International variations in mortality among diabetic patients: the WHO Multinational Study of Vascular Disease in Diabetics
International variations in mortality a m o n g diabetic patients: the W H O Multinational Study o f Vascular D i s e a s e in D i a b e t i c s *
J. H e a d a n d J. H . F u l l e r 0 1 2
Oklahoma 0 1 2
London 0 1 2
Tokyo 0 1 2
0 Kong , Switzerland
1 of the original centres (Warsaw , Berlin, Havana, Arizona , USA
2 Department of Comunity Medicine, University College and Middlesex School of Medicine , London , UK
Summary. Mortality a m o n g 4740 diabetic men and w o m e n aged 35-55 years participating in the W H O Multinational Study of Vascular Disease in Diabetics has been studied. Ten Zagreb) were able to identify the life/death status of their study participants on 1 January 1983, giving an average follow-up period of six to seven years. All-cause mortality rates in males varied about threefold among the ten participating centres with the highest rates in Warsaw, Berlin and H a v a n a and the lowest rates in Tokyo and Zagreb. All-cause mortality rates for females varied about fourfold with the highest rates in Warsaw and O k l a h o m a and the lowest rates in Tokyo. The proportion of deaths ascribed to circulatory disease varied a m o n g the centres ranging from 32% for males and 0% for females in Tokyo to 67% for males and 47% for females in London. There was an excess all-cause mortality in males c o m p a r e d to females for all centres except Zagreb. This excess also applied to circulatory diseases in general, ischaemic heart disease in particular and occurred in both insulin-dependent and non-insulin-dependent diabetic patients. D e a t h rates for insulin-dependent diabetic patients were generally higher than those for non-insulin-dependent patients. sulin-dependent diabetes, circulatory disease, epidemiology.
Mortality; insulin-dependent diabetes; non-in-
9 Springer-Verlag 1990
H o n g
cigarette s m o k i n g a n d h y p e r l i p i d a e m i a [
3
]. T h e
W H O
Multinational S t u d y of Vascular D i s e a s e in Diabetics
( M S V D D ) [
4
] c o n f i r m e d p r e v i o u s studies in showing a
m a r k e d variation in the p r e v a l e n c e of indicators of
cardiovascular disease such as e l e c t r o c a r d i o g r a p h i c ( E C G )
abnormalities and chest p a i n s y m p t o m s a m o n g 14 samples
o f m i d d l e - a g e d diabetic patients t h r o u g h o u t the w o r l d [
5
].
Ten of the original 14 participating centres h a v e b e e n
able to c a r r y out a follow-up study for m o r t a l i t y in their
diabetic cohorts, including a standardised assessment of
cause o f death. We h a v e e x a m i n e d variations in m o r t a l i t y
rates a m o n g the centres for all causes of death, circulatory
disease and renal disease. We have also analysed all cause
mortality by t y p e of diabetes: i n s u l i n - d e p e n d e n t or n o n
insulin-dependent.
Subjects and methods
Sample
In each centre, samples were selected from patients aged 35-55
already under treatment for diabetes who had been diagnosed at least
one year earlier [
4
].
In each participating centre, a method for identifying a sampling
frame of diabetic patients was chosen which would best reflect the
population with diabetes in that area. Further details of the sampling
method are given in previous publications of the WHO MSVDD [4,
* Prepared by the authors for the WHO multinational study group.
5]. Some centres sampled from all patients attending diabetic
outpatient clinics (London, Warsaw, Hong Kong and Tokyo). In Berlin,
Zagreb and Havana, the samples were selected from registers of all
diabetic patients resident within a defined geographical area. The
Arizona and Oklahoma patients were recruited from registers of
diabetic patients belonging to identifiable ethnic groups i.e., the
Pima Indians in Arizona and the American Indians in Oklahoma.
The Swiss sample was selected randomly from the diabetic patients
of a large number of local practitioners from all over the country.
Patients were grouped according to age (3541; 42-48;
4955 years), duration of diabetes (1-6; 7-13; 14 or more years) and
gender. A random sample was selected from each of these 18
agegender-duration strata. The aim was to select 28 patients from each
stratum which would provide a total sample of approximately 500
patients from each centre. In this study, an insulin-dependent diabetic
patient is defined on a clinical basis as one receiving insulin therapy
within one year of diagnosis. The small proportion (7.5%) of patients
from Arizona thereby classed as insulin-dependent reflects the false
positive error rate of this definition, since it is thought that
insulin-dependent diabetes does not occur among the Pima Indians [
6, 7
].
Mortalityfollow-up
Each centre was asked to ascertain the life/death status of all the
original study participants on 1 January, 1983. This information was
reported to the co-ordinating centre using a standard questionnaire.
If the life/death status was unknown, the reason (e. g., emigration)
(...truncated)