Socioeconomic inequalities in weight, height and body mass index from birth to 5 years.

BACKGROUND/OBJECTIVES: Studies in high-income countries show that despite the positive association of weight with socioeconomic position at birth, an inverse socioeconomic gradient in overweight (OW) appears later in childhood. The objectives were to understand the natural history of socioeconomic inequalities in weight, height and body mass index (BMI), by investigating their associations with maternal educational level between birth and 5 years, separately in boys and girls. SUBJECTS/METHODS: A published work of growth modelling between birth and 5 years allowed us to calculate predicted weight, height and BMI at 1 month, 6 months, 1, 3 and 5 years for 1735 children from the French EDEN mother-child cohort. Associations between maternal education and predicted measures of body size were analysed with marginal linear and logistic models, stratified by sex. RESULTS: In girls, despite a positive association between maternal education and birthweight, an inverse socioeconomic gradient was observed as early as 1 month for BMI. Girls whose mothers had low education levels were shorter on the whole than their counterparts with better-educated mothers, despite their similar weights. In boys, no socioeconomic gradient in BMI was observed at any age, including birth, but positive associations were found as early as 1 month for both weight and height. CONCLUSIONS: The emergence of an inverse socioeconomic gradient in BMI and OW apparently results from a complex pattern of socioeconomic inequalities in weight and height from 1 month onwards. The very start of life thus appears to be an important window of opportunity for addressing socioeconomic inequalities in growth.

To which mixtures are French pregnant women mainly exposed? A combination of the second French total diet study with the EDEN and ELFE cohort studies.

Pregnant women and their unborn child are exposed to a large number of substances during pregnancy. Some of these substances may cross the placenta, resulting in exposure of the foetus. There is growing evidence that certain substances could interact to produce a mixture effect. It is therefore essential to identify the main mixtures mothers are exposed to. This study aimed to identify the major mixtures French pregnant women included in EDEN and ELFE cohorts were exposed to, on the basis of the 441 substances analysed in the second French total diet study. Exposure systems and the composition of substances were identified from co-exposures using sparse non-negative matrix under-approximation to generate the main mixtures. Individuals were clustered to define clusters with similar co-exposure profiles. Six clusters associated with eight mixtures were identified. For example in ELFE, cluster 2 comprising 10% of the population was characterised by mixtures "Pest-1" mainly contains pesticides and "TE-F-PAH″ contains trace elements, furans and polycyclic aromatic hydrocarbons. Five other clusters were also described with their associated mixtures. Similar results were observed for EDEN. This study helps to prioritise mixtures for which it is crucial to investigate possible toxicological effects and to recommend epidemiological studies concerning health effects.

Influence of infant feeding patterns over the first year of life on growth from birth to 5 years.

OBJECTIVES: As early-life feeding experiences may influence later health, we aimed to examine relations between feeding patterns over the first year of life and child's growth in the first 5 years of life. METHODS: Our analysis included 1022 children from the EDEN mother-child cohort. Three feeding patterns were previously identified, i.e. 'Later dairy products introduction and use of ready-prepared baby foods' (pattern-1), 'Long breastfeeding, later main meal food introduction and use of home-made foods' (pattern-2) and 'Use of ready-prepared adult foods' (pattern-3). Associations between the feeding patterns and growth [weight, height and body mass index {BMI}] were analysed by multivariable linear regressions. Anthropometric changes were assessed by the final value adjusted for the initial value. RESULTS: Even though infant feeding patterns were not related to anthropometric measurements at 1, 3 and 5 years, high scores on pattern-1 were associated with higher 1-3 years weight and height changes. High scores on pattern-2 were related to lower 0-1 year weight and height changes, higher 1-5 years weight and height changes but not to BMI changes, after controlling for a wide range of potential confounding variables including parental BMI. Scores on pattern-3 were not significantly related to growth. Additional adjustment for breastfeeding duration reduced the strength of the associations between pattern-2 and growth but not those between pattern-1 and height growth. CONCLUSION: Our findings emphasize the relevance of considering infant feeding patterns including breastfeeding duration, age of complementary foods introduction as well as type of foods used when examining effects of early infant feeding practices on later health. © 2017 World Obesity Federation.

