Tehran Lipid And Glucose Study Research Articles

Patterns of change in obesity indices and other cardiometabolic risk factors before the diagnosis of type 2 diabetes: two decades follow-up of the Tehran lipid and glucose study

BackgroundIdentifying patterns of variation in obesity indices and other cardiometabolic risk factors before the diagnosis of type 2 diabetes could provide insight into the critical period when drastic changes occurred and facilitate targeted interventions for the prevention of diabetes. Therefore, this study sought to explore patterns of change in obesity indices and other cardiometabolic risk factors before diabetes diagnosis.MethodsWe investigated 6305 participants (43.7% men) aged 20–65 from the Tehran Lipid and Glucose Study (TLGS) who were free of diabetes at baseline. First, we jointly estimated developmental multi-trajectories of obesity indices using multivariate latent class growth mixed model, and then patterns of cardiometabolic risk factors within the identified multi-trajectories were assessed using mixed-effects models.ResultsThree patterns of change in obesity indices were identified. Most participants belonged to the “progressing” group (83.4%; n = 742), with a slight but steadily rising in obesity indices until diagnosis in both men and women. All multi-trajectory groups showed similar exponential increases in fasting and 2-h plasma glucose concentrations 6 years before diagnosis and linear increases in blood pressure and total and LDL cholesterol throughout follow-up. Patterns of triglyceride and HDL cholesterol accompanied each group’s patterns of change in obesity indices.ConclusionThree patterns of the joint progression of obesity indices before diabetes diagnosis were accompanied by similar blood glucose patterns and other cardiometabolic risk factors. These findings suggest the impact of the increasing trend of obesity indices and other metabolic factors on the incidence of diabetes and emphasize the importance of assessing the metabolic risk factors at each visit.

BMI variability and incident diabetes mellitus, Tehran Lipid and Glucose Study (TLGS)

Previous epidemiologic studies debated the association of body mass index (BMI) trends with cardiovascular disease and mortality. This study aimed to evaluate the association of BMI variability and slope with the incidence of Type 2 diabetes mellitus (T2DM) in a sex-stratified 15.8-year follow-up in the population-based Tehran Lipid and Glucose Study (TLGS). Of 10,911 individuals aged 20–60 years, 4981 subjects were included and followed for 15.8-years. The slope coefficient of BMI in the linear regression model represented individuals’ BMI trends up to the incidence of DM. The root mean squared error (RMSE) of the BMI linear trend was selected to reflect BMI variability through six follow-ups. Cox proportional hazards regression was used to investigate the association of the baseline BMI, BMI slope and RMSE with the incidence of T2DM among men and women. Multivariable-adjusted HRs of T2DM for each SD increment in BMI slope was 1.18 (95% CI: 0.94–1.48, p = 0.161) in normal weight men and 1.26 (95% CI: 1.10–1.44, p = 0.001) in overweight and obese men. However, in women, each SD increment in BMI slope increased the risk of T2DM with a HR of 1.19 (95% CI: 1.01–1.40, p = 0.039) in normal weight, and 1.14 (95% CI: 1.08–1.19, p < 0.001) in women with BMI ≥ 25 kg/m2. In men with a baseline BMI ≥ 25 kg/m2, BMI-RMSE was associated with a decreased risk of T2DM (HR: 0.71, 95% CI: 0.53–0.93, p = 0.015). Baseline BMI was not associated with the risk of diabetes in men and women. Positive BMI slope is associated with the development of diabetes in both sexes. The association of BMI variability with incident T2DM differs according to sex and baseline BMI. BMI variability is associated with a lower risk of T2DM in overweight and obese men. BMI variability in women and baseline BMI in both gender are not related to the risk of T2DM.

High-Fat Dairy Products May Decrease the Risk of Chronic Kidney Disease Incidence: A Long-Term Prospective Cohort Study

The association between consumption of dairy products and risk of chronic kidney disease (CKD) is under debate. We aimed to determine the potential effects of total and subtypes of dairy intake on the occurrence of CKD. This study was conducted within the Tehran Lipid and Glucose Study (TLGS) on 2416 CKD-free adults. At baseline, consumption of dairy products was estimated using a validated 168-items semiquantitative food frequency questionnaire. Adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) of CKD were calculated in tertile categories of dairy products. Also, the CKD risk was estimated with multivariable Cox regression to substitute total dairy with other dietary protein sources. During 8.4years of follow-up, the incidence rate of CKD was 21%. The participants' mean (±SD) age was 38 (±13) years and 46% were men. Dietary intakes of total dairy, low-fat dairy, and fermented dairy were not associated with CKD risk. There were significant lower risks of CKD in the highest compared to the lowest tertiles of high-fat dairy (HR=0.76, 95% CI=0.60-0.95) and high-fat milk (HR=0.75, 95% CI=0.59-0.96). However, no significant associations were found between other categories of dairy products and CKD incidence. Substitutions of total dairy with other dietary protein sources were not associated with CKD risk. In this study, higher intakes of high-fat dairy and high-fat milk were associated with lower risks of CKD. No significant associations were found between other dairy products and CKD. More prospective and clinical trials are needed to clarify the issue.

