Interaction of Diet/Lifestyle Intervention and TCF7L2 Genotype on Glycemic Control and Adiposity among Overweight or Obese Adults: Big Data from Seven Randomized Controlled Trials Worldwide

Objective. The strongest locus which associated with type 2 diabetes (T2D) by the common variant rs7903146 is the transcription factor 7-like 2 gene (TCF7L2). We aimed to quantify the interaction of diet/lifestyle interventions and the genetic effect of TCF7L2 rs7903146 on glycemic traits, body weight, or waist circumference in overweight or obese adults in several randomized controlled trials (RCTs).Methods. From October 2016 to May 2018, a large collaborative analysis was performed by pooling individual-participant data from 7 RCTs. These RCTs reported changes in glycemic control and adiposity of the variant rs7903146 after dietary/lifestyle-related interventions in overweight or obese adults. Gene treatment interaction models which used the genetic effect encoded by the allele dose and common covariates were applicable to individual participant data in all studies.Results. In the joint analysis, a total of 7 eligible RCTs were included (n = 4,114 ). Importantly, we observed a significant effect modification of diet/lifestyle-related interventions on the TCF7L2 variant rs7903146 and changes in fasting glucose. Compared with the control group, diet/lifestyle interventions were related to lower fasting glucose by -3.06 (95% CI, -5.77 to -0.36) mg/dL (test for heterogeneity and overall effect: I 2 = 45.1 % , p < 0.05 ; z = 2.20 , p = 0.028 ) per one copy of the TCF7L2 T risk allele. Furthermore, regardless of genetic risk, diet/lifestyle interventions were associated with lower waist circumference. However, there was no significant change for diet/lifestyle interventions in other glycemic control and adiposity traits per one copy of TCF7L2 risk allele.Conclusions. Our findings suggest that carrying the TCF7L2 T risk allele may have a modestly greater benefit for specific diet/lifestyle interventions to improve the control of fasting glucose in overweight or obese adults.

Once-weekly subcutaneous semaglutide treatment for persons with type 2 diabetes: Real-world data from a diabetes out-patient clinic

AIMS

The once-weekly administered glucagon-like peptide 1 (GLP-1) receptor agonist (GLP-1RA) semaglutide, has, in clinical trials, demonstrated significant reductions in glycated haemoglobin A_1c (HbA_1c ) and body weight in persons with type 2 diabetes. We evaluated the real-world clinical effects of semaglutide once weekly in a hospital-based diabetes out-patient clinic.

METHODS

This retrospective observational cohort study included persons with type 2 diabetes (n = 119) on a broad range of antidiabetic medicine: GLP-1RA naïve persons (n = 37) and GLP-1RA-experienced persons (n = 82). Person characteristics at inclusion: age [median (quartiles)]: 65 (57, 72) years; body weight 99 (86, 118) kg; body mass index (BMI) 33 (29, 38) kg/m²; HbA_1c 61 (54, 69) mmol/mol/(7.7 (7.1, 8.5) %). Data were collected at baseline and after 3, 6 and 12 months of semaglutide treatment. Data were analysed using a general linear mixed model for repeated measurements.

RESULTS

After 12 months, the reductions in HbA_1c were (mean [95% confidence interval]: GLP-1RA naïve: -12.8 [-17.0, -8.5] mmol/mol/ -1.2 [-1.6, -0.8]% (p < 0.01) and GLP-1RA experienced: -6.4 [-9.0, -3.8] mmol/mol/ -0.6 [-0.8, -0.4]% (p < 0.01), respectively. Body weight reductions in GLP-1RA naïve: -5 [-6.9, -3.1] kg (p < 0.01) and GLP-1RA experienced: -3.2 [-4.4, -2.0] kg (p < 0.01), respectively. Seventy-five percent received 1 mg QW semaglutide.

CONCLUSION

We observed effects of semaglutide once weekly on HbA_1c and body weight comparable with the effects observed in clinical studies with fewer persons in our cohort receiving maximum dose of semaglutide.

Genetic markers of abdominal obesity and weight loss after gastric bypass surgery

Background

Weight loss after bariatric surgery varies widely between individuals, partly due to genetic differences. In addition, genetic determinants of abdominal obesity have been shown to attenuate weight loss after dietary intervention with special attention paid to the rs1358980-T risk allele in the VEGFA locus. Here we aimed to test if updated genetic risk scores (GRSs) for adiposity measures and the rs1358980-T risk allele are linked with weight loss following gastric bypass surgery.