Prospective associations between energy balance-related behaviors at 2 years of age and subsequent adiposity: the EDEN mother-child cohort.

BACKGROUND/OBJECTIVES: Sedentary behavior, physical activity and dietary behavior are formed early during childhood and tend to remain relatively stable into later life. No longitudinal studies have assessed the independent influence of these three energy balance-related behaviors during toddlerhood on later adiposity. We aimed to analyze the associations between TV/DVD watching time, outdoor play time and dietary patterns at the age of 2 years and child adiposity at the age of 5 years, in boys and girls separately. SUBJECTS/METHODS: This study included 883 children from the French EDEN mother-child cohort. TV/DVD watching time, outdoor play time and dietary intakes were reported by parents in questionnaires when the child was aged 2 years. Two dietary patterns, labeled 'Guidelines' and 'Processed, fast foods', were identified in a previous study. The percentage of body fat (%BF) based on bioelectrical impedance analysis and body mass index were measured at the age of 5 years. RESULTS: In boys, TV/DVD watching time at the age of 2 years was positively associated with %BF at the age of 5 years (ß=0.50 (95% confidence interval: 0.001, 1.00) for those boys with ⩾60 min per day of TV/DVD watching time vs those with ⩽15 min per day, P-value for trend 0.05). In girls, outdoor play was inversely associated with %BF (ß=-0.96 (95% confidence interval: -1.60, -0.32) for those in the highest tertile of outdoor play time vs those in the lowest tertile, P=0.001). Overall, at the age of 2 years, dietary patterns were associated with both TV/DVD watching time and outdoor play time, but no significant and independent association was observed between dietary patterns and later adiposity. CONCLUSION: This study shows longitudinal and gender-differentiated relations between both TV/DVD watching time and outdoor play time in toddlerhood and later adiposity, whereas evidence for a relation between dietary patterns and subsequent fat development was less conclusive. Early childhood-by the age of 2 years-should be targeted as a critical time for promoting healthy energy balance-related behaviors.

Maternal weight prior and during pregnancy and offspring's BMI and adiposity at 5-6 years in the EDEN mother-child cohort.

BACKGROUND: Beyond pre-pregnancy BMI, maternal weight change before and during pregnancy may also affect offspring adiposity. OBJECTIVE: To investigate the relationship between maternal weight history before and during pregnancy with children's adiposity at 5-6 years. METHODS: In 1069 mother-child dyads from the EDEN Cohort, we examined by linear regression the associations of children's BMI, fat mass and abdominal adiposity at 5-6 years with maternal pre-pregnancy BMI, pre-pregnancy average yearly weight change from age 20 and gestational weight gain. The shapes of relationships were investigated using splines and polynomial functions were tested. RESULTS: Children's BMI and adiposity parameters were positively associated with maternal pre-pregnancy BMI, but these relationships were mainly seen in thin mothers, with no substantial variation for maternal BMI ranging from 22 to 35 kg/m2 . Gestational weight gain was positively associated with children's BMI Z-score, but again more so in thin mothers. We found no association with pre-pregnancy weight change. CONCLUSIONS: Before the adiposity rebound, maternal pre-pregnancy thinness explains most of the relationship with children's BMI. The relationship may emerge at older ages in children of overweight and obese mothers, and this latency may be an obstacle to early prevention.

Early factors related to carbohydrate and fat intake at 8 and 12 months: results from the EDEN mother-child cohort.

