Eur. J. Epidemiol. 0392-2990
September 1992, p. 693-701
EUROPEAN JOURNAL
Vol. 8, No. 5
OF EPIDEMIOLOGY
CHOLESTEROL AND OTHER CARDIOVASCULAR RISK FACTORS IN A WORKING POPULATION IN ILE-DE-FRANCE (FRANCE): FIRST RESULTS OF THE PCV-METRA STUDY D. LAURIER*, N.P. CHAU .1, P. SEGOND** and the PCV-METRA Group *Unitb de R e c h e r c h e s B i o m a t h ~ m a t i q u e s et B i o s t a t i s t i q u e s - University Paris 7 - l n s e r m U263 2, P l a c e J u s s i e u - 75251 - Paris - C e d e x - 05. **The P C V - M E T R A Group - B o u l o g n e - France.
Key words: Cholesterol - Lipoproteins - Screening - CDV risk factors In 1989, the French PCV-METRA Group (PCV-METRA = Prrvention Cardio-Vasculaire en Mrdecine du Travail) started a large prospective survey of cardiovascular (CDV) morbidity and mortality and of CVD risk factors, especially cholesterol, in a working population in Ile-de-France, a region including Paris. This report presents the first results of this study, based on a sample of 5758 men and 2603 women, aged 18-65 years. The variables examined included the levels of total cholesterol (TC), High-density-liproprotein cholesterol (HDL-C), low-density-lipoprotein cholesterol (LDL-C), and the other major CVD risk factors (smoking, sedentary way of life, hypertension, hypercholesterolemia, use of oral contraceptives and familial history of CVD risks). TC and LDL-C significantly increased with age. The changes with age were significantly different in men and women. The levels were similar in both sexes at less than 30 yrs, increased sharply for men after age 30 and were significanlty higher in men than in women from 30 to 55 yrs. Beyond 55 yrs, no difference was obserbed between the two sexes. In contrast, HDL-C was higher in women at all age ranges. In the total sample, 35% of men and 210/0of women were hypercholesterolemic (TC > 2.4 g/ L). Our observations fully confirm and refine previous findings in the US and in other Efi-?opean countries. In addition, a substantial set of data on CVD risk factors for the working population in France, especially for female subjects for whom data are scanty, is now available.
INTRODUCTION
Serum cholesterol has long been recognized as one of the benchmark measures for the risk of development of cardiovascular diseases (CVD). Since the beginning of the 1960's several longitudinal surveys, such as the Framingham Study (15), the Multiple Risk Factor Intervention Trial (5) and the Pooling Project (20), have demonstrated that plasma cholesterol level was signicantly correlated to the prevalence of CVD. In addition, the Lipid Research Clinic's Coronary Primary Prevention Trial (19) has demonstrated a reduction in coronary heart disease 1 Corresponding author.
(CHD) incidence associated with cholesterol lowering. Also, randomized multifactorial primary prevention trial (14) and community programs (21) have shown a direct relationship between a lowering of cholesterol level and the reduction of CHD. In France, some epidemiological studies on risk factors of cardiovascular morbidity and mortality have been published, but none specifically addressed the role of cholesterol. The Paris Prospective Study, initiated in 1967, investigated cardiovascular risk factors in a cohort of 7700 civil servants (10). This study involved exclusively male subjects. The PreClinical Investigations Study, started in 1970, also explored cardiovascular risk factors in male subjects only (6). More recently, the WHOs Monitoring
693
Laurier D. et aL
Eur. J. Epidemiol.