Methods

Five hundred seventy six patients with morbid obesity underwent Roux-en-Y gastric bypass. A GRS for BMI and a GRS for waist-hip-ratio adjusted for BMI (proxy for abdominal obesity), respectively, were constructed. All patients were genotyped for the rs1358980-T risk allele. Associations between the genetic determinants and weight loss after bariatric surgery were evaluated.

Results

The GRS for BMI was not associated with weight loss (β = -2.0 kg/100 risk alleles, 95% CI -7.5 to 3.3, p = 0.45). Even though the GRS for abdominal obesity was associated with an attenuated weight loss response adjusted for age, sex and center (β = -14.6 kg/100 risk alleles, 95% CI -25.4 to -3.8, p = 0.008), it was not significantly associated with weight loss after adjustment for baseline BMI (β = -7.9 kg/100 risk alleles, 95% CI -17.5 to 1.6, p = 0.11). Similarly, the rs1358980-T risk allele was not significantly associated with weight loss (β = -0.8 kg/risk allele, 95% CI -2.2 to 0.6, p = 0.25).

Discussion

GRSs for adiposity derived from large meta-analyses and the rs1358980-T risk allele in the VEGFA locus did not predict weight loss after gastric bypass surgery. The association between a GRS for abdominal obesity and the response to bariatric surgery may be dependent on the association between the GRS and baseline BMI.

Genetic Determinants of Weight Loss After Bariatric Surgery

BACKGROUND

The weight loss after bariatric surgery shows considerable individual variation. Twin studies of response to dietary interventions and studies of bariatric surgery patients suggest that genetic differences may play a role. This study aimed to examine the effect of three genetic risk scores on the inter-individual variation in excess body mass index loss (EBMIL) after Roux-en-Y gastric bypass. Furthermore, we searched among known adiposity-related single nucleotide polymorphisms (SNPs) for genetic determinants of the inter-individual variation in EBMIL.

METHODS

Patients with morbid obesity underwent Roux-en-Y gastric bypass and were genotyped (n = 577). Two genetic risk scores for weight loss after bariatric surgery and a genetic risk score for body mass index were calculated. Associations between the genetic risk scores and EBMIL were evaluated. Lasso regression was performed on 126 SNPs known to be associated with adiposity.

RESULTS

The average EBMIL was 76.9% (range 21.7-149.2%). EBMIL was 81.1% (SD 20.6) and 73.9% (SD 21.7) in the high and low tertile groups of a genetic risk score for weight loss. Patients with a low genetic risk score for body mass index (in the lowest 5% percentile) had an EBMIL of 68.8% (SD 20.6, p = 0.018). Thirteen adiposity-related SNPs were identified to associate with EBMIL through lasso regression.

DISCUSSION

A genetic risk score was associated with EBMIL after bariatric surgery, but may not yet be applicable to clinical practice. Patients genetically predisposed to low body mass index had lower weight loss after bariatric surgery.

Is abdominal obesity at baseline influencing weight changes in observational studies and during weight loss interventions?

Background

Body fat distribution is a marker of metabolic health independent of body size. Visceral fat accumulation has been suggested to result from a decreased expandability of the subcutaneous fat depots. Furthermore, the visceral fat may be easier to mobilize than the peripheral fat. We examined whether differences in abdominal obesity at baseline influenced prospective body-weight changes.

Objective

In this study we examined whether body-fat distribution at baseline was associated with long-term and short-term weight changes.

Design

We included 3 observational studies (ntotal = 7271) with mean follow-up times of 5-9 y and two 8-10-wk weight loss intervention studies (ntotal = 1091). We examined the association between baseline waist circumference and weight changes in a substitution regression model, where body weight, height, and fat-free mass were fixed so that a difference in waist circumference would reflect a difference in body fat distribution alone. The results were summarized in meta-analyses.

Results

In the observational studies, we found no associations between baseline waist circumference and subsequent weight change in men (β: 0.03 kg; 95% CI: -0.01, 0.08 kg; P = 0.19), but a negligible inverse association in women (β: -0.05 kg; 95% CI: -0.08, -0.01 kg; P = 0.01). There was no association between baseline waist circumference and weight loss in the intervention studies (men: β: -0.05 kg; 95% CI: -0.13, 0.03 kg; P = 0.25; women: β: -0.00 kg; 95% CI: -0.03, 0.03 kg; P = 0.84). However, in all studies, the SDs of the weight change residuals were greater, the greater the waist circumference at baseline. This trend was statistically significant in women in most studies as well as in men in 1 of the studies.