BACKGROUND/OBJECTIVES: Few studies have examined the factors explaining the variability in fat and carbohydrate intake during infancy. We aimed to describe infants' fat and carbohydrate intake and analyse the associations with infant and maternal characteristics and feeding practices. SUBJECTS/METHODS: This study included 1275 infants aged 8 months from the French EDEN mother-child cohort. Carbohydrate intake, fat intake, added fat (vegetable oils and animal fats) and added sugar (honey, white sugar, brown sugar, jam and sweetened beverages) consumption were calculated at 8 and 12 months. Associations between these variables and infant and maternal characteristics as well as maternal dietary patterns during pregnancy, breast-feeding duration and age at complementary feeding introduction were analysed using multivariable linear and logistic regressions. RESULTS: Less than 5% of non-breast-fed infants reached the recommendation of consuming at least 40% of total energy from fat, whereas more than 95% of them reached 45% of energy from carbohydrates. Overall, infant and maternal characteristics and maternal diet during pregnancy were marginally associated with both carbohydrate/added sugar and fat/added fat intake. Nevertheless, age at complementary feeding introduction was associated with all outcomes. CONCLUSIONS: Our results suggest that only a small proportion of non-breast-fed infants at 8 and 12 months reached the recommendations for fat intake, whereas a majority of them reached the recommendations for carbohydrate intake. As subgroups of infants with a higher risk of inadequate diet were not identified, the present results call for an improved dissemination of information regarding infant-specific dietary fat needs in the entire population.

Dietary acrylamide intake during pregnancy and anthropometry at birth in the French EDEN mother-child cohort study.

BACKGROUND AND AIM: Acrylamide is a contaminant formed in a wide variety of carbohydrate-containing foods during frying or baking at high temperatures. Recent studies have suggested reduced foetal growth after exposure to high levels of acrylamide during pregnancy. OBJECTIVE: To study the relationship between maternal dietary acrylamide intake during pregnancy and their offspring's anthropometry at birth. DESIGN: In our population of 1471 mother-child pairs from two French cities, Nancy and Poitiers, dietary acrylamide intake during pregnancy was assessed by combining maternal food frequency questionnaires with data on food contamination at the national level, provided by the second "French Total Diet Study". Newborns weighing less than the 10th percentile, according to a customised definition, were defined as small for gestational age (SGA). Linear and logistic regression models were used to study continuous and binary outcomes respectively, adjusting for the study centre, maternal age at delivery, height, education, parity, smoking during pregnancy, the newborn's gestational age at birth and sex. RESULTS: The median and interquartile range of dietary acrylamide intake were 19.2µg/day (IQR, 11.8;30.3). Each 10µg/day increase in acrylamide intake was associated with an odds-ratio for SGA of 1.11 (95% Confidence Interval: 1.03,1.21), birth length change of -0.05cm (95% CI: -0.11,0.00) and birth weight change of -9.8g (95% CI: -21.3,1.7). CONCLUSIONS: Our results, consistent with both experimental and epidemiological studies, add to the evidence of an effect of acrylamide exposure on the risk of SGA and suggest an effect on foetal growth, for both weight and length.

Infant feeding patterns over the first year of life: influence of family characteristics.

BACKGROUND/OBJECTIVES: Early eating patterns and behaviors can determine later eating habits and food preferences and they have been related to the development of childhood overweight and obesity. We aimed to identify patterns of feeding in the first year of life and to examine their associations with family characteristics. SUBJECTS/METHODS: Our analysis included 1004 infants from the EDEN mother-child cohort. Feeding practices were assessed through maternal self-report at birth, 4, 8 and 12 months. Principal component analysis was applied to derive patterns from breastfeeding duration, age at complementary food (CF) introduction and type of food used at 1 year. Associations between patterns and family characteristics were analyzed by linear regressions. RESULTS: The main source of variability in infant feeding was characterized by a pattern labeled 'late CF introduction and use of ready-prepared baby foods'. Older, more educated, primiparous women with high monthly income ranked high on this pattern. The second pattern, labeled 'longer breastfeeding, late CF introduction and use of home-made foods' was the closest to infant feeding guidelines. Mothers ranking high on this pattern were older and more educated. The third pattern, labeled 'use of adults' foods' suggests a less age-specific diet for the infants. Mothers ranking high on this pattern were often younger and multiparous. Recruitment center was related to all patterns. CONCLUSIONS: Not only maternal education level and age, but also parity and region are important contributors to the variability in patterns. Further studies are needed to describe associations between these patterns and infant growth and later food preferences.

