:
1. General Description
R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
A general description of the multinational study under the auspices of the World Health Organization is
presented. The purpose was to compare the prevalence of vascular disease in representative samples of
diabetic subjects of different ethnic and cultural habitus. Standardized methods of investigation were
devised. Fourteen centers participated, and the data collected so far indicate that the characteristics of
the populations studied varied as to age composition, adiposity, cigarette smoking, treatment, age at
diagnosis, and duration of diabetes. Thus, taking these factors into consideration and while awaiting
completion of the data, conclusions must be drawn with reserve, DIABETES CARE2.- 175-186, MARCH-APRIL 1979.
R
eported differences in certified causes of death,
particularly coronary artery disease, between
Japanese and Western diabetic populations1
prompted a projected standardized comparison of
the prevalence of large vessel and microvascular disease in
Tokyo with that in London. This project was discussed at a
meeting of the European Diabetes Epidemiology Study
Group in Berne in 1972, in the presence of an invited representative of the Unit for Non-Communicable Diseases of the
World Health Organization, Dr. V. Khatchatourov. This
meeting and further discussion with interested national
groups eventually led to an extension of the projected study to
a multinational comparison under the auspices of WHO, 2
which provided facilities for central coordination for the
study, data collection and analysis, and support for meetings
of participants. Individual national centers sought and
received support from local sources to which acknowledgment will be fully expressed elsewhere. This preliminary
report of the study and some of its results is presented on
behalf of the multinational group of investigators, the individuals being listed in Appendix 1. Any comments or
views, however, are the responsibility of the authors, and
this presentation does not constitute an official publication or
document of the World Health Organization.
potential investigators at the CIBA Foundation, London, in
January 1974. The purpose of the study was to compare the
prevalence of vascular disease in representative samples of
diabetic subjects of different ethnic and cultural habitus and,
when necessary, to devise and evaluate standardized methods
of investigation.
STUDY PROTOCOL
MICROVASCULAR DISEASE
A draft protocol was prepared in Geneva by Drs. D. A.
Pyke, H. Keen, and R. J. Jarrett and finalized at a meeting of
For diabetic microvascular disease, specifically retinopathy
and nephropathy, no validated standardized methodology
ARTERIAL DISEASE
For large vessel disease, certain methods had already been
standardized and evaluated in nondiabetic populations. These
included the questionnaire for angina pectoris, history of
myocardial infarction and intermittent claudication,3 and the
Minnesota coding of electrocardiograms.4 The WHO questionnaire was incorporated unchanged into the Study Questionnaire, and appropriate translations were made where
none was already available. Twelve lead electrocardiograms
were included in the protocol, with arrangements for central
reading and coding of all ECGs by two independent observers,
differences being resolved by a single arbitrator. The other
centrally standardized measurements were of serum cholesterol
and creatinine by the WHO reference laboratories in Atlanta
and Moscow. Urinary protein was measured locally using a
standard semiquantitative salicylsulphonic acid technique
(see Appendix 2).
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
175
WHO STUDY OF VASCULAR DISEASE AND DIABETES, PART 1/R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
TABLE 1
Participating centers and state of completion at 25 September 1978 (i.e.
number of questionnaires coded on the Geneva computer)
Center
1.
3.
4.
5.
6.
7.
8.
9.
10.
11.
13.
14.
15.
17.
United Kingdom (UK)
Switzerland (SWI)
Belgium (BEL)
U.S.S.R. (USSR)
Poland (POL)
East Berlin (GDR)
Yugoslavia (YUG)
India (IND)
Hong Kong (HK)
Japan (JAP)
Cuba (CUB)
Oklahoma (OKL)
Pima (PHOE)
Bulgaria (BULG)
Male
Female
Total
254
278
157
110
241
285
222
263
198
235
255
262
25
196
243
256
113
143
245
275
180
237
224
200
260
373
54
497
534
270
253
486
560
402
500
422
435
515
635
79
418
222
suitable for comparative epidemiologic studies has yet been
devised. Direct ophthalmoscopy of the fundus through well
dilated pupils was the major procedure for identifying microangiopathy and clearly raises problems of observer variation.
Ideally, an independent team of observers should have
carried out eye examinations in all participating centers, but
this would have been prohibitively expensive in time and
money as well as raising considerable logistic problems. It was
agreed at the CIBA Foundation meeting that, to avoid
variable preconceptions about the descriptions "microaneurysm" and "hemorrhage," the simple descriptive term
"red lesion" would be used, and observers were requested to
describe them in terms of size and number (see Appendix 2).
To minimize timing bias by observers who might vary duration of examination according to expectation of abnormality,
the protocol specified a precise 2-min examination of each
eye, the observers being allowed additional time to count
but not to seek lesions. This system was rehearsed in a
selected group of diabetic patients at the 1974 planning
meeting and proved to be reasonably satisfactory. Standard
photographs illustrating individual fundus abnormalities were
later prepared and circulated to participants.
Other measurements made on each subject included
height, weight, and blood pressure, and the conditions for
measuring these were specified in the protocol (see
Appendix 2).
SAMPLING
Equal numbers were selected from each sex and the choice
of the age range for the study was determined by two factors.