Trends and Determinants in Cardiovascular Disease (MONICA) project examined the prevalence of cardiovascular morbidity and mortality in HauteGaronne (Toulouse), Bas-Rhin (Strasbourg) and the Urban Community of Lille (22). Its objective was to explore the relationships between temporal variations of CVD incidence, risk factor levels and medical care in the 3 regions. However, cholesterol is not routinely screened for in France and is rarely assayed before 40 yrs of age. All these facts demonstrate a need for further study of cholesterol in France, especially in female subjects. Screeening of cholesterol and other risk factors for cardiovascular diseases can be best fulfdled by the national "m6decine du travail", the workers health care system in France. In 1989, the PCV-METRA group initiated a large prospective study on cholesterol and other CVD risk factors in the workers of several firms in lie-de-France, a region that includes Paris. This report presents the first results of this survey. MATERIALS AND METHODS
Design of the PCV-METRA Cholesterol Survey The PCV-METRA group is composed of occupational health physicians and cardiovascular researchers from Paris Universities and the National Institute for Health and Medical Research (INSERM) (note 1). The objectives of the PCV-METRA cholesterol survey are: 1. to investigate the prevalence and distribution of cholesterol and cardiovascular risk factors in male and female employees of several national finns in Ile-de-France; 2. to provide complementary clinical and laboratory examinations in cases of borderline high or high CVD risks and; 3. to follow the subjects for several years on the basis of an annual health visit (follow-up of the factors under study, behavior changes, and the occurrence of cardiovascular diseases). ~The PCV-METRA includes 23 occupational health physicians of 20 large firms in Ile-de-France. In France, a regular, annual medical visit is obligatory for all workers. Health visits were performed at the worksite. The random selection of participants was carried out by the physicians from the list of persons elegible for ther annual exam. An anonymous questionnaire was administered by the physician and a blood sample was taken for biological measurements. The questionnaire and procedures to guarantee the anonimity of the data fully explained to the selected subjects. The latter procedures were examined and approved by the National Committee for Informatics and Liberty (CNIL). After being informed of the procedures, refusal were the exception (< 1%) and all participants gave their informed consent. However, only about 80% of the participants accepted the blood sample.
Subjects whose cholesterol level or blood pressure were high were referred to a hospital in Paris. There, additional examinations were carried out, including electrocardiogram, vascular echography, nutritional assessment and dietetic advice by a dietetician and pharmacological treatment when necessary. Questionnaire Principal variables of interest on the questionnaire are: age at examination, sex, occupational categories (see below), smoking habits (currently smoking, number of cigarettes, cigar or pipes per day, duration of smoking), sedentary way of life (defined as having no or less than one hour of physical activity per week, not including walking to work), use of contraceptives, and a family history for CVD (positive history was defined as having at least one history of sudden death, vascular disease, arterial hypertension, hypercholesterolemia, diabetes or heart disease among parents, grandparents, sisters, brothers, aunts and uncles aged 65 yrs or less). To define occupational categories, we used an abbreviated four categories classification system (manager-engineer, supervisor-technician, employee, manual labor) of the National Institute for Statistics and Economic Studies (INSEE). However, in the analysis the two first categories were grouped as "manager" and the two last categories were grouped as "Employee". Measurements Weight and height were measured in light clothes and without shoes. Systolic blood pressure (SBP, mm Hg) and diastolic blood pressure (5th Korotkoffphase, DBP, mm Hg) were determined by standard sphygmomanometer, with the subject in a supine position after 10 minutes at rest. A venous blood sample was withdrawn after the subjects had fasted for 12 hours and with the subjects in the supine position for at least 10 minutes (12). Lipid measurements were performed in four different laboratories. These laboratories (as all laboratories in France) were reviewed four times a year by the National Health Laboratory. Before the beginning of this study, a sample was sent to the four laboratories to test the comparability of the measurements. The interlaboratory variation was less than 5%. Serum total cholesterol (TC, mmol/L) and triglycerides (TG, mmol/L) were measured with enzymatic methods (11). High-density lipoprotein cholesterol (HDL-C, mmol/L) was measured by an enzymatic method after the precipitation of low-density lipoprotein and verylow-lipoprotein by phosphotungstic acid amgnesium chloride (26). Blood glucose level (mmol/L) was measured after overnight fast. Calculations and statistics The body mass index: BMI = weight/height 2 (kg/ m2), was used to assess body corpulence. Obesity was
694
Vol. 8, 1992
PCV-METRA study on cardiovascular risk factors
defined as a BMI of 30 kg/m 2 or above for men and a BMI of 28.6 kg/m 2 or higher for women. These figures correspond to 1200/0 o f the ideal weight (3). Cigarette-equivalent was calculated by the following formula: Cigarette-equivalent = (number of cigarettes/day) + 4 (number of cigars/day) + 1.5 (number of pipes/day). Lifelong smoking dose was estimated by multiplying the number of years smoking and cigarette-equivalent/20, and expressed in pack-years. The level o f LDL cholesterol (LDL-C) was calculated by the formula (12): LDL cholesterol = TC - H D L cholesterol - (TG/2.2). Subjects were classified into one o f three groups according to their TC level: normal group: TC < 5.2 m m o l / L (2.0 g/L), borderline high cholesterol group: TC from 5.2 to 6.19 m m o l / L (2.00 to 2.39 g/L) and high cholesterol group: TC _> 6.2 mmol/L (2.40 g/L) (12).