Weight fluctuation, mortality, and cardiovascular disease in adults in 18years of follow-up: Tehran Lipid and Glucose Study.

Controversies exist in the effect of body weight loss and fluctuation on cardiovascular disease (CVD) and mortality. This study aims to assess the effect of weight variability on CVD and all-cause and cardiovascular mortality in the Tehran Lipid and Glucose Study (TLGS) cohort. Participants aged ≥ 40year at the baseline period with at least 3 BMI measurements were included in this study. After excluding individuals with cancer, CVD, end-stage renal disease, systemic use of glucocorticoids, pregnancy, and missing covariates at the baseline, a total of 3461 participants were enrolled and followed for 18years. BMI variability was defined using root mean squared error (RMSE) and average successive variability (ASV). In the RMSE method, BMI variability was calculated using the best-fitting model for BMI trend of each subject. Multivariate Cox proportional hazard models were applied to assess BMI variability's effect on CVD and mortality. Among the 3461 participants in this study, the group with the highest weight variability had an increased risk of death for all-cause (HR 1.65; 95% CI 1.21-2.25), non-cardiovascular (HR 1.77; 95% CI 1.24-2.53), and non-cancer (HR 1.77; 95% CI 1.25-2.50) mortality. However, BMI variability showed to be protective against CVD (HR 0.76; 95% CI 0.6-0.97). These findings were significant in males, non-smokers, participants with age ≤ 60year, BMI < 30, negative BMI slope, and both diabetic and non-diabetic subjects. High BMI variability is associated with increased risk of all-cause, non-CVD, and non-cancer mortality, although protective for the CVD event. Appropriate strategies for body weight maintenance after weight loss could be adopted to avoid weight variability, particularly in non-obese subjects.

Factors Associated with Depression, Anxiety, and Stress in Men and Women: Findings from a Population-Based Study in Iran

Background: This study aimed to investigate the relationship between personal and clinical characteristics of adults in the Tehran Lipid and Glucose Study (TLGS) with depression, anxiety, and stress. Methods: Data of 2272 adults participating in the 6th phase of TLGS were used for univariate analysis to investigate the association between socio-demographic, behavioral, and clinical characteristics and participants’ emotional states. Thereupon, the predictors with a P value&lt;0.20, at least for one of depression, anxiety, and stress in the primary analysis, were included in the model for multivariate modeling. Results: The mean age of participants was 47.23±14.87. The mean scores of depression, anxiety, and stress were higher in women (P value:&lt;0.001,&lt;0.001, and 0.004) than in men. Higher age was associated with lower anxiety (β=-0.04, P=0.004) and stress (β=-0.13, P&lt;0.001) in men, but only lower stress in women (β=-0.07, P=0.001). Highly educated participants experienced lower depression (β=-2.26, P=0.01, β=-2.26, P=0.003). Although married men robustly experienced lower depression than others (β=-1.69, P=0.009), a less powerful relationship existed between being married and depression in women (β=-1.37, P=0.05). All cigarette smokers experienced higher depression, anxiety, and stress, but only female hookah smokers had higher anxiety and stress. Physical activity and chronic disorders had no relationship with emotional states. Obesity was associated with stress in women (β=1.95, P=0.001). Conclusion: This study provides an update on factors associated with mental health outcomes in a large general population. Examining the factors associated with depression, anxiety, and stress through a sex-sensitive lens could help clinicians diagnose and plan the best preventive and therapeutic approach.

Visceral adiposity-related dietary patterns and the risk of cardiovascular disease in Iranian adults: A population-based cohort study.