As coronary heart disease is uncommon before the age of 40, it
MEN
100
80
60
WOMEN
40
20
JJR^&XWj&SSyiSil
I
I
1
20
40
60
80
100
20
40
60
80
100
BEL I
I
^J^^^^g^SfiiKcS&a 1 CUB I
100
80
60
40
20
WHO 10/-78
<20
20-39
40*
FIG. I. Distribution of age at diagnosis of diabetes in the national samples by sex. (See Table 1 for abbreviations.)
176
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
WHO STUDY OF VASCULAR DISEASE AND DIABETES, PART 1 /R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
perhaps over-represent patients with problems related to the
study endpoints. To avoid this bias, the sampling was carried
out from a complete list of all patients in the appropriate sex,
age, and duration ranges who had attended one or more times
in the year before the commencement of the study (see
Appendix 2).
was necessary to have a lower age limit which was set at 35 yr.
It was also decided to set an upper age limit (of 55 yr) since
the prevalence in older age groups might be distorted by
"premature" mortality. Microvascular disease is predominantly related to duration of diabetes, so the sampling
procedure was devised to provide a range of duration within
the selected age groups (see Appendix 2).
The other potential problem that affected sampling was
related to the necessity for deriving the study subjects from
diabetes clinics. It was decided that, to avoid the problem of
defining diabetes in terms of glucose tolerance testing, no
population screening should be undertaken to recruit subjects
for the study and that, for the purposes of the study, diabetic
subjects should be defined as patients under treatment for the
disorder at a defined date. It was recognized that some
clinics might derive a substantial number of patients referred
because of complications or from the screening of special
groups, such as survivors of a myocardial infarction. These
possible sources of bias have been avoided and allowed for, as
far as possible, by selecting from broadly based clinical groups
and by the inclusion of specific questions concerning mode of
discovery of diabetes and previous medical history in the
questionnaire.
Casual acquisition of subjects would be likely to give selection advantage to frequent attenders to the clinic, and thus
PARTICIPANTS AND TIMING
F
ourteen centers (see Appendix 1) have participated
in the study and, by September 1978, most had
completed their data collection (Table 1). The
groups represented include a diverse geographical,
ethnic, and cultural mix, though we were unfortunately
unable to find a black African center both willing and able to
participate. Nevertheless, we have two Asian groups—Hong
Kong and Tokyo—representing countries where the population frequency of atherosclerotic disease is reputedly low.
Time taken to completion—just over 3 yr—has been
approximately 18 months longer than originally anticipated.
A number of factors have contributed to this, one of the more
important being the difficulties encountered by many groups
in recruiting sufficient medium- and long-term diabetic
subjects in the younger age groups. Despite intensive efforts
to supplement the original sampling population, several
WOMEN
MEN
00
80
40
60
\
i
1
•niiiiiniiu
0
20
!
UK
1
1
|
)
1
1
• - : . : "a
1
•IIIIIIIIIIIIKCKQ
1
[
BEL
f
1
i
1
1
i
1
1
1
1
j
1
i
iiiiiiiiiiiiiiiiii lillIIIIIH
mini!
00
i
I
80
•
1
60
I
:l
GDR
f"
j
YUG
:\
IND
i
HK
1
1
1
1
1
1
1
1
' "1
'. i
CUB
'
80
!
'
100
'
m&mv\UUIIIIIIIIIIIIIIIHH
1
IS^e^^SS^^IIIIIIIIIi
1
1
•»^^v!iiiiiiiiiiiiiiiiiin
i
1
:
1
i
. ..
I,' ",
I '
L, '
j
1
f
20
0
0
0-9
10-19
1
•
<
Ki«&%VXV'ft?&lill
.ftWuVt'fwSft^JIII
i
i
1
. '." r^^TKssasssiniiiiiB
I
1
-
•
•
i
•
20-29
1 1 1
i
^-
Milllllli iili
1
'.
1...'
1
. • : " ; . : ] BULG
%
i
1
1
I'1," '
1
i
tt^Kxm^>iiiiiiiiiiiiiiiiin
.K^^AVSSSJSBIIIIIIIIIIIH
OKL
K.VVSS'.UJNI
40
1
t ,
1
IIIIIK^W'AV?
i
POL
i JAP
Bllllllll£ra3&)
1
60
!
USSR ['
1
.-•.
1
I
iiiiirara -&}» « S K « ! l . " . . • •
1
'
1
"&£&&>.
T
'
1
;
i
40
1
•r
•IIIIIIIIIIIKMM&Y&3M
i
t'. •'
SWI
•iiiiiiiiniiiiK
lIHIIIlRSJs?'?^
20
i
il.. .'r .
tm$&:
0
%
20
'
1
Jf/Vi'V.>>'i?(v?'txJ^V,W^/«\^lllllll
1
i
I
4C
60
.
i
80
i
i
100
WHO 10/'78
30+
FIG. 2. Distribution of duration of diabetes from diagnosis at time of W H O examination, by sex. (See Table I for abbreviations.)
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
177
WHO STUDY OF VASCULAR DISEASE AND DIABETES, PART 1 /R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
groups, in particular Hong Kong and Tokyo, have been
forced to conclude their investigations without being able to
reach the planned number of patients in these categories.
TABLE 2
Frequency of regular cigarette smoking
Center
CORE AND NON-CORE INVESTIGATIONS
All participants in the study undertook to complete for
each patient a basic set of data—the Core Study—which has
been outlined above and is dealt with in more detail in
Appendix 2. Participating groups were also encouraged to add
non-core investigations either on a unilateral basis or as part
of joint enterprises. In particular, they were encouraged to
organize a comparison, especially in relation to coronary heart
disease prevalence, between the diabetic sample and a local
control, nondiabetic population. In addition, participants
were asked to consider the possibility of extending the
prevalence study in their diabetic sample into a prospective
study of morbidity and/or mortality. Non-core investigations
have been carried out in several centers but are outside the
remit of the WHO-sponsored investigation and will be
reported separately.