Pre-pregnancy body mass index and weight gain during pregnancy: relations with gestational diabetes and hypertension, and birth outcomes.

To study the relationship between pre-pregnancy body mass index (BMI) and weight gain during pregnancy with pregnancy and birth outcomes, with a focus on gestational diabetes and hypertension and their role in the association with fetal growth. We studied 1,884 mothers and offspring from the Eden mother-child cohort. Weight before pregnancy (W1) and weight after delivery (W2) were collected and we calculated BMI and net gestational weight gain (netGWG = (W2 - W1)/(weeks of gestation)). Gestational diabetes, hypertension gestational age and birth weight were collected. We used multivariate linear or logistic models to study the association between BMI, netGWG and pregnancy and birth outcomes, adjusting for center, maternal age and height, parity and average number of cigarettes smoked per day during pregnancy. High BMI was more strongly related to the risk of giving birth to a large-for-gestational-age (LGA) baby than high netGWG (odds ratio OR [95% CI] of 3.23 [1.86-5.60] and 1.61 [0.91-2.85], respectively). However, after excluding mothers with gestational diabetes or hypertension the ORs for LGA, respectively weakened (OR 2.57 [1.29-5.13]) for obese women and strengthened for high netGWG (OR 2.08 [1.14-3.80]). Low in comparison to normal netGWG had an OR of 2.18 [1.20-3.99] for pre-term birth, which became stronger after accounting for blood pressure and glucose disorders (OR 2.70 [1.37-5.34]). Higher net gestational weight gain was significantly associated with an increased risk of LGA only after accounting for blood pressure and glucose disorders. High gestational weight gain should not be neglected in regard to risk of LGA in women without apparent risk factors.

Influence of fetal and parental factors on intrauterine growth measurements: results of the EDEN mother-child cohort.

OBJECTIVE: In small-for-gestational-age neonates, parental and fetal characteristics can be used to distinguish between constitutionally small size and growth restriction, which is associated with a higher risk of morbidity and mortality. The aim of this study was to quantify relationships of parental and fetal characteristics with fetal ultrasound measurements. METHODS: The EDEN mother-child cohort included 2002 pregnant women with singleton pregnancies attending one of two university hospitals. Data from two routine ultrasound examinations for fetal biometry were recorded, at 20-25 and 30-35 weeks of gestation. Biparietal diameter (BPD), head circumference (HC), femur length (FL), abdominal circumference (AC) and estimated fetal weight (EFW) were studied as a function of prepregnancy maternal body mass index (BMI), maternal height, paternal height, fetal sex and gestational age. RESULTS: Data were obtained at the first scan from 1833 women and at the second scan from 1752 women. Parental anthropometric characteristics were significantly associated with ultrasound measurements at both scans. Maternal BMI was more strongly associated with AC and EFW, whereas both maternal and paternal height were more strongly associated with FL. An association was also found between fetal sex and all ultrasound measurements other than FL. CONCLUSION: Maternal and paternal anthropometric characteristics are significantly associated with ultrasound measurements in mid to late pregnancy. These relationships provide support for the use of these characteristics in ultrasound fetal size reference charts.

Confirmation of associations between ion channel gene SNPs and QTc interval duration in healthy subjects.