Hypertension was defined as an SBP _> 160 m m Hg and/or a DBP _> 95 m m Hg. Hyperglycemia was defined as a blood glucose level > 6.1 mmol/L (1.10 g/L). Data management and analyses were performed using the SAS statistical packages (25). The Chisquare test was used to compare percentages (Table 1) and one-way analysis of variance was used to compare the levels of a risk factor among age groups (Table 2 and 3). Duncan's multiple range test was used to determine if cholesterol level in an age group was significantly different from the other age groups while multivariate analysis of variance was used to test differences between the sexs, when age groups and sex age interaction were taken into accoint. Comparison of cholesterol levels for each age group (Fig. 1) were performed by analysis of variance. Since multiple comparisons were performed, we used the global residual sum of squares.
TABLE 1. - Distribution (% of the total sample) of CVD risk factors, by sex and age groups. 18-65 years %
18-34 yeras %
35-44 years %
45-54 years %
55-65 years %
(n = 5758)
(n = 1207)
(n = 2107)
(n = 1810)
(n = 634)
Occupation (Manager)
83
74
83
87
90
0.001
Smoking
35
36
38
33
30
0.001
Hypertension b
14
5
12
18
22
0.001
Hyperglycemiac
9
2
7
12
13
0.001
Obesityd
6
3
6
7
4
0.001
Sedentary way of life
44
37
43
48
50
0.001
Family history of CHD risk factors °
64
66
65
63
61
NS
(n = 2603)
(n = 789)
(n = 978)
(n = 614)
(n = 222)
Occupation (Manager)
43
32
45
50
49
0.001
Smoking
27
36
27
20
18
0.001
Hypertension b
5
1
4
9
13
0.001
Hyperglycemiac
3
2
2
5
4
0.008
Obesity d
8
3
8
12
12
0.001
Sedentary way of life
55
48
55
62
65
0.001
Family history of CHD risk factors°
73
71
72
75
74
NS
Oral contraceptive use
24
44
21
9
1
MEN
WOMEN
a b c 'd
P_valuea
0.001
Chi-square test of differences by age groups, NS = not significant; Hypertension is defined as SBP > 160 m m Hg and/or DBP > 95 m m Hg; Hyperglycemia is defined as bloo-d glucose > 6.10 mmol/L (1~10 g/L); Obesity is defined as a BMI > 30 kg/m2 for men and BMI > 28.6 kg/m2 for women; Family history of CHD risk factors include family history of sudd--en death, vascular disease, hypertension, hypercholesterolemia, diabetes or heart disease.
695
Laurier D. et al.
Eur. J. Epidemiol.