UK
SW1
BEL
USSR
POL
GDR
YUG
IND
HK
JAP
CUB
OKL
PHOE
BULG
Men
(%)
Women
(%)
37.8
33.1
30.7
29.1
50.2
42.5
43.7
24.0
51.5
63.0
67.8
46.9
44.0
32.1
30.4
20.3
15.0
7.7
20.4
26.5
10.6
0
11.6
16.5
31.5
32.7
7.4
2.7
See Table 1 for abbreviations.
duration exceeding 20 yr and, in particular, 30 yr. Only Cuba
among the non-Europeans has a similar spread of duration
DIFFERENCES BETWEEN POPULATIONS
groups.
These differences between the groups represented in the
Apart from the ethnic differences, there are several other
WHO
study must be borne in mind in the analysis and
notable differences in the characteristics of the populations
interpretation
of the variables and endpoints to be discussed.
that will have to be taken into account in the analysis. These
Until
data
are
completed for all centers and more sophisinclude age composition, adiposity, cigarette smoking habit,
ticated
analyses
have been performed, interpretations must
type of treatment, age at diagnosis, and duration of diabetes.
be
made
and
conclusions
drawn with reserve.
Age at diagnosis (Figure I). The notable finding illustrated
is the paucity or absence of subjects diagnosed below the
age of 20 yr in the three Asian centers, in Oklahoma, and TABLE 3
Type of treatment
among women in several other centers.
Body mass index (BMl) (Figure 2). This is directly related to
Men
adiposity. Here the Asian men and the Japanese women
are notable for their low BMI. By contrast, the two American
Insulin
Oral
injections
agents
Indian groups and, to a lesser extent, the Russian and Polish
(%)
(%)
women, are notable for their subjects with BMI greater than Center
36, an index of considerable corpulence.
28.3
UK
56.3
Cigarette smoking (Table 2). The prevalence of current SW1
38.5
51.8
cigarette smokers differs widely between centers and, in some BEL
24.2
51.0
centers, between the sexes. Thus, no Indian lady admitted to USSR
30.9
59.1
38.2
smoking in comparison with the 32.7% Oklahomans who did. POL
58.1
30.2
42.5
Type of treatment (Table 3). The proportion of people GDR
40.5
YUG
37.4
receiving insulin varies from as little as 14.3% of Indian
66.9
23.2
women to 61.3% of U.K. women. There is a similar wide IND
59.6
38.4
variation in the use of oral agents, with Indian men and HK
43.0
29.8
JAP
women having the highest proportion.
29.0
55.3
CUB
Duration of diabetes (Figure 3). Although the sampling OKL
54.2
21.0
method was intended to produce similar distributions of PHOE
36.0
28.0
diabetes duration, the figure clearly demonstrates that the BULG
44.9
46.9
methodology was only partly successful. The European
centers generally have substantially greater numbers with See Table 1 for abbreviations.
178
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
Women
Diet
only
(%)
Insulin
injections
(%)
Oral
agents
(%)
Diet
only
(%)
15.4
9.7
24.8
10.0
3.7
27.4
22.1
9.9
2.0
27.2
15.7
24.8
36.0
8.2
61.3
60.5
46.0
51.0
60.8
50.2
33.3
14.3
29,5
32.5
24.2
19.8
29.6
55.9
30.5
32.0
25.7
42.0
31.8
26.2
45.0
73.0
69.2
49.0
64.2
52.0
33.3
39.2
8.2
7.4
28.3
7.0
7.3
23.6
21.7
12.7
1.3
18.5
11.5
28.2
37.0
5.0
WHO STUDY OF VASCULAR DISEASE AND DIABETES, PART 1/R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
MEN
100
80
60
40
100
JliUHi 11111111111
1/M-.-illlllliii
IKta
Illllllllllllllll
iiiiiiiimiii
!••
iiiiiiiiiiiiiiiii IIIIIIIIIIIIHM
BEaBsa_:
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIH
WEBBT
-Y:,;.V-/^;>V|||||||||IIIIII
[mill
^W^lllllllllllllllll
^'Jlllllllllllll IIIIHIIIIIIIIIIIIII H I "
;
i
PHOE ||&fe #;
100
80
60
20
40
60
80
100
WHO 17/-78
?
<22
22-27
28-35
36*
FIG. 3. Distribution of body mass index (BMl) —weight in kilograms/square of the height in meters—at time of examination, by sex. (See Table I
for abbreviations.)
From the Department of Community Medicine and the Unit for
Metabolic Medicine, Guy's Hospital, London SE1 9RT, England,
and the Division of Non-Communicable Diseases, World Health
Organisation, Geneva, Switzerland.
Address reprint requests to H. Keen, Unit for Metabolic
Medicine, Guy's Hospital, London S.E.I, 9RT U.K.
REFERENCES
1
Kuzuya, N., and Kosaka, K.: Diabetes in Japan. In Diabetes
Mellitus in Asia. ExcerptaMed. Int. Congr. Ser. No. 221: 11, 1971.