Population-based association studies have identified several polymorphic variants in genes encoding ion channel subunits associated with the electrocardiographic heart-rate-corrected QT (QTc) length in healthy populations of Caucasian origin (KCNH2 rs1,805,123 (K897 T) and rs3,815,459, SCN5A rs1,805,126 (D1,819D), 1,141-3 C>A, rs1,805,124 (H558R), and IVS24+116 G>A, KCNQ1 rs757,092, KCNE1 IVS2-128 G>A and rs1,805,127 (G38S), and KCNE2 rs2,234,916 (T8A)). However, few of these results have been replicated in independent populations. We tested the association of SNPs KCNQ1 rs757,092, KCNH2 rs3,815,459, SCN5A IVS24+116 G>A, KCNE1 IVS2-128 G>A and KCNE2 rs2,234,916 with QTc length in two groups of 200 subjects presenting the shortest and the longest QTc from a cohort of 2,008 healthy subjects. All polymorphisms were in Hardy-Weinberg equilibrium in both groups. The minor allele SCN5A IVS24+116 A was more frequent in the group of subjects with the shortest QTc, whereas the minor alleles KCNQ1 rs757,092 G and KCNH2 rs3,815,459 A were more frequent in the group with the longest QTc. There was no significant difference for KCNE1 IVS2-128 G>A and KCNE2 rs2,234,916 between the two groups. Haplotype analysis showed a twofold increased risk of QTc lengthening for carriers of the haplotype, combining alleles C and A of the two common KCNE1 SNPs, IVS2-129 C>T (rs2,236,609) and rs1,805,127 (G38S), respectively. In conclusion, our study confirms the reported associations between QTc length and KCNQ1 rs757,092 and KCNH2 rs3,815,459.

Two hour plasma glucose is not unequivocally predictive for early death in men with impaired fasting glucose: more results from the Paris Prospective Study.

AIMS/HYPOTHESIS: We examined whether the 2-h plasma glucose (2 hPG) concentration after a 75 g OGTT is predictive of death in men with a diabetic, an impaired or a normal fasting plasma glucose concentration (DM-FPG: > or =7.0 mmol/l; IFG: 6.1-6.9 mmol/l; normal-FPG: <6.1 mmol/l). METHODS: The 17-year mortality of 7018 men, aged 44 to 55 years, from the Paris Prospective Study, who were not known to be diabetic at baseline was studied. RESULTS: The 2 hPG was not associated with early mortality in men with a DM-FPG in contrast to men with an IFG or a normal-FPG; for an increase from 10 to 11 mmol/l in the 2 hPG, the age-adjusted hazards ratios were 1.01 (95% CI 0.95-1.08), 1.15 (1.03-1.28) and 1.24 (1.18-1.31) respectively. Coronary heart disease mortality and within this category sudden death but not ischaemic heart disease death, were related with 2 hPG but only in the men with normal FPG. However, the prediction by 2 hPG did not differ between the men with DM-FPG, an IFG or a normal-FPG: the overall age-adjusted hazards ratios for these three causes of death were 1.09 (1.00-1.18), 1.13 (1.02-1.26) and 1.13 (0.99-1.29), respectively. CONCLUSION/INTERPRETATION: 2 hPG is unequivocally prognostic for all-cause mortality only in men with normal FPG. Screening men with an IFG by using a 75 g OGTT is of limited benefit.

Alcohol intake and fasting insulin in French men and women. The D.E.S.I.R. Study.