TABLE 2. - CHD risk factor levels by sex and age groups: means ± standard deviation. 18-65 years
18-34 years
35-44 years
45-54 years
55-64 years
MEN
(n = 5758)
(n = 1207)
(n = 2107)
(n = 1810)
(n = 634)
Glycemia (mmol/L)
5.22+
4.88___ 0.55
5.16___ 0.67
5.38___ 1.00
5.38___ 0.89
Systolic Blood Pressure (mm Hg)
0.83
133.0 ___ 15.0 129.1 ___ 11.5 132.1 _+ 14.0 135.0 ___ 15.3 138.6 ___ 16.9
P-value a
0.001 0.001
Diastolic Blood Pressure (mmHG) 81.0 ___ 11.0
77.9 ___ 8.6
81.2 ___ 10.7
83.5 ___ 11.3
84.6 ± 11.1
0.001
Body Mass Index (kg/m 2)
24.7 -+- 3.2
23.4 _
3.0
24.7 ___ 3.3
25.3 _
3.1
25.2 ___ 3.0
0.001
Cigarette-equivalent
17.3 ___ 11.9
15.8 ___ 9.0
17.9 + 12.1
17.8 ___ 13.3
15.9 _ 11.9
NS
Lifelong Smoking Dose (pack/year)
18.4 ___ 16.3
8.7 ___ 6.3
17.2 ± 13.2
24.1 ± 18.2
28.0 __+ 23.1
0.001
WOMEN
(n = 2603)
(n = 789)
(n = 978)
(n = 614)
(n = 222)
Glycemia (mmol/L)
4.88___ 0.78
4.77___ 0.72
4.83 +
Systolic Blood Pressure (mm H G )
124.0
0.55 5 . 0 0 _
1.00
5.00___ 0.72
± 14.0 119.9 ___ 11.2 122.6 ___ 12.9 129.0 ± 14.2 132.3 + 15.5
0.001 0.001
Diastolic Blood Pressure ( m m H G )
76.0 + 10.0
73.5 ___ 8.4
75.6 --- 9.6
79.1 ___ 10.0
81.0 __+ 10.9
0.001
Body Mass Index (kg/m 2)
22.8 ±
3.8
21.6 ±
2.9
22.8 _
3.8
23.9 __+ 4.1
24.3 ___ 4.2
0.001
Cigarette-equivalent
14.9 ___ 9.2
14.2 +
8.0
15.4 ±
9.7
16.0 ___ 10.2
13.8 _ 11.1
NS
Lifelong Smoking Dose (pack/year)
11.9 _± 11.0
1 3 . 6 _ 10.1
16.9 ___ 12.3
20.9 _+ 22.3
0.001
7.1 ___ 5.5
a Analysis o f variance through age groups, NS = not significant.
RESULTS
Subjects The present analysis was performed on data from the first examination and includes 5758 men (70%) and 2603 women (30%), aged 18-65 years. CVD risk factors Distribution of the main CVD risk factors in the total sample are displayed in Table 1. The proportion o f managers was higher in m e n that in women (830/0 versus 43°/0, p < 0.001). The percentage of smokers was higher in men than in women (350/0 versus 270/0, p < 0.001) and decreased with age in both sexes. Hyperglycemia and hypertension were more frequent (about 3 times) in m e n than in women (p < 0.001 for both factors). The proportion of obese subjects was higher in women than in m e n (p < 0.001). The percentage o f obesity increased with age from 18 to 45 696
years in both sexes, and then plateaued after this age range. The percentage of subjects without physical activities was lower in men than in women (44% versus 55%, p < 0.001). All these factors showed a significant increase with age. Familial cardiovascular history was declared by 64°/0 o f the men and 74°/0 o f the women (p < 0.0001). At age _< 45 yrs, 31% o f the women used oral contraceptives. Table 2 displays mean levels o f the studied CVD risk factors. Glycemia, blood pressures, BMI and lifelong smoking dose all increased with age in both sexes. Cigarette-equivalent was not related to age in either sex. For all o f these factors, mean values were significantly higher in men than in women (p < 0.001, for each factor).