2
Jarrett, R. J.: The WHO multinational study of vascular disease
in diabetics. IDF Bull. 19 (2): 4, 1974.
3
Rose, G., and Blackburn, H.: Cardiovascular Survey Methods,
WHO Monograph Ser. No. 56, WHO Geneva, 1968.
4
Blackburn, H., Keys, A., Simonson, E., Rautaharju, P., and
Punser, S.: The electrocardiogram in population studies: a classification system. Circulation 2 J: 1160-1175, 1960.
APPENDIX 1
List of Centers and Investigators
Coordinating center: WHO, Geneva
Dr. V. Khatchatourov, Dr. V. Grabauskas, Dr. B. Grab
Center 01: London, United Kingdom
Dr. R. J. Jarrett, Prof. H. Keen, Dr. P. J. Watkins,
Dr. R. B. Smith, Dr. J. H. Fuller
Center 03: Berne, Switzerland
Dr. A. Teuscher and physicians
Center 04: Brussels, Belgium
Dr. J. Pirart, Dr. Golard, Prof. Englert, Dr. Dineur
Center 05: Moscow, U.S.S.R.
Dr. A. Mazovetsky, Dr. L. Slavina, Dr. G. Klushina,
Dr. T. Zukashina, Dr. Z. Dudnikova, Dr. S. Arapova,
Dr. P. Zavadsky
Center 06: Warsaw, Poland
Prof. A. Czyzyk, Dr. A. S. Krolewski, Asst. Prof. J.
Kopczynski, Dr. D. Janeczko, Dr. B. Puncewicz,
Dr. A. Ostrowska, Dr. M. Karpowicz, Dr. O. Malinowska,
Dr. J. Kodejszko, Dr. J. Mianowicz, Dr. J. Gorecki
Center 07: Berlin, GDR
Dr. V. Schliack, Dr. Gorr, Dr. Raskovitsch
Center 08: Zagreb, Yugoslavia
Dr. I. Aganovic, Prof. Z. Skrabalo, Dr. A. Stayljenic,
Dr. M. Ugrinovic, Dr. N. Raic
Center 09: New Delhi, India
Prof. M. M. S. Ahuja
Center 10: Hong Kong
Prof. R. T. T. Yeung, Dr. L. K. F. Chan, P. K. Lee,
Dr. C. K. Yeung
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
179
WHO STUDY OF VASCULAR DISEASE AND DIABETES, PART 1/R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
Center 11: Tokyo, Japan
Prof. K. Kosaka, Dr. E. Miki, Dr. N. Kuzuya, Dr. K.
Yoshizawa, Prof. T. Mizuno, Dr. T. Motegi, Dr. K. Ooto,
Dr. E. Takatori, Asst. Prof. M. Fukuda
Center 13: Havana, Cuba
Dr. O. Mateo de Acosta, Dr. S. Amaro, Dr. O. Diaz,
Dr. M. Hang, Dr. X. Quesada
Center 14: Oklahoma Indians, U.S.A.
Dr. K. M. West, Dr. T. C. Coniglione, Dr. M. E. Sanders
Center 15: Pima Indians, U.S.A.
Dr. P. H. Bennett, Dr. P. J. Savage, Dr. Carryl Webner
Center 17: Sofia, Bulgaria
Prof. D. Andreev, Dr. W. Denev, Dr. O. Zlatarev,
Dr. L. Pompulov, Dr. K. Polonov, Dr. I. I. Pencev,
Dr. N. Veleva, Dr. B. Osunova, Dr. L. Pejceva
Notes
For local reasons, certain of the collaborating centers departed
from the protocol in assembling their populations or in data
collection from them. These departures from protocol were considered permissible in that they would not seriously bias comparisons or conclusions.
The Swiss sample was selected from a central register constructed from a compilation of diabetic patients under the care
of physicians in the area of Berne. Physicians of the patients
selected randomly from the central register each completed the
WHO history and examination document. In Belgium, patients
were drawn from the records of a single physician (J.P.) who had
maintained careful clinical data on these for many years; when
Selected,patients were unavailable, data on clinical symptomatology
were taken from the records, making this the only center where the
data are not completed at face-to-face interview. The symptomatology data for this group are thus not entirely comparable with
other centers. The GDR sample probably contains a higher
proportion of subjects discovered in systematic diabetes detection
surveys, which, for many years, have been a health-care feature of
this area. India, Hong Kong, and Japan found particular difficulty
in filling the youngest age/longest duration cells and were permitted
to extend recruitment of these patients outside the original
defined clinic populations. The Cuban sample was selected from a
population living in a denned area of Havana, Cuba, rather than
recruited from a clinic clientele.
APPENDIX 2
(Condensed from the Study Protocols)
(a) Sample Size
This has been determined by the need to show statistically
significant differences of >33% and is estimated to be 250
men and 250 women in the stated age range. It follows that
smaller samples cannot be included in the main study,
although it is possible that certain comparisons between
groups using smaller samples may be possible outside the
general auspices of the main investigation.
180
(b) Sample Composition
As far as possible the group from which the sample is drawn
should be representative of all diabetic individuals in that
country. In countries where there are small minorities of
different ethnic composition or with radically different
cultural characteristics, individuals belonging to such groups
should be excluded. These are the only grounds for exclusion
from the total group before sampling and, if this proves to be
necessary, the grounds for exclusion should be reported
to WHO.