BackgroundVisceral obesity is a significant predictor of cardiovascular disease (CVD). Diet may associate with CVD risk through its effects on visceral adiposity. This study aimed to find dietary patterns (DPs) related to indicators of visceral adiposity and to determine whether the DPs were associated with CVD risk.MethodsThis prospective study included 2,496 participants of the Tehran Lipid and Glucose Study (TLGS) without CVD, who were followed from the third study examination (2005–2008; baseline) to March 2018. DPs at baseline were determined using reduced rank regression (RRR) and partial least squares regression (PLS). The response variables were age and BMI-adjusted waist circumference (WC) and age-adjusted visceral adiposity index (VAI).ResultsTwo and three DPs were retained with RRR and PLS, respectively. The first patterns of each method were mainly characterized by adjusted-WC (RRR: 10.8%, PLS: 8.6%); none of them were associated with CVD risk. The second pattern of RRR and the third pattern of PLS were mainly explained by adjusted-VAI (RRR: 3.3, PLS: 2.1%). After adjusting for CVD risk factors, the hazard ratios [95% confidence intervals (CI)] for CVD in the second and third tertiles of the RRR-pattern 2 were 1.76 (1.15, 2.69) and 1.55 (1.00, 2.43) vs. the first tertile (p-trend: 0.058). This pattern had high positive loadings for non-leafy vegetables, pickled vegetables, fried vegetables, and bread and high negative loadings for eggs, cakes, butter, jam-honey, red meat, poultry, fish, juice, non-fermented dairy, and fruits. Per one SD increase in PLS-pattern 3 score, the risk of CVD was 19% higher (95%CI = 3–38%). This positive association was also observed across tertiles of the pattern (p-trend: 0.032). This pattern was characterized by high intakes of leafy vegetables, non-leafy vegetables, organ meat, soft drinks, olive oil, pickled vegetables, fried vegetables, and bread and low intakes of biscuits, cakes, butter, eggs, and non-fermented dairy.ConclusionFor each of the RRR and PLS approaches, a visceral-related DP that was positively linked to CVD was identified. These two patterns had a modest correlation. The pattern generated by PLS explained more variations in food groups and offered stronger evidence of association with CVD than the RRR-derived pattern.

Changes in ideal cardiovascular health among Iranian adolescents: 2007–2008 to 2015–2017

BackgroundAssessment of both behavior and factors of health as ideal cardiovascular health (iCVH) in adolescence could contribute to cardiovascular disease prevention in adulthood.AimsTo explore the changes in the prevalence of iCVH and its components during a decade among Tehranian adolescents.MethodsThe 12–19 years old adolescents were selected from the Tehran Lipid and Glucose Study(TLGS). The iCVH score was calculated in the study period 1 (2007–2008; n = 267) and 2 (2015–2017; n = 336). To calculate iCVH, body mass index(BMI), systolic and diastolic blood pressure (BP), total cholesterol, fasting plasma glucose (FPG), physical activity, smoking status, and dietary intakes were measured by standard protocols. The changes in iCVH components between the two study periods were reported by prevalence (95% confidence interval). A logistic regression model was conducted to test the effects of study periods, sex, and age groups on the iCVH(≥ 6 scores).ResultsOverall, there was a reduction in the prevalence of ideal FPG (97.4 vs. 91.1%) and ideal BP(91.8 vs. 82.7%). Girls had a decrease in the prevalence of ideal BP(91.2 vs. 79.4%) as well as an increase in non-smoking status(77.6 vs. 89.7%). However, the prevalence of ideal FPG (96.5 vs. 88.5%) and ideal BP(92.2 vs 85.0%) decreased in boys. Study period 2, compared to period 1 was associated with lower odds of having iCVH. Furthermore, boys were 1.57 folds more likely to have ideal CVH factors than girls.ConclusionThere was a decrease in the prevalence of ideal CVD metrics, including FPG and BP, after one decade. Generally, adolescent boys had higher odds of having ideal CVH compared to girls.

A pragmatic multi-setting lifestyle intervention to improve leisure-time physical activity from adolescence to young adulthood: the vital role of sex and intervention onset time.