Cholesterol levels Table 3 displays TC, HDL-C, LDL-C and T G levels, by sex and age group. All four levels increased with age in both sexes (p < 0.001). W h e n controlling for age, TC, LDL-C and T G L levels were higher in
Vol. 8, 1992
PCV-METRA study on cardiovascular risk factors
TABLE 3. - Cholesterol and triglycerides levels (mmol/L) by sex and age groups: mean _ standard deviation. 18-65 years
18-34 years
35-44 years
45-54 years
55-65 years
P-value a
MEN
(n = 5758)
(n = 1207)
(n = 2107)
(n = 1810)
(n = 634)
Total Cholesterol level (mmol/L)
5.82___ 1.14
5.20___ 1.09
5.82___ 1.14
6.08 + 1.06
6.10___ 1.06
0.001
HDL-C level (mmol/L)
1.32 ___ 0.34
1.29 ___ 0.36
1.32 ___ 0.34
1.32 ___ 0.34
1.37 _ 0.36
0.001
LDL-C level (mmol/L)
3.90 ___ 1.03
3.41 ___ 0.98
3.90 + 1.01
4.11 __ 0.96
4.11 ___ 0.98
0.001
Triglycerides level (mmol/L)
1.32 ___ 0.95
1.08 _+ 0.69
1.35 ___ 1.08
1.42 + 0.95
1.30 ___ 0.76
0.001
WOMEN
(n = 2603)
(n = 789)
(n = 978)
(n = 614)
(n = 222)
Total Cholesterol level (mmol/L)
5.48 ___ 1.03
5.04 ___ 1.01
5.32 ___ 0.90
5.79 + 1.03
6.23 ___ 1.06
0.001
HDL-C level (mmol/L)
1.60 + 0.39
1.55 _+ 0.39
1.58 ___ 0.36
1.63 _+ 0.41
1.65 _ 0.46
0.001
LDL-C level (mmol/L)
3.49 _ 0.98
3.13 _ 0.98
3.36 ___ 0.85
3.72 _+ 0.93
4.14 ___ 1.03
0.001
Triglycerides level (mmol/L)
0.87 _ 0.43
0.81 ___ 0.38
0.83 ___ 0.36
0.93 _+. 0.53
0.97 ___ 0.41
0.001
a Analysisi of variance through age groups; HDL-C: High-Density Lipoproteins Cholesterol; LDL-C: Low-Density Lipoproteins Cholesterol.
men than in women (p < 0.001), whereas the contrary was observed for HDL-C (p < 0.001). A refined analysis o f the changes in cholesterol levels with age is depicted in Figure 1. TC and LDL-C curves had similar shapes. Interaction between age and sex was significant for TC (p < 0.001) and LDL-C (p < 0.001). This indicates that the evolution of TC or LDL-C with age was different in men and women. At age < 30 yrs, no difference was observed in TC and LDL-C levels between men and women. In men, a steepper increase can be seen at an earlier age and total cholesterol level plateaued from age 45 to 65. In women, the increase of total cholesterol level was moderate until age 45, and then became steeper. From 30-55 yrs, TC and LDL-C were significantly higher in men than in women. Beyond age 55, no significant
differences could be seen between the two sexes. The shape of the HDL-C curve was very different from those o f TC and LDL-C. HDL-C was higher in women than in men for all age groups, however the interaction between age and sex was not statistically significant. Figure 2 depicts the percent o f subjects with borderline or high cholesterol for different age groups. The percentage o f subjects with high TC increased steeply in men and more slowly in women in the younger age groups. The two trends of increase were significant (13 < 0.001). In both sexes, the percentage of subjects with borderline TC increased at the early ages and plateaued above age 35 yrs. W h e n all age groups were pooled together, 35% of the men and 21% of women (19 < 0.001) had high TC levels.
697
Laurier D. et al.
Eur. J. Epidemiol.
TABLE 4. - Total cholesterol level in France: mean (mmol/L) and (standard deviation). PCV-METRA Paris
Paris Prospective Study" Paris
35-44 years 45-54 years 55-64 years
5.82 (1.14) 6.08 (1.06) 6.10 (1.06)
5.77 (1.11)
Women 35-44 years 45-54 years 55-64 years
5.33 (0.91) 5.79 (1.03) 6.23 (1.06)
MONICA-France b Haute-Garonne Bas-Rhin
Men
a
5.87 (1.52) 5.92 (1.06) 5.97 (1.01)
5.48 (1.01) 5.61 (1.19) 5.69 (1.06)
5.30 (0.90) 5.79 (1.11) 6.51 (1.11)
4.89 (0.91) 5.61 (1.11) 6.10 (1.01)
b M O N I C A - F r a n c e Study, 1985-87 (8).
Paris prospective study, 1967-72 (9);
6.7. Men
g --
6.2-
~
~ 5.7-
e
50--
***
40
**
~ 30-
5.2. [ [...
4.7
n
,
I
Women[
S
10-
<'25 25'-29 30"34 35'-39 40'-44 45"49 50'-54 55'-59 >'60 Age groups (yrs)
~
~b~i~e~lt~?e~:glho;h°lester°l ]
0 <30
30134
35'39 40144 45~49 50154 Age groups (yrs)
~55
4.4. ***
3.9.
-$
***
~
3.4.
e~
t~ ~ 2.9
Women
50 •
[
*
en W. . . .