BACKGROUND: To study the relation between alcohol consumption and the fasting insulin concentration in a French population with a range of alcohol intakes. METHODS: 2.406 men and 2.500 women, aged 30 to 65 years who were not known as diabetic and with a non-diabetic fasting plasma glucose<7.0 mmol/l were studied. Insulin was assayed by a specific micro-enzyme immunoassay and alcohol intake was from a self-questionnaire. RESULTS: Fasting insulin concentration showed an inverse linear association with alcohol consumption, after adjustment for age and possible confounding factors (p for trend<0.0001 men; p<0.002 women), with a 29% higher insulin in non-drinkers compared to very heavy drinkers (> 80 g/day) in men (p<0.0001) and a 23% and 26% difference when compared to heavy drinkers (41-80 g/day) in men and women respectively (p<0.0001, p<0.003). This relation did not differ significantly according to whether the alcohol was consumed as wine, beer/cider or spirits. Fasting plasma glucose modified the relation between alcohol and insulin in men: while the negative relation alcohol-insulin was strong for fasting plasma glucose<6.0 mmol/l (p<0.0001), there was no association above 6.0 mmol/l (p=0.4). CONCLUSION: There is an inverse relation between alcohol consumption and fasting insulin concentrations. Some studies have found a U shaped relation, and this is probably due to the inclusion of diabetic subjects. As hyperinsulinemia has been shown to be positively associated with cardiovascular disease, it may be one of the variables that explains the protective effect of moderate alcohol consumption on cardiovascular disease.

Is microalbuminuria an integrated risk marker for cardiovascular disease and insulin resistance in both men and women?

BACKGROUND: Microalbuminuria is associated with some cardiovascular risk factors. However, it is presently unclear whether microalbuminuria is associated with other cardiovascular risk factors and markers of insulin resistance in both men and women. DESIGN: The baseline parameters of the on-going prospective D.E.S.I.R. Study (Data from an Epidemiological Study on the Insulin Resistance syndrome) were analysed. METHODS: The subjects (n = 3878, age 30-64) had a medical and biological checkup including arterial pressure, ECG, lipids, haematocrit, leukocyte count, insulin and urinary albumin concentration. RESULTS: Arterial pressure was significantly higher in microalbuminuric than in normoalbuminuric men and women. The heart rate and the prevalence of smoking were higher in microalbuminuric than in normoalbuminuric men, but not in women. Total- and LDL-cholesterol, triglycerides, apolipoprotein-B, leukocyte count, uric acid, haematocrit, haemoglobin and red cell volume were greater in microalbuminuric than in normoalbuminuric men. Total- and HDL-cholesterol were lower whereas triglycerides and uric acid were significantly greater in microalbuminuric than in normoalbuminuric women. The 9-year calculated absolute cardiovascular risk was higher in microalbuminuric than in normoalbuminuric men (4.18% vs. 2.79%, respectively, P < 0.0001) but was similar in women (0.75% vs. 0.69%, respectively, NS). Fasting blood glucose and insulin were higher in microalbuminuric than in normoalbuminuric men, but not in women. A multiple regression showed that microalbuminuria was associated with systolic arterial pressure, fasting glucose, leukocyte count and haematocrit in men, and with systolic arterial pressure and triglycerides in women. CONCLUSIONS: Microalbuminuria is associated with most cardiovascular risk factors and markers of insulin resistance in men, but no consistent association is observed in women. The determinants of microalbuminuria appear different in men and women, suggesting a different pathophysiology and perhaps significance of microalbuminuria in men and women.

Hypertension in four African-origin populations: current 'Rule of Halves', quality of blood pressure control and attributable risk of cardiovascular disease.

OBJECTIVE: To assess the public health burden from high blood pressure and the current status of its detection and management in four African-origin populations at emerging or high cardiovascular risk. DESIGN: Cross-site comparison using standardized measurement and techniques. SETTING: Rural and urban Cameroon; Jamaica; Manchester, Britain. SUBJECTS: Representative population samples in each setting. African-Caribbeans (80% of Jamaican origin) and a local European sample in Manchester. MAIN OUTCOME MEASURES: Cross-site age-adjusted prevalence; population attributable risk. RESULTS: Among 1,587 men and 2,087 women, age-adjusted rates of blood pressure > or =160 or 95 mmHg or its treatment rose from 5% in rural to 17% in urban Cameroon, despite young mean ages, to 21% in Jamaica and 29% in Caribbeans in Britain. Treatment rates reached 34% in urban Cameroon, and 69% in Jamaican- and British-Caribbean-origin women. Sub-optimal blood pressure control (> 140 and 90 mmHg) on treatment reached 88% in European women. Population attributable risks (or fractions) indicated that up to 22% of premature all-cause, and 45% of stroke mortality could be reduced by appropriate detection and treatment. Additional benefit on just strokes occurring on treatment could be up to 47% (e.g. in both urban Cameroon men and European women) from tighter blood pressure control on therapy. Cheap, effective therapy is available. CONCLUSION: With mortality risk now higher from non-communicable than communicable diseases in sub-Saharan Africa and elsewhere, systematic measurement, detection and genuine control of hypertension once treated can go hand-in-hand with other adult health programmes in primary care. Cost implications are not great. The data from this collaborative study suggest that such efforts should be well rewarded.