Conclusions

With narrow CIs in 3 observational studies and 2 weight loss interventions, we did not find any clinically or epidemiologically relevant association between baseline abdominal obesity and weight change. However, the present study suggests that a greater baseline abdominal obesity is a marker for greater weight fluctuations. The CCHS trial was registered at www.clinicaltrials.gov as NCT02993172. The Health2006 trial was registered at www.clinicaltrials.gov as NCT00316667. The ORG study was conducted before trial registration was required. The NUGENOB trial was registered at www.isrctn.com as ISRCTN25867281. The DiOGenes trial was registered atwww.clinicaltrials.gov as NCT00390637.

Prospective Studies Exploring the Possible Impact of an ID3 Polymorphism on Changes in Obesity Measures

OBJECTIVE

Changes in fat mass depend on adipogenesis and angiogenesis, mechanisms regulated by the inhibitor of differentiation-3 (ID3). Id3 knockout mice showed attenuated increases in BMI and visceral fat mass. We hypothesized that the ID3 missense variant (rs11574-A) would lead to an attenuated increase over time in fat mass, BMI, waist circumference (WC), and waist-hip ratio (WHR) in humans.

METHODS

The genotyped study populations included the Obesity Research Group - Genetics (ORGGEN) cohort, a cohort of men with obesity (N = 716) and of randomly selected men (N = 826) from the Danish draft register who were examined at mean ages of 20 and 46 years, and the Inter99 (N = 6,116) and Health2006 (N = 2,761) cohorts, two population-based samples of middle-aged people, followed up after 5 years.

RESULTS

In meta-analyses of all data, no association was found between rs11574-A and changes in BMI, WC, WHR, or fat mass. We found an association between rs11574-A and cross-sectional BMI (N = 10,359, β: -0.16 kg/m² per allele, 95% CI: -0.30 to -0.01, P = 0.033) and fat mass (N = 4,188, β: -0.52 kg/m² per allele, 95% CI: -1.03 to -0.01, P = 0.046).

CONCLUSIONS

No consistent impact of the genetic variant on changes in fat mass, BMI, or fat distribution was found in three Danish cohorts.

Effect of the interaction between diet composition and the PPM1K genetic variant on insulin resistance and β cell function markers during weight loss: results from the Nutrient Gene Interactions in Human Obesity: implications for dietary guidelines (NUGENOB) randomized trial

Background: Circulating branched-chain amino acids (BCAAs) and aromatic amino acids (AAAs) have been shown to be associated with insulin resistance and diabetes risk. The common rs1440581 T allele in the protein phosphatase Mg2+/Mn2+ dependent 1K (PPM1K) gene has been related to elevated BCAA concentrations and risk of type 2 diabetes.Objective: In the present study, we tested whether dietary fat and carbohydrate intakes influenced the association between the rs1440581 PPM1K genetic variant and glucose-metabolism traits during weight loss.Design: The rs1440581 PPM1K genetic variant was genotyped in a total of 757 nondiabetic individuals who were randomly assigned to 1 of 2 energy-restricted diets that differed in macronutrient composition (low-fat diet: 20-25% fat, 15% protein, and 60-65% carbohydrate; high-fat diet: 40-45% fat, 15% protein, and 40-45% carbohydrate). The changes in fasting glucose, fasting insulin, insulin resistance (homeostasis model assessment of insulin resistance) and homeostasis model assessment of β cell function (HOMA-B) were measured after a mean ± SD weight loss of 6.8 ± 3.4 kg over 10 wk and analyzed according to the presence of the T allele of rs1440581.Results: The rs1440581 T allele was associated with a smaller improvement in glucose concentrations after the 10-wk dietary intervention (β ± SE: 0.05 ± 0.02 mg/dL; P = 0.03). In addition, significant gene-diet interactions were shown for the rs1440581 PPM1K genetic variant in relation to changes in insulin and HOMA-B (P-interaction = 0.006 and 0.002, respectively). In response to the high-fat diet, the T allele was associated with a higher reduction of insulin (β ± SE: -0.77 ± 0.40 μU/mL; P = 0.04) and HOMA-B (β ± SE: -13.2 ± 3.81; P = 0.003). An opposite effect was observed in the low-fat diet group, although in this group the T allele was marginally (P = 0.10) and not significantly (P = 0.24) associated with insulin and HOMA-B, respectively.Conclusion:PPM1K rs1440581 may affect changes in glucose metabolism during weight loss, and this effect is dependent on dietary fat and carbohydrate intakes. This trial was registered at controlled-trials.com as ISRCTN25867281.