~ 30. ~ 20.
i
<25 25-29 30-34 35-39 40-44 45-49 50-54 55-59 >60 Age groups (yrs)
10. 0 <30
-$ 1.8E
g
***
~
I~_._..U-----t
I
50'-54
>55
Figure 2. - Proportion of subjects with borderline high cholesterol and with high cholesterol, by sex and age groups.
.3
"~ 1.4 ----~ 1.0
30'-34 35'-39 40'-44 45'-49 Age groups (yrs)
Men Women
[] '
Borderline high cholesterol level (5.17 _< T C < 6.20 mmol/L)
[] "
high cholesterol level (TC > 6.20 m m o l / L ) .
<'25 25"29 30134 35"39 40'44 45-'49 50154 55"59 >'60 Age groups (yrs) DISCUSSION
Representativeness
Figure 1. - Cholesterol levels ( m e a n _ SEM, m m o l / L ) , by sex a n d age groups. Difference b e t w e e n sexes: * = p < 0.05, ** = p < 0.01, *** = < 0.901. [] Men; --Women.
The PCV-METRA survey is being carried out by occupational health physicians at the work site. Since all workers must undergo an obligatory annual medical visit, no selection was made on either clinical
698
Vol. 8, 1992
PCV-METRA study on cardiovascular risk factors
or biological aspects. The studied sample represents about 17% of the total working population of the 20 firms. We hope to include about 50% of the total population in the next 3 years. Mean age was the same in the population and in the sample for both sexes. However, managers-technicians were somewhat overrepresented (69% in the sample versus 63% in the population), most probably due to random sampling variations.
and more action to sensitize the public to the risks factors of CVD. Sex, age and cholesterol levels
Studied factors Each of the studied factors has been shown to be related to CVD: cholesterol levels (4, 13), triglycerides level (7), smoking, blood pressure (20), obesity (16), blood glucose level, family history of CHD (12), sedentary way of life (2), and use of oral contraceptives (23). Our data indicates that the percentage of smokers decreased with age in both sexes, but the number of cigarette-equivalent was the same at different ages. This means that if less people are smoking when getting older, the quantity of cigarettes consumed remained the same. Also, we observed fewer histories of cardiovascular risk factors in men than in women. This difference could be due to an unwillingness to discuss this topic or lack of interest for cardiovascular issues in men. The prevalence of the three main CVD risk factors (hypercholesterolemia, hypertension and smoking) was significantly higher in men than in women. The proportion of subjects being at risk for these three factors was 3 times higher in men than in women (2.2% versus 0.8%). This observation is important with regards to the predominance of cardiovascular mortality in men as compared to women (17). Cholesterol levels Table 4 compares mean total cholesterol levels in three different surveys in France, namely the Paris Prospective Survey (9), the French MONICA Project (8), and the PCV-METRA Study. The TC levels we observed in women were similar to those of the MONICA-France survey (8), which was a unique epidemiological survey of CVD risk factors in French women. For men, our TC levels were slightly higher than those of the other reports (8, 9). This might be explained by regional differences but differences in date of follow-up might also be a cause. Our HDL-C levels were similar to those observed in the MONICAFrance survey, for both sexes (8). The prevalence of hypercholesterolemic subjects was comparable to that of the United States (NHANES II) (12) as well as two cholesterol screenings performed in two large American cities in 1985 and 1986 (27). In contrast, our prevalence was lower than reported in Finland (24). We found 35% of men and 21% of women with high cholesterol levels. These figures confirm the need for large scale cholesterol screening in France