(i) Representativity. It is clearly vital that biases for or
against vascular disease in the selection of the samples should
be minimized and measurable. For example, frequent clinic
attenders, problem cases, and in-patients would be overrepresented unless special precautions are taken. Furthermore, clinics which attract their diabetic patients because of
special reasons, such as an acknowledged interest in diabetic
retinopathy or some other complication of diabetes, are
clearly unable to provide a representative sample. The same
applies to clinics whose policy is to refer for treatment elsewhere their patients with "mild" diabetes. Groups wishing to
participate, therefore, should be prepared to establish the
representative nature of the diabetic population from which
they would sample.
(ii) Sampling procedure. The suggested procedure is to make
a complete listing of all patients in the age range 35-54 yr
attending the clinic/center over a time period long enough
to include the least frequently seen patients. In most centers,
it seems likely that 1 yr should suffice. These patients should
be further stratified into the age categories 35-41, 42—48,
and 49-54, and into known duration of diabetes: 1-6 yr,
7-13 yr, and 14 or more years. For each sex, there will,
therefore, be nine cells—three age groups subdivided into
three duration groups. For the purposes of the study, 28
patients are required in each cell. As it is unlikely that all
patients selected will actually be examined, 35 patients should
be selected for each cell using a random method of selection
from all patients eligible for that particular cell. It is
possible that certain cells will be difficult to fill. If this is the
case, WHO should be consulted before changing the basis or
source of recruitment.
(c) Clinical Procedures
The completion of the questionnaire and examination will
take longer than is usually possible in a routine diabetic
clinic. In order to maintain the quality of observations for
this study, two possible arrangements are suggested. Either a
special clinic session is dedicated to the study or one or more
doctors are designated to do only this work within an ordinary
clinic session.
(ii) In any multicenter study or trial, observer variation
is a major problem. To minimize this, the following procedures are proposed:
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
WHO STUDY OF VASCULAR DISEASE AND DIABETES, PART 1 /R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
Within each center a limited number of named doctors
should be responsible for all Study procedures. These
doctors should expect to spend at least 1 yr at the Study
center and should be explicitly trained in the standard
Study procedures before participating. The principal
investigators will be responsible for ensuring that these
standards are observed in their own center and for training
observers. For the ophthalmoscopic examination, observers must be both experienced and competent but need
not necessarily be specialist ophthalmologists.
INSTRUCTIONS FOR COMPLETION OF
QUESTIONNAIRE AND EXAMINATION METHODS
Previous Medical History
Answers to some of these questions will depend to a large
extent on existing records. If these are inadequate to provide
a precise answer, record as "not known."
Present Treatment
13. Diet. Information is sought concerning the kind of
diet prescribed for the patient. This may well be different
from the diet actually consumed.
Three kinds of dietary prescription are assumed: none;
advice to restrict a given type of food, without specifying an
actual amount; and an actual specification of an amount or a
maximum amount of a given type of food.
Interview
18-20. These questions are taken from "Cardiovascular
Survey Methods" by G. A. Rose and H. Blackburn, WHO
Monograph Series No. 56, Geneva, 1968 (Annex 6, pages
172-174). Problems in the application of these questions are
discussed in pages 101-128 of "Training in Cardiovascular
Epidemiology" by G. A. Rose and J. R. T. Colley, distributed
by the Regional Office for Europe, WHO Copenhagen. This
part of the questionnaire has been extensively used in
cooperative international studies and has been standardized
and validated. It is important that the questions be asked
as printed. The use of specified supplementary questions
designed to define patient's replies and which can only be
used in specified circumstances is explained in the training
manual (Rose and Colley, above), and these specifications
must also be rigorously adhered to.
2 i. History of ' 'stroke.'' Ask patient if he or she has had a
stroke, or use whatever is the lay term in local use.
22. Visual disability. The question asks for the patient's
own assessment of whether or not there is a visual disability
which is not corrected by wearing appropriate spectacles or
contact lenses. If the patient admits to such a disability, the
doctor makes the assessment of degree and cause (question 30).
23. Smoking questionnaire. This is a modification of that in
Rose and Blackburn, pages 160-162. Information concerning cigarette smoking only is sought.
Examination
26. Blood pressure. A standard mercury sphygmomanometer
should be used and should be regularly maintained. Before
use, check that the mercury is at zero level. When reading
the level, the mercury column should be at the examiner's
eye level.
The blood pressure should be measured immediately after
the application of the questionnaire, before the patient is
subjected to any of the other examination procedures and
after he/she has been sitting for 10 min or more. Eating or
smoking should not be allowed for the half-hour before. The
measurement should be made in the sitting position in a chair,
in either arm. The cuff bladder should measure at least
12 X 22 cm. Any restrictive clothing should be removed and
the cuff applied to the upper arm 2-3 cm above the antecubital fossa with the cuff bladder on the medial side of the
arm. The arm should be comfortable, supported at an angle
of 0-45° from the trunk:
(a) Rapidly inflate the cuff to a pressure above the radial
palpatory pressure.
(b) Start the run-down of the mercury column promptly,
at a rate of fall of 2 mm per second.
(c) Systolic pressure is recorded at first perception of
successive sounds.
(d) Diastolic fourth-phase level (diastolic 1) is recorded as
the point at which sounds cease to be tapping in
quality and are fully muffled. Changes in loudness are
not counted. If there are two points of change, record
the lower.
(e) Diastolic fifth-phase level (diastolic 2) is recorded as
the point at which sounds disappear.