Association between heart rate-corrected QT interval and coronary risk factors in 2,894 healthy subjects (the DESIR Study). Data from an Epidemiological Study on the Insulin Resistance syndrome.

In the prospective Data from an Epidemiological Study on the Insulin Resistance Syndrome, 2,894 healthy subjects aged 30 to 64 years had determinations of fasting glucose, insulin, serum lipid and fibrinogen concentrations, blood pressures, body mass index, and waist-hip ratio, as well as tobacco and alcohol consumptions and physical activity. A 12-lead electrocardiogram with automatic measurement of the QT interval was recorded and the formula used for heart rate correction was based on the best-fit regression between QT and heart rate. The QT duration was influenced by glucose homeostasis in both sexes, and increased in men with physical activity; there was a dose-effect relation for men who smoked.

[Totally reimbursed care for type 2 diabetes in 1994: characteristics and follow-up]. / Le diabète de type II pris en charge à 100 % en 1994 : caractéristiques et suivi.

Non-insulin-dependent diabetes mellitus (NIDDM) is an important health issue in France. According to projections from current data, the problem should increase over the upcoming years. The aim of this study was to determine the characteristics of patients with NIDDM who are enrolled on the list of 30 long term diseases (ALD30) and examine their follow-up and disease-related complications. This study was performed with the data base established by the French National Health Insurance for Salaried Workers (CMANTS) in 1994. Diagnosis and medical information were coded by official physicians, who either examined the patients or consulted their primary care physician (69% of the cases), or relied on available medical records. Eight thousand two hundred forty-seven patients (sex ration=0.95) with NIDDM enrolled on the long term disease list were included. Mean age was 65.1+/- 0.13 years (m+/- SEM). More than one blood glucose level per year was available for 89% of the patients and an HA1c assay in only 34%. In addition, for patient follow-up at least once a year, 80%, 47%, and 42% of the patients had a serum creatinine assay, a funus examination, or an electorcariogram respectively. Irrespective of the criteria retained (fasting glucose=1.26 g/l; fasting glucose=1.40 g/l) there was a high percentage of patients whose level was well above the limit (75% and 63% respectively). For HbA1c, only a quarter of the patients had values lower than the limits defined by the ANAES and AFFSAPS guidelines. The percentage of diabetic patients without complications was 41% (hypertension excluded). The most frequently observed complications isolated peripheral vascular conditions (31%). These results were obtained from a database created in 1994, before guidelines on type 2 diabetes were officially published. They can serve as a reference to follow medical practices.

[Pharmacoepidemiology of diabetes: assessment of good use of oral antidiabetic drugs]. / Pharmacoépidémiologie du diabète : évaluation du bon usage des antidiabétiques oraux.