The increase of cholesterol level with age in both sexes has been amply reported in the literature (12, 27). Our study emphazises the fact that the TC and LDC-C levels increased more steeply in men than in women at early ages (from 20 to 40 years), and more steeply in women than in men at older ages (see Fig. 1). From 30 to 55 yrs, TC and LDL-C levels were significantly higher in men than in women, whereas the HDL-C level was significantly lower in men than in women. These sex differences might be explained, in part, by a depressing effect of female hormones on cholesterol levels (27). As seen in Figure 2, about 300/0 of men had high TC at age 40 yrs, while the same percentage was reached only at age 50 in women. This result is of importance in relation with the Framingham Heart Study which demonstrated a 10year difference in the CHD mortality rate between males and females at the middle decades of life (17)i CVD risk factors and public health Cardiovascular diseases are responsible for approximately one third of all deaths in France. Ischemic diseases a)~one cause more than 50,000 deaths per year and the cost has been estimated to be about 30 billion francs (18). The importance of cardiovascular prevention, especially a survey of cholesterol levels, is recognized by all health authorities. In the US, a 10% reduction of cholesterol in the general population has been estimated to represent an economical saving of 10 billion dollars (1). The preliminary results of this study demonstrate high cholesterol levels in a large percentage of workers in 20 large firms in France. The precise values of cholesterol and other CVD risk factors in different segments of this population might be of great use in the estimation of CVD in the country and in the planning of health care policies. This study combines a screening for CVD risk factors and active health care of the workers through dietary advise, help in stopping cigarette consumption and treatment of cases at high risk. Cholesterol screening coupled with physician follow-up and treatment can have a substantial impact in lowering cholesterol levels and the attendant risks of TC (27). The study is an ongoing study and the results obtained thus for have been in accordance with data from the literature. The cholesterol levels of the subjects are being followed and it will soon be possible to evaluate the results of different treatments of subjects with high CVD risks. The relatively low CHD mortality rate and a substantial prevalence of risk factors is known in France as the "French paradox". This study showed that levels of various risk factors in the sample population are high, it will be necessary to wait for follow-up data in order to document this French paradox.
699
Laurier D. et al.
Eur. J. Epidemiol.
Acknowledgements
We greatly thank the participation of Mafia Taki in the management of the survey. We thank the National Bank of Paris, l'Orral-Paris, France, Matra SA, Sextant Avionique, and Procter and Gambe-France for sponsoring this PCVMETRA study.
APPENDIX
The Prevention Cardio-Vasculaire en M6decine du Travail PCV-METRA Group: Patrick Segond, MD (chairman); Marie Fran~oise Bourillon, MD; Mich~le Chenet, MD; Philippe Corteel, PhD; Bernard Demure, MD, Marie Thrr~se Douguet, MD; Thierry Drumare, MD; Dani~le Esteve, MD; Olivier Galamand, MD; Anne Marie Giard, MD; Richard Gitel, MD; Catherine Guilbert, MD; Henry Hage, MD; Catherine Lanoiselee, MD; Isabelle Leprince, MD; Denise-Dani~le Miara, MD; Marie Christine Pasteau, MD; Martine Picard, MD; Marie Rose Pupponi, MD; Fr6drrique Szabason, MD; Patrick Taine, MD; Catherine Tarin, MD; Bernard Thaureaux, M D .
REFERENCES
1. American Heart Association and the National Heart, Lung, and Blood Institute (1990): The cholesterol facts: a summary of the evidence relating dietary fats, serum cholesterol, and coronary heart disease Circulation 81: 1721-1733. 2. Berlin J.A. and Colditz G. (1990): A meta-analysis of physical activity in the prevention of coronary heart disease - Am. J. Epidemiol. 132: 612-628. 3. Black D., James W.P.T., Besser G.M., et al. (1983): Obesity. A report of the Royal College of Physicians J. Royal College of Physicians of London 17: 5-65. 4. Bloch C. and Richard J.L. (1985): Les facteurs de risque des maladies par athrrosclrrose dans l'Etude Prospective Parisienne I. Comparison avec les 6tudes 6trangrres - Rev. Epidrmiol. et Sant6 Publ. 33: 108120. 5.
6.
7.