(f) All readings are recorded to the nearest 2 mm Hg
pressure, reading the calibration below the meniscus.
Training procedures for blood pressure registration
should be undertaken (see Rose and Blackburn, pages
92-93), usjng the standard tape recordings.
Calibration of instruments used should be made at the
outset of the Study and at regular intervals thereafter.
27-28. Record height (without shoes) and weight (without coat/jacket and shoes) in metric units (centimeters and
kilograms). Those centers normally using other units should
change to metric equipment or use conversion tables.
If, for any reason, such as bilateral amputation, height
or weight cannot be directly measured, the best estimate
should be recorded.
29. Fundus examination. Mydriatic drops should be instilled
— unless glaucoma is present—as soon as the patient arrives
at the clinic, so that adequate dilatation will occur by the
time the questionnaire has been applied and other examinations performed. If the pupil has not dilated to a minimum of
3 mm, further mydriasis should be applied. If dilatation above
3 mm still does not occur, examination should be attempted.
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
181
WHO STUDY OF VASCULAR DISEASE AND DIABETES, PART 1/R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
If a poor view of the fundus is obtained, the fact should be
recorded on the questionnaire.
Examination should be performed with a direct ophthalmoscope in a darkened room. The patient's gaze should be fixed
(and the examiner move his head to survey the fundus)
except for examination of the macula, when the patient
should look directly at the ophthalmoscope light.
Examination of one fundus should be completed in 120 s,
excluding the counting of any lesions observed. A timing
device (e.g. as used in photographic developing) should be
used to indicate the elapsed time.
Definitions
(a) Small red lesion: having a diameter certainly less than
that of the retinal artery at the optic disc.
(b) Medium red lesion: larger than (a) but with its largest
diameter less than that of the optic disc.
(c) Large red lesion: having its largest diameter greater than
that of the optic disc.
(d) Hard exudate: These should be distinguished from
"colloid" or "Drusen" bodies. A hard exudate is white with a
sharp outline. If they are large or extensive, with a total
area greater than that of the optic disc, include in "six or
more" category.
(e) Soft exudates: Enumerate as for (d).
(f) New vessel systems. It is sometimes uncertain whether
only a few fine vessels are present or whether they are normal
retinal components. The three categories designated are
therefore "nil," "doubtful," and "definite." The last category
includes all definite new vessels, whether in the retinal or
preretinal plane, and also retinitis proliferans in all its forms.
Proteinuria
This should be assessed using the salicylsulphonic acid test,
as follows:
182
5-6 ml of centrifuged urine are placed in a 13 X 100
mm test tube.
1-2 ml of 30% (w/v) salicylsulphonic acid in methanol
is layered onto the urine without mixing. After a minute the
urine/acid interface is examined in good lighting against a
dark background. If no turbidity—or a barely perceptible
turbidity—is seen, grade as 1. If proteins are present, a
cloudy precipitate appears at the junction of the two fluids.
Grade slight turbidity as 2 and distinct or heavy turbidity
as 3. A heavy, flocculent precipitate should be graded as 4.
The reagent salicylsulphonic acid in methanol should be
prepared thus:
Dissolve 300 g reagent grade salicylsulphonic acid
crystals (2-hydroxybenzoic, 5-sulphonic acid dihydrate) in
absolute methanol and dilute (with methanol) to 1 liter.
The solution should be clear; if cloudy, allow to stand until
settled and draw off clear supernate into another container.
Resting Electrocardiogram
Twelve lead resting electrocardiograms are required, and
these will be centrally coded. The Minnesota coding procedure requires good quality recordings and at least six
complexes per lead. Duplicate recordings (two series as long
strips) should be made and one retained at the participating
center.
Procedure. The recording should be made with the patient
supine at least 2 h after the last meal: Smoking and vigorous
exercise should be forbidden for half an hour before the
recording. Careful preparation of skin electrode contact is
important. All recordings should include a 1 mv calibration
signal. With a steady base line, six complexes are adequate
for coding. However, if the base line is unsteady, longer
recordings should be made. If interference suppression is
used, the fact should be recorded on the electrocardiogram.
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
WHO STUDY OF VASCULAR DISEASE AND DIABETES, PART 1/R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
APPENDIX 3
MULTINATIONAL
STUDY
CORE
OF
VASCULAR
DISEASE
IN
DIABETICS
QUESTIONNAIRE
Study Number
u
8
Centre Number
1-2
3-4
Serial N umber
5-8
N ame of patient:
All forenames
Family name
PLEASE COMPLETE IN PENCIL
You are asked to insert the appropriate number in a box. T o facilitate the coding and to avoid error, please make the
DOCUMENT B SHOULD
figures easily
Also
write 4 ; for 'four' write H ; for 'seven' write
e s i y readable.
e b e .
lso for
f 'one'
e w
EE CONSULTED FOR ITEMS MARKED WITH AN *
Was the diabetes first diagnosed because of specific
Basic data
features such as one or more of the following Day
Month
Year
thirst, polyuria, weight loss, genital pruritus,
1. Date of birth:
9-14
19
pre-coma, coma?
2.
Place of birth:
1 = yes; 2 = no; 9 = unknown
Town:
If YES, state time between onset of symptoms
and subsequent diagnosis:
C ountry:
Local identifying number:
3^
Sex
4.
Marital status:
1 = male;
2 = female
1 = single;
2 = married;
4 = widowed; 5 = other;
3 = divorced;
9 = unknown
16
6.