Diabetes' treatment is specific; therefore diabetic patients can be identified by the prescription of antidiabetic drugs. This pathology is characterized by chronicity and occurrence of complications which need additional medications. All the drugs have official use guidelines due to drug interactions and/or physiological status of the patients. To evaluate how these guidelines are followed in general practice, we used a data base from the French Health care System. Two thousand eight hundred and ninety eight (2 898) prescriptions on which, at least, one oral antidiabetic drug was noticed, were analysed. The number of drugs on each prescription was between 1 and 18 (mean (5,6 +/- 2,9)). Considering the drugs which need precautions of use in association with sulfonylureas, coprescription occurred with B-bloquants and non steroid anti-inflammatory drugs in 18 and 17 % of the cases respectively. The drugs which need precautions of use in association with biguanides were diuretics, oestrogens and/or progestatives and steroids, coprescribed in 28, 1, 1 % of the cases respectively. Percentages of coprescriptions which need precautions of use were quite high. The relevance of detected coprescriptions. and the compatibility between recommendations for "good prescription" and "real prescription practice" should be taken into account considering diabetic patients' characteristics.

Analysis of mortality in French diabetic patients from death certificates: a comparative study.

This study was implemented in France to determine the causes of death in diabetic patients, whether diabetes was mentioned or not on the death certificate, and to assess the underestimation of the prevalence of diabetes at death. Two stratified random samples of death certificates were selected in the national mortality data base. The first included certificates mentioning diabetes as a cause of death (cases). The second, included certificates with no mention of diabetes (controls). For each certificate, a record form was sent to the certifying physician to ascertain diabetes in the first group and to trace unrecorded diabetes in the second group. In case of diabetes, the characteristics of the patient and his disease were collected (age at onset, treatment, complications ...). We obtained complete data for 325 cases and 959 controls. Among cases, 1% of the subjects were not confirmed as diabetic, while almost 10% of the controls were identified as having diabetes. The corresponding ratio of the corrected prevalence at death to that provided by the French statistics was estimated to 4.0 in men and 3.1 in women. Particular features are that 2% of the total diabetic decedents died from acute metabolic complications (diabetic or hyperosmolar coma, acidoketosis, or acute hypoglycemia), and that 33% of the unreported diabetic decedents under 45 died from trauma or poisoning. These results show that in France, the death rates published in the statistics for diabetes dramatically underestimate the impact of diabetes. A high risk of death is linked to this disease, particularly in people aged under 45, a problem that health deciders should address.

Relationships between abdominal body fat distribution and cardiovascular risk factors: an explanation for women's healthier cardiovascular risk profile. The D.E.S.I.R. Study.

OBJECTIVE: To assess whether the extent of abdominal fat distribution, as measured by the waist to hip ratio (WHR), might account for the sex differences in the levels of cardiovascular risk factors. DESIGN: Cross-sectional age-matched study. SUBJECTS: 1264 men and 1264 premenopausal women, aged 30-49 y, free from known cardiovascular diseases and diabetes, included in the prospective study, D.E.S.I.R. MEASUREMENTS: (1) body mass index (BMI), WHR and blood pressures; (2) fasting concentrations of blood glucose, insulin, lipids and lipoprotein subfractions, and apolipoproteins; and (3) smoking status, physical activity, and alcohol consumption. RESULTS: After taking into account age and BMI, there were gradual relationships, within and across sexes, between WHR and the levels of most lipids and lipoproteins, of fasting glucose and insulin, and, to a lesser extent, of blood pressures. In particular, men and women with similar BMI and WHR had similar levels of triglycerides. Multivariate regression analysis showed that the variance of cardiovascular risk factors explained by the model was increased when sex was included, after controlling for age, BMI and lifestyle habits (all P<0.01). If WHR was included in the model, sex had no additional effect on total cholesterol (P>0.09 for change in total r2 ) or triglycerides (P>0.40 for change in total r2). In contrast, for other cardiovascular risk factors, adjustment for covariates and WHR did not fully eliminate the sex differences, although WHR increased the variance explained with or without additional control for sex (all P<0.01). CONCLUSION: The continuous increase of cardiovascular risk factors with WHR, especially for lipids and lipoproteins, suggests that the abdominal body fat distribution may partially explain the relative unhealthier cardiovascular risk profile of men.