A genome-wide association study of thyroid stimulating hormone and free thyroxine in Danish children and adolescents

Background

Hypothyroidism is associated with obesity, and thyroid hormones are involved in the regulation of body composition, including fat mass. Genome-wide association studies (GWAS) in adults have identified 19 and 6 loci associated with plasma concentrations of thyroid stimulating hormone (TSH) and free thyroxine (fT4), respectively.

Objective

This study aimed to identify and characterize genetic variants associated with circulating TSH and fT4 in Danish children and adolescents and to examine whether these variants associate with obesity.

Methods

Genome-wide association analyses of imputed genotype data with fasting plasma concentrations of TSH and fT4 from a population-based sample of Danish children, adolescents, and young adults, and a group of children, adolescents, and young adults with overweight and obesity were performed (N = 1,764, mean age = 12.0 years [range 2.5−24.7]). Replication was performed in additional comparable samples (N = 2,097, mean age = 11.8 years [1.2−22.8]). Meta-analyses, using linear additive fixed-effect models, were performed on the results of the discovery and replication analyses.

Results

No novel loci associated with TSH or fT4 were identified. Four loci previously associated with TSH in adults were confirmed in this study population (PDE10A (rs2983511: β = 0.112SD, p = 4.8 ∙ 10⁻¹⁶), FOXE1 (rs7847663: β = 0.223SD, p = 1.5 ∙ 10⁻²⁰), NR3C2 (rs9968300: β = 0.194SD), p = 2.4 ∙ 10⁻¹¹), VEGFA (rs2396083: β = 0.088SD, p = 2.2 ∙ 10⁻¹⁰)). Effect sizes of variants known to associate with TSH or fT4 in adults showed a similar direction of effect in our cohort of children and adolescents, 11 of which were associated with TSH or fT4 in our study (p<0.0002). None of the TSH or fT4 associated SNPs were associated with obesity in our cohort, indicating no pleiotropic effects of these variants on obesity.

Conclusion

In a group of Danish children and adolescents, four loci previously associated with plasma TSH concentrations in adults, were associated with plasma TSH concentrations in children, suggesting comparable genetic determinants of thyroid function in adults and children.

Patients With Long-QT Syndrome Caused by Impaired hERG-Encoded Kv11.1 Potassium Channel Have Exaggerated Endocrine Pancreatic and Incretin Function Associated With Reactive Hypoglycemia

Supplemental Digital Content is available in the text.

Background:

Loss-of-function mutations in hERG (encoding the K_v11.1 voltage-gated potassium channel) cause long-QT syndrome type 2 (LQT2) because of prolonged cardiac repolarization. However, K_v11.1 is also present in pancreatic α and β cells and intestinal L and K cells, secreting glucagon, insulin, and the incretins glucagon-like peptide-1 (GLP-1) and GIP (glucose-dependent insulinotropic polypeptide), respectively. These hormones are crucial for glucose regulation, and long-QT syndrome may cause disturbed glucose regulation. We measured secretion of these hormones and cardiac repolarization in response to glucose ingestion in LQT2 patients with functional mutations in hERG and matched healthy participants, testing the hypothesis that LQT2 patients have increased incretin and β-cell function and decreased α-cell function, and thus lower glucose levels.

Methods:

Eleven patients with LQT2 and 22 sex-, age-, and body mass index–matched control participants underwent a 6-hour 75-g oral glucose tolerance test with ECG recording and blood sampling for measurements of glucose, insulin, C-peptide, glucagon, GLP-1, and GIP.

Results:

In comparison with matched control participants, LQT2 patients had 56% to 78% increased serum insulin, serum C-peptide, plasma GLP-1, and plasma GIP responses (P=0.03–0.001) and decreased plasma glucose levels after glucose ingestion (P=0.02) with more symptoms of hypoglycemia (P=0.04). Sixty-three percent of LQT2 patients developed hypoglycemic plasma glucose levels (<70 mg/dL) versus 36% control participants (P=0.16), and 18% patients developed serious hypoglycemia (<50 mg/dL) versus none of the controls. LQT2 patients had defective glucagon responses to low glucose, P=0.008. β-Cell function (Insulin Secretion Sensitivity Index-2) was 2-fold higher in LQT2 patients than in controls (4398 [95% confidence interval, 2259–8562] versus 2156 [1961–3201], P=0.03). Pharmacological K_v11.1 blockade (dofetilide) in rats had similar effect, and small interfering RNA inhibition of hERG in β and L cells increased insulin and GLP-1 secretion up to 50%. Glucose ingestion caused cardiac repolarization disturbances with increased QTc intervals in both patients and controls, but with a 122% greater increase in QTcF interval in LQT2 patients (P=0.004).