Caggiula A.W., Christakis G., Farrand M , et al. (1984): Multiple Risk Factors Intervention Trial (MRFIT), IV: intervention on blood lipids - Prey. Med. 10: 443-475. Cambien F., Chretien J.M., Ducimetibre P., Guize L. and Richard J.L. (1985): Is the relationship between blood pressure and Cardiovascular risk dependant on Body Mass Index? - Am. J. Epidemiol. 122: 434-442. Castelli W.P. (1986): The triglyceride issue: a view from Framingham - Am. Heart J. 112: 432-437. 700
8. Douste-Blazy P., Ruidavest J.B., Arveiller D., et al. (1988): Facteurs de risque cardio-vasculaire darts la population de deux rrgions couvertes par les registres MONICA-FRANCE: Strasbourg et Toulouse - Rev. Epidrmiol. et Sant6 Publ. 36: 342-349. 9. Ducimetibre P., Richard J.L, Cambien F., Rakotovao R. and Claude J.R. (1980): Coronary heart disease in middle-aged Frenchmen; comparisons between Paris Prospective Study, Seven Countries Study, and Pooling Project - The Lancet 1: 1346-1350. 10. Ducimetibre P., Richard J.L., Claude J.R., Warnet J.M. (1981): Les cardiopathies ischrmiques: incidence et facteurs de risque: l'Etude Prospective Parisienne Ed INSERM, Paris pp. 149. 11. Ellefson R.D. and Carawat W.T. (1987): Nutrition. In: Tietz N.W. ed. Fundamentals of clinical chemistry. Philadelphia, Pa: WB Saunders Co. 477-479. 12. The Expert Panel (1988): Report of the National Cholesterol Education Program Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults - Arch. Intern. Med. 148: 36-69. 13. Gordon D.J., Probstfield J.l., Garrison R.J., et al. (1989): High-density lipoprotein cholesterol and cardiovascular disease: four prospective American studies - Circulation 79: 8-15. 14. Hjermann I., Felve-Byre K., Holme I. and Leren P. (1981): Effect of diet and smoking intervention on the incidence of coronary heart disease - Lancet 2: 13031310. 15. Kannel W.B., Castelli I'F.P., Gordon T. and McNamara P.M. (1971): Serum cholesterol, liproproteins, and the risk of coronary heart disease: the Framingham Study - Am. Intern. Med. 74: 1-12. 16. Kannel W.B., Cupples L.A., Ramaswami R., Stokes III J., Kreger B.E. and Higgins M. (1991): Regional obesity and risk of cardiovascular disease; the Framingham study - J. Clin. Epidemiol. 44: 183-190. 17. Lerner D.J. and Kanner W.B. (1986): Patterns of coronary heart disease morbidity and mortality in the sexes: a 26-year follow-up of the Framingham population - Am. Heart J. 111: 383-390. 18. Levy E. and Le Pen C. (1990): Le cofit des maladies coronariennes. Projection - La sant6 au futur 2: 135148. 19. Lipid Research Clinics Program (1984): The Lipid Research Clinic's Coronary Primary Prevention Trial results: I. Reduction in incidence of coronary heart disease - JAMA 251: 351-364. 20. The Pooling Research Group (1978): Relationship of blood pressure, serum cholesterol, smoking habit, relative weight and ECG abnormalities to incidence of major coronary events: Final report of the Pooling Project - J. Chronic Dis. 31: 201-306.
Vol. 8, 1992
PCV-METRA study on cardiovascular risk factors
21. Puska P., Nissinen A., Salonen J.T. and Tuomilehto J. (1983): Ten years o f the North Karelia Project: results with community-based prevention o f coronary heart disease - Scand. J. Soc. Med. 11: 65-68.
frequency and cholesterol awareness in three geographical areas of Finland - Eur. Heart J. 11: 294301. 25. SAS Institute Inc (1990): SAS user's Guide, version 6, third edition, Car),, NC: SAS Institute Inc.
22. Richard J.L. (1988): Le projet MONICA. Un projet OMS de recherche cardio-vasculaire - Rev. Epid6m. et Sant6 Publ. 36: 325-334.
26. Stein E.A. (1987): Nutrition. In: Tietz N.W. ed. Fundamentals o f clinical chemistry - Philadelphia, Pa: WB Saunders Co. 471-472.
23. Rosenberg L., Hennekens C.H., Rosner B., Belanger C., Rothman K.J. and Speizer F.E. (1980): Oral contraceptive use in relation to non fatal myocardial infarction - Am. J. Epidemiol. 111: 59-66.
27.
24. Salomaa V., Korhonen H.J., Tuomilehto J., et al. (1990): Serum cholesterol distribution, measurement
701
Wynder E.L., Harris R.E. and Haley N.J. (1989): Population screening for plasma cholesterol: community-based results from Connecticut - Am. Heart J. 117: 649-656.