1 = less than one week;
2 = one week or more but less
than one year;
3 = one year or more but less
than three years;
4 = three years or more;
9 = unknown
Year of first diagnosis of diabetes by
a doctor (not necessarily of onset of
symptoms)
1 = yes; 2 = no; 9 = unknown
,~
Was vascular disease originally responsible
for bringing the diagnosis of diabetes to light?
1 = yes; 2 = no; 9 = unknown
17-18
Routine urine test
Routine blood test
•
19
D iabetes detection programme
O ther (please specify)
If Y ES, were the following conditions responsible?
1 = yes; 2 = no; 9 = unknown
Ischaemic heart disease
Peripheral vascular disease
28
•
D
•
•
•
•
•
•
•
Is the diabetes secondary to or associated with the following?
1 = yes; 2 = no; 9 = unknown
•
20
Cushing's Syndrome
21
Acromegaly
Cerebral vascular disease
22
Diabetic retinopathy
23
Diabetic renal disease
24
H ypertension
25
Haemochromatosis
Pancreatic calcification
Chronic pancreatitis
O ther (please specify)
•
If the answers to questions 6 and 7 are NO, specify
circumstances leading to the diagnosis:
Previous medical history
5.
•*
•
Steroid therapy
26
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
(Continued)
183
WHO STUDY OF VASCULAR DISEASE AND DIABETES, PART 1/R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
Page 2 for Serial Number
MULTINATIONAL STUDY OF VASCULAR DISEASE IN DIABETICS
10. At the time of diagnosis were the following present?
1 = yes; 2 = no; 9 = unknown
Congestive heart failure
Angina pectoris
History of cardiac infarct
History of stroke
History of intermittent claudication
History of hypertension
An ischaemic foot
Diabetic retinopathy of any degree
Proteinuria due to diabetic nephropathy
•
•
•
•
•
•
D
D
•
what is the total duration of
insulin therapy? (disregard
short breaks in treatment)
39
15. Is the patient receiving oral antidiabetic drugs?
40
41
42
1
2
3
4
= sulphonylureas;
= biguanides;
= both;
= neither
•
72
16. Is the patient receiving the following types of drugs?
1 = yes; 2 = no; 9 = unknown
D
43
Oral diuretics
44
•
•
•
•
Oral combined contraceptives or
other oestrogens
45
Cortisone or related steroids
46
Blood lipid lowering drugs
47
Blood pressure lowering drugs
11. Has the patient suffered from ischaemic gangrene or
had an amputation of toe, foot or leg for arterial
obstruction?
1 = yes; 2 = no; 9 = unknown
70-71
years
Card No.
48
73
74
75
76
77
0
80
•
10
Interview - ask questions directly of patient
17.* "Have you ever had any pain or discomfort in
your chest?"
Present treatment
1 = yes; 2 = no;- 9 = unknown
12. * Current diet as prescribed
(a) If NO, "have you ever had any pressure or
heaviness in your chest?"
Enter actual quantity specified. If no amount specified,
enter 9999 or 999; if restricted but not quantified, enter
8888 or 888.
1 = yes; 2 = no; 9 = unknown
Total calories/day
cal.
49-52
Carbohydrate/day
grammes
53-55
Fat/day
grammes
56-58
Protein/day
grammes
59-61
Alcohol/day
grammes
62-64
14. Is the patient receiving insulin?
1 = yes; 2 = no; 9 = unknown
If YES:
what is the total daily dose?
184
units
•
•
(b) "Do you get it when you walk uphill or hurry?"
1 = yes; 2 = no;
3 = never hurry or walk uphill
I
|
I
| 11
If NO, proceed to question 18.
(c) "Do you get it when you walk at an ordinary
pace on the level?"
12
1 = yes; 2 = no
13. Does the dietary prescription include advice
to increase proportion of unsaturated to saturated fat?
1 = yes; 2 = no; 9 = unknown
If NO, proceed to question 19.
(d) "What do you do if you get it while you are walking?"
65
66
1 = stop or slow down;
2 = carry on
|
|
I
| 13
(Record "stop or slow down" if subject carries
on after taking nitroglycerine)
If CARRY ON, proceed to question 18.
•67-69
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
(Continued)
WHO STUDY OF VASCULAR DISEASE AND DIABETES, PART 1/R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
Page 3 for Serial Number
MULTINATIONAL STUDY OF VASCULAR DISEASE IN DIABETICS
(e) "If you stand still, what happens to it?"
1 = relieved; 2 = not relieved
(c) "In what part of your leg do you feel it?"
I
14
If NOT RELIEVED, proceed to question 18.
I
I
Sternum (upper or middle)
Sternum (lower)
Left anterior chest
Left arm
Other
27
•
28
(d) "Do you get it if you walk uphill or hurry?"
If MORE THAN 10 MINUTES, proceed to question 18.
(g) "Will you show me where it was?" Put code
1 = yes? 2 = no in each box.
•
If calves not mentioned, ask "anywhere else?"
If PAIN DOES NOT INCLUDE CALF/CALVES,
proceed to question 20.
( 0 "How soon?"
1 = 10 minutes or less;
2 <= more than 10 minutes
1 = pain includes calf/calves;
2 = pain does not include calf/calves
1 = yes; 2 = no;
3 = never hurry or walk uphill
If NO, proceed to question 20.
•
•
•
•
16
17
18
(e) "Do you get it if you walk at an ordinary pace
on the level?"
•
•
1 = yes; 2 = no
(f) "Does the pain ever disappear while you are
walking?"
19
1 = yes; 2 = no
If YES, proceed to question 20.
20
1 = yes; 2 = no
30
(g) "What do you do if you get it when you are walking?"
1 = stop or slow down;
2 = carry on
(h) "Do you feel it anywhere else?"
29
I I
II
31
|
|
If CARRY ON, proceed to question 20.
(If Y ES, record additional information above)
(h) "What happens to it if you stand still?"
(i) "Did you see a doctor because of this pain (or discomfort)?"
1 = yes; 2 = no
|
|
1 = relieved; 2 = not relieved
22
(i) "How soon?"
If Y ES, "what did he say it was?"
1 = 10 minutes or less;
2 = more than 10 minutes
20.* "Have you ever had a stroke?"
18.* "Have you ever had a severe pain across the front of
your chest lasting for half an hour or more?"
1 = yes; 2 = no
If YES, ask:
"Did you see a doctor because of this pain?"
1 = yes; 2 = no
If YES, "what did he say it was?"
1 = yes; 2 = no
•
•
23
If NO, "have you ever had weakness or loss of
strength in an arm or leg lasting for 24 hours
or more?"
1 = yes; 2 = no
24
•
•
•
•
•
32
33
34
35
If the answer to either question is YES, ask,
"did this happen before your diabetes was discovered?
i?"
1 = yes; 2 = no; 9 = unknown
21.* "Do you have any difficulties with your eyesight other
than the ordinary need for glasses (spectacles)?"
1 = yes; 2 = no
36
37
19.* (a) "Do you get pain in either leg on walking?"
25
1 = yes; 2 = no
1 = yes, regularly;
2 = yes, occasionally (less
than 1 per day);
3 = no
If NO, proceed to question 20.
(b) "Does this pain ever begin when you are standing still
or sitting?"
1 = yes; 2 = no
22.* (a) "Do you smoke cigarettes now?"
f
[
38
If NO, proceed to question 23.
26
If YES, proceed to question 20.
(Continued)
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
185
WHO STUDY OF VASCULAR DISEASE AND DIABETES, PART 1/R. J. JARRETT, H. KEEN, AND V. GRABAUSKAS
MULTINATIONAL STUDY OF VASCULAR DISEASE IN DIABETICS
Page 4 for Serial Number
If NO, it is because of:-
(b) "Do you inhale?"
1 = yes; 2 = no
39
(c) "How many cigarettes do you
usually smoke per day?"
40-41
(d) "How many cigarettes did you
smoke per day a year ago?"
42-43
(e) "How old were you when you began
to smoke regularly?"
Age
in yrs
1 = poor pupillary dilatation;
2 = lens opacity;
3 = vitreous opacity/haemorrhage;
4 = combination of 2 and 3;
5 = corneal opacity;
6 = glaucoma;
7 = eye absent;
8 = other
Right
eye
Left
eye
D •
28-29
Is lens present?
44-45
1 = yes; 2 = no
After asking this question proceed to item 24.
30-31
23.* (a) "Did you ever smoke cigarettes?"
•
1 = yes, regularly;
2 = yes, occasionally;
3 = no, never
1
2
3
4
If NO, proceed to item 24.
(b) "What is the maximum number of cigarettes
you ever smoked per day for as long as a
year?"
= none;
= one;
= two to five;
= six or more
49-50
b) soft
(d) "How old were you when you
stopped smoking?"
Age
in yrs
51-52
New vessel systems
1 = none; 2 = doubtful; 3 = definite
Month
25. Name of interviewer:
(please print)
Card No.
Systolic
9-11
1
29.* Fundus examination
Is pupillary dilatation 3mm or more?
15-17
(b) central.laboratory
18-2( 33.* Plasma creatinine:
(a) local laboratory
^n
28. * Weight (without shoes and coat/jacket)
21-23
kg
Right
eye
(b) central laboratory
mg/dl
46 -4E
mg/dl
49-51
mg/dl
52-54
mg/dl
55-57
•
1 = none;
2 = slight turbidity;
3 = distinct or heavy turbidity;
4 = flocculent precipitate
35.* Resting electrocardiogram - Minnesota code
1-11-2 1-3 2 3 4 5 6
Is retinal detail easily visible?
1 = yes; 2 = no
186
•
U44"45
U
34* Proteinuria:
Left
eye
24-21
1 = yes; 2 = no
42-43
31. Name of eye observer:
(please print)
Investigations
12-14 32.* Plasma cholesterol:
(a) local laboratory
Diastolic 1
Diastolic 2
40-41
30. Visual assessment:
Year
If patient answered YES to question 21,
assess visual disability
53-58
1 = totally blind
2 = light perception only
3 = can count.fingers
4 = can read newspaper headlines
2 80
(characters 2 cm in height)
Examination
27. *Standing height (without shoes)
|38-39
.
19
26* Blood pressure: mmHg
(sitting)
36-37
a) hard
in yrs
Day
• •
• •
• •
Large red lesions
Exudates
17-48
32-33
Medium red lesions
(c) "How old were you when you
first began to smoke regularly?"
24. Date of interview
D
Small red lesions
46
7
8
58
9
59-69
J26-2'
DIABETES CARE, VOL. 2 NO. 2, MARCH-APRIL 1979
Card No.
80