Conclusions:

Besides a prolonged cardiac repolarization phase, LQT2 patients display increased GLP-1, GIP, and insulin secretion and defective glucagon secretion, causing decreased plasma glucose and thus increased risk of hypoglycemia. Furthermore, glucose ingestion increased QT interval and aggravated the cardiac repolarization disturbances in LQT2 patients.

Clinical Trial Registration:

URL: http://clinicaltrials.gov. Unique identifier: NCT02775513.

FTO genotype and weight loss: systematic review and meta-analysis of 9563 individual participant data from eight randomised controlled trials

OBJECTIVE

 To assess the effect of the FTO genotype on weight loss after dietary, physical activity, or drug based interventions in randomised controlled trials.

DESIGN

 Systematic review and random effects meta-analysis of individual participant data from randomised controlled trials.

DATA SOURCES

 Ovid Medline, Scopus, Embase, and Cochrane from inception to November 2015.

ELIGIBILITY CRITERIA FOR STUDY SELECTION

 Randomised controlled trials in overweight or obese adults reporting reduction in body mass index, body weight, or waist circumference by FTO genotype (rs9939609 or a proxy) after dietary, physical activity, or drug based interventions. Gene by treatment interaction models were fitted to individual participant data from all studies included in this review, using allele dose coding for genetic effects and a common set of covariates. Study level interactions were combined using random effect models. Metaregression and subgroup analysis were used to assess sources of study heterogeneity.

RESULTS

 We identified eight eligible randomised controlled trials for the systematic review and meta-analysis (n=9563). Overall, differential changes in body mass index, body weight, and waist circumference in response to weight loss intervention were not significantly different between FTO genotypes. Sensitivity analyses indicated that differential changes in body mass index, body weight, and waist circumference by FTO genotype did not differ by intervention type, intervention length, ethnicity, sample size, sex, and baseline body mass index and age category.

CONCLUSIONS

 We have observed that carriage of the FTO minor allele was not associated with differential change in adiposity after weight loss interventions. These findings show that individuals carrying the minor allele respond equally well to dietary, physical activity, or drug based weight loss interventions and thus genetic predisposition to obesity associated with the FTO minor allele can be at least partly counteracted through such interventions.

SYSTEMATIC REVIEW REGISTRATION

 PROSPERO CRD42015015969.

Genetic risk scores link body fat distribution with specific cardiometabolic profiles

OBJECTIVE

Forty-nine known single nucleotide polymorphisms (SNPs) associating with body mass index (BMI)-adjusted waist-hip-ratio (WHR) (WHRadjBMI) were recently suggested to cluster into three groups with different associations to cardiometabolic traits. Genetic risk scores of the clusters on the risk of incident diabetes and associations with detailed cardiometabolic phenotypes were tested.

METHODS

In a prospective study of 6,121 Inter99 individuals, the risk of incident diabetes using Cox proportional hazards regression was evaluated. Using linear regession, the associations between genetic risk scores and anthropometry and blood samples at fasting and during an oral glucose tolerance test were tested. Analyses were adjusted for age, sex, and BMI.

RESULTS

Cluster 1 associated with an increased risk of diabetes (HR = 1.05, P = 2.74 × 10(-) (4) ) and with a poor metabolic profile, including fasting serum triglyceride (β = 0.98% mmol/L, P = 3.33 × 10(-) (8) ) and Matsuda index (β = -0.74%, P = 1.29 × 10(-) (4) ). No similar associations for Clusters 2 and 3 were found. The three clusters showed different patterns of association with waist circumference, hip circumference, and height.

CONCLUSIONS

Our results suggest that the 49 WHRadjBMI-associated SNPs affect metabolic health differently depending on the cluster of SNPs. The clusters further associate differently with anthropometric measures.