BackgroundThe long-term effectiveness of healthy lifestyle interventions on improving leisure-time physical activity (LTPA) in adolescents and its factors in low- and middle-income communities is unclear. This study is the first to investigate LTPA trends in a population of Iranian adolescents who underwent a multi-setting lifestyle intervention, considering sex and the time of intervention onset.MethodsParticipants were 2374 adolescents (57.2% girls), aged 12–18 years, who participated in the Tehran Lipid and Glucose Study (TLGS) during 1999–2001 and followed for a median follow-up of 15.9 over five data points every 3 years. Adolescent participants were assigned to the intervention or control groups based on their residential areas. Boys and girls were categorized into 12–15 or 16–18 years old to minimize participants’ bio-psychological differences, facilitate environmental interventions by more compliance with the Iranian educational system and identify the best time to start the intervention. All adolescents in the intervention area received healthy lifestyle interventions via the school-, family-, and community-based programs. LTPA was assessed using the reliable and validated Iranian Modifiable Activity Questionnaire (MAQ) version over the five data points. The Generalized Estimating Equations method was used to evaluate educational intervention’s effect on LTPA in adolescents during the follow-up.ResultsIn boys who experienced the early onset of intervention (12–15 years), the interaction effect of follow-up examinations and the intervention was significant where the impact of the intervention differed over time. In this group, LTPA was higher in the control group than in the intervention group at the first follow-up examination (β = − 1088.54). However, an increasing trend of LTPA was observed in the intervention group until the third follow-up examination (β = 1278.21, p = 0.08, and β = 1962.81, p = 0.02, respectively), with borderline significance levels at the 2nd (P = 0.08) and the 4th (P = 0.08) measurements. The interaction terms and main effects of intervention and follow-up examinations were not significant in boys with late intervention onset. Although older boys in the intervention group had higher LTPA than the control group, there were no significant differences among study groups in all follow-up examinations. Regarding girls, LTPA did not differ significantly between intervention and control groups in all follow-up examinations (P > 0.05).ConclusionOur results showed that a multi-setting practical lifestyle intervention could improve long-term energy expenditure in LTPA in adolescent boys who have experienced an early onset intervention. Findings emphasized the vital role of gender and the onset of these interventions. The current results would be valuable to plan tailored interventions to improve LTPA and community health.Trial registrationThis study is registered at Iran Registry for Clinical Trials (IRCT), a WHO primary registry (http://irct.ir). The TLGS clinical trial was the very first registration in the IRCT (Iran Registry of Clinical Trials). it was registered on 2008-10-29 by the registration number IRCT138705301058N1. Based on the international committee of medical journal Editors (ICMJE), “retrospective registration” is acceptable for trials that began before July 1, 2005.

The Long-Term Effectiveness of a Multisetting Lifestyle Intervention on Tobacco-Related Habits in Adolescent Boys and Girls: Tehran Lipid and Glucose Study.

This study assesses the long-term effectiveness of a lifestyle intervention on cigarette, passive, and hookah smoking in adolescent boys and girls. One thousand one hundred and fifty-nine adolescent participants of Tehran Lipid and Glucose Study (TLGS) (2001-2004) were followed for 12 years. Participants in the intervention area received lifestyle interventions in family, schools, and community settings, with the largest part implemented in schools. After excluding those with missing baseline parental data (n=66), complete parental data of 1093 adolescents was used for cluster analysis. Families were classified as low- and high-risk. Two hundred and ninety-six individuals who did not complete intervention/follow-ups were excluded, GEE was performed on 797 adolescents (369 boys, 192 intervention) to assess the intervention effect on tobacco-related habits. Adolescents' mean baseline age was 15.21 ± 1.95. Those living in high-risk families were more at risk of cigarette, passive, and hookah smoking. Intervention decreased odds of cigarette and passive smoking by 38% and 57%, respectively. Intervention did not reduce hookah smoking in adolescents. After sex-specific analysis, the intervention reduced current cigarette smoking by 40% only in boys and passive smoking in both sexes. Targeting lifestyle behaviors in school-aged adolescents and their families reduces cigarette smoking in boys and protects both sexes from secondhand smoke.

Serum adiponectin and cortisol levels are not affected by studied ADIPOQ gene variants: Tehran lipid and glucose study

BackgroundObesity is a major public health concern in developed and even developing countries worldwide. Adiponectin is a protein secreted by adipose tissue that modulates many metabolic processes and plays a vital role in obesity. This study aimed to determine the association of four variants of the ADIPOQ gene with serum adiponectin, cortisol levels and obesity status.MethodsThis case-control study was performed on 164 obese individuals compared by 156 control from the Tehran Lipid and Glucose Study (TLGS). Standard procedures obtained anthropometric measures and metabolic parameters. Cortisol and adiponectin levels were measured by ELISA method. rs1501299, rs266729, rs17300539, and rs17366743 on the ADIPOQ gene were genotyped using the PCR-RFLP. The correlation between adiponectin gene SNPs and obesity were calculated by Additive, dominant, and recessive genetic models. Pearson’s or Spearman’s found correlations between adiponectin levels and metabolic and anthropometric variables. Data were analyzed using SPSS software Version 20.ResultsAdiponectin and cortisol levels were significantly lower in obese subjects compared to the control group (p < 0.05). There was a significant negative correlation between serum adiponectin level and BMI, waist circumference (WC), waist-hip ratio, hip circumference (HC), Fasting blood sugar (FBS) Triglyceride (TG), Total cholesterol (TC), Systolic blood pressure (SBP), Diastolic blood pressure (DBP) (r = − 0.147, r = − 0.324, r = 0.371, r = − 0.179, r = − 0.299, r = − 0.277, r = − 0.041, r = − 0.134, and r = − 0.149, respectively). A positive correlation was found between adiponectin and high-density lipoprotein cholesterol (HDL-C) (r = 0.29), but no significant correlations were found between adiponectin and Low-density lipoprotein cholesterol(LDL-C) and cortisol. ADIPOQ variant rs1501299 was significantly associated with cortisol levels in subjects with BMI ≥ 25 (P-value =0.039).ConclusionsAdiponectin and cortisol levels were associated with obesity. No ADIPOQ gene variants and haplotypes were associated with cortisol, Adiponectin, and obesity.

Body Composition Assessment by Bioelectrical Impedance Analysis in Prediction of Cardio-Metabolic Risk Factors: Tehran Lipid and Glucose Study (TLGS)

Background:We aimed at evaluating the best body mass index (BMI) and percent body fat (PBF) cutoffs related to cardio-metabolic risk factors and comparing the discriminative power of PBF and BMI for predicting these risk factors.Methods:In this cross-sectional study in phase V (2012–2015), 1271 participants (age ≥ 20 yr; 54.3% women) were enrolled. Bioelectrical impedance analysis (BIA) was used to estimate PBF. Joint Interim Statement criteria were used for defining metabolic syndrome (MetS). We compared PBF with BMI through logistic regression and area under the curve of the receiver operating characteristic (ROC) curve. Percent body fat cutoff points were > 25 in men and >35 in women.Results:Percent body fat and BMI cutoff points for predicting MetS were 25.6% and 27.2 kg/m2 in men and 36.2% and 27.5 kg/m2 in women, respectively. There were no significant differences between BMI and PBF area under the ROC curves for predicting MetS and its components, except for abdominal obesity in men and low high-density lipoprotein (HDL) in women in favor of BMI. Logistic regression analysis indicated that BMI in women was better for predicting MetS and its components, except for abdominal obesity. Moreover, BMI was equal or superior to PBF in men, except for low HDL and high triglyceride levels.Conclusion:Comparison of PBF with BMI showed that the use of PBF is not significantly better than BMI in predicting cardio-metabolic risks in the general population.

The trend of 10-year cardiovascular risk among diabetic and non-diabetic participants in Tehran Lipid and glucose study: 1999\u20132018

BackgroundAssessing the risk of cardiovascular disease (CVD) is crucial in preventive cardiology. We aimed to determine the trend of CVD risk among individuals with and without diabetes during two decades of follow-up in a Middle Eastern cohort.MethodsWe studied 8,450 individuals (55.5% women) aged 40–75 years who participated in the Tehran Lipid and Glucose Study (TLGS). Diabetes status and CVD risk factors were evaluated in six examinations from 1999 to 2018. The individual 10-year CVD risk score was calculated using the ACC/AHA recommended risk equation. We used generalized estimating equation models (GEE) to assess the time trends of CVD risk factors and CVD risk scores in diabetic and non-diabetic groups separately.ResultsThe age-adjusted ACC/AHA risk score significantly decreased in non-diabetic women and men (from 3.2% to 1.6% in women and 6.8% to 5.0% in men; p for trend < 0.001). Whereas the risk significantly decreased among diabetics men (from 13.8% to 11.5%), it increased somehow among diabetics women (from 5.3% to 5.5%). Furthermore, in both sexes, diabetic individuals compared to non-diabetic ones had better control on their systolic blood pressure, total cholesterol, and fasting plasma glucose during the last two decades.ConclusionsThe CVD risk and most CVD risk factors improved in individuals with and without diabetes in the past two decades; however, they have not reached the targets yet. So, more stringent lifestyle modifications and treatment strategies are needed, especially for primary prevention in the general population.

Spot urinary microalbumin concentration, metabolic syndrome and type 2 diabetes: Tehran lipid and glucose study

AimThis study aimed to determine the association of urinary microalbumin concentrations with type 2 diabetes mellitus (T2DM), metabolic syndrome (MetS), and its phenotypes. The optimum cut-off values of urinary microalbumin and microalbumin-to-creatinine ratio (MCR) for predicting the chance of having T2DM and MetS were also defined.MethodsAdult men and women (n = 1192) participated in the sixth phase (2014-2017) of the Tehran Lipid and Glucose Study (TLGS), with completed data, were included in the analyses. Odds ratios (ORs) (and 95% confidence intervals (CIs)) of T2DM, MetS, and its components across tertile categories of urinary microalbumin concentrations were estimated using multivariable logistic regressions. The optimal cut-off points of urinary microalbumin and MCR were determined using the receiver operator characteristic (ROC) curve analysis.ResultsParticipants’ mean (±SD) age was 44.9 (±14.0) years, and 44.6% of the participants were men. The prevalence of microalbuminuria was 14.4%. Chance of having T2DM was significantly higher in the highest tertile of urinary microalbumin concentration (OR = 2.29, 95% CI = 1.43-3.67) and MCR (OR = 1.82, 95% CI = 1.15-2.89). Subjects with the highest urinary microalbumin concentration were more likely to have MetS (OR = 1.66, 95% CI = 1.17-2.35), hypertension (OR = 1.63, 95% CI = 1.16-2.30) and hyperglycemia (OR = 1.78, 95% CI = 1.24-2.56). No significant association was observed between urinary microalbumin concentrations and other components of MetS. The optimal cut-off points of urinary microalbumin for predicting the chance of having T2DM and MetS were 14.0 and 13.6 mg/L, respectively.ConclusionsElevated spot urinary microalbumin, below the values defined as microalbuminuria, was associated with the chance of having T2DM and MetS.

The association of the age, period, and birth cohort with 15-year changes in body mass index and waist circumference in adults: Tehran lipid and glucose study (TLGS)

ObjectiveTo examine the association of age, period, and birth cohort with body mass index (BMI) and waist circumference (WC) changes among the participants of the Tehran Lipid and Glucose Study from 1999 to 2015.MethodsThis prospective cohort study included 4895 participants aged ≥20 years (41.3% men), who were divided into twelve gender stratified groups, having a ten-year age difference between them. Analyses were conducted to explicitly assess the association of age vs. period on BMI and WC changes. In addition, we evaluated BMI and WC changes among different birth cohorts.ResultsUpon 15 years of follow-up, the mean BMI of men and women increased from 26.0 ± 3.9 to 27.5 ± 4.3 kg/m2 and from 27.5 ± 4.8 to 29.9 ± 5.4 kg/m2 (P trend < 0.001), and this trend was accompanied by an increase in WC from 88.8 ± 10.9 to 97.8 ± 10.4 cm and from 87.3 ± 12.4 to 95.8 ± 12.1 cm, respectively (P trend < 0.001). Men and women in all age cohorts tended to have a rise in their BMI and WC with aging throughout the follow-up period. For men, this trend was more prominent in younger birth cohorts at phase III for BMI and at phases III and V for WC (indicating a significant negative association with birth cohort). For women, this trend was more prominent in older birth cohorts at both phases III and V for BMI and WC (indicating a significant positive association with the birth cohort).ConclusionThe rise in BMI and WC was strongly associated with age in both sexes. The men born in the recent birth cohorts and the women born in earlier birth cohorts had the most alarming BMI and WC trends. More efforts must be spent on obesity prevention policies, especially for younger men.

Diabetes mellitus risk prediction in the presence of class imbalance using flexible machine learning methods

BackgroundEarly detection and prediction of type two diabetes mellitus incidence by baseline measurements could reduce associated complications in the future. The low incidence rate of diabetes in comparison with non-diabetes makes accurate prediction of minority diabetes class more challenging.MethodsDeep neural network (DNN), extremely gradient boosting (XGBoost), and random forest (RF) performance is compared in predicting minority diabetes class in Tehran Lipid and Glucose Study (TLGS) cohort data. The impact of changing threshold, cost-sensitive learning, over and under-sampling strategies as solutions to class imbalance have been compared in improving algorithms performance.ResultsDNN with the highest accuracy in predicting diabetes, 54.8%, outperformed XGBoost and RF in terms of AUROC, g-mean, and f1-measure in original imbalanced data. Changing threshold based on the maximum of f1-measure improved performance in g-mean, and f1-measure in three algorithms. Repeated edited nearest neighbors (RENN) under-sampling in DNN and cost-sensitive learning in tree-based algorithms were the best solutions to tackle the imbalance issue. RENN increased ROC and Precision-Recall AUCs, g-mean and f1-measure from 0.857, 0.603, 0.713, 0.575 to 0.862, 0.608, 0.773, 0.583, respectively in DNN. Weighing improved g-mean and f1-measure from 0.667, 0.554 to 0.776, 0.588 in XGBoost, and from 0.659, 0.543 to 0.775, 0.566 in RF, respectively. Also, ROC and Precision-Recall AUCs in RF increased from 0.840, 0.578 to 0.846, 0.591, respectively.ConclusionG-mean experienced the most increase by all imbalance solutions. Weighing and changing threshold as efficient strategies, in comparison with resampling methods are faster solutions to handle class imbalance. Among sampling strategies, under-sampling methods had better performance than others.

Development and validation of dietary and lifestyle insulinemic indices among Iranian adult population

BackgroundThere is no study regarding developing a valid index to predict insulin-related disorders in the Iranian population based on their dietary habits and lifestyle. In the current study, we aimed to develop and validate insulinemic potential indices of diet and lifestyle in Iranian adults.MethodsIn this cross-sectional study, we analysed data of 1063 men and women aged ≥ 25 years among participants of the examination three of Tehran lipid and glucose study (TLGS) (2006–2008). Dietary intakes were assessed using a valid semi-quantitative food frequency questionnaire. Dietary and lifestyle indices were developed using stepwise linear regression analysis based on dietary intakes, body mass index, and physical activity data. Fasting serum insulin concentration and homeostatic model assessment for insulin resistance (HOMA-IR) were used as biomarkers of hyperinsulinemia (HI) and insulin resistance (IR). Validation analyses were performed in examination four of TLGS.ResultsWe developed four indices related to insulin homeostasis, including the dietary index for HI (DIH), the dietary index for IR (DIR), the lifestyle index for HI (LIH), and the lifestyle index for IR (LIR). Based on multivariable-adjusted models, the relative values of the biomarker in subjects in the highest quartile of indices were 45% for LIH (95% CI 1.36–1.55, Ptrend < 0.001), 28% for DIR (95% CI 1.13–1.42, Ptrend = 0.019), and 51% for LIR (95% CI 1.41–1.61, Ptrend < 0.001), higher than those in the reference quartile, respectively.ConclusionWe designed and validated indices to determine the insulin potential of diet and lifestyle for the Iranian population, according to Iran’s demographic and dietary intake characteristics.

Impact of Polycystic Ovary Syndrome on Silent Coronary Artery Disease and Cardiovascular Events; A Long-term Population-based Cohort Study

The existing data regarding the impact of Polycystic Ovary Syndrome (PCOS) on the risk of developing cardiovascular disease (CVD) are conflicting. To explore the effect of PCOS status on the occurrence of silent coronary artery disease (CAD)/CVD. A total of 1591 women without CVD at baseline, aged 18-45 years, including 356 PCOS patients (defined by the Rotterdam criteria) and 1235 eumenorrheic non-hirsute women without polycystic ovarian morphology (controls), were selected from the Tehran Lipid and Glucose Study (TLGS). The median follow-up was 15.4 years, and most participants were in their late reproductive years at the end of the study. Silent CAD and CVD outcomes in PCOS and control groups were compared according to the multivariable-adjusted hazard ratios (HRs) and cumulative hazard functions. There was no difference in CVD risk factors between the PCOS and control groups. After controlling for confounders, PCOS status did not increase the risk of silent CAD (HR: 0.96, 95% CI 0.86-1.08). Regardless of PCOS status, women with a history of silent CAD showed 2.25 times higher CVD events than those without this history (95% CI 1.63-3.10). PCOS status reduced the CVD incidence by 42%, independently of silent CAD or traditional risk factors (HR: 0.58, 95% CI 0.35-0.98). Whereas silent CAD, regardless of PCOS, accelerated CVD, PCOS preserved it, most likely due to a combination of protective factors, including the endocrine pattern in the late reproductive period, environmental/social elements, and recruiting additional counseling and lifestyle modifications.

Age at natural menopause in women with a history of chronic diseases–A population-based cohort study

ObjectiveThis study was carried out to determine whether chronic diseases can influence age at natural menopause (ANM). Study design and main outcome measuresThis population-based prospective study recruited all eligible women from participants in the Tehran Lipid and Glucose Study (TLGS). The Cox proportional-hazards model (unadjusted and adjusted) was used to assess associations of ANM with chronic diseases, including diabetes mellitus (DM), hypertension (HTN), dyslipidemia, metabolic syndrome (MetS), cardiovascular disease (CVD), chronic kidney disease (CKD), and thyroid disorders, which were not included simultaneously in the model because there were correlations between chronic diseases. ResultsA total of 4662 women, aged 15–50 years, who had not reached menopause at initiation of the TLGS were included. Of these women, 1220 (26.2%) experienced menopause during the study follow-up. In the multivariable Cox regression model adjusted for potential confounders, women with DM [HR:1.75; 95%CI: (1.32, 2.32)], HTN [HR: 2.11; 95%CI: (1.64, 2.72)], MetS [HR: 2.01; 95%CI: (1.57, 2.58)], CVD [HR: 3.02; 95%CI: (1.93, 4.73)], CKD [HR: 2.64; 95%CI: (2.16, 3.22)], and thyroid disorders [HR: 1.41; 95%CI: (1.06, 1.88)] reached menopause earlier, while dyslipidemia was not associated with ANM. ConclusionThis study demonstrates that women with a history of chronic diseases, such as metabolic disorders, cardiovascular events, and CKD, experience menopause earlier than their counterparts without such a history.

Changes in dairy product consumption and subsequent type 2 diabetes among individuals with prediabetes: Tehran Lipid and Glucose Study

BackgroundPeople with prediabetes can postpone or even reverse progression to type 2 diabetes (T2D) by making dietary changes. This study aimed to examine the association of changes in consumption of total and specific types of dairy products with the subsequent risk of incident T2D among individuals with prediabetes.MethodThis cohort study included 639 individuals (50% female, mean age 47.3 years) of the Tehran Lipid and Glucose Study (TLGS) who had prediabetes at baseline. We assessed 3-year changes in the consumption of dairy products using a food frequency questionnaire. Using multivariable logistic regression, odds ratios (OR) and 95% confidence intervals (CI) were calculated for the association of changes in intake of total and subtypes of dairy products during a 3-year interval with the risk of incident T2D in the subsequent 3 years.ResultsAfter almost 9 years of follow-up, the incidence of T2D was 25.2%. Compared with individuals whose intake remained relatively stable over 3 years, those who decreased consumption of total dairy (> 0.5 servings/day) had a higher T2D risk (OR = 1.56; 95% CI: 1.02 to 2.41). Increasing low-fat dairy consumption by 0.50 serving/d was associated with a lower risk of T2D (OR = 0.56; 95% CI: 0.35 to 0.90) compared with stable consumption. Those who increased consumption of low-fat milk (OR = 0.59; 95% CI: 0.37 to 0.92) and low-fat yogurt (OR = 0.55; 95% CI: 0.33 to 0.93) had a lower risk of T2D than those who were relatively stable in their consumption. Replacing low-fat milk and yogurt with regular cheese was associated with 66 and 47% higher risk of T2D, respectively.ConclusionIn individuals with prediabetes, increasing consumption of low-fat dairy, low-fat milk, and low-fat yogurt had reduced risk of subsequent T2D. These data suggest a role of low-fat dairy products in the prevention of T2D among prediabetes patients.

Health-related quality of life in men and women who experienced cardiovascular diseases: Tehran Lipid and Glucose Study

BackgroundCardiovascular diseases (CVDs) are among the most common causes of death worldwide, including in Iran. Considering the adverse effects of CVDs on physical and psychosocial health; this study aims to investigate the association between experience of CVDs and health-related quality of life (HRQoL) in adult participants of the Tehran Lipid and Glucose Study (TLGS).MethodsThe participants of this cross-sectional study were 7009 adults (≥ 20 years) who participated in the TLGS during 2014–2017. Demographic information and HRQoL data was collected through validated questionnaires by trained interviewers. HRQoL was assessed by the Iranian version of the SF-12 questionnaire. Data was analyzed using the SPSS software.ResultsThe mean age of participants was 46.8 ± 14.6 years and 46.1% of them were men. A total of 9.0% of men and 4.4% of women had CVDs. In men, the mean physical HRQoL summary score was significantly lower in those with CVDs compared to those without CVDs (46.6 ± 0.8 vs. 48.5 ± 0.7, p > 0.001). In women, the mean mental HRQoL summary scores was significantly lower in those with CVDs compared to those without CVDs (42.8 ± 1.0 vs. 45.2 ± 0.5, p = 0.009). In adjusted models, men with CVDs were more likely to report poor physical HRQoL compared to men without CVDs (OR(95%CI): 1.93(1.32–2.84), p = 0.001); whereas for women, the chance of reporting poor mental HRQoL was 68% higher in those with CVDs than those without CVDs (OR(95%CI): 1.68(1.11–2.54), p = 0.015).ConclusionThe findings of the current study indicate poorer HRQoL in those who experienced CVDs compared to their healthy counterparts with a sex specific pattern. While for men, CVDs were associated with more significant impairment in the physical dimension of HRQoL, women experienced a similar impairment in the mental dimension of HRQoL.