The potential role of inhibitor of differentiation-3 in human adipose tissue remodeling and metabolic health

Metabolic health in obesity is known to differ among individuals, and the distribution of visceral (VAT) and subcutaneous adipose tissue (SAT) plays an important role in this regard. Adipose tissue expansion is dependent on new blood vessel formation in order to prevent hypoxia and inflammation in the tissue. Regulation of angiogenesis in SAT and VAT in response to diet is therefore crucial for the metabolic outcome in obesity. Knowledge about the underlying genetic mechanisms determining metabolic health in obesity is very limited. We aimed to review the literature of the inhibitor of differentiation-3 (ID3) gene in relation to adipose tissue and angiogenesis in humans in order to determine whether ID3 could be involved in the regulation of adipose tissue expansion and metabolic health in human obesity. We find evidence that ID3 is involved in regulatory mechanisms in adipose tissue and regulates angiogenesis in many tissues including adipose tissue. We discuss how this might influence obesity and metabolic health in obesity and further discuss some potential mechanisms by which ID3 might regulate visceral and subcutaneous adipose tissue expansion. The combined results from the reviewed literature suggest ID3 to play a potential role in the underlying regulatory mechanisms of metabolic health in human obesity. The literature is still sparse and further studies focusing on human ID3 in relation to the nature of obesity are warranted.

Increased orosomucoid in urine is an independent predictor of cardiovascular and all-cause mortality in patients with type 2 diabetes at 10 years of follow-up

AIMS

To evaluate whether increased urinary orosomucoid excretion rate (UOER) is an independent predictor of cardiovascular and all-cause mortality in type 2 diabetes (T2DM) and type 1 diabetes (T1DM) at 10years of follow-up.

METHOD

We followed 430 patients with T2DM and 148 patients with T1DM until emigration, death or November 2011. We measured UOER levels in overnight urine samples.

RESULTS

Descriptive data are given in the article. In patients with T2DM and T1DM, all-cause mortality (log-rank test, p<0.01 for both types) and cardiovascular mortality (log-rank test, p<0.01 for T2DM and p=0.04 for T1DM) were significantly higher in patients with increased UOER. Normoalbuminuric patients with T2DM and increased UOER levels had higher all-cause and cardiovascular mortality (log-rank test, p<0.01 for both types). UOER was independently predictive of all-cause (HR 1.52; 95% CI 1.10-2.09; p=0.01) and cardiovascular (HR 2.31; 95% CI 1.46-3.66; p<0.01) mortality in patients with T2DM, but not in patients with T1DM.

CONCLUSION

UOER is an independent predictor of all-cause and cardiovascular mortality even in normoalbuminuric patients with T2DM at 10years of follow-up. Further studies are needed in order to evaluate the prognostic and clinical relevance.

Stagnation in body mass index in Denmark from 1997/1998 to 2004/2005, but with geographical diversity

INTRODUCTION

We analyzed the trend in body mass index (BMI) as well as in the prevalence of overweight and obesity among Danish adults, mainly women, from 1997/1998 to 2004/2005 and evaluated any regional differences.

MATERIAL AND METHODS

Data were drawn from two cross-sectional population-based studies conducted in parallel in Aalborg and Copenhagen from 1997/1998 and 2004/2005. Height and weight were measured in a total of 7,487 participants in the two cohorts.

RESULTS

In the total cohort, we found no significant difference in BMI from 1997/1998 to 2004/2005 (p = 0.828). There was an increase in BMI in Aalborg of 0.32 (p = 0.030), while in Copenhagen we observed a statistically significant decrease in BMI of 0.30 (p = 0.017). The difference in change over time in BMI between the two regions was significant (p = 0.002). Also the difference in the trend in prevalence of overweight and obesity was statistically significant between the two cities (p = 0.010).

CONCLUSION

Our results indicate that the obesity epidemic is leveling off - at least among women - and that it may even be receding in Copenhagen. Nevertheless, the absolute average BMI values and the prevalence of overweight and obesity in both cities are high which underlines the need for further initiatives to prevent obesity-related health risks in the future.

FUNDING

The DanThyr studies were supported by funding from the Tømmerhandler Vilhelm Bang Foundation, the Copenhagen Corporation Research Foundation, the 1991 Pharmacy Foundation, the Danish Medical Foundation, the Health Insurance Foundation, the Agnes and Knut Mørk Foundation, the Wedel Wedelsborg Foundation, the Ortho-Clinical Foundation and BRAHMS Diagnostica.

TRIAL REGISTRATION

not